Emerging Topics in Ecotoxicology

Principles, Approaches and Perspectives

Volume 4
Series Editor
Lee R. Shugart
L.R. Shugart and Associates, Oak Ridge, TN, USA

For further volumes:

http://www.springer.com/series/7360

Bryan W. Brooks Duane B. Huggett

!

Editors

Human Pharmaceuticals
in the Environment
Current and Future Perspectives

Editors
Bryan W. Brooks
Baylor University
Waco, Texas, USA

Duane B. Huggett
University of North Texas
Denton, Texas, USA

ISSN 1868-1344
ISSN 1868-1352 (electronic)
ISBN 978-1-4614-3419-1
ISBN 978-1-4614-3473-3 (eBook)
DOI 10.1007/978-1-4614-3473-3
Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 201293197
Springer Science+Business Media, LLC 2012
All rights reserved. This work may not be translated or copied in whole or in part without the written
permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York,
NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in
connection with any form of information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they
are not identified as such, is not to be taken as an expression of opinion as to whether or not they are
subject to proprietary rights.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Contents

Perspectives on Human Pharmaceuticals in the Environment ...................

Bryan W. Brooks, Jason P. Berninger, Alejandro J. Ramirez,
and Duane B. Huggett
Environmental Risk Assessment for Human Pharmaceuticals:
The Current State of International Regulations ..........................................
Jrg Oliver Straub and Thomas H. Hutchinson

17

Regulation of Pharmaceuticals in the Environment: The USA ..................

Emily A. McVey

49

Environmental Fate of Human Pharmaceuticals .........................................

Alistair B.A. Boxall and Jon F. Ericson

63

Environmental Comparative Pharmacology: Theory

and Application ...............................................................................................
Lina Gunnarsson, Erik Kristiansson, and D.G. Joakim Larsson

85

A Look Backwards at Environmental Risk Assessment:

An Approach to Reconstructing Ecological Exposures ............................... 109
David Lattier, James M. Lazorchak, Florence Fulk, and Mitchell Kostich
Considerations and Criteria for the Incorporation of
Mechanistic Sublethal Endpoints into Environmental
Risk Assessment for Biologically Active Compounds .................................. 139
Richard A. Brain and Bryan W. Brooks
Human Health Risk Assessment for Pharmaceuticals in the
Environment: Existing Practice, Uncertainty, and Future Directions ....... 167
E. Spencer Williams and Bryan W. Brooks

vi

Contents

Wastewater and Drinking Water Treatment Technologies ......................... 225

Daniel Gerrity and Shane Snyder
Pharmaceutical Take Back Programs ........................................................... 257
Kati I. Stoddard and Duane B. Huggett
Appendix A. Take Back Program Case Studies ........................................... 287
Index ................................................................................................................. 297

Contributors

Jason P. Berninger Department of Environmental Science, Center for Reservoir

and Aquatic Systems Research, Institute of Biomedical Studies, Baylor University,
Waco, TX 76798, USA
Office of Research and Development, National Health and Environmental Effects
Research Laboratory, U.S. Environmental Protection Agency, Duluth, MN 55804, USA
Alistair B.A. Boxall Environment Department, University of York, Heslington,
York YO10 5DD, UK
Richard A. Brain Ecological Risk Assessment, Syngenta Crop Protection LLC,
Greensboro, NC 27409, USA
Bryan W. Brooks Department of Environmental Science, Center for Reservoir
and Aquatic Systems Research, Institute of Biomedical Studies, Baylor University,
Waco, TX 76798, USA
Jon F. Ericson Pfizer Global Research and Development, Worldwide PDM,
Environmental Sciences, MS: 8118A-2026, Groton, CT 06340, USA
Florence Fulk National Exposure Research Laboratory, Ecological Exposure
Research Division, US Environmental Protection Agency, Office of Research and
Development, Cincinnati, OH 45268, USA
Daniel Gerrity Water Quality Research and Development Center, Southern
Nevada Water Authority, River Mountain Water Treatment Facility, Henderson,
NV 89015, USA
Lina Gunnarsson Department of Neuroscience and Physiology, Institute of
Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg,
405 30 Gteborg, Sweden
Duane B. Huggett Department of Biological Sciences, University of North Texas,
Denton, TX 76203, USA

vii

viii

Contributors

Thomas H. Hutchinson CEFAS Weymouth Laboratory, Centre for Environment,

Fisheries and Aquaculture Sciences, Weymouth, Dorset DT4 8UB, UK
Mitchell Kostich National Exposure Research Laboratory, Ecological Exposure
Research Division, US Environmental Protection Agency, Office of Research and
Development, Cincinnati, OH 45268, USA
Erik Kristiansson Department of Neuroscience and Physiology, Institute of
Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg,
405 30 Gteborg, Sweden
Department of Zoology, University of Gothenburg, 405 30 Gteborg, Sweden
D.G. Joakim Larsson Department of Neuroscience and Physiology, Institute of
Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg,
405 30 Gteborg, Sweden
David Lattier National Exposure Research Laboratory, Ecological Exposure
Research Division, US Environmental Protection Agency, Office of Research and
Development, Cincinnati, OH 45268, USA
James M. Lazorchak National Exposure Research Laboratory, Ecological
Exposure Research Division, US Environmental Protection Agency, Office of
Research and Development, Cincinnati, OH 45268, USA
Emily A. McVey Office of Pharmaceutical Science, Center for Drug Evaluation
and Research, U.S. Food and Drug Administration, Silver Spring, MD 20993,
USA
WIL Research, 5203DL s-Hertogenbosch, The Netherlands
Alejandro J. Ramirez Mass Spectrometry Center, Mass Spectrometry Core
Facility, Baylor University, Baylor Sciences Building, Waco, TX 76798, USA
Shane Snyder Chemical and Environmental Engineering, University of Arizona,
Tucson, AZ 85721, USA
Jrg Oliver Straub F.Hoffmann-La Roche Ltd, Group SHE, LSM 49/2.033,
Basle CH-4070, Switzerland
Kati I. Stoddard Department of Biological Sciences, University of North Texas,
Denton, TX 76203, USA
E. Spencer Williams Department of Environmental Science, Institute of
Biomedical Studies, Center for Reservoir and Aquatic Systems Research, Baylor
University, Waco, TX 76798-7266, USA

Perspectives on Human Pharmaceuticals

in the Environment
Bryan W. Brooks, Jason P. Berninger, Alejandro J. Ramirez,
and Duane B. Huggett

Background
Human interaction with the environment remains one of the most pervasive facets
of modern society. Whereas the anthropocene is characterized by rapid population growth, unprecedented global trade and digital communications, energy
security, natural resource scarcities, climatic changes and environmental quality,
emerging diseases and public health, biodiversity and habitat modifications are
routinely touted by the popular press as they canvas global political agendas and
scholarly endeavors. With a concentration of human populations in urban areas
B.W. Brooks (*)
Department of Environmental Science, Center for Reservoir and Aquatic Systems Research,
Institute of Biomedical Studies, Baylor University, One Bear Place, #97266,
Waco, TX 76798, USA
e-mail: [email protected]
J.P. Berninger
Department of Environmental Science, Center for Reservoir and Aquatic Systems Research,
Institute of Biomedical Studies, Baylor University, One Bear Place, #97266,
Waco, TX 76798, USA
National Health and Environmental Effects Research Laboratory, National Research Council
Research Associates Program, Office of Research and Development, U.S. Environmental
Protection Agency, 6201 Congdon Boulevard, Duluth, MN 55804, USA
e-mail: [email protected]
A.J. Ramirez
Mass Spectrometry Center, Mass Spectrometry Core Facility, Baylor University,
Baylor Sciences Building, One Bear Place, #97046, Waco, TX 76798, USA
e-mail: [email protected]
D.B. Huggett
Department of Biological Sciences, University of North Texas,
1155 Union Circle, #305220, Denton, TX 76203, USA
e-mail: [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_1, Springer Science+Business Media, LLC 2012

B.W. Brooks et al.

unlike any other time in history, the coming decades will be defined by A New
Normal, as proposed by Postel [1], where the interplay among sustainable
human activities and natural resource management will inherently determine the
regional fates of human societies.
In recent years, few topics have captured the publics attention like the presence of human pharmaceuticals in environment. Fish on Prozac [2, 3]. Male fish
becoming female [4, 5]? Drugs found in drinking water [6, 7]. Indias drug
problem [8]. Chances are you have seen these headlines or read related reports.
Pharmaceuticals and trace levels of other contaminants (e.g., antibacterial agents,
flame retardants, perfluorinated surfactants, harmful algal toxins) are increasingly
reported in freshwater and coastal ecosystems. In the developed world, many of
these chemicals are released at very low levels (e.g., parts per trillion) from wastewater effluent discharges to surface and groundwaters. But why were citizens so
engaged by stories about fish on Prozac [3] and drugs in drinking water [7]?
Because pharmacotherapy is now entrenched in everyday life, a realization that
common drugs were found in the water we drink or the fish we eat likely produces
a boomerang effect, where our daily reliance on well-accepted therapies was concretely linked in a new way with their potential consequences to the natural world.
On an increasingly urban planet, pharmaceutical residues and traces of other
contaminants of emerging concern represent signals of the rapidly urbanizing
water cycle and harbingers of the New Normal.
Over the past 2 decades the implications of endocrine disruption and modulation have permeated public consciousness, scientific inquiry, regulatory frameworks, and management decisions in the environmental and biomedical sciences.
Publication of Colburn, Dumanoski, and Myers Our Stolen Future [9], which
is often referred to as the second coming of Rachel Carsons Silent Spring [10],
stimulated the public, scientific, and regulatory attention given to endocrine disruptors and ultimately influenced the environmental studies of human pharmaceuticals [11]. For example, human reproductive developmental perturbations
elicited by the estrogenic human pharmaceutical diethylstilbestrol and feminization of male fish exposed to municipal effluent discharges represent examples of
causal relationships among endocrine active substances and biologically important
adverse outcomes [12].
In the late 1990s, research in the area of endocrine disruption was taking off,
particularly to identify constituents of effluents or other environmental matrices that
were potentially responsible for endocrine perturbations in wildlife and humans.
Because many xenoestrogens are present in effluent discharges, initial investigations in the UK employed toxicity identification evaluation studies to fractionate
and identify causative components of the complex mixtures inherent with effluents
[13]. At the same time in the USA, Arcand-Hoy et al. [14] highlighted the importance of considering human estrogen agonist and veterinary androgen agonist pharmaceuticals as potential causative toxicants from point and nonpoint source
effluents. Also in 1998, two of the first review papers on pharmaceuticals in the
environment, by Halling-Sorensen et al. [15] and Ternes [16], appeared in the literature. In 1999, another review paper, by Daughton and Ternes [17], considered

1.0

2500

0.8

2000

0.6

1500

0.4

1000

0.2

500

0.0

0
1998

2000

2002

2004

2006

2008

Cumulative Frequency of Citations

Relative Cumulative Frequency of Citations

Perspectives on Human Pharmaceuticals in the Environment

2010

Year

Fig. 1 Representative increase in peer-reviewed publications related to pharmaceuticals in the

environmental through 2010, summarized by the cumulative and relative cumulative citation
frequency of early review papers by Halling-Sorensen et al. [15], Ternes [16], and Daughton and
Ternes [17]. Citation information from Web of Knowledge

Pharmaceuticals and Personal Care Products (PPCP) in the environment and by

doing so coined the PPCP acronym, which remains pervasive. Subsequently, a precipitous number of workshops, symposia, special meetings, and publications related
to pharmaceuticals in the environment have occurred. For example, Fig. 1 describes
citation frequencies of just the Halling-Sorensen et al. [15], Ternes [16], and
Daughton and Ternes [17] papers as surrogates for the trajectory of scientific inquiry
in this important area of environmental science and public health.
Some of the most important developments related to pharmaceuticals in the environment included special issues of Toxicology Letters in 2002 and 2003, Pellston
workshops by the Society of Environmental Toxicology and Chemistry (SETAC) on
human pharmaceuticals (in 2003 [18]) and veterinary medicines (in 2007 [19]),
formation of the SETAC Pharmaceuticals Advisory Group (in 2005; http://www.
setac.org/node/34) and the Water Environment Federations Microconstituents
Community of Practice (http://www.wef.org), International Conferences on the
Occurrence, Fate, Effects, and Analysis of Emerging Contaminants in the
Environment (e.g., htpp://www.EmCon2011.com), the International Water
Associations MicroPol conferences (e.g., htpp://www.micropol2011.org), and a
special issue of Environmental Toxicology and Chemistry entitled Pharmaceuticals
and Personal Care Products in the Environment in 2009. Following an editorial by
Brooks et al. [20] entitled Pharmaceuticals and Personal Care Products: Research
Needs for the Next Decade, an international workshop entitled Effects of
Pharmaceuticals and Personal Care Products in the Environment: What are the Big
Questions? was held by Health Canada/SETAC in April 2011 [21]. In 2012, the
SETAC Pharmaceutical Advisory Group is planning another Pellston conference on
antimicrobial resistance, which represents a major threat to global public health.
Though the information in this timely area continues to rapidly expand, it appears

B.W. Brooks et al.

critically important to now consider the lessons learned from the study of human
pharmaceuticals in the environment and formulate directions for future efforts.

Environmental Analysis and Exposure

To date, the majority of information for human pharmaceuticals in the environment
is related to occurrence in various environmental matrices, which largely accounts
for publication trends summarized in Fig. 1. Perhaps the most influential paper on
occurrence was published by Kolpin et al. [22]. In 2002, this landmark article provided the first national reconnaissance study of a variety of contaminants of emerging concern, including a number of pharmaceuticals, in water [22] and promises to
be the most heavily cited paper published in the history of the journal Environmental
Science & Technology. In Table 1, we provide an overview of the representative
literature related to the environmental analysis and occurrence of pharmaceuticals
in the environment. Instead of performing an exhaustive survey and synthesis here,
we instead relay some perspectives on environmental analysis and refer readers to
the recent review of occurrence information for human pharmaceuticals by Monteiro
and Boxall [23].
Table 1 Representative recent reviews on pharmaceutical analysis in various environmental
matrices
Target analytes
Matrix
Type of review
Pharmaceuticals

Water

Solidsa
Water, solids

Analytical methods [64], multiresidue

methods [65], LCMS/MS methods [66],
basic pharmaceuticals [67], antibiotics
[68], anti-inflammatory drugs [69],
recent advances [70]
LCMS/MS [71], tetracycline antibiotics [72]
Analytical methods [73], LCMS/MS
methods [74]

Conventional and/or
contaminants of
emerging concern,
including
pharmaceuticals

Water
Water, solids
Various environmental
matrices

Analytical methods [75, 76]

LCMS in environmental analysis [77]
Analytical methods [78, 79], methods
applied to fate [80], environmental mass
spectrometry [81], recent advances [82]

Pharmaceuticals
and/or degradation
products

Water

Advanced MS techniques [83], LCMS

methods [84], methods applied to fate
and removal [85]
Mass spectrometry [86], analytical problems
and sample preparation [87]

Various environmental
matrices
Other reviews related
to pharmaceutical
analysis and
general occurrence
information
Sediment, biosolids and soil

Multivariate analysis [88, 89], sampling

and/or extraction [9094], chiral analysis
[95], general occurrence [23], biological
tissues [28, 29, 96]

Perspectives on Human Pharmaceuticals in the Environment

Gas chromatographymass spectrometry (GCMS) was the primary analytical

tool used to assess the environmental occurrence of PPCPs in initial studies (Table 1).
The popularity of GCMS in early work was due to its widespread availability and
historical use in contract service laboratories for historical industrial chemical
contaminants. The availability of electron-impact spectral libraries was initially
important, as they increased confidence in analyte identification. Further, the distinctive nonpolar operating range of GCMS was consistent with analysis of most
personal care products (PCPs). In contrast, the use of GCMS for analysis of pharmaceuticals, which are relatively polar compared to most PCPs, typically requires
derivatization prior to analysis. For example, Brooks et al. [3] employed GCMS
with derivatization for initial identification of the antidepressants sertraline and
fluoxetine in fish tissue. However, derivatization reactions are often unpredictable
for complex samples and can limit the quality of quantitative data. Consequently,
liquid chromatographymass spectrometry (LCMS) has become the technique of
choice for analyzing pharmaceuticals in environmental samples.
Numerous studies have demonstrated the distinct advantages of LCMS for
analysis of pharmaceuticals (Table 1). LCMS enables identification and
quantification without derivatization and typically results in lower detection limits
(below 1 ng/L and 1 ng/g for liquid and solid samples, respectively) and better
precision than comparable GCMS methodologies. In environmental applications,
LC is typically combined with tandem MS (or MS/MS) to promote enhanced
selectivity and sensitivity for target analytes. In a routine MS/MS analysis, a
molecular ion is selected and subsequently fragmented to produce one or more
distinctive product ions that enable both qualitative and quantitative monitoring.
Recently introduced ultraperformance liquid chromatography (UPLC) provides a
novel approach to chromatographic separation. UPLC differs from regular LC by
the implementation of chromatographic columns with smaller particle diameters
(i.e., sub-2-mm particles), which generates elevated back pressures and narrower
chromatographic peaks. The overall effect is resolved peaks in shorter periods of
time with increased sensitivity. UPLC requires fittings and pumps designed to support high back pressures, which increases the price of the LC system. An important
feature of UPLC is the need of a fast detector to account for small peak widths
(ca. 10 s). In other words to acquire enough data points through chromatographic
peaks, selected mass spectrometer need to collect data points at high sampling
rates. Q-TOF mass spectrometers are often coupled with UPLC systems due to
their fast sampling rates. It is important to note, however, that LCMS is not exempt
from limitations. One of the limitations of LCMS is that atmospheric pressure
ionization (API) processes are influenced by coextracted matrix components.
Matrix effects typically result in suppression or less frequent enhancement of analyte signal. There have been a number of methods proposed to compensate for
matrix effects, including the method of standard addition, surrogate monitoring,
and isotope dilution (Table 1). Although isotope dilution is the most highly recommended approach for analysis of human pharmaceuticals in environmental matrices, isotopically labeled standards are not always readily available for these target
analytes. A further limitation is the paucity of available isotopically labeled standards

B.W. Brooks et al.

for therapeutic metabolites. An alternative approach involves the use of an

appropriate internal standard (i.e., a structurally similar compound expected to
mimic the behavior of a target analyte(s)) with or without matrix-matched calibration. However, a given internal standard is typically effective over a limited retention time window. Accordingly, the use of more than one internal standard is
recommended to compensate for matrix effects throughout the chromatographic
run. Finally, it is important to point out that strategies to compensate for matrix
effects should take into account the variability of matrix within each set of samples
to be analyzed (e.g., surface water, effluent, sediment, fish tissue).
Due to potential regulatory implications of human pharmaceuticals in the environment, environmental analyses typically include rigorous quality assurance and
quality control (QA/QC) metrics to confirm reliability of analytical data. Initial
method validation provides essential performance parameters, such as method
recoveries, precision, and limits of detection (LODs). Recurring analysis of quality
control (QC) samples (e.g., method blanks, matrix spikes, laboratory control samples) is important to verify performance of the method over time, and to assess
potential matrix effects. Considering the unpredictable nature of matrix interference
in LCMS analysis and the lack of effective strategies to deal with this difficulty, it
has become imperative to use QA/QC data to document and qualify analytical
results for human pharmaceuticals in environmental matrices. This is particularly
important when reporting concentrations at or near the limit of detection for a given
analytical method.
In this volume, an overview of global environmental regulatory activities relevant to human pharmaceuticals is provided in Chaps. 2 and 3. In Chap. 4, Boxall
and Ericson examine important considerations for understanding the environmental
fate of therapeutics. Below we provide some perspectives on bioaccumulation and
effects of human pharmaceuticals in the environment.

Environmental Bioaccumulation and Effects

Though the potential for uptake of veterinary medicines by animals reared in aquaculture were understood for some time (see [24, 25]), Boxall et al.s [26] study of
the uptake of veterinary medicines from soils to plants highlighted the importance
of considering potential accumulation of human medicines in terrestrial organisms
because biosolids and effluents from wastewater treatment plants can be applied
to agricultural fields. Such observations are particularly relevant for antibiotics.
In fact, developing an understanding of the influences of human antibiotics and
antimicrobial agents on antibiotic resistance was recently identified as critical areas
of research need for environmental science and public health [21].
In aquatic systems, Larsson et al. [27] likely provided the first report of bioaccumulation of a human pharmaceutical, 17a-ethinylestradiol, in bile of fish exposed
to Swedish effluent discharges. Brooks et al.s [3] findings of the antidepressants
fluoxetine and sertraline (and their primary metabolites) in brain, liver, and muscle

Perspectives on Human Pharmaceuticals in the Environment

tissues of three fish species from an effluent-dominated stream (a.k.a. fish on

Prozac) appear to represent the second report in the literature of accumulation of
human pharmaceuticals in wildlife and the first observation from North America.
Such observations stimulated research related to the accumulation and effects of
human pharmaceuticals in the environment and subsequently shaped the National
Pilot Study of PPCPs in Fish Tissue by the US Environmental Protection Agency
[28]. This study by Ramirez et al. [28] provided the first evidence of bioaccumulation of a number of human pharmaceuticals in fish collected across a broad geographic area. A summary of research on bioaccumulation of pharmaceuticals in
aquatic organisms recently highlighted the need to understand thresholds of drug
accumulation associated with adverse effects [29]. Unfortunately, an understanding of human pharmaceuticals accumulating in terrestrial wildlife is poorly understood [20] but has been recently identified as a major research question [21].
Several recent publications have started to further our understanding of the bioconcentration/bioaccumulation potential of pharmaceuticals in a laboratory setting, as
well as publications aimed at understanding pharmaceutical metabolism in wildlife
and its role in the accumulation of drugs [3039]. Below we introduce important
considerations for understanding relationships between pharmaco(toxico)kinetics
and -dynamics of human medications in aquatic and terrestrial organisms. A more
thorough examination of comparative pharmacological approaches for environmental
applications is provided by Gunnarsson et al. in Chap. 5.
Understanding the environmental risks posed by historical contaminants has
been challenged by the paucity of toxicity information available for most industrial
chemicals [40]. In the case of human pharmaceuticals, however, intensive investigations occur prior to distribution, which yields a wealth of pharmacological and toxicological data compared to other industrial contaminants. To illustrate available
data, Table 2 provides a summary of common characteristics for hundreds of pharmaceuticals. During the design of therapeutics, careful consideration is given to
target-specific biomolecules (e.g., receptors, enzymes) and pathways to elicit
beneficial outcomes. Because side effects are not desirable and large margins of
safety (relationship between therapeutic and toxic doses) are ideal, pharmaceutical
development often results in therapeutics with relative well-understood mechanisms/modes of actions (MOAs) and very low acute toxicity in mammals. For
example, a recent study predicted that less than 8% of all pharmaceuticals are
expected to be classified as highly acutely toxic to rodent models [41]. Similarly,
Berninger and Brooks [41] predicted that less than 6% of all pharmaceuticals are
acutely toxicity to fish below 1 mg/L.
As noted previously, concentrations of individual human pharmaceuticals in
surface water of developed countries rarely exceed parts per billion levels; thus,
limited acute toxicity is expected in surface waters of the developed world.
Unfortunately, most studies to date have only examined acute toxicity in standard
aquatic organisms [42]. However, chronic adverse responses resulting from therapeutic MOAs are more likely to be observed in the environment [41], particularly
in systems with instream flows dominated by continuous release of effluent discharges [43] leading to longer effective exposure durations [11]. Early investigators

8.6

797

145,781

1,042

Max

0.95
2.03
5.01

164
346
732

9.4

10th
50th
90th

6.94

1,035

56,000

77
971
12,283

0.00075

832

330

0.0017
0.1300
9.91

7.5 106

741

4.7 108

97
8,127
681,657

1.6

936

1,070

0.49
3.71
27.9

0.0029

979

87,600

0.77
5.01
32.6

0.033

944

2,348

0.15
1.03
6.96

0.035

831

9.1 109

2.9 105
0.0446
69.4

8.4 1010

MW molecular weight (g/mol); log P octanolwater partitioning coefficient; LD50 median oral lethal dose for rat model (mg/kg); Cmax human peak plasma
concentration (or therapeutic dose; mg/mL); ATR acute to therapeutic ratio margin of safety analog (LD50/Cmax; see Berninger and Brooks [41]); Cl clearance rate (mg/min/kg); T half-life of elimination (hour); Vd apparent volume of distribution (L/kg); AqET is the aqueous effect threshold (mg/L) where
fish plasma BCF/Cmax = aquatic exposure concentration at the point in which Cmax = fish plasma concentration and fish plasma BCF exposure concentration = fish plasma concentration [29]

Centiles

Min

Table 2 A summary of the minimum and maximum values and 10th, 50th, and 90th centiles of common properties associated with pharmaceuticals
MW
log P
LD50
Cmax
ATR
Cl
T
Vd
AqET

8
B.W. Brooks et al.

Perspectives on Human Pharmaceuticals in the Environment

recognized the importance of leveraging mammalian pharmacological safety data

to help understand various pharmaceutical effects in the environment, because
many MOAs of human therapeutics appear to be evolutionarily conserved, particularly
in vertebrates [14, 4446].
In 2003, Huggett et al. [47] proposed a screening approach to identify pharmaceuticals in water that may result in fish plasma levels (or internal doses) human
therapeutic levels (e.g., Cmax). Huggetts plasma model was based on three core
assumptions: (1) Evolutionary conservation of structure and function of drug targets
among mammals and fish species; (2) Internal fish doses approaching mammalian
Cmax levels would result in similar therapeutic outcomes; and (3) A gill uptake model
[48] for predicting rainbow trout plasma concentrations following waterborne exposure to nonionizable chemicals [48]. Subsequently, several recent studies have
employed the Huggett et al. plasma model approach [4951] or conceptually similar
variations to account for ionization influences on bioavailability [29, 52, 53]. Of
particular importance, Valenti et al. [53] recently provided an independent validation of the Huggett et al. [47] plasma model when ionization of the weak base sertraline [54] and an alternative gill uptake model [48] was considered. Valenti et al.
[53] also employed an adverse outcome pathway (AOP) design [55], which included
quantification of binding at the therapeutic target and anxiety-related behavioral
responses stereotypical of the therapeutic efficacy of this model antidepressant. In
the Valenti et al. [53] study, adult male fathead minnow were exposed via aqueous
exposure to sertraline for 21 days. Fish plasma concentrations were accurately predicted from water exposures when pH influences on ionization and lipophilicity
were considered [29, 52, 54]. When these plasma levels in fish exceeded the human
therapeutic dose (Cmax) of sertraline, binding to the serotonin reuptake transporter
and antianxiety behavior were significantly affected [53]. The AOP approach was
recently proposed by Ankley et al. [55] for linking molecular initiation events, such
as those related to pharmaceutical interactions with a target site (e.g., a receptor),
with cascading events leading to adverse outcomes at the individual and population
level, which can be used as measures of effect in risk assessments. As demonstrated
by Valenti et al. [53], linking predictions of uptake from surface waters to fish
plasma with conceptual AOP models appear to represent a sound foundation from
which potentially hazardous human pharmaceuticals may be identified.
Probabilistic hazard assessment approaches, which are commonly used to support environmental and public health decision making, can use existing mammalian
pharmacological safety data to develop predictive models for various parameters
[41]. These predictive tools can support prioritization activities for testing hypotheses regarding pharmacological parameters of various drug classes or chemical
specific computational attributes that may result in hazards to wildlife [41]. For
example, Table 2 presents the minimum and maximum values and 10th, 50th and
90th centiles of probabilistic pharmaceutical distributions (PPD) of molecular
weight, logP, acute LD50, Cmax, acute to therapeutic ratio margin of safety analog
(LD50/Cmax; see [41]), clearance rate, half-life of elimination, apparent volume of
distribution (Vd), and the aqueous effect threshold (AqET; see [52]) based on data
from hundreds of pharmaceuticals. PPD approaches can be used to predict the

10

B.W. Brooks et al.

99.99
99.9
99

Percent Rank

90
70
50
30
10
1
0.1
0.01
10-3

10-2

10-1

100

101

102

103

104

Apparent Volume of Distribution (L/kg)

Fig. 2 Probabilistic pharmaceutical distribution of apparent volume of distribution (L/kg) for 944
pharmaceuticals. Reference lines relate to the 10th, 50th and 90th centiles (Table 2), which correspond to 0.15, 1.03, and 6.96 L/kg, respectively. For example, apparent volume of distribution is
predicted by this model to be at or above 6.96 L/kg for 10% of all pharmaceuticals

likelihood of encountering another therapeutic with attributes of interest. To illustrate the utility of PPD analyses, Fig. 2 depicts a PPD for Vd. Briefly, Vd data were
ranked and converted to probability percentages then plotted against respective
probability ranks on a log-probability scale; centiles were determined by regression
(see [30] for a complete description of methods). Using this approach, we predict
that 10% or less of all pharmaceuticals would have Vd values of 0.15 L/kg. In Fig. 3,
we extend the PPD assessment to predict the likelihood of encountering a pharmaceutical in surface waters exceeding the AqET value, which is based here on the
specific assumptions of Huggett et al.s [47] plasma model. For example, 10% of all
pharmaceuticals are predicted to result in internal fish plasma concentrations equaling the human Cmax value at or below an environmentally relevant surface water
concentration of 29 ng/L (Fig. 3, Table 2).
Based on the current state of the science, it appears critical to develop an advanced
understanding of the risks associated with human pharmaceuticals in the environment. In Chaps. 6 and 7, Lattier et al. consider mechanistic characteristics of drugs
for reconstructing environmental exposure scenarios and Brain and Brooks provide
perspectives for incorporating non-standard endpoints in environmental risk assessments, respectively. In Chap. 8, Williams and Brooks examine human health risk
assessment considerations for environmental exposures to therapeutics. When the
outcome of an environmental risk assessment identifies unacceptable risks to wildlife
or humans, risk management decisions and practices serve as interventions to
protect public health and the environment. In the case of pharmaceuticals and other

Perspectives on Human Pharmaceuticals in the Environment

11

99.99
99.9
99

Percent Rank

90
70
50
30
10
1
0.1
0.01
10-11 10-9 10-7 10-5 10-3 10-1 101

103

105

107

109

Aqueous Effect Threshold (mg/L)

Fig. 3 Probabilistic pharmaceutical distribution of aqueous effect threshold (AqET; mg/L) for 831
pharmaceuticals. Reference lines relate to the 10th, 50th, and 90th centiles (Table 2), which correspond to 29 ng/L, 44.6 mg/L, and 66.4 mg/L, respectively. For example, an aquatic concentration
leading to a plasma concentration in fish above the mammalian Cmax value is predicted by the
AqET model to be at or below 29 ng/L for 10% of all pharmaceuticals

contaminants in treated wastewater effluents, a number of treatment approaches,

including appropriately designed and maintained constructed wetlands [56], appear
viable for supporting risk management of indirect and direct potable water reuse.
In this volume, Chaps. 9 and 10 examine timely issues related to environmental risk
management. In Chap. 9, Gerrity and Snyder examine the available information
related to the efficacy of various wastewater and drinking water treatment technologies for human pharmaceuticals. In Chap. 10, Stoddard and Huggett conclude this
volume with an interesting perspective on pharmaceutical take back programs,
which promise to divert unused medications from down the drain discharges and
drug abuse by and poisonings of unintended users.
Lessons learned from human pharmaceuticals in the environment will continue to
advance our understanding of the environmental risks of chemicals. For example, a
number of organic contaminants are chiral, which remains an important environmental
consideration because fate and effects often differ among enantiomers [57]. Herein,
studies of chiral pharmaceuticals have advanced our understanding of risks posed by
other chiral chemicals [58]. Similarly, many environmental contaminants, including
metabolites and degradates, are weak acids and weak bases. Because site-specific pH
influences environmental fate, uptake and toxicity, the study of ionizable therapeutics
(~70% of all drugs are weak bases) has advanced our understandings of the impacts of
climatic changes on bioaccumulation and toxicity of moderately polar and ionizable
chemicals [59, 60]. Interestingly, lessons learned from the study and design of less-toxic

12

B.W. Brooks et al.

pharmaceuticals, often described as benign by design [61], can be extended to advance

green chemistry principles by developing sustainable molecular design guidelines for
reducing the toxicity of other industrial contaminants [62, 63]. To the fields of aquatic
toxicology and environmental risk assessment in particular, understanding the toxicity
of human pharmaceuticals in the environment is beginning to advance our understanding of toxicity pathways. To date, relatively few toxicity pathways have been defined in
ecological systems, but hundreds of pharmaceuticals targets are evolutionarily conserved across the various kingdoms. Developing an understanding of pharmaceutical
MOAs and associated AOPs will improve prospective and retrospective diagnosis and
management of environmental risks posed by industrial contaminants. Clearly a number of timely research questions remain unanswered [21].

References
1. Postel S (2010) Water: adapting to a new normal. In: Heinberg R, Lerch D (eds) The Post
Carbon Reader: managing the 21st Centurys Sustainability Crises. Watershed Media/
University of California Press, California
2. Fish on Prozac. http://news.sciencemag.org/sciencenow/2003/11/04-01.html
3. Brooks BW, Chambliss CK, Stanley JK, Ramirez AJ, Banks KE, Johnson RD, Lewis RJ
(2005) Determination of select antidepressants in fish from an effluent-dominated stream.
Environ Toxicol Chem 24:464469
4. Male fish becoming female? Researchers worry about estrogen and pollutants in the water.
http://www.msnbc.msn.com/id/6436617/ns/nightly_news/t/male-fish-becoming-female
5. Woodling JD, Lopez EM, Maldonado TA, Norris DO, Vajda AM (2006) Intersex and other
reproductive disruption of fish in wastewater effluent dominated Colorado streams. Comp
Biochem Phys Part C 144:1015
6. Drugs found in drinking water. http://hosted.ap.org/specials/interactives/pharmawater_site
7. Benotti MJ, Trenholm RA, Vanderford BJ, Holady JC, Stanford BD, Snyder SA (2009)
Pharmaceuticals and endocrine disrupting compounds in U.S. drinking water. Environ Sci
Technol 43:597603
8. Indias drug problem. http://www.nature.com/news/2009/090204/full/457640a.html
9. Colburn T, Dumanoski D, Myers JP (1996) Our stolen future. Dutton, Peguin Books, New York
10. Carson R (1962) Silent spring. Houghton Mifflin Company, New York
11. Ankley GT, Brooks BW, Huggett DB, Sumpter JP (2007) Repeating history: pharmaceuticals
in the environment. Environ Sci Technol 41:82118217
12. Hotchkiss AK, Rider CV, Blystone CR, Wilson VS, Hartig PC, Ankley GT, Foster PM,
Gray CL, Gray LE (2008) Fifteen years after Wingspreadenvironmental endocrine disrupters and human and wildlife health: where we are today and where we need to go. Toxicol
Sci 105:235259
13. Desbrow C, Routledge E, Brighty G, Sumpter J, Waldock M (1998) Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening.
Environ Sci Technol 32:15491558
14. Arcand-Hoy L, Nimrod AC, Benson WH (1998) Endocrine-modulating substances in the environment: estrogenic effects of pharmaceutical products. Int J Toxicol 17:139158
15. Halling-Sorensen B, Nielsen SN, Lanzky PF, Ingerslev F, Lutzhoft HCH, Jorgensen SE (1998)
Occurrence, fate and effects of pharmaceutical substances in the environmenta review.
Chemosphere 36:357394
16. Ternes TA (1998) Occurrence of drugs in German sewage treatment plants and rivers. Water
Res 32:32453260

Perspectives on Human Pharmaceuticals in the Environment

13

17. Daughton CG, Ternes TA (1999) Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspect 107(suppl 6):907938
18. Williams RT (ed) (2005) Human pharmaceuticals: assessing impacts on aquatic ecosystems.
SETAC Press, Pensacola, Florida
19. Crane M, Barrett K, Boxall A (eds) (2008) Veterinary medicines in the environment. SETAC
Press, Pensacola, Florida
20. Brooks BW, Huggett DB, Boxall ABA (2009) Pharmaceuticals and personal care products:
research needs for the next decade. Environ Toxicol Chem 28:24692472
21. Boxall ABA, Rudd M, Brooks BW, Caldwell D, Choi K, Hickmann S, Innes E, Ostapyk K,
Staveley J, Verslycke T, Ankley GT, Beazley K, Belanger S, Berninger JP, Carriquiriborde P,
Coors A, DeLeo P, Dyer S, Ericson J, Gagne F, Giesy JP, Gouin T, Hallstrom L, Karlsson M,
Larsson DGJ, Lazorchak J, Mastrocco F, McLaughlin A, McMaster M, Meyerhoff R, Moore
R, Parrott J, Snape J, Murray-Smith R, Servos M, Sibley PK, Straub JO, Szabo N, Tetrault G,
Topp E, Trudeau VL, van Der Kraak G (2012) Pharmaceuticals and personal care products in
the environment: what are the big questions? Environ Health Perspect (in press)
22. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton HT (2002)
Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams,
19992000: a national reconnaissance. Environ Sci Technol 36:12021211
23. Monteiro SC, Boxall ABA (2010) Occurrence and fate of human pharmaceuticals in the environment. Rev Environ Contam Toxicol 202:53154
24. Nordlander I, Johnsson H, Osterdahl B (1987) Oxytetracycline residues in rainbow trout analyzed by a rapid HPLC method. Food Addit Contam 4:291296
25. Capone DG, Weston DP, Miller V, Shoemaker C (1996) Antibacterial residues in marine sediments and invertebrates following chemotherapy in aquaculture. Aquaculture 145:5575
26. Boxall ABA, Johnson P, Smith EJ, Sinclair CJ, Stutt E, Levy LS (2006) Uptake of veterinary
medicines from soils into plants. J Agric Food Chem 54:22882297
27. Larsson DGJ, Adolfsson-Erici M, Parkkonen J, Pettersson M, Berg AH, Olsson PE, Forlin L
(1999) Ethinyloestradiolan undesired fish contraceptive? Aquat Toxicol 45:9197
28. Ramirez AJ, Brain RA, Usenko S, Mottaleb MA, ODonnell JG, Stahl LL, Wathen JB, Snyder
BD, Pitt JL, Perez-Hurtado P, Dobbins LL, Brooks BW, Chambliss CK (2009) Occurrence of
pharmaceuticals and personal care products (PPCPs) in fish: results of a national pilot study in
the U.S. Environ Toxicol Chem 28:25872597
29. Daughton CG, Brooks BW (2011) Active pharmaceuticals ingredients and aquatic organisms.
In: Meador J, Beyer N (eds) Environmental contaminants in wildlife: interpreting tissue concentrations, 2nd edn. Taylor and Francis, Boca Raton, pp 281341
30. Zhang X, Oakes KD, Cui S, Bragg L, Servos MR, Pawliszyn J (2010) Tissue-specific in vivo
bioconcentration of pharmaceuticals in rainbow trout (Oncorhynchus mykiss) using spaceresolved solid-phase microextraction. Environ Sci Technol 44:34173422
31. Paterson G, Metcalfe CD (2008) Uptake and depuration of the anti-depressant fluoxetine by
the Japanese medaka (Oryzias latipes). Chemosphere 74:125130
32. Nallani G, Paulos P, Vanables B, Constantine L, Huggett DB (2011) Bioconcentration of
Ibuprofen in Fathead minnow (Pimephales promelas) and Channel catfish (Ictalurus punctatus). Chemosphere 84:13711377
33. Nallani G, Paulos P, Vanables B, Constantine L, Huggett DB (2011) Tissue specific uptake and
bioconcentration of the oral contraceptive, Norethindrone, in two freshwater fishes. Arch
Environ Contam Toxicol 62(2):306313
34. Smith EM, Chu S, Paterson G, Metcalfe CD, Wilson JY (2010) Cross-species comparison of
fluoxetine metabolism with fish liver microsomes. Chemosphere 79:2632
35. Gomez C, Constantine L, Moen M, Vaz A, Huggett DB (2010) The influence of gill and liver
metabolism on the predicted bioconcentration in fish. Chemosphere 81:11891195
36. Gomez CF, Constantine L, Moen M, Vaz A, Wang W, Huggett DB (2011) Ibuprofen metabolism in the
liver and fill of rainbow trout, Oncorhynchus mykiss. Bull Environ Contam Toxicol 86:247251
37. Schultz MM, Painter MM, Bartell SE, Logue A, Furlong ET, Werner SL, Shoenfuss HL (2011)
Selective uptake and biological consequences of environmentally relevant antidepressant pharmaceutical exposures on male fathead minnows. Aquat Toxicol 104:3847

14

B.W. Brooks et al.

38. Nakamura Y, Yamamoto H, Sekizawa J, Kondo T, Hirai N, Tatarako N (2008) The effects of
pH on fluoxetine in Japanese medaka (Oryzias latipes): acute toxicity in fish larvae and bioaccumulation in juvenile fish. Chemosphere 70:865873
39. Zhou SN, Oakes KD, Servos MR, Pawliszyn J (2008) Application of solid-phase microextraction for in vivo laboratory and field sampling of pharmaceuticals in fish. Environ Sci Technol
42:60736079
40. Environmental Defense Fund (1997) Toxic ignorance: the continuing absence of basic health
testing for top-selling chemicals in the United States. Environmental Defense Fund, New York
41. Berninger JP, Brooks BW (2010) Leveraging mammalian pharmaceutical toxicology and
pharmacology data to predict chronic fish responses to pharmaceuticals. Toxicol Lett
193:6978
42. Brausch JM, Connors KA, Brooks BW, Rand GM (2012) Human pharmaceuticals in the
aquatic environment: a critical review of recent toxicological studies and considerations for
toxicity testing. Rev Environ Contam Toxicol 218:199
43. Brooks BW, Riley TM, Taylor RD (2006) Water quality of effluent-dominated stream ecosystems: ecotoxicological, hydrological, and management considerations. Hydrobiologia
556:365379
44. Seiler JP (2002) Pharmacodynamic activity of drugs and ecotoxicology: can the two be connected? Toxicol Lett 131:105115
45. Huggett DB, Brooks BW, Peterson B, Foran CM, Schlenk D (2002) Toxicity of select betaadrenergic receptor blocking pharmaceuticals (b-blockers) on aquatic organisms. Arch Environ
Contamin Toxicol 42:229235
46. Brooks BW, Foran CM, Richards S, Weston JJ, Turner PK, Stanley JK, Solomon K, Slattery M,
La Point TW (2003) Aquatic ecotoxicology of fluoxetine. Toxicol Lett 142:169183
47. Huggett DB, Cook JC, Ericson JF, Williams RT (2003) A theoretical model for utilizing mammalian
pharmacology and safety data to prioritize potential impacts of human pharmaceuticals to fish.
Hum Ecol Risk Assess 9:17891799
48. Fitzsimmons PN, Fernandez JD, Hoffman AD, Butterworth BC, Nichols JW (2001) Branchial
elimination of superhydrophobic organic compounds by rainbow trout (Oncorhynchus mykiss).
Aquat Toxicol 55:2334
49. Brown JN, Paxeus N, Forlin L, Larsson DGJ (2007) Variations in bioconcentration of human
pharmaceuticals from sewage effluents into fish blood plasma. Environ Toxicol Pharmacol
24:267274
50. Fick J, Lindberg RH, Parkkonen J, Arvidsson B, Tysklind M, Larsson DGJ (2010) Therapeutic
levels of levonorgestrel detected in blood plasma of fish: results from screening rainbow trout
exposed to treated sewage effluents. Environ Sci Technol 44:26612666
51. Fick J, Lindberg RH, Tysklind M, Larsson DGJ (2010) Predicted critical environmental concentrations for 500 pharmaceuticals. Regul Toxicol Pharmacol 58:516523
52. Berninger JP, Du B, Connors KA, Eytcheson SA, Kolkmeier MA, Prosser KN, Valenti TW,
Chambliss CK, Brooks BW (2011) Effects of the antihistamine diphenhydramine to select
aquatic organisms. Environ Toxicol Chem 30:20652072
53. Valenti TV, Gould GG, Berninger JP, Connors KA, Keele NB, Prosser KN, Brooks BW (2012)
Human therapeutic plasma levels of the selective serotonin reuptake inhibitor (SSRI) sertraline
decrease serotonin reuptake transporter binding and shelter seeking behavior in adult male
fathead minnows. Environ Sci Technol 46:24272435
54. Valenti TW, Perez Hurtado P, Chambliss CK, Brooks BW (2009) Aquatic toxicity of sertraline
to Pimephales promelas at environmentally relevant surface water pH. Environ Toxicol Chem
28:26852694
55. Ankley GT, Bennett RS, Erickson RJ, Hoff DJ, Hornung MW, Johnson RD, Mount DR,
Nichols JW, Russom CL, Schmieder PK, Serrano JA, Tietge JE, Villeneuve DL (2010) Adverse
outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment. Environ Toxicol Chem 29:730741

Perspectives on Human Pharmaceuticals in the Environment

15

56. Mokry L, Brooks BW, Chambliss CK, Knight R, Keller C, Sedlak DL (2011) Evaluate wetland
systems for treated wastewater performance to meet competing effluent quality goals.
WateReuse Research Foundation, Alexandria, VA. 153 p
57. Garrison AW (2006) Probing the enantioselectivity of chiral pesticides. Environ Sci Technol
40:1623
58. Stanley JK, Brooks BW (2009) Perspectives on ecological risk assessment of chiral compounds. Integr Environ Assess Manag 5:364373
59. Valenti TW, Taylor JT, Back JA, King RS, Brooks BW (2011) Influence of drought and total
phosphorus on diel pH in wadeable streams: implications for ecological risk assessment of
ionizable contaminants. Integr Environ Assess Manag 7:636647
60. Brooks BW, Valenti TW, Cook-Lindsay BA, Forbes MG, Scott JT, Stanley JK, Doyle RD
(2011) Influence of Climate change on reservoir water quality assessment and management:
effects of reduced inflows on diel pH and site-specific contaminant hazards. In: Linkov I,
Bridges TS (eds) Climate: global change and local adaptation. NATO science for peace and
security series C: environmental security. Springer, New York, pp 491522
61. Kmmerer K (2007) Sustainable from the very beginning: rational design of molecules by life
cycle engineering as an important approach for green pharmacy and green chemistry. Green
Chem 9:899907
62. Voutchkova AM, Kostal J, Steinfeld JB, Emerson JW, Brooks BW, Anastas P, Zimmerman JB
(2011) Towards rational molecular design: derivation of property guidelines for reduced acute
aquatic toxicity. Green Chem 13:23732379
63. Voutchkova AM, Kostal J, Connors KA, Brooks BW, Anastas P, Zimmerman JB (2012) Towards
rational molecular design for reduced chronic aquatic toxicity. Green Chem 14:10011008
64. Fatta D, Nikolaou A, Achilleos A, Meric S (2007) Analytical methods for tracing pharmaceutical residues in water and wastewater. Trends Anal Chem 26:515533
65. Gros M, Petrovic M, Barcelo D (2006) Multi-residue analytical methods using LC-tandem MS
for the determination of pharmaceuticals in environmental and wastewater samples: a review.
Anal Bioanal Chem 386:941952
66. Hao C, Clement R, Yang P (2007) Liquid chromatographytandem mass spectrometry of bioactive pharmaceutical compounds in the aquatic environment a decades activities. Anal
Bioanal Chem 387:12471257
67. Hernando MD, Gomez MJ, Aguera A, Fernandez-Alba AR (2007) LC-MS analysis of basic
pharmaceuticals (beta-blockers and anti-ulcer agents) in wastewater and surface water. Trends
Anal Chem 26:581594
68. Hernandez J, Sancho JV, Ibaez M, Guerrero C (2007) Antibiotic residue determination in
environmental waters by LC-MS. Trends Anal Chem 26:466485
69. Wong CS, MacLeod SL (2009) JEM spotlight: recent advances in analysis of pharmaceuticals
in the aquatic environment. J Environ Monit 11:923936
70. Kim SC, Carlson K (2005) LCMS2 for quantifying trace amounts of pharmaceutical compounds in soil and sediment matrices. Trends Anal Chem 24:635644
71. OConnors S, Aga DS (2007) Analysis of tetracycline antibiotics in soil: advances in extraction, clean-up, and quantification. Trends Anal Chem 26:456465
72. Buchberger WW (2007) Novel analytical procedures for screening of drug residues in water,
waste water, sediment and sludge. Anal Chim Acta 593:129139
73. Petrovic M, Hernando MD, Diaz-Cruz MS, Barcelo D (2005) Liquid chromatographytandem
mass spectrometry for the analysis of pharmaceutical residues in environmental samples: a
review. J Chromatogr A 1067:114
74. Giger W (2009) Hydrophilic and amphiphilic water pollutants using advanced analytical
methods for classic and emerging contaminants. Anal Bioanal Chem 393:3744
75. Richardson S (2009) Water analysis: emerging contaminants and current issues. Anal Chem
81:46454677
76. Barcelo D, Petrovic M (2007) Challenges and achievements of LC-MS in environmental
analysis: 25 years on. Trends Anal Chem 26:211

16

B.W. Brooks et al.

77. Hao C, Zhao X, Yang P (2007) GC-MS and HPLC-MS analysis of bioactive pharmaceuticals
and personal-care products in environmental matrices. Trends Anal Chem 26:569580
78. Morley MC, Snow DD, Cecrle C, Denning P, Miller L (2006) Emerging chemicals and analytical methods. Water Environ Res 78:10171053
79. Kot-Wasik A, Debska J, Namiesnik J (2007) Analytical techniques in studies of the environmental fate of pharmaceuticals and personal-care products. Trends Anal Chem 26:557568
80. Richardson S (2008) Environmental mass spectrometry: emerging contaminants and current
issues. Anal Chem 80:43734402
81. Rubio S, Perez-Bendito D (2009) Recent advances in environmental analysis. Anal Chem 81:
46014622
82. Perez S, Barcelo D (2007) Application of advanced MS techniques to analysis of human and
microbial metabolites of pharmaceuticals in the aquatic environment. Trends Anal Chem 26:
494514
83. Petrovic M, Barcelo D (2007) LC-MS for identifying photodegradation products of pharmaceuticals in the environment. Trends Anal Chem 26:486493
84. Radjenovic J, Petrovic M, Barcelo D (2007) Advanced mass spectrometric methods applied to
the study of fate and removal of pharmaceuticals in wastewater treatment. Trends Anal Chem
26:11321144
85. Kosjek T, Heath E, Petrovic M, Barcelo D (2007) Mass spectrometry for identifying pharmaceutical biotransformation products in the environment. Trends Anal Chem 26:10761085
86. Kostopoulou M, Nikolaou A (2008) Analytical problems and the need for sample preparation
in the determination of pharmaceuticals and their metabolites in aqueous environmental matrices. Trends Anal Chem 27:10231035
87. Escandar GM, Faber NM, Goicochea HC, Muoz de la Pea A, Olivieri AC, Poppi RJ (2007)
Second- and third-order multivariate calibration data, algorithms and applications. Trends
Anal Chem 26:752765
88. Galera MM, Gil Garcia MD, Goicochea HC (2007) The Application to wastewaters of
chemometric approaches to handling problems of highly complex matrices. Trends Anal Chem
26:10321042
89. Lambropoulou DA, Konstantinou IK, Albanis TA (2007) Recent developments in headspace
microextraction techniques for the analysis of environmental contaminants in different matrices. J Chromatogr A 1152:7096
90. Pavloivc DM, Babic S, Horvat AJM, Kastelan-Macan M (2007) Sample preparation in analysis of pharmaceuticals. Trends Anal Chem 26:10621075
91. Pichon V, Chapuis-Hugon F (2008) Role of molecularly imprinted polymers for selective
determination of environmental pollutantsa review. Anal Chim Acta 622:4861
92. Rodriguez-Mozaz S, Lopez de Alda MJ, Barcelo D (2007) Advantages and limitations of online solid phase extraction coupled to liquid chromatographymass spectrometry technologies
versus biosensors for monitoring of emerging contaminants in water. J Chromatogr A 1152:
97115
93. Soderstrom H, Lindberg RH, Fick J (2009) Strategies for monitoring the emerging polar
organic contaminants in water with emphasis on integrative passive sampling. J Chromatogr
A 1216:623630
94. Wardencki W, Curylo J, Namiesnik J (2007) Trends in solventless sample preparation techniques for environmental analysis. J Biochem Biophys Methods 70:275288
95. Perez S, Barcelo D (2008) Applications of LC-MS to quantitation and evaluation of the environmental fate of chiral drugs and their metabolites. Trends Anal Chem 27:836846
96. Ramirez AJ, Mottaleb MA, Brooks BW, Chambliss CK (2007) Analysis of pharmaceuticals in
fish tissue using liquid chromatographytandem mass spectrometry. Anal Chem 79:
31553163

Environmental Risk Assessment for Human

Pharmaceuticals: The Current State
of International Regulations
Jrg Oliver Straub and Thomas H. Hutchinson

Introduction
An overview is given on environmental risk assessment for pharmaceuticals (ERA),
with a description of the current regulatory requirements for human pharmaceuticals ERA in Europe and the USA as well as developments worldwide. In addition,
further developments on national levels concerning the environmental safety of
pharmaceuticals are presented. Also, a short comparison with international veterinary pharmaceuticals guidelines and with biocides ERA is given.
As long as human population density is low and excreta are spread diffusely over
a large area, no significant levels of PAS or metabolites are expected in the environment. But when population density increases, when excreta collect in sewage and
the latter is discharged, after wastewater treatment or not, to receiving waters, measurable to significant concentrations in surface waters may be reached. With strong
population growth in industrialised societies from the nineteenth century onward,
with sewage collection systems in the growing cities and with the increase in the
number of pharmaceutical companies and their biologically active products, a rise
in environmental concentrations of at least certain PAS followed during the past
century. A parallel development in analytical methods and power, expressed as
constantly decreasing limits of detection and quantitation, inevitably led to determinations of PAS in environmental matrices.

J.O. Straub (*)

F.Hoffmann-La Roche Ltd, Group SHE,
LSM 49/2.033, Basle CH-4070, Switzerland
e-mail: [email protected]
T.H. Hutchinson
CEFAS Weymouth Laboratory, Centre for Environment, Fisheries and Aquaculture Sciences,
The Nothe, Barrack Road, Weymouth, Dorset DT4 8UB, UK
e-mail: [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_2, Springer Science+Business Media, LLC 2012

17

18

J.O. Straub and T.H. Hutchinson

The first analytical detections of PAS and metabolites in environmental media

are reported from the USA in the 1970s [33, 37], where among others salicylic
acid, the main metabolite of acetylsalicylic acid was detected in sewage works
effluent. These initial detections initiated a rapidly growing list of similar publications and reviews covering sewage treatment effluent, surface, estuarine, marine,
ground and tap water over the following decades (e.g. Richardson and Bowron
[64], Aherne and Briggs [1], Ayscough et al. [4] and Thomas and Hilton [77] in the
UK; Heberer et al. [35] and Ternes et al. [73] in Germany; Halling-Srensen et al.
[34] in Denmark; Buser et al. [10] and Tixier et al. [77] in Switzerland; Belfroid
et al. [6] in the Netherlands; Stumpf et al. [72] in Brazil; Zuccato et al. [84] and
Calamari et al. [11] in Italy; Farr et al. [28] and Fernndez et al. [29] in Spain;
Kolpin et al. [48] and Barnes et al. [5] in the USA; Metcalfe et al. [54] in Canada;
Vieno et al. [81] in Finland; Nakada et al. [57] in Japan; Rabiet et al. [63] in France;
Kim et al. [47] in South Korea). Note this is not meant to be a complete list but
rather an illustration of the worldwide increase in publications in the 1990s and
2000s. Again, the scope of detections widened with massively refined analytical
instruments and methods.
In parallel to these ubiquitous detections in environmental media, the question of
possible adverse effects caused by PAS to environmental organisms and ecosystems
also gained importance. Initial environmental risk assessments (ERAs), comparing
environmental concentrations with known effects, began in the 1980s. The concerns
about environmental safety of PAS, alone and in particular in combinations, strongly
increased with accruing evidence for widespread endocrine disruption in wild fish
[44], in particular downstream of sewage treatment works effluents and also with
experimental adverse effects seen with a few PAS at very low concentrations (e.g.
[19, 30, 43]), which in some cases were close to or within the range of measured
environmental concentrations (MECs). In parallel, the use of PAS or similar substances has played an important role in other areas of aquatic research, including
aquaculture [31, 40] and marine antifoulant paints [38, 50, 61].
In view of mounting evidence for widespread environmental exposure and potential or probable environmental effects of PAS, enquiries and investigations into
environmental hazards and risks due to PAS began in the 1980s (e.g. [1, 18, 34, 36,
46, 65, 78]). In parallel to these often government-sponsored investigations, the
necessity for and development of formal ERAs specifically for PAS (pharmaceuticals ERA or PERA) was recognised by regulators on both sides of the Atlantic,
which led to legal requirements and, with some delay, to guidelines for such PERAs
as part of the registration dossier from the 1990s onwards. Formal guidelines were
developed and published in 1998 in the USA and in 2006 in the European Union
(EU). In other countries, PERAs are requested (e.g. Australia) or formal own guidelines
are in the making (Canada, Japan). In addition, Sweden led the way with a system
for the ERA of old PAS already on the market. But even beyond the formal
requirements for PERAs in the context of registration, PAS in the environment (PIE)
may be the subject of other legislation than registration, which, however, may still
require some kind of ERA. These developments and current states will be outlined in
the following paragraphs.

Current State of Regulations for Human Pharmaceuticals ERA

19

Current State of PERA Regulation in Various

Regions or Countries
PERA started in the USA and EU in the 1980s or early 1990s. Much of the methodology seems to derive from pesticides ERA, which came into focus and developed
appropriate methodologies earlier than pharmaceuticals in general. All of the ERA
procedures have in common a comparison between predicted (or measured) environmental concentrations (PECs or MECs) with predicted no effect concentrations
(PNECs), both per environmental compartment under consideration. Such compartments may be wastewater treatment, surface waters, sediments, groundwaters, tidal
and coastal/marine waters, soils (through landspreading of surplus sewage sludge,
called biosolids in North American terminology) and, rarely, the atmosphere. PECs
are derived from either predicted use or maximum daily use multiplied by a default
use or penetration factor in the population, integrating human metabolism and depletion during sewage treatment or in the environment, sorption and distribution to other
environmental compartments, dilution and advection (off-transport by the medium) in
the receiving compartments. PNECs are mostly derived from either acute or chronic
ecotoxicity tests, normally with standard organism groups representative for the compartment, by dividing by assessment factors (AFs) which are dependent on the character and number of ecotoxicity results available. In higher tiers of the ERA, the above
deterministic procedure using AFs can be replaced by probabilistic methodology,
where the distributional characteristics of a number of ecotoxicity test results
(normally at least ten chronic datapoints) are used to derive a PNEC. PECs and PNECs
are compared per compartment, in general through forming the PEC/PNEC ratio.
If this ratio is <1, i.e. if the expected concentration is below the one predicted to cause
no adverse effect, and there are no other concerns for all the compartments under
consideration, there is no indication for significant risk and the ERA may be finalised.
In case the PEC/PNEC ratio is 1, risk cannot be excluded and therefore the ERA
must be refined by reappraisal of the PEC and/or PNEC through better, more in-depth
methodology. An ERA may thereby progress from a relatively simple and crude
assessment based on little data to a much more realistic assessment that, in turn, needs
and incorporates far more experimental data and often also advanced models. However,
even with a highly refined assessment there is never any guarantee that the outcome
will be no significant risk. A refined ERA can only characterise a possible risk
better than a crude ERA, but it cannot make risks or concerns disappearon the
other hand, it certainly will identify compartments at potential risk, allowing the
development of targeted risk management strategies if indicated.

PERA in the USA

Based on the 1969 US National Environmental Policy Act (NEPA) as amended, the
Code of Federal Regulations (CFR) Title 21 Part 25 as amended details environmental

20

J.O. Straub and T.H. Hutchinson

assessments (EAs in US legal terminology) within the US Food and Drugs legislation
(21 CFR 25; current version available at http://www.accessdata.fda.gov/scripts/cdrh/
cfdocs/cfcfr/CFRSearch.cfm?CFRPart=25). By this, all applications or petitions
requesting Agency action must be accompanied by either an EA or a claim of categorical exclusion; failure to submit one or the other is sufficient grounds for refusing to
file or approve the application (cited from Environmental Impact Review at CDER,
http://www.fda.gov/AboutFDA/CentersOffices/CDER/ucm088969.html). In 1998 the
US Center for Drug Evaluation and Research (CDER) and the Center for Biologics
Evaluation and Research (CBER) within the US Food and Drug Administration published a Guidance for Industry, Environmental Assessment of Human Drug and
Biologics Applications, revision 1 [14], which is still current today.
The Guidance describes in which cases an EA can be waived and how to proceed
with an EA in the remainder. Waivers, the so-called categorical exclusions, may be
invoked in the following cases:
If the application does not increase the use of active moiety (i.e. in case of extensions or additional applications by third parties for PAS already on the market).
If the application may lead to increased use but the estimated concentration of
the AS at the point of entry into the environment is less than 1 part per billion
(ppb). This means that the entry into the environment concentration (EIC) of a
particular PAS from US publicly owned treatment works (POTWs) must be
below 1 mg/L, discounting all metabolism; calculating back from an EIC of
1 mg/L and the average annual total effluent of all POTWs results in a maximum
annual amount of approximately 44 metric tonnes of PAS per year for the whole
continental USA, based on daily POTW inflow data given in the Guidance ([14];
p 4). Hence, if the predicted annual use of a new PAS is below 44 tonnes/annum
there is no need for an EA, except if the applicant has information to suggest that
the use of even a lesser quantity may significantly affect the quality of the
human environment ([14]; p 3).
For biological PAS if their use will not lead to significant concentrations in the
environment.
For investigational new drugs still under development in clinical research.
For specific biological products for blood or plasma transfusion.
In all other cases, the applicant needs to prepare an EA following a tiered, stepwise approach that follows the course of a PAS from human excretion into the
environment. Hence, in a first basic step, if there is experimental evidence that a
new PAS is rapidly depleted, e.g. through biodegradation in a POTW, and not inhibitory to microorganisms, the EA can be stopped and finalised with a Finding of No
Significant Impact (FONSI). If the PAS is not rapidly depleted and if it is lipophilic
(with an n-octanol/water distribution coefficient logDOW 3.5 at a relevant environmental pH of approximately 7), suggesting bioaccumulation, the applicant should
initiate chronic testing in tier 3; note the tier numbering is given according to the
Guidance [14]. Further details as to depletion (degradation, hydrolysis or partitioning to other environmental compartments) and to interpretation of these fate
processes are given.

Current State of Regulations for Human Pharmaceuticals ERA

21

In all other cases, the effects testing starts with one acute test in tier 1. If the ratio
of the 50% effect or 50% lethal concentration (EC50 or LC50) in this test divided
by the EIC or predicted (or expected in US terminology) environmental concentration (PEC or EEC), whichever is higher, is 1,000 and there were no adverse effects
observed at the higher of EIC or EEC (termed maximum expected environmental
concentration or MEEC), the EA can be stopped and finalised. This ratio corresponds to a margin of safety (MOS) in general ERA terminology. If there were
effects at MEEC, the applicant should initiate chronic testing in tier 3.
If the tier 1 MOS is <1,000, acute base set testing in tier 2 is specified. For
aquatic EA the base set consists of acute algal, aquatic invertebrate and fish tests, for
terrestrial testing of plant growth, earthworm and soil microbial toxicity. The lowest
EC50 or LC50 from the effects base set is again divided by the MEEC. If the
obtained tier 2 MOS is 100 and there were no adverse effects observed at MEEC,
the EA can be stopped and finalised. If there were such effects, the applicant should
initiate chronic testing in tier 3.
In tier 3, an unspecified number and selection of aquatic or terrestrial species
should be tested chronically; applicants are advised to contact CDER/CBER for test
selection. If the obtained tier 3 MOS between the (lowest) chronic EC50 or LC50
and the MEEC is 10 and there were no adverse effects observed at MEEC, the EA
can be stopped and finalised. If the MOS is <10 or if there were effects at MEEC,
the applicant should contact CDER/CBER for further advice and strategy.
Overall, the US Guidance is characterised by a comparatively high threshold of
1 mg/L as the EIC (POTW effluent), respectively, as 0.1 mg/L as an average EEC,
using the standard dilution factor of 10 ([14], p 19), which in turn translates to the
above 44 tonnes/annum below which an EA can normally be waived. If this threshold or trigger is surpassed, the actual EA proceeds logically from excretion to sewage treatment and into further compartments along traditional methodology,
comparing PECs and effect concentrations. Lower-tier PNECs are based on only
one (tier 1) or a base set of three (tier 2) acute ecotoxicity tests. In case of only one
acute ecotoxicity test in tier 1, the tier 1 MOS must be 1,000 for the EIC (POTW
effluent), which corresponds to an implicit AF of 10,000 for surface waters, including the ten times default dilution from POTW effluent. For tier 2, with an acute
ecotoxicity base set comprising three different groups of organisms, the surface
water AF drops to 1,000, while for tier 3 with an unstated number of chronic ecotoxicity data the implicit AF is 100 for surface waters, based on chronic EC50s or
LC50s, which is unusual and in contrast with other guidelines that use the chronic
NOECs for PNEC derivation.
While the very first detections of single PAS in environmental media in the USA
date from the 1970s [33, 37], it took a long time before a report of widespread
detections in sewage works effluents, surface and groundwaters in the USA [48]
brought the topic of pharmaceuticals in the environment (PIE) to scientific and regulatory, later also to public attention. A series of syndicated articles from Associated
Press journalists in the late 2000s with a focus on PIE, specifically PAS in drinking
water [3], attracted and widened public and political attention to the topic of PIE
and tap water. Comparable reports continue being published from various States

22

J.O. Straub and T.H. Hutchinson

(e.g. [55], for Delaware drinking waters). Within half a year starting from the first
AP report, according to the AP [3] site, a US Congressional Panel discussed monitoring and potential impacts of micropollutants including PAS in environmental
waters, which are currently not regulated by the US Environmental Protection
Agency (EPA) either as a group or as single substances in the USA. The discussion
seemed to focus mainly on potential human risks from PIE through water abstraction, treatment and consumption as drinking water, but less on risks for environmental organisms or ecosystems. Also, questions on PIE and the safety of PAS in
drinking waters were raised in the US Senate Committee on Environment and Public
Works (http://epw.senate.gov/public/index.cfm, search for pharmaceuticals and
water). Some investigations on potential human health risks from PIE via drinking water were published in the previous decade (e.g. [9, 12, 16, 17, 45, 67, 83, 84]),
all of which have found no significant risks based on the available evidence.
In addition, on July 7, 2010, the Great Lakes Environmental Law Center and the
Natural Resources Defense Council as petitioners submitted a Citizen Petition to
the US Food and Drugs Administration Commissioner. A Citizen Petition in the US
is a legal means to challenge existing regulations. In this Citizen Petition concerning
an amendment to the current US PERA Guidance [14], the repealing of the categorical exclusion threshold of 1 ppt (1 mg/L, corresponding to approximately 44 metric
tonnes of PAS per annum) EIC is requested, because the current regulation does
not reflect a safe standard supported by current scientific information. In case the
threshold for a categorical exclusion is indeed repealed, this would mean that nearly
all new human PAS would need an EA for registration.
It will remain to be seen whether the parliamentary discussions and legal motions
in the USA will eventually have effects on US regulations, on PERA in general, on
the US PERA Guideline, possibly also for old PAS already on the market, or for
the regulation of water contaminants by the EPA.

PERA in the European Union

First requirements for PERA were laid down in EU Directive 93/39/EEC, which
asked to give indications of any potential risks presented by the medicinal product
to the environment. The development of the PERA guideline in the EU took 13
years in all, with several draft guidelines published during that time [68, 69]. In
2006, the European Medicines Agency (EMA, London, UK; note that the former
abbreviation EMEA for European Medicines Evaluation Agency is not being used
any longer) published the first definitive Guideline for Environmental Risk
Assessment of Human Medicines [26]. This guideline describes a tiered procedure,
from categorical exclusion or direct referral, to a simple, worst-case exposure estimation of a pharmaceutical active substance to the investigation of fate and effects
in sewage works and surface waters, up to a refined assessment for these or other
environmental compartments.

Current State of Regulations for Human Pharmaceuticals ERA

23

A PERA is required for new registrations (Medicines Authorisation

Application or MAA in EU terminology) and for all repeat registrations by the
same applicant, termed variations in the EU, that may lead to significantly
increased environmental exposure to the PAS; note that significant is not
defined or quantified in this context. In the basic Phase 1 of the PERA, certain
categories of PAS are excluded from PERA (amino acids, proteins, peptides,
carbohydrates, lipids, electrolytes, vaccines and herbal medicines), while other
PAS are directly referred to special ERA. Highly lipophilic PAS with a log KOW
> 4.5 are directly referred to a persistence, bioaccumulation and (high eco)
toxicity (PBT) assessment, where these properties are to be tested and evaluated
in that order, following the methodology of the EU Technical Guidance
Document (TGD, [75]), now replaced by the REACH Technical Guidance
Document [24]. As a second direct referral category, potential endocrine disrupters, viz. those PAS that may affect the reproduction of vertebrate or lower
animals at concentrations lower than 0.01 mg/L, should be assessed using a
tailored strategy that addresses the specific mode of action. Note that there is
no technical guidance for assessing potential endocrine disrupters at present, the
applicant should justify all actions taken and, to be on the safe side, would be
well advised to contact the EMA Committee for Human Medicinal Products for
scientific advice.
All remaining PAS in Phase 1 undergo a prescreening that involves a rigid
worst-case PEC prediction which is compared with a threshold value or action
limit in EMA terminology. The maximum daily dose of the PAS is multiplied
with a default penetration factor (Fpen) of 0.01 or 1%, which was derived by
probabilistic methods to model a reasonable-worst-case use of a medicine in the
population [26], and divided by a default 200 L of wastewater per person per day
and a default surface water dilution factor of 10, to give the Phase I surface water
PEC. If this surface water PEC is <0.01 mg/L (i.e. <10 ng/L) and there are no
other grounds for direct referral, the PERA can be finalised. Backcalculating
with all the default values, a PAS would need to have a maximum daily dose of
<2 mg for the surface water PEC to remain below 10 ng/L. If the PEC is 10 ng/L,
the PERA has to go into Phase 2 Tier A for an initial ERA based on experimental
data. Note that no metabolism, human or environmental, may be factored in the
Phase 1 PEC. Further, the Fpen may only be changed in Phase 1 based on published epidemiology data for the medical indication(s) addressed by the PAS in
question, but not by marketing predictions or other indicators. Both in case of a
categorical exclusion and if the PEC is <0.01 mg/L, a justification letter for not
producing an ERA Expert Report should be prepared and included with the registration dossier.
In Phase 2 Tier A a prescribed set of experimental environmental fate and effects
data must be elaborated under GLP quality assurance. The results are then used to
derive PEC/PNEC ratios for various compartments and for comparison with given
threshold values, which will inform on the necessity or not of further evaluation of
potential risks in certain environmental compartments in Phase 2 Tier B. Modelled

24

J.O. Straub and T.H. Hutchinson

data, e.g. by quantitative structure activity/property relationship algorithms (QSAR

or QSPR) are not acceptable. The Phase 2 Tier A experimental data set consists of:
n-Octanol/water partition coefficient (log KOW, determined using OECD107,
OECD117, OECD 123 or draft OECD122 technical guidelines)
Adsorption constants to the organic carbon fraction in soils or activated sludges
(KOC and Kd; OECD106, OECD121 or OPPTS 835.1110)
Ready biodegradability (OECD301) as a facultative test; if not readily biodegradable, a transformation test in aquatic sediment systems (OECD308) is mandatory
An algal growth inhibition test (OECD201) with green algae, in case of antimicrobials with cyanobacteria
A daphnid reproduction test (OECD211) with Daphnia sp. (meaning not with
Cerodaphnia dubia, which has a shorter generation time)
A fish early life stage toxicity test (OECD210)
An activated sludge respiration inhibition test (OECD209)
These data are used for the following decision tree:
If the substance is lipophilic with a KOW > 1,000 (log KOW >3), it is assumed that
the PAS may bioaccumulate, which is why such a PAS is directed to an experimental bioaccumulation study in Phase 2 Tier B. Note that this provision is
redundant to the Phase 1 direct referral for lipophilic PAS with a logKOW > 4.5.
The latter stems from concerns from the EU OSPAR panel (the Oslo-Paris
Commission on PBT substances in the North Sea, http://www.ospar.org/), which
uses a logKOW threshold value of 4.5 for screening potential B substances.
However, the more lipophilic a PAS is, the lower on the whole is its bioavailability (Roche own unpublished data); this entails an increase of daily dosage in
order to attain pharmacologically active levels of the PAS. Thereby the action
limit of 2 mg PAS per day will be breached and the substance will go into Phase
2 Tier A, with a sediment/water study (persistence), testing for bioaccumulation
and the three base set chronic tests (toxicity). Hence, the whole PBT package is
performed for all PAS with a logKOW >3, anyway.
If the substance adsorbs strongly with a KOC >10,000 L/kg (logKOC >4), it is
assumed that it would be removed during wastewater treatment by adsorption
to activated sludge, unless it proves to be readily biodegradable. As surplus
sludge is often spread on arable land after treatment (dewatering, anaerobic
digestion, etc.), strongly sorbing PAS are assumed to reach the terrestrial
compartment, which is why such a PAS is directed to a terrestrial ERA in
Phase 2 Tier B.
If the substance is not readily biodegradable and the sediment/water environmental fate study shows >10% in the sediment at any time after 13 days, it is
assumed that the PAS will partition to the sediment to a relevant degree, which
is why such a PAS is directed to sediment toxicity testing and a sediment ERA in
Phase 2 Tier B.
The PNEC for surface waters is derived from the lowest chronic algal, daphnid
or fish NOEC, dividing by an AF of 10. If the PEC/PNEC ratio for surface water

Current State of Regulations for Human Pharmaceuticals ERA

25

is below 1, the PAS is unlikely to represent a risk to the aquatic compartment. If

the ratio is 1, a refinement preferably of the PEC should be made in Phase 2 Tier
B. Note that there is no specific mention of further refining the PNEC, e.g.
through probabilistic methods.
The microorganism PNEC for wastewater treatment is derived from the NOEC
of the activated sludge respiration inhibition test, dividing by an AF of 10. If the
ratio of surface water PEC (which is extrapolated from the wastewater PEC with
a dilution factor of 10) divided by the microorganism PNEC is <0.1 (i.e. if the
implicit ratio of wastewater PEC divided by the microorganism PNEC is <1), the
PAS is unlikely to represent a risk for wastewater treatment. If the ratio is 0.1,
a refinement of the fate of the PAS in wastewater treatment or the effect on
microorganisms should be made in Phase 2 Tier B.
An initial groundwater assessment should be made, except for those PAS that are
readily biodegradable or that have a 90% dissipation time (DT90) in the sediment/water study of <3 days or that have an average KOC >10,000 L/kg, all of
which would be largely removed during sewage treatment. The PNEC for
groundwater is derived from the chronic daphnid NOEC (as the only potential
higher organisms in groundwater are invertebrates, in contrast to green algae or
fish) by dividing by an AF of 10. The groundwater PEC is approximated as surface
water PEC 0.25. If the groundwater PEC/PNEC ratio is <1, the PAS is unlikely
to represent a risk to the groundwater compartment. If the ratio is 1, a refinement
preferably of the (surface water) PEC should be made in Phase 2 Tier B.
In Phase 2 Tier B of the EU PERA, referrals from Phase 1 or potential risks
identified in Phase 2 Tier A should be investigated. Whereas in earlier phases Type
II Variation For the Treatment of Granulomatosis With Polyangiitis (Wegeners)
and Microscopic Polyangiitis only the parent compound was investigated based on
a total residue approach (meaning that no demonstrable metabolism or degradation
could be factored in), evidenced human metabolism may be used to refine PECs in
Phase 2 Tier B, but then relevant (>10% of parent PAS) metabolites are to be
assessed by PEC and PNEC as well:
The surface water PEC may be further refined using sewage works modelling with
the spreadsheet application SimpleTreat that is integrated in the EU substance
assessment model EUSES (downloadable from http://ecb.jrc.ec.europa.eu/euses/).
The sludge adsorption (KOC) value from the Phase 2 Tier A adsorption test and
ready biodegradability (if attained) must be entered into SimpleTreat, respectively,
EUSES. In addition, a so-called local PEC can be calculated for refinement. Again,
PEC/PNEC ratios as above are to be derived and the risk for the given compartments (wastewater treatment, surface waters, groundwater) characterised.
For sediment ERA, a sediment PEC is to be calculated based on the TDG (2003)
[75], respectively the REACH TGD [24] algorithms, based on surface water PEC,
adsorption and default EU sediment parameters. The sediment PNEC is based on
at least one chronic test with sediment-dwelling organisms (the crustacean
Hyalella, the oligochaete worm Lumbriculus and the larvae of the insect
Chironomus are specifically mentioned). In case of one chronic NOEC available,

26

J.O. Straub and T.H. Hutchinson

the AF is 100; for two chronic NOECs the AF is 50 and for three the AF is 10, to
derive the sediment PNEC.
For refined wastewater treatment microorganism risk assessment, the PAS concentration in the aeration tank of a standard sewage works should be calculated
using SimpleTreat. This should be compared with a PNEC refined based on
microorganisms testing and AFs as set out in the TDG (2003) [75], respectively,
REACH TGD [24]. If refinement does not result in a PEC/PNEC ratio <1, further
PNEC refinement should be undertaken.
Terrestrial assessment should be performed with a soil PEC calculated with a
combination of SimpleTreat modelling to generate a sludge PEC and the derivation of the soil PEC from sludge spreading and the results, in particular the soil
half-life, from an obligatory soil transformation test (OECD307), using the algorithm in the TGD [75], respectively, the REACH TGD [24]. The soil PNEC is
derived from the lowest (no) effect value from the following obligatory terrestrial
ecotoxicity tests soil microorganisms (nitrogen transformation test, OECD 216),
terrestrial plants growth test (OECD 208), earthworm acute toxicity test (OECD
207), Collembola soil insects reproduction test (ISO 11267), again based on the
above TGDs. Soil risk is then characterised with the PEC/PNEC ratio.
The above Phase 2 Tier B assessment concludes the EU PERA. The whole
assessment is to be compiled in an Expert Report with all conclusions, with all references and test reports, and with the curriculum vitae and signature of the expert
who produced the report. In case there remains residual risk in one or more compartments, this may not keep the medicine from the market, as patient benefit is
given priority before environmental concerns [22]. However, to minimise environmental exposure from unused medicines, the following phrasing should be inserted
in package/patient information leaflets: Medicines should not be disposed of via
wastewater or household waste. Ask your pharmacist how to dispose of medicines
no longer required. These measures will help to protect the environment. Note that
it is recommended that this phrase be included even for medicines that do not require
special disposal measures. Also, in case of residual risk, the Expert Report should
contain evaluation of precautionary and safety measures to be taken with a view to
minimising environmental exposure both from disposal of unused medicines and
from patient use; this information should also become part of the Specific Product
Characteristics information.
The 2006 EMA PERA Guideline has a ten times lower threshold compared with
the US EA guideline; moreover, if a PAS is directed to Phase 2 Tier A or B, much
more experimental data must be elaborated for initial and in particular for refined
assessment, notably water/sediment fate and chronic effects testing. Based on currently available knowledge, both aspects may be defended with good scientific reasons. However, there are still some shortcomings in the EMA approach.
In Phase 2 Tier A the data set for environmental fate seems somewhat imbalanced. On the one hand, only a facultative ready biodegradability study is listed, but
if that does not meet the criteria for ready biodegradability, no additional higherlevel biodegradation information is requested, even though wastewater treatment is

Current State of Regulations for Human Pharmaceuticals ERA

27

by far the most important entry pathway of a PAS into the environment and removal
in sewage works is often the most significant fate process. On the other hand,
adsorption and sediment/water fate studies are requested, both of which (at least for
OECD106 and 308) are exacting and expensive studies that normally use radiolabelled substance. But basically the results are only used for deciding on the necessity of a terrestrial ERA or of a sediment ERA, respectively, in Phase 2 Tier B. With
the exception of PAS with either a logKOC >4 in the adsorption test or a systems
half-life <3 days in the sediment/water study (in both of which cases the substance
need not be assessed for groundwater risk), those two assays are not utilised any
further. While half-lives must be stated in the Expert Report, they are not actually
processed in a PEC refinement or used in the ERA. It is not easy to see why sophisticated sediment/water fate data should be determined if they are not really used; it
is not easy to see, either, why the distribution to sediment cannot be read from the
adsorption test, in particular as all sediment risk should be normalised to standard
sediment parameters with a specified organic carbon content, anyway. Instead of the
water/sediment fate test, which was developed to model a small ditch beside a field
for pesticides ERA and never meant to be an assay for surface water fate, there is an
OECD-validated alternative that really does test for surface water fate, the OECD309
surface water degradation test, where the biodegradation of a test substance in natural water with a small concentration of suspended natural sediment is investigated.
As Richard Murray-Smith (pers. comm.) commented in several workshops and conferences, this test would give more realistic and useful information on surface water
fate. Human PAS do not normally end up in small ditches close to fields, but they
will show up in surface waters.
On the ecotoxicity side, the EMA [26] PERA guideline consequently addresses
chronic effects. This is based on the realisation that (nearly) all PAS in surface
waters, whether rapidly degradable or not, show a phenomenon termed pseudopersistence, viz. relatively constant concentrations due to more or less continuous
input or replenishment from human use (e.g. [18]). Hence, environmental organisms are exposed in a constant manner, which can only be scientifically evaluated
using chronic-based PNECs. However, the EMA guideline does not give any guidance on how a chronic aquatic PNEC (normally based on a traditional deterministic
approach) could be further refined. For example, a more refined approach may be
useful in some cases through the use of probabilistic assessment methodologies
originally developed to support pesticides risk assessment (e.g. [13]). Indeed, probabilistic approaches have been recently applied for PERA of a few old human
PAS during the past years [70, 71].
In the EU, Guidelines are to be revisited and updated if necessary on a regular
basis. The EMA [26] PERA guideline was only 4 years old at the time of writing,
hence a revision may be somewhat premature. However, in view of some uncertainties in the guideline, the CHMP Safety Working Party of the EMA prepared and in
March 2011 published a Question and Answer Document (Q&ADoc; [27]) to
provide clarification and harmonise the use of the EMA [26] PERA guideline.
This Q&ADoc has due to the time passed since originally writing the manuscript,
this is now official become the official companion to the guideline. It gives pertinent

28

J.O. Straub and T.H. Hutchinson

information on how the regulators want to handle PERA in the EU in the next years.
Only selected items deemed important will be shortly highlighted in the following
paragraphs:
Generics are not exempted from providing an ERA and crossreference to the
ERA of the original applicant is not possible. Hence, a new applicant for a generic
PAS, also in combination with a new PAS, must provide a full PERA following
the EMA [26] guideline.
Significant increase in environmental exposure due to a variation remains
undefined.
The sediment/water fate test remains compulsory (except in the case of a positive
ready biodegradability test, as already stated). It may not be waived even if the applicant presents a sediment ERA under the assumption of all substance distributing to
the sediment. However, the testing of fully anaerobic systems of the water/sediment
fate test is not considered necessary in general, as even the aerobic systems will
develop anaerobic parts. Similarly, for environmental fate testing in soil (OECD307),
only aerobic systems are required.
For combination products the ERA should be performed separately for each PAS.
Metabolites are included up to Phase 2 Tier A in the total residue approach
adopted in the EMA [26] guideline. They may be subtracted for refining the
surface water PEC in Phase 2 Tier B, but then a full ERA is requested for all
metabolites excreted as 10% of the applied dose. PECs and PNECs are then
calculated separately for all substances investigated, and all PEC/PNEC ratios
are added together for the evaluation of the whole product.
Many PAS are ionisable compounds and present as charged acids, bases or zwitterions at environmentally relevant pH range (commonly accepted as pH 59).
Yet it still is the logKOW, measured for acids and bases at a nondissociating pH
value, that decides on bioaccumulation testing in Phases 1 and 2 Tier A. In the
Q&ADoc only the KOC from an OECD106 adsorption test is recognised to be a
possible function of the ionisability of a substance.
In the sediment/water fate test, the so-called bound residues are commonly
formed, which cannot be extracted even with appropriate solvents. However, the
bound residue fraction may not be subtracted from the sediment PEC, i.e. bound
residues are regarded as (ultimately) bioavailable.
The draft Q&ADoc does give more definition to the EMA [26] guideline, but it
also maintains the same highly precautionary approach to PERA. With the publication of the 2006 guideline it was the regulators clear statement that over the coming
years they wanted to collect PERAs to analyse them also for the scientific content
and usefulness of the guideline, and to review the scheme based thereon, but obviously this time has not come yet. The Precautionary Principle being a nondefined
and very controversially handled concept [32], it would seem that for the time being
the EMA considers it has not sufficient scientific information to include a weight of
evidence analysis and therefore remains on the conservative side.

Current State of Regulations for Human Pharmaceuticals ERA

29

PERA in Switzerland
In Switzerland, which is not a member of the EU, the relevant Medicines Registration
Ordinance (Arzneimittel-Zulassungsverordnung, AMZV; [2]) only requires information and documentation on ecotoxicity for human pharmaceuticals ([2], article
4,2,d), while for veterinary pharmaceuticals both data on ecotoxicity and potential
risks for the environment are required ([2], articles 9,2,b and 9,1,b, respectively).
There is no specific guideline for PERA nor any detailed requirements for ecotoxicity
basic data mentioned in the AMZV [2]. Swiss regulators accept EU PERAs following
the EMA [26] guideline.

PERA Developments in Canada

For the time being, pharmaceuticals in Canada are regulated under the New Substances
Notification Regulation (Chemicals and Polymers) [58], respectively, the New
Substances Notification Regulation (Organisms) [59], based on the Canadian
Environmental Protection Act [8, 15]. In the NSNR/C&P, all kinds of chemical substances or organisms imported into or manufactured in Canada that are not already on
the Canadian Domestic Substances List (DSL; http://www.ec.gc.ca/subsnouvellesnewsubs/default.asp?lang=En&n=47F768FE-1) must be notified to the authorities.
For substances, the substance-related information content of the notification package
depends on the total amount brought to the Canadian market in one calendar year. For
a chemical or biochemical substance (including PAS) not on the DSL, below a first
threshold of 100 kg per annum, no assessment is necessary, while increasing, defined
substance information base sets (schedule X information) become necessary in
higher tonnage bands (>1,000, >10,000, >50,000 kg/a). If the chemical or biochemical is already on the DSL, the first threshold (requiring no notification) is 1,000 kg/a,
with the same schedule X information necessary in higher tonnage bands.
However, a proper PERA guideline is currently (2012) under development in Canada.
For the time being there seems to be no official draft document available to the public.
Based on an earlier, nonattributable crude sketch that circulated a couple of years ago
(and which may not be relevant any longer), it is possible that certain experimental tests
that are not among the lower-tier studies in US or EU PERA schemes might become
standard first-tier studies in the Canadian scheme. It is expected that a final draft of the
Canadian PERA scheme will be published for a short public discussion and comment
phase in 2012/2013 and that a definitive version may be adopted in the same year.

PERA Developments in Japan

Japan has been developing a PERA guideline for some years, according to Yasuyoshi
Azuma (pers. comm.) of AstraZeneca, who presented on these activities at an

30

J.O. Straub and T.H. Hutchinson

international conference on PERA in Barcelona in 2009. PIE have been a topic for
public news and scientific investigations in Japan, with increasing concern about the
environmental safety of PAS in the recent past. In view of ongoing developments and
of the language barrier, only little information is available as to the probable contents
of a draft guideline. Still, an intermediate report from a mixed PERA study group led
by the Ministry of Health, Labor and Welfare in 2008 (cited by Dr Azuma) suggests
that PERA will become mandatory for new PAS, that categorical exclusions will
apply, that the actual PERA would be risk based (i.e. not only hazard based) and that
a tiered approach was preferred. In early tiers, a simple PEC would be calculated,
with the possibility of refinement in higher tiers, while effects characterisation would
be through chronic testing. Also, it perspired that a negative outcome of the PERA
would not be sufficient reason to deny registration and marketing approval.
Based on this information, assuming it is still current, Japan seems to be set on
developing a PERA scheme generally in line with existing guidelines elsewhere.
While no precise dates are known, a draft guideline for public comment and
finalisation is generally expected by about the year 2012/2013.

PERA Requirements in Australia

Australia has a requirement for a PERA to be submitted with new medicines registration in Annex I to Module 1 of the Common Technical Document issued by the
Australian Therapeutic Goods Administration [74]: Applications to register prescription medicines for human use should include [] an indication of any potential
risks presented by the medicine for the environment. This requirement is particularly applicable to new active substances and live vaccines. Applications for new
active substances may include [] an indication of relevant environmental hazards,
making reference to standard physicochemical tests and any appropriate testing
they have conducted on biodegradability, including some testing in sensitive species. [] The risk assessment overview should include an evaluation of possible
risks to the environment from the point of view of use and/or disposal and make
proposals for labelling provisions that would reduce this risk. ([74]; Annex I to
Module 1). There is no specific Australian PERA guideline nor is there information
about such a guideline being developed; however, the EU EMA [26] Guideline is
linked on the TGA homepage (link: http://www.tga.gov.au/docs/pdf/euguide/
swp/444700en.pdf, which directly opens the EMA Guideline). Based on own experience as an environmental risk assessor with an international research pharmaceuticals company, an EU PERA is acceptable to the Australian regulators.

Further PERA Requirements

Based on own experience, there are a few sporadic cases of further countries that
have started requiring PERAs, e.g. in South America. These have so far accepted

Current State of Regulations for Human Pharmaceuticals ERA

31

Spanish translations of the respective EU PERAs. It is not known whether these

requests were based on established national legislation or on a wish on the side of
environmental regulators to receive more pertinent information on new PAS. It may
be assumed that such requests will increase in numbers and that some countries will
establish formal legal requirements for PERAs, in view of developments in other
countries and regions.

Other PERA Initiatives: The Swedish Environmental

Classification and Simplified ERA of Old PAS Already
on the Market
At the EnvirPharma Conference on Human and Veterinary Pharmaceuticals in the
Environment in Lyon, France, in 2003, Prof. ke Wennmalm of the Stockholm
County Council (SCC) presented his concept of Environmental Classification of
Pharmaceuticals, by assessing the environmental hazards of PAS by three criteria,
persistence, bioaccumulability and ecotoxicity (PBT). The SCC is one of the biggest healthcare providers (including pharmaceuticals distribution) in Sweden. In
2003 the SCC started assessing the hazards of old PAS that were already on the
market, which are not covered by PERA guidelines in the EU or USA. This hazard
assessment proceeds by assigning numerical values from 0 to 3 to indicators or
substitutes for PBT properties (ready, inherent or nonbiodegradability; bioaccumulation or logKOW; ecotoxicity data); in case of no available data for a category, the
maximum, worst-case of 3 points will be applied. This results in a total between 0
and 9 points per PAS. Updated results of this PAS hazard assessment are available
as a printable booklet (current 2012 version at: http://www.janusinfo.se/Global/
Miljo_och_lakemedel/miljobroschyr_engelsk_2012_uppslag.pdf).
The pharmaceutical industry, which has delivered the PAS basic data for the SCC
classification since 1993, suggested to improve the classification by not only considering hazard but also relating hazard to exposure, i.e. extending the SCC hazard
assessment to a simplified PERA for PAS already on the Swedish market. In this
so-called Voluntary Environmental Drug Classification System [53], substance-relevant
data on physicochemical properties, (bio)degradability, persistence, bioaccumulability and (acute or chronic) ecotoxicity are contrasted with surface water PECs for
Sweden based on actual annual use of the respective PAS. The results are expressed
in three different formulations, on level 1 as a simple phrasing of the risk for lay
people, mainly patients (e.g. use of the medicine has been considered to result in
insignificant/low/moderate/high environmental risk, with the qualifiers as appropriate
for the PAS in question). On a second level intended for prescribers of the medicines, the environmental risk is given as in level 1, with additional information as to
environmental degradation/persistence, to bioaccumulability or to PBT characteristics, all as appropriate for the PAS in question. On level 3, the full information
available is given for specialists to assess and judge themselves. However, all levels
of information are open to the public.

32

J.O. Straub and T.H. Hutchinson

All PAS on the Swedish market are going to be integrated into this scheme;
moreover, existing assessments are updated with recent pharmaceutical use data and
new substance-specific data if available, every 3 years. Available risk classifications
can be searched at FASS.se (http://www.fass.se/LIF/miljo/miljoinfo.jsp; search by
ATC code or substance name (substans, e.g. sulfamet), select one single PAS
(e.g. sulfametoxazol in Swedish), then one single product (e.g. Bactrim forte),
click on FASS on top of the product window, scroll down to subheading
Miljpverkan, then click on Ls mer >> to see the detailed environmental
information).
The Swedish hazard and PERA systems address, at least in part, questions about
old PAS already on the market. Current US and EU PERA guidelines do not primarily address old PAS but mostly new ones. Hence, the Swedish classification is
an important step towards a broader base for a risk overview for PIE. As a consequence, several other EU member states have shown interest in the Swedish
classification as a model for themselves or for the whole of the EU for existing PAS.
In particular the Nordic countries with Norway and Denmark (beside Sweden), but
also Germany, the Netherlands and the UK are looking into the Swedish model,
possibly also into elevating it with additions to EU level.

Other, Non-PERA Regulations that Still Have

an Indirect Influence on PIE and PERA
Is PERA Beyond REACH?
Registration, Evaluation, Authorisation and Restriction (which is less often mentioned but still part of the full name) of Chemicals, the new chemicals management named REACH in the European Community [23], aims at regulating the
production, marketing and use of all chemicals not covered by other pertinent legislation. REACH intends to improve chemicals safety throughout, by assessing hazard
and risk in function of annual amounts put on the market on the one hand and of
hazardous properties marking the chemical as a substance of very high concern
(SVHC), viz. carcinogenic or mutagenic or reprotoxic (CMR) or persistent and
bioaccumulative and highly ecotoxic (PBT) substances, on the other.
Registration under REACH is not necessary for non-hazardous substances used
in amounts below one metric tonne per year; all other chemicals may only be used
after they have been duly (pre-)registered. However, a notification of Classification
and Labelling is required for all chemicals, irrespective of amounts. For low tonnages the dossier is comparatively simple, but with increasing amounts or in case of
SVHCs, additional prescribed data sets become necessary, including physicochemical, toxicological and environmental substance basic data, defined use scenarios for
the chemical in question and chemical safety reports based thereon (e.g. [66]).
Human pharmaceuticals (also veterinary medicines, medical devices, cosmetics,

Current State of Regulations for Human Pharmaceuticals ERA

33

food or feedstuffs) in general are exempt from REACH, as they are assessed under
different legislation. But REACH is still highly relevant for the pharmaceutical
industry, as all starting materials, intermediates and also ancillary compounds like
solvents fall under the chemicals legislation. However, in some avowedly rare cases,
even a PAS may fall under full REACH coverage. This is the case where a PAS is
formulated as a prodrug, mostly for reasons of improved absorption and bioavailability, and is only metabolised back to the actual PAS in the body. If this PAS
already occurs in the chemical synthesis, to be esterified or otherwise chemically
converted to the prodrug as dispensed, the PAS technically is an intermediate and
therefore falls under REACH.
On the other hand, according to the EMA [26] ERA guideline, in case of prodrugs,
the actual PAS should be assessed in an ERA. This leads to a situation where the
same PAS may be investigated and assessed following two different, non-congruent
guidelines. In the EMA scheme, both a base set including a water-sediment fate test
(OECD test guideline 308) and chronic ecotoxicity studies with algae, daphnia and
fish are needed, while in the REACH scheme [25], up to a high tonnage, comparatively simple ready biodegradability and acute ecotoxicity studies suffice. On the
other hand, under REACH chemicals may be restricted or even denied marketing
(unless their necessity can substantiated) based on SVHC characteristics, which
is not possible following the human medicines legislation [22], where patient
benefit takes precedence before environmental concerns. Hence, there may be
double legislative coverage, inconsistent ERA and contradictory regulatory options
for some few PAS.

The European Water Framework Directive and PAS

The EU Water Framework Directive (WFD; [21]) aims at achieving enhanced protection and improvement of the aquatic environment, which covers inland surface
waters, ground, transitional and coastal marine waters, and (re-)establishing good
ecological status of aquatic ecosystems. One express means of attaining these goals
is through specific measures for the progressive reduction [] and the cessation or
phasing-out of discharges, emissions and losses of the priority hazardous substances, thereby ensur[ing] the progressive reduction of pollution of groundwater
and prevent[ing] its further pollution. Besides these priority substances, there may
well be additional Environmental Quality Standards (EQS), corresponding to legal
limit values, for other pollutants. EQS are set based on a compilation and interpretation of basic substance data including in particular environmental fate and toxicity,
mammalian toxicity and bioaccumulation information. PAS are not exempt from
the WFD; on the contrary, several PAS were included in a first list of candidate EQS
substances at a relatively late stage in development of that list, end of 2009. In a
proposal for an update of the WFD dating to January, 2012, oestradiol, ethinyloestradiol and diclofenac were identified as candidate- for an official EQS within
the scope of the WFD. Further PAS may be included in a proposed watchlist of

34

J.O. Straub and T.H. Hutchinson

additional substances that should be analysed in surface waters and some of them
may also warrant the future development of EQS values.
However, the question may be, what would the regulators decide in case an EQS
for PAS were regularly breached at one or more sampling sites? Theoretically, the
WFD has the power to ban substances of concern from further use. But banning
pharmaceuticals might not be that simple given the human health benefits in prophylactic and disease treatment contexts. Indeed, the current Human PAS EU
Directive 2001/83/EC [24] specifically states that human PAS may not be kept from
the market, even in case of a negative PERA (in contrast to veterinary PAS, biocides
or pesticides, which may be restricted or even banned in such a case; [68]). Possibly,
the best way forward would be a specific improvement of sewage treatment works
with a view to increase the removal of PAS, through biological or physicochemical
means, as already advocated years ago by OBrien and Dietrich [60].

ERA for the Production of Pharmaceuticals?

Subsequent to a first Swedish publication [52] evidencing very high concentrations
of PAS, particularly antibiotics, in the effluent of a wastewater treatment plant serving an industrial park near Hyderabad, India, the production of pharmaceuticals
(both of PAS and of formulated products) came under further scrutiny. In a later
publication from the same group [51] it was shown that many PAS contained in
medicines marketed in Sweden were originally produced in India, most of them
with the same insufficient control and treatment of production effluents. One of the
consequences of these investigations was a proposal by the Swedish Medical
Products Agency (MPA) that a requirement for an environmental certification of
the production facilities be introduced into the international regulations on Good
Manufacturing Practice and further that the current EU legislation for the authorisation of medicinal products for humans should be changed so that an ERA [of the
production of the PAS] is also included in the approval [56]. If so, a production
ERA of the PASs and of the finished medicines would also become part of the
PERA. However, this would be in contradiction to both the EU REACH legislation
[23], which already covers all intermediates of pharmaceuticals production, and to
the EMA [26] PERA Guideline, which expressly excludes the production from the
scope of the PERA.
What the Swedish initiative certainly does is to point the finger at preventable
environmental exposures to PAS that have no therapeutical, palliative or preventative benefit. While some exposures from patients excretions may prove to be
difficult or even impossible to be prevented, these uses at least have important medical benefits, while exposure through insufficient retention or treatment of wastewaters have none. On the other hand, not all productions of PAS lead to inacceptable
environmental exposure, as shown by Boegrd et al. [7] and Hoerger et al. [39] for
two production sites in Europe, but a recent publication does show high levels of
PAS downstream of US formulation facilities [62]. It will remain to be seen whether

Current State of Regulations for Human Pharmaceuticals ERA

35

the reports showing high receiving water concentrations of PIE due to production
will entail changes in EU or US regulations for PAS.

A Short Comparison with PERA

for Veterinary Pharmacueticals
The situation for veterinary medicinal products (VMP) ERA is formally different
from that for human PAS. VMP ERA, which had been developed independently in
several countries (similar to the current situation for human PERA), has been harmonised between the EU, Japan and USA in the so-called International Cooperation
for the Harmonisation of technical requirements for the registration of Veterinary
medicinal product (VICH) process. Canada, Australia and New Zealand have
adopted (and further countries may accept) the VICH PERA guidelines, which are
split in to two phases.
Based on the concept that VMPs with very limited use or a high rate of metabolisation will have limited environmental exposure and effects, Phase I [80] is mainly
a decision tree for filtering out those VMPs where no ERA is needed. Such categorical exclusions comprise:
VMPs that are formally exempt from ERA by legislation
VMPs that are natural substances, the use of which will not alter the background
concentrations in the environment
VMPs for exclusively non-food animals (i.e. pets or companion animals)
VMPs for use in a minor species that is handled and treated similarly to a major
species for which a VICH PERA already exists
VMPs that are only used to treat a small number of animals in a herd
VMPs that are extensively metabolised in the treated animal
Then the decision tree splits into two branches, depending on whether the treated
species are aquatic or terrestrial. For VMPs used in aquaculture, the categorical
exclusions comprise:
VMPs that are not released into the aquatic compartment (e.g. aquarium species)
by disposal of the aquatic waste matrix
VMPs that are used in a confined facility and that are not endo- or ecto-parasiticides
and where the entry into the EIC from the aquaculture facilities is <1 mg/L (or
where this EIC is mitigated to <1 mg/L by installations or proven degradation
mechanisms)
On the terrestrial side, the categorical exclusions comprise
VMPs that are not released into the terrestrial compartment by disposal of the
terrestrial waste matrix
VMPs that are used in animals reared on pasture and that are not endo- or ectoparasiticides and where the soil PEC is <100 mg/kg soil (or where this PEC is
mitigated to <100 mg/kg by installations or proven degradation mechanisms)

36

J.O. Straub and T.H. Hutchinson

VMPs that are used in animals not reared on pasture and where the soil PEC
from spreading manure is <100 mg/kg soil (or where this PEC is mitigated to
<100 mg/kg by installations or proven degradation mechanisms)
For all the above exclusions, the VICH PERA stops with a Phase I report that
discusses the basis for this decision. The Phase I document [79] also contains a
lengthy Q&A part where explanations for certain questions are given. In case a
VMP would normally go into Phase II but the PEC can be brought below the aquatic
or terrestrial threshold, discussion with the regulators is necessary before deciding
how to proceed.
All other VMPs must be assessed in Phase II [80], which is a two-tiered procedure. Tier A is a simpler, more conservative assessment; if a conclusion of no
significant risk cannot be reached in Tier A, the assessment must progress into Tier
B with more demanding data. In Phase I a total residue approach is taken, metabolites are not normally considered and PECs have to be calculated based on the
applied dose; note that in Phase II Tier A, PNECs are calculated for every single
group of test organisms.
For practical reasons, Phase II is divided into three major branches, aquaculture,
intensively reared terrestrial animals and pasture animals, due to different entry
pathways into the environment. Detailed guidance for the three branches is given
regarding types of environmental exposure, experimental data for Tiers A and B and
calculation and refinement of PECs for the respective compartments.
General data requirements for Phase II Tier A comprise:

Water solubility (OECD105)

Dissociation constants (OECD112)
UVvisible absorption spectrum (OECD101)
Melting point/range (OECD102)
Vapour pressure (OECD104), normally by QSPR calculation, except if there is
evidence that the vapour pressure is >105 Pa at 20C, in which case it should be
determined experimentally
n-Octanol/water partition coefficient (OECD107 or OECD117); note that for
ionisable substances there is a cryptic footnote that if appropriate, the logKOW
for such substances should be measured on the non-ionised form at environmentally relevant pHs
Soil adsorption/desorption (OECD106) reporting both Kd and KOC values
In case of primary exposure to soil, soil biodegradation (OECD307)
In case of primary exposure to the aquatic compartment, degradation in water/
sediment systems (OECD308); note that for marine applications, the possibility
of doing the water/sediment study with seawater should be discussed with the
regulators
(Optional) photolysis in water (OECD316) for aquaculture (consider seawater
photolysis test for marine applications) or on soil (OECD guideline in preparation) for terrestrial branches
(Optional) hydrolysis (OECD111)

Current State of Regulations for Human Pharmaceuticals ERA

37

Acute aquatic ecotoxicity base set, for both terrestrial and aquaculture applications (consider seawater tests for marine applications), with appropriate endpoint
and AF to derive the group-specific PNEC

Freshwater algal growth inhibition (OECD201), EC50, AF = 100 or

Seawater algal growth inhibition (ISO10253), EC50, AF = 100
Freshwater Daphnia immobilisation (OECD202), EC50, AF = 1,000 or
Saltwater crustacean acute toxicity (ISO14669), EC50, AF = 1,000
Freshwater fish acute toxicity (OECD203), LC50, AF = 1,000 or
Seawater fish acute toxicity (seek guidance), LC50, AF = 1,000

Terrestrial ecotoxicity base set, for terrestrial branches/soil exposures, with

appropriate endpoint and AF to derive the group-specific PNEC
Microbial nitrogen transformation (OECD216), to be tested at 1 and 10 the
soil PEC; note no AF, but pass if difference in nitrate formation is 25%
compared with controls at any time before 28 days; else the study should be
prolonged to 100 days in Phase II Tier B
Terrestrial plants seedling emergence and growth test (OECD208), EC50,
AF = 100
Earthworm subacute toxicity (OECD220) or reproduction test (OECD222),
NOEC, AF = 10
Specifically for endo- or ecto-parasiticides used in pasture treatments, the following tests on dung fauna are also recommended:
Dung fly larvae acute toxicity test (OECD228), EC50, AF = 100
Dung beetle larvae acute toxicity test (OECD test in preparation, seek guidance),
EC50, AF = 100
For Phase II Tier A risk characterisation, the initial PEC for soil or the aquatic
compartment from Phase I is to be compared with all appropriate PNECs derived
as a PEC/PNEC risk quotient (RQ) in VICH terminology. If all RQs are <1 and
there is no risk of accumulation of the VMP in the environment, based on persistence data, there is no significant risk and the VICH PERA can be concluded. If
the RQ is 1 for any organism tested, the initial PEC should be refined with
information on metabolism/excretion and on environmental degradation
(OECD307, 308). If the RQ is still 1 for any organism tested, the PERA should
progress to Phase II Tier B and chronic testing should be done on the organism
concerned refine the PNEC. In the case of pasture applications, if the RQ is 1
for a dung organism, the initial dung PEC, which assumes the whole dose of
VMP excreted in one day, should be refined based on realistic excretion patterns;
if the refined RQ is 1, regulatory guidance should be sought. Further, if the
logKOW is 4, bioaccumulation should be investigated in Phase II Tier B. If the
aquatic invertebrate RQ is 1, an initial sediment assessment based on equilibrium partitioning is recommended; if the RQ is still 1, a refinement of the calculated sediment PNEC through a preferably chronic sediment toxicity study is
recommended.

38

J.O. Straub and T.H. Hutchinson

In Phase II Tier B, the following organism- or compartment-specific assessments

are described:
For bioaccumulation, a fish bioaccumulation study (OECD305) should be performed (normally with radio-labelled material). If the bioconcentration factor is
>1,000, regulatory guidance should be sought.
For chronic aquatic effects testing, the following are recommended for those
organisms with a Tier A RQ 1
Algal growth inhibition (freshwater and marine), but use NOEC and an AF of
10 from the tests already performed in Tier A
Freshwater Daphnia reproduction test (OECD210), NOEC, AF = 10,
For seawater seek guidance as to test guideline, NOEC; AF = 10
For freshwater fish early life stage test (OECD211), NOEC, AF = 10
For saltwater seek guidance for a chronic fish test, NOEC, AF = 10
For freshwater sediment, invertebrate chronic toxicity (OECD219 if entry
into the environment is through water, OECD218 if it is through sediment or
adsorbed to soil in surface run-off), NOEC, AF = 10
For seawater sediment, invertebrate chronic toxicity (seek guidance), NOEC,
AF = 10
For terrestrial long-term effects, the following are recommended for those organisms with a Tier A RQ 1
Terrestrial plants seedling emergence and growth test (OECD208), repeat test
with two additional species from the most sensitive group and the most sensitive species in the first Tier A test, NOEC, AF = 10
Earthworm test (neither test guideline nor endpoint, respectively, AF given,
hence seek guidance)
Microbial nitrogen transformation (OECD216) prolonged to 100 days in
Phase II Tier B; note no AF, but pass if difference in nitrate formation is
25% compared with controls
If after Tier B testing any RQ, including dung fauna RQ, is still 1 or if the pass
level was not reached in the soil nitrification test, regulatory guidance should be
sought. Also, the specific guidance for Phase II approaches for the three branches
helps to identify and deal with common problems. If there are still unresolved questions at the end of Phase II Tier B, the applicant should contact the regulators for
discussing risk mitigation strategies (e.g. use only under specified conditions to
minimise environmental exposure), which would result in a registration with restrictions on use and mandatory labelling of the product regarding the environmental
hazards.
The VICH [79, 80] guidelines present a highly detailed scheme of investigating
environmental risks from VMPs. Many applications are excluded in Phase I, where
a non-significant environmental exposure is assumed to result. In view of the discontinuous administration, in contrast to many human PAS, Phase II Tier A relies

Current State of Regulations for Human Pharmaceuticals ERA

39

on acute data for an initial RQ assessment. However, this assessment is rendered

specific for the three branches, respectively, for the main (and secondary) environmental compartments exposed by the aquaculture, pasture and intensively
reared applications. The VICH procedure differs from other PERA schemes in that
additional and often highly important environmental degradation pathways, e.g.
aquatic or soil surface photodegradation are specifically mentioned. Phase II Tier B
then is more conventional on the effects side, with chronic-based PNECs. With
restricted registration looming at the end of the process, VICH applicants have very
good reasons to refine their PECs and PNECs to the maximum possible to attain
RQs < 1.

Outlook and Conclusion

Through regional development, current international PERA guidelines (and probably the ones in preparation) are different in many details, agreeing only on very
broad levels or methodologies. On the other side, the environment is not that different on both sides of the Atlantic or the Pacific Oceans, PAS are often exactly the
same, entry pathways into and fate processes within that environment are highly
comparable and so, basically, are environmental organisms and ecological functions
within compartments.
Moreover, at higher tiers or levels in the PERA process, the published guidelines
(including the VICH Phase II) are much better comparable than at lower tiers, where
evident discrepancies exist. A comparison of the existing human and veterinary
ERA guidelines, together with the EU Biocides [20] guideline as an outgroup representative, is given in Table 1. On the other hand, all stakeholders in the investigation of environmental risks from pharmaceuticals would be expected to have the
same expectations: clear, transparent, scientifically sound and comparable (if not
identical) schemes to assess substances, with confidence and decreased uncertainty,
on a local, national and international level. This, for human PERA at least, is still
missing.
A side glance to veterinary PERA regulations and the VICH process shows the
way forward: Even though some of the human PERA guidelines are only being
developed or finalised, all of them may already now be in need of revision with a
view to international harmonisation. This would facilitate the work of both applicants and regulators and, most of all, it would render PERA much more transparent
on an international level.
PERA promises to remain both scientifically interesting, economically and societally important but also at times somewhat politically confusing. The scientific
lessons learned in PERA are also likely to provide positive benefits for other areas
where chemicals are used to support diverse industries from antifoulant paints to
food production.

[AU7]

Threshold PEC/ Yes. 1 mg/L EIC

trigger value
(sewage works
effluent);
0.1 mg/L surface
waters

Yes

Tiered ERA

0.01 mg/L surface

waters

Yes

NI

Yes

No

No

No

NI

Yes. logKOW >4.5

or potential
endocrine
disruptor/
reproductive
effects at
<10 ng/L
Yes

(Yes) if the PAS

may
significantly
affect the
human
environment

Yes (chemical Yes (chemical

groups)
groups)

Yes (chemical
groups)

No

Old PAS GL
Sweden

Yes (chemical
groups)

NI

Draft GL
Japan

No (EU member
states)

EU EMA [26]

No

Categorical
inclusions
(special
assessment)

Internationally
harmonised
GL
Categorical
exclusions

US (1998)
VICH II [81]

Yes
Yes. 1 mg/L
NA (higher tier)
EIC for
surface
waters or
100 mg/kg
EIC for soil

Yes

Yes (EU, USA, Yes (EU, USA, J,

J, CAN,
CAN, AUS,
AUS, NZ)
NZ)
Yes (regulation NA (higher tier)
and
application
based)
Yes. Endo-/
NA (higher tier)
ectoparasiticides

VICH I [80]

(Yes) partly,
refinement
possible
No

(No) all biocides

must de assessed

No

No (EU member
states)

EU biocides [20]

Table 1 Comparison of existing guidelines (GLs) for the environmental risk assessment (ERA) of pharmaceuticals, with EU biocides ERA as an outgroup
Human PAS ERA GLs
Veterinary medicinal products GLs Outgroup GL

40
J.O. Straub and T.H. Hutchinson

(No) data increasing with every

tier, but no
fixed genuine
base set

No. Only in highest Yes. Phase 2

tier
10,000 (one acute
10 (three chronic
EC/LC50
NOECs phase 2
tier 1),
tier A)
1,000 (three acute
EC/LC50s
tier 2),
100 (unknown
number of
chronic EC/
LC50s tier 3)

Defined base set

Chronic testing
compulsory
Assessment
factors for
surface water
PNEC

Draft GL
Japan
VICH I [80]

1,000 (three
NA (first tier)
acute EC/
LC50s),
100 (one, fish
or daphnia,
chronic
NOEC),
50 (two chronic
NOECs),
10 (three
chronic
NOECs)

NI

NA (First tier)

(No) exposure
threshold
approach

No

(Yes)
insufficient
data
situation
will be
noted

(Yes) straight- Yes. straightforward for

forward
PEC and
flow
acute-based
scheme
PNEC, little
guidance
for
refinements

Old PAS GL
Sweden
Yes. EU TGD [76]
and TGD-based
EUSES
application

EU biocides [20]

(continued)

Yes. Both for

Yes
phase II tier
A and
compartment/
organism
specific in
phase II tier B
Yes (higher tier) No. only in higher
tier
10 (chronic
1,000 (three acute
NOEC), no
EC/LC50s),
overall PNEC
lower for
for surface
higher-tier
water or soil,
chronic tests: 10
but single(three chronic
species risk
NOECs)
quotients

Yes. Detailed for

the three
branches
aquaculture,
intensively
reared and
pasture
animals

VICH II [81]

Veterinary medicinal products GLs Outgroup GL

Yes

Yes.
NI
Straightforward
for phases 1 and
2 tier A,
insufficient for
phase 2 tier B
and reproductive/endocrine
assessment
Yes. Phase 2 tier A NI
base set and in
part for phase 2
tier B

Yes.
Straightforward
flow scheme,
some lack of
guidance for
highest tier 3

EU EMA [26]

Technical
guidance
available

US (1998)

Human PAS ERA GLs

Current State of Regulations for Human Pharmaceuticals ERA

41

Yes

ERA applicable
for new
substances
only

(Yes) except repeat Yes

registrations/
variations
leading to
higher exposure,
generics for new
applicants
No. Compulsory
No
labelling and
proposal for
minimising
exposure may
apply

VICH I [80]

No

No. specific
programme
for old
PAS

NA

(No)

(No) application
restrictions
and compulsory labelling
may apply

NA (higher tier)

Phase II tier B

Yes. logKOW 4
in phase II
tier A

VICH II [81]

Yes. Limited
authorisation
with use
restrictions and
compulsory
labelling and
requirement of a
safety data sheet
may apply

(No) only acute

testing required
in annex IIA/VII
(No)

Yes. Part of the base

set in Annex
IIA/VII

EU biocides [20]

Veterinary medicinal products GLs Outgroup GL

NA (first tier)
(No) but PAS
are regarded
as
bioaccumulating if
logKOW 3
No
NA (first tier)

Old PAS GL
Sweden

Based in part on table III in Straub [69]. As there is no information available on the draft Canadian PERA GL, this was not used in the comparison. For details
and in case of uncertainties refer to the original guideline
Note (no) or (yes) in brackets means this question/criterion cannot be unambiguously decided/applied, but the tendency is indicated
EIC entry into the environment concentration (before dilution); NA not applicable; NI no information

Registration
dependent
on result
of ERA

NI

Possibly tier 3

Comprehensive
ERA

Phase 2 tier B

NI

EU EMA [26]

Draft GL
Japan

Bioaccumulation No. logDOW 3.5 at Yes. logKOW 4.5

testing/trigger
pH of ~7 only
in phase 1,
requires chronic logKOW 3 in phase
2 tier A
testing

US (1998)

Table 1 (continued)
Human PAS ERA GLs

42
J.O. Straub and T.H. Hutchinson

Current State of Regulations for Human Pharmaceuticals ERA

43

Acknowledgements Our thanks to our colleagues in Europe and the USA (I Radtke; plus many
others), Canada (L Jack, A Beck), Japan (Y Azuma, T Tosaka) and Australia (L Justice) for discussions, help and support.

References
1. Aherne GW, Briggs R (1989) The relevance of the presence of certain synthetic steroids in the
aquatic environment. J Pharm Pharmacol 41:735736
2. AMZV (2001) Verordnung des Schweizerischen Heilmittelinstituts vom 9. November 2001
ber die Anforderungen an die Zulassung von Arzneimitteln (ArzneimittelZulassungsverordnung, AMZV), Stand am 12. September 2006. SR 812.212.22, The Swiss
Confederation, Berne. http://www.admin.ch/ch/d/sr/c812_212_22.html. Accessed Aug 2010
3. AP (2009) An Associated Press Investigation: pharmaceuticals found in drinking water. http://
hosted.ap.org/specials/interactives/pharmawater_site/index.html. Accessed July 2010
4. Ayscough NJ, Fawell J, Franklin G, Young W (2000) Review of human pharmaceuticals in the
environment. R&D Technical Report P390. Environment Agency, Bristol, UK
5. Barnes KK, Kolpin DW, Focazio MJ, Furlong ET, Meyer MT, Zaugg SD, Haack SK, Barber
LB, Thurman EM (2008) Water-quality data for pharmaceuticals and other organic wastewater
contaminants in ground water and in untreated drinking water sources in the United States,
200001: US Geological Survey Open-File Report 20081293, 7 p. plus tables. http://pubs.
usgs.gov/of/2008/1293/pdf/OFR2008-1293.pdf. Accessed July 2010
6. Belfroid AC, Van der Horst A, Vethaak AD, Schfer AJ, Rijs BJ, Wegener J, Cofino WP
(1999) Analysis and occurrence of estrogenic hormones and their glucuronides in surface
water and wastewater in The Netherlands. Sci Total Environ 225:101108
7. Boegrd C, Coombe V, Holm G, Taylor D (1998) The concentrations and potential environmental impact of pharmaceuticals in the effluent from a major AstraZeneca manufacturing
facility. Poster, KNAPPE Conference, Nmes (France), February 19th20th, 2008
8. Breton R, Boxall A (2003) Pharmaceuticals and personal care products in the environment:
regulatory drivers and research needs. QSAR Comb Sci 22:399409
9. Bruce GM, Pleus RC, Snyder SA (2010) Toxicological relevance of pharmaceuticals in drinking water. Environ Sci Technol 44:56195626
10. Buser HR, Mller MD, Theobald N (1998) Occurrence of the pharmaceutical drug clofibric
acid and the herbicide mecoprop in various Swiss lakes and in the North Sea. Environ Sci
Technol 32:188192
11. Calamari D, Zuccato E, Castiglioni S, Bagnati R, Fanelli R (2003) Strategic survey of therapeutic drugs in the rivers Po and Lambro in Northern Italy. Environ Sci Technol 37:12411248
12. Caldwell DJ, Mastrocco F, Nowak E, Johnston J, Yekel H, PFeiffer D, Hoyt M, DuPlessie BM,
Anderson PD (2010) An assessment of potential exposure and risk from estrogens in drinking
water. Environ Health Perspect 118(3):338344
13. Campbell PJ, Arnold DJS, Brock TCM, Grandy NJ, Heger W, Heimbach F, Maund SJ, Streloke
M (1999) Guidance document on higher-tier aquatic risk assessment for pesticides (HARAP).
SETAC Europe, Brussels (Belgium)
14. CDER (1998) Guidance for industryenvironmental assessment of human drug and biologics
applications. US Dept of Health and Human Services, Food and Drug Administration, US
Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and
Research (CBER). July 1998, CMC 6, Revision 1. http://www.fda.gov/downloads/Drugs/
GuidanceComplianceRegulatoryInformation/Guidances/ucm070561.pdf. Accessed July 2010
15. CEPA (1999) Canadian Environmental Protection Act, 1999. S.C.1999, c.33. Current version
at http://laws-lois.justice.gc.ca/eng/C-15.31/20100726/page-0.html?rp2=HOME&rp3=SI&rp
1=environmental%20protection%20act&rp4=exact&rp9=cs&rp10=L&rp13=50#idhit1 .
Accessed July 2010

44

J.O. Straub and T.H. Hutchinson

16. Christensen FM (1998) Pharmaceuticals in the environmenta human risk? Regul Toxicol
Pharmacol 28(3):212221
17. Cunningham VL, Binks SP, Olson MJ (2009) Human health risk assessment from the presence
of human pharmaceuticals in the aquatic environment. Regul Toxicol Pharmacol 53(1):3945
18. Daughton CG, Ternes TA (1999) Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspect 107:907938
19. De Lange HJ, Noordoven W, Murk AJ, Lrling M, Peeters ETHM (2006) Behavioural
responses of Gammarus pulex (Crustacea, Amphipoda) to low concentrations of pharmaceuticals. Aquat Toxicol 78:209216
20. EC (1998) Directive 98/8/EC of the European Parliament and of the Council of 16 February
1998 concerning the placing of biocidaly products on the market. Off J L123:163
21. EC (2000) Directive 2000/60/EC of the European Parliament and of the Council of 23 October
2000 establishing a framework for Community action in the field of water policy. Off J
L327:172
22. EC (2001) Directive 2001/83/EC of the European Parliament and of the Council of 6
November 2001 on the Community code relating to medicinal products for human use. Off J
L311:67128. Current (end-2009) consolidated version available at http://eur-lex.europa.eu/
LexUriServ/LexUriServ.do?uri=CONSLEG:2001L0083:20091005:EN:PDF. Accessed 13
Jan 2010
23. EC (2006) Regulation (EC) No 1907/2006 of the European Parliament and of the Council of
18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of
Chemicals (REACH), establishing a European Chemicals Agency, amending Directive
1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation
(EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives
91/155/EEC, 93/67/EEC, 93/105/EC and 2000/21/EC. Off J L396:1849. Current (end-2009)
consolidated version available at http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CO
NSLEG:2006R1907:20090220:EN:PDF. Accessed 11 Jan 2010
24. ECHA (2008) European Chemicals Agency, Guidance for the implementation of REACH.
Chapter R11, Guidance on information requirements and chemical safety assessment, Part C:
PBT Assessment. http://guidance.echa.europa.eu/docs/guidance_document/information_
requirements_part_c_en.pdf?vers=20_08_08. Accessed July 2010
25. ECHA (2010) European Chemicals Agency, REACH and Classification & Labelling Guidance.
http://guidance.echa.europa.eu/guidance_en.html. Accessed 11 Jan 2010
26. EMA (2006) Note for guidance on environmental risk assessment of medicinal products for
human use. CPMP/SWP/4447/00. European Medicines Agency, London. http://www.ema.
europa.eu/pdfs/human/swp/444700en.pdf. Accessed 11 Jan 2010
27. EMA (2011) Questions and answers on Guideline on the environmental risk assessment of
medicinal products for human use, 17 March 2011. EMA/CHMP/SWP/44609/2010, EMA,
London (UK). http://www.ema.europa.eu/docs/en_GB/document_library/Other/2011/04/
WC500105107.pdf. Accessed April 2012
28. Farr M, Ferrer I, Ginebreda A, Figueras M, Olivella L, Tirapu L, Vilanova M, Barcel D
(2001) Determination of drugs in surface water and wastewater samples by liquid chromatographymass spectrometry: methods and preliminary results including toxicity studies with
Vibrio fischeri. J Chromatogr A 938:187197
29. Fernndez C, Gonzlez-Doncel M, Pro J, Carbonell G, tarazona JV (2010) Occurrence of
pharmaceutically active compounds in surface waters of the Henares-Jarama-Tajo river system
(Madrid, Spain) and a potential risk characterization. Sci Total Environ 408:543551
30. Fong PP (1998) Zebra mussel spawning is induced in low concentrations of putative serotonin
reuptake inhibitors. Biol Bull 194:143149
31. Fong PP, Molnar N (2008) Norfluoxetine induces spawning and parturition in estuarine and
freshwater bivalves. Bull Environ Contam Toxicol 81(6):535538
32. Foster KR, Vecchia P, Repacholi MH (2000) Science and the precautionary principle. Science
288:979980

Current State of Regulations for Human Pharmaceuticals ERA

45

33. Garrison AW, Pope JD, Allen FR (1976) GC/MS analysis of organic compounds in domestic
wastewaters. In: Keith LH (ed) Identification & analysis of organic pollutants in water. Ann
Arbor Science Publications, Ann Arbor, MI, pp 517556
34. Halling-Srensen B, Nors Nielsen S, Lanzky PF, Ingerslev F, Holten Ltzhft HC, Jrgensen
SE (1998) Occurrence, fate and effects of pharmaceutical substances in the environmenta
review. Chemosphere 36(2):357393
35. Heberer T, Butz S, Stan H-J (1995) Analysis of phenoxycarboxylic acids and other acidic
compounds in tap, ground, surface and sewage water at the low ng/l level. Int J Anal Chem
58:4353
36. Henschel K-P, Wenzel A, Diedrich M, Fliedner A (1997) Environmental hazard assessment of
pharmaceuticals. Regul Toxicol Pharmacol 25:220225
37. Hignite C, Azarnoff DL (1977) Drugs and drug metabolites as environmental contaminants:
chlorophenoxyisobutyrate and salicylic acid in sewage water effluent. Life Sci 20:337342
38. Hilvarsson A, Hallsdorsson HP, Granmo A (2007) Medetomidine as a candidate antifoulant:
sublethal effects on juvenile turbot (Psetta maxima L.). Aquat Toxicol 83(3):238246
39. Hoerger CC, Drr B, Schlienger C, Straub JO (2009) Environmental risk assessment for the
galenical formulation of solid medicinal products at Roche Basle, Switzerland. Integr Environ
Assess Manag 5(2):331337
40. Honkoop PJC, Luttikhuizen PC, Piersma T (1999) Experimentally extending the spawning
season of a marine bivalve using temperature change and fluoxetine as synergistic triggers.
Mar Ecol Prog Ser 180:297300
41. Huffman MA, Seifu M (1989) Observations on the illness and consumption of a possibly
medicinal plant Vernonia amygdalina (Del.), by a wild chimpanzee in the Mahale Mountains
National Park, Tanzania. Primates 30(1):5163
42. Huffman MA, Wrangham RW (1996) Diversity of medicinal plant use by chimpanzees in the
wild. In: Wrangham RW, McGrew WC, de Waal FBM, Heltne PG (eds) Chimpanzee cultures.
Harvard University Press, Cambridge, MA, pp 129148
43. Huggett DB, Brooks BW, Peterson B, Foran CM, Schlenk D (2002) Toxicity of select beta
adrenergic receptor-blocking pharmaceuticals (b-blockers) on aquatic organisms. Arch Environ
Contam Toxicol 43:229235
44. Jobling S, Nolan M, Tyler CR, Brighty G, Sumpter JP (1998) Widespread sexual disruption in
wild fish. Environ Sci Technol 32:24982506
45. Jones OA, Lester JN, Voulvoulis N (2005) Pharmaceuticals: a threat to drinking water? Trends
Biotechnol 23(4):163167
46. Kalbfus W, Kopf W (1998) Erste Ansatze zur kologischen Bewertung von Pharmaka in
Oberflchengewssern. Mnchner Beitrge zur Abwasser Fischerei und Flussbiologie 51:
628652
47. Kim SD, Cho J, Kim IS, Vanderford BJ, Snyder SA (2007) Occurrence and removal of pharmaceuticals and endocrine disruptors in South Korean surface, drinking, and waste waters.
Water Res 41:10131021
48. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton HT (2002)
Pharmaceuticals, hormones and other organic wastewater contaminants in US streams, 1999
2000: a national reconnaissance. Environ Sci Technol 36:12021211
49. Kmmerer K (ed) (2008) Pharmaceuticals in the environment: sources, fate, effects and risks,
3rd edn. Springer, Berlin, Heidelberg, p 521
50. Krang AS, Dahlstrom M (2006) Effects of a candidate antifoulant compound (medetomidine)
on pheremone induced mate search in the amphipod Corophium volutator. Mar Pollut Bull
52(12):17761783
51. Larsson DGJ, Fick J (2009) Transparency throughout the production chaina way to reduce
pollution from the manufacturing of pharmaceuticals? Regul Toxicol Pharmacol 53:161163
52. Larsson DGJ, de Pedro C, Paxus N (2007) Effluent from drug manufacturers contains
extremely high levels of pharmaceuticals. J Haz Mater 148:751755
53. Mattson B, Nsman I, Strm J (2007) Swedens voluntary environmental classification
system. Raj Pharma 153158. http://www.lif.se/cs/Medlemsservice/Sidinnehall/Nyheter/

46

54.
55.
56.

57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.

70.
71.

J.O. Straub and T.H. Hutchinson

Dokument/RAJ%202007%20Mar%20-%20Sweden%20Environment%20FINAL.pdf .
Accessed July 2010
Metcalfe CD, Koenig BG, Bennie DT, Servos M, Ternes TA, Hirsch R (2003) Occurrence of
neutral and acidic drugs in the effluents of Canadian sewage treatment plants. Environ Toxicol
Chem 22(12):28722880
Montgomery J (2010) Delaware drinking water at risk: prescription drugs on tap from major
suppliers. News J. Delaware Online, http://www.delawareonline.com/article/20100804/
NEWS02/8040343. Accessed 4 Aug 2010.
MPA (2009) Lkemedelsverket/Medical Products Agency. Redovisning av regeringsuppdrag
gllande mjligheten att skrpa miljkrav vid tillverkning av lkemedel och aktiv substans.
Report from the Swedish MPA (with an English abstract), 16th December 2009. http://www.
lakemedelsverket.se/upload/nyheter/2009/2009-12-16_rapport_milj%C3%B6kravl%C3%A4kemedel.pdf. Accessed July 2010
Nakada N, Tanishima T, Shinohara H, Kiri K, Takada H (2006) Pharmaceutical chemicals and
endocrine disrupters in municipal wastewater in Tokyo and their removal during activated
sludge treatment. Water Res 40:32973303
NSNR/C&P (2005) Canadian New Substances Notification Regulation (Chemicals and
Polymers). SOR/2005247. Current version available at http://laws.justice.gc.ca/PDF/
Regulation/S/SOR-2005-247.pdf. Accessed July 2010
NSNR/O (2005) Canadian New Substances Notification Regulation (Organisms). SOR/2005
248. Current version available at http://laws.justice.gc.ca/PDF/Regulation/S/SOR-2005-248.
pdf. Accessed July 2010
OBrien E, Dietrich DR (2004) Hindsight rather than foresight: reality versus the EU draft
guideline on pharmaceuticals in the environment. Trends Biotechnol 22(7):326330
Peterson SM, Batley GE, Scammell MS (1993) Tetracycline in antifouling paints. Mar Pollut
Bull 26(2):96100
Phillips PJ, Smith SG, Kolpin DW, Zaugg SD, Buxton HAT, Furlong ET, Esposito K, Stinson
B (2010) Pharmaceutical formulation facilities as sources of opioids and other pharmaceuticals to wastewater treatment plant effluent. Environ Sci Technol 44:49104916
Rabiet M, Togola A, Brissaud F, Seidel J-L, Budzinski H, Elbaz-Poulichet F (2006) Consequences
of treated water recycling as regards pharmaceuticals and drugs in surface and ground waters of
a medium-size Mediterranean catchment. Environ Sci Technol 40:52825288
Richardson ML, Bowron JM (1985) The fate of pharmaceutical chemicals in the aquatic environment. J Pharm Pharmacol 37:112
Rmbke J, Knacker T, Stahlschmidt-Allner P (1996) Umweltprobleme durch Arzneimittel
Literaturstudie. UBA-Forschungsbericht 96060. Umweltbundesamt, Berlin (Germany)
Rudn C, Hansson SO (2010) Registration, evaluation and authorization of chemicals
(REACH) is but the first stephow far will it take us? Six further steps to improve the European
chemicals legislation. Environ Health Perspect 118(1):610
Schwab BW, Hayes EP, Fiori JM, Mastrocco FJ, Roden NM, Cragin D, Meyerhoff RD, DAco
VJ, Anderson PD (2005) Human pharmaceuticals in US surface waters: a human health risk
assessment. Regul Toxicol Pharmacol 42:296312
Straub JO (2002) Environmental risk assessment for new human pharmaceuticals in the
European Union according to the draft guideline/discussion paper of January 2001. Toxicol
Lett 135:231237
Straub JO (2005) Environmental risk assessment of new human pharmaceuticals in the European
Union according to the 2003 draft guideline. In: Dietrich DR, Webb SF, Petry T (eds) (2005) Hot
spot pollutants; pharmaceuticals in the environment. Elsevier Academic Press, Burlington, MA,
pp 299317
Straub JO (2008) Deterministic and probabilistic environmental risk assessment for diazepam.
In: Kmmerer K (ed) Pharmaceuticals in the environment; sources, fate, effects and risks, 3rd
edn. Springer, Heidelberg, pp 343383
Straub JO, Stewart KM (2007) Deterministic and probabilistic acute-based environmental risk
assessment for naproxen for western Europe. Environ Toxicol Chem 26(4):795806

Current State of Regulations for Human Pharmaceuticals ERA

47

72. Stumpf M, Ternes TA, Wilken RD, Rodrigues SV, Baumann W (1999) Polar drug residues in
sewage and natural waters in the State of Rio de Janeiro, Brazil. Sci Total Environ 225:135141
73. Ternes TA, Stumpf M, Schuppert B, Haberer K (1998) Simultaneous determination of antiseptics
and acidic drugs in sewage and river water. Vom Wasser 90:295309
74. TGA (2008) Australian Government, Department of Health and Aging, Therapeutic Goods
Administration. Module 1, Administrative Information and Prescribing Information For
Australia; Notice to Applicants; CTD-Module 1, TGA Edition November 2008. http://www.
tga.gov.au/docs/pdf/euguide/tgamod1.pdf. Accessed August 2010
75. TGD (2003) Technical Guidance Document on Risk Assessment in support of Commission
Directive 93/67/EEC on Risk Assessment for new notified substances, Commission Regulation
(EC) No 1488/94 on Risk Assessment for existing substances, Directive 98/8/EC of the
European Parliament and of the Council concerning the placing of biocidal products on the
market; Part II, Environmental Risk Assessment. European Commission; 2003. http://ecb.jrc.
ec.europa.eu/tgd/. Accessed 12 Jan 2010
76. Thomas KV, Hilton MJ (2004) The occurrence of selected human pharmaceutical compounds
in UK estuaries. Mar Pollut Bull 49:436444
77. Tixier C, Singer HP, Oeller S, Mller SR (2003) Occurrence and fate of carbamazepine,
clofibric acid, diclofenac, ibuprofen, ketoprofen and naproxen in surface waters. Environ Sci
Technol 37(6):10611068
78. van der Heide EF, Hueck-van der Plas EH (1984) Geneesmiddelen en milieu. Pharm Weekbl
119:946947
79. VICH (2000) Environmental impact assessment (EIA) for veterinary medicinal products
(VMPs)phase I. VICH GL 6 (Ecotoxicity Phase I). http://www.vichsec.org/pdf/2000/Gl06_
st7.pdf. Accessed July 2010
80. VICH (2004) Environmental impact assessment for veterinary medicinal productsphase II
guidance. VICH GL 38 (Ecotoxicity Phase II). http://www.vichsec.org/pdf/10_2004/GL38_
st7.pdf. Accessed July 2010
81. Vieno NM, Tuhkanen T, Kronberg L (2005) Seasonal variation in the occurrence ogf pharmaceuticals in effluents from a sewage treatment plant and in the recipient river. Environ Sci
Technol 39:82208226
82. Watts C, Maycock D, Crane M, Fawell J, Goslan E (2007) Desk based review of current
knowledge on pharmaceuticals in drinking water and estimation of potential levels. Final
Report for the Drinking Water Inspectorate in DEFRA (UK Department of the Environment,
Food and Rural Affairs) by WCA, contract number CSA7184/WT02046/DWI 70/2/213. http://
randd.defra.gov.uk/Default.aspx?Menu=Menu&Module=More&Location=None&ProjectID
=14717&FromSearch=Y&Publisher=1&SearchText=wt02046&SortString=ProjectCode&So
rtOrder=Asc&Paging=10. Accessed July 2010
83. Webb S, Ternes T, Gibert M, Olejniczak K (2003) Indirect human exposure to pharmaceuticals
via drinking water. Toxicol Lett 142:157167
84. Zuccato E, Calamari D, Natangela M, Fanelli R (2000) Presence of therapeutic drugs in the
environment. Lancet 355:17891790

Regulation of Pharmaceuticals
in the Environment: The USA
Emily A. McVey

Introduction
The fate of pharmaceuticals in the environment has been studied for more than
50 years, with the presence and potential effects acknowledged shortly thereafter
[15]. It has gradually become apparent that risk assessments developed for the
usual chemical contaminants cannot be applied carte blanche to pharmaceuticals,
because they are developed to be highly active and specific in biological systems at
low levels [69]. Therefore, when applying risk assessment models to the environmental assessment of pharmaceuticals, regulators must take into account not only
the complexity of the entity to be protected (the ecosystem at large) but also the
complexity of the regulated article (pharmaceuticals).
Regulation and policy are, at best, a merging of science, politics, social science,
and stakeholder input. Environmental regulation is further complicated by the complex nature of the entity to be regulated. Environmental regulations and regulators
have larger, long-term goals of protection of human health and ecosystems (large
and small) from damage as a result of environmental exposures to contaminants.
What this means in practice and in theory may depend on the interpretation of existing policies at any given time, but the most overarching goal for environmental
regulators is promotion and protection of harmony (sustainability) within ecosystems. This is no small task, particularly when you consider how to define an ecosystem. The traditional view of the ecosystem is the interacting organisms and
biophysical components in a particular place, focusing on relationships and processes of the living and nonliving components [10]. This definition could be seen as
E.A. McVey (*)
Office of Pharmaceutical Science, Center for Drug Evaluation and Research,
U.S. Food and Drug Administration, 10903 New Hampshire Avenue, Silver Spring,
MD 20993, USA
WIL Research, P.O. Box 3476, 5203DL s-Hertogenbosch, The Netherlands
e-mail: [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_3, Springer Science+Business Media, LLC 2012

49

50

E.A. McVey

relatively distinct from typical public health goals; on its surface, however, ecosystem
protection directly intersects and impacts with the protection of health, as defined
by the World Health Organization (WHO) as the state of complete physical, mental
and social well-being (rather than simply an absence of disease) [10, 11]. From this
definition, a sustainable ecosystem encompasses all impacts on human health,
including social and economic, as well as the health of the natural world, as was
recently recognized in the 2003 Millennium Ecosystem Assessment [12]. Indeed,
where ecosystems are concerned, sustainability means maintaining a wide variety
of complex (and not fully understood) systems that support health and life [13].
Currently, the European Medicines Agency (EMA), United States Food and Drug
Administration (FDA), and Health Canada oversee the assessment of environmental
risk from pharmaceuticals at the time of registration in the European Union (EU),
USA and Canada, respectively. For each of these, the assessment of hazard and
exposure is used to come to a conclusion regarding the risk of a particular substance.
A common problem with environmental regulation to protect ecosystem sustainability
is the development of risk characterization (hazard assessment) testing schemes,
since the majority of environmental exposures occur chronically, over long periods
of time and (particularly in the case of pharmaceuticals) possibly at very low levels
[7, 1417]. In addition, exposure is simultaneously to a wide variety of entities, and
a multitude of different vectors may be exposed in a multitude of ways [1821].
Extrapolation from acute toxicity testing has been the most typical method utilized,
as many acute toxicity tests are well established and validated using particular organisms as representative examples, they require less time and money, and a much larger
amount of data for acute tests is already available [7, 22]. The use of acute toxicity
studies for environmental risk assessment in general has been criticized, because of
their focus on immediate endpoints such as lethality, which may not be appropriate
when trying to assess risk, especially for highly potent and biologically specific contaminants such as pharmaceuticals [2325]. Indeed, some studies have found high
acute to chronic ratios for certain pharmaceuticals (mostly for estrogenic or hormonally active compounds) [2628]. Recently, new testing schemes have been proposed
and devised, and work is ongoing on developing and validating chronic toxicity tests
or testing schemes to assess long-term risk of low-level environmental contaminants.
However, whatever testing scheme is utilized, the outcome of the risk assessment is
what really matters. What to do if an environmental impact is expected? In the case
of human pharmaceuticals the risk/benefit analysis highly favors human health over
any potential ecotoxicological effects, making it doubtful that registration, approval,
or use of a drug would be limited based on ecological concerns.

Regulation in the USA

The regulation of pharmaceuticals in the environment falls under the purview of
both the US Environmental Protection Agency (EPA), which has authority over
most environmental media through the Clean Water Act (CWA), Clean Air Act
(CAA), Toxic Substances Control Act (TSCA), and Resource Conservation and

3 Regulation of Pharmaceuticals in the Environment: The USA

51

Recovery Act (RCRA), among others, and the FDA. FDA has authority over
pharmaceuticals through the Federal Food, Drug, and Cosmetics Act (FFDCA) and
has a regulatory responsibility to investigate the environmental impact of pharmaceuticals through one of the oldest environmental statutes, the National Environmental
Policy Act of 1969 (NEPA).
NEPA requires any US federal governmental entity to assess the environmental
impact of its actions. It is overseen by the White Houses Council on Environmental
Quality (CEQ) and the US EPA. It is important to note that NEPA itself does not
give any federal agency extra authoritywhatever the outcome of a NEPA process,
the federal agency does not have any additional regulatory authorities. For this reason, it may be said that NEPA is a procedural statute, an interpretation which has
been established by case law in the US Supreme Court (Vermont Yankee Nuclear
Power Corp. v. Natural Resources Defense Council1, Kleppe v. Sierra Club2). By
this it is meant that NEPA places a statutory requirement upon federal agencies to
perform certain procedural tasks, but does not provide them with any authority
related to that task, other than those which may be applicable from their own authoritative statutes. When a federal entity is sued, it may be sued for not performing the
NEPA procedure correctly (under the Administrative Procedures Act (APA)). (It is
important to note that the Environmental Analysis under NEPA is directly tied to an
action by the government.) Although NEPA is only required for actions by federal
agencies, or actions that have significant federal involvement, in the USA there are
a number of State Environmental Policy Acts (SEPAs) that require similar assessments for State Governmental Agencies.

The NEPA Process

Under NEPA, before performing any action, a federal agency must assess the environmental impact of that action by preparing an Environmental Assessment (EA),
followed by a Finding of No Significant Impact (FONSI) or an Environmental
Impact Statement (EIS). An EA is a concise (i.e., 1012 pages) document that
assesses the potential environmental impact of an action. From the EA, regulators
then determine whether a more in-depth EIS is required or whether there can be a
FONSI. If it is determined that an EIS is needed, a longer process begins to create a
(in most cases) massive document which takes into account all possible environmental impacts of an action, including cultural and social impacts and those relating
to environmental justice. The EIS process can be summed up by Fig. 1 and often
takes years to complete (not to mention significant funds). An EIS is required for an
action which is expected to have a significant impact or that is expected to impact
a significant resource (such as an endangered species or the habitat of an endangered
species). In practice, many governmental agencies produce more comprehensive

1
2

435 US 519, 558 (1978).

427 US 390 (1976).

52

E.A. McVey

Fig. 1 EIS process flowchart

Determine Lead Agency

Prepare Environmental Assessment (optional)
Publish Notice of Intent
Conduct Scoping Process
Prepare Draft EIS
Circulate Draft EIS for Review
File Draft EIS with EPA
Hold public hearing if required or desired
Prepare Final EIS
Circulate Final EIS
File with EPA
Adopt Final EIS
Make Agency Decision
Prepare Record of Decision

EAs to assess potential environmental impacts fully before determining if an EIS is

required. For this reason, most EAs nowadays are not particularly concise, though
they are still dwarfed by the size of EISs.
Of course, if every federal entity had to perform even an EA for every action
(payroll approvals, carpet changes, etc.) the entire federal government would grind
to a halt. To circumvent this potential problem, NEPA and the CEQ provide a way
for the respective agencies to exclude certain very common actions from having to
perform NEPA analyses, through the use of agency-promulgated Categorical
Exclusions (CEs or Cat Exes).
Agencies write implementing regulations which express their understanding of a
particular statute, and how they plan to actually implement or enforce that statute.
Therefore, while regulations are not law (but rather an Agencys interpretation of
law), they are what govern an Agencys everyday actions. In the case of NEPA, each
Agencys implementing regulations, including Categorical Exclusions they intend
to use, are vetted by the CEQ and the US EPA and subject to public commentary,
before they become common practice.

History of NEPA and the US FDA

To put the environmental regulation of pharmaceuticals by the FDA under NEPA
into context, it is worthwhile to summarize the history of NEPA implementation by
the FDA, since the advent of NEPA in 1969. Briefly, 3 years after NEPA passed, the

3 Regulation of Pharmaceuticals in the Environment: The USA

53

FDA performed an EIS on the use of plastic bottles for food and drugs. At this time,
federal agencies were still trying to understand what NEPA meant for their actions,
particularly when they did not have statutory authority to address environmental
issues. Following the plastic bottles EIS, the FDA promulgated a regulation which
(very basically) said that FDAs legal interpretation of NEPA was that an adverse
EIS does not permit the FDA to act if the adverse impact identified does not involve
a threat to public health, adulteration, or misbranding or some other factor already
identified by the FFDCA, and therefore the FDA did not consider any of its actions
to be under the purview of NEPA. The interpretation was challenged (EDF, Inc. v.
Mathews3) and from this case it was established that drug approvals and withdrawals are considered agency actions and are therefore subject to NEPA. The FDA
therefore created implementing regulations to express how it intended to meet the
NEPA burden to perform environmental analyses of Agency actions. In 1985, the
FDA issued implementing regulations that detailed when the Agency would prepare
or require an EA, when it would prepare an EIS, and what actions were categorically excluded from NEPA analysis [29].
In 1995, the Presidents National Performance Review issued a report
Reinventing Regulation of Drugs and Medical Devices, under the reinventing
government (REGO) initiatives. One of the initiatives in the report was a proposal
to reduce the number of environmental assessments (EAs) required to be submitted
and, consequently, the number of reviews performed by the Agency. As a result, the
FDA proposed to reduce the number of EAs by creating additional categorical
exclusions from the EA requirements. The final rule for these new exclusions was
published on July 29, 1997 and became effective August 28, 1997. The regulations
that were promulgated at this time are the ones that are currently followed within the
Agency and can be found in Title 21 of the Code of Federal Regulations, Part 25 (21
CFR 25).

Current Practice in Environmental Assessment at the US FDA

Since NEPA and CEQ allow an agency to contract the preparation of an EA, the US
FDA requires sponsors and applicants to submit EAs as a part of the application
package, similar to the requirement to submit safety and efficacy data. These EAs
are then reviewed as part of the approval process for the pharmaceutical in
question.
The document Guidance for Industry: Environmental Assessment of Human
Drug and Biologics Applications (finalized in July 1998) summarizes the current
practice for Environmental Assessment at the Center for Drug Evaluation and
Research (CDER) and the Center for Biologics Evaluation and Research (CBER) at
the US FDA [30]. A Guidance document describes the FDAs current thinking on a
particular issue, to inform the regulated industry regarding the information the

410F. Supp. 336 (D.D.C. 1976).

54

E.A. McVey

Agency will be expecting and how they will be reviewing and interpreting it. As
regulations are interpretations of statutes, so Guidances are interpretations of
regulations.
Under the current regulations, an EA is required and received when the expected
introductory concentration (EIC) into the environment will be greater than or equal
to 1 mg/L (ppb), and/or when a noncultivated plant or animal is used in the production of the drug, and/or when the drug is expected to adversely impact the environment of any endangered species. An EA can also be required under the extraordinary
circumstances provision. This provision allows the Agency to override its own
categorical exclusions at times when it considers the weight of evidence to suggest
that a significant affect on the environment might be expected (21 CFR 25.21).
All applications to the FDA must have either an EA or a claim of categorical exclusion. If they do not, this is considered grounds for refusing to file or approve the
application. An EA that is adequate for filing addresses relevant environmental
issues with sufficient information to allow the FDA to determine whether the proposed action may affect the quality of the human environment [30].
In practice, both Investigational New Drug (IND) applications and Abbreviated
New Drug Applications (ANDAs) typically claim and are granted exclusions: INDs
because their EIC is under 1 ppb, and ANDAs because there is expected to be no
increased use over the amount which was assessed in the EA for the originators
NDA for that product (if that amount was over 1 ppb to start with). These two are
the major categorical exclusions that the majority of applications fall under.
Increased use of an active moiety occurs if the drug will be administered at a
higher dosage, for a longer duration, or for a different indication than previously, or
if the drug is a new molecular entity [30]. The Guidance provides lists of examples
of actions that would or would not be considered to be increased use and suggests
that if a sponsor has a question they can contact the Agency to find out whether they
should provide an EA.
The equation used to calculate the concentration of a substance at the point of
entry into the aquatic environment to determine qualification for the under 1 ppb
categorical exclusion is calculated by multiplying the kg/year of the active moiety
produced for direct use by 1/L water per day entering publicly owned treatment
works. Conversion factors are included to convert to mg/L and from day to year. The
number of L/day entering publicly owned treatment works (POTWs) is published in
the Needs Survey, Report to Congress and can be found at http://www.epa.gov/
owm. It is updated periodically and is at 1.321 1011 L/day currently.
The calculation assumes that all drug products produced in a year are used and
enter the POTWs, that the drug product usage occurs throughout the USA in proportion to the population and the amount of waste generated, and that there is no
metabolism. The estimate of the kg/year active moiety is based on the highest
quantity of the active moiety expected to be produced for direct use in any of the
next 5 years, excluding any quantity produced for inventory buildup or nonuse purposes. It includes the quantity of active moiety used in all dosage forms and strengths
included in the application and the quantity used in an applicants related applications. This calculation is an out of pipe calculation, meaning that the calculation

3 Regulation of Pharmaceuticals in the Environment: The USA

55

is at the point of entry into the environment, and no dilution factor is added for
dilution in surface or groundwater. If the applicants wish to factor in metabolism,
depletion or dilution, they must document these in detail (see below) [30].
Once it has been established that the EA should be submitted, the Guidance
provides specifics of what should be included: The first step is identification of the
substance of interest, its chemical and physical characterization, and a discussion
of the environmental fate of the substance. This section includes potential environmental depletion mechanisms. If the depletion mechanism is going to be used to
reduce the expected introduction concentration (EIC) or eliminate effects testing,4
a detailed analysis of the depletion mechanism is provided, otherwise a summary
is acceptable.
If a moiety is expected to significantly partition into biosolids (Koc 1,000), a
terrestrial EIC is calculated and terrestrial fate and effects testing undertaken. In the
same manner, if a moiety is expected to have significant introduction into the atmospheric environment, this is considered and discussed in the EA. If no rapid and
complete depletion mechanism is identified it is assumed that the substance will
persist in the environment and toxicity testing will be required. The toxicity tests
required for the EA follow a tiered system laid out in the Guidance document and
summarized in Fig. 2.
Briefly, acute ecotoxicity testing (Tier 1) is performed on a minimum of one suitable test organism, and if the effect concentration for 50% (EC50) or lethal concentration for 50% (LC50) divided by the minimum expected effect concentration
(MEEC) is greater than or equal to 1,000, no further testing is completed unless
sublethal effects are observed at MEEC.5 If the EC50 or LC50 divided by the MEEC
is less than 1,000, Tier 2 testing is performed [30].
Tier 2 is acute ecotoxicity testing on the minimum base set of aquatic and/or terrestrial organisms (depending on where the active moiety or active metabolites are
expected to accumulate), typically an acute fish toxicity test, an aquatic invertebrate
acute toxicity test, and an algal species bioassay constitute the aquatic base testing;
plant early growth tests, earthworm toxicity tests, and soil microbial toxicity tests
constitute the terrestrial base tests. If the EC50 or LC50 for the most sensitive organism in the base set divided by the MEEC is greater than or equal to 100, no further
testing is conducted unless sublethal effects are observed at the MEEC. If EC50 or
LC50 divided by the MEEC is less than 100, Tier 3 testing should be performed.
Chronic toxicity testing (Tier 3) is considered if the compound has the potential to
bioaccumulate or bioconcentrate, if indicated based on Tier 1 and/or 2 testing or
if there are indications that the compound biotransforms into more toxic compounds.
If the logarithm of the octanolwater partition coefficient (log Kow) is greater than or

If a rapid (defined in the Guidance document) and complete depletion mechanism is identified
(with simple, polar by-products), no testing to determine environmental effects is necessary, except
a microbial inhibition test (or other appropriate test to determine potential for effects on waste
treatment processes).
5
Sublethal effects at the MEEC indicate that chronic testing (Tier 3) should be performed.
4

56

E.A. McVey

Fig. 2 CDER/CBER tiered approach to fate and effects testing. FDA CDER/CBER [30]

equal to 3.5 under relevant environmental conditions (e.g., neutral pH), chronic
testing is required. If the EC50 or LC50 divided by the MEEC from the Tier 3 testing
is greater than or equal to 10, no further testing is conducted unless sublethal effects
are observed at the MEEC. The applicant consults with the Agency on how to proceed if the result is less than 10 or sublethal effects are observed at the Tier 3
MEEC.
While the information above summarizes the typical process for the submission
and review of Environmental Assessments (EA)s for pharmaceutical applications
at CDER and CBER, other Centers within the FDA have their own specific
regulations for NEPA compliance. In particular, the Center for Food Safety and
Nutrition (CFSAN) and the Center for Veterinary Medicine (CVM) have separate

3 Regulation of Pharmaceuticals in the Environment: The USA

57

regulations and Guidances specific to the substances they are likely to encounter
and the manner in which those substances may be introduced into the environment.
The CVM participated in the Veterinary International Conference on Harmonization
(VICH) for environmental assessment of animal pharmaceuticals to harmonize
environmental assessment in animal medicines with requirements in the EU and
Japan.

Regulation of Pharmaceuticals in the Environment

in the European Union
In 2006, the European Medicines Agency/Committee for Medicinal Products for
Human Use (EMA/CHMP) released a final Guideline on the environmental risk
assessment of medicinal products for human use. This guideline, like the FDA
Guidance from a decade earlier, outlines when an environmental risk assessment
(ERA) is required before marketing approval for a human drug, and if an assessment is required, what data should be submitted and reviewed.
An exposure assessment is the first step to determine if a certain action limit
has been reached and an environmental assessment should be submitted. The action
limit for the surface water predicted environmental concentration (PECsurface water) for
human pharmaceuticals in water was arrived at by dividing the 1 ppb threshold from
the US FDA implementing regulations by 100, to take the precautionary principle
into account and is therefore set at 10 ng/L [25].
The PECsurface water is calculated using the daily dose of the pharmaceutical in question, default values for wastewater production per capita, and estimated sales of the
pharmaceutical product in question (market penetration factor), assuming no metabolism and no biodegradation or retention in the sewage treatment plant [31, 32]. In
addition, the EMA specifies that highly lipophilic or potential endocrine-disrupting
substances that may affect organisms below the threshold value must be evaluated
via a risk assessment strategy [31]. A persistence, bioaccumulation, toxicity (PBT)
assessment is required if log Kow is 4.5 [31].
If any of the above situations exists, Phase II of the risk assessment is begun.
Phase II is divided into Tier A and Tier B testing. In Tier A, a quantitative risk
assessment is conducted for surface water, groundwater, and microorganisms in
water (the predicted environmental concentration (PEC) value is compared with the
respective predicted no effect concentrations (PNECs)). If one or more of the risk
quotients suggest a potential for harm to the environment, Tier B studies are required.
Tier B is known as the extended environmental fate and effects analysis, as the PEC
can be refined according to specific sorption effects in sewage treatment plants
(STPs), and water, sediment, microorganism, and terrestrial effects testing is performed. If there appears to be a potential for harm to the environment upon completion of the review of the ERA data, labeling, including proper disposal and
highlighting of environmental risks can be employed [31, 32]. A potential risk to the
environment is not grounds to block marketing approval.

58

E.A. McVey

Regulation and Environmental Risk Assessment

Beyond challenges of available data and appropriate testing for hazard identification,
environmental risk assessment for regulatory purposes entails a number of other
difficulties. For example, each regulatory body may define ecosystem protection
differently, creating wide variety in regulatory goals. Therefore, harmonization of
environmental protection goals and definitions, standards, and risk assessment models is essential for efficient and appropriate environmental assessment, whether on a
local, national, or international level [33]. In systems with large variability (such as
environmental and biological systems) it is important to identify and distinguish
between variability and uncertainty in data used for the assessment. It is possible to
reduce the level of uncertainty of data, by improvement of testing and collection of
greater amounts of data, so this is often the focus of improved risk assessment, but
it will always be necessary to account for variability within the risk assessment.
Finally, a risk assessment must be science-based, with validated testing schemes
and data, but must also be flexible enough to allow for a variety of new data and new
models to be accommodated and considered, particularly with regard to the environmental compartments to be tested and addressed.
The FDAs approach for the 1998 CDER/CBER Guidance on EA was to achieve
the goal of appropriate environmental assessment by requiring the information that
was deemed necessary to identify potential hazards and from that perform an appropriate, science-based risk assessment. At the time of its creation, the Guidance was
unique, and it continues to allow the data collection and flexibility that it was
designed to achieve. The FDA continues to monitor regulations and developments
elsewhere in the world and review current literature with eye toward ensuring that
the EAs that the Agency receives and reviews are appropriate information for modern environmental risk assessment.
Disclaimer The views expressed in this chapter are those of the authors and should not be interpreted as the official opinion or policy of the US Food & Drug Administration, Department of
Health and Human Services, or any other agency or component of the US government.
Acknowledgements The author would like to thank Dr. Nakissa Sadrieh for her thoughtful and
extremely helpful comments and edits during the development of this manuscript.

Appendix: Table of abbreviations

FDA
EPA
EMA
WHO
CWA
CAA

United States Food and Drug Administration

United States Environmental Protection Agency
European Medicines Agency
World Health Organization
Clean Water Act
Clean Air Act

3 Regulation of Pharmaceuticals in the Environment: The USA

TSCA
RCRA
FFDCA
NEPA
CEQ
APA
SEPA
FONSI
EIS
EA
CFR
CDER
CBER
IND
ANDA
POTW
EIC
EC50
LC50
MEEC
PNEC
CVM
CFSAN
VICH
EMEA/CHMP
ERA
PEC
PBT
STP

59

Toxic Substances Control Act

Resource Recover and Conservation Act
Federal Food, Drug, and Cosmetics Act
National Environmental Policy Act
Council on Environmental Quality
Administrative Procedures Act
State Environmental Policy Act
Finding of No Significant Impact
Environmental Impact Statement
Environmental Assessment
Code of Federal Regulations
Center for Drug Evaluation and Research
Center for Biologics Evaluation and Research
Investigational New Drug
Abbreviated New Drug Application
Publicly Owned Treatment Works
Expected Introductory Concentration
Effect Concentration for 50% of population
Lethal Concentration for 50% of population
Minimum Expected Effect Concentration
Predicted No Effect Concentration
Center for Veterinary Medicine
Center for Food Safety and Nutrition
Veterinary International Conference on Harmonization
European Medicines Agency/Committee for Medicinal Products for Human Use
Environmental Risk Assessment
Predicted Environmental Concentration
Persistence, Bioaccumulation, Toxicity
Sewage Treatment Plant

References
1. Grimes DJ, Singleton FL, Colwell RR (1984) Allogenic succession of marine bacterial communities in response to pharmaceutical waste. J Appl Bacteriol 57(2):247261
2. Lee WY, Arnold CR (1983) Chronic toxicity of ocean-dumped pharmaceutical wastes to the
marine amphipod Amphithoe-valida. Mar Pollut Bull 14(4):150153
3. Peele ER et al (1981) Effects of pharmaceutical wastes on microbial-populations in surface
waters at the Puerto-Rico dump site in the Atlantic-Ocean. Appl Environ Microbiol
41(4):873879
4. Soulides DA, Pinck LA, Allison FE (1962) Antibiotics in soils: V. Stability and release of soiladsorbed antibiotics. Soil Sci 94(4):239244
5. Voets JP et al (1976) Degradation of microbicides under different environmental-conditions.
J Appl Bacteriol 40(1):6772
6. Daughton CG, Ternes TA (1999) Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspect 107(suppl 6):907938
7. Halling-Sorensen B et al (1998) Occurrence, fate and effects of pharmaceutical substances in
the environmenta review. Chemosphere 36(2):357394

60

E.A. McVey

8. Stan HJ, Heberer T (1997) Pharmaceuticals in the aquatic environment. Analusis

25(7):M20M23
9. Sybesma W (1992) Do veterinary pharmaceuticals affect the environment. Tijdschrift Voor
Diergeneeskunde 117(9):274275
10. Marten GG (2001) Human ecology: basic concepts for sustainable development. Earthscan,
London
11. WHO (1946) Preamble to the Constitution of the World Health Organization. In International
Health Conference. Official Records of the World Health Organization, New York
12. Alcamo J, Bennet EM et al (2003) Ecosystems and human well-being: a framework for assessment. Millennium Ecosystem Assessment. Island Press, Washington DC
13. McMichael AJ (2006) Population health as the bottom line of sustainability: a contemporary
challenge for public health researchers. Eur J Public Health 16(6):579581
14. Daughton CG (2000) Pharmaceuticals in the environment: overarching issues and concerns.
Abs Papers Am Chem Soc 219:U622
15. Daughton CG (2001) Emerging pollutants, and communicating the science of environmental
chemistry and mass spectrometry: pharmaceuticals in the environment. J Am Soc Mass
Spectrom 12(10):10671076
16. Kolpin DW et al (2002) Pharmaceuticals, hormones, and other organic wastewater contaminants in US streams, 19992000: a national reconnaissance. Environ Sci Technol
36(6):12021211
17. Ternes TA, Joss A, Siegrist H (2004) Scrutinizing pharmaceuticals and personal care products
in wastewater treatment. Environ Sci Technol 38(20):392a399a
18. Daughton CG (2004) Non-regulated water contaminants: emerging research. Environ Impact
Assess Rev 24(78):711732
19. Lu GH et al (2010) Active biomonitoring of complex pollution in Taihu Lake with Carassius
auratus. Chemosphere 79(5):588594
20. Pomati F et al (2006) Effects of a complex mixture of therapeutic drugs at environmental levels
on human embryonic cells. Environ Sci Technol 40(7):24422447
21. Inoue D et al (2010) Contamination with retinoic acid receptor agonists in two rivers in the
Kinki region of Japan. Water Res 44(8):24092418
22. Floate KD et al (2005) Fecal residues of veterinary parasiticides: nontarget effects in the pasture environment. Annu Rev Entomol 50:153179
23. Bound JP, Voulvoulis N (2004) Pharmaceuticals in the aquatic environmenta comparison of
risk assessment strategies. Chemosphere 56(11):11431155
24. Montforts M (2005) The trigger values in the environmental risk assessment of pharmaceuticals in the European Union: a critical appraisal. Bilthoven, RIVM
25. Schmitt H, Boucard T, Garric J, Jensen J, Parrott J, Pery A, Rombke J, Straub JO, Hutchinson
TH, Sanchez-Arguello P, Wennmalm A, Duis K (2010) Recommendations on the environmental risk assessment of pharmaceuticals: effect characterization. Integr Environ Assess Manag
6(S1):588602
26. Ferrari B et al (2003) Ecotoxicological impact of pharmaceuticals found in treated wastewaters: study of carbamazepine, clofibric acid, and diclofenac. Ecotoxicol Environ Saf
55(3):359370
27. Forbes VE, Calow P (2002) Extrapolation in ecological risk assessment: balancing pragmatism
and precaution in chemical controls legislation. Bioscience 52(3):249257
28. Webb SF (2001) A data-based perspective on the environmental risk assessment of human
pharmaceuticals. In: Kummerer K (ed) Pharmaceuticals in the environment. Springer, New
York, pp 203219
29. Vincent PG (1993) Environmental assessmentus requirements in new drug applications.
J Hazard Mater 35(2):211216
30. CDER/CBER F (1998) Guidance for industryenvironmental assessment of human drugs
and biologics applications, U.D.o.H.a.H. Services, Editor. United States Food and Drug
Administration, Center for Drug Evaluation and Research and Center for Biologics Evaluation
and Research

3 Regulation of Pharmaceuticals in the Environment: The USA

61

31. EMA/CHMP (2006) Guideline on the environmental risk assessment of medicinal products
for Human Use, E.M.A.C.f.M.P.f.H. Use, Editor. EMEA/CHMP/SWP/4447/00, London
32. Kuster A, Alder AC, Escher BI, Duis K, Fenner K, Garric J, Hutchinson TH, Lapen DR, Pery
A, Rombke J, Snape J, Ternes T, Topp E, Wehrhan A, Knacker T (2010) Environmental risk
assessment of human pharmaceuticals in the European Union: a case study with the betablocker atenolol. Integr Environ Assess Manag 6(S1):514524
33. Montforts M (2001) Regulatory and methodological aspects governing the risk assessment for
medicinal products; need for research. In: Kummerer K (ed) Pharmaceuticals in the environment. Sources, fate, effects and risks. Springer, Heidelberg, pp 439462

Environmental Fate of Human Pharmaceuticals

Alistair B.A. Boxall and Jon F. Ericson

Introduction
Over the past 10 years, a wealth of data has been generated on the inputs, occurrence, transport, and fate of human pharmaceuticals in the environment. In addition,
significant changes have occurred in the regulation of pharmaceuticals in the environment with guidelines being developed on which environmental fate studies need
to be performed and on how to interpret these in terms of environmental risk. In this
chapter, we attempt to draw upon the development in our scientific understanding of
the fate and transport of pharmaceuticals in order to assess the suitability of current
environmental risk assessment schemes for pharmaceuticals. We also attempt to
provide some guidance on those factors that should be considered when interpreting
existing regulatory fate studies as well as provide some ideas on alternative testing
or modeling strategies for assessing environmental fate and exposure. The ultimate
goal of the chapter is to contribute to the advancement of the understanding of environmental fate, and hopefully realign the context of regulatory guidance to relevant
testing conditions, appropriate endpoints, and interpretation of depletion mechanism, as we have seen with ecotoxicity testing with adoption of chronic testing, and
the use of relevant endpoints related to the mode of action of an API. The intent is
not to provide a comprehensive review. The focus of subsequent chapters is on post
consumer use rather than manufacturing operations.

A.B.A. Boxall
Environment Department, University of York, Heslington, York YO10 5DD, UK
e-mail: [email protected]
J.F. Ericson (*)
Pfizer Global Research and Development, Worldwide PDM, Environmental Sciences,
MS: 8118A-2026, Eastern Point Road, Groton, CT 06340, USA
e-mail: [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_4, Springer Science+Business Media, LLC 2012

63

64

A.B.A. Boxall and J.F. Ericson

Physicochemical Characteristics
The physicalchemical properties of pharmaceuticals have been presented and discussed elsewhere in detail [1, 2] clearly demonstrating that they are large complex
molecules (MW 3001,000), typically with several functional/ionizable groups,
highly ionic in nature (cations, anions, zwitterions) with relative low solubility
(mg/L-mg/L range) with respect to other chemicals. From an organic chemistry
perspective, pharmaceuticals are multifunctional diverse compounds composed
individually or in combination of amines (primary, secondary, or tertiary), carboxylic acids, alcohols, polycyclic, aromatic/aliphatic, conjugated systems to only name
a few. Most are solids and are designed as salts to enhance aqueous solubility
(rather than volatility); and present in the environment either in a dissolved or
sorbed state rather than as micelles or other biphasic solutions. Their partitioning
into lipids, adsorption to environmental matrices, and bioavailability at trace levels
are perhaps some of the more important fate and transport characteristics that determine their ultimate fate. Generally, the more we know about something, the better
off we are to be able to predict and characterize its environmental behavior. But
quite often the outcome is that we are overwhelmed with information without
knowing what is pivotal, what properties we need to measure, and what is OK to
estimate or model. The following section provides a perspective of what are the key
chemical properties and why.

Nature and Extent of Ionization

The nature and extent of ionization is one of the more significant factors in determining a chemicals disposition in the environment and subsequent transport. It is
also important in determining the bioavailability of a compound and is relevant to
current discussions as to whether bound residues may be considered as depleted.
The following section provides a rationale around how ionization impacts ultimate
environmental disposition of pharmaceuticals when present at trace levels in the
environment.

Equilibrium Processes
Most pharmaceuticals are either an acid or base, present as a cation, anion, or a
combination of both with respective charges of +, , or a net charge of 0 [3]. The
extent of the ionization is dependent on the type of functional group(s) present,
and the pH of the surrounding environment. Very few pharmaceuticals are neutral
or hydrophobic compounds, perhaps less than 510% all together. Generally, the
more neutral a compound is, whether hydrophobic in nature or functionally

Environmental Fate of Human Pharmaceuticals

65

neutral by carrying a net charge of 0, the greater the extent of partitioning into
lipids found in the biomass (sludge) of wastewater treatment plants and/or in living
organisms, such as fish. Conversely, the greater extent of ionization observed the
greater extent of ionic complexation to minerals and clays typically found in suspended particulates, sediments, and soils as with cations; or greater dissolution
in surface waters as with anions. Examples of ionic mechanisms of binding
include ionic bonds, charge-transfer complexes, van der Waals forces, and H bonds
to name a few [4].

Nonequilibrium Processes
This may potentially infer everything else that is not included in the above, such as
what is entrapped in sediment and soil pores as part of the aging process. But of
particular interest is what is found as irreversibly bound to soils and sediments.
Perhaps not as significant for other chemicals, it is of interest for pharmaceuticals as
it is one of the predominant end pathways as residues become either potentially
depleted and/or inactivated as they become incorporated into the humic acid cycle.
Though still debated, there is considerable evidence from other chemical sectors to
suggest that such bound residues are not bioavailable [5, 6] and are essentially
removed. Examples of pesticides and the link of covalent binding through amine
functionalities [7] is of particular relevance for pharmaceuticals.and significance
as most pharmaceuticals are cations containing amine functional groups.
Physicalchemical properties of pharmaceuticals are well characterized early on
in development necessary to support active pharmaceutical ingredient (API) synthesis and formulation activities. Pharmaceutical APIs are usually: solids, available as
a salt form, mp > 90C with aqueous solubility in mg/L range. They have to be stable with an acceptable shelf life, even at elevated temperatures. When you think
about it, they are bound by the conditions that are required to make an acceptable
pharmaceutical product. Methods for determining physicalchemical properties are
well established. Perhaps questions remain as to what information is really needed
and what if any may be modeled. The following section is offered in the context of
solid APIs which represent most pharmaceutical APIs.

Solubility and Melting Point

Both parameters are typically well characterized as part of drug development.
Differential scanning calorimetry provides melting point endpoint and solubility is
often determined for both aqueous and organic systems. Solubility information is
typically more helpful in the handling and preparation of standard solutions used in
environmental studies, than being a key determinant in its environmental fate.

66

A.B.A. Boxall and J.F. Ericson

Vapor Pressure and Density

Vapor pressure of drug substances and its potential impact to the atmospheric compartment is not a primary concern for most APIs found in the final product, except
for rare cases where the API is a liquid or gas. Salt forms of APIs inherently raise
the MP and increase aqueous solubility [8], thereby limiting vapor pressure and the
ability to partition from water phase into the gas phase. Most vapor pressures are
below the criteria established by the FDA that triggers an assessment of the risks to
the atmosphere compartment (107 torr) as shown by several examples presented by
Elder [9, 10]. When one also considers that decomposition upon melting is a common observation from differential scanning calorimetry [9], APIs as an atmospheric
concern is not likely as a result of post consumer use. It is reasonable to assume
vapor pressure determinations may be more relevant to manufacturing operations
where drying operations at elevated temperatures and/or reduced pressure may
make sublimation control a process control requirement. Estimations of vapor pressure may be made by several approaches [11, 12], including the EPIWIN predictive
software, and these predictions may be used as a way of identifying circumstances
where an experimental determination should be made.
Density falls into the same category as vapor pressure and also is not a significant
parameter for consideration. Pharmaceuticals discharged as a result of post consumer
use are at trace levels in the environment that are either full dissolved or partitioned
onto solids and are not expected to be biphasic in nature or present as micelles. As part
of a post consumer assessment, density is not an important parameter to measure.

OctanolWater Partitioning
Octanolwater partition coefficient (Kow) has been used historically in the general
chemical industry to depict the distribution of a chemical between an oil and water
phase and is useful in assessing how it may partition from the water into biomass of
soils, sediments and sludge, or perhaps into the lipid of biota. This has been very useful
for predicting behaviour of chemicals that are neutral in charge and applicable to many
older pesticides, as well as many other general chemicals that dont ionize. Quite often
one sees this expressed as log Kow or log P and this has been a key parameter found in
many regulations, especially those pertaining to bioaccumulation (log P > 3.0). When
one looks at pharmaceuticals, there is a need for a different expression. We believe
that the log D is a more appropriate descriptor. Log D is defined as the distribution
coefficient which accounts for the partitioning for all of the ionic species present in
addition to the neutral form. When one considers how all of the ionic species for a
given compound partitions from water into octanol, biomass, or lipids, one will see
that log Kow will often overpredict partitioning, as it does not account for the ionic
species present and assume they all behave as the neutral form. The difference between
these two values is directly related to the extent of ionization as predicted by the pH

Environmental Fate of Human Pharmaceuticals

67

Table 1 Impact of ionization on O/W partitioning; log P vs. log D; modeled vs. measured values
log D (OECD 107)
Modeled log D
Measured log D
Compound
Ionic specie
Modeled log Pa
@ pH 7.0a
@ pH 7.0b
Exemestane
Amlodipine
Pregabalin

Neutral
Cation
Zwitterion

3.3
3.7
1.1

3.3
2.0
1.38

2.5
2.15c
1.35

ACD labs
OECD 107
c
pH 7.4
a

of the water and the pKa (s) of the compound. Table 1 compares the values of log P
and log D for selected neutral, cationic, and zwitterionic compounds noting that they
are essentially the same for neutral compound as the neutral species is the only species
present; and considerable different for those that are ionic in nature as they have one
or more ionic species present in addition to the neutral form. It is worth mentioning
that while many of the fate parameters are difficult to model via a QSAR approach,
log D is one parameter that has a fairly good history of modeling for pharmaceuticals
with a range of software available including ALOGPS, Pallas Prolog D or ACD Labs
software [13]. When determined experimentally, log D should be assessed at three
different pHs in the range of 4.010. One pH around 7.0 should be tested as environmentally relevant, and one above and one below 7.0 to fully assess the impact of pH
on extent of ionization. Actual values selected may depend on the pKa value(s)
(dissociation constant) of the compound to assure that all potential ionic species are
present when tested and preferably different from those at pH of 7.0.

Partitioning and Persistence in Environmental Systems

Partitioning Between Water and Air
The Henrys Law Constant (dimensionless) relates the concentration of a compound
in the gas phase to that in the liquid phase (H = Csg/Csl). It may also be expressed in
terms of vapor pressure and solubility as H = Pvp/S [12]. As noted in our earlier
discussion on vapor pressure, most pharmaceuticals are salts to enhance solubility
and bioavailability, thereby inherently diminishing its ability to partition into the gas
phase. As a result, with a few exceptions (e.g., some of the pharmaceuticals used as
local anesthetics) the Henrys Law Constants for pharmaceuticals are extremely low
so contamination of the atmospheric environment is typically not a concern for most
pharmaceuticals and will not be discussed further. It is not clear why vapor pressure
data is requested in some environmental assessment guidance given the overall contribution it has on its overall environmental disposition. While many fugacity and other
models require this data to run the model, quite often all that is needed is a default
value indicating that no volatilization is likely (Vp < 1 107 mmHg).

68

A.B.A. Boxall and J.F. Ericson

Partitioning Between Water and Sludge, Sediment or Soil

The water-solid distribution coefficient (Kd) is key for understanding the mobility of a
pharmaceutical through environmental systems and its availability for degradation
and something where experimental data is required. Kd relates the amount of a chemical sorbed to a solid compared to that dissolved in solution at equilibrium. It is fundamental to understanding the overall disposition of a chemical once discharged into the
environment. Pharmaceuticals display a wide range of sorption behavior and sorption
of the same compound in different soil or sediment types can vary significantly (e.g.,
[14]). While quite often we see this normalized to the organic content expressed as
Koc, there is need to proceed with precaution when using this data. As with log D,
ionization is a predominant factor that is also influential when it comes to mechanisms
for sorption. For neutral compounds, Koc may be used to normalize distribution to the
amount of organic matter, and for estimating distribution from one matrix to another,
such as from soil to sludge. In this specific case, partitioning into the organic matter is
mainly driven by one mechanism, the compounds hydrophobicity or conversely its
lipophilicity. But for all other pharmaceuticals that are ionizable, there is the potential
for many mechanisms of sorption acting at any one time, including association with
organic matter (OM), ion exchange, surface adsorption to mineral constituents, hydrogen bonding, and formation of complexes with ions such as Ca2+, Mg2+, Fe3+, or Al3+.
For these compounds, normalizing data to the organic content will not necessarily
compensate for the other mechanisms.
The sorption behavior is also influenced by the properties of the environmental
system being studied, including pH, organic carbon content, metal oxide content,
ionic strength, and cationic exchange capacity (e.g., [1517]). The complexity of
the sorbatesorbent interactions means that modeling approaches developed for
predicting the sorption of other groups of chemicals (e.g., pesticides and neutral
organics) are inappropriate for use on pharmaceuticals in sludge, sediment, and soil.
It is therefore dangerous to extrapolate sorption results across different matrices
(e.g., sludge to soil) or extrapolating across the same matrix (e.g., one soil to another
soil). The mismatch between sorption coefficients across matrices is demonstrated
in Table 2 which compares sludge and soil Koc data for three types of compounds.
While some researchers have proposed models for understanding the sorption
processes and estimating sorption behavior of pharmaceuticals in aquaticsolid systems, these are not yet in a state where they can be applied routinely in the environmental risk assessment process, as a consequence, sorption coefficient should be
measured for sludge and for soils/sediments respectively with each matrix characterized for their properties.
For soils, the presence of biosolids may also alter the sorption behavior of pharmaceuticals. Recent studies have demonstrated that the addition of these matrices
can either increase or reduce the sorption coefficients of pharmaceuticals in soils [18]
Table 3. The reasons for the sludge effect are still unclear but similar work with veterinary medicines that have explored the effect of the animal manures on sorption,
attribute the changes to changes in pH or the nature of dissolved organic carbon in
the soil/manure system (e.g., [19, 20]).

Environmental Fate of Human Pharmaceuticals

69

Table 2 Impact of ionization on mechanism of sorption for neutral and cationic compounds; comparison of sludge vs. soil measured values OECD 106
Compound
Ionic specie
Sludge Koc
Soil Koc
Exemestane
Azithromycin
Varenicline

Neutral
Cation
Cation

2,285
40
62

1,5946,533; n = 5
22,80059,600; n = 5
6,50015,000; n = 4

Table 3 Effect of the presence of sludge on sorption coefficients (Kd) and persistence (DT50) for selected pharmaceuticals in soils (Monteiro and Boxall [18, 54])
Soil only
Soil + sludge
Max

Min

Max

Min

Sorption (Kd)
Carbamazepine
Naproxen
Fluoxetine
Sulfamethazine

4.7
10.1
134
1.7

32.8
253
235
98.2

6.6
7.2
123
5.0

27.8
149
218
44.9

Degradation (DT50)
Naproxen

3.1

6.9

3.9

15.1

Persistence in Environmental Matrices

Since the first notable detection of pharmaceuticals in the environment, clofibric
acid in the North Sea in the late 90s [21], and subsequent studies determining the
occurrence of pharmaceuticals in the environment such as the work of Kolpin [22],
questions remain around the disposition of pharmaceuticals and their ultimate biodegradability in the environment. What is clear is how they primarily enter into the
environment from post consumer use with other minor sources from disposal and
manufacture. The extent they enter into the environment is mitigated first by their
metabolism in man and then secondarily by removal during wastewater treatment,
both from sorption and biodegradation. From that point on, residues either enter the
environment through land applied biosolids, or get dissolved in wastewater effluents.
Concentrations of pharmaceuticals are generally in the ng/l range in river waters,
variable based on overall volume of use, level of human metabolism, and the type
of treatment plant and/or operating efficiencies of such [2325].

Elimination Rates
Whether for the risk assessment or for classification purposes, elimination rates
(half-lives) are needed to better characterize the fate of pharmaceuticals as they are
transported through the environment and to better understand where persistence

70

A.B.A. Boxall and J.F. Ericson

may be an issue. Historically, the fate of pharmaceuticals has been characterized by

methods slated for general chemicals that are still found in current risk assessment
guidance today. Many still follow ready and inherent methods that have been around
since the late 1990s that are run at high test substance concentrations, low biomass
concentrations and nonchemical specific endpoints such as DOC or mineralization.
What is missing, however, are biodegradation methods that truly represent the trace
conditions that pharmaceuticals are introduced into the environment, and methods
that characterize the predominate mechanisms of biodegradation that some of these
methods miss. It is only in recent times that we have begun to see some more methods specific for these conditions, as with the OECD 314 B [26]. Trace level of test
substance, realistic biomass solid levels typical of wastewater treatment plants, with
specific chemical and CO2 analysis, are needed in sludge, soil, and sediment tests.
For those cases where residues are transferred from one compartment to another,
such as is with the land application of biosolids from wastewater treatment plants,
there is a need to develop more rigorous protocols to specify how and when test
substance is spiked to the soil. For example, to amend soils with biosolids, should
the test substance be taken through the anaerobic digester process before the biosolids are amended to the soils; or is it sufficient to add the test material to biosolids at
the start of the soil biodegradations study? Very little is known about the potential
binding of pharmaceuticals during anaerobic digestion and its impact on its bioavailability to subsequent degradation in soil. Similarly, very little is known about
the bioavailability of bound residues to sediments and whether the unextractable
residues are truly depleted as they enter the humification process; or whether at
some point in time they may be released. Also needed from these studies is more
specific guidance on how to calculate elimination rates such when residues are
highly bound. Simple, quick screening tests that assess the biodiversity present in
the environment from the various compartments are also not available and require
further development. A more detailed description of approaches for assessing elimination in different environmental matrices is given below.

Activated Sludge
The occurrence of pharmaceuticals in sludge and their removal during subsequent
wastewater treatment is of interest and has been well studied. Historically, the biodegradability of compounds has been screened using a variety of ready and inherent
tests. While many are economical and easy to perform, the outcome for many pharmaceuticals is that they are not readily biodegradable and require further testing. As
shown in Table 4 , the results of the ready biodegradation test show the three compounds fail the ready biodegradation test and are assigned default rate constants of
0. For the same three compounds, using the sludge die-away test (conducted at
more realistic biomass and test concentrations) one is able to determine more meaningful elimination rate kinetics. As seen with exemestane and eplerenone, the ke
values of 1.8 and 0.08 result in significant reduction of predicted surface water

Environmental Fate of Human Pharmaceuticals

71

Table 4 Comparison of ready biodegradation and sludge die-away endpoints and data output for
assessing the biodegradation of pharmaceuticals
OECD 301 ready biodegradation
OECD 314B sludge die-away
2,500 mg/L sludge solids, 1003 mg/L
test material concentration, loss of
parent,% CO2
Endpoints
% CO2 @
% CO2 @
Ke hr1
parent Ke
% Removed in
28 days
hr1
28 days
effluenta
Exemestane
15.2
Default 0
1.8
80.6
99.9
Eplerenone
2.8
Default 0
0.08
0.5
38
Varenicline
15.7
Default 0
0.01
0.68
6
a
Ke used as first order rate constant in modeling removal for 6 h hydraulic retention time
Conditions

25 mg/L sludge solids, mg/L test

material concentration,% CO2

concentrations by 99.9% and 38% respectively. Such information may be very

helpful in understanding the extent of removal during wastewater treatment and in
refining the predicted environmental concentration (PEC) for the risk assessment.
Data from these studies also provide a degradation profile illustrating the number of
biotransformation products present and the sequence of formation of such over
time. As discussed above, since biotransformation is a key pathway in its overall
biodegradation, one can see how tests that provide this additional kinetic information
are a good fit for risk assessment needs.

Water-Sediment or Soil
Occurrence of pharmaceuticals in surface waters is also very well studied [22] as
with activated sludge. Their detection in rivers and surface waters, and in some
cases in lakes and seas, would generally infer that their overall biodegradation rate
in the water is somewhat slower than in other systems. It is reasonable that these
observed elimination rates are not as great as activated sludge, or for that matter
sediment just based on the low abundance of microorganisms found in surface
waters [27]. Because of the microbial diversity encountered from one compartment to the next, it is also hard to translate elimination rates or type of transformation products seen in the sludge compartment to that found in sediment or soil as
noted with diclofenac [25, 27] for example. Occurrence of pharmaceuticals in
sediments is somewhat less published than what is seen is surface waters, and even
more so around lab studies investigating the fate in sediment systems [28, 29].
This is an area where more research is needed, especially when one considers the
sink conditions that sediments offer for most pharmaceuticals, especially cationic pharmaceuticals that are more likely to become highly bound. Table 5 shows
the elimination rates for one pharmaceutical, exemestane in sludge, mixing zone
and surface water determined in lab scale test systems (see Fig. 1 for structures of
chemicals in Tables, 1, 2, 4 and 5). One observes a general decrease in the elimination rate with the overall amount of biomass present in each of these systems.

72

A.B.A. Boxall and J.F. Ericson

Table 5 Comparison of biodegradation potential in sludge, mixing zone, river water, and watersediment systems: half-life for parent
Sludge die-away Mixing zone
River water modified Water-sediment
Exemestane
OECD 314B
OECD 314C
OECD 314C
OECD 308
DT-50 h
Ke hr1
Relative DT-50
Solids mg/L

0.39
1.8
1
2,500

58
0.012
149
15

191
0.0036
490
12.3

Generic Name

362
0.0019
923

Structure

Trade Name
CAS#

exemestane

Chiral

Aromasin

107868-30-4

H H

eplerenone
Inspra
107724-20-9

H
O
H

Champix

OH

N+

COO

azithromycin

HOOC
OH

O
O

O
N

H
O

O
N
O

Chiral

Lyrica

148553-50-8

amlodipine
Norvasc

O
O

pregabalin

Fig. 1 Structures of compounds

found in Tables 1, 2, 4 and 5

Chiral

Zithromax
83905-01-5

varenicline

375815-87-5

88150-42-9

O
O

O
O

O
Cl

Environmental Fate of Human Pharmaceuticals

73

While this is oversimplistic, it is also noted that some of this is also likely due to
changes in the overall microbial community structure and diversity found in these
samples. When comparing the DT50 values from sludge and water to that of watersediment system, we also see a progression to longer half lives. This may be
explained in part by the role sediment sorption has on the overall dissipation from
water and the resulting decrease in bioavailability in the water-sediment system.
The interface between the overlying water, pore water, and the microorganisms
found in sediment also plays a role in overall rates, suggesting that water flow may
enhance transport of pharmaceuticals to the sediment environment where degradation may occur, thereby increasing the overall observed degradation rate [27]. It is
not clear what role bound or unextractable residues have on the ultimate fate of
pharmaceuticals, and how that may impact their overall bioavailability [29, 30].
Further work is needed in developing better test methods applicable to pharmaceuticals, their route of entry into the environment and subsequent release environment, and guidance as to how to apply such data in the risk assessment.
As biotransformation is a key process in the depletion of pharmaceuticals, one
must be careful not to extrapolate its relative biodegradability from one compartment or matrix to the next. For each compartment (sludge, sediment, soil) the extent
of biodiversity, the availability of other carbon sources and the extent of residue
bioavailability will determine the overall rate of depletion and the types of potential
metabolites formed. Consequently, there is a potential need to develop elimination
rates for each of these matrices. Sludge elimination rate is pivotal to most risk
assessments, as wastewater treatment plants are common to most discharge systems
and provide some degree of removal relevant to the subsequent release environment.
The need for water, sediment, and soil elimination rates is more contingent upon the
specific target compartment of the analysis and the refinement of the risk assessment required, whether a worst case local assessment, a more general regional
assessment or perhaps a more dynamic watershed analysis (High/Low/Mean Flow).
A further discussion on how future needs and how these elimination rates should be
used in risk assessment is found in Future Needs section.
Environmental factors such as soil type, temperature, and moisture are also very
important in determining degradation rates of pharmaceuticals [31, 32]. Like, sorption, the presence of the biosolid matrix also seems to affect degradation rates compared to soil only (Table 3). For example, caffeine degradation rates in soils increased
with addition of aerobically digested sewage sludge, whereas addition of anaerobically treated sewage sludge did not accelerate caffeine mineralization [33]. The
degradation rate of naproxen was also reported to be increased by the addition of
biosolids [31]. In only a few studies has the formation of metabolites been investigated [31, 32]. No detectable transformation products were found for naproxen or
the hormones estrone and 17b-estradiol [31, 32].
Pharmaceuticals at trace levels are generally insufficient to support microbial
growth as the sole carbon source. That is not to say that one does not see any mineralization, as quite often there are minor microbial communities capable of this.
But for the most part, this is not the predominant acting mechanism, especially in
carbon source rich environment such as is found in a wastewater treatment plant.

74

A.B.A. Boxall and J.F. Ericson

This is perhaps the most overlooked factor in assessing whether something is readily biodegradable or not, and whether something may mineralizes or not. Unless
structurally similar to other carbon food sources, pharmaceuticals are unlikely to
rapidly biodegrade and completely mineralize to CO2 as is found with many other
chemicals, such as detergents, where for example biodegradation is significant not
only during wastewater treatment but also during transit to the wastewater facility
and post discharge in the mixing zone. Without an higher exposure concentration to
drive enzyme induction, rapid mineralization is unlikely to occur by microorganisms typically found in the wastewater treatment to plants that are rich in other
carbon sources.
Transformation pathways in the environment, as with metabolic pathways in
humans, are key to the degradation and elimination of pharmaceuticals. Xenobiotics
are typically detoxified by the liver via P450 and other routes of metabolism and then
excreted as more polar metabolites via the kidneys. Likewise, xenobiotics found in
the environment are transformed by an abundant sources of P450 and other enzymes
associated with microbial metabolism [34]. Many of the Phase I reactions [35] typical of human metabolism are are also observed in environmental transformations in
environmental microorganisms. As a result, these transformations become a
significant pathway in the overall depletion of pharmaceuticals in the environment
[3639]. The rate and extent of biotransformation are limited by its bioavailability, as
well as its bioaccessibilty. Rather than being an issue of uptake from the gut, it is a
question as to whether residues bound to suspended particulates, soils, and sediments
are truly available to the microorganisms for subsequent biotransformation.
Microorganisms unable to directly use pharmaceutical substrate as a carbon source
quite often require several biotransformation steps to yield something more similar
to other carbon sources before mineralization is observed.

Uptake into Organisms

Bioconcentration and bioaccumulation refer to the concentration of compound
found in plasma and/or tissue of environmental organisms relative to the concentration found in the ambient water environment, either from direct exposure or, respectively, from direct exposure and other sources, such as food uptake. Values greater
than 1 infers that the compound is bioconcentrating in the organism, and when values exceed 2,000, a compound is classified as bioaccumulative (B) or very bioaccumulative (vB) when exceeding 5,000 [40]. Historically both of these have been
mainly discussed in the context of lipophilic compound and their ability to partition
into the lipid fractions of biota from the surrounding water environment. Great
examples are found with older chlorinated hydrocarbons such as DDT and polycyclic hydrocarbons [12] for example that have log Kow values greater than 3.0 and
some with values greater than 4.5. For pharmaceuticals, however, that are predominately ionic in nature, most do not have log Kow values greater than 3.0 nor fall under
the classification of B or vB. No current pharmaceutical has exceeded a log Kow

Environmental Fate of Human Pharmaceuticals

75

of 4.5 nor triggered a PBT assessment. What is interesting though is the focus on
some pharmaceuticals such as fluoxetine, gemfibrozil, and diclofenac, for example,
that are not classified as B but may have BCF values up to 500 [41, 42]. Most of
these examples result in bioconcentration from other mechanisms than lipophilic
partitioning, such as those mechanisms that either enhance uptake relative to the
water, or diminish clearance. Compounds such as fluoxetine that have a pKa value
close to environmental pH may appear to have an enhanced uptake of a drug as
slight changes in environmental pH dramatically changes the extent of the neutral
specie present [43, 44], thereby enhancing partitioning and uptake. Compounds that
show cytological effects [45], such as with diclofenac, or alternatively enzyme inhibition [46] of diclofenac, gemfibrozil and some antidepressants, may result in
decreased renal or hepatic clearance respectively and result in a slightly elevated
BCF. It is not clear for any of these mechanisms how easy it would be to read-across
to nonclinical drug safety studies as a means of predicting some of these more
subtle BCF effects in environmental species.
Uptake of pharmaceuticals by soil organisms is also possible [4749]. Antibiotics
including florfenicol, trimethoprim, enrofloxacin, sulfamethazine, and chlorotetracycline have been shown to be taken up by plants from soils and sludge or manureamended soils [4852]. The occurrence of anthropogenic waste indicators, including
the pharmaceutical trimethoprim, has also been reported in earthworm tissue. It is
however difficult to develop a clear relationship between uptake and pharmaceutical
properties, such as hydrophobicity, as some pharmaceuticals are taken up by some
organisms and not by others and uptake into similar organisms in the different environments can vary. This is perhaps not surprising, as data for other environmental
processes (e.g., sorption to soil) indicate that the behavior of pharmaceuticals in the
environment is poorly related to hydrophobicity but is determined by a range of factors including H-bonding potential, cation exchange, cation bridging at clay surfaces, and complexation. Residues of fluoroquinolones have also recently been
reported in the eggs of vultures and kites and associated with effects on the developing
embryo (e.g., [53]). While the authors of this study indicated that the route of exposure was most likely from the consumption of carcasses of animals that have been
treated with the drugs, there is a possibility that other environmental routes of exposure may be important. A more detailed understanding of the movement of pharmaceuticals through food chains would help to address this. Through controlled
experimental studies it may be possible in the future to begin to understand those
factors and processes affecting the uptake of veterinary medicines into plants and to
develop modeling approaches for predicting uptake.

Transport of Pharmaceuticals Around the Environment

Pharmaceuticals released from sewage treatment works will typically be transported downstream and if not degraded will ultimately end up in marine systems.
Depending on the hydrology of the system, pharmaceuticals may infiltrate into

76

A.B.A. Boxall and J.F. Ericson

aquifers. The water may also be abstracted for use in drinking water or for
irrigation of crops. A number of recent studies have detected pharmaceuticals in
drinking waters (e.g., [54]). Our understanding of the fate of pharmaceuticals in
common drinking water treatment processes is however not well developed, and
this is an area where further research is required.
Following entry into the wastewater system, many pharmaceuticals will adsorb
to the sludge phase and this may be subsequently be applied as a fertilizer to agricultural land [55, 56]. It is therefore not surprising that a plethora of pharmaceuticals
and personal care products (including hormones and steroids, stimulants, antiepileptics, antidepressants, antibiotics, and musks) have been detected in biosolids
(e.g., [55, 5759]). These compounds will then be released to the soil environment
when the biosolids are used as a fertilizer. Alternatively, pharmaceuticals may also
enter soils from the irrigation of soils with contaminated wastewater [60]. In the
current risk assessment process for pharmaceuticals, consideration of terrestrial risk
is required if the sludge Koc for the compound is 10,000 or greater. This approach
however is unlikely to provide a realistic indication of which compounds are likely
to be present in biosolids applied to land, the reason being that the approach does
not consider information on the volume of the drug entering the sewage treatment
plant. If we take two hypothetical chemicals, the first has a very high sludge Koc
(>10,000) and a low usage volume, and one with a medium sludge Koc and a high
usage volume, the first compound would require terrestrial assessment whereas the
second compound would not even though the application rate to soil is in fact very
similar for both substances. We would therefore advocate that instead of using a
single Koc trigger, an alternative trigger is developed that combines information on
drug usage and sorption to sludge.
Contaminants applied to soil can be transported to aquatic systems via surface
runoff, subsurface flow and drainflow. Most work to date on contaminant transport
from agricultural fields has focused on pesticides, nutrients, and bacteria, but recently
a number of studies have explored the fate and transport of some pharmaceuticals.
For example, runoff of pharmaceuticals from soils amended with sewage sludge
has been reported [61]. In a field work performed in Canada, sewage sludge was
applied using two common practices: broadcast and injection application. In this
study, it was concluded that the pharmaceuticals studied, such as carbamazepine,
ibuprofen, acetaminophen, and naproxen, do run off with wet weather from a
broadcast application [61]. Studies into the leaching behavior of antibiotics have
shown that selected compounds have the potential to leach to groundwaters (e.g.,
[62]), and these data fit with groundwater monitoring campaigns that have detected
a number of pharmaceuticals in groundwaters [6365]. The extent of transport via
any of the processes discussed above is determined by a range of factors, including the following: the solubility, sorption behavior, and persistence of the contaminant; the physical structure, pH, organic carbon content, and cation exchange
capacity of the soil matrix; and climatic conditions such as temperature and rainfall volume and intensity.
The surface water exposure profile via these routes of exposure is likely to be
very different from the exposure profile arising from releases from wastewater

Environmental Fate of Human Pharmaceuticals

77

Fig. 2 Conceptual diagram of the aquatic exposure patterns for pharmaceuticals released to aquatic
systems from wastewater discharges (dashed line) and via the soil environment (solid line)

treatment works. When substances move from the soil environment to surface
waters, they will tend to enter in a series of pulses (corresponding to periods of
rainfall), whereas emissions from wastewater treatment processes are more or less
continuous, although with small variations in concentrations (Fig. 2). The modeling of aquatic exposure and subsequent risk assessment of pharmaceuticals that
enter the soil environment therefore probably needs to be addressed differently
from pharmaceutical releases from wastewater treatment plants.
Currently, the regulatory guidance for pharmaceuticals recommends that estimates of aquatic exposure, arising from releases from biosolid-amended soils, is
estimated using a very simple algorithm from the soil sorption coefficient. More
sophisticated models are available for predicting the movement of chemicals from
soils to surface waters. These models generally originate from the pesticide risk
assessment area. For example, the Forum for Coordination of Pesticide Fate Models
and their Use (FOCUS) have established a suite of models for predicting the concentrations of pesticides in surface waters (PRZM, MACRO, and TOXSWA) and
groundwaters (PEARL, PELMO) to support regulatory risk assessments (e.g., [66]).
FOCUS has also developed a set of scenarios covering the different climatic, soil,
and cropping characteristics encountered in the different member states. The
FOCUS modeling framework may provide a basis for assessing the aquatic exposure arising from the application of pharmaceuticals to the terrestrial environment,
although the climate/soil/crop scenarios may need to be adjusted to better reflect the
characteristics of areas where biosolids are applied (e.g., [67]). Moreover, while the
models have not been extensively evaluated against real monitoring data, the data
that are available indicate that in some instances the models may greatly underestimate exposure (e.g., Fig. 3; [68]). One possible explanation for the mismatch
between model outputs and measured concentrations is that key fate and transport
processes that are important for pharmaceuticals are not covered in the model; one
example of such a process would be facilitated transport of the pharmaceutical to

78

A.B.A. Boxall and J.F. Ericson

SCP (g/L)

0
0

10

12

14

Time after treatment (days)

Fig. 3 Comparison of modeled (square points) and measured (diamonds) concentrations of the
sulphonamide antibiotic, sulfachloropyiradzine (SCP), in a lysimeter system (adapted from
Blackwell et al. [68])

groundwater or surface waters in biosolid particulate material or biosolid derived

dissolved organic carbon. We would therefore advocate that much more work needs
to be done to evaluate the suitability of these models and to adapt them to address
some of these important fate and transport processes.

Future Needs
It is clear from the previous sections that the fate of pharmaceuticals in the environment is complex and highly dependent on the structure of the pharmaceutical, the
nature of the receiving environment, and the mode of entry into the environment.
However, current risk guidance for pharmaceuticals is often based on existing regulations for other chemical classes (e.g., industrial chemicals). Until we recognize
that pharmaceuticals are fundamentally very different from many of these chemical
classes, and begin to develop mechanisms explaining the environmental behavior of
pharmaceuticals and refine the risk assessment process accordingly, we will not
fully understand environmental risks of pharmaceutical products. There are a number of areas that we believe need attention:
1. A better understanding of those factors and processes affecting the persistence of
pharmaceuticals. Ideally we should be able to predict microbial transformation
(route and rate) of pharmaceuticals in environmental matrices (sludge, soil, surface
waters, sediment) based on models or screening. In order to achieve this we need to

Environmental Fate of Human Pharmaceuticals

2.

3.

4.

5.

6.

79

understand the key factors (species diversity, abundance, properties of the microenvironment, interactions with other chemicals, biochemical processes) modulating
biotransformation. It would also be valuable to explore whether data from studies
with mammals (e.g., ADME) can be used to inform our understanding of the rates
and routes off degradation in the environment. We also urgently need to develop
approaches to assess the environmental relevance, if any, of bound residues.
Understanding of fate in treatment processes. While we have a good understanding of the fate of pharmaceuticals in some treatment processes (e.g., activated
sludge), our understanding of fate in other treatment processes is less well developed (e.g., anaerobic digestion, drinking water treatment processes). As we
develop a better understanding of these processes, we should work to evaluate
some of the more complex wastewater treatment models and drinking water
treatment models (e.g., from engineering area) and begin to apply these in the
environmental risk assessment processes.
Modeling of transformation product exposure. When assessing risks, we tend to
focus on the parent pharmaceutical yet these may be transformed to other compounds in the different environmental compartments. In some instances, the fate
of the transformation product (and hence the exposure) may be very different
from that of the parent compound [69]. We should begin to develop an understanding of the formation and fate of transformation products and look at ways
in which we can better incorporate transformation products into the risk assessment process. Potential approaches for doing this are described in [70].
Understanding of sorption mechanisms. Data from sorption studies show that
existing models for prediction of sorption behavior or extrapolating to different
matrices are probably not appropriate for pharmaceuticals. The reason being that
sorption is determined by a range of mechanisms. Systematic studies are required
to develop an understanding of the different sorption mechanisms as well as the
effect of environmental properties for predicting sorption in different matrices.
Ultimately, we should develop quantitative structure property relationships for
estimating sorption from underlying chemical properties and environmental
properties.
Exposure model evaluation and development. A range of exposure models are
available for estimating concentrations of pharmaceuticals in different media.
While some of these have been evaluated for some substances (e.g., [71]), the
accuracy of many of the models for use on pharmaceuticals has yet to be established. We should work to use the wealth of monitoring data that are available for
different matrices to evaluate the suitability of these models. Where models are
found to fall down we should work to adapt them to cover key fate and transport
processes not currently considered (e.g., DOC-facilitated transport).
Bioconcentration and transport through food webs. Very little work has been done
to understand the uptake of pharmaceuticals into organisms and the potential for
movement of pharmaceuticals through food chains. Work is required to explore the
mechanisms of uptake (active and passive) of pharmaceuticals into plants, invertebrates and vertebrates as well as the potential movement through food webs, e.g.,
soilplantsmall mammalred kite. In addition, we need to better understand

80

A.B.A. Boxall and J.F. Ericson

biotransformation in aquatic and terrestrial organisms and its potential role in mitigating bioconcentration and bioaccumulation. Like biodegradation, it is possible
that information on the behavior of pharmaceuticals in mammalian systems (e.g.,
ADME data) may help to inform this issue.
7. Dealing with a changing landscape. It is important to recognize that the uses and
risks of pharmaceuticals in the future could be very different from today due to
changes in the environment as well as the application of new technologies. For
example, climate change may well affect fate and transport processes and effects
of pharmaceuticals in the environment as well as change drug usage patterns
(e.g., [72]), meaning that risks are very different from today. The growth of new
technologies, such as nanotechnology, also raise challenges for fate and exposure
assessment.
Finally, to ensure the safe and sustainable use of pharmaceuticals, it is critical
that regulatory fate testing guidelines are regularly updated in the light of new
scientific knowledge.

References
1. Williams RT (ed) (2003) Human pharmaceuticalsassessing the impacts on aquatic ecosystems. SETAC Press, Pensacola
2. Kmmerer K (ed) (2008) Pharmaceuticals in the environment. Springer, New York
3. Kulshrestha PJ, Giese RF, Aga DS (2004) Investigating the molecular interactions of oxytetracycline in clay and organic matter: insights on factors affecting its mobility in soil. Environ Sci
Technol 38(15):40974105
4. Gevao B, Semple KT, Jones KC (2000) Bound pesticide residues in soils: a review. Environ
Pollut 108:314
5. Alexander M (2000) Aging, bioavailability, and overestimation of risk from environmental
pollutants. Environ Sci Technol 34(20):42594265
6. ECPA (2000) ECPA position paper on soil non-extractable residues. European Crop Protection
Association, Brussels
7. Northcott GL, Jones KC (2000) Experimental approaches and analytical techniques for determining organic compound bound residues in soil and sediment. Environ Pollut 108:1943
8. Giron D (2003) Characterization of salts of drug substances. J Therm Anal Calorim
73:441457
9. Elder JP (1997) Sublimation measurements of pharmaceutical compounds by isothermal
thermogravimetry. J Therm Anal Calorim 49:897905
10. Heng JYY et al (2005) The physiochemical and surface properties of solid state pharmaceutical solids. World Congress of Chemical Engineering, 7th Glowgow, UK, pp 110
11. ASC (1990) Handbook of chemical property estimation methods, 14-120. In: Lyman WJ,
Reehl WF, Rosenblatt DH (eds) Vapor pressure. ASC, Washington
12. Boethling RS, Mackay D (eds) (2000) Handbook of property estimation methods for chemicals. Lewis Publishers, Boca Raton
13. Tetko IV, Poda GI (2004) Application of ALOGPS 2.1t predict log D distribution coefficient
for Pfizer proprietary compounds. J Med Chem 47(23):56015604
14. Tolls J (2001) Sorption of veterinary pharmaceuticals in soils: a review. Environ Sci Technol
35:33973406
15. Ter Laak TL, Gebbink WA, Tolls J (2006) The effect of pH and ionic strength on the sorption of
sulfachloropyridazine, tylosin and oxytetracycline to soil. Environ Toxicol Chem 25(4):904

Environmental Fate of Human Pharmaceuticals

81

16. Strock TJ, Sassman SA, Lee LS (2005) Sorption and related properties of the swine antibiotic
carbadox and associated n-oxide reduced metabolites. Environ Sci Technol 39:3134
17. Sassman SA, Lee LS (2005) Sorption of three tetracyclines by several soils: assessing the role
of pH and cation exchange. Environ Sci Technol 39:7452
18. Monteiro SC, Boxall ABA (2009) Factors affecting the degradation of pharmaceuticals in
agricultural soils. Environ Toxicol Chem 28(12):25462554
19. Boxall ABA, Blackwell PA, Cavallo R, Kay P, Tolls J (2002) The sorption and transport of a
sulphonamide antibiotic in soil systems. Toxicology Letters 131:1928
20. Thiele-Bruhn S, Aust MO (2004) Effects of pig slurry on the sorption of sulfonamide antibiotics in soil. Arch Environ Contam Toxicol 47(1):31
21. Buser H et al (1998) Occurrence of the pharmaceutical drug clofibric acid and the herbicide
mecoprop in various Swiss lakes and in the North Sea. Environ Sci Technol 32(1):188192
22. Kolpin D et al (2002) Pharmaceuticals, hormones, and other organic wastewater contaminants in
U.S. streams, 19992000: a national reconnaissance. Environ Sci Technol 36(6):12021211
23. Strenn B et al (2003) The elimination of selected pharmaceuticals in wastewater treatment-lab
scale experiments with different sludge retention times. Water Resour Manage II 8:227236
24. Strenn B et al (2004) Carbamazepine, diclofenac, ibuprofen and bezafibrate-investigations on
the behavior of selected pharmaceuticals during wastewater treatment. Water Sci Technol
5:269276
25. Swiener C et al (2003) Short-term tests with a pilot sewage plant and biofilm reactors for the
biological degradation of the pharmaceutical compounds clofibric acid, ibuprofen and
diclofenac. Sci Total Environ 309(13):201211
26. OECD (2008) OECD guideline for the testing of chemicals. TG 314B biodegradation in activated sludge guideline. OECD, Paris
27. Loffler D et al (2005) Environmental fate of pharmaceuticals in water/sediment systems.
Environ Sci Technol 39:52095218
28. Kunke U, Radke M (2008) Biodegradation of acidic pharmaceuticals in bed sediments: insight
from a laboratory experiment. Environ Sci Technol 42:72737279
29. Ericson J (2007) An evaluation of the OECD 308 water/sediment system for investigating the
biodegradation of pharmaceuticals. Environ Sci Technol 41(16):58035811
30. Diaz-Cruz MS et al (2003) Environmental behavior and analysis of veterinary and human
drugs in soils, sediments and sludge. Trends Anal Chem 22(6):340351
31. Topp E, Hendel J, Lapen D, Chapman R (2008) Fate of the non-steroidal anti-inflammatory
drug naproxen in agricultural soil receiving liquid municipal biosolids. Environ Toxicol Chem
27(10):20052010
32. Collucci M, Bork H, Topp E (2001) Persistence of estrogenic hormones in agricultural soils
(I17-beta estradiol and estrone). J Environ Qual 30:20702076
33. Topp E, Hendel JG, Lu Z, Chapman R (2006) Biodegradation of caffeine in agricultural soils.
Can J Soil Sci 86(23):533544
34. Rosazza JP (ed) (1982) Microbial transformations of bioactive compounds, vol I & II. CRC
Press, Boca Raton
35. Perez S, Barcelo D (2007) Application of advanced MS techniques to analysis and identification
of human and microbial metabolites of pharmaceuticals in the aquatic environment. Trends
Anal Chem 26(6):494514
36. Schulz M, Loffler D, Wagner M, Ternes T (2008) Transformation of the X-ray contrast medium
iopromide in soil and biological wastewater treatment. Environ Sci Technol 42(19):72077217
37. Perez S, Aga D, Barcelo D (2006) Biodegradation of pharmaceuticals in the environment. Top
Issues Appl Microbiol Biotechnol 113144
38. Farrre ML, Perez S, Kantiani L, Barcelo D (2008) Fate and toxicity of emerging polutants,
their metabolites and transformation products in the aquatic environment. Trends Anal Chem
27(11):9911007
39. Aga DS (2008) Fate of pharmaceuticals in the environment and in water treatment systems.
Advances in the analysis of pharmaceuticals in the aquatic environment. CRC Press, Boca
Raton, pp 5380

82

A.B.A. Boxall and J.F. Ericson

40. European Chemicals Bureau (2003) Technical guidance document on risk assessment. Part II.
European Communities, Luxembourg, p 164
41. Brown JN et al (2007) Variations in bioconcentration of human pharmaceuticals from sewage
effluents into fish blood plasma. Environ Toxicol Pharmacol 24:267274
42. Mimeault C et al (2005) The human lipid regulator, gemfibrozil bioconcentrates and reduces
testosterone in the goldfish, Carassisu auratus. Aquat Toxicol 73:4454
43. Zurita JOL et al (2007) Toxicological effects of the lipid regulator gemfibrozil in four aquatic
systems. Aquat Toxicol 81:106115
44. Nakamura Y et al (2008) The effects of pH on fluoxetine in Japanese medaka (Oryzias latipes):
acute toxicity in fish larvae and bioaccumulation in juvenile fish. Chemosphere 70:865873
45. Schwaiger J et al (2004) Toxic effects of the non-steroidal anti-inflammatory drug diclofenac
Part I: histopathological alterations and bioaccumulation in rainbow trout. Aquat Toxicol
68:141150
46. Thibaut R et al (2006) The interference of pharmaceuticals with endogenous and xenobiotic
metabolizing enzymes in carp liver: an in-vitro study. Environ Sci Technol 40(16):51545160
47. Jjemba PK (2002) The potential impact of veterinary and human therapeutic agents in manure and
biosolids on plants grown on arable land: a review. Agric Ecosyst Environ 93(13):267308
48. Boxall ABA, Johnson P, Smith EJ, Sinclair CJ, Stutt E, Levy L (2006) Uptake of veterinary
medicines from soils into plants. J Agric Food Chem 54(6):22882297
49. Kinney CA, Furlong ET, Kolpin DW, Burkhardt MR, Zaugg SD, Werner SL, Bossio JP, Benotti
MJ (2008) Bioaccumulation of pharmaceuticals and other anthropogenic waste indicators in
earthworms from agricultural soil amended with biosolid or swine manure. Environ Sci
Technol 42(6):18631870
50. Migliore L, Cozzolino S, Fiori M (2003) Phytotoxicity to and uptake of enrofloxacin in crop
plants. Chemosphere 52:12331244
51. Dolliver H, Kumar K, Gupta S (2007) Sulfamethazine uptake by plants from manure-amended
soil. J Environ Qual 36:12241230
52. Kumar K, Gupta SC, Baidoo SK, Chander Y, Rosen CJ (2005) Antibiotic uptake by plants
from soil fertilized with animal manure. J Environ Qual 34:20822085
53. Lemus JA, Blanco G, Arroyo B, Martinez F, Grande J (2009) Fatal embryo chondral damage
associated with fluoroquinolone in eggs of threatened avian scavengers. Environ Pollut
157:24212427
54. Monteiro SC, Boxall ABA (2010) Occurrence and fate of human pharmaceuticals in the environment. Rev Environ Contam Toxicol 202:53154
55. Golet EM, Xifra I, Siegrist H, Alder AC, Giger W (2003) Environmental exposure assessment
of fluoroquinolone antibacterial agents from sewage to soil. Environ Sci Technol 37(15):
32433249
56. Oppel J, Broll G, Lffler D, Rmbke J, Meller M, Ternes T (2004) Leaching behaviour of
pharmaceuticals in soil-testing-systems: a part of an environmental risk assessment for groundwater protection. Sci Total Environ 328:265273
57. Kinney CA, Furlong ET, Zaugg SD, Burkhardt MR, Werner SL, Cahill JD, Jorgensen GR
(2006) Survey of organic wastewater contaminants in biosolids destined for land application.
Environ Sci Technol 40(23):72077215
58. Lindberg RH, Wennberg P, Johansson MI, Tysklind M, Andersson BAV (2005) Screening of
human antibiotic substances and determination of weekly mass flows in five sewage treatment
plants in Sweden. Environ Sci Technol 39:34213429
59. Reddersen K, Heberer T, Dnnbier U (2002) Identification and significance of phenazone
drugs and their metabolism in ground- and drinking water. Chemosphere 9:539544
60. Ternes T, Bonerz M, Herrmann N, Teiser B, Andersen HR (2007) Irrigation of treated wastewater in Braunschweig, Germany: an option to remove pharmaceuticals and musk fragrances.
Chemosphere 66:894904
61. Topp E, Monteiro SC, Beck A, Coelho BB, Boxall ABA, Duenk PW, Kleywegt S, Lapen DR,
Payne M, Sabourin L, Li H, Metcalfe CD (2008) Runoff of pharmaceuticals and personal care

Environmental Fate of Human Pharmaceuticals

62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.

83

products following application of biosolids to an agricultural field. Sci Total Environ

396:5259
Blackwell PA, Kay P, Boxall ABA (2007) The dissipation and transport of veterinary antibiotics in a sandy loam soil. Chemosphere 62(2):292299
Heberer T, Verstraeten IM, Meyer MT, Mechlinski A, Reddersen K (2001) Occurrence and
fate of pharmaceuticals during bank filtration- preliminary results from investigations in
Germany and the United States. Water Resour 120:417
Holm JV, Rgge K, Bjerg PL, Christensen TH (1995) Occurrence and distribution of pharmaceutical organic compounds in the groundwater downgradient of a landfill (Grinsted, Denmark).
Environ Sci Technol 29(5):14151420
Ternes, T (2001) Pharmaceuticals and metabolites as contaminants of the aquatic environment.
In: Daughton CG, Jones-Lepp T (eds) American Chemical Society, Symposium series 791.
Washington, DC, pp 3954
FOCUS (2000) FOCUS groundwater scenarios in the EU plant protection product review process. Report of the FOCUS Groundwater Scenarios Workgroup, EC document reference
Sanco/321/2000. p 197
Schneider MK, Stamm C, Fenner K (2007) Selecting scenarios to assess exposure of surface
waters to veterinary medicines in Europe. Environ Sci Technol 41:46694676
Blackwell PA, Kay P, Ashauer R, Boxall ABA (2009) Effects of agricultural conditions on the
leaching behaviour of veterinary antibiotics in soils. Chemosphere 75(1):1319
Boxall ABA, Sinclair CJ, Fenner K, Kolpin DW, Maund S (2004) When synthetic chemicals
degrade in the environment. Environ Sci Technol 38(19):369A375A
Boxall ABA (2009) Transformation products of synthetic chemicals in the environmental.
Springer, Germany
Metcalfe C, Boxall A, Fenner K, Kolpin D, Servos M, Silberhorn E, Staveley J (2008) Exposure
assessment of veterinary medicines in aquatic systems. CRC Press, Boca Raton
Boxall ABA, Hardy A, Beulke S, Boucard T, Burgin L, Falloon PD, Haygarth PM, Hutchinson
T, Kovats RS, Leonardi G, Levy LS, Nichols G, Parsons SA, Potts L, Stone D, Topp E, Turley
DB, Walsh K, Wellington EMH, Williams RJ (2009) Impacts of climate change on indirect
human exposure to pathogens and chemicals from agriculture. Environ Health Perspect
117(4):508514

Environmental Comparative Pharmacology:

Theory and Application
Lina Gunnarsson, Erik Kristiansson, and D.G. Joakim Larsson

Introduction
Pharmaceuticals are intentionally selected or designed to interact with specific target proteins at relatively low doses. Similarly, their physicochemical characteristics
often allow for their efficient uptake across biological membranes. As drugs most
often are present in the environment at very low concentrations, high-affinity
interactions are likely to mediate any adverse effects in wild life species. It can
therefore be assumed that nontarget organisms with conserved drug targets have a
higher risk of being affected by residual drugs, compared with species lacking conserved targets. Furthermore, the molecular mechanisms behind uptake, distribution,
metabolism, excretion and pharmacological effects can be conserved between the
organism that the drug is intended to affect (usually humans) and potential nontarget
organisms in the environment. Accordingly, certain pharmaceuticals pose an environmental risk at very low concentrations [13].
The vast knowledge base of a new drugs molecular, pharmacokinetic, and pharmacodynamic properties in humans and other mammalian models, derived during
its development, provides a basis for an expanded understanding of the potential
action of residual pharmaceuticals in exposed nontarget species [4]. However, a
comprehensive understanding of the physiology of the exposed wildlife species is
also necessary in order to make well-founded predictions, and for the vast majority

L. Gunnarsson (*) D.G.J. Larsson

Department of Neuroscience and Physiology, Institute of Neuroscience and Physiology,
The Sahlgrenska Academy, University of Gothenburg, Box 434, 405 30 Gteborg, Sweden
e-mail: [email protected]
E. Kristiansson
Department of Neuroscience and Physiology, Institute of Neuroscience and Physiology,
The Sahlgrenska Academy, University of Gothenburg, Box 434, 405 30 Gteborg, Sweden
Department of Zoology, University of Gothenburg, Box 463, 405 30 Gteborg, Sweden
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_5, Springer Science+Business Media, LLC 2012

85

86

L. Gunnarsson et al.

of species, this is currently a hampering factor. The recent advances in sequencing

and characterization of genomes and transcriptomes have opened up new possibilities to advance the field of comparative pharmacology with an ecotoxicological
focus. Information gained from deeper comparative efforts has the potential to aid
in the prioritizing of drugs that need further attention for assessment of their environmental risks. Such data can also guide the selection of appropriate test species
and methodologies (e.g. endpoints) [57]. Furthermore, genomic information could
also indicate possible test-combinations of drugs and species which are not likely to
be protective for others. In other words, toxicity data that are generated from a
species that is lacking a human drug target ortholog might not be protective for
a species with a conserved drug target. The copepod, Nitocra spinipes, does for
example not have an ortholog to the estrogen receptor. Consequently, the NOEC
value of 0.05 mg/L for chronic toxicity of EE2 in the copepod [8] is not protective
for fish, even if an assessment factor of 1000 was applied.
Much of the best ecological comparative pharmacology work today has been a
result of examining the literature on drugs target proteins, pathways, mechanisms,
etc. in nontarget species. The European Centre for Ecotoxicology and Toxicology
of Chemicals (ECETOC) has published a review on the use of intelligent test strategies in ecotoxicology [9]. They suggest and exemplify how information about the
mode-of-action for specifically acting chemicals can be used in the environmental
risk assessment. Many of the examples and case studies include pharmaceuticals.
Other reviews, focusing on the comparative pharmacology in fish for selective
serotonin reuptake inhibitors (SSRIs) and adrenoreceptorantagonists (beta-blockers), have also been published. Kreke and Dietrich (2008) summarize the current
knowledge about the comparative pharmacology of SSRIs with an emphasis on
possible physiological endpoints of potential SSRI interactions in fish, and conclude that the serotonergic system plays a modulatory role in several physiological
processes in fish and that serotonin signaling transduction may be mediated by
neuronal, endocrine and paracrine pathways. The influence of serotonin on different target tissues appears to be species-specific and may also depend on the gender
and/or the development and reproductive status of the individual. Because of this
complexity, it is difficult to assess the potential consequences of prolonged exposure of fish populations to SSRIs [10]. Owen et al. [11] describe the current knowledge about the comparative physiology, pharmacology and toxicology of
b-blockers: to date, the full repertoire of b-adrenergic receptors has not been
reported for any fish species, and even less is known about their expression and
specificities for b-blockers. Another paper by Brain et al. [12] reviews the effects
and risks of exposure to pharmaceuticals in aquatic plants. Plants provide a number
of evolutionarily conserved target sites for antibiotic drugs, resulting from the bacterial ancestry of plastid organelles and conservation of certain metabolic pathways. The statin type of blood lipid regulators is a group of pharmaceuticals with
a human target that also are conserved in plants. Indeed, measuring the downstream metabolites (sterols) of the target enzyme (HMG-CoA reductase) provided
a specific biomarker in Lemna gibba, an aquatic plant, with a sensitivity 2 or 3
times lower than that of fresh weight. Apart from antibiotics and statins, there are

Environmental Comparative Pharmacology: Theory and Application

87

few other classes of pharmaceuticals that are known to exert a strong toxicity in
plants. Recently, Winter et al. [4] have published a review on the usage of drug
development data in the environmental risk assessment of pharmaceuticals, discussing challenges associated with read across. Access to data from the drug development process and established strategies for how to perform comparative
predictions is however lacking. Taken together, the different reviews conclude that
future toxicological testing should encompass and reflect the known pharmacological effects of the substances studied, and should therefore focus more strongly
on specific molecular targets. By identifying potentially affected pathways, it may
be possible to identify sensitive endpoints.
In this chapter, we would like to start from a theoretical point-of-view and discuss ways to predict the conservation of proteins known to interact with drugs in the
human body. Although such an approach has limitations, and of course must be followed up by empirical studies, it might enable predictions of both pharmacokinetic
and pharmacodynamic properties for a large set of drugs in wildlife species with
relatively limited effort. We therefore put some focus on how to predict proteins
with a conserved function and the downstream pathways in nontarget organisms.
Without any attempt at comprehensiveness, we also give some selected examples on
both theoretically and empirically derived pharmacokinetics and pharmacodynamic
data in nontarget species.

Information on Pharmacokinetics and Pharmacodynamics

of Human Pharmaceuticals
The physiochemical, pharmacological and toxicological properties of an active
pharmaceutical ingredient (API) are extensively studied during the development
of a new drug. For most approved APIs, such information is easily accessible in
different public databases, of which some of the most important are listed
below.
ChEMBL (www.ebi.ac.uk/ChEMBL) is a chemogenomic database for drug-like
molecules that brings together chemical, bioactivity, and genomic data. To date,
it contains more than 500 000 compounds.
DrugBank database (http://www.drugbank.ca/) is an example of a bioinformatics
and chemoinformatics resource combining detailed drug data with metabolizing
enzyme and drug target information.
KEGG drug (http://www.genome.jp/kegg/drug/) is an information resource for
all approved drugs in Japan and the USA. Based on the drugs chemical structure, but also contains information about drug targets and pathways.
Pharmacogenetics and Pharmacogenomics Knowledge base (PharmGKB; http://
www.pharmgkb.org/) contains rather few drugs and drug targets, but has information about the relationships between drugs, affected pathways and genes
therein, diseases and genes, including their variations and gene products. It aims

88

L. Gunnarsson et al.

to aid researchers in understanding how genetic variation among individuals

contributes to differences in reactions to drugs.
RxList (http://www.rxlist.com/script/main/hp.asp) is a comprehensive drug information database aiming to assist and support clinical decisions.
For ecologically relevant nontarget organisms, empirically derived pharmacokinetic and pharmacodynamic data are much more fragmented. Human drugs have
been used for a long time, in order to gain insight into the physiology of different
organisms. Thus, even literature with a very different purpose in mind, published
before the environmental effects of pharmaceuticals became an issue of concern,
can prove useful. During the past 1015 years, large datasets, primarily on microalgae, daphnia, and to some extent fish, have been developed for the environmental
risk assessments required for the approval process of new medical products in the
EU [13] and USA [14]. The Swedish Association for Pharmaceutical Industries is
responsible for the Web site www.fass.se, where they publish such data for products on the Swedish market. For most of these entries, there is, however, no pharmacokinetic information on nontarget species. Also, most of the data are based on
results from short-term tests, and mainly cover gross endpoints such as lethality. In
the EU, acute tests are no longer considered sufficient for environmental risk
assessments of pharmaceuticals as short-term lethality (or inhibited growth) may
not be reflective of the specific mode-of-action that could be expected to dominate
at low, environmentally relevant concentrations [13]. Accordingly, moving to more
chronic tests in general appears to have changed the species sensitivity distribution, with fish more often becoming the most sensitive organism [15]. Indeed, this
is in agreement with a higher degree of conservation of drug targets in fish compared with daphnia and microalgae [6]. This also means that the value of the vast
dataset on short-term toxicities for comparative pharmacology purposes may be
relatively limited.

Predicting Conserved Function of Proteins

Translating pharmacological information from humans to other nontarget species is
challenging and not without pitfalls. Such efforts are often based on inter-species
comparisons of the human proteins affected by the drug, such as the primary target
and the downstream pathway. If the biochemical properties of a protein in a nontarget species are similar to those of a human protein associated to a given drug, it is
likely that the two proteins share similar pharmacological characteristics. Thus, the
developmental history and evolution of proteins that are affected by APIs can provide information valuable in an ecotoxicological context.
The pharmaceutical industry has started to put more focus on the evolutionary
aspects within the drug development. One reason for this is the rather low success
rate in the pharmaceutical pipeline, which could partly be explained by the difficulties
in successfully translating results from safety and efficacy studies in animal models

Environmental Comparative Pharmacology: Theory and Application

89

Fig. 1 Homologous proteins have a shared common ancestry. (a) An ancestral gene G undergoes
one speciation event which is followed by a gene duplication event. The different variants of G
appearing in different species are called orthologs (G: frog; G1 or G2: human). The different variants within each species are called paralogs (G1 and G2: human). Paralogs appearing as a result of
a gene duplication event occurring after the latest speciation event are called recent or in-paralogs.
(b) The ancestral gene G undergoes one gene duplication event followed by two speciation events.
The different variants between the different species are still referred to as orthologs, but the paralogs in this case are called ancient or out-paralogs

to humans [16]. It is argued that the extrapolation of safety and efficacy could be
improved if the evolutionary history, including information regarding functional
similarities and discrepancies, is known for the target proteins of a drug. One apparent example is the choice of an appropriate animal model, which could be founded
on information regarding the evolutionary conservation of the target pathway.
Another example is minimizing the risk of unexpected off-target effects, which may
be due to drug interaction with additional targets that have a shared ancestry with
the primary drug target protein.

Homologs, Orthologs, and Paralogs

Functionally similar proteins between different species are generally identified
based on sequence homology, a concept that is outlined in Fig. 1a, b. Two proteins
are called homologous if they are evolutionarily related such that they stem from a
common ancestral protein. The homologous proteins can be further classified as
orthologs and/or paralogs, where orthologs are homologous proteins that exist in
different species as a result of a former speciation event (Fig. 1), while paralogs, on
the other hand, originate from gene duplication events within the genome of a single
species. Although paralogous proteins often retain similar biochemical functions,
they generally diverge after the duplication event. Paralogs can be divided into two
groups, in-paralogs (Fig. 1a) and out-paralogs (Fig. 1b), depending on whether the
duplication event occurred before or after the latest speciation event. These two
groups of paralogs are therefore sometimes denoted as recent and ancient paralogs.

90

L. Gunnarsson et al.

Since homologous proteins have a common evolutionary history, they often have
similar biological functions. Even though orthology does not guarantee functional
equivalence, the concept can be utilized to extrapolate the biochemical properties of
drug target proteins from humans to nontarget species [17].
The identification of orthologous proteins between diverse species is in general
a nontrivial undertaking, which is both conceptually and computationally challenging. In the eukaryotic kingdoms, this problem is particularly delicate since a substantial part of the genetic variation stems from recombination events giving rise to
extensive functional redundancy. Reliable and accurate orthology predictions
between different eukaryotic species are therefore dependent on the ability to detect
both orthologs and paralogs [18]. Several different approaches are described in the
literature, and these can be loosely dived into those that are phylogenetically based
and those without an explicit tree structure. For reviews of existing approaches for
predicting orthologs, please refer to [19, 20].

Orthology Predictions
Phylogenetics is the study of evolutionary relatedness based on molecular sequence
data [21]. The phylogenetic topology, usually assumed to have the form of a tree,
can be estimated from a group of related proteins. This structure encodes the causality of the evolutionary events, including speciation and gene duplication, and hence
the orthologous proteins can be identified. Even though phylogenetically based
approaches are generally considered to be relatively accurate and have a high resolution, they also exhibit number of weaknesses when applied to the en masse prediction of orthologs [22]. Inferring reliable phylogenetic trees is in general dependent
on high-quality multiple sequence alignments, which in turn, depends on the a priori
selection of relevant proteins [23]. Creating multiple alignments may also require
manual intervention to achieve an optimal result, especially for environmental risk
assessment purposes, where human orthologs are predicted in distantly related species. Furthermore, computing phylogenetic trees is also a computationally intensive
task, which will limit the number of species that can be searched for orthologs.
Popular methods for predicting orthologs based on phylogenetic trees are the
EnsemblCompara Gene Trees [24] and PhiGs [25].
The identification of proteins with conserved functions can also be achieved
without assuming an explicit evolutionary tree-like structure. A forthright approach
is to use one-way comparisons, where a protein in one species is compared to all the
proteins in a divergent species and those with a sequence similarity above a
predefined threshold are defined as orthologs. For computational efficiency, these
comparisons are usually performed using heuristic sequence comparison procedures, such as FASTA [26] or Basic Local Alignment and Search Tool (BLAST)
[27], and the sequence similarity is usually based on a generated alignment score
(e.g., E-value). Unfortunately, one-way sequence comparisons are often plagued by
false positives. Since this approach does not take in-paralogs and out-paralogs into

Environmental Comparative Pharmacology: Theory and Application

91

account, they are both usually identified as orthologs, a problem that is particularly
severe in eukaryotic genomes exhibiting a large amount of paralogs [28]. Sequence
comparisons from local alignments can also be too optimistic, such that divergent
proteins sharing a single preserved functional domain can get a high sequence similarity score. The number of false positives from BLAST-based one-way comparisons was evaluated by Chen et al. [19], who estimated that 50% of the predicted
orthologs between divergent eukaryotic species were false positives, while only 4%
were false negatives. Nevertheless, the one-way comparison approach has still
proven to be of some use, and was applied by Kostich and Lazorchak [7] to identify
several conserved human drug targets in nontarget eukaryotic species.
It is possible to generalize from one-way comparisons, comparing one single
protein against all proteins in a divergent species, to a symmetric approach, where
all proteins in both species are compared against each other. In the most conspicuous setting, proteins from two species are called orthologous if they match each
other, or in other words, they are each others reciprocal best hit (RBH) [17, 29].
This approach has long been used to identify orthologs between prokaryotes and
can easily be extended to include several genomes [30]. Since each pair of proteins
needs to match each other, the RBH-approach is much more conservative than oneway comparisons, and the proportion of false positives was estimated by [19] to
decrease to 8%, while that of false negatives increased to 30%. The reason for the
high number of false negatives is the inability to recognize many-to-many and
many-to-one ortholog relationships, i.e., in-paralogs and multiple orthologs.
There are several procedures that extend the basic RBH approach to also incorporate multiple orthologs and in-paralogs. The InParanoid algorithm applies clustering to identify orthologs between pairs of species [31]. One major drawback of
the InParanoid algorithm is its inability to identify orthologs between several
genomes simultaneously. Ortholog predictions from InParanoid are available at the
InParanoid Web site (http://inparanoid.sbc.su.se/cgi-bin/index.cgi), which currently
comprises 100 organisms with more than 1,600,000 proteins.
OrthoMCL is another algorithm for the identification of orthologs between
eukaryotic genomes [18, 19]. In contrast to InParanoid, OrthoMCL can identify
both multiple orthologs and in-paralogs in any number of species simultaneously.
The algorithm works in a two-staged manner, where first all proteins are compared
against each other using BLAST. These results are then interpreted as a graph where
the nodes corresponds to proteins and the weighted edges their pair-wise sequence
similarity. The graph is then partitioned into sub-graphs using a technique called
Markov Clustering (MCL), a fast and efficient algorithm for clustering large graphs
[32, 33]. The results are several clusters of proteins, each containing the orthologs
and the in-paralogs for each species (Fig 2). The OrthoMCL algorithm was estimated to perform well (16% false positives and 7% false negatives) on a divergent
set of eukaryotic species by Chen et al. [19], and the procedure has been used to
predict orthologs for 1,318 human drug targets in 16 species of which many are
relevant to ecotoxicological testing [6].
There are also several other considerations that can improve the power of orthology
prediction. One example is secondary and tertiary protein structure, and another is

92

L. Gunnarsson et al.

Fig. 2 Identification of both paralogs and orthlogs is necessary to understand the evolutionary
history of a gene. The figure shows an example of an ancestral gene G, which appears as different
variants in human, rat and frog. Since the reciprocal best BLAST hit (RBH) of G1 in frog is G1 in
human and G2 in rat, while the RBH of G1 in human is G1 in rat, algorithms solely using information from the RBHs would not identified orthlogs for G in all three species. However, the clustering
approach utilized by OrthoMCL can detect all these nontrivial relationships and correctly assign
all variants within a single cluster containing both the in-paralogs and orthologs

synteny, i.e., evolutionarily preserved chromosomal localization of genes. The latter

has been combined with BLAST-based sequence similarity measures in the NCBI
Homologene database (http://www.ncbi.nlm.nih.gov/homologene).

Comparative Pharmacokinetics
Pharmacokinetics describes how the body affects a specific drug after administration, and can be separated into absorption, distribution, metabolism, and excretion
(ADME). This information is critical to understand how, for how long, and at what
concentration an API gains access to its molecular target(s) in the organism.

Absorption and Distribution

The uptake of APIs in humans occurs most commonly through ingestion, while the
uptake route can be very different in other species. A summary of the different
uptake routes of pharmaceuticals in fish, invertebrates, plants and algae are available
in [9]. For most pharmaceuticals, the major uptake route in fish and amphibians is
likely to be through the gill/lung and skin. Randall et al. [34] showed that fish take
up lipophilic xenobiotics (log Kow >3) mainly across the gills, and the substances

Environmental Comparative Pharmacology: Theory and Application

93

then enter the bloodstream directly. The amount of uptake via ingestion appears to
be negligible for most pharmaceuticals in view of the normally low feeding rate and
water intake of fish compared with their much larger breathing volume [34]. For
invertebrates, the uptake route is heavily dependent on the life stage of the animal
and the environment they are living in [35], while the uptake in plants and aquatic
macrophytes depends on the type of plant tissue that is in contact with the drug [9].

Passive Diffusion and Carrier-Mediated Uptake

The uptake of pharmaceuticals is generally regulated by passage through the cell
membrane in all organisms. A drug can permeate by passive diffusion or by an
active or carrier-mediated uptake. Passive diffusion through the lipid bilayer of the
cell is often considered to be the dominant process by which a drug is taken up by
the cell. In drug development, Lipinskis rule of five is used to identify drug candidates with poor absorption, and it assumes that drugs are mainly taken up by
passive diffusion. The rule of five predicts that poor absorption is more likely
when there are more than five hydrogen bond donors, ten hydrogen bond acceptors,
the molecular weight is greater than 500 and the calculated Log P is greater than
five [36]. Huggett et al. [37] have similarly proposed a model, assuming passive
diffusion, to predict the risk for fish to be affected by pharmaceuticals. The only
inputs in this model are the water concentration of the drug, the LogP of the drug,
and the human therapeutic plasma concentration. Empirical results suggest that a
rather simplistic model like this could be valuable for identifying APIs with the
potential to affect aquatic organisms at environmentally relevant concentrations
[3742]. Many pharmaceuticals are ionisable and the pH of the environment can
affect the uptake and toxicity [43, 44]. Therefore, consideration of the ionisation of
an API can be important for environmental risk assessments.
In many cases, drugs do not behave as expected based on passive diffusion alone,
and other factors need to be considered in predicting the absorption and distribution
of a drug. One factor may be carrier-mediated or active uptake of drugs. Dobson and
Kell [45] argue that carrier-mediated and active uptake of pharmaceuticals may be
more common than traditionally assumed. Indeed, many drugs are taken up by
carriers in those specific cases where it has been studied (for a comprehensive list of
examples see supplementary information S1 in [45]).
Carrier-mediated and active drug uptake might for example explain why certain
drugs concentrate in specific tissues and also bioaccumulate in some aquatic organisms. Dobson and Kell [45] give examples of drug uptake by three of the most
significant families of transporters. We used the NCBI Homologene database to
predict the evolutionary conservation of these transporters. The solute carrier
organic anion transporter family, member 1B1 (SCLO1B1), was the least conserved
transporters, while four of the transporters were predicted to be conserved among
all eukaryotes (Table 1). A conserved transporter protein might indicate a potential

94

L. Gunnarsson et al.

Table 1 Evolutionary conservation of selected human

based on the NCBI Homologene database
Transporter
Pharmaceuticals
HUGO symbol
Amoxicillin, Cefaclor, Cefalexin,
SLC15A1
Bestatin, Amoxicillin, Ampicillin,
Cefadroxil, Cefixime, Enalapril,
Midodrine, Valacyclovir,
Valganciclovir, Ceftibuten
Amoxicillin, Cefaclor, Cefadroxil,
SLC15A2
Bestatin, Valganciclovir
Zidovudine, Acyclovir, Ganciclovir,
SLC22A1
Metformin, Cimetidine
Memantine, Metformin, Propranolol, SLC22A2
Cimetidine, Zidovudine,
Pancuronium, Quinine
Cimetidine
SLC22A3

Quinidine, Verapamil

SLC22A4

Quinidine, Verapamil, Valproate,

Cephaloridine
Adefovir, Acyclovir, Zalcitabine,
Didanosine, Stavudine,
Trifluridine, Ganciclovir,
Lamivudine, Zidovudine,
Methotrexate, Ketoprofen,
Ibuprofen, Cimetidine,
Tetracycline, Cephaloridine
Zidovudine, Tetracycline, Salicylate,
Methotrexate, Erythromycin,
Theophyline
Valacyclovir, Zidovudine,
Methotrexate, Salicylate,
Cimetidine, Cephaloridine
Zidovudine, Cephaloridine

SLC22A5
SLC22A6

transport proteins for pharmaceuticals

Description
Oligopeptide
transporter

Conserved according
to Homologene
Eukaryota

H+/peptide
transporter
Organic cation
transporter
Organic cation
transporter

Eukaryota

Extraneuronal
monoamine
transporter
Organic cation
transporter
Organic cation
transporter
Organic anion
transporter

Amniotaa

Amniotaa
Euteleostomib

Eukaryota
Eukaryota
Euteleostomib

SLC22A7

Organic anion
transporter

Euteleostomib

SLC22A8

Organic anion
transporter

Eutheriac

SLC22A11

Organic anion/
cation
transporter
Organic anion
transporter
Organic anion
transporter

Eutheriac

Organic anion
transporter

Amniotaa

Fexofenadine, Rocuronium, Enalapril, SLCO1A2

Temocaprilat, Rosuvastatin
Benzylpenicillin, Pravastatin,
SLCO1B1
Rifampicin, Atorvastatin,
Capsofungin, Cerivastatin,
Fexofenadine, Fluvastatin,
Pitavastatin, Methotrexate
Digoxin, Rifampicin, Fexofenadine,
SLCO1B3
Fluvastatin, Pitavastatin,
Rosuvastatin, Methotrexate

Amniotaa
Homo/Pan/Gorilla
group

(continued)

Environmental Comparative Pharmacology: Theory and Application

95

Table 1 (continued)
Pharmaceuticals
Pravastatin, Glibenclamide,
Atorvastatin, Benzylpenicillin,
Fluvastatin, Rosuvastatin
Methotrexate, Digoxin

Transporter
HUGO symbol
SLCO2B1

SLCO4C1

Description
Organic anion
transporter
Organic anion
transporter

Conserved according
to Homologene
Amniotaa

Bilateriad

The selection of drugs and transport proteins is based on Dobson and Kell [45]
Mammals, reptiles, and birds
b
Bony vertebrates
c
Placental mammals
d
All animals with bilateral symmetry
a

for bioaccumulation of the drugs taken up by the transporter. However, drugs could
also concentrate in species that lack transporter orthologs and the significance of
conserved carrier proteins needs to be evaluated. Data on uptake and distribution of
pharmaceuticals, as well as molecular characterization of transporters in nontarget
species, are needed.

Plasma-Protein Binding
Binding to proteins in the blood plasma is another factor that can affect the uptake
and distribution of pharmaceuticals [46]. If a drug is bound to a plasma protein, it
limits the drugs free motion, reduces its volume of distribution as well as its renal
excretion, liver metabolism, and tissue penetration. Binding of plasma proteins can
also increase the absorption and the half-life of the drug. Most drugs commonly
bind to serum albumin (ALB) and orosomucoid (ORM1 and 2, alpha acid glycoprotein) in humans [47]. Neither serum albumin nor orosomucoid has orthologs in fish
according to the NCBI Homologene database but an albumin-like protein has been
described [48] although not predicted to be orthologous to human ALB. The plasma
protein profile is also different in fish and the total protein levels are generally lower
than in human plasma [49]. Thus, the characteristic binding of drugs to plasma
proteins may not extrapolate to fish. For example, the antibiotic drug sulfadimethoxine and the antimicrobial ormetrophine are to a great extent associated to proteins in human plasma while the binding is very limited in trout [49]. Other drugs
bind however in a similar manner. Sex hormone-binding globulin is conserved in
euteleostomi (bony vertebrates) and it is the major transport protein for sex steroids
in the blood both in humans and in fish [50]. It was shown that sex hormone-binding
globulin controls the flux of sex steroids across fish gills and that its function can be
hijacked, for example, by 17a-ethinylestradiol (EE2) [51]. The synthetic progestin
levonorgestrel bioconcentrates from water into blood plasma of trout considerably
more than what is expected from its log P [40]. The high potency of this drug in fish

96

L. Gunnarsson et al.

[3] is likely linked to a high bioconcentration factor facilitated by binding to sex

hormone-binding globulin in the fish [40, 52]. Plasma binding proteins could thus
be an important factor to consider both when uptake and bioavailability at the drug
target should be predicted.

Metabolism and Excretion

Phase I and phase II drug metabolizing enzymes play a central role in the metabolism of drugs. Pharmaceuticals are often hydrophobic and need to be biotransformed
to become more polar and water soluble so that they can be excreted. Biotransformation
not only promotes drug elimination but can also change the overall biological properties of the drug, leading to the activation or inactivation of the pharmacological
activity. Phase I metabolizing enzymes often catalyze oxidation, reduction, and
hydrolysis reactions, while phase II enzymes catalyze conjugation reactions, which
add polar functional groups to the drug. The metabolites generated by phase I and
II reactions can be excreted from the body with the aid of membrane efflux pumps
such as the multidrug resistance associated proteins [53]. Drug metabolism can
occur in many diverse cell and organ systems, but the liver, intestine, kidney, and
gill/lung play the largest role in vertebrates. In mammals, the kidney is by far the
most important organ for excreting drugs. Short summaries of the metabolism and
excretion of chemicals in fish, invertebrates, plants, and algae are available in [9].
The main phase I metabolizing enzymes are the cytochrome P450 protein superfamily, which catalyzes the incorporation of one oxygen atom from molecular O2
into a substrate. This cytochrome reaction can be observed in virtually all living
organisms, from bacteria to mammalian species [54]. In human liver, the most
important drug metabolizing P450 enzymes are CYP1A2, CYP2B6, CYP2C9,
CYP2C19, CYP2D6, and CYP3A4/CYP3A5, which metabolize about 95% of the
drugs in clinical use [55, 56].
There are several problems with predictions of P450 mediated drug metabolism
in nontarget species based on human data. It is very difficult to predict the orthology
of P450s across distantly related species using sequence similarity based prediction
methods. The great diversity of the P450 superfamily has arisen by extensive processes of gene duplication, conversions, genome duplications, gene loss and lateral
transfers. This have created a large number of P450 paralogs, and many out-paralogs
and in-paralogs are present in almost all eukaryotic genomes [54]. Mammalian
CYP1A1 and CYP1A2 are believed to have diverged 250 million years ago by a
duplication event [57]. Fish diverged from the mammalian line prior to that and
does consequently only have one CYP1A gene [58]. The CYP2 gene family is the
most diverse CYP gene family with 13 know CYP2 subfamilies in fish. None of the
human CYP2 enzymes that are the most important for drug metabolism are believed
to have orthologs in fish [58]. The ortholog relationships between various CYP3A
enzymes are more unclear. To date, 13 teleost CYP3A genes have been identified,
but the current nomenclature for the CYP3 gene family does not reflect orthologous

Environmental Comparative Pharmacology: Theory and Application

97

relationship between organisms [58]. Despite the clear ortholog relationships and
high conserved function of individual P450s in mammals, there are significant differences between species. For example, orthologous P450 enzymes in closely
related species can have very different basal expression levels, different levels of
induction by APIs, and also the enzyme substrate specificity may differ [59]. An
example is omeprazole, a CYP1A inducer in humans that has little effect on CYP1A
forms in rats, mice, and rabbits [55]. Another example is CYP1A mediated metabolism in fish and mammals. The fish CYP1A has similar substrate preferences, but
the oxidation rate for some substrates can differ by orders of magnitude between
fish and mammalian species [60]. Developmental differences in protein expression
and activity of P450 enzymes are also important factors to consider. CYP1A is for
example expressed in zebrafish 15h post fertilization but no protein expression or
ethoxyresorufin O-deethylase activity could be detected until after hatching [61].
Taken together, we believe that high-throughput orthology predictions to extrapolate P450-mediated metabolism of drugs between distantly related species suffer
from major limitations.
Nevertheless, in cases where the orthologous relations are clear there are also
examples of functional similarities between P450 enzymes in, for example,
mammals and fish. The fungicide ketoconazole inhibits CYP3A in both mammals [62] and fish [63], for example. A useful base for comparisons could be
Lee et al. ([62]; Table IV) that summarizes drugs and xenobiotics that are inhibitors of one or more human cytochrome P450 enzymes involved in drug
metabolism.
Phase II metabolizing enzymes often catalyze conjugation reactions, which add
more polar functional groups to the drug. Sulfotransferases (SULTs), UDPglucuronosyltransferases (UGTs), and glutathione S-transferases (GSTs) are examples of phase II enzymes that catalyze the addition of sulfate, glucuronate, and
glutathione respectively. Methyltransferase, NAD(P)H:quinone oxidoreductase
(DT-diaphorase), and acetyltransferase are other examples of phase II metabolizing
enzymes [53]. The UGTs, SULTs, and GSTs are all superfamilies of proteins, with
homologs present in almost all eukaryotic genomes. As many as 117 mammalian
UGT genes and 56 distinct eukaryotic SULT isoforms and have been identified to
date [6466]. The UGTs are the most important phase II drug metabolism enzymes,
and have therefore been extensively studied in humans. However the research on
UGTs in lower vertebrates and invertebrates is much more limited [58].
Today, the lack of understanding of the detailed function and specificities of
different phase I and II drug metabolism enzymes are two of several factors hampering the accurate prediction of the kinetics of drugs in nontarget species. There is a
great need for empirical studies on ADME of APIs in wildlife species, both for
generating and evaluating different predictive models. Predictions on the pharmacokinetics of drugs in different species may be a valuable component in identifying
species at risk. For example, differences in pharmacokinetics of diclofenac between
different vulture species may be an explanation behind the rather large difference in
sensitivity between relatively closely related birds [67]. Computational models are
frequently and successfully used in the drug discovery pipeline to predict ADME of

98

L. Gunnarsson et al.

a substance [6870]. In the future, when we have better knowledge of the function
of individual transporters and metabolizing enzymes, such in silico approaches
might be possible to use in nontarget species as well.

Pharmacodynamics
Pharmacodynamics is the study of the biochemical and the physiological effects of
drugs on the body. This includes the interaction between the drug and the target
protein (i.e., activation or inhibition), the downstream mode-of-action and the
affected physiological endpoints. Pharmacodynamics also encompasses drug interactions with off-targets that induce side effects and potentially toxic responses.

Drug Targets
A drug target can be defined as a molecular structure that undergoes a specific interaction with a pharmaceutical and the interaction has a connection with a clinical
effect. The great majority of drug targets are proteins, and these can be classified
into different broad functional groups, the most important of which are enzymes,
receptors, ion channels and transporters. Drugs are selected or designed to induce
their intended clinical effect and to cause a minimal amount of side effects at relatively low doses.
In Gunnarsson et al. [6], we performed an ortholog prediction of all human drug
targets available in Drugbank [71, 72]. The study shows that roughly 80% of the
human drug targets are conserved in the aquatic vertebrates Xenopus tropicalis,
Danio rerio, and Gasterosteus aculeatus, roughly 60% are conserved in the invertebrates Daphnia pulex, Drosophila melanogaster, and Caenorhabditis elegans, and
less than 40% are conserved in the plant Arabidopsis thaliana and the alga
Chlamydomonas reinhardtii (Fig. 3). The protein sequence similarity of the
orthologs show a similar pattern, with the vertebrate orthologs sharing about 60%
sequence similarity with the human drug targets, whereas the invertebrates have
about 40% sequence similarity (Fig. 3).
Another interesting pattern was revealed when different functional groups of the
drug targets were analyzed. Receptors constitute an important group of validated
pharmacological targets, and as much as 40% of all FDA approved drugs elicit their
effects through receptors [73]. We show that the proportion of receptors decrease
significantly while the proportion of enzymes significantly increase with the evolutionary distance to man (Fig. 4). Thus, the choice of environmental test species is
particularly important for drugs that have receptors as targets, and effects on nonvertebrates would not be expected for most drugs from this class. Enzyme drug
targets, on the other hand, were more ubiquitously present. Generally speaking, one
could therefore expect drugs targeting enzymes to affect a wider range of species,

Environmental Comparative Pharmacology: Theory and Application

99

Fig. 3 The number of evolutionary conserved drug target proteins differs between species. The
figure shows on the x-axis the median sequence similarity to the corresponding human drug targets
and on the y-axis the number of conserved drug targets for 16 nontarget species. The boxes indicate
25 and 75% quantiles (Reprinted from Gunnarsson et al. [6] with permission from the American
Chemical Society)

including invertebrates. The full list of ortholog predictions of 1,318 human drug
targets in 16 species is available as supplementary information together with the
published paper [6].
As already mentioned, the presence of a drug target ortholog in a nontarget species does not guarantee that a functional interaction with the drug can occur.
However, in Gunnarsson et al. [6], we presented literature data supporting that an
ortholog prediction often can indicate the ability of a conserved drug target protein
to interact with the human drug. A more precise prediction of a potential drug target
interaction might be possible with better knowledge about drug binding domains.
Sakharkar et al. [74] summarized protein family (Pfam) domains related to druggable domains. However, many of these domains are too general to provide additional useful information to predict the ability of a protein to interact with a drug in
an evolutionarily distant species. The database Supersite has recently been released.
It contains 3D protein structures and ligand-binding site information for over 1,300
medically active compounds, as well as some evolutionary information [75]. Given
that the structure of the proteins in nontargets species can be accurately modeled,
this could improve current predictions.

100

L. Gunnarsson et al.

Fig. 4 The evolutionary conservation depends on the type of drug target. The figure shows evolutionarily conserved human drug targets in 16 nontarget species divided into functional categories
(Reprinted from Gunnarsson et al. [6], with permission from the American Chemical Society)

Mode-of-Action
An interaction between a drug and its primary target generally leads to a number of
subsequent reactions within the cell. This course of events can be described by one
or several pathways, chains of causal biochemical reactions and interactions. Even
though pathways, strictly speaking, are theoretical models, they can provide means
for identification of affected physiological endpoints. The pathways affected by
pharmaceuticals have not been as extensively studied as the drug targets themselves.
An absolute identification of the therapeutic mode of action is not a requirement to
market a new drug. Nevertheless, there is a considerable amount of human drug
targets associated to pathways that at least partly describe the mode-of-action of the
drug [76, 77].
The evolutionary process of a pathway is complex and not yet fully understood.
Many of the pathways currently described are based on mammalian biology and
can therefore be completely different or even nonexistent in nontarget species.
Stimulation of the same target protein in two different species may indeed lead to
very different outcomes depending on the physiology of the species. Knowledge
about the evolutionary conservation of a pathway can therefore be helpful for
identification of physiological endpoints affected by drugs in nontarget species.

Environmental Comparative Pharmacology: Theory and Application

101

Pathway Prediction
As for drug targets, predictions of conserved pathways are based on identification of
orthologs, but for a pathway, information on all its members needs to be combined.
The Reactome Project (www.reactome.org) uses OrthoMCL to predict the conserved counterparts of human pathways in 22 species, including mouse, rat and
chicken, but also more distant species such as worm, fly and yeast. The information
on pathways in nontraditional model organisms is however sparse.
The activation of a pathway often results, directly or indirectly, in a change in
gene expression profiles. This is also true for pathways mediated by drug target
proteins, and if the mode-of action is evolutionarily conserved between human and
nontarget species, then it is likely that their transcriptional responses also have similarities. Large-scale gene expression analysis can thus provide information regarding the pharmacodynamics of drugs. These experiments are typically performed
using DNA microarrays, which have been widely available for model species such
as mouse, rat, and zebrafish. The introduction of more versatile platforms, such as
Agilent or Nimblegen custom microarrays (www.aglient.com; www.nimblegen.
com) and febit RT-analyzer (www.febit.de), has made the microarray technique
more applicable to nonmodel organisms without a well-categorized genome [78,
79]. Even though the quality of data generated from microarrays was initially questioned [80, 81], the technique has matured and is today approaching the higher
sensitivity and accuracy of methods such as quantitative PCR and high-throughput
mRNA sequencing [8284].

Microarray Analysis
Compared to more focused approaches, microarrays can be used in an explorative
manner, i.e., without a clear hypothesis on how the nontarget species could be
affected by the investigated drug. Some combinations of species and drugs will
inevitably cause toxicity in unexpected ways via novel mechanisms, even at rather
low concentrations. Thus, microarray analysis is an important complement of
mode-of-action based tests relying on our ability to a priori identify relevant and
sensitive endpoints.
Gene expression data generated by microarrays can be used to identify affected
pathways. This can be used both to confirm hypotheses based on mammalian pharmacodynamics and to discover novel and unexpected effects. Heckmann et al. [85]
showed, for example, that several of the genes involved in eicosanoid (for instance
prostaglandin) metabolism were differentially expressed in D. magna exposed to
the NSAID ibuprofen. In mammals, NSAIDs inhibit prostaglandin synthesis, and
thus the mode-of-action for this drug is partly conserved. However, the physiological endpoints are different, as differences in eicosanoid metabolism may have direct
consequences for reproduction in crustaceans.

102

L. Gunnarsson et al.

Pathway analysis can thus be used to understand and interpret gene expression
data in a pharmacological context. In general, this kind of analysis is performed by
testing whether the genes from a given pathway have a higher tendency to be differentially expressed than other genes on the microarray [86]. The procedure is
usually repeated for all pathways in a database, such as KEGG, and the models best
describing the gene expression pattern can thus be identified. Several software
applications performing pathway analysis of gene expression data have been developed, e.g., GenMAPP [87], Pathway Miner [88] and the SkyPainter tool at the
Reactome Project [89], among which the latter allows for the identification of evolutionary conserved human pathways in microarray data from other species.
Even though pathway analysis has the ability to provide information regarding the
molecular-level mechanistic of drugtarget interactions, there are a number of potential issues that can make interpretations difficult. All proteins associated with a pathway affected by a certain drug will generally not be regulated at the transcriptional
level. Indeed, later regulatory steps, such as post-transcriptional and post-translational modifications, may be more suitable in many cases, and such actions cannot be
directly detected by microarray-based gene expression analysis. A pathway can also
have one or a few specific bottlenecks, and hence increasing or decreasing the abundance of these proteins might be enough to change the activity of the entire pathway.
Testing the entire pathway for upregulation or downregulation at the transcriptional
level may thus not be the most appropriate solution. Another significantly hampering
factor for pathway analysis in nonmodel organism is the lack of species-specific
information about protein function, interaction, and thus pathways [89].

Transcription-Factor Proteins
Microarray data can also be used for identification of the activation or inhibition of
transcription factor proteins. The synthetic estrogen used in contraceptives, EE2,
binds to the estrogen receptor, which is consequently relocated into the nucleus and
initiates the transcription of hundreds of genes. This mechanism is conserved in all
vertebrates [90], and the gene expression patterns produced also show similarities in
many species [91]. Many pathways activate transcription of genes, which then can
be measured using microarrays. The regulation of gene expression is generally performed by transcription factor proteins, which can initiate transcription by binding
to short DNA sequences called cis-regulatory elements. Information regarding the
cis-regulatory elements within the promoters of differentially expressed genes can
therefore be used to untangle which transcription factors and biological process
are responsible for the observed changes in gene expression [92]. Hence, if a
cis-regulatory element is overrepresented among the differentially expressed genes,
it is likely that the corresponding transcription factor regulates the transcription of
these genes. However, the identification of cis-regulatory elements is dependent on
information on the promoter regions, which is often lacking even for species with
completely sequenced genomes [93]. The overrepresentation of cis-regulatory

Environmental Comparative Pharmacology: Theory and Application

103

elements among differentially expressed genes can therefore yet only be performed
reliably in a limited number of model species. The applicability of these approaches
will, however, continue to improve as the number of sequenced species increases.
The evolutionary conservation of the mode-of-action of a specific drug can provide valuable information and facilitate the extrapolation of pharmacodynamics to
nontarget species. Orthology predictions of drug targets and pathways are fundamentally connected to genomics data, and therefore continue to improve as more
and more nontarget species become sequenced. Increased knowledge about protein
function, interactions and the general physiology are equally important for reliable
comparative pharmacodynamics. However, molecular-level responses identified by
large scale comparative pharmacology approaches can aid in the identification of
relevant and sensitive endpoints. Combining those endpoints with measurements of
adverse effects can provide valuable information about possible chronic consequences of pharmaceutical exposure and for population health.

Conclusion
The currently available ecotoxicity data for most pharmaceuticals are insufficient, but
the vast knowledge base derived in drug development could provide insights on possible effects of residual pharmaceuticals in exposed nontarget species. The rapid
advances in genomics have opened up new possibilities to develop the field of comparative pharmacology in species of interest for ecological risk assessments. Accurate
predictions of the conservation of proteins and pathways may provide important input
to our understanding of pharmacokinetics and pharmacodynamics in nontarget species. However, a comprehensive understanding of the physiology of the exposed
wildlife species is equally important to make well-founded predictions, and for the
vast majority of species, this is currently a hampering factor. Nevertheless, large-scale
comparisons of conserved proteins and pathways can still aid in the prioritization of
which drugs need further assessment of their environmental risks, which organisms
should be prioritized for testing, and what endpoints are most appropriate.
Acknowledgments The authors wish to thank the two anonymous reviewers for valuable
comments and the Foundation for Strategic Environmental Research (MISTRA) and the Swedish
Research Council (VR) for financial support.

References
1. Garric J, Vollat B, Duis K, Pery A, Junker T, Ramil M, Fink G, Ternes TA (2007) Effects of the
parasiticide ivermectin on the cladoceran Daphnia magna and the green alga Pseudokirchneriella
subcapitata. Chemosphere 69:903910
2. Kidd KA, Blanchfield PJ, Mills KH, Palace VP, Evans RE, Lazorchak JM, Flick RW (2007)
Collapse of a fish population after exposure to a synthetic estrogen. Proc Natl Acad Sci U S A
104:88978901

104

L. Gunnarsson et al.

3. Zeilinger J, Steger-Hartmann T, Maser E, Goller S, Vonk R, Lange R (2009) Effects of synthetic

gestagens on fish reproduction. Environ Toxicol Chem 28:26632670
4. Winter MJ, Owen SF, Murray-Smith RM, Panter GH, Hetheridge MJ, Kinter LB (2009) Using
data from drug discovery and development to aid the aquatic environmental risk assessment of
human pharmaceuticals: concepts, considerations and challenges. Integr Environ Assess
Manag 6:3851
5. Ankley GT, Brooks BW, Huggett DB, Sumpter JP (2007) Repeating history: pharmaceuticals
in the environment. Environ Sci Technol 41:82118217
6. Gunnarsson L, Jauhiainen A, Kristiansson E, Nerman O, Larsson DGJ (2008) Evolutionary
conservation of human drug targets in organisms used for environmental risk assessments.
Environ Sci Technol 42:58075813
7. Kostich MS, Lazorchak JM (2008) Risks to aquatic organisms posed by human pharmaceutical
use. Sci Total Environ 389:329339
8. Breitholtz M, Bengtsson BE (2001) Oestrogens have no hormonal effect on the development
and reproduction of the harpacticoid copepod Nitocra spinipes. Mar Pollut Bull 42:879886
9. ECETOC (2007) Intelligent testing strategies in ecotoxicology: mode-of action approach for
specifically acting chemicals. ECETOC Technical Report No.102. http://www.ecetoc.org/
index.php?mact=MCSoap,cntnt01,details,0&cntnt01by_category=16&cntnt01template=displ
ay_list_science&cntnt01document_id=281&cntnt01returnid=101. Accessed Sept 2008
10. Kreke N, Dietrich DR (2008) Physiological endpoints for potential SSRI interactions in fish.
Crit Rev Toxicol 38:215247
11. Owen SF, Giltrow E, Huggett DB, Hutchinson TH, Saye J, Winter MJ, Sumpter JP (2007)
Comparative physiology, pharmacology and toxicology of beta-blockers: mammals versus
fish. Aquat Toxicol 82:145162
12. Brain RA, Hanson ML, Solomon KR, Brooks BW (2008) Aquatic plants exposed to pharmaceuticals: effects and risks. Rev Environ Contam Toxicol 192:67115
13. EMEA (2006) Guideline on the environmental Risk Assessment of Medical Products for
Human Use. The Committee for Medicinal Products for Human Use (CHMP), European
Medicines Agency, CHMP/SWP/4447/00. http://www.tga.gov.au/docs/pdf/euguide/swp/
444700en.pdf. Accessed Sept 2009
14. FDA (1998) Guidance for environmental assessment of human drug and biologics applications. Food and Drug Administration, Center for Drug Evaluation and Research, Center for
Biologics Evaluation and Research. http://www.fda.gov/cder/guidance/1730fnl.pdf. Accessed
Sept 2009
15. Maack G, Adler N, Bachmann J, Ebert I, Hickmann S, Kster A, B. R (2010) Environmental
risk assessment of medicinal products for human use: does the risk assessment reflect the reality?
SETAC Europe: 20th annual meeting. http://www.eventure-online.com/eventure/publicAbstractView.do;jsessionid=abcrY0Urm3inMfjVSloOs?id=115911&congressId=3358 .
Accessed June 2010
16. Holbrook JD, Sanseau P (2007) Drug discovery and computational evolutionary analysis.
Drug Discov Today 12:826832
17. Bork P, Dandekar T, Diaz-Lazcoz Y, Eisenhaber F, Huynen M, Yuan Y (1998) Predicting function: from genes to genomes and back. J Mol Biol 283:707725
18. Li L, Stoeckert CJ Jr, Roos DS (2003) OrthoMCL: identification of ortholog groups for
eukaryotic genomes. Genome Res 13:21782189
19. Chen F, Mackey AJ, Vermunt JK, Roos DS (2007) Assessing performance of orthology detection strategies applied to eukaryotic genomes. PLoS One 2:e383
20. Dolinski K, Botstein D (2007) Orthology and functional conservation in eukaryotes. Annu Rev
Genet 41:465507
21. Felsenstein J (1988) Phylogenies from molecular sequences: inference and reliability. Annu
Rev Genet 22:521565
22. Altenhoff AM, Dessimoz C (2009) Phylogenetic and functional assessment of orthologs inference projects and methods. PLoS Comput Biol 5:e1000262

Environmental Comparative Pharmacology: Theory and Application

105

23. Gabaldon T (2008) Large-scale assignment of orthology: back to phylogenetics? Genome Biol
9:235
24. Vilella AJ, Severin J, Ureta-Vidal A, Heng L, Durbin R, Birney E (2009) EnsemblCompara GeneTrees:
complete, duplication-aware phylogenetic trees in vertebrates. Genome Res 19:327335
25. Dehal PS, Boore JL (2006) A phylogenomic gene cluster resource: the Phylogenetically
Inferred Groups (PhIGs) database. BMC Bioinformatics 7:201
26. Pearson W (2004) Finding protein and nucleotide similarities with FASTA. Curr Protoc
Bioinformatics Chapter 3:Unit3.9
27. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997)
Gapped BLAST and PSI-BLAST: a new generation of protein database search programs.
Nucleic Acids Res 25:33893402
28. Ding G, Sun Y, Li H, Wang Z, Fan H, Wang C, Yang D, Li Y (2008) EPGD: a comprehensive
web resource for integrating and displaying eukaryotic paralog/paralogon information. Nucleic
Acids Res 36:D255D262
29. Tatusov RL, Koonin EV, Lipman DJ (1997) A genomic perspective on protein families. Science
278:631637
30. Raymond J, Zhaxybayeva O, Gogarten JP, Gerdes SY, Blankenship RE (2002) Whole-genome
analysis of photosynthetic prokaryotes. Science 298:16161620
21. Remm M, Storm CE, Sonnhammer EL (2001) Automatic clustering of orthologs and in-paralogs from pairwise species comparisons. J Mol Biol 314:10411052
32. van Dongen S (2000) Graph clustering by flow simulation. PhD thesis, University of Utrecht,
Utrecht
33. Enright AJ, Van Dongen S, Ouzounis CA (2002) An efficient algorithm for large-scale detection of protein families. Nucleic Acids Res 30:15751584
34. Randall DJ, Connell DW, Yang R, Wu SS (1998) Concentrations of persistent lipophilic compounds in fish are determined by exchange across the gills, not through the food chain.
Chemosphere 37:12631270
35. Barnes RSK, Calow P, Olive PJW, Golding DW, Spicer JI (2001) The invertebratesa synthesis. Blackwell Science, Malden
36. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ (1997) Experimental and computational
approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev 46:325
37. Huggett BD, Cook CJ, Ericson FJ, Williams RT (2003) A theoretical model for utilizing mammalian pharmacology and safety data to prioritize potential impacts of human pharmaceuticals
to fish. Hum Ecol Risk Assess 9:17891799
38. Brown JN, Paxeus N, Forlin L, Larsson DGJ (2007) Variations in bioconcentration of human
pharmaceuticals from sewage effluents into fish blood plasma. Environ Toxicol Pharmacol
24:267274
39. Owen SF, Huggett DB, Hutchinson TH, Hetheridge MJ, Kinter LB, Ericson JF, Sumpter JP
(2009) Uptake of propranolol, a cardiovascular pharmaceutical, from water into fish plasma
and its effects on growth and organ biometry. Aquat Toxicol 93:217224
40. Fick J, Lindberg RH, Parkkonen J, Arvidsson B, Tysklind M, Larsson DGJ (2010) Therapeutic
levels of levonorgestrel detected in blood plasma of fish: results from screening rainbow trout
exposed to treated sewage effluents. Environ Sci Technol 44:26612666
41. Fick J, Lindberg RH, Tysklind M, Larsson DGJ (2010) Predicted critical environmental
concentrations for 500 pharmaceuticals. Regul Toxicol Pharmacol 58(3):516523
42. Roos V, Gunnarsson L, Fick J, Larsson DGJ, Ruden C (2012) Prioritising pharmaceuticals for
environmental risk assessment: Towards adequate and feasible first-tier selection. Science of
the Total Environment 421422:102110
43. Rendal C, Kusk KO, Trapp S (2011) The effect of pH on the uptake and toxicity of the bivalent
weak base chloroquine tested on Salix viminalis and Daphnia magna. Environ Toxicol Chem
30:354359

106

L. Gunnarsson et al.

44. Valenti TW, Perez-Hurtado P, Chambliss KC, Brooks BW (2009) Aquatic toxicity of sertraline
to Pimephales promelas at environmentally relevant surface water pH. Environ Toxicol Chem
28:268594
45. Dobson PD, Kell DB (2008) Carrier-mediated cellular uptake of pharmaceutical drugs: an
exception or the rule? Nat Rev Drug Discov 7:205220
46. Benet LZ, Kroetz DL, Sheiner LB (1996) Pharmacokinetics. The dynamics of drug absorption
distribution and elimination. McGraw Hill, New York.
47. Kratochwil NA, Huber W, Muller F, Kansy M, Gerber PR (2002) Predicting plasma protein
binding of drugs: a new approach. Biochem Pharmacol 64:13551374
48. Gray JE, Doolittle RF (1992) Characterization, primary structure, and evolution of lamprey
plasma albumin. Protein Sci 1:289302
49. Kleinow KM, Nichols JW, Hayton WL, Mckim JM, Barron MG (2008) The toxicokinetics in
fish. In: Di Gulio RT, Hinton DE (eds) The toxicology of fishes. CRC Press, Boca Raton
50. Bobe J, Guiguen Y, Fostier A (2010) Diversity and biological significance of sex hormonebinding globulin in fish, an evolutionary perspective. Mol Cell Endocrinol 316:6678
51. Solange M-Q, Geoffrey LH (2008) Sex Hormone-Binding Globulin in Fish Gills is a Portal for
Sex Steroids Breached by Xenobiotics. Endocrinology 149:42694275
52. Miguel-Queralt S, Hammond GL (2008) Sex hormone-binding globulin in fish gills is a portal
for sex steroids breached by xenobiotics. Endocrinology 149:42694275
53. Iyanagi T (2007) Molecular mechanism of phase I and phase II drug-metabolizing enzymes:
implications for detoxification. Int Rev Cytol 260:35112
54. Werck-Reichhart D, Feyereisen R (2000) Cytochromes P450: a success story. Genome Biol
1(6):reviews3003
55. Graham MJ, Lake BG (2008) Induction of drug metabolism: species differences and toxicological relevance. Toxicology 254:184191
56. Yang X, Zhang B, Molony C, Chudin E, Hao K, Zhu J, Gaedigk A, Suver C, Zhong H, Leeder
JS et al (2010) Systematic genetic and genomic analysis of cytochrome P450 enzyme activities
in human liver. Genome Res. doi:10.1101/gr.103341.109
57. Nebert DW, Gonzalez FJ (1987) P450 genes: structure, evolution, and regulation. Annu Rev
Biochem 56:945993
58. Schlenk D, Celander M, Gallagher EP, George S, James M, Kullman SW, van den Hurk P,
Willett K (2008) Biotransformation in Fishes. In: Hinton DE (ed) Di Giulio DR. The Toxicology
of Fishes CRC Press, Boca Raton (FL)
59. Fink-Gremmels J (2008) Implications of hepatic cytochrome P450-related biotransformation
processes in veterinary sciences. Eur J Pharmacol 585:502509
60. Prasad JC, Goldstone JV, Camacho CJ, Vajda S, Stegeman JJ (2007) Ensemble modeling of
substrate binding to cytochromes P450: analysis of catalytic differences between CYP1A
orthologs. Biochemistry 46:26402654
61. Mattingly CJ, Toscano WA (2001) Posttranscriptional silencing of cytochrome P4501A1
(CYP1A1) during zebrafish (Danio rerio) development. Dev Dyn 222:645654
62. Lee MD, Ayanoglu E, Gong L (2006) Drug-induced changes in P450 enzyme expression at the
gene expression level: a new dimension to the analysis of drug-drug interactions. Xenobiotica
36:10131080
63. Hegelund T, Ottosson K, Radinger M, Tomberg P, Celander MC (2004) Effects of the antifungal imidazole ketoconazole on CYP1A and CYP3A in rainbow trout and killifish. Environ
Toxicol Chem 23:13261334
64. Blanchard RL, Freimuth RR, Buck J, Weinshilboum RM, Coughtrie MW (2004) A proposed
nomenclature system for the cytosolic sulfotransferase (SULT) superfamily. Pharmacogenetics
14:199211
65. Mackenzie PI, Bock KW, Burchell B, Guillemette C, Ikushiro S, Iyanagi T, Miners JO, Owens
IS, Nebert DW (2005) Nomenclature update for the mammalian UDP glycosyltransferase
(UGT) gene superfamily. Pharmacogenet Genomics 15:677685
66. Blanchette B, Feng X, Singh BR (2007) Marine glutathione S-transferases. Mar Biotechnol
(NY) 9:513542

Environmental Comparative Pharmacology: Theory and Application

107

67. Rattner BA, Whitehead MA, Gasper G, Meteyer C, Link WA, Taggart MA, Meharg AA, Pattee
OH, Pain DJ (2008) Apparent tolerance of turkey vultures (Cathartes aura) to the non-steroidal
anti-inflammatory drug Diclofenac. Environ Toxicol Chem 27(11):23415
68. van de Waterbeemd H, Gifford E (2003) ADMET in silico modelling: towards prediction paradise? Nat Rev Drug Discov 2:192204
69. Mohan CG, Gandhi T, Garg D, Shinde R (2007) Computer-assisted methods in chemical toxicity
prediction. Mini Rev Med Chem 7:499507
70. Chang C, Duignan DB, Johnson KD, Lee PH, Cowan GS, Gifford EM, Stankovic CJ, Lepsy
CS, Stoner CL (2008) The development and validation of a computational model to predict rat
liver microsomal clearance. J Pharm Sci 98(8):285767
71. Wishart DS, Knox C, Guo AC, Shrivastava S, Hassanali M, Stothard P, Chang Z, Woolsey J
(2006) DrugBank: a comprehensive resource for in silico drug discovery and exploration.
Nucleic Acids Res 34:D668D672
72. Wishart DS, Knox C, Guo AC, Cheng D, Shrivastava S, Tzur D, Gautam B, Hassanali M
(2008) DrugBank: a knowledgebase for drugs, drug actions and drug targets. Nucleic Acids
Res 36:D901D906
73. Overington JP, Al-Lazikani B, Hopkins AL (2006) How many drug targets are there? Nat Rev
Drug Discov 5:993996
74. Sakharkar MK, Sakharkar KR, Pervaiz S (2007) Druggability of human disease genes. Int
J Biochem Cell Biol 39:11561164
75. Bauer RA, Gunther S, Jansen D, Heeger C, Thaben PF, Preissner R (2009) SuperSite: dictionary of metabolite and drug binding sites in proteins. Nucleic Acids Res 37:D195D200
76. Gong L, Owen RP, Gor W, Altman RB, Klein TE (2008) PharmGKB: an integrated resource
of pharmacogenomic data and knowledge. Curr Protoc Bioinformatics Chapter 14:Unit14 17
77. Frolkis A, Knox C, Lim E, Jewison T, Law V, Hau DD, Liu P, Gautam B, Ly S, Guo AC et al
(2010) SMPDB: the Small Molecule Pathway Database. Nucleic Acids Res 38:D480D487
78. Garcia-Reyero N, Adelman I, Liu L, Denslow N (2008) Gene expression profiles of fathead
minnows exposed to surface waters above and below a sewage treatment plant in Minnesota.
Mar Environ Res 66:134136
79. Kristiansson E, Asker N, Forlin L, Larsson DGJ (2009) Characterization of the Zoarces
viviparus liver transcriptome using massively parallel pyrosequencing. BMC Genomics
10:345
80. Woo Y, Affourtit J, Daigle S, Viale A, Johnson K, Naggert J, Churchill G (2004) A comparison
of cDNA, oligonucleotide, and Affymetrix GeneChip gene expression microarray platforms.
J Biomol Tech 15:276284
81. Kristiansson E, Sjogren A, Rudemo M, Nerman O (2005) Weighted analysis of paired microarray experiments. Stat Appl Genet Mol Biol 4:Article30
82. Kuo WP, Liu F, Trimarchi J, Punzo C, Lombardi M, Sarang J, Whipple ME, Maysuria M,
Serikawa K, Lee SY et al (2006) A sequence-oriented comparison of gene expression measurements across different hybridization-based technologies. Nat Biotechnol 24:832840
83. Shi L, Reid LH, Jones WD, Shippy R, Warrington JA, Baker SC, Collins PJ, de Longueville F,
Kawasaki ES, Lee KY et al (2006) The MicroArray Quality Control (MAQC) project shows
inter- and intraplatform reproducibility of gene expression measurements. Nat Biotechnol
24:11511161
84. t Hoen PA, Ariyurek Y, Thygesen HH, Vreugdenhil E, Vossen RH, de Menezes RX, Boer JM,
van Ommen GJ, den Dunnen JT (2008) Deep sequencing-based expression analysis shows
major advances in robustness, resolution and inter-lab portability over five microarray platforms. Nucleic Acids Res 36:141
85. Heckmann LH, Sibly RM, Connon R, Hooper HL, Hutchinson TH, Maund SJ, Hill CJ,
Bouetard A, Callaghan A (2008) Systems biology meets stress ecology: linking molecular and
organismal stress responses in Daphnia magna. Genome Biol 9:R40
86. Curtis RK, Oresic M, Vidal-Puig A (2005) Pathways to the analysis of microarray data. Trends
Biotechnol 23:429435

108

L. Gunnarsson et al.

87. Salomonis N, Hanspers K, Zambon AC, Vranizan K, Lawlor SC, Dahlquist KD, Doniger SW,
Stuart J, Conklin BR, Pico AR (2007) GenMAPP 2: new features and resources for pathway
analysis. BMC Bioinformatics 8:217
88. Pandey R, Guru RK, Mount DW (2004) Pathway Miner: extracting gene association networks
from molecular pathways for predicting the biological significance of gene expression microarray data. Bioinformatics 20:21562158
89. Matthews L, Gopinath G, Gillespie M, Caudy M, Croft D, de Bono B, Garapati P, Hemish J,
Hermjakob H, Jassal B et al (2009) Reactome knowledgebase of human biological pathways
and processes. Nucleic Acids Res 37:D619D622
90. Thornton JW, Need E, Crews D (2003) Resurrecting the ancestral steroid receptor: ancient
origin of estrogen signaling. Science 301:17141717
91. Gunnarsson L, Kristiansson E, Forlin L, Nerman O, Larsson DG (2007) Sensitive and robust
gene expression changes in fish exposed to estrogena microarray approach. BMC Genomics
8:149
92. Kristiansson E, Thorsen M, Tamas MJ, Nerman O (2009) Evolutionary forces act on promoter
length: identification of enriched cis-regulatory elements. Mol Biol Evol 26:12991307
93. Schmid CD, Perier R, Praz V, Bucher P (2006) EPD in its twentieth year: towards complete
promoter coverage of selected model organisms. Nucleic Acids Res 34:D82D85

A Look Backwards at Environmental Risk

Assessment: An Approach to Reconstructing
Ecological Exposures
David Lattier, James M. Lazorchak, Florence Fulk, and Mitchell Kostich

Introduction
The primary goal for environmental protection is to eliminate or minimize the
exposure of humans and ecosystems to potential contaminants. With the number of
environmental contaminants increasing annually, more than 2,000 new chemicals
are manufactured or imported each year for use in the USA, understanding the
sources of contaminants, the movement of contaminants through environmental
media, and the contact of contaminants with humans and ecosystems is critical to
advancing environmental protection in the USA. A shift in emphasis from detection
of chemical exposure to reconstruction of exposure scenarios will enhance the ability to assess the effectiveness of current environmental regulations and to improve
environmental risk assessment for both humans and ecosystems. Exposure reconstruction is a concept that can guide this shift in research focus. Exposure
reconstruction, as defined in this chapter, is the characterization of exposures,
environmental concentrations, and/or sources from internal biological measurements
that are used to inform environmental decision-making (Fig. 1).

This document has been reviewed in accordance with US Environmental Protection Agency policy
and approved for publication. Approval does not signify that the contents necessarily reflect the
views or policies of the Agency nor does mention of trade names or commercial products constitute endorsement or recommendation for use.
D. Lattier (*) J.M. Lazorchak F. Fulk M. Kostich
National Exposure Research Laboratory, Ecological Exposure Research Division,
US Environmental Protection Agency, Office of Research and Development,
26 W. Martin Luther King Drive, Cincinnati, OH 45268, USA
e-mail: [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_6, Springer Science+Business Media, LLC 2012

109

110

D. Lattier et al.

Fig. 1 Exposure reconstruction links an internal biological measurement to an external exposure

and environmental concentration.

Background
Currently, information on the exposure to humans and ecosystems by environmental contaminants is primarily limited to biomonitoring studies which mainly collect
data on the occurrence of a predetermined list of contaminants in environmental and
biological samples (i.e., urine, blood, and tissues). Typically, the goals of national
scale biomonitoring studies are to detect contaminant exposure and establish baseline measures, monitor exposure trends and identify geographic hotspots. Little to
no information is collected that can contribute to elucidating where the contaminants originated, how they were transported through the environment, what pathways the contaminant took to reach living organisms (i.e., drinking water, food
consumption, ambient air) or the routes of exposure by which people and other
organisms came into contact with contaminants in question (i.e., ingestion, inhalation, adsorption). These studies are fundamentally unlike investigations of occupational exposure, wherein epidemiological events, usually industrial in nature [1],
result in adverse health effects to specific members of a well-defined human population. Retrospective studies in human health, particularly relating to industrial exposures, presumes that the causative agent(s) are known and based on personal usage,
proximity to exposures, historical presence and supportable interviews with affected
individuals. Human health investigations in the realm of radiation dose metrics [2],
and associated risk assessment such as inhalation studies [3, 4] lend support for
efforts in exposure reconstruction. Epidemiological retrospectives, the underpinning of which is exposure reconstruction, are made possible by the limitless compilation of knowledge about a well characterized species. This cumulative assessment
includes behavior and habits of individuals, psychology, medical histories, work
histories, and the inestimable physiological, toxicological, cellular, and whole
genome information. Ecological exposure reconstructionillustrated in following
pagesis predicated on identical specific aims; however, this undertaking is entirely
deficient in the extensive knowledge bases readily accessible to human health
investigators.
The Centers for Disease Control and Prevention (CDC) published the First,
Second, and Third Reports on Human Exposure to Environmental Chemicals [5,
6]. The 2005 report provides exposure biomonitoring data for a representative
sample of the US population and targets 148 environmental chemicals commonly
found in the environment including lead, pesticides, herbicides, phthalates,
polychlorinated biphenyls (PCBs), and polycyclic aromatic hydrocarbons (PAHs).

A Look Backwards at Environmental Risk Assessment

111

Of these 148 chemicals, only 25 have established reference values for safe levels
of exposure. The current CDC survey gathers very limited exposure data and
does not allow for exposure estimates by location or permit identification of
sources of contaminants. The USEPA National Study of Chemical Residues in
Lake Fish Tissue [7] conducted in 2005 faced similar challenges. One goal of this
study was to develop national estimates of the mean levels of 268 persistent, bioaccumulative and toxic chemicals (PBTs) in fish, establishing a national baseline for
tracking reductions in PBTs in freshwater fish as a result of pollution control
activities. Information on the potential sources of the chemicals or the timing of
the exposures of fish populations to PBTs is, by approach and design, absent from
the study.
A major finding of a recent 2006 National Research Council (NRC) report on
Human Biomonitoring for Environmental Chemicals [8] is the need for more extensive exposure information. The NRC report states that the collection of biomonitoring data and development of biomarkers has outpaced our ability to interpret the
data with respect to both potential health effects and in retrospective source tracking. Biomonitoring is an important tool for understanding the linkages between
external chemical exposures, internal doses, and potential health outcomes. However,
biomarker data independently shows only that humans or organisms were exposed
to a chemical at some point in time.

Exposure Reconstruction
Exposure reconstruction is the characterization of exposures, environmental concentrations, and/or sources from internal biological measurements to inform environmental decision-making (Fig. 1). The ability to reconstruct exposure scenarios
requires a basic understanding of the relationship between an external exposure
concentration and an internal biological measurement. The quantitative relationship
between human and ecosystem exposures and biomarkers are estimated using a
number of computational tools including physiologically based pharmacokinetic
(PBPK) models and empirically based regression models. Exposure reconstruction
as a research concept can guide the development of new biomarkers and the design
of future biomonitoring studies and ultimately provide a critical component of environmental protection as well as the identification of important sources, pathways,
and routes of exposure. There are two broad areas of research to support exposure
reconstruction: (1) to leverage existing biomonitoring studies by collection of additional data and enhanced modeling techniques to aid the reconstruction of exposures and (2) to develop new biomarkers that can inform the what, when, where, and
how much exposure.
A recent publication [9] provides an example of the first approach. The goal of
this study was to evaluate plausible exposure scenarios of humans to chloroform
from the activity of showering, consistent with measured concentrations of

112

D. Lattier et al.

chloroform in human biomarkers. The authors approach included multiple steps:

combined a PBPK model with an exposure model for showering to estimate the
intake concentration of chloroform; evaluated the combined model using data from
existing biomonitoring studies; developed potential exposure regimens based on
typical levels of chloroform in residential water and accounting for multiple exposure routes (i.e., inhalation, dermal, ingestion); estimated distributions of exposure
consistent with measured levels of chloroform in human biomarkers by a reverse
dosimetry approach with the combined model. The foregoing study highlights the
capability and the difficulty in reconstructing exposures from internal biological
measurements. The authors were able to demonstrate that inhalation and dermal
exposure substantially contributed to total chloroform exposure. However, sources
of variability in model output from exposure conditions and from pharmacokinetics
were significant.
The second research area that will further the capability to reconstruct exposure scenarios is the development of biomarkers of exposure. A biomarker of
exposure, as defined in the NRC report, is the chemical or its metabolite or the
product of an interaction between a chemical and some target molecule or cell that
is measured in a compartment or an organism. Beyond the NRC definition, in
order for an exposure biomarker to be useful in reconstructing exposures it must
possess a number of characteristics: (1) specificity; the biomarker must identify a
specific compound or class of compounds, if possible by way of mode of action
(MOA), (2) sensitivity; the biomarker must be capable of measuring exposures
above background levels and at concentrations that are environmentally relevant,
(3) reproducibility; biomarker values are reproducible for both environmental
sampling and laboratory analysis, (4) validated concentration response; the relationship between the biomarker and the external concentration is validated across
a concentration gradient in single or multiple species, the relationship should be
validated in both laboratory and field studies which account for physical conditions of the exposure, and (5) knowledge of the exposure kinetics; the biomarker
or set of biomarkers are accompanied by an understanding of the kinetics of the
exposure (i.e., biomarker level relative to the concentration and timing of exposure). These five characteristics establish a gold standard for biomarkers that will
provide critical information to identify the contaminant, the exposure concentration and the timing of exposure. The NRC report recognized that new technologies in biomonitoring have the potential to transform the nations capacity to track
exposure to pollutants and understand their impact on human (and ecosystem)
health [8]. Biomarkers of exposure developed using advances in genomics technology have the capability to meet the criteria for informing exposure reconstruction. In the area of ecosystem exposure, biomarkers for exposure of aquatic
organisms to estrogenic endocrine disrupting chemicals (EDCs) were developed
using genomic endpoints [10, 11]. In combination, an upregulated gene followed
by production of a protein [12] allow for discrimination between an ongoing
exposure (hours to days) from a recent exposure (days to weeks). Applied within
the spatial context of a watershed, these biomarkers can inform the timing and the
source of exposures to EDCs.

A Look Backwards at Environmental Risk Assessment

113

The Challenges of Reconstructing Ecological Exposures

An ecological exposure reconstruction program will aim to accurately identify conditions under which target or other indigenous species in aquatic ecosystems might
have been receptors of geospatial point or nonpoint exposures from xenobiotic or
natural stressors. This approach, which is not fundamentally dissimilar from that
used in human health, presumes identification and quantification of specific stressors and exposure pathways (route and source, or the what and how), attempts to
delineate exposure chronology and duration, employs strategies and models that
will make possible recreation of acute and persistent exposures, integrates exposure
and population information, and evaluates historical, current and prospective
exposures.
Stepwise, the general approach in ecological exposure reconstruction in aquatic
systems employs the same logic tree as do retrospective studies in human health.
Identification of suspect xenotoxicants; subsequent to determining that an exposure has occurred, use available resources including analytical chemistry, histology, obvious physiological or behavioral aberrancies, ecological and community
assemblages and toxicological data bases to limit the possible exposure agents
the assembled data conceivably leading to source.
Identify the relevant ecological pathways and media of interest; if pathway not
immediately obvious, such as known point source, invokes use of GIS and other
spatial data including fate and transport models to reconstruct toxicant entry into
watershed, such as the case of agrichemicals entering a watershed during a precipitation event. Presumption that application rates of agricultural chemicals
comply with listed standards.
Estimate external concentrations of toxicants in media at target sites and, if timedependent bioindicators (molecular and cellular, histopathology, community
assemblages) are available, approximate window of time(s), including duration
of exposure.
Establish routes of exposure; organisms in aquatic ecosystems, ingestion or
absorption by way of physical contact.
If indicated, by point source release or nonpoint source entry into water course
(rain events, inadvertent spillage) develop cumulative exposure estimates during
speculative time periods.
Calculate internal dose for suspect xenotoxicant or, if determined, toxicant mixtures; given multifarious physical parameters, described presently, determination
of internal dose presents the most rigorous scientific burden to the reconstructive
process; however, using all available resources, and replicating physical and geochemical conditions in surrogate ecosystems, internal dose can be resolved using
precise quantification of induced mRNA transcripts (real-time quantitative
PCR) from a narrow set of mode of action-specific genes.
Molecular indicators (genomics, transcriptomics, proteomics and metabolomics)
will play a crucial role in reconstruction of exposure events; however, these courses

114

D. Lattier et al.

of action and associated technologies are incapable of functioning as the stand

alone analytical scheme. The limiting factor for applied molecular biology in the
realm aquatic ecosystems is the shortfall of knowledge with respect to genomes of
nearly all species that inhabit inland surface waters; however, as cellular and
molecular profiles for species of interest become more inclusive, use of molecular
indicators for reconstructing ecological exposures will evolve into a pivotal asset.
The nature of exposure reconstruction makes a case for a multifaceted scientific
partnership, using analytical and descriptive assessments to accurately recreate an
exposure episode. Meticulous reconstruction must first begin with analytical chemistry in order to exclude causal factors such as natural stressors or habitat demise.
Included in the initial evaluation will also be watershed and ecological characterization such as analyses of community assemblages and Index of Biotic Integrity (IBI),
remote imagery and GIS, and measures of genetic diversity within communities
(biodiversity metrics, Fig. 2). Additionally, this undertaking must rely on available
models (hydrogeologic, soil physics, fate and transport) for spatial depiction of
contaminant depositionparticularly when considering exposure from agricultural
chemicals and other possible nonpoint sources. In addition, the ER framework could
be used to determine whether a chemical compound, detectable by analytical
methods, contributes to observed physical and biologic effects in an aquatic ecosystem
and, if so, the relative bioavailability of the suspect compound in a given physical or
trophic state. As cited anecdotally in abundant environmental reviews, the average
person is exposed, by way of dermal contact, diet, inhalation, etc., to about 10,000
discrete chemicals per day. The analogous argument could also be made regarding
scores of organisms that inhabit aquatic ecosystems throughout the globe. One of
the primary challenges in the area of ecosystems biomonitoring is to consider
disparity in species sensitivities to myriad stressors and choose the appropriate
organism with which to apply the correct suite of analyses for addressing a specific
or suspected condition.
Fish populations have been steadily declining in second order Canyon River, but
not in nearby Cottonwood River, also a second order stream located in the same
watershed. Histology and biochemistry show signs of thyroid dysfunction in Canyon
River fish, but not fish resident in Cottonwood River. Prolonged laboratory exposures using water from Canyon River and the standard aquatic toxicological model
Pimephales, indicate thyroid histopathology and changes in thyroid hormone levels
in fish. Concurrent laboratory exposures with water collected from Cottonwood
River exhibit normal thyroid histology and function. Data suggests that xenobiotic
stressor(s) in the water of Canyon River is damaging thyroid tissue in native fish
populations and may contribute to the observed population declines. Chemical analysis of suspect water fails to reveal the presence of chemicals with known thyroid
toxicity. Chemical scans show the presence of multiple minor peak differences
between Canyon River and Cottonwood River, but the exact identity and potential
toxicity of these peaks is not known. Exposure to primary cultures of fish thyroid
cells of water from Canyon River, using a microarray platform and 2D gel protein
analyses suggest induction of genes associated with cell death (apoptosis). Based on
these first approximation results, narrow-spectrum molecular methods, such as

A Look Backwards at Environmental Risk Assessment

115

Conceptual Model for Using GIS, Biological (molecular to community) and

Chemistry Tools in Aquatic Exposure Assessment and Exposure Reconstruction
Aquatic
Life Use
Unattained

Stressor
Identification
Procedures

Cause known

Risk
management

Cause unknown

GIS/Landscape
Assessment
(Land use, Riparian
Field -level
condition, Point sources)
Exposure
Assessments
ID Candidate Sources
and Stressors
Pesticide use, CAFOs,
WWTPs, septic systems
industrial sources
mining sources

Reconstructive
Exposures
(laboratory)

Bioassessment
(fish/invert/algae)
Molecular Markers
(gene/protein/biochemistry)

Water
chemistry
assessment

ID Candidate indicators
Biological diagnostic indicators
Molecular/biochemical markers

Conduct Lab Reconstructive Exposures

Pesticides, EDCs, PPCPs, POPs , Metals

Develop molecular and protein fingerprints of exposure

and ID important life stages

Field
Exposure
Confirmation

Deploy fish / inverts (appropriate

life stages) at times/locations
indicated by GIS

Collect indigenous fish / inverts

at times/locations
indicated by GIS

Compare molecular/protein fingerprints,

confirm field exposure

Fig. 2 Scenarios in exposure reconstruction that depict specific applications of molecular indicators in lotic environments

real-time PCR using transcription-specific synthetic oligonucleotides and apoptotic

antibodies, are developed to track the newly identified biomarker genes in the
thyroid cell assay. Successive chemical fractionation is performed on the water from
Canyon River, with the focused biomolecular assays applied to monitor the active
biologic fraction at each step, until individual active peaks are identified recovered.
Analytical chemistry is used to identify the structure of the material in each active
peak. The identity is confirmed by synthesizing the compound and demonstrating
an identical profile using analytical chemistry, and replicating toxic effects at relevant concentrations, first in the inexpensive cell-based assay, followed by whole

116

D. Lattier et al.

animal models. The suspect toxicant is added to water from Cottonwood River
at concentrations originally identified in Canyon River, and the toxic potency of
the spiked control is compared to the toxic potency of Canyon River water, first in
cell-based assays, then in whole animal tests. The goal here is to determine what
fraction of the total bioactivity related to Canyon River is attributable to the analytical
concentrations of the suspect toxicant. Subsequent to establishment of the toxicity
factor, source trackingby way of detection chemistrywould comprise the next
step, thereby paving the way for resource management teams to make informed
decisions about hazard identification, mitigation or remediation. Note that, for this
scenario, biomolecular methods are used to follow biologic activity, and not to
determine chemical structure directly. Chemical analyses perform the complementary
role of chemical structure determination, but have nothing to say about biologic
activity or bioavailability. The two methods must be used in tandem to solve the
problem. Molecular readouts are used in preference to whole animal readouts of
biological activity because of (a) reduced sample quantities requireda critical factor during chemical fractionation, in which many fractions will be produced with
limited volumes of active materiala volume vastly insufficient to reconstitute the
liters of exposure medium needed for whole animal testing; cell-based methods
require reconstitution of 1 mL or less (6+ orders of magnitude less medium required);
(b) reduced cost of a 24-well or 96-well based assay compared to a whole animal
assay; (c) reduced whole animal testinga worthy goal in itself, and the original
objective of the United States Environmental Protection Agency Computational
Toxicology initiative.
A similar scenario can be imagined where tumors are seen in fish from a particular river, and analytical chemistry fails to detect any of the usual suspects in the
water collected from the site of biologic impairment. A similar fractionation/biomolecular readout approach can be used, except that this time the molecular readout
will likely be different. Perhaps the comet assay, differential expression of one or
more DNA repair enzymes, or induction of S-phase associated genes, will serve as
a more appropriate indicator. The initial wide-spectrum molecular scans will facilitate selection of the indicator genes of interest, which then focal by employing lowcost, high-throughput assays for future tracking of relevant activity.
The foregoing two environmental monitoring scenarios aim to associate observed
adverse outcomes to specific environmental exposures. When histopathology or any
number of aberrant physiological traitsincluding behavior of individuals and populationscan be inexorably linked to a toxicant or xenobiotic stressor, then a point
of phenotypic anchoring [13, 14] can be established as one component biomarker
for exposure reconstruction in future monitoring activities. Environmentally induced
temporal changes in gene expression in the context of conventional toxicology endpoints can make possible the anchoring of phenotypes as a consistent, valid biomarker in occurrences of ecological exposures.
A persistent, real world problem is posed by the estrogenic activity that are frequently a complement to posttreatment effluent released by waste water treatment
plants (WWTPs), as well as other point and nonpoint sources. This is of concern at
least in part because of the observation of declining populations, and histopathology

A Look Backwards at Environmental Risk Assessment

117

suggestive of feminizing effects observed in fish downstream of some WWTPs compared to upstream organisms. Samples of WWTP effluent transported to laboratories
have been shown to inhibit fish fertility in short-term reproductive assays. Additionally,
as an initial screen for estrogenic potential in surface waters, we have developed the
simple approach of thermally amplifying (PCR) estrogen induced transcripts of the
vitellogenin gene in numerous species of freshwater teleosts [15, 16]. Because the
vitellogenin gene is normally quiescent in males, detection of the gender-specific
transcribed gene product (i.e., vtg mRNA1) provides a sensitive exposure marker for
environmentally present estrogenic compounds. The gene for vitellogenin is categorically unique in expression and function, and discovery of other bioindicators with
similar biologic profiles, such as gender-specific expression induced by a restricted
mode of action, will likely be few and far between at best.
Chemistry analysis often indicates the presence of several known estrogenic substances in WWTP effluents, including the natural estrogens, estrone (E1), 17b-estradiol
(E2), and the synthetic contraceptive estrogen, 17a-ethynylestradiol (EE2). These
substances have been shown to result in adverse reproductive effects in laboratory
validations using concentrations at which they occur in effluents. Does this mean we
should embark on a strategy to reduce the levels of these three compounds in WWTP
effluents? Will the substantial investment pay off in terms of improved wildlife
health? One important question that needs to be answered is what fraction of
the total estrogenic activity, released in effluent by the WWTP, is attributable to
customary estrogenic substances. Several studies have suggested presence of other
estrogenic materials in WWTP effluents, including nonylphenol and associated
ethoxylates, congeners of PCB, certain metals, and dioxins. Additionally, estrogenic compounds, not amenable to analytic detection, might also be present in a
given effluent. For the sake of all stakeholders involved, it is important to determine
the fractional contribution of each conditional estrogen until the greater part of
aggregate estrogenic activity in effluents has been clarified. Only then can an appropriate mitigation strategy can be implemented. Otherwise, a method may be chosen
that addresses only a fraction of the problem (i.e., the structurally related steroidal
estrogens E1, E2, and EE2, but not the structurally distant nonsteroidal estrogenic
compounds), costing much, but producing minimal management benefit. A current
approach to addressing the question of fractional analyses exploits estrogen-responsive
cell lines transfected with reporter genes driven by estrogen responsive cis-active
elements. This reporter system can effectively be used to measure the intrinsic
estrogenicity of chemicals alone and in combination. The anticipated outcome is
that in vitro systems can be used to characterize the fractional contribution of each
known estrogenic agent, and also be used for isolation of previously uncharacterized estrogenic agents present in effluents. The main advantages offered by these

Designation of macromolecular products: vtg, transcribed vitellogenin gene product (mRNA) and
Vtg, circulating vitellogenin protein, follows the Zebrafish Nomenclature Guidelines, based on
Trends in Genetics Genetic Nomenclature Guide (1998), found at http://zfin.org/zf_info/nomen.
html#1.

118

D. Lattier et al.

cell-based molecular methods, when compared to whole animal studies, are low
cost, speed of analysis, and limiting volumes of media required for analyses.
Besides this sort of retrospective exposure research, cell-lines may also prove
useful for first pass high-throughput screening of new HPV industrial chemicals for
estrogenic activity. Eventually, the development of batteries of molecular assays
covering the most frequently encountered modes of toxicity seems like a sensible
approach to preliminary screening of new chemicals (giving some hint of the relevance of predicted exposure levels). Such batteries of molecular assays can also
help quickly and cheaply identify modes of toxicity and relevant molecular indicators for use during forensic and retrospective analyses intended to characterize the
exposures causing observed population effects as noted in the previous three
illustrations.
The argument for scaled-down and focused schemes of analyses would benefit
greatly by use of embryos and early developmental stages (fish, amphibians, etc.)
as a point of departure for developing exposure biomarkers. There are substantial
numbers of similarities between fish and mammals in relation to developmental
pathways, with approximately 75% of developmentally specific genes being
homologous across metazoa. In addition to exploiting developmental plasticity for
biomarker development and effects forecasting, physical exposures to early developmental stages require minimal experimental resources and reagents, resulting in
significant cost savings. Additionally, this approach will comply with the charge of
moving away from whole animal testing and use of adult animals as models for
exposure.

Vitellogenin, the Answer in an Egg Shell;

Once in a Genome Opportunity
Vitellogenin is an established and sensitive endpoint for analysis of exposure to
estrogens, androgens and respective mimics in fish [10, 15, 17, 18]. There are several studies that have demonstrated links between high level induction of vtg and
effects in fish [1922]. One of the most popular test fish species for assessing chemical effects is the fathead minnow (Pimephales promelas, FHM), which is now used
widely for studies into endocrine disruption [23, 24]. There is now unequivocal
evidence showing that EDCs can have long-term effects on reproduction and subsequent population development in natural fish populations [21, 25].
Male fish downstream of some wastewater outfalls produce vitellogenin protein
(Vtg), a protein normally synthesized by females during oocyte maturation, in addition
to early stage eggs in their testes, and this feminization has been attributed to the
presence of estrogenic substances such as natural estrogens (estrone or 17b-estradiol
(E2)), the synthetic estrogen used in birth control pills (17a-ethynylestradiol (EE2)),
or weaker estrogen mimics such as nonylphenol in the water. Despite widespread
evidence that male fishes are being feminized, it is not known whether these
low-level, chronic exposures adversely impact the sustainability of wild populations.

A Look Backwards at Environmental Risk Assessment

119

Vg / 18S

Vitellogenin Expression by QPCR

Fathead Minnow Adult Male Livers
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
DMSO

Labline

Lake 114

5 ng/L EE2

Lake 260**

Treatment Group

Fig. 3 Vitellogenin gene expression results from exposing male fathead minnows for 24 h to water
collected from the Experimental Lake Area EE2 study, Lake 114 nondosed lake and Lake 260 EE2
dosed lake. Compared to a 5 ng L1 (nominal) EE2 positive control **, n = 4

A 7-year whole-lake experiment was conducted at the Experimental Lakes Area

(ELA) in northwestern Ontario, Canada [21], which demonstrated that chronic
exposure of fathead minnow (P. promelas) to low concentrations (5-6 ng L1) of the
potent synthetic estrogen EE2, led to feminization of males through the production
of vtg mRNA and protein, impacts on gonadal development as evidenced by intersex in males and altered oogenesis in females, and ultimately, a near extinction of
this species from the lake. These observations demonstrate that the concentrations
of estrogens and estrogenic mimics detected in freshwaters can impact the sustainability of wild fish populations.
There were several confirmation studies that were performed concurrent with the
ELA study to demonstrate the utility of using vtg as a gene marker for exposure to
estrogens in male fathead minnows. Figure 3 presents the vtg expression results of
laboratory (USEPA Cincinnati Aquatic Facility) reared male fathead minnows
exposed to water shipped from to the Cincinnati facility during the first year of lake
dosing with EE2. Lake 114 was one of two control lakes where no EE2 was introduced and Lake 260 is the lake dosed with EE2. A positive control concentration of
5 ng L-1 EE2 was also tested. The target EE2 concentration for Lake 260 was 5 ng L1
and measured concentrations during the first year were 6.1 ng L1 (SD 2.8 ng L1).
Males exposed to lab water, water with DMSO, and Lake 114 water showed no
expression of the vitellogenin gene (Fig. 3). Males exposed to Lake 260 water
showed extensive increase in levels of vitellogenin gene expression, even higher in
some cases than males exposed to 5.0 ng L1 EE2. Males had variable response to
both the 5.0 ng L1 water and the Lake 260 water. Two of the fish exposed to
5.0 ng L1 EE2 showed no increase in expression of vitellogenin. One male exposed
to water from Lake 260 showed no expression, while two had expression levels
comparable to that of fish exposed to 5.0 ng L1 EE2. Gene expression results were
found to be very similar in male fathead minnows exposed to both pure EE2 at
5.0 ng L1 and water from Lake 260 at a concentration of 6.18 ng L1 EE2.

120

D. Lattier et al.
1.2
1

Vg/(Vg+18S)

0.8
0.6
0.4
0.2
0
-0.2

Lake Lake
114 260
Day 1 Day 1

Lake Lake
260
114
Day 3 Day 3

Lake Lake
114 260
Day 7* Day 7

Lake
260
Day
13**

Fig. 4 Results of deploying indigenous fathead minnows from Lake 114* (only 7 fish left) non
dosed lake into Lake 260 EE2 dosed Lake for 13 days** (only 4 fish left)

A second experiment was conducted in which indigenous fathead minnows were

collected using minnow traps in reference Lake 114 two days prior to deployment.
Males and females were housed together until the day of deployment, at which time
the sexes were separated. Only males were used in the deployment study. Males
were deployed in cages in Lakes 114 and 260. The cages were suspended several
feet below the water surface and held in place by anchors and buoys. Fish were
provided no food during the period of deployment. The cages allowed for the free
movement of water and suspended materials. Minnows were retrieved from cages
on days 1, 3, 7 and 13 of deployment. Agarose gel-based RT-PCR was performed on
these samples. There was an insufficient number of fish to continue the study in
Lake 114 through day 14, so all seven remaining fish were removed on day 7. The
experiment was also terminated early in Lake 260 since only four fish were remaining on day 13. Male fathead minnows exhibited an increase in vitellogenin mRNA
levels after only 1 day of deployment in Lake 260 (Fig. 4). Vitellogenin mRNA
levels remained high throughout the study to day 13. Response to EE2 by males was
variable, with some fish showing high levels of expression and others showing very
little expression. The standard deviations for these samples are quite high. Males in
Lake 114 showed no significant expression on days 3 and 7 (Fig. 4). However, on
day 1 there was a single fish in Lake 114 that had elevated levels of vitellogenin
mRNA. The other four fish showed no Vg expression.
In order to determine the kinetics of vitellogenin expression during the initial
period of exposure in 2001, male FHM were collected from Lake 260 after 7 weeks,
9 weeks and 3 months of dosing. Male fathead minnow were collected at the same
time from reference Lake 114. Agarose gel-based RT-PCR was performed on samples and vitellogenin expression quantified relative to 18S ribosomal RNA (rRNA)

A Look Backwards at Environmental Risk Assessment

121

0.8

0.6

0.4

0.2

0
Lake 114 Lake 260
July 9
July 9

-0.2

Lake 114 Lake 260 Lake 114

July 25
July 25
July 25
Females

Lake 114 Lake 260

Sept
Sept

Fig. 5 2001 Summer and fall results of indigenous male and female fathead minnows

4
3.5
3
2.5
2
1.5
1
0.5
0
-0.5

Lake 114

Lake 442

Lake 260

Fig. 6 Vitellogenin gene expression in male pearl dace collected in May 2003

expression. Vitellogenin was induced in males collected from Lake 260 at all time
points (Fig. 5). Males collected from Lake 114 had little to no vitellogenin mRNA.
Vitellogenin expression in males was comparable to that of females collected from
Lake 114 on July 25. The level of expression of vitellogenin in male fathead minnows collected from Lake 260 was statistically different from that of males from
Lake 114.
Similar results were found in male Pearl Dace collected from reference lakes
114 and 440 and dosed Lake 260 in 2003 (Fig. 6). One interesting result found in
both female Pearl Dace and fathead minnows was an increased level of Vg gene

122

D. Lattier et al.
2.5
2
1.5
1
0.5
0
Lake 114 May

Lake 442 May

Lake 260 May

-0.5

Fig. 7 Vitellogenin gene expression in female pearl dace collected in 2003

3.5
3
2.5
2
1.5
1
0.5
0
-0.5

Lake 114 Lake 114 Lake 114 Lake114

May 27
June 3
June 19 June 21

Lake 114
Sept 27

Lake 442 Lake 442 Lake 442

May 27 Sept 24
Oct 1

Lake 260 Lake 260 Lake 260

May 27 June 19
Oct 1

-1
-1.5

Fig. 8 Vitellogenin gene expression in female fathead minnows collected in 2002

expression compared to that observed in reference lakes (Figs. 7 and 8). Female
FHM from Lake 260 had elevated Vg levels beyond the spawning season, September
and October 2002 (Fig. 8).
Kidd et al. [21] published results of whole fish homogenates analyzed for vitellogenin protein for fish collected during the same time as those analyzed for
vitellogenin gene expression. Whole fish vitellogenin protein analyses showed
results similar to found using the gene expression assay. Male and female fathead
minnows collected from the EE2 dosed lake had elevated levels of vitellogenin protein when compared to reference Lakes, 114 and 442most strikingly during the
second and third year of whole lake dosing, years 2002 and 2003.
Kidd et al. [21] also presented histological results of fish collected during similar
times as those analyzed for vitellogenin gene expression and whole fish homogenate vitellogenin protein. Testicular tissues of all of the male fathead minnow collected the first spring after EE2 additions began displaying delayed spermatogenesis,

A Look Backwards at Environmental Risk Assessment

123

widespread fibrosis and malformations of the tubules. Testicular germinal tissue

from all EE2 exposed males consisted primarily of spermatogonia instead of the
spermatocytes that would be the norm during the given time of year. Gonad size in
the spring of 2002 averaged 0.40 0.21% (n = 10) of the body weight for fathead
minnow from Lake 260. This was approximately one third of the mean value for
reference Lake 114 fish (1.39 0.38%; n = 15) and only one fifth of that from reference Lake 442 fish (2.27 0.41%; n = 10) collected at the same time of year [21].
Kidd et al. [21] published results that showed the fathead minnow population in
Lake 260 collapsed in the fall of 2002, after the second season of EE2 additions,
because of a loss of young-of-the-year. This reproductive failure was also observed
in the third season of amendments and continued for an additional 2 years after the
EE2 additions had ceased, although a few small individuals were caught each year,
indicating some reproduction was occurring. The loss of smaller size classes of
fathead minnow was not observed in reference Lake 442, a system with similar species composition, water volume, and trophic status.
The whole lake dosing experiment demonstrates the relationship of estrogenic
exposure from the molecular level to population effects. It illustrates that gene
expression can be exploited as an earlier indicator of exposure and an ecologically
relevant tool that can be used to reconstruct exposures to endocrine disrupting
compounds.
In 2005 EPA examined the results of exposure to 17a-ethynylestradiol in the
Experimental Streams Facility (ESF) located in Milford, Ohio. ESF has channels
that are fed continuously with water from the East Fork of the Little Miami River,
southwestern Ohio. Streams were dosed with three concentrations of EE2; 2.5, 12.5,
and 62.5 ng L1. Male FHM were placed in minnow traps in the tail tanks (located
at the terminus of each stream), of each dosed stream, and a non dosed stream, for
a period of 4 days. In addition, male fathead minnows were placed in minnow traps
in a ditch outside the facility that receives water from all ESF channels to assess
whether EE2 was being discharged to the East Fork of the Little Miami River by
way of the receiving stream. This ditch discharges into the effluent of a nearby
WWTP, so fish were also placed in the effluent of the WWTP downstream following mixture with ditch water, and in the East Fork of the Little Miami River below
the WWTP discharge. Figure 9 contains vitellogenin gene expression profiles of
liver-specific mRNA collected from this study. Vitellogenin gene transcription indicates a dose-dependent response to experimental concentrations of EE2, intermediate response from exposure to ditch water and no expression in the WWTP or
receiving stream.
Our final example demonstrates that exposure-induced vitellogenin gene expression has potential to assess estrogenicity in receiving streams below potential
sources of estrogenic compounds. Figure 10 contains the results of a study conducted in the Eagle Creek Watershed, near Indianapolis Indiana. Male fathead minnows were caged for 7 days below two effluents, sites 2 and 4, and an animal feedlot,
Site 1 during spring and fall of 2008. The two dotted lines illustrate the EE2 equivalents of 2.5 and 5.0 ng L1 as estimated from laboratory studies. The results are not
indicative of the presence of estrogenic compounds. Chemical analyses were

124

D. Lattier et al.

Male fathead minnows

Four day in-stream and combined effluent exposure
1.2
1

Vg/18S

0.8
0.6
0.4
0.2
0
Control
EE2

2.5 ng/L
EE2

12.5 ng/L
EE2

62.5 ng/L
EE2

ESF
effluent

WWTPLittle Miami

Fig. 9 Results of a 4-day EE2 exposure to male fathead minnows in EPAs Experimental Stream
Facility and outside deployments to monitor potential estrogenic discharge to Little Miami River
0.00002

vtg /18S

0.000016
0.000012
0.000008
0.000004
0

Time 0 Site 4b

Site 5 Site 2
May 22

Site 1 Site 4a

Site 5 Site 2
June 3

Site 1

Fig. 10 Results of 7-day deployments of male fathead minnows within the Eagle Creek Watershed
in spring and summer. Top dashed line indicates 5 ng EE2 L1 equivalent exposure lower dashed
line indicates 2.5 ng EE2 L1 equivalent exposure

performed on water samples collected during these exposures; however neither EE2
nor estradiol (E2) was detected. The ramification of this study is the possibility for
biologic detection of estrogenicity in the absence of detectable chemical analyses.
This could indicate mixtures of estrogenic compounds that might be less than limits
of chemical detection but not below the necessary level to induce vitellogenin gene
expression.
The studies outlined in this section are the support and foundation for the concept of using gene expression as an approach to reconstructing exposures to estrogenic compounds. Current research is underway using a toxicity identification

A Look Backwards at Environmental Risk Assessment

125

evaluation (TIE) approach to identify estrogenic compounds using gene arrays.

Specific gene fingerprints of these compounds may be developed that can be used
on effluents and surface waters found to be estrogenic to identify the specific chemical or chemicals. In turn this information can be used to help develop a control
strategy to reduce concentrations of estrogenic compounds in the environment.

Acknowledging the Dynamics and Complexities

Exposure is the result of a stressor intersecting with a receptor. To understand the
phenomenon of exposure we must consider the characteristics and behavior of
the stressor and the characteristics and behavior of the receptor. For example, once the
stressor is released into the environment physical conditions and variables move it
across and through the landscape, during which time the stressor may be transformed by myriad processes, including chemical, photo-, and microbiological
influences. Conversely the receptor might also move across the landscape and
acquire morphological or physiological adaptations at various life stages that confer
protection from stressors by way of induced polyphenisms or, in some cases in the
course of early development, succumb to greater vulnerability and mortality.
Environmentally induced polyphenisms in early development [26, 27], can yield a
consequence of population-specific exposure threshold. These phenomena, in conjunction with unknown and possible highly variable genetic backgrounds of geographically distinct communities, further confounds molecular measurements of
ecological exposure to parallel stressorsespecially in populations where polyphenisms modulate susceptibility to exposure. Reconstructing ecological exposure
is to attempt an understanding of the history, genetics and inestimable processes that
influenced the fateful juncture of a deleterious stressor and populations of biologic
receptors. Subsequent to exposure, the risk of an adverse outcome to wildlife communities and populations is a function of magnitude, frequency, duration, and
cumulative aspects of exposureall of which are amenable to retrospective inference using existing models.
For exploitation of any ecologically based biomarker, there is an absolute need
to consider geophysical parameters and nutrient conditions when applying indicators to risk analyses of environmental toxicants. Attempting to establish a relationship between biomarker and external concentration of contaminant, for retrospective
analysis or immediate survey, investigators have an obligation to consider and characterize the oftentimes disregarded temporal, spatial and geochemical and other
physical conditions of the proximal study area [28]. Physical conditions that have
overwhelming implication on bioavailability of xenobiotic stressors, in field and
laboratory studies [29] include, but are not limited to, total nitrogen and phosphorus, pH, dissolved oxygen, alkalinity, dissolved minerals, microbiologic communities, temperature, turbidity and regional meteorology.
In a recent study using surrogate ecosystems, we investigated the effect on
relative trophic levels on EE2 induced expression of the gene for vitellogenin in

126

D. Lattier et al.

fathead minnows [30]. Male fathead minnows, were exposed to a single dose of
17a-ethynylestradiol at a nominal concentration of 20 ng EE2 L1 in fiberglass
mesocosm tanks containing either a carrier vehicle (DMSO) or water control or
secondary nutrient treatment reflecting oligotrophic (0.012 mg L1 total phosphorous
(TP)), mesotrophic (0.025 mg L1 TP), or eutrophic systems (0.045 mg L1 TP).
A total of 21 tanks were used in a random treatment/control design of three replicates per treatment/control. In preconditioned mesocosms that were designed to
replicate 17a-ethynylestradiol (EE2) exposure in respective trophic systems, results
suggested that the level of vitellogenin gene expression was inversely related to the
nutrient load, with the highest expression observed in mesocosms lacking in plant
nutrients and having an abundance of dissolved oxygen.
Given the dynamic nature of primary productivity in aquatic ecosystems, investigators who embark upon exposure reconstruction and retrospective quantification
of external dose must rely heavily on fate and transport models and climatological
records, in addition to taking into account the physicalchemical parameters of the
system (watershed). Such might be the case in a scenario of recent application of
one of the most widely used herbicides in the United States, followed closely by a
precipitation event. Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine)
is arguably one of the most pervasive stressors detected in groundwater and aquatic
ecosystems throughout lower 48 states [31]. The regulated concentration of this
herbicide in US drinking water is 3 ppb (3 mg L1); however, in stream levels on the
order of 224 ppb have been reported in areas dominated primarily by agricultural
activities.
Although previous studies have suggested that atrazine is not estrogenic by virtue of failure to induce vitellogenin transcription in mature male goldfish (Carassius
auratus) [32], the controversial chemical nevertheless has been determined to be an
endocrine disrupting compound (EDC). Recent investigations have indicated that
atrazine binds directly to steroidogenic factor-1 (SF-1), an orphan receptor, facilitating enhanced SF-1 binding to the aromatase promoter [33], resulting in over
expression of the aromatase gene. Data supports the long held speculation that atrazine functions as an endocrine disruptor in wildlife, with possibilities of reproductive impairment and skewed gender ratios, and is capable of initiating reproductive
neoplasia in experimental animals and multiple human cell lines. By way of latest
data, reconstructing what is presumed to be atrazine exposure might transpire similarly to the previously described lotic environment scenarios, again using GIS, fate
models, chemistry and aromatase (CYP19) expression as the initial screening
biomarker.

Down the Primrose Pathway

Use of nucleic acid technologies, including thermal cycle amplification (PCR) and
microarray hybridization, to inform ecological exposure and hazard assessment has
added significantly to the observations rendered by toxicological testing over the

A Look Backwards at Environmental Risk Assessment

127

past 3 decades; however, the collective benefits might not shift the balance from
pitfalls and indiscreet interpretation of biologic observations. The promise to inform
ecological risk assessment with a molecular crystal ball that would allow us to faithfully divine clear-cut linkages between molecular triggers initiating toxic pathways,
and biologically relevant adverse endpoints, has largely been caught up in myriad
complex interactions that occur as organisms simply attempt to make it through
another day in an environmental mlange. This combined with the anthropocentric
supposition that gene expression is necessarily the lone inevitable consequence of
antecedent environmental events resulting in all biologic and phenotypic outcomes.
Exposure outcomes represent the complex interplay between genetics, the action of
many genes, behavior and the environment.
Earlier the suggestion was made that when attempting to describe biologic processes at any level of organization, investigators are advised to consider combined
hydrogeologic and geophysical parametersreplicating them in laboratory/mesocosm studies to the extent feasible. In combination, and absence of xenotoxicant
agents, physical and geological factors are, indeed, a mixture of individual stressorscontinually altering molecular responses based on the ever-changing nature
of ecological systems. Adding to this potential exposure medium are one or more
synthetic xenoestrogens; then, exposure-specific transcript patterns in any number
of aquatic species arising from microarray analysis become theoretically untenable.
The challenge that ecotoxicogenomics has yet to meet is discrimination among the
biologic networks and gene sets that account for preponderance of physiological
stress, directing a homeostatic condition to conclude in allostatic overload. This is
no small informatics feat given the number of biologic influences. The above
difficulties do not include all too common technical discrepancies such as reproducibility, use of multiple array platforms, and transformation of raw data.
In the sphere of ecological genomics, much has been noted regarding compensatory and adaptive responses to xenobiotic exposure. Recent finding in Daphnia
magna suggest [34] that local environmental conditions can lead to genetic adaptation of natural populations. Demonstrable selection pressure occurring in a local
habitat suggests that physical conditions with added stress of xenotoxicants might
significantly contribute to genetic attrition in natural populations of Daphnia. This
phenomenon further exemplifies investigators need to be judicious when parsing
data from differential gene expression profiles.
The concept of hormesis has in recent years resurfaced as a general model for
physiological response to exposure. There is vociferous argument that most, if not
all, experimental exposures in animal models occur as a function of hormesisthe
endpoints described by either resulting in either a J-shaped or an inverted-U dose
response curve [35]. Results from experiments using synthetic estrogenic compounds
dispute the classic notion of hormesis as an adaptive response [36]. The investigators
argue that in the case of manmade xenoestrogens exposure, observed apical endpoints result from an adverse stimulatory response, radically dissimilar from the
notion that toxicity pathways, elicited by exposures over a range of concentrations,
make corrections for low level internal doses. Outcomes of adverse stimulatory
responses are detrimental to populations and communities, redirecting energy

128

D. Lattier et al.

essential for homeostatic processes which results in diminished fitness. If such

adverse response is the case in human health as well as ecological exposures; this
concept will undoubtedly confront current means of performing risk assessments.
Mechanisms accounting for such dose related biologic activity have been described
for numerous pharmacologic agents at the level of intercellular receptors in mammalian models. Research finding to date have yet to suggest that there is a specific
hormetic mechanism; however, such conclusions offer strong inference for an allpurpose strategy the aims of which are conservation of resources across biological
systems. Considering increasing numbers of natural habitats in decline throughout
past decades, resource conservation would have far reaching implications for xenobiotic exposures in aquatic and terrestrial ecosystems. Hormesis indicates that potential lack of correspondence among gene networks at low and high doses might lead
to altered strategies for biomarker development, in addition to a radical change
regarding the way in which the business of risk assessment is conducted.
The path from ecological genomics to resource management and regulatory policy is hindered for the time being by a seemingly insurmountable gap of uncertainty.
Mindful ecological monitoring and retrospective ecological analyses will serve to
reduce uncertainty in exposure science. Linking available knowledge bases including those containing characterizations on physiology, cellular and molecular mechanisms, histopathology and behavior, in conjunction with spatial information, will
assist in narrowing the number of candidate toxicants, in addition to providing possible scientific insights regarding timing and duration of exposure. As a point of
departure, investigators engaged in ecological monitoring could begin to assemble
an inventory of preliminary molecular biomarkers by exposing aquatic test model
organisms, in mesocosms, to high production volume chemicals (agrochemicals,
etc.) or chemicals of emerging concern (CECs; pharmaceuticals) to discriminate
exposure-specific gene expression markers. Exposure media that reflects known
degrees of primary productivity (total nitrogen and phosphorus; oligo-, meso- and
eutrophic systems), might make possible identification of a restricted suite of primary and correlative mode of action gene-based biomarkers. Using microarray platforms to discern patterns of transcription or for isolation of individual gene products
from which synthetic PCR oligonucleotides could be generated; this collection of
indicators could then serve as the initial screening line up to taper suspected agents
of exposure in a reconstructive scenario. The use of whole genome microarrays for
initial site-specific exposure surveys might, by virtue of environmental complexity,
portray more qualitative profiles; however, distillation of results will in all probability lead to platforms designed for more targeted transcriptomic read outs.

Beyond Genes and Proteins

One of the main objectives of exposure reconstruction is estimation of the temporal
aspects of exposure. In all likelihood, only a small set of molecular processes have
the capability to reveal the duration component of ecological exposurethe gene

A Look Backwards at Environmental Risk Assessment

129

and protein for vitellogenin, a structural precursor to yolk protein in oviparous

animals, being one of the current most notable examples. As seen in Figs. 2 and 3,
induced transcription of the vitellogenin (vtg) gene detects the immediacy of exposure to environmental estrogens or estrogen mimics. Circulating vitellogenin protein, following downstream translation of the processed message, is detected
approximately 14 days post transcription. Relative numbers of vitellogenin transcripts increase proportionally with concentration of estrogenic compound and with
time. Magnitude of response can be indicative of either. Absolute quantification of
induced vitellogenin transcripts permits investigators to back calculate from biomarker to presumptive concentration in EE2 equivalents. If there is a presumption
that a xenoestrogenic exposure event has occurred in an aquatic system and the
liver-specific gene indicates no active transcription, then investigators would test for
the circulating protein using an available ELISA technique. This then allows for a
possible first approximation of external concentration in addition to time at which
exposure to the estrogenic compound occurred. Again, this temporal estimate must
be made in context of other available data, such as identified point-sources, recent
meteorological events and regional land use information such as seasonal application of agrichemicals. In cases such as this, investigators and modelers must assume
the roles of ecosleuths.

Epigenetics; Methylation and microRNA

Epigenetics represents most recently observed cellular phenomena destined to
transform the landscape of exposure science, particularly as related to ecotoxicology and biomarker discovery in aquatic sentinels. The term epigenetic refers to
heritable changes in gene expression that occur in the absence of structural
modifications in sequence of DNA. These heritable instructions in transcriptional
machinery can become fixed in the genome mitotically, meiotically, or during both
genetic events [37, 38]. Epigenetic processes, which are reflected in levels of gene
expression, comprise the following known mechanisms; nucleotide-specific DNA
methylation, modifications in chromatin structure mediated by both acetylation and
methylation of histone proteins, and the expression of noncoding, small (micro)
RNAs (miRNAs) in posttranscriptional regulatory control programs.
Given the suite of diverse epigenetic mechanisms, the one most likely to yield
experimental clues to immediate or preceding xenobiotic exposure, is methylation
of DNA at the cytosine-5 site within CpG dinucleotides (CpG islands)particularly in 5 upstream promoter regions of transcriptional units, and cis-regulatory
elements that often function distally with respect to the site of transcription initiation. Methylated CpG dinucleotides establish distinguishable epigenetic features
that are commonly associated with transcriptionally silent, condensed chromatin.
Such biochemical modifications generate genomically spatial and functional
controls that harmonize with trans-regulatory mechanisms. A straightforward
description of methylated promoters would posit that methyl groups projecting from

130

D. Lattier et al.

abundant cytosine nucleotides act to sterically impede the binding activity of soluble
transcription factors and cofactors; therefore, it follows that a state of hypomethylation results in accessible conformation and increased levels of gene transcription. The analogous structural modification on chromatin is histone
hyperacetylationthe epigenetic switch usually associated with an upsurge in
transcriptional activity.
Epigenetic features have long been known to play an important role in developmental plasticity [39, 40]. Molecular mechanisms responsible for bringing about
epigenetic modifications, that are manifest in phenotypes, are becoming increasingly well characterized. Methylation of nucleic acids and the DNA packaging histone proteins is established as a primary orchestrator in early embryogenesis in
metazoa and is the principal mechanism of X-chromosome inactivation (Barr bodies) in addition to the phenomenon of imprinting during early development [41].
Ongoing epigenetic investigations in the area of human health point to incontrovertible data indicating that environmental exposures, particularly during early development, can provoke long-term epigenetic changes. These fixed structural alterations
which can be transgenerationally inherited, appear to be primary etiology in various
disease states that arise in later life stages [42]. Recent investigation into the stable
and lasting consequences of genomic DNA methylation support the long held strong
inference that presence or absence of cytosine-5 methyl groups has substantial
influence regarding aspects of physiology and behavior. Detrimental postpartum
events imposed upon mice suggests that early life stress (ELS) was associated with
sustained DNA hypomethylation [43] of a critical DNA regulatory region that was
shown to withstand age-dependent conversion in methylation states. This state of
hypomethylation resulted in hypersecretion of the glucocorticoid corticosterone,
resulting in changes of ability to effectively manage stress and with memory function. One might presume that analogous ELS, resulting from unintended contact
with environmental stressors and xenobiotics, could occur in wildlife populations
leading to long-term effects with consequences ranging from individuals to
communities.
Methylation states of DNA, in context of environmental exposure, will provide
the most accessible biologic window into the inadvertent loss, or gain, of gene
function, not only during early development but at any life stagein any aquatic
organism. Altered states of DNA methylation triggered by environmental exposure, analogous to the identical trends in disease progression, will lead to unscheduled transcription and premature initiation, or deactivation, of certain genes. Since
diet, xenobiotics and behavior can produce changes in the assorted epigenetic
organization [44], it follows that environmental causation might systematically
change epigenetic profiles and influence future responses to environment stressors. Some endocrine disrupters have been shown to exert genome-wide effects on
the state of DNA methylation [45]. In one study that doubtless has ramifications
for all oviparous animals and ecological analyses, hormone treatment of immature White Leghorn roosters resulted in a demethylation of upstream estradiolreceptor binding site of the gene for vitellogenin [46]. This change of methylation
state, observed on only one of the two strands of DNA, is referred to as

A Look Backwards at Environmental Risk Assessment

131

hemimethylation and was directly correlated with immediate onset of the

vitellogenin gene primary transcript.
The epigenetic phenomenon of DNA methylation, because of stability over time,
will provide direct linkage for interpretation of ecotoxicogenomic data. Although
the majority of studies in area of epigenetic methylation focus on single gene events
[47], postexposure global and anonymous methylation patterns might yield insight
not only into the initiating stressor but also, through transgenerational analysis, into
the temporal range wherein the observed exposure-driven methylation states
occurred. Investigating the interplay between the environment and epigenome [48]
has the potential to yield MOA-specific exposure biomarkers in aquatic and terrestrial ecosystems.
The second mode of epigenesis that has implications for ecological monitoring,
and possibly exposure reconstruction, is the recently described gene regulatory process mediated by a class of small, noncoding RNA molecules [49] that function in
genomes of eukaryotes. Most small RNA candidates identified to date, in human,
mouse and rat, exhibit conservation in other vertebrates, including dog, cow,
chicken, opossum, and zebrafish [50], and conceivably throughout the animal kingdom. Two primary categories of these small RNAs have thus far been described:
short interfering RNAs (siRNAs) and microRNAs (miRNAs) [51]. MicroRNAs
comprise a genus of 2030 nucleotide moieties that bind to sequence-specific, complementary regions of processed mRNA transcripts in a double-stranded conformation, appropriating the message, thereby decreasing or eliminating production of the
corresponding protein product. This posttranscriptional mode of regulating gene
expression gene is mediated not only by formation of miRNA-target hybrids, but
also target mRNA degradation [52].
Predictions have been made that higher Eukaryotes express thousands of miRNAs, and although only a fraction of those have been identified, there is corroborating data to suggest that this class of small RNAs play an important role in numerous
developmental processes and critical response pathways. As the numbers of characterized small RNAs in diverse genomes continue to expand, modeled conjecture
suggests that miRNAs can regulate a substantial fraction of the genome. In 2005,
computational predictions held that 10% of all protein-coding transcripts were subject to regulatory control by miRNAs; however, recent data indicates that this fraction will likely expand considerably. Single miRNAs are hypothesized to regulate
multiple gene products, and there are suggestions that, in genomes of higher eukaryotes, the functional importance of miRNAs in regulatory programs of gene expression could surpass that of soluble trans-acting factors.
One ecological investigation endeavored to link stressor induced effects observed
in a wild population of teleosts, to epigenetic causation. Investigators observed that
Fundulus heteroclitus (killifish) inhabiting a creosote-contaminated system in the
Elizabeth River, Virginia, exhibit what is termed, refractory CYP1A phenotype.
Contrary to conventional wisdom, the population lacked the expected induction of
cytochrome P4501A (CYP1A) mRNA [53], an essential enzymatic component of
phase I xenobiotic and drug metabolism induced by aromatic hydrocarbons.
Additionally, the population lacked immunodetectable catalytic P450 protein.

132

D. Lattier et al.

Although the refractory CYP1A phenotype indicted heritability, strictly genetic

bases did not seem to be linked to causation. The hypothesis that cytosine methylation at CpG sites in the promoter region of CYP1A underlies the refractory CYP1A
phenotype was tested using the technique of bisulfite sequencing. Liver-specific
genomic DNA was isolated from wild-caught adult killifish and from pools of laboratory reared F1 embryos. Analyses of DNA isolated from indigenous fish taken
from both the contaminated and reference site indicated that there was no detectable
cytosine-5 methylation at any of the 34 CpG sites examined, including three regions
that are considered integral to the putative xenobiotic response element (XRE). The
investigators also noted that the refractory CYP1A phenotype gradually diminished in the course of development in laboratory reared F1 generation fish.
Although in the above study, promoter methylation has been excluded as a causal
factor for the described phenotype, an epigenetic program might well be at play.
The deficiency of functional catalytic protein and a time-dependent gain of function
in F1 generation and in subsequent life stages are consistent with and offer a compelling argument that small RNAs might serve as the regulatory mechanism for the
refractory CYP1A phenotype.
The study of microRNAs, and roles in posttranscriptional regulation, is still considered to be in preliminary stages, although the small RNA network has already
been recognized in the area of human health as targets for biomarker and therapeutic development. If it is the case that miRNAmRNA hybrids are determined to
maintain stability over timethat is, greater duration than xenobiotic-induced
translatable messagesor if, under given environmental pressures, microRNAs are
constitutively expressed with developmental or tissue specificity, then these molecules will offer another inroad into ecological retrospective exposure analysis.

Otolith Geochemistry
Eco investigators taking yet a different approach to exposure monitoring make a
forceful argument for using the novel approach of otolith geochemistry [54] as a
tool for making strong inference for environmental conditions and exposure history
in individual and populations of teleosts. Otoliths are structures of the inner ear of
fish located just behind the eyes, also referred to as ear bone or ear stone.
Calcium carbonate (generally aragonite), the primary constituent of otoliths, is
derived from water and components present therein which bind to the mineral structure otoliths with continual deposition. Because of acellular biomineralization, otolith structures are not subject to resorption during periods of starvation or stress and
eliminate confounding variables such as size, age, and gender. As the otolith expands
in mass, new calcium carbonate crystals form and, as with most crystal structures,
lattice vacancies are a consequence of crystal formation. Analyses of the trace elemental composition or isotopic signatures of trace elements within a fish otolith provide
insight into the water bodies, and associated conditions, in which individuals have
previously been inhabitants. The otolith method, analyzing for organochlorine

A Look Backwards at Environmental Risk Assessment

133

pesticides and PCBs, has recently been exploited to distinguish spatiotemporal

variability and origins between North Atlantic and Mid Atlantic populations of
Bluefin tuna [55]. Results of this study essentially describe a successful effort to
reconstruct the toxicant profile to which populations had been exposed.
Additionally, fish scaleswhich develop from dermal mesenchymemight
offer a path to examine a xenobiotic gradient, revealing an ordered history of
toxicant exposure. This method has been applied to detect presence of mercury in
largemouth bass [56], comparing relationship between total Hg concentration
in scale samples and muscle tissue in the same organism.

Coda: Reconstructing Ecological Exposures

In most cases, selection of the right biomonitoring assays as well as back-prediction
of exposure patterns, environmental concentrations, or sources of contaminants will
be heavily dependent on data from other sources. As methods for ecological monitoring proliferate, it will become increasingly inefficient to run every possible
biomonitoring assay in every case. In some cases, lists of usual suspects will be
targeted for monitoring, while at other times site-specific information will suggest
particular assays to be exploited. Usually, the measurements from these assays will
be consistent with a range of exposure routes, timings, and concentrations. Additional
data will then be needed in order to narrow these ranges sufficiently to provide
actionable information for environmental decision-making.
The development and continued update of usual suspects inventory can support exposure reconstruction in cases where suspected contaminant stressors are not
immediately evident. Previous successful reconstructions as well as existing chemical monitoring data can provide convenient and reasonable starting points for developing lists of candidate toxicants. In addition, data on import, production,
distribution, usage, and disposal allows estimation of possible introduction rates of
different contaminants into the environment through a variety of pathways. Measured
and predicted physiochemical properties of potential contaminants can be used to
predict transport and eventual fate, including important exposure processes such as
biomagnification, thereby transforming estimated environmental introduction rates
into potential external concentrations and rates of biologic exposure. Preexisting
data on potency, differential susceptibility, and modes of action can be combined
with potential exposure rates to prioritize contaminants for placement on candidate
contaminant lists, based on the likelihood of harmful outcome resulting from exposure. This information can also be used to subselect biomonitoring approaches
based on site-specific conditions, such as observations on modes of toxicity, range
of species affected, local transport processes, and proximity to potential sources of
contamination.
Once assembled, biomonitoring results are typically found to be consistent with
a range of exposure scenarios, and additional information often plays a critical role
in narrowing this range to the point that useful science policy decisions can be

134

D. Lattier et al.

made. Any given set of biomonitoring results should be consistent with a range of
durationconcentration combinations, the size of which is dependent on the shape
of the doseresponse curve, the frequency of sampling, and the kinetics of the
biomonitoring signal following exposure. Individual durationconcentration combinations may in turn be consistent with a variety of potential contaminant sources
and routes of environmental fate and transport. Information on potential introduction rates from near and remote sources, as well as fate and transport properties will
often provide the information pivotal to correctly identify the source and route, as
well as provide important corroboration of the proposed contaminant identity and
the exposure durationconcentration profile. Monitoring for chemicals or manufactured products may depend less on the risks posed by the end-state product than on
the risks induced by extraction, processing, and transportation of raw materials or
by the wastes generated during manufacturing processes. In such situations, biomonitoring assessments should integrate risks from the entire product life cycle.
Identifying and minimizing exposures to contaminants plays a critical role in
environmental protection. Current biomonitoring studies provide minimal information to identify the what, when, where, and how much of exposures. The concept of
exposure reconstruction can provide a framework to guide the development of new
biomarkers and the design of future biomonitoring studies that will shift the research
focus from detection of exposures to elucidating the mechanisms of exposures from
sources to internal measurements. Human and ecosystem exposures can be better
understood through the strategic development of biomarkers of exposure that can
inform the exposure reconstruction process. Significant research in the area of
exposure reconstruction is necessary to advance the protection of humans and ecosystems, and research in this area presents numerous opportunities and challenges.
A recent publication [57] highlights a number of these issues including the variation
in performance and computational complexity of inversion techniques, the multiplicity of potential real-world exposure scenarios, and the impact of biochemical
properties and sampling characteristics related to biomarkers.
Geneenvironment interactions are extremely complex and irrefutably nonlinear.
No existing ecological risk models are informed with the capability of predicting
exposure dose relationships and outcomes that arise from the intersection of
inestimable environmental conditions and complex biologic responses; however, as
the number of well-characterized genomes becomes greater, our understanding of
byzantine processes will enhance not only predictive risk assessment but also the
ability to describe retrospective exposures.
Exposure reconstruction demands that we essentially shift modes of thinking
from what was previously deductive reasoning to the strong inference inductive
interpretation, the flow of which is depicted below [58].
Observation; determination of chemical and biologic patterns
Speculative multiple hypotheses based on incremental, retrospective data
from multiple sources
Strong inference; extrapolative external concentration; hypothesis elimination and causal reconstruction

A Look Backwards at Environmental Risk Assessment

135

The above approach will permit the formulation of conditional inductive trees that
provide the foundation for rebuilding an exposure phenomenon. If the cumulative
information that arises from a reconstruction scenario is sufficient, then a bench-scale
experimental reconstruction can be designed with replicated ecological parameters,
using mesocosms, artificial streams, or other surrogate ecosystems. This will facilitate
further development of genomic indicators for continued monitoring and site surveys.

References
1. Aylward LL, Brunet RC, Starr TB, Carrier G, Delzell E, Cheng H, Beall C (2005) Exposure
reconstruction for the TCDD-exposed NIOSH cohort using a concentration- and age-dependent
model of elimination. Risk Anal 25:945956
2. Bauchinger M (1998) Retrospective dose reconstruction of human radiation exposure by
FISH/chromosome painting. Mutat Res 404:8996
3. Roy A, Weisel CP, Gallo MA, Georgopoulos PG (1996) Studies of multiroute exposure/dose
reconstruction using physiologically based pharmacokinetic models. Toxicol Ind Health
12:153163
4. Williams PR, Paustenbach DJ (2003) Reconstruction of benzene exposure for the Pliofilm
cohort (1936-1976) using Monte Carlo techniques. J Toxicol Environ Health A 66:677781
5. CDC (2003) Second national report on human exposure to environmental chemicals. US Department
of Health and Human Services, Center for Disease Control and Prevention, Atlanta, GA
6. CDC (2005) Third national report on human exposure to environmental chemicals. US Department
of Health and Human Services, Center for Disease Control and Prevention, Atlanta, GA
7. USEPA (2005) National lake fish tissue study. US Environmental Protection Agency,
Washington, DC
8. NRC (2006) Human biomonitoring for environmental chemicals. National Research Council
of the National Academies, The National Academies Press, Washington, DC
9. Tan YM, Liao KH, Conolly RB, Blount BC, Mason AM, Clewell HJ (2006) Use of a physiologically based pharmacokinetic model to identify exposures consistent with human biomonitoring data for chloroform. J Toxicol Environ Health A 69:17271756
10. Lattier DL, Gordon DA, Burks DJ, Toth GP (2001) Vitellogenin gene transcription: a relative quantitative exposure indicator of environmental estrogens. Environ Toxicol Chem 20:19791985
11. Korte JJ, Kahl MD, Jensen KM, Pasha MS, Parks LG, LeBlanc GA, Ankley GT (2000) Fathead
minnow vitellogenin: Complementary DNA sequence and messenger RNA and protein expression after 17 beta-estradiol treatment. Environ Toxicol Chem 19:972981
12. Schmid T, Gonzalez-Valero J, Rufli H, Dietrich DR (2002) Determination of vitellogenin
kinetics in male fathead minnows (Pimephales promelas). Toxicol Lett 131:6574
13. Paules R (2003) Phenotypic anchoring: linking cause and effect. Environ Health Perspect
111:A338A339
14. Moggs JG, Tinwell H, Spurway T, Chang HS, Pate I, Lim FL, Moore DJ, Soames A, Stuckey R,
Currie R, Zhu T, Kimber I, Ashby J, Orphanides G (2004) Phenotypic anchoring of gene expression changes during estrogen-induced uterine growth. Environ Health Perspect 112:15891606
15. Lattier DL, Reddy TV, Gordon DA, Lazorchak TM, Smith ME, Williams DE, Wiechman B,
Flick RW, Miracle AL, Toth GP (2002) 17 a-ethynylestradiol-induced vitellogenin gene transcription quantified in livers of adult males, larvae, and gills of fathead minnows (Pimephales
promelas). Environ Toxicol Chem 21:23852393
16. Biales AD, Bencic DC, Lazorchak JL, Lattier DL (2007) A quantitative real-time polymerase
chain reaction method for the analysis of vitellogenin transcripts in model and nonmodel fish
species. Environ Toxicol Chem 26:26792686

136

D. Lattier et al.

17. Arukwe A, Goksoyr A (2003) Eggshell and egg yolk proteins in fish: hepatic proteins for the
next generation: oogenetic, population, and evolutionary implications of endocrine disruption.
Comp Hepatol 2:4
18. Sumpter JP, Jobling S (1995) Vitellogenesis as a biomarker for estrogenic contamination of the
aquatic environment. Environ Health Perspect 103(suppl 7):173178
19. Ankley GT, Bencic DC, Breen MS, Collette TW, Conolly RB, Denslow ND, Edwards SW,
Ekman DR, Garcia-Reyero N, Jensen KM, Lazorchak JM, Martinovic D, Miller DH, Perkins
EJ, Orlando EF, Villeneuve DL, Wang RL, Watanabe KH (2009) Endocrine disrupting
chemicals in fish: developing exposure indicators and predictive models of effects based on
mechanism of action. Aquat Toxicol 92:168178
20. Herman RL, Kincaid HL (1988) Pathological effects of orally administered estradiol to rainbow
trout. Aquaculture 72:165172
21. Kidd KA, Blanchfield PJ, Mills KH, Palace VP, Evans RE, Lazorchak JM, Flick RW (2007)
Collapse of a fish population after exposure to a synthetic estrogen. Proc Natl Acad Sci USA
104:88978901
22. Schwaiger J, Spieser OH, Bauer C, Ferling H, Mallow U, Kalbfus W, Negele RD (2000)
Chronic toxicity of nonylphenol and ethinylestradiol: haematological and histopathological
effects in juvenile Common carp (Cyprinus carpio). Aquat Toxicol 51:6978
23. Hutchinson TH, Barrett S, Buzby M, Constable D, Hartmann A, Hayes E, Huggett D,
Laenge R, Lillicrap AD, Straub JO, Thompson RS (2003) A strategy to reduce the numbers of fish used in acute ecotoxicity testing of pharmaceuticals. Environ Toxicol Chem
22:30313036
24. Panter GH, Hutchinson TH, Lange R, Lye CM, Sumpter JP, Zerulla M, Tyler CR (2002) Utility
of a juvenile fathead minnow screening assay for detecting (anti-)estrogenic substances.
Environ Toxicol Chem 21:319326
25. Scholz S, Mayer I (2008) Molecular biomarkers of endocrine disruption in small model fish.
Mol Cell Endocrinol 293:5770
26. Gilbert SF (2002) The genome in its ecological context: philosophical perspectives on interspecies epigenesis. Ann N Y Acad Sci 981:202218
27. Nijhout HF (2003) Development and evolution of adaptive polyphenisms. Evol Dev 5:918
28. Petersen DG, Sundbck K, Larson F, Dahllf I (2009) Pyrene toxicity is affected by the nutrient
status of a marine sediment community: implications for risk assessment. Aquat Toxicol
95(1):3743
29. Persoone G, Van de Vel A, Van Steertegem M, De Nayer B (1989) Predictive value of laboratory
tests with aquatic invertebrates: influence of experimental conditions. Aquat Toxicol 14:149167
30. Gordon DA, Toth GP, Graham DW, Lazorchak JM, Reddy TV, Knapp CW, deNoyelles J,
Frank CS, Lattier DL (2006) Effects of eutrophication on vitellogenin gene expression in male
fathead minnows (Pimephales promelas) exposed to 17a-ethynylestradiol in field mesocosms.
Environ Pollut 142:559566
31. Barbash JE, Thelin GP, Kolpin DW, Gilliom RJ (2001) Major herbicides in ground water:
results from the National Water-Quality Assessment. J Environ Qual 30:831845
32. Span L, Tyler CR, Rv A, Devos P, Mandiki SNM, Silvestre F, Thom J-P, Kestemont P (2004)
Effects of atrazine on sex steroid dynamics, plasma vitellogenin concentration and gonad
development in adult goldfish (Carassius auratus). Aquat Toxicol 66:369379
33. Fan W, Yanase T, Morinaga H, Gondo S, Okabe T, Nomura M, Komatsu T, Morohashi K,
Hayes TB, Takayanagi R, Nawata H (2007) Atrazine-induced aromatase expression is SF-1
dependent: implications for endocrine disruption in wildlife and reproductive cancers in
humans. Environ Health Perspect 115:720727
34. Coors A, Vanoverbeke J, De Bie T, De Meester L (2009) Land use, genetic diversity and toxicant tolerance in natural populations of Daphnia magna. Aquat Toxicol 95(1):7179
35. Calabrese EJ, Baldwin LA (2003) Toxicology rethinks its central belief. Nature 421:691692
36. Weltje L, vom Saal FS, Oehlmann J (2005) Reproductive stimulation by low doses of xenoestrogens contrasts with the view of hormesis as an adaptive response. Hum Exp Toxicol
24:431437

A Look Backwards at Environmental Risk Assessment

137

37. Metivier R, Gallais R, Tiffoche C, Le Peron C, Jurkowska RZ, Carmouche RP, Ibberson D,
Barath P, Demay F, Reid G, Benes V, Jeltsch A, Gannon F, Salbert G (2008) Cyclical DNA
methylation of a transcriptionally active promoter. Nature 452:4550
38. Goldberg AD, Allis CD, Bernstein E (2007) Epigenetics: a landscape takes shape. Cell
128:635638
39. Dolinoy DC (2007) Epigenetic gene regulation: early environmental exposures.
Pharmacogenomics 8:510
40. Dolinoy DC (2008) The agouti mouse model: an epigenetic biosensor for nutritional and
environmental alterations on the fetal epigenome. Nutr Rev 66(suppl 1):S7S11
41. Vire E, Brenner C, Deplus R, Blanchon L, Fraga M, Didelot C, Morey L, Van Eynde A,
Bernard D, Vanderwinden JM, Bollen M, Esteller M, Di Croce L, de Launoit Y, Fuks F
(2006) The Polycomb group protein EZH2 directly controls DNA methylation. Nature
439:871874
42. Tost J (2009) DNA methylation: an introduction to the biology and the disease-associated
changes of a promising biomarker. Methods Mol Biol 507:320
43. Murgatroyd C, Patchev AV, Wu Y, Micale V, Bockmuhl Y, Fischer D, Holsboer F, Wotjak CT,
Almeida OF, Spengler D (2009) Dynamic DNA methylation programs persistent adverse
effects of early-life stress. Nat Neurosci 12(12):15591566
44. Cooney CA (2007) EpigeneticsDNA-based mirror of our environment? Dis Markers
23:121137
45. Tabb MM, Blumberg B (2006) New modes of action for endocrine-disrupting chemicals. Mol
Endocrinol 20:475482
46. Saluz HP, Jiricny J, Jost JP (1986) Genomic sequencing reveals a positive correlation between
the kinetics of strand-specific DNA demethylation of the overlapping estradiol/glucocorticoidreceptor binding sites and the rate of avian vitellogenin mRNA synthesis. Proc Natl Acad Sci
U S A 83:71677171
47. Reamon-Buettner SM, Mutschler V, Borlak J (2008) The next innovation cycle in toxicogenomics: environmental epigenetics. Mutat Res 659:158165
48. Kovalchuk O (2008) Epigenetic research sheds new light on the nature of interactions between
organisms and their environment. Environ Mol Mutagen 49:13
49. Carthew RW, Sontheimer EJ (2009) Origins and mechanisms of miRNAs and siRNAs. Cell
136:642655
50. Berezikov E, Guryev V, van de Belt J, Wienholds E, Plasterk RH, Cuppen E (2005) Phylogenetic
shadowing and computational identification of human microRNA genes. Cell 120:2124
51. Valeri N, Vannini I, Fanini F, Calore F, Adair B, Fabbri M (2009) Epigenetics, miRNAs, and
human cancer: a new chapter in human gene regulation. Mamm Genome 20(910):573580
52. Hudder A, Novak RF (2008) miRNAs: effectors of environmental influences on gene expression and disease. Toxicol Sci 103:228240
53. Timme-Laragy AR, Meyer JN, Waterland RA, Di Giulio RT (2005) Analysis of CpG methylation
in the killifish CYP1A promoter. Comp Biochem Physiol C Toxicol Pharmacol 141:406411
54. Campana SE, Thorrold SR (2001) Otoliths, increments, and elements: keys to a comprehensive understanding of fish populations? Can J Fish Aquat Sci 58:3038
55. Dickhut RM, Deshpande AD, Cincinelli A, Cochran MA, Corsolini S, Brill RW, Secor DH,
Graves JE (2009) Atlantic bluefin tuna (Thunnus thynnus) population dynamics delineated by
organochlorine tracers. Environ Sci Technol 43:85228527
56. Lake JL, Ryba SA, Serbst JR, Libby AD (2006) Mercury in fish scales as an assessment
method for predicting muscle tissue mercury concentrations in largemouth bass. Arch Environ
Contam Toxicol 50:539544
57. Georgopoulos PG, Sasso AF, Isukapalli SS, Lioy PJ, Vallero DA, Okino M, Reiter L (2009)
Reconstructing population exposures to environmental chemicals from biomarkers: challenges
and opportunities. J Expo Sci Environ Epidemiol 19:149171
58. Chamberlin TC (1965) The method of multiple working hypotheses: with this method
the dangers of parental affection for a favorite theory can be circumvented. Science
148:754759

Considerations and Criteria for the

Incorporation of Mechanistic Sublethal
Endpoints into Environmental Risk Assessment
for Biologically Active Compounds
Richard A. Brain and Bryan W. Brooks

Introduction
Awareness about the presence and unintended consequence of biologically active
organic contaminants in the environment was largely borne out of seminal observational works concerning pesticides in the early 1960s [19], eventually culminating in the establishment of protective legislation, government regulatory bodies
and a rigorous, continually improving risk assessment paradigm [79, 80]. As a
result of associated and necessary scientific advancement, a mounting inventory of
novel sublethal endpoints has materialized, which has also precariously and inadvertently highlighted the critical lack of comprehensive and cohesive regulatory
position and process with respect to consideration of these metrics in environmental
risk assessment (ERA). Although recent attention concerning this issue has been
championed largely by a relatively new term; biomarkers, the concept and importance of sublethal endpoints is certainly not new [18]. Moreover, the fundamental
concept of biological context and causality concerning sublethal effects has been
emphasized for decades, seemingly in concert with the environmental awareness
movement itself [7] with reviews on the subject dating back to the early 1970s [76].
Yet with nearly 50 years of knowledge and experience, the maturing field of sublethal effects appears to be impeded by a corresponding ecological risk paradigm
that is still comparatively less developed [89]. Although the conceptual barriers for
consideration are well known [29, 36, 56, 76, 79], the actual process and criteria

R.A. Brain (*)

Ecological Risk Assessment, Syngenta Crop Protection LLC, 410 Swing Road,
Greensboro, NC 27409, USA
e-mail: [email protected]
B.W. Brooks
Department of Environmental Science, Center for Reservoir and Aquatic Systems Research,
The Institute of Ecological, Earth and Environmental Sciences, and Institute of Biomedical
Studies, Baylor University, One Bear Place #97266, Waco, TX 76798, USA
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_7, Springer Science+Business Media, LLC 2012

139

140

R.A. Brain and B.W. Brooks

for consideration, incorporation, and integration of mechanistic sublethal effects

remain poorly defined, and in many cases completely lacking, leading to considerable uncertainty and subjectivity.
The term sublethal endpoint(s) encompasses a vast array of effects representing
a spectrum of biological complexity ranging from biochemical to physiological,
and as stated by Sprague [76] nearly 40 years ago: understanding physiological
action of a toxicant is the key to predicting important sub-lethal effects. Moreover,
Sprague [76] also asserted that biochemistry, considered as a basic level (biological strata), should be related to higher levels of organization whenever possible in
order to address the question of whether performance of an individual, or ultimately success of groups of individuals, is in turn affected. Echoed by Walker [90]
some 20 years later, if the molecular mechanism of toxicity is known then the
degree to which the chemical interacts with the target site can potentially be evaluated and possibly related to the nature and degree of toxic effect. However, within
this seemingly intuitive and logical exercise in extrapolation and correlation lies
the fundamental proviso that must be satisfied in order for sublethal effects to be
formally and effectively considered for inclusion in ERA; causally and plausibly
relating effects across biological strata. This is a critical consideration, and as
identified by Bradbury et al. [12], while studies at lower biological strata are used
to determine mechanism of action (MOA), interpretation of the relevant toxicological events for risk-management decisions is typically associated with adverse
responses observed at higher biological strata. Among the pantheon of synthetically produced chemicals, no classes have more intensively characterized and
exploited biochemical pathways than those affected specifically by pesticides and
pharmaceuticals.
Pesticides and pharmaceuticals are both classified as biologically active; however, substantial disparity among ERA paradigms exists, owing primarily to fundamental differences in nature, development, application and use, which ultimately
dictates how these compounds enter the environment; conceptual and perceptual
differences also exist. Notwithstanding these obvious differences, which do need to
be considered, certain fundamental issues are more systematic in nature and do not
segregate uniquely or exclusively based on class. Consequently, due to their collective, yet respective, biologically active natures, conceptual issues and challenges
common to both pharmaceutical and pesticide ERA should be considered in concert. And of particular joint interest is the recurring issue concerning utilization and
incorporation of sublethal effects, particularly MOA-specific data. Although the
foundation for any stressor induced effects cascade is first manifested at the biochemical level, knowing and understanding the pathway, causal linkages, and ultimate consequence across the spectrum of biological complexity requires intensive
experimental characterization. Consequently, few classes of compounds demonstrate the requisite data intensive biological profile to support such analyses; pesticides and pharmaceuticals are unique exceptions. Thus, here we will explore
considerations and criteria for the incorporation of MOA-specific sublethal effects
into ERA, focusing particularly on biologically active compounds with well established pharmacological or toxicological MOAs.

Considerations and Criteria for the Incorporation of Mechanistic

141

Overview of Sublethal Effects in Risk Assessment

Sublethal effects language and utilization has begun to permeate the risk assessment
lexicon for biologically active compounds. As a component of the US pesticide
registration process, select sublethal effects are currently considered for the purpose
of risk assessment, though as outlined in Table 1, these effects are largely gross
morphological, physiological, pathological, or histological. However, the US
Environmental Protection Agency (EPA) does exercise the option to consider additional
sublethal effects on a case-by-case basis, with the caveat that careful consideration
of the nature of the sublethal effect measure is provided and that a plausible clear
causal relationship has been established with the assessment endpoint, namely survival and reproduction [79]. This language is commensurate with the Bradford Hill
criteria for establishing causality [44] and consistent with current opinion regarding
Table 1 List of current sublethal effect measures considered by the US Environmental Protection
Agency (EPA) for use in ecological risk assessment (ERA) of pesticides for the purposes of registration as outlined in the ERA overview document
Organism
Test-type
Sublethal measurement endpoints
Invertebrate
Life-cycle
Production of young by first generation
Length of first generation
Fish

Early life-stage

Life-cycle

Embryo hatch rate

Time to hatch
Time to swim-up
Growth (length and weight)
Pathological or histological effects
Observations of other clinical signs
Embryo hatch rate
Time to hatch
Growth (length)
Exposed adult egg production
Second generation hatch rate
Second generation growth

Birds

Reproduction

Maternal weight
Eggs laid/hen
Eggs cracked
Eggshell thickness
Viable embryos
Hatchling number 14-day survivors
Gross necropsy (organ lesions, fat and muscle
deterioration)
Observations of other clinical signs

Mammals

Two-generation
reproduction

Total panel of reproduction parameters including:

histopathology, parental and offspring growth,
weight, mating, lactation, gonadal development
milestones, sexual organ performance, and
offspring production

Adapted from the USEPA [79]

142

R.A. Brain and B.W. Brooks

biomarkers [36]. Not surprisingly, the Agency anticipates further advancement in this
area as part of the continually improving state-of-the-science [79]. However, beyond
those outlined in Table 1, there has been little to no consideration, detail or guidance
concerning the nature or characteristics of potential sublethal endpoints, particularly concerning those related to MOA. That is not to say of course that MOAspecific endpoints are not considered at all, as is evidenced upon review of the
toxicological data requirements for acetylcholinesterase (AChE) inhibiting compounds under 40 CFR Part 158 [80]; developmental neurotoxicity (DNT) studies
are conditionally required based on weight of evidence. Based on a review of 20
DNT studies, 13 of which evaluated AChE inhibition, this sublethal endpoint was
found to be the most sensitive metric [80]. The MOA and physiological consequences of organophosphate-mediated inhibition of AChE are well understood [53]
satisfying the causal criteria. Moreover, as a requirement of FIFRA Part 158 Section
3(c)(2)(B), endangered species assessments are currently being required for all new
pesticide registrations and registration reviews [80], where the impacted action
area can potentially be defined based on sublethal endpoints [83] under consultation with the US Fish and Wildlife Service and National Marine Fisheries [79].
Borne out of considerations stemming from both the Food Quality Protection
Act (FQPA) and amendments to the Safe Drinking Water Act (SDWA) passed in
1996, the Endocrine Disruptor Screening Program (EDSP) was developed based on
provisions calling for the screening and testing of chemicals and pesticides for possible endocrine disrupting effects. Not surprisingly, considering the highly specific
receptor-mediated nature of endocrine active compounds, current protocols under
the EDSP incorporate MOA-specific sublethal endpoints as an integral component
of whole organism testing [82]. As outlined in OPPTS EDSP Test Guideline
890.1350 [82] measurement of plasma vitellogenin content, an indicator of estrogenic agonists when expressed in males, is required in conjunction with histological/physiological endpoints and mortality.
According to European guidance concerning birds and mammals [31] under
Relevance of endpoints in long-term toxicity tests and regarding prediction of
effects at the population level, only endpoints which are related to survival rate,
reproduction rate and development (collectively termed key factors of population
dynamics) are considered ecotoxicologically relevant. Stated vaguely in the same
section, although some sublethal endpoints assessed in mammalian tests are not
ecologically relevant, it is suggested that before disregarding biochemical effects,
lab to field extrapolative uncertainty should be considered [31]. However, no insight
regarding specific criteria for inclusion are outlined, though it is stated that transient
or reversible sublethal effects are less relevant than those that are continuous or
irreversible after exposure termination [31]. The validity of terminology within this
contention, however, falls under question in circumstance where irreversible effects
sustained subsequent to exposure termination at higher biological strata are incurred
as a consequence of a reversible sublethal endpoint such as enzyme inhibition (e.g.,
paralysis as a result of AChE inhibition). Under such a circumstance, the inclusion
of a reversible sublethal effect could be convincingly argued.

Considerations and Criteria for the Incorporation of Mechanistic

143

Under the EU aquatic ecotoxicology guidance [32], if short-term exposure leads

to sublethal effects, which are not covered by acute toxicity testing, further evaluations might be needed in such special cases. However, again, no specific criteria
concerning the nature, orientation, or appropriateness of sublethal mechanistic
effects (MOA) is provided. Only under the EU terrestrial ecotoxicology guidance
[33] is any mention or consideration of MOA made. For nontarget arthropod testing
it is recommended that for substances suspected to have a special MOA (e.g.,
insect growth regulators; IGRs) tests should include sub-lethal endpoints and may
need modifications [33], though again no specific details or guidance concerning
the criteria mentioned above are offered rendering decision nebulous and incorporation circumstantial.
For pharmaceuticals sublethal considerations are even less well defined. For
example, in the USA, if a trigger value (1 mg/L) is exceeded and an EA is
required [84], based on a number of assumptions, albeit somewhat vague, guidance does explicitly state that alternative, scientifically justified approaches can
also be used [84], though it is not stated whether this includes MOA-specific
sublethal considerations. Based on the inverse RQ methodology employed
(PNEC/PEC), if the appropriate assessment factor (AF; 1,000, 100, and 10 for
Tiers 1, 2, and 3, respectively) is not exceeded then no further testing is required,
unless sub-lethal effects are observed at the maximum expected environmental
concentration (MEEC) [84]. Furthermore, it is recommended that sublethal
effects (observed effects) at the MEEC indicate that chronic toxicity testing (Tier
3) should be performed [84]. Under Tier 3 US EPA and OECD (or other peerreviewed literature) methods and organisms are cited for reference [84], though
unfortunately no further guidance is offered, particularly concerning MOA. With
respect to European guidance concerning human medicinal products no specific
mention is made concerning sublethal effects outside of what is contained in
cited test protocols [28]; the same is true for the internationally harmonized
guidance on veterinary medicinal products [87, 88].
In addition to these efforts, US EPA has recently examined use of MOA-specific
sublethal endpoints under the statutory requirements of the US Clean Water Act.
Specifically, US EPA developed a white paper examining the role of biomarkers
related to MOA when developing National Ambient Water Quality Criteria
(NAWQC; [81]). Historically, NAWQC derivation relied on standardized ecotoxicity responses such as survival, short-term growth or invertebrate reproduction [77].
Minimum data requirements for acute and chronic NAWQC include toxicity data
from eight different organisms representing various trophic levels of an aquatic food
web [77]. The impetus for developing this document included considering the use
of MOA related endpoints for contaminants of emerging concern, which include
pharmaceuticals and endocrine disrupting chemicals (EDCs). EPA specifically
stated that the Good Science clause of the Guidelines (e.g., [77]) provides the
flexibility to adopt procedures that will produce a technically rigorous and protective criterion [81]. They further concluded,
Chronic test data and other data should be examined to determine whether, for the specific
chemical or MOA, endpoints beyond those traditionally used for criteria derivation may

144

R.A. Brain and B.W. Brooks

have intrinsic biological importance and therefore could be used as a basis for defining
threshold of effect (e.g., sex ratio). Specifically, in the context of EDCs:

Other endocrine-sensitive endpoints (e.g., VTG, testis-ova) should be examined to determine whether they can be relied upon as definitive indicators of
other biologically important endpoints (e.g., reproduction), with the idea that
they may be incorporated into calculation of the criterion. Important sources of
this information would include full lifecycle tests in which these other endpoints
were measured alongside traditional chronic endpoints, and may include tests
with other chemicals with the same MOA (e.g., E2 for EE2).
If endpoints, such as VTG or testis-ova, are used as direct or indirect indicators
of effect, it is critically important that the baseline condition (e.g., variation during normal development) be understood sufficiently to define when changes are
biologically meaningful.
Selection of appropriate endpoints (and their associated effect thresholds) may,
in some instances, transcend biological importance (the focus of the Guidelines)
to reflect societal concerns (e.g., physical appearance of wild-caught fish) [81].
Following a favorable review by the US EPA Science Advisory Board, the US
EPA plans on developing a technical support document on deriving aquatic life criteria for CECs.

Establishing Causality and Confidence

Forty years ago Sprague [76] indicated that it is relatively easy to document small
changes within an animal, but there is often a question whether the changes are deleterious, or merely within the normal range of adaptation of the animal. Given that the
targets of protection in ecological risk assessment are characteristically populations,
communities, and ecosystems, but only rarely individuals, sublethal mechanistic
responses must be consistently, systematically, and plausibly linked to responses at
these higher strata if such metrics are to be used as reliable and consequential effects
indicators [36]. Bridging this span of uncertainty in the effects spectrum requires convincingly demonstrating causality, the principals for which were first established by
Bradford Hill and Sir Richard Doll [25, 44]. Commensurate with these principals, the
US EPA maintains that clear, reasonable, and plausible links between sublethal effects
and survival or reproductive capacity of organisms in the field must be firmly established in order to influence the confidence of the overall risk assessment conclusions
[79]. However, addressing this caveat biologically requires substantial improvements
in our understanding of how mechanistic processes at each level are functionally integrated in terms of whole-organism performance [36]. This is the so called burden of
proof, and the more intensively characterized the response is mechanistically, the
greater the potential weight of evidence for establishing causal linkages.
For any given toxicological response, effects cascades are initially manifested at
the biochemical level regardless of how specific (e.g., receptor mediated) or generic
(e.g., narcosis) the MOA. If biological complexity is conceptualized as an inverted

Considerations and Criteria for the Incorporation of Mechanistic

145

Fig. 1 Inverted biological

cascade pyramid and
associated effect measures
demonstrating the concept of
causality in weight of
evidence approaches
employed with ecological risk
assessments of biologically
active molecules (e.g.,
pharmaceuticals, pesticides)

pyramid (Fig. 1) with the most fundamentally basic organic building blocks (e.g.,
biomolecules) at the apex, progressing upwards from this foundation, tissues are
derived, followed by organs, organ systems, organisms, and finally collections of
similar and diverse organisms [12]. Consequently, effects at the apex or hypothetical biological foundation have the potential to cascade or reverberate across higher
levels of biological organization. However, the degree to which higher strata are
affected depends entirely upon the significance (biological consequence) of effects
at lower strata and ultimately how tightly these tiered or stratified effects are related
biologically; this in effect defines the biological cascade (Fig. 1). The term sublethal can be defined at various points along the inverted biological cascade pyramid (i.e., below individual; Fig. 1), and the greater the separation of strata along
this axis the more difficult it is convincingly, consistently and reliably to establish
causality. Not surprisingly, the greater the number of strata measured in a given
assay, the greater the potential to establish causality, thus dictating the strength in
weight-of-evidence, and consequently, the confidence in the sublethal endpoint(s)
of interest. Moreover, testing effects at different strata over a range of exposures
facilitates establishment and comparison of concentrationresponse trends, which
accordingly also dictates the strength of causality based on the characteristics of the
aforementioned trends (shape, range, comparative sensitivity, etc.). In this context,
we identify essentially five critical or core elements that need to be established in
order to foster confidence in sublethal effects for ERA consideration, (1) biological
plausibility (causality), (2) sensitivity, (3) biological consequence, (4) effect concurrence, and (5) diagnostic capacity (Fig. 2). First, the more closely response trends
evaluated at multiple strata simultaneously track one another (respond or scale proportionally with concentration as either congruent or inverse functions), the greater
the causal strength of their collective association (for comparison see Fig. 3a, c).
Second, the greater the resolution afforded by a sublethal response in comparison to
higher biological strata (with the caveat that differential sensitivity becomes bound
maximally by relevance), the greater the utility in predictive power (for comparison
see Fig. 3a, b). Third, the greater the biological consequence conveyed to higher
strata (biologically meaningful effects) resultant from the sublethal response, the
more consequential the endpoint. Fourth, the greater the synchronization in response
manifestation between effects at different biological strata (synchronized temporally and/or concentration-dependently), the stronger the relationship. Finally, the

146
Fig. 2 Five general criteria or core elements
required to confer confidence in selecting a
sublethal endpoint of interest (e.g., mechanism
of action (MOA)) in relation to effects at higher
biological strata during ecological risk
assessments of biologically active molecules
(e.g., pharmaceuticals, pesticides)

R.A. Brain and B.W. Brooks

Biological
Plausibility

Confidence in
Mechanistic SubLethal Endpoint

more unique in nature the sublethal biological effects signature the greater the
potential diagnostic capacity, and consequently, the greater the utility within the risk
assessment paradigm. If a sublethal concentrationresponse trend does not proportionally reflect, and concurrently manifest with those derived at higher strata and
with convincing biological consequence, or if the sublethal effect measure does not
afford greater sensitivity (predictive ability), it is of little value to RA beyond potential toxicological diagnostics, which, however, can be extremely important. The
sensitivity criterion is somewhat subjective and requires judgment concerning the
upper bound where relevance (biological consequence) becomes questionable, particularly when considering transient or fully reversible effects. For example, reasonable lower and upper bounds of sensitivity between MOA specific- and higher-strata
effects (survival and reproduction as a reference point) may be considered as 1.5
and 10 for enzyme inhibition, based on the subset of examples detailed subsequently in the present evaluation. For downstream metabolite reduction or upstream
metabolite accumulation as a surrogate of enzyme inhibition the range may be considered broader, for example 1.5 and 50, though the broader the range the greater
the burden of strength in the causal relationship. Clearly such proposed criteria are
useful for developing testable hypotheses and thus require further study for validation for various MOAs and organisms. Moreover, depending on the nature of the
MOA-specific effect the dynamic range of sensitivity vs. relevance may vary, and
requires expert judgment.
Nowhere else has this paradigm of causal scrutiny been highlighted more than in
the arena of biomarkers. The term biomarker is somewhat ambiguous and can be
applied broadly and generally but encompasses biochemical, physiological, or ecological structures or processes (including MOA) that have been correlated or causally linked to biological effects measured at one or more levels of biological
organization [56]. No biomarker can by itself offer a complete solution and a battery
of biomarkers evaluated across the spectrum of biological resolution will likely be
necessary in order to convincingly evaluate chemical hazards [29]. Thus, how precisely an effect can be identified and/or characterized depends upon a multiparametric approach which includes biomarkers of general stress and more specific

Considerations and Criteria for the Incorporation of Mechanistic

a 120
100
Percent Response

Fig. 3 Hypothetical
concentrationresponse
curves for effect measures
representing biological strata
extremes (e.g., enzyme
inhibition vs. mortality)
demonstrating differential
sensitivity with proportional
response tracking (a), similar
sensitivity with proportional
response tracking (b), and
differential sensitivity with
unproportional response
tracking (c)

147

Enzyme
Inhibition

80

Mortality

60
40
20

100

10

1000

120

Percent Response

100
80

Enzyme
Inhibition

60

Mortality

40
20

10

100

1000

Percent Response

120
100
80

Enzyme
Inhibition

60

Mortality

40
20
0

10

100

Log10 Concentration (ug/L)

1000

biomarkers such as MOA [34]. However, measuring a suite of biomarkers will only be
useful if they are integrated into a mechanistic model with obvious links to fitness [36]
fulfilling the conditions of causality and conferring requisite weight-of-evidence
relating effects along the inverted biological cascade pyramid (Fig. 1).

148

R.A. Brain and B.W. Brooks

Mechanistic Causality in Environmental Toxicology

As eloquently stated by Bartholomew [7] and highlighted by Sprague [76] there
are a number of levels of biological integration andeach level finds its explanation
of mechanisms in the levels below, and its significance in the levels above. Due to
the hierarchy in biological complexity, investigations concerning MOA require
higher resolution sophisticated experimental procedures and equipment, and are
typically not measured concurrently with effects at higher levels of biological organization, but rather autonomously from tissue extracts, cell cultures or purified proteins. As a consequence. it is often difficult to convincingly demonstrate that
biochemical ripples result in physiological, population or community waves. Much
like an object which breaks the waters surface, the origin of the ripples can be
traced, but the impact will not be fully understood and appreciated unless the resultant waves are measured on the shoreline, simultaneously. Implicit in this analogy is
the ability to not only identify the source of the ripples, but to also chronicle and
characterize the process or processes turning them into waves. Although numerous
studies have evaluated MOA at the biochemical level or measured impacts at higher
biological strata, comparatively few studies have actually evaluated both in tandem,
rendering causality more often than not speculative or extrapolative in nature due to
fundamental gaps in the weight-of-evidence case. Often, the paucity of mechanistic
data (MOA) owes to inherent difficulties in elucidating, characterizing, and even
measuring effects at the biochemical level. Case in point, a few years after the publication of Silent Spring [19] the relationship between eggshell thinning and exposure to chlorinated organics (most notoriously dichlorodiphenyltrichloroethane;
DDT) was first established in two landmark studies [43, 66]. However, the actual
causal agent (diphenyldichloroethylene; DDE) was not identified until nearly a
decade later [54], and to this day the exact causative biochemical mechanism underlying the eggshell thinning phenomena is not completely understood, nearly 50
years after the first anecdotal reports of population decline. Thus, even when visually distinct impacts at the population level are recognized and the causative agent
identified, the underlying process or mechanism(s) triggering the biological cascade
can prove elusive even with the highest profile and most intensively researched
examples. Notwithstanding, examples concurrently or pseudoconcurrently (identical experimental conditions) measuring and relating effects at the extremes of biological strata do exist in the realm of ecotoxicology. Reviews concerning
biomarkers have been compiled for both plants [13, 29, 34] and animals [1, 36,
5658, 90]; however, the following examples will focus exclusively on MOAspecific sublethal endpoints in relation to higher biological strata.

Examples in Vertebrates (Fish) and Invertebrates

Perhaps the most prominent and compelling example causally demonstrating
weight of evidence in the biological effects cascade, indicative of the inverted complexity pyramid (biochemical peak and population base), concerns vitellogenin,

Considerations and Criteria for the Incorporation of Mechanistic

149

17a-ethinylestradiol (EE2), and fish. Mechanistically, vitellogenin is a lipoprotein

precursor of egg yolk proteins occurring naturally in the liver of oviparous vertebrate female fishes, activated through estrogen receptors by 17b-estradiol [59]. Not
expressed under normal physiological conditions in males, induction of vitellogenin (mRNA and protein) can signal exposure to estrogen-receptor agonists,
which may result in renal pathology, death, and hypothesized at the time, reproductive consequences [35, 92]. Although the proposed biological cascade made logical sense mechanistically, a convincing example causally linking vitellogenin
induction to survival, development, reproduction (fecundity), and ultimately population-level impacts remained elusive until a group of scientists decided to dose an
entire experimental lake with EE2 in Northern Canada [50]. During 3 years of
seasonal exposure to 56 ng/L EE2, male fathead minnows (Pimephales promelas)
demonstrated sustained vitellogenin protein levels three orders of magnitude
greater than male reference samples, with mRNA levels over a magnitude higher
in exposed males than in reference females [50]. Moreover, numerous associated
histological abnormalities were documented in exposed male testicular tissues
including delayed spermatogenesis, fibrosis, tubule malformations, arrested testicular development, and increased incidence of ova-testes (intersex) with the presence of primary-stage oocytes [50]. As a consequence, subsequent to the second
seasonal application, the fathead minnow population in the EE2 exposed lake collapsed, with reproductive failure continuing for an additional 2 years after EE2
additions had ceased [50]. Although it can be argued in this case that vitellogenin
induction is not an explicit MOA-related biochemical product toxicologically, the
response is unequivocally manifested from an estrogen receptor-mediated cascade
resultant from EE2 exposure, leading to histological malformations (intersex); ultimately reflected across the biological continuum at the population-level. Given
that only one concentration of EE2 was tested, an artifact of experimental design,
it is not possible to establish definitive concentrationresponse trends for comparison of relative sensitivity among biological strata within the inverted pyramid
scheme (Fig. 1). However, the associated causal linkages are robust, even though
not all five major criteria for inclusion (Fig. 2) were fulfilled.
Estrogen antagonists have also been showed to influence vitellogenin content
with causal linkages to reduced fecundity and model-forecast population declines in
P. promelas [59]. In female fish exposed to estrogen antagonists the vitellogenin
response is conceptually inverted compared to the response in males exposed to
estrogens, typified by a reduction in content, which can potentially forecast reproductive success [59]. Miller et al. [59] found that exposing fathead minnows to several estrogen antagonists (17 -trenbolone, 17 -trenbolone, prochloraz, fenarimol,
and fadrozole) resulted in concentration-dependent inhibition of vitellogenin plasma
content with strongly associated concentration-dependent reductions in fecundity.
Population modeling (based on fecundity) revealed ominous projections commensurate with those identified by Kidd et al. [50], where a 25% reduction in female
vitellogenin content would potentially exhibit a nearly 35% population decrease
after just 2 years of exposure [59]. Although population declines were model forecast, and histology was not assessed thereby weakening the causal strength along
the axis of the inverted biological effects pyramid (Fig. 1), the concomitant, and strongly

150

R.A. Brain and B.W. Brooks

correlated, concentration-dependent reductions in vitellogenin and fecundity provides strong causal evidence between these effects strata. Although the sublethal
response (vitellogenin content) proved to be of similar sensitivity to higher strata
effects (fecundity) in this case, the response tracked nearly identically and
afforded valuable diagnostic information thus largely fulfilling the five major criteria for inclusion. In contrast to the study by Kidd et al. [50], vitellogenin content in
this case can be considered directly related to MOA. Estrogen antagonist induced
reduction of vitellogenin in exposed female fish resulting in reduced lipoprotein
content, and consequently reduced egg production and viability, is mechanistically
consequential, whereas in males exposed to estrogen agonists, increased vitellogenin production is not directly MOA related or necessarily consequential; oocyte
formation is ultimately the mechanistic manifestation of exposure in male fish.
Among biologically active compounds, perhaps no MOA has been more intensively studied than AChE inhibition, and the corresponding body of literature
concerning MOA-specific effects for organophosphates (OPs) alone is immense.
However, relatively few studies actually measure AChE inhibition relative to
survival concurrently in the same test (see [38] for a comprehensive review), even
with mechanistic evolutions in vivo (in fish) dating back over 50 years [91]. In vertebrates and invertebrates AChEs (and in some cases butyrylcholinesterases; BChEs)
are critically responsible for deactivating the neurotransmitter acetylcholine (ACh)
via a hydrolysis reaction into choline and acetic acid [41]. In vertebrates, ACh performs numerous functions as a neurotransmitter; excitatory action in the somatic
nervous system involved in voluntary muscle control, preganglionic and postganglionic functions in the parasympathetic nervous system, and preganglionic functions
in the sympathetic nervous system [41]. The function of ACh in invertebrates is
comparatively less well characterized, though its primary function is as a neurotransmitter for afferent nerve fibers [38]. Excess build-up of ACh in the synaptic
cleft can results in overstimulation (excitation) of the post-synaptic neuron eventually leading to paralysis and potentially death [41]. However, considerable tissuespecific (e.g., brain vs. muscle) variability in sensitivity can exist [61, 78].
Perhaps the earliest study relating AChE inhibition with survival was performed
with Sheepshead minnows (Cyprinodon variegatus) exposed to Guthion, phorate,
parathion, phosphamidon, Cygon, malathion, EPN, Dursban, dichlorvos, diazinon,
Dibrom, and methyl parathion, where inhibition to below 20% functionality was
associated with median (4060%) mortality [22]. Similarly exposure of several
estuarine fish species including spot, Leiostomus xanthurus; Atlantic croaker,
Micropogon undulatus; sheepshead minnows; and pinfish, Lagodon rhomboids to
malathion, naled, Guthion, and parathion at median lethal concentrations (LC4060)
resulted in mean AChE inhibition between 70 and 96% [23]. Coppage and Matthews
[23] also found that pinfish (L. rhomboids) exposed to malathion exhibited constantly measured AChE inhibition at 7279% in replicate exposed groups with
4060% lethality at 3.5, 24, 48, and 72 h at mean exposure concentrations of 575,
142, 92, and 58 mg/L, respectively. Moreover, mean AChE activity was found to
decrease in a concentration-dependent manner with increasing exposure at multiple
time-points [24]. Although these studies do not convincingly fulfill the five major

Considerations and Criteria for the Incorporation of Mechanistic

151

criteria for inclusion, they represent pioneering studies relating MOA-specific

effects (AChE inhibition) with consequences at higher biological strata.
One caveat worth mentioning in the context of the current discussion with AChE
inhibition is that although the relationship between effects at varying biological
strata may be causally unequivocal, mechanistically speaking, based on decades of
research, disparity in sensitivity is not always uniform. For example, Fulton [37] as
described in [38] found a nearly 40-fold difference between concentrations resulting in acute lethality (96-h LC50 = 32.16 mg/L) and brain AChE inhibition (24-h
EC50 = 0.81 mg/L) in the estuarine mummichog Fundulus heteroclitus exposed to
azinphosmethyl. Thus, even when sensitivity exceeds the subjective divide or
threshold such that biological consequence cannot be convincingly demonstrated,
the mechanistic cascade may in fact be sound. In such cases a collective multispecies weight of evidence approach may be necessary, acknowledging tissue and species specific variability.
Considering the degree of historically intensive mechanistic study, there are surprisingly few current examples relating AChE inhibition beyond behavior to survival; however, Rao [65] has recently demonstrated a strong relationship in
Oreochromis mossambicus (Tilapia) exposed to chlorpyrifos. In fish exposed to
respective 24, 48, 72, and 96-h LC50s of 44, 36, 31, and 26, AChE median inhibitory
times (IT50s) increased sequentially with concentration (8, 18, 27, and 35 h, respectively) indicating temporal dependence of inhibition on concentration. Moreover ,
inversely related uniform increases in AChE activity across four recovery durations
(3, 7, 14, and 21 days, respectively) were also found, where time to recovery was
strongly and negatively correlated (r2 = 0.99) to increasing chlorpyrifos concentration [65]. A separate AChE inhibition experiment evaluated over varying durations
(4, 8, 12, 16, 20, and 24 h, respectively) of a single exposure concentration (40 mg/L)
also indicated strong temporal dependence (r2 = 0.99), where percent inhibition
increased sequentially with exposure duration [65]. Although this example does not
establish proportional biological concurrence in effect manifestation and biological
consequence across strata directly, the weight of evidence indirectly satisfies these
criteria. It is not possible, however, to compare relative sensitivity due to the differing metrics utilized (time vs. concentration), though the diagnostic capacity is
excellent.
Utilizing an alternative strategy Muniswamy et al. [61] measured ACh accumulation in relation to AChE inhibition in the freshwater fish Labeo rohita exposed to
fenvalerate. Exposure of L. rohita to experimentally determined lethal (LC50 = 6 mg/L)
and sublethal (one eighth of the LC50 = 0.75 mg/L) exposures were found to result in
time- and concentration-dependent inhibition in the activity of AChE and consequent accumulation of ACh in multiple tissues (brain, gill, liver, and muscle).
Although only two concentrations were evaluated over differing durations (1, 2, 3,
and 4 days for lethal, and 1, 5, 10, 15, and 20 days for sublethal exposures, respectively) in separate experiments, the difference in effect response was proportionally
consistent for both AChE inhibition and resultant Ach accumulation. Thus, much
like the previous example [65], several criteria for MOA-specific sublethal effects
consideration/inclusion are addressed indirectly, while others are satisfied directly.

152

R.A. Brain and B.W. Brooks

As consistently illustrated with AChE inhibition, numerous studies have

evaluated the relationship between MOA and behavioral endpoints [48, 49, 71],
though the incorporation of these metrics into ERA is somewhat contentious [68].
Justifiable arguments concerning the relevance, diagnostic capacity, and comparative
sensitivity of behavioral endpoints are evident [68], however, relating such effects
to those accepted as consequential (survival and reproduction) remain a fundamental
requirement. With this caveat in mind Kavitha and Rao [48, 49] demonstrated
time-dependent reduced locomotor behavior (distance moved) in mosquito fish
(Gambusia affinis) exposed to median lethal concentrations of monocrotophos
(LC50 = 20.49 mg/L) and chlorpyrifos (LC50 = 297 mg/L) for 96 h in separate experiments. During subsequent recovery evaluations similar time-dependent increases
in AChE activity and swimming speed were concurrently detailed for both compounds [48, 49]. In addition, for both chlorpyrifos and monocrotophos exposures,
antioxidant enzymes (catalase, superoxide dismutase, and glutathione reductase)
were all found to recover from initial inhibition in a time-dependent manner concurrently with AChE; lipid peroxidation also decreased concurrently with AChE
recovery in a time-dependent manner [48, 49]. The concurrent and proportional
response of multiple related sublethal effects after exposure to median lethal concentrations of AChE inhibitors again provides indirect evidence concerning causal
criteria and validates the utility of behavioral endpoints in the requisite context of
biological consequence.
An exemplary study demonstrating multistrata effect-correspondence in aquatic
invertebrates was conducted by Duquesne [26], where the stated goal was to
specifically address the issue of translating effects through different levels of biological organization. Duquesne [26] exposed Daphnia magna to paraoxon-methyl, the
metabolite of parathion, for 24 h and monitored AChE activity, survival, body size,
reproductive performance, and population growth rate, representing multiple strata
in the inverted biological effects cascade pyramid scheme (Fig. 1). Exposure of D.
magna to paraoxon-methyl was found to pseudoconcurrently (separate experiments
but identical conditions) inhibit AChE activity and decrease survival in a concentration-dependent manner [26]. At exposures of 1.0 mg/L paraoxon-methyl or above
AChE activity was reduced by 70%, whereas survival decreased significantly at or
above exposures of 2.2 mg/L; respective EC50 and LC50 values after 24 h of exposure
were 0.7 and 2.3 mg/L, comparatively, indicating that the sublethal response (AChE
inhibition) was three times more sensitive than survival. Exposure to 1.5 mg/L resulted
in significantly reduced body size 6 days after 24-h exposure and reproduction (number of offspring per surviving individual) was impaired 89 days subsequent with an
associated 14-day EC50 of 2.0 mg/L [26]. Hence, as indicated by Duquesne [26] the
suborganismal effects (e.g., transient inhibition of AChE) were concomitantly
accompanied by effects at the organismal (survival, reduction in reproductive performance, decrease in body size) and population (reduced population growth rate) levels. In this case AChE inhibition was measured in a separate experiment than survival
and reproduction, and although the experiments were done with the same test organism under the same conditions, ideally it is preferable to evaluate effects at multiple strata simultaneously in the same test. Nevertheless, by evaluating effects at multiple

Considerations and Criteria for the Incorporation of Mechanistic

153

biological strata in a concentration-dependent and physiologically plausible manner

and demonstrating differential sensitivity the causality case is compelling and four of
the major criteria for sublethal effects (MOA) inclusion are robustly fulfilled.
A Similar example of multistrata effect-correspondence has also been demonstrated in terrestrial invertebrates [62]. In brown planthoppers (Nilaparvata lugens)
exposed to azadirachtin (0.25, 0.5, and 1.0 ppm), mortality in adult females was
found to increase in a concentration-dependent manner with an LC50 of 0.47 ppm.
In a separate exposure series (0.1, 0.25, and 0.5 ppm) under the same conditions
AChE activity, female body weight, and fecundity all demonstrated concentrationdependent reductions [62], although AChE activity of adult females was significantly
inhibited at only 0.25 and 0.5 ppm, whereas fecundity and female-weight were
found to be significantly reduced at all exposure concentrations. In addition, evaluation of ovary histology indicated disruption to follicle epithelial cells at 0.25 ppm
and destruction at 0.5; morphological abnormalities were also noted at 0.5 and
1 ppm azadirachtin. In this case, the results indicate that the biochemical endpoint
(AChE inhibition) demonstrated sensitive similar to that of survival, histology and
gross morphology, though less sensitive than somatic or reproductive endpoints.
Thus, the mechanistic endpoint does not convincingly fulfill all five criteria for
inclusion here; however, the causal relationship across biological strata is sound and
the diagnostic capacity is valuable.

Examples in Plants
Commensurate with AChE inhibition in vertebrates and invertebrates, inhibition of
photosynthesis is arguably the most intensively studied MOA in plants. Research
concerning the MOA of photosystem II (PSII) inhibiting herbicides such as s-triazines
date back to the 1950s [60], where it was first posited that these compounds disrupted
the Hill reaction (photoreduction of an electron acceptor by electrons and protons
originating from water and resulting in the evolution of oxygen). Subsequent research
specifically suggested inhibition of noncyclic photophosphorylation [75]; however,
the ultimate target site was not confirmed until 1980s [47] as blockage of electron
flow between the primary acceptor (Q) and the secondary acceptor (B), now formally
known as the QB binding site of the D1 protein and plastoquinone, respectively.
Considering the previously mentioned protracted timeframe concerning MOA discovery,
there are few examples actually detailing concurrent measurement of PSII and morphological (growth) inhibition, until recently. Utilizing a relatively new technique
(chlorophyll fluorescence) to evaluate PSII inhibition Magnusson et al. [55] found very
similar concentrationresponse curve shapes and slopes among growth rate, biomass
and PSII efficiency (effective quantum yield) for estuarine diatoms (Navicula sp.)
and green algae (Nephroselmis pyriformis) exposed to diuron, hexazinone, and atrazine. Photosystem II efficiency is a parameter which essential measures the proportion of chlorophyll absorbed light associated with PSII that is used in photochemistry
and provides a measure of the linear electron transport Maxwell and Johnson [94].

154

R.A. Brain and B.W. Brooks

Among the three metrics evaluated, biomass was the most sensitive for Navicula sp.
whereas PSII efficiency was the most sensitive for N. pyriformis; growth rate was the
least sensitive in all cases [55]. After 72 h of exposure the relationships between
growth rate, biomass and PSII efficiency were linear and correlated with r2 0.90 for
each species and herbicide concentration [55]. Moreover the correlation regression
slopes were near unity for both species indicating good agreement between endpoints over concentrations spanning three orders of magnitude and for two very different organisms [55]. As stated by Magnusson et al. [55] these results directly link
inhibition of PSII mechanistically with declines measured in endpoints at higher
strata (growth rate and biomass), which are used routinely by industry and regulators.
These results also convincingly satisfy the five consideration criteria, particularly
concurrence, consequence and biological plausibility of effect. Diagnostic resolution
is afforded by measuring a mechanistic surrogate for plastoquinone competition
(electron transfer), and although sensitivity of the mechanistic endpoint was comparable to biomass, PSII efficiency was up to 1.8-fold, and on average 1.5-fold more
sensitive than the standard regulatory endpoint (growth rate) based on comparison of
EC50s [55].
Perhaps considered comparatively less intuitive in nature than the case with
herbicides, fairly robust examples systematically linking metabolite accumulation
and depletion to growth inhibition have recently been demonstrated in plants
exposed to pharmaceuticals. Much like the previous example [55], Brain et al. [14]
characterized very similar concentrationresponse curve shapes between inhibition of sterol biosynthesis (stigmasterol and b-sitosterol) and biomass production
in Lemna gibba exposed to statin blood lipid regulators (atorvastatin and lovastatin). In higher plants the target enzyme 3-hydroxy-3-methylglutaryl coenzyme-A
reductase (HMGR; analogous to the human receptor) regulates cytosolic isoprenoid
biosynthesis in the mevalonic acid (MVA) pathway [6], ultimately responsible for
the synthesis of sterols, which are critical components of plant membranes also
regulating morphogenesis and development [39, 73]. As a consequence of statininduced, concentration-dependent reductions in sterol production in exposed plants
in vivo, biomass (growth) was similarly and concurrently inhibited; however, the
mechanistic endpoint (sterol reductions as a result of HMGR inhibition) was two
(atorvastatin exposed) to nearly ten times (lovastatin exposed) as sensitive based
on comparison of respective EC50s [14]. This relationship established between
biologically stratified endpoints [14] is supported by a thoroughly characterized
biochemical pathway [74] with known implications resulting from disruption [39,
73] thus satisfying the remaining confidence and incorporation criteria of plausibility and diagnostics.
In a conceptually similar study Brain et al. [16] showed an inverse relationship
between metabolite production and growth inhibition in L. gibba exposed to the
sulfonamide antibiotic sulfamethoxazole. In bacteria sulfonamides specifically target the enzyme dihydropteroate synthase (DHPS) in the folate biosynthetic pathway, which was recently established as identical to that of plants [8]. In plants, as in
all living organisms, folates (Vitamin B9) are responsible for a host of functions [42,
67], particularly as essential molecules mediating the transfer of one-carbon units

Considerations and Criteria for the Incorporation of Mechanistic

155

(C1 metabolism) in metabolic pathways that are of paramount importance to cellular

viability. As an indirect diagnostic surrogate measure of DHPS inhibition, Brain
et al. [16] quantitatively assessed an upstream metabolite (p-aminobenzoic acid;
pABA), and found accumulation levels that were 18 and 39 times more sensitive to
sulfamethoxazole exposure in vivo than biomass and frond number morphological
endpoints, respectively, based on EC50s [16]. Moreover, the characterized exponential rise in pABA accumulation was manifested concurrently and proportionally
with growth inhibition thus satisfying biological consequence and rounding out satisfaction of the five confidence and incorporation criteria.

Approach and Strategy: Orienting the ERA

According to the decision tree outlined in Fig. 4, MOA must first be characterized
biochemically, and second identified (or likely present) in the nontarget organism(s)
of interest, prior to further consideration in the ERA process (see [4, 15]). In the
context of predictive modeling and intelligent testing [21], pragmatically orienting,
focusing, and prioritizing testing efforts requires an in-depth understanding of
MOA [12]. However, implicitly underlying this contention is the qualification of
adequately discerning MOA, ranging in nature from receptor-mediated to narcosis
(baseline toxicity). As a generality, the more specific the MOA the narrower the
biological range of potentially affected nontarget organisms; conversely, the more
general the MOA, the broader the biological range of potential nontarget organisms. Consequently, a more fundamentally important question is whether or not the
nontarget organism of interest expresses an appropriate and susceptible receptor
(see [46]). Perhaps the most definitive methodology to address this question lies
within the technology known as omics (e.g., genomics and proteomics) [52].
Detailing the intensive process of MOA discovery during chemical development is
beyond the scope of the current discussion; hence we shall focus explicitly on
methodological strategies for MOA identification in potentially susceptible nontarget organisms here. Integral to this process is the concept of evolutionary conservation (receptor homology; genetic similarity), within and among different taxa [20,
21, 27, 40, 51, 52]. As outlined by Kwekel et al. [52], the availability of complete
genome sequences for multiple species provides unprecedented opportunities for
comprehensive comparative analysis in support of mechanistic and predictive toxicology However, simply comparing gene sequences is not enough; rather,
investigations of orthologous gene relationships based on sequence similarity
(reciprocal BLAST (Basic Local Alignment Search Tool) best hit), synteny (conserved order of genes), phylogenetic tree matching (organism-level relatedness),
and functional complementation (conservation of molecular function) across species are a recommended approach [52]. This process is most pragmatically pursued
using focused, gene-specific, and hypothesis-driven investigations, and several
queriable genetic and proteomic resources are currently available for this purpose
and reviewed by Kwekel et al. [52].

156

R.A. Brain and B.W. Brooks

Is the mechanism of action known?
Yes
No
Identify major taxonomic groups likely to
have homologous receptor (MoA) and
identify appropriate sub-lethal response
to measure (e.g. biomarker)

Characterize
biochemistry
of pathway
(identify MoA)

Is appropriate species (taxonomic

group) represented in core test set?
No
Yes
Focus testing intensity on
likely impacted species
(taxonomic groups)

Supplemental tests on
appropriate non-core
representative species

Are effects measured and observed at higher biological strata

according to the inverted biological cascade pyramid scheme
(Figure 1; e.g. survival, reproduction, and energy flow)?
Yes
No
Is the mechanistic sub-lethal response: Designate
1) Plausibly and consistently linked to - No Effect
2) More sensitive than 3) Biologically consequential for 4) Diagnostic towards 5) Manifested concurrently with -effects at higher biological strata?
No
Yes
Incorporate mechanistic sub-lethal
response into ERA:
1) Compliment traditional endpoints
2) Reduce uncertainty factors
3) Establish cause and effect
4) Provide biological insight

Conduct ERA
with traditional
parameter(s)

Fig. 4 Decision flowchart concerning the incorporation of MOA-specific considerations and sublethal effects into the collective ecological risk assessment process for biologically active compounds (e.g., pharmaceuticals, pesticides)

An exemplary illustration of receptor homology is provided by Gunnarsson

et al. [40], where best BLASTP hit (protein-based BLAST search) was utilized to
evaluate possible sequence alignments for 1,318 unique human target orthologs
(135 drug targets for a total of 1,152 drugs) in 16 species with representation from
vertebrates, invertebrates, arthropods, plants, yeast, and bacteria. Among 5 functional categories (enzyme, receptor, ion channel, transporter, and other) enzyme
groups were found to have the largest intersections, sharing 53 drug targets, suggesting these targets are highly conserved and ecotoxicologically relevant [40].

Considerations and Criteria for the Incorporation of Mechanistic

157

Among classes, vertebrates were found to have the highest similarity in target
homology with human drug target orthologs. In the context of species currently
used for aquatic environmental risk assessments, fish and frogs were predicted to
have by far the greatest number with the highest degree of similarity [40], which
has also been corroborated by Christen et al. [20]. For example, Zebrafish had
orthologs to 86% of the drug targets, whereas only 61% were conserved in Daphnia
and 35% in green algae [40]. Thus, orthology prediction can be used as an effective
tool to identify potential receptors in nontarget organisms, and additionally refine
and orient effects testing efforts to the most relevant species. As with any technology, however, caution must be exercised as low sequence homology is not an absolute predictor of receptor-mediated effect or lack of effect. As an example
HMG-CoA reductase was only found to be 30% homologous between humans and
plants [40], yet research suggests conservation of function as observed for strong
inhibition by human targeted drugs [14]. Thus, after identifying whether a conserved drug target is present in nontarget organisms, it is critical to define whether
the function is also conserved [4]. Notwithstanding, -omics approaches hold promise and will only become more refined and specific as advances are made. In fact,
the Adverse Outcome Pathways (AOP) concept represents a robust approach for
ecological risk assessments of pharmaceuticals, pesticides and other contaminants
[5, 45]. For example, Villeneuve and Garcia-Reyero [89] further explore the utility
of -omics and AOP in several predictive ecotoxicology applications.
Prior to pursuing omics technology, however, addressing the more obvious and
intuitive consideration of taxonomy (or life history) provides an initial, albeit coarse,
point of reference. For example, nontarget effects of herbicides would generally be
most logically pursued among plants, whereas the effects of estrogenic compounds
would more appropriately be pursued among nontarget animals (e.g., vertebrates).
Fortunately, between these two extremes lie intermediate methodologies to further
aid and address these questions, for example, quantitative structure activity relationships (QSARs) and acute to chronic ratios (ACRs).
If the MOA is unknown QSARs afford a means to at least suggest the nature of
effect by classifying a compound into four major potential classes: nonpolar narcotic compounds (I), polar narcotic compounds (II), reactive compounds (III), and
compounds with a specific mode of action (IV; e.g., pharmaceuticals, pesticides)
[86]. This general classification strategy has been applied to numerous databases
representing a diverse array of compounds [1012, 30, 70, 85, 86]. Using toxic
ratios (the ratio of predicted and measured toxicity) Vaal et al. [85] found that reactive compounds and compounds with a specific mode of action were a factor of
10100,000 more toxic than predicted, which was found to be seemingly independent of corresponding log Kow values, a critical determinant in QSAR modeling.
Thus for a given compound, higher toxic ratios are potentially indicative or at least
suggestive of a specifically acting MOA. Moreover, toxic ratios applied to a base
subset of organisms (e.g., fish, invertebrates, and plants) can potentially provide
insight into which major taxonomic groups are likely to be impacted, or those most
likely to contain an appropriate receptor. Similar in concept to QSARs, chemical
read-across approaches, which facilitate inferences about potential toxicity based

158

R.A. Brain and B.W. Brooks

on structural similarity to chemicals with known toxicity profiles, are also gaining
favor, particularly for prioritization [21] and potentially regulatory efforts in the
USA [45] and in the EU under REACH [93].
In a conceptually similar fashion, ACRs (e.g., ratio of the LC50 and NOEC or
LOEC) have demonstrated utility in distinguishing between specific-acting and
generally acting MOAs [69]. In an analysis of nonpolar narcosis (baseline toxicity),
polar narcosis, a-specific reactivity, specific reactivity (receptor mediated), and
heavy metals, Roex et al. [69] demonstrated that specifically acting chemicals typically have larger average ACRs, though variability was also large. Ahlers et al. [2]
found less definitive trends; although narcosis (polar and particularly, nonpolar narcosis) was considered a useful predictor for low ACRs, a nonnarcotic MOA was not
considered a reliable indicator of high ACR. However, partitioning the dataset
according to specific structural alerts (SAs; defining chemical groups such as phenols, amines, esters etc.), compounds containing at least one SA had a substantially
increased probability for a high ACR [2]. Thus it was suggested that a scheme combining both MOA and SA knowledge could potentially better discriminate between
low and high ACRs [2].
When an ACR value is unknown, default values are often employed for regulatory purposes. For example, Raimondo et al. [63] identified a 90th centile ACR
value of 79.5 for aquatic contaminants. In fact, Rand [64] suggested that the larger
the size of an ACR, the greater the likelihood of a chemical acting through a specific
MOA. When ACRs were considered for pharmaceuticals, Sanderson and Thomsen
[72] suggested that an ACR of 100 may be adequate for estimating the chronic
responses of Daphnia sp. and algae because nonspecific, narcosis MOAs may be
appropriate. In these organisms, nonspecific, narcosis MOA are more likely to be
observed based on relatively lower conservation of drug targets than aquatic vertebrates [4, 17, 40, 46], though, as noted above, the effects of antibiotic effects to
plants and algae represent noticeable exceptions [15, 16]. However, it is critical to
note that much higher ACR values (e.g., >1,000,000) have been reported for some
pharmaceuticals when sublethal chronic fish responses (e.g., not 7 day juvenile P.
promelas growth) are plausibly linked to pharmacological MOAs [3]. Berninger
and Brooks [9] noted that a default ACR value of 100 could represent just a 20th
centile value for chronic effects of pharmaceutical on fish, when MOA related endpoints are used to calculate an ACR. Here again, it may be possible to leverage
mammalian therapeutic information using biological read-across approaches,
which could identify classes of pharmaceuticals presenting the greatest potential
hazards to fish. For example, maximizing the pharmacological margin of safety
(MOS) is an important consideration during the development of pharmaceuticals.
However, as demonstrated by Berninger and Brooks [9], compounds with larger
MOS values are often more potent such that higher MOS values may be predictive
of larger ACRs in fish.
Once a chemicals MOA has been characterized and candidate nontarget species with known or suspected (target) susceptibility have been identified, effects
testing can be taxonomically focused and the relationship nature between MOA

Considerations and Criteria for the Incorporation of Mechanistic

159

and effects at higher biological strata established (Fig. 4). Establishing or defining
relationship nature requires satisfaction of the five criteria of MOA incorporation outlined previously (Fig. 2) and providing a robust causal weight-of-evidence
case according to the biological effects cascade pyramid scheme (Fig. 1). If the
sublethal mechanistic response is purely transient and without relevant consequence for higher biological strata (e.g., fitness and survival) consideration for
regulation, or establishment of life-criteria cannot be justified. Ultimately, this
debate must address the so what question; if an affect measured at lower strata
is temporary and/or reversible without direct consequence at the community, population, or even organismal level why is it important? Unfortunately, the threshold
at which sublethal responses become consequential or relevant is unavoidably
subjective, but can be defined as the concentration beyond which irreparable
impact to higher biological strata are predicted to occur via biological chain-reaction
or cascade. As a conservative estimate suggested here, a bracketing range of MOA
sensitivity between 1.5 and 10 is subjectively considered as being predictively
useful for, yet still consequentially relevant to, effects at higher biological strata
(based on enzyme inhibition). However, justification for basing ERA thresholds
on sublethal mechanistic effects instead of lethality is fundamentally and causally
limited, and thus direct and absolute substitution for traditional threshold is not
explicitly recommended. Mechanistic sublethal effects, or for the sake of argument any sublethal effect in general, cannot singularly be used to replace traditionally measured effects at higher-strata (survival and reproduction) for the
purposes of ERA given the requisite need to establish cause-and-effect relationships (relate) to these metrics. Thus, sublethal responses, by virtue of requiring
validation in effects of regulatory consequence (survival and fecundity), are not
capable of circumventing traditional test metrics, rather their predictive and diagnostic capability provides a powerful and invaluable foundation for understanding
process, cause-and-effect, and addressing uncertainty surrounding true threshold
tolerance. Toxicity values typically employed in ERA range considerably in terms
of tolerable or acceptable impact criteria from highly conservative (e.g., NOAELs
and LOAELs or LC5s and LC10s) to less conservative (e.g., LC25s and LC50s)
depending on the nature and goals of the assessment. Thus the case could be made
that simply using a more conservative value based on traditional metrics and traditional uncertainty assessment could be an equally effective approach, depending
on the magnitude of the uncertainty factors used for extrapolation from acute to
chronic effects. However, this contention systematically ignores the value in
mechanistically based cause-and-effect relationships across multiple biological
strata. In a practical sense then, incorporation of well defined sublethal effects
values could more appropriately be considered for the purposes of reducing uncertainty (addressing and alleviating arbitrary application factors) rather than strictly
replacing the traditional endpoint out of principal to further proliferate unnecessary and unrealistic conservative precaution. The fundamental caveat inherent in
the previous statement of course is explicitly contingent on convincingly demonstrating the five hypothetical criteria for inclusion outlined here. If ACR values

160

R.A. Brain and B.W. Brooks

are developed for various MOAs, then ERAs of biologically active molecules will
benefit from the application of science-based, rather than simple default, uncertainty factors.

Conclusions
Under the current ERA paradigm for biologically active compounds accepted and
required sublethal effects data are largely composed of fecundity-based, grossnecropsy, and pathology measures. Conversely, incorporation of sublethal effects
data from lower biological strata remains a highly contentious issue due to lack of
established guidance concerning formal criteria for acceptance. In the present
assessment a methodological framework is proposed which consists of five formal
criteria, intended as acceptability considerations concerning the causal weight of
evidence supporting the incorporation of MOA-specific data into ERA process;
plausibility and consistent linkage, comparative sensitivity, biological consequence,
diagnostic capacity, and temporal, concentrationresponse concurrence. In fact, the
approach presented here is consistent with the AOP concept just recently developed
by Ankley et al. [5]. Although our criteria manifested specifically in consideration
of MOA, the methodology can be broadly applied to effects assessed at lower
biological strata in general. Depending on the adequacy or degree to which the
suggested criteria are satisfied experimentally dictates the strength of the causal
weight-of-evidence case, ultimately providing justification for incorporation. The
underlying fundamental premise underlying this methodology derives from the concept of biological effects cascading. It is argued that invariably every effect realized
and measured at higher biological strata is first manifested at the biochemical level
and essentially resonated and magnified up through higher biological tiers conceptually analogous to an inverted pyramid. If effects measured at lower biological tiers
cannot be relevantly and consequentially linked to those measured at higher strata,
there is effectively no justification for incorporation or further pursuit of the effect(s)
in question. Once causal strength has been established, the nature of sublethal
effects incorporation is suggested to be predictive, pre-emptive and diagnostic.
Rather than empirically and systematically replacing traditional endpoints for the
purposes of conservatism, utilization of sublethal effects data is recommended to
reduce uncertainty, address and alleviate arbitrary application factors, and emphasize cause-and-effect relationships across multiple biological strata. The process
suggested here can be preemptively refined based on intelligent testing methodologies where taxonomy, QSARs, ACRs, and more specifically genomics and proteomics technologies can be effectively utilized to orient, focus, and refine testing
efforts by predicting candidate nontarget organisms or groups of organisms most
likely to be susceptible to a given stressor of interest. Ultimately, the fundamental
goal of sublethal effects generation and incorporation should center on uncertainty
reduction not propagation.

Considerations and Criteria for the Incorporation of Mechanistic

161

References
1. Adams SM, Giesy JP, Tremblay LA, Eason CT (2001) The use of biomarkers in ecological risk
assessment: recommendations from the Christchurch conference on biomarkers in ecotoxicology. Biomarkers 6:16
2. Ahlers J, Riedhammer C, Vogliano M, Ebert R-U, Kuhne R, Schuurmann G (2006) Acute and
chronic ratios in aquatic toxicity-variation across trophic levels and relationship with chemical
structure. Environ Toxicol Chem 25:29372945
3. Ankley GT, Black MC, Garric J, Hutchinson TH, Iguchi T (2005) A framework for assessing
the hazard of pharmaceutical materials to aquatic species. In: Williams R (ed) Science for
assessing the impacts of human pharmaceutical materials on aquatic ecosystems. SETAC
Press, Pensacola, pp 183238
4. Ankley GT, Brooks BW, Hugget DB, Sumpter JS (2007) Repeating history: pharmaceuticals
in the environment. Environ Sci Technol 41:82118217
5. Ankley GT, Bennett RS, Erickson RJ, Hoff DJ, Hornung MW, Johnson RD, Mount DR,
Nichols JW, Russom CL, Schmieder PK, Serrano JA, Tietge JE, Villeneuve DL (2010) Adverse
outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment. Environ Toxicol Chem 29:730741
6. Bach TJ, Lichtenthaler HK (1982) Inhibition of mevalonate biosynthesis and of plant growth
by the fungal metabolite mevinolin. In: Wintermanns JFGM, Kuiper PJC (eds) Biochemistry
and metabolism of plant lipids. Elsevier Biochemical Press, Amsterdam, pp 515521
7. Bartholomew G (1964) The roles of physiology and behavior in the maintenance of homeostasis in the desert environment. In: Hughes GM (ed) Symposia of the society for experimental
biology, vol 18, Homeostasis and feedback mechanisms (p 460). Academic, New York,
pp 729
8. Basset GJC, Quinlivan EP, Gregory JF, Hanson AD (2005) Folate synthesis and metabolism in
plants and prospects for biofortification. Crop Sci 45:449453
9. Berninger JP, Brooks BW (2010) Leveraging mammalian pharmaceutical toxicology and
pharmacology data to predict chronic fish responses to pharmaceuticals. Toxicol Lett
193:6978
10. Bradbury S (1994) Predicting modes of toxic action from chemical structure: an overview.
SAR QSAR Environ Res 2:89104
11. Bradbury S (1995) Quantitative structure-activity relationships and ecological risk assessment:
an overview of predictive aquatic toxicology research. Toxicol Lett 79:229237
12. Bradbury S, Feijtel T, van Leeuwen C (2004) Meeting the scientific needs of ecological risk
assessment in a regulatory context. Environ Sci Technol 38:463A470A
13. Brain RA, Cedergreen N (2009) Biomarkers in plants: selection and utility. Rev Environ
Contam Toxicol 198:49110
14. Brain RA, Reitsma TS, Lissemore LI, Bestari B-J, Sibley PK, Solomon KR (2006) Herbicidal
effects of statin pharmaceuticals in Lemna gibba. Environ Sci Technol 40:51165123
15. Brain RA, Hanson ML, Solomon KR, Brooks BW (2008) Aquatic plants exposed to pharmaceuticals: effects and risks. Rev Environ Contam Toxicol 192:67115
16. Brain RA, Ramirez AJ, Fulton BA, Chambliss CK, Brooks BW (2008) Herbicidal effects of
sulfamethoxazole in Lemna gibbaG3: using pABA as a biomarker of effect. Environ Sci
Technol 42:89658970
17. Brooks BW, Ankley GT, Hobson JF, Lazorchak JM, Meyerhoff RD, Solomon KR (2008)
Assessing the aquatic hazards of veterinary medicines. In: Crane M, Barrett K, Boxall A (eds)
Effects of veterinary medicines in the environment. CRC Press, Boca Raton, pp 97128
18. Cairns J (1966) Dont be half-safe-the current revolution in bio-assay techniques. In:
Proceedings of the 21st industrial waste conference, Purdue University, Engineering Extension
series, Richmond, pp 559567
19. Carson R (1962) Silent spring. Houghton Mifflin, Boston

162

R.A. Brain and B.W. Brooks

20. Christen V, Hickmann S, Rechenberg B, Fent K (2010) Highly active human pharmaceuticals
in aquatic systems: a concept for their identification based on their mode of action. Aquat
Toxicol 96:167181
21. Combes R, Barratt M, Balls M (2003) An overall strategy for the testing of chemicals for
human hazard and risk assessment under the EU REACH system. Altern Lab Anim 31:719
22. Coppage D (1972) Organophosphate pesticides: specific level of brain AChE inhibition related
to death in sheepshead minnows. Trans Am Fish Soc 101:534536
23. Coppage D, Matthews E (1974) Short-term effects of organophosphate pesticides on cholinesterases of estuarine fishes and pink shrimp. Bull Environ Contam Toxicol 11:483488
24. Coppage D, Matthews E, Cook G, Knight J (1975) Brain acetylcholinesterase inhibition in fish
as a diagnosis of environmental poisoning by malathion, O, O-dimethyl s-(1,2-dicarbethoxyethyl) phosphorodithioate. Pestic Biochem Physiol 5:536542
25. Doll R, Hill A (1950) Smoking and carcinoma of the lung: preliminary report. Br Med J
4682:739748
26. Duquesne S (2006) Effects of an organophosphate on Daphnia magna at suborganismal and
organismal levels: implications for population dynamics. Ecotoxicol Environ Saf 65:145150
27. Eads B, Andrews J, Colbourne J (2008) Ecological genomics in Daphnia: stress responses and
environmental sex determination. Nat Heredity 100:184190
28. EMEA (2005) Guideline on the environmental risk assessment of medicinal products for
human use. The European Agency for the Evaluation of Medicinal Products: Committee for
Medicinal Products for Human Use, London
29. Ernst WHO, Peterson PJ (1994) The role of biomarkers in environmental assessment (4).
Terrestrial plants. Ecotoxicology 3:180192
30. Escher B, Hermens J (2002) Modes of action in ecotoxicology: their role in body burdens,
species sensitivity, QSARs, and mixture effects. Environ Sci Technol 36:42014217
31. EUC (2002) Working document, Guidance document on risk assessment for birds and mammals under Council Directive 91/414/EEC. SANCO/4145/2000final. European Commission
Health & Consumer Protection Directorate-General, Directorate EFood safety: plant health,
animal health and welfare, international questions E1Plant health, Brussels
32. EUC (2002) Working document, Guidance document on aquatic ecotoxicology in the context
of the Directive 91/414/EEC. Sanco/3268/2001 rev.4 (final). European Commission Health &
Consumer Protection Directorate-General, Directorate EFood safety: plant health, animal
health and welfare, international questions E1Plant health, Brussels
33. EUC (2002) Working document, Guidance document on terrestrial ecotoxicology under
Council Directive 91/414/EEC. SANCO/10329/2002 rev 2 final. European Commission
Health & Consumer Protection Directorate-General, Directorate EFood safety: plant health,
animal health and welfare, international questions E1Plant health, Brussels
34. Ferrat L, Pergent-Martini C, Romo M (2003) Assessment of the use of biomarkers in aquatic
plants for the evaluation of environmental quality: application to seagrasses. Aquat Toxicol
65:187204
35. Folmar L, Gardner G, Schreibman M, Magliulo-Cepriano L, Mills L, Zaroogian G, GutjahrGobell R, Haebler R, Horowitz D, Denslow N (2001) Vitellogenin-induced pathology in male
summer flounder (Paralichthys dentatus). Aquat Toxicol 51:431441
36. Forbes VE, Palmqvist A, Bach L (2006) The use and misuse of biomarkers in ecotoxicology.
Environ Toxicol Chem 25:272280
37. Fulton M (1989) The effects of certain intrinsic and extrinsic variables on the lethal and sublethal toxicity of selected organophosphorus insecticides in the mummichog, Fundulus heteroclitus, under laboratory and field conditions. Thesis, University of South Carolina, Columbia
38. Fulton M, Key P (2001) Acetylcholinesterase inhibition in estuarine fish and invertebrates as
an indicator of organophosphorous insecticide exposure and effects. Environ Toxicol Chem
20:3745
39. Grandmougin-Ferjani A, Schuller-Muller I, Hartmann MA (1997) Sterol modulation of the
plasma membrane H+-ATPase activity from corn roots reconstituted into soyabean lipids.
Plant Physiol 113:163174

Considerations and Criteria for the Incorporation of Mechanistic

163

40. Gunnarsson L, Jauhiainen A, Kristiansson E, Nerman O, Larsson J (2008) Evolutionary conservation of human drug targets in organisms used for environmental risk assessments. Environ
Sci Technol 42:58075813
41. Habig C, DiGiulio R (1991) Biochemical characteristics of cholinesterases in aquatic organisms. In: Mineau P (ed) Cholinesterase inhibiting insecticides: their impact on wildlife and the
environment, vol 2, Chemicals in agriculture. Elsevier, New York, pp 1934
42. Hanson AD, Roje S (2001) One-carbon metabolism in higher plants. Annu Rev Plant Physiol
52:119137
43. Hickey J, Anderson D (1968) Chlorinated hydrocarbons and eggshell changes in raptorial and
fish-eating birds. Science 162:271273
44. Hill A (1965) The environment and disease: association or causation? Proc R Soc Med 58:295300
45. Hoff D, Lehmann W, Pease A, Raimondo S, Russom C, Steeger T (2010) Predicting the toxicities of chemicals to aquatic animal species (White Paper). U.S. Environmental Protection
Agency, Washington, DC, 127 p
46. Hugget DB, Cook JC, Ericson JF, Williams RT (2003) A theoretical model for utilizing mammalian pharmacology and safety data to prioritize potential impacts of human pharmaceuticals
to fish. Hum Ecol Risk Assess 9:17891799
47. Jursinic P, Stemler A (1983) Changes in [14C]atrazine binding associated with the oxidationreduction state of the secondary quinone acceptor of photosystem II. Plant Physiol 73:703708
48. Kavitha P, Rao V (2007) Oxidative stress and locomotor behaviour response as biomarkers for
assessing recovery status of mosquito Wsh, Gambusia aYnis after lethal eVect of an organophosphate pesticide, monocrotophos. Pestic Biochem Physiol 87:182188
49. Kavitha P, Rao V (2008) Toxic effects of chlorpyrifos on antioxidant enzymes and target
enzyme acetylcholinesterase interaction in mosquito fish, Gambusia affinis. Environ Toxicol
Pharmacol 26:192198
50. Kidd K, Blanchfield P, Mills K, Palace V, Evans R, Lazorchak J, Flick R (2007) Collapse of a fish
population after exposure to a synthetic estrogen. Proc Natl Acad Sci U S A 104:88978901
51. Khler H-R, Kloas W, Schirling M, Lutz I, Reye A, Langen J-S, Triebskorn R, Nagel R,
Schnfelder G (2007) Sex steroid receptor evolution and signalling in aquatic invertebrates.
Ecotoxicology 16:131143
52. Kwekel J, Burgoon L, Zacharewski T (2008) Comparative toxicogenomics in mechanistic and
predictive toxicology. In: Brown J (ed) Comparative genomics basic and applied research.
CRC Press, Taylor and Francis Group, Boca Raton, p 400
53. Levine R (1991) Recognized and possible effects of pesticides in humans. In: Hayes W, Laws
E (eds) Handbook of pesticide toxicology, vol 1. Academic, New York, pp 275360
54. Lincer J (1975) DDE-induced eggshell-thinning in the American Kestrel: a comparison of the
field situation and laboratory results. J Appl Ecol 12:781793
55. Magnusson M, Heimann K, Negri A (2008) Comparative effects of herbicides on photosynthesis and growth of tropical estuarine microalgae. Mar Pollut Bull 56:15451552
56. McCarty LS, Munkittrick KR (1996) Environmental biomarkers in aquatic toxicology: fiction,
fantasy, or functional? Hum Ecol Risk Assess 2:268274
57. McCarthy JF, Shugart LR (1990) Biomarkers of environmental contamination. Lewis
Publishers, Boca Raton
58. McCarty LS, Power M, Munkittrick KR (2002) Bioindicator vs biomarkers in ecological risk
assessment. Hum Ecol Risk Assess 8:159164
59. Miller D, Jensen K, Villeneuve D, Kahl M, Makynen E, Durhan E, Ankley G (2007) Linkage
of biochemical response to population-level effects: a case study with vitellogenin in the fathead minnow (Pimephales promelas). Environ Toxicol Chem 26:521527
60. Moreland D, Gentner W, Hilton J, Hill K (1959) Studies on the mechanism of herbicidal action
of 2-chloro-4,6-bis(ethylamino)-S-triazine. Plant Physiol 34:432435
61. Muniswamy D, Patil V, Chebbi S, Marigoudar S, Chittaragi J, Halappa R (2009) Gas-liquid
chromatography for fenvalerate residue analysis: in vivo alterations in the acetylcholinesterase
activity and acetylcholine in different tissues of the fish, Labeo rohita (Hamilton). Toxicol
Mech Methods 19:410415

164

R.A. Brain and B.W. Brooks

62. Nathan S, Choi M, Seo H, Paik C, Kalaivani K, Kim J (2008) Effect of azadirachtin on acetylcholinesterase (AChE) activity and histology of the brown planthopper Nilaparvata lugens
(Stal). Ecotoxicol Environ Saf 70:244250
63. Raimondo S, Montague BJ, Barron MG (2007) Determinants of the variability in acute to chronic
toxicity ratios for aquatic invertebrates and fish. Environ Toxicol Chem 26:20192023
64. Rand GM (1995) Fundamentals of aquatic toxicology: effects, environmental fate, and risk
assessment. Taylor & Francis, Washington, DC
65. Rao V (2008) Brain acetylcholinesterase activity as a potential biomarker for the rapid assessment of chlorpyrifos toxicity in a euryhaline fish, Oreochromis mossambicus. Environ Bioindic
3:1122
66. Ratcliffe D (1967) Decrease in eggshell weight in certain birds of prey. Nature 215:208210
67. Rbeill FD, Macherel D, Mouillon JM, Garin J, Douce R (1997) Folate biosynthesis in higher
plants: purification and molecular cloning of a bifunctional 6-hydroxymethyl-7,8-dihydropterin pyrophosphokinase/7,8-dihydropteroate synthase localized in mitochondria. EMBO Rep
16:947957
68. Robinson P (2009) Behavioural toxicity of organic chemical contaminants in fish: application
to ecological risk assessments (ERAs). Can J Fish Aquat Sci 66:11791188
69. Roex E, van Gestel C, Van Wezel AP, Van Straalen NM (2000) Ratios between acute aquatic
toxicity and effects on population growth rates in relation to toxicant mode of action. Environ
Toxicol Chem 19:685693
70. Russom C, Bradbury S, Broderius S, Hammermeister D, Drummond R (1997) Predicting
modes of toxicity action from chemical structure: acute toxicity in the fathead minnow
(Pimephales promelas). Environ Toxicol Chem 16:948967
71. Sandahl J, Baldwin D, Jenkins J, Scholz N (2005) Comparative thresholds for acetylcholinesterase inhibition and behavioral impairment in coho salmon exposed to chlorpyrofos. Environ
Toxicol Chem 24:136145
72. Sanderson H, Thomsen M (2009) Comparative analysis of pharmaceuticals versus industrial
chemicals acute aquatic toxicity classification according to the United Nations classification
system for chemicals: assessment of the (Q)SAR predictability of pharmaceuticals acute
aquatic toxicity and their predominant acute toxic mode-of-action. Toxicol Lett 187:8493
73. Schaller H (2004) New aspects of sterol biosynthesis in growth and development of higher
plants. Plant Physiol Biochem 42:465476
74. Schwender J, Seemann M, Lichtenthaler HK, Rohmer M (1996) Biosynthesis of isoprenoids
(carotenoids, sterols, prenyl side-chains of chlorophylls and plastoquinone) via a novel pyruvate/glyceraldehyde 3-phosphate non-mevalonate pathway in the green algae Scenedesmus
obliquus. Biochem J 316:7380
75. Shimabukuro R, Swanson H (1969) Atrazine metabolism, selectivity, and mode of action.
J Agric Food Chem 17:199205
76. Sprague J (1971) Measurement of pollutant toxicity to fish-III sublethal effects and safe
concentrations. Water Res 5:245266
77. Stephan CE, Mount DI, Hansen DJ, Gentile JH, Chapman GA, Brungs WA (1985) Guidelines
for deriving numerical national water quality criteria for the protection of aquatic organisms
and their uses. US EPA ORD ERL. PB 85-227049, Duluth, pp 197
78. Strum A, Wogram J, Segner H, Liess M (2009) Different sensitivity to organophosphates of
acetylcholinesterase and butyrylcholinesterase from three-spined stickleback (Gasterosteus
aculeatus): application in biomonitoring. Environ Toxicol Chem 19:16071615
79. USEPA (2004) Overview of the ecological risk assessment process in the Office of Pesticide
Programs, U.S. Environmental Protection Agency: endangered and threatened species effects
determinations. United States Environmental Protection Agency, Office of Prevention,
Pesticides and Toxic Substances Office of Pesticide Programs, Washington, DC
80. USEPA (2007) Federal Register. Part II. Environmental Protection Agency. 40 CFR Parts 9,
152, 156, 159 et al. Pesticides; data requirements for conventional chemicals, technical amendments, and data requirements for biochemical and microbial pesticides; final rules. United
States Environmental Protection Agency, Washington, DC, 56 p

Considerations and Criteria for the Incorporation of Mechanistic

165

81. USEPA (2008) Aquatic life criteria for contaminants of emerging concern. Part I: general challenges and recommendations. U.S. Environmental Protection Agency, Washington DC. http://
water.epa.gov/scitech/swguidance/waterquality/standards/upload/2008_06_03_criteria_sabemergingconcerns.pdf. Accessed 04/2010
82. USEPA (2009) Endocrine disruptor screening program test guidelines OPPTS 890.1350: fish
short-term reproduction assay. Office of Prevention, Pesticides and Toxic Substances (7101),
Washington, DC, 93 p
83. USEPA (2009) Federal Register/Vol. 74, No. 84/Monday, May 4, 2009/Rules and regulations,
p 20422. http://edocket.access.gpo.gov/2009/pdf/E9-10203.pdf. Accessed 04/2010
84. USFDA (1998) Guidance for industry; environmental assessment of human drug and biologics
applications. U.S. Department of Health and Human Services, Food and Drug Administration,
Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and
Research (CBER), Rockville, CMC 6, 39 p
85. Vaal M, Van Leeuwen C, Hoekstra J (2000) Variation in sensitivity of aquatic species to toxicants: practical consequences for effect assessment of chemical substances. Environ Manage
25:415423
86. Verhaar H, van Leeuwen C, Bol J, Hermens J (1994) Application of QSARs in risk management of existing chemicals. SAR QSAR Environ Res 2:3958
87. VICH (2000) Environmental impact assessment (EIAS) for veterinary medicinal products
(VMPS)phase I. VICH GL6 (ecotoxicity phase I). VICH International Cooperation on
Harmonisation of Technical Requirements for Registration of Veterinary Medicinal Products,
Brussels
88. VICH (2004) Environmental impact assessment for veterinary medicinal products: phase II
guidance. VICH-GL38 (ecotoxicity phase II). VICH International Cooperation on Harmonisation
of Technical Requirements for Registration of Veterinary Medicinal Products, Brussels
89. Villeneuve DL, Garcia-Reyero N (2011) Predictive ecotoxicology in the 21st century. Environ
Toxicol Chem 30:19
90. Walker C (1995) Biochemical biomarkers in ecotoxicologysome recent developments. Sci
Total Environ 171:189195
91. Weiss C (1958) The determination of cholinesterase in the brain tissue of three species of fresh
water fish and its inactivation in vivo. Ecology 39:194199
92. Wheeler J, Gimeno S, Crane M, Lopez-Juez E, Morritt D (2005) Vitellogenin: a review of analytical methods to detect (anti) estrogenic activity in fish. Toxicol Mech Methods 15:293306
93. Williams ES, Panko J, Paustenbach DJ (2009) The European Unions REACH regulation: a
review of its history and requirements. Crit Rev Toxicol 39:553575
94. Maxwell K, Johnson GN (2000) Chlorophyll fluorescencea practical guide. J Exp Bot
51:659668

Human Health Risk Assessment for

Pharmaceuticals in the Environment: Existing
Practice, Uncertainty, and Future Directions
E. Spencer Williams and Bryan W. Brooks

Abbreviations
5-FU
ADI
API
BCF
CBZ
COPC
CPA
E2
EDC
EE2
ERA
GAC
GREAT-ER
HHRA
HQ
LOAEL
LOEL
MEC
MOA
MOS

5-Fluorouracil
Acceptable daily intake
Active pharmaceutical ingredient
Bioconcentration factor
Carbemazepine
Contaminant of potential concern
Cyclophosphamide
Estradiol
Endocrine-disrupting compound
Ethinylestradiol
Ecological or environmental risk assessment
Granular activated carbon
Geography-referenced regional exposure assessment tool for
European rivers
Human health risk assessment
Hazard quotient
Lowest observed adverse effects level
Lowest observed effects level
Measured or monitored environmental concentration
Mode of action
Margin of safety

E.S. Williams (*) B.W. Brooks

Department of Environmental Science, Institute of Biomedical Studies,
Center for Reservoir and Aquatic Systems Research, Baylor University,
One Bear Place, #97266, Waco, TX 76798-7266, USA
e-mail: [email protected]; [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_8, Springer Science+Business Media, LLC 2012

167

168

NOAEL
NOEL
OTC
PEC
PhATE
PIE
PNEC
POD
RfD
TTC
UF
WWTP

E.S. Williams and B.W. Brooks

No observed adverse effects level

No observed effects level
Over the counter
Predicted environmental concentration
Pharmaceutical Assessment and Transport Evaluation
Pharmaceuticals in the environment
Predicted no-effect concentration
Point of departure
Reference dose
Threshold of toxicologic concern
Uncertainty factor
Wastewater treatment plant

Introduction
Globally, several thousand substances are produced for pharmaceutical and biomedical applications in humans. The production tonnage of these compounds is
astronomical, ranging to hundreds of tons annually. Based on data collected by the
National Center for Health Statistics, individuals who visited their physician
recorded an average of almost seven medications taken per person [1]. As expected,
this number increases dramatically in older persons to almost 20 medications per
person after age 65. As the global population ages, the use of pharmaceuticals to
alleviate age-related conditions can reasonably be expected to increase. Further, the
ongoing development of large markets such as China and India will further increase
the magnitude of pharmaceutical consumption.
Scientists have been aware of the presence of active pharmaceutical ingredients
(APIs) in the environment since the late 1970s [2]. Efforts to monitor the occurrence
of these APIs more comprehensively began in earnest in the early 1990s, focused
primarily on substances that appeared to modulate the activity of endocrine systems
in humans and aquatic receptors (i.e., endocrine-disrupting compounds or EDCs).
Copious effort has been devoted to understanding the potential risks to the environment associated with EDCs and other types of APIs, including analgesics, neuroactive substances, and cardiovascular drugs. As of the end of 2009, over 39,000 articles
were found in a search of the ScienceDirect database using the keywords pharmaceuticals, risk, and water [3]. The focus of most of these studies has centered
on risk to ecological receptors, and many environmental impacts of APIs have been
identified [46]. In general, it has been believed that the environmental concentrations of APIs are too low to constitute a risk to human health in developed countries,
and several studies have been conducted to assess this perspective. However, a
recent poll among expert stakeholders reported that 62% of those interviewed
believed that pharmaceuticals in the environment (PIE) represent a risk to human
health [7].

Human Health Risk Assessment for Pharmaceuticals in the Environment

169

Several investigations have been conducted to determine the concentrations of

limited sets of substances in environmental compartments. Most of the studies that
report measured concentrations of API in environmental compartments (PIE) focus
only on a few analytes, though a handful of studies pursued more robust data sets.
The most exhaustive of these studies, conducted by the USGS, analyzed water samples from 30 states for 48 prescription and nonprescription drugs, along with caffeine [8]. A comprehensive list of all detections of PIE would be difficult to compile,
as the literature has expanded exponentially on this topic in recent years. APIs have
been observed in a number of different environmental compartments and subcompartments, including surface, ground, and drinking water, as well as wastewater
treatment plant (WWTP) influent and effluent. APIs have been also detected in soil
and leachate from landfills.
Colborn et al. coined the term endocrine disruption in the early 1990s to
describe the effects of some chemical substances found in the Great Lakes ecosystem. The observations indicated that diverse substances could impact reproductive
development and health in ecological receptors [9]. It is perhaps not surprising that
the wide variety of pharmaceutical substances designed to modulate the activity of
human reproductive systems should also carry this potential and thus be described
as an environmental endocrine-disrupting compound (EDC). Chief among these
products are hormones designed to prevent pregnancy or to alleviate ongoing symptoms of menopause. Naturally, the presence of EDCs in the environment has been a
source of concern with regard to public health [10]. More research on this area is
definitely warranted, and one of the avenues for this research is by using risk assessment tools.
An increasing body of research is available on ecological impacts resulting from
exposure to pharmaceutical substances. The effects associated with EDCs are best
characterized and include reduced fertility and delayed embryonic development in
fish [11, 12]. Another well-documented example of unanticipated toxicity in an
ecological receptor is the deaths of large numbers of vultures in Central Asia linked
to the presence of diclofenac in cattle carcasses on which the vultures were feeding
[13]. For the most part however, APIs are present in the environment at concentrations that are orders of magnitude lower than that which would be expected to cause
acute toxicity, even for those APIs designated as ecological hazards [5]. There have
been notable exceptions, as eco-risks have been suggested for acetylsalicylic acid,
paracetamol, ibuprofen, amoxicillin, oxytetracycline, and mefenamic acid, among
others [6, 14, 15]. Currently, there are no monitoring or regulatory requirements
pertaining to APIs in surface and drinking water, though the US EPA is moving in
that direction [16].
Several studies have been conducted to assess the risk to human health arising
from APIs in the environment [1731]. These investigators have focused primarily
on exposures through ingestion of drinking water and fish. The available studies
have assessed risk from a limited set of PIE. This is to be expected with the huge
number of PIE and the relatively small amount of data on the concentrations of the
substances.

170

E.S. Williams and B.W. Brooks

As a possible hazard, pharmaceuticals is a terribly broad term to employ in a

risk assessment complex, analogous to using chemicals for an occupational risk
assessment. According to Drugs@FDA, there are 5,986 approved substances for
human use in the USA, representing a wide variety of different chemical structures
and properties, even among those that have the same molecular targets (http://www.
accessdata.fda.gov/scripts/cder/drugsatfda/). Some of these substances are also
used for veterinary applications [32]. It is also worth noting that substances not
approved by the FDA can enter the environment through disposal as a result of
manufacture or testing, but the concentrations would be expected to be minimal in
the developed world [33, 34]. There are many classes of pharmaceuticals, designed
for many different purposes and with tremendous variation in physicochemical
characteristics and structures.
The EPA does not currently regulate the concentrations of APIs in drinking
water. However, several APIs have been listed in the newest Contaminant
Candidate List (CCL3), including equilenin, equilin, estradiol (E2), ethinylestradiol (EE2), estrone, estriol, mestranol, norethindone, quinolin, and erythromycin
(http://www.epa.gov/ogwdw000/ccl/ccl3.html). Nine of the ten APIs added to
the list are hormones, believed to act as environmental EDCs. After further regulatory review, the EPA may determine that the presence of APIs in drinking
water, beginning with these candidates, will be regulated under the Safe Water
Act. The USGS has included 43 veterinary and human pharmaceuticals in their
list of emerging contaminants for national reconnaissance studies in water bodies
in the USA [35].

Detections of APIs in the Environment

Surveys of bodies of water in the USA, and sources of untreated drinking water,
were conducted by the USGS in 19992001 [8, 36]. Kolpin et al. sampled 139
streams across the USA, and analyzed the samples for 31 human and veterinary
antibiotics, 15 prescription drugs, 7 nonprescription drugs, and 18 steroids and
hormones [8]. This study targeted bodies of water that were likely to be contaminated [37]. Many of these were detected in the ng/L range, and at relatively low
frequency (130% of samples). Higher frequency of detection was observed for
caffeine and nicotine metabolites, which were labeled as nonprescription drugs in
this study. A follow-up study in 2001 by Focazio et al. analyzed pharmaceuticals
in untreated surface and groundwater that are used for drinking water [36].
Unsurprisingly, the detection frequency for most analytes was lower than was seen
in surface waters from the prior study. As with the prior study, the highest frequency of detection was observed for nonprescription drugs including caffeine
(and its metabolite 1,7-dimethylxanthine) and cotinine.
It is perhaps to be expected that the concentrations of various APIs will vary
across seasons. The most obvious example of this phenomenon is the expectation

Human Health Risk Assessment for Pharmaceuticals in the Environment

171

that antibiotics will be more likely to be released into the environment during
cold and flu seasons in the spring and fall. However, these fluctuations may be
difficult to track, as sampling between wet and dry seasons in some parts of the
USA demonstrates [38].
Beyond over the counter (OTC) and prescription pharmaceuticals, illegal substances may be present in water that contributes to drinking water supplies [39]. In
particular, methamphetamine was widely detected in a study of Nebraska wasteand surface waters [40]. Published concentrations of cocaine, heroin, morphine,
amphetamine, methamphetamine, and LSD are summarized by Petrovic et al. [41]
and Zuccato and Castiglioni [39]. The usage rates of these materials are not well
understood, and thus there has been some interest in using effluent concentrations
of illicit substances to characterize consumption and perhaps also to track consumers [42]. For example, Kasprzyk-Hordern et al. calculated the consumption of
cocaine and amphetamine in South Wales by analyzing the concentrations of these
substances in raw wastewater [43, 44].
A number of other API occurrence studies have been conducted around the
world [174]. This research has focused on European countries such as Italy [42, 46],
Germany [47, 48], France [49, 50], Switzerland, Greece, Sweden, Denmark,
Finland, and the UK [15, 5153]. Beyond Europe, studies are available for Australia
[5456], India [57, 58], Brazil [59], Korea [60], Japan [6163], China [64, 65],
Vietnam [66], and Taiwan [67].

Routes of Pharmaceutical Introduction into the Environment

The concentration of PIE may or may not be related to the total mass manufactured
or the mass prescribed or purchased by consumers. This information may be
obtained via several routes, including governmental agencies, manufacturers, and
consulting groups like IMS Health [17, 21, 2325, 30, 34]. Pharmaceutical substances are then marketed and sold through a number of avenues, primarily through
over-the-counter (OTC) sales and prescriptions. Prescription drugs obviously are
sold directly to pharmacies, who then sell them to consumers who have been prescribed these drugs by a physician. According to the National Association of Chain
Drug Stores, 3.4 billion prescriptions were filled in the USA in 2006, amounting to
$716 billion in sales (http://www.nacds.org/).
A conceptual model of the dominant route of APIs through their lifecycle is
detailed in Fig. 1. There are several possible routes of APIs into the environment,
including excretion from consumers (urinary, fecal, or dermal), disposal of
medications, and manufacturing waste streams. An interesting analysis of the route
of ibuprofen and metaprolol through usage and disposal in the environment was
presented by Bound and Voulvoulis [68].

172

E.S. Williams and B.W. Brooks

Fig. 1 The major lifecycle

pathways of human active
pharmaceutical ingredients
from manufacture to potential
points of human exposure

Pharmaceutical
Manufacturing

Over the
counter

Pharmacies
Prescription

Consumers
Unused Medicines Excretion

Trash

WWTP
Land Application

Surface
water

Fish

Landfill
Biosolids

Leaching

Groundwater

Drinking water

Excretion
It is generally believed that the primary route through which these substances enter
the environment is through excretion of native substance, metabolites, or conjugates
[4, 33, 69]. This perspective is supported by the observation that temporal variations
in the API mass emitted are accompanied by similar variations in nitrogen [69].
A major source of APIs in the environment is urine, to the extent that separate
wastewater collection for urine has been recommended by at least one author [70].
Once excreted, these substances and their metabolites enter waste streams that pass
through WWTPs and then into other aquatic compartments. The importance of
environmental contamination by disposal of unused medications or as a result of
manufacturing processes is unclear at this point, but these routes should not be
ignored as examined closely in Chapter 10 of this volume. Also, medications enter
the terrestrial environment and groundwater as a result of disposal of solid and
semisolid wastes from WWTPs.
Though excretion is believed to be the primary route of APIs to the environment,
understanding the nature of this route on a substance-by-substance basis is very
difficult. The pharmacological reality of these substances is that a range of metabolites and conjugates will be generated, and these by-products and the parent
compound will be excreted in varying magnitudes through urine and feces [42, 43,
68, 71]. Jjemba classified a large number of APIs based on the extent to which they
were excreted as parent compound [72]. The author noted a variety of compounds that
are excreted as greater than 70% parent, including amoxicillin, atenolol, cimetidine,

Human Health Risk Assessment for Pharmaceuticals in the Environment

173

ciprofloxacin, codeine, furosemide, and valsartan [73]. Metoprolol, for example, is

excreted primarily in urine, 77% as parent compound [74, 75]. However, even
among the b-blockers, there is wide variation in the route of excretion (urine vs.
feces) and whether parent or metabolites or conjugates are excreted [76]. Similar
variation is noted among cancer chemotherapeutic drugs 5-fluorouracil (5-FU),
cyclophosphamide (CPA), and doxorubicin [24]. The percentage of API excreted as
the parent compound correlates strongly with its occurrence in the environment
[77]. Further, APIs that are excreted as polar metabolites (such as glucuronides) can
be cleaved in sewage treatment or in ecological compartments to the parent compound [78]. This variability is further compounded by differences in metabolism
and excretion due to individual-specific factors such as gender, age, nutrition, endocrine function, and preexisting disease [79, 80].
Beyond urine and feces, Daughton suggest that excretion through skin may be a
significant route for some medications [81]. The authors posit that these substances,
especially those applied topically, may be washed from the skin and to wastewater.
There is also the possibility of excretion of APIs in sweat [81]. Absorbent patches
have been used to test for illicit drugs. Sweat has already been characterized as a
mechanism to monitor for the use of illicit substances including cocaine, marijuana,
amphetamines, and heroin (for a review, see ref. [82]). Daughton and Ruhoy suggest
that for some APIs that are extensively metabolized, sweat may contribute three
times as much to the total excreted load compared to urine [83].
Improper Disposal
It has been theorized that APIs which are primarily disposed of directly to wastewater (e.g., via sinks or toilets) could contribute a disproportionate amount of the parent compound to the environmental load, as these APIs would bypass the metabolic
processes in the body [68, 84]. In fact, the loadings of some parent compounds by
this practice relative to excretion could be very important if a therapeutic undergoes
near complete metabolism and thus is excreted as metabolites. Some healthcare
professionals endorse this method of disposal, but increasingly it is recognized that
these disposals can constitute a significant contribution to the concentrations of API
in aquatic environments [34, 80, 85]. Poison control centers often counsel callers to
dispose of medications in sinks or toilets, to prevent children from coming into
contact with the APIs. Bound and Voulvoulis reported the results of a survey of API
consumers in the UK, which suggested that overall disposals to wastewater (i.e.,
using a sink or toilet) were relatively minor (016.7% of respondents, based on different drug types) and a significant portion (21.8%) were returned to the pharmacy
[68]. Ibuprofen, acetaminophen, and diclofenac are among the unused drugs most
frequently returned to pharmacies. However, a survey of Americans suggested that
slightly more than 1% of medications were returned to the pharmacy, while 54%
were disposed into trash, and 35% into sinks or toilets [86]. The low level of return
is not surprising as in most US states, pharmacies are not allowed to accept returned
medication from patients [68]. Beyond that, it has also been observed that medications

174

E.S. Williams and B.W. Brooks

returned to the pharmacy may be disposed of in similar ways used by consumers.

In the case of contraceptive patches containing EE2, special disposal advisories
were added to the packaging and safety leaflet to urge consumers to either return
unused patches to the pharmacy and to seal used patches back in the original packaging before disposing in solid waste [87].
Canada has established the Medications Return Program (MRP), and similar
programs have been enacted in Italy, France, and Australia [80]. Another role of this
program is to identify the reasons why medications go unused and ultimately to
tailor the prescriptions to both cut down on overprescribing and excessive disposal.
In 2007, the White House Office of National Drug Control Policy (ONDCP) issued
federal guidance on drug disposal by consumers, and the US Fish and Wildlife
Service and the American Pharmacists Association initiated a SMARxT Disposal
program, each of which recommended disposing of unused medications in the trash,
after removing any labeling and mixing the substances with unpalatable materials
like kitty litter (presumably to deter those who would accidentally or intentionally
come into contact with the APIs) [85]. However, the ODNCP guidance still recommends sink or toilet disposal for 13 APIs that are either highly toxic or likely to be
abused. For illegal drugs, direct disposal to sinks or toilets may be a more significant
pathway for parent compounds or by-products in the synthesis [33]. Specific information on pharmaceutical Take Back Programs is found in Chap. 10 of this book.
Manufacturing Waste Streams
There are hundreds of companies that manufacture pharmaceutical substances. In
the USA, the largest and best known of these include AstraZeneca, Bristol-Myers
Squibb, GlaxoSmithKline, Eli Lilly, Merck, Pfizer, Procter & Gamble, Roche,
Schering-Plough, and Wyeth. Each of these companies manufactures an array of
proprietary substances intended to treat a wide variety of conditions through prescriptions or over the counter. Formulations of these substances of course vary as
widely as their uses and the associated process can be expected to result in some
type of waste stream. It is also to be expected that these companies have multiple
manufacturing facilities, spread across multiple continents, consistent with their
need to provide pharmaceutical substances to a global community.
Until recently, it was generally understood that the mass of pharmaceutical substances discharged to surface waters streams is relatively in the ng/L to low mg/L
range in the developed world [32, 34, 88]. However, observations in developing
countries are less available. For example, a study conducted by Larsson et al. indicated that wastewater streams emanating from a WWTP serving 90 drug manufacturing facilities in India contained 21 pharmaceutical substances over 1 mg/L, some
reaching as high as 31 mg/L [58, 89]. Further examination of the surface and
groundwater connected to this WWTP revealed concentrations of several APIs,
including cetirizine, ciprofloxacin, metoprolol, and trimethoprim [57]. The authors
also demonstrated that many APIs were detectable in drinking water wells in the area
in the ng to mg/L range. In China, Li et al. reported concentrations of oxytetracycline

Human Health Risk Assessment for Pharmaceuticals in the Environment

175

and related compounds at levels in excess of 1 mg/L in treated wastewater arising

from pharmaceutical production facility and that these substances may be causing
the proliferation of resistant microbial strains [90]. From this information, it can be
inferred that API manufacturing facilities, when not managed properly, have the
potential to contribute a significant proportion of total load to aquatic environments.
In the USA, a recent study by Philips et al. [171] identified elevated levels of several
APIs in surface waters receiving discharges from manufacturing facilities.
Waste streams from hospitals, which are most likely an amalgamation of patient
excreta and disposed APIs, may also be a significant source of APIs in the environment. In a study in Sweden, as much as 12% of the total load of acetaminophen
entering WWTPs was contributed by hospital effluent, while many other APIs fell
in the 24% range [91]. A study in Australia suggested that hospital effluent may
contribute as much as 25% of the total load of roxithromycin, 10% of trimethoprim,
and approximately 5% of furosemide, ibuprofen, acetaminophen, ranitidine, and
salicylic acid [56]. Hospitals may also be expected to be associated with significant
output of anticancer medications, though these courses of treatment now frequently
occur on an outpatient basis [73].

Seasonal Variability
The use patterns of individual pharmaceutical substances vary based on their intended
target. This naturally leads to a seasonal variability of consumer usage of APIs, especially antibiotics and anti-inflammatory drugs, which would be expected to be used
much more during winter months [78, 92]. Castiglioni et al. demonstrated that the
wastewater loads of several APIs including ibuprofen, ciprofloxacin, ofloxacin, and
sulfamethoxazole were lower during the summer [92]. Some pharmaceuticals are
intended to moderate symptoms and thus are taken for extended periods [33]. Thus,
it is not surprising that the loads of b-blockers, diuretics, and antiulcer medications,
or the nonpharmaceutical substance caffeine, do not vary seasonally [92, 93]. In the
absence of seasonal variations in usage, the concentrations at various times of the
year could also be significantly affected by the flow conditions; concentrations of
APIs will be higher in summer months when wastewater effluent represents a larger
fraction of total flow [38, 50]. Degradation is also affected by seasonal factors such
as irradiance, temperatures, and microbial activity, both in the natural environment
and in WWTP systems [63, 9496]. Of course, seasonal variations will therefore not
be observed in all studies for all APIs [77, 97, 98].

APIs in Biosolids
Biosolids from WWTPs are frequently used to fertilize croplands. This use may allow
both runoff of the APIs enriched in the sludge into surface waters or uptake into edible
foodstuffs [99101]. These biosolids can contain relatively high concentrations of

176

E.S. Williams and B.W. Brooks

APIs [83, 102105]. Due to the analytical challenges associated with quantification of
APIs in biosolids, the significance of such materials with regard to hazard, exposure,
and risk is poorly understood and thus remains a significant research need [34].

APIs in MSW/Landfills
As mentioned above, disposal of APIs leads the substances not only to aquatic environments; they also are disposed into landfills [68]. Aside from disposed medications, sewage sludge (which contains APIs) from WWTP may also be disposed of
at a landfill [33]. Several studies have indicated that APIs that enter landfills can
leach into surrounding groundwater [53, 106, 107]. In particular, Holm et al.
observed in samples of landfill leachate several substances originating from waste
from the pharmaceutical industry [53]. Clofibric acid, ibuprofen, and prophenazone
were identified in leachate from a domestic landfill in Germany [108]. An understanding of APIs in leachate from landfills in the developing world is not known.

Veterinary Pharmaceuticals
Occasionally, APIs are used both for human and veterinary applications. These primarily include anti-infectives and hormones [32]. The routes for veterinary APIs (vAPIs)
into the environment can include emissions from manufacturing and from disposal, as
with human APIs. However, excreted urine and feces from livestock animals which
contain vAPIs are directly discharged to land and thus have the potential to contaminate soil or surface waters (through runoff) [32]. It appears that hazard information for
vAPI may be more readily available in some cases than for human APIs [109].
The presence of vAPIs in the environment has been observed [8]. Hamscher et al.
noted the presence of tetracycline and chlortetracycline, APIs used in veterinary
applications, in animal manure and in soil fertilized with manure [110]. Ivermectin,
a substance commonly used to deworm livestock animals, has been found to persist
in soil [87]. It has also been reported that tylosin, a veterinary antibiotic, has been
detected in drinking water [46]. An important activity that appears to introduce vAPIs
into the environment is that of aquaculture, the practice of raising aquatic animals.
Often, these substances are given in food pellets. The majority of these vAPIs ultimately leave the aquaculture area and enter the surrounding aquatic environments.
This pathway into the environment is well reviewed by Boxall et al. [111].

Fate and Behavior of APIs in the Environment

The route of APIs mostly passes through excretion by consumers, and thus it is to
be expected that a large proportion of these APIs will move through a WWTP

Human Health Risk Assessment for Pharmaceuticals in the Environment

177

before entering other aquatic compartments. WWTPs are variably effective in

removing APIs from wastewater; this may be a function of the technology type.
Once the parent APIs, metabolites, or conjugates enter the environment, they are
subject to normal processes of transport and degradation. For example, acetylsalicylic acid is degraded to salicylic acid following deacetylation and can also be
converted to ortho-hydroxyhippuric acid and gentisic acid [78]. These metabolites
have all been detected in wastewater influent [47]. However, it is worth noting that
salicylic acid can also arise from nonpharmaceutical uses [78]. Degradation processes can best be divided into biotic (i.e., biotransformation) and abiotic (e.g.,
photodegradation, hydrolysis). Manufacturers of APIs have in some instances
taken steps to retard biotic degradation processes so that their product will last
longer [112].
Loffler et al. studied the fate and behavior of ten common APIs in water and sediment they gathered from a waterway in Germany [113]. The bottom line of these
experiments was that, in the absence of the possibility of photodegradation, parent
compounds in many cases seemed to persist in the environment, while their metabolites (as expected) remained in the system for shorter periods. Carbemazepine
(CBZ) was found to be recalcitrant in the model system, with a 50% dissipation
time (DT50) of 328 days and moderate affinity for sediment. A metabolite, CBZdiol, was also suspected of persisting in aquatic environments. Clofibric acid was
also stable in the experiment (DT50 = 119 day), but had low sediment affinity.
Diazepam also persisted strongly in the environmental model, though its human
metabolite oxazepam degraded somewhat more quickly. Ivermectin sorbed strongly
into sediments and the potential for accumulation in that compartment appeared
high. Ibuprofen and its metabolite 2-hydroxibuprofen however were converted
almost completely to CO2 by the end of the experiment. Acetominophen was also
degraded relatively rapidly. Further experimentation suggested that the rapid degradation of these two APIs was due to biotic processes.
Photodegradation appears to be an important process for APIs in aquatic environments, both through direct and indirect pathways [112, 114, 115]. In some locations, wastewater treatment processes include a UV irradiation step. Under these
techniques, it appears that some types of APIs will be completely degraded, while
others may be somewhat resistant [116, 117]. The experiments of Lin and Reinhard
highlight the importance of other factors in degradation, as photolysis experiments
conducted in river water produced much faster degradation rates than experiments
conducted in purified water [115]. Under environmentally relevant conditions, the
natural photosensitizers may hasten the photodegradation of APIs, possibly as a
result of the formation of reactive oxygen species which react with the compound
[51]. This type of indirect mechanism is also important to the degradation of
cimetidine, clofibric acid, and ibuprofen [114, 118]. However, direct photochemical
degradation is also in play, for ranitidine, naproxen, CBZ, and diclofenac [114].
As noted by Brooks et al. [173], clearly this is an area requiring more attention.
Consideration of the environmental fate of APIs are more thoroughly examined in
Chapter 4 of this volume.

178

E.S. Williams and B.W. Brooks

Removal of APIs in WWTP and Drinking Water Treatment

WWTP treatment is somewhat effective at removing or at least lowering the concentrations of a broad spectrum of APIs, though undoubtedly a number of APIs pass
through standard treatment without significant reductions in concentration [103,
119, 120]. The rate of removal varies widely between APIs, though the relationship
between removal rate and structure or physicochemical properties is unclear [69].
Removal rate may also vary seasonally [92]. The technology and techniques used
also dictate the removal efficiency during treatment, to a significant extent on a
substance-by-substance basis [92, 102, 121]. There appears to be a level of variability between experiments, as well, as removal rates from 0 to 69% have been published for diclofenac [102].
Water treatment processes appear to remove hydrophobic substances most
efficiently [122]. This is not surprising, as hydrophobic APIs would be expected to
bind to organic material that would predominately be removed by flocculation and
sedimentation processes. Biosolids in the WWTP system can be a significant reservoir of APIs, however, which is relevant both for resuspension/dissolution of the
substance and for soil contamination should the biosolids be deposited on land or in
a landfill later [103]. Regardless, the processes of clarification, disinfection, and
filtration through granular activated carbon (GAC) can reduce concentrations of
APIs in source water by 75% or to a level below detection limits [122]. A study of
the behavior of 13 APIs in a drinking water plant suggested that while many were
eliminated by a combination of ferric salt coagulation, sand filtration, ozonation,
GAC filtration, and UV disinfection, ciprofloxacin was able to pass through into
finished drinking water [123]. Ozonation, a method first used for the removal of
coliform bacteria and enteric viruses, appears to be a very highly effective process
for the removal of APIs during processing [124126].
It is worth noting that while APIs are consumed in every corner of the globe,
wastewater treatment is not uniform in all countries. It would be interesting to study
whether the lack of sophisticated wastewater treatment or a lower rate of API consumption plays a larger role in the concentrations of API in the environments of
developing nations. Treatment technologies for APIs in drinking source waters and
wastewater influents are examined in Chap. 9 of this book.

Human Exposure
A summary of potential major pathways for human exposure to environmental APIs
is presented in Fig. 2. Some of the pathways described are not expected to be complete. For instance, it is unlikely that any APIs will volatilize sufficiently to produce
an inhalation dose, though there are some indications of potential inhalation exposure to antibiotics and thus potentially other APIs sorbed to particulate matter in
some circumstances [110]. Also, dermal exposures though possible are not likely to

Human Health Risk Assessment for Pharmaceuticals in the Environment

179
Ingestion
Drinking water

Disposal

Drinking water

Landfill

Surface water
Soil

On-site Septic

Fish/shellfish

Groundwater

Foodstuffs

Manufacturing
Surface water

CSO

Dermal
Drinking water

Soil

WWTP

Surface water
Soil

Excretion
Sludge

Sediment

Sediment

Inhalation
Veterinary

Manure

Foodstuffs

Drinking water
Surface water
Soil

Fig. 2 Potential major exposure pathways for human active pharmaceutical ingredients to human
receptors

reach a level of concern. Intuitively, ingestion of APIs in drinking water is expected

to be the most relevant exposure pathway. Thus, sampling and analysis of potential
sources of drinking water, as well as surface waters, seem to be the most informative. It is worth noting that sampling of these sources may not generate a reasonable
measure of exposure, as sampling tends to focus on areas where contamination is
suspected [22]. Indeed, risk assessments performed to date have focused primarily
on this pathway, though exposure concentrations used in these exercises are more
likely to rely on surface water data as a more conservative value. It seems likely that
the magnitude of doses would be greatest in this pathway, but the importance of
other potentially complete pathways must be a focus of future research efforts.
Several studies have also assessed the potential for exposure through ingestion of
fish caught in contaminated waters. Measured environmental and predicted environmental concentrations (MECs, PECs) used in all of the human health risk assessment (HHRA) exercises published to date are summarized in Table 1.

Measured Environmental Concentrations

Analytical measurements of APIs in drinking water are relatively scarce [73]. The
available data do suggest that concentrations are lower in finished drinking water
than in prior stages, suggesting that various processing technologies have varying
effectiveness, but the vast majority of APIs are not detected in drinking water [122].
Webb et al. using data from monitoring studies in Germany estimated the exposure
from 64 APIs in drinking water [31, 127, 128]. However, these monitoring studies

180

E.S. Williams and B.W. Brooks

Table 1 Measured and predicted environmental concentrations of active pharmaceutical ingredients in risk assessment exercises published to date
Pharmaceutical substance
Concentration type
Concentrations
Unit
References
17a-Ethinylestradiol

PECDW-local
PECDW-regional
MECmax
Acetominophen
MECSW-max
PECDW-max
PECSW-max
Acetylsalicylate
MECDW-max
MECSW-max
MECSW-max
MECmax
Albuterol
MECSW-max
PECDW-max
PECSW-max
MECmax
Alkylating chemotherapeutic PECmax-mean flow
agentsa
PECmax-high flow
Anthracycline antibioticsb
PECmax-mean flow
PECmax-high flow
PECmax-mean flow
Antimetabolite
chemotherapeuticsc
PECmax-high flow
Atenolol
MECDW-max
MECmax
Atomoxetine
PECusage
PhATE 99th PEC
Atorvastatin
MECDW-max
Benzylpenicillin
MECmax
Betaxolol
MECmax
Bezafibrate
MECmax
Bisoprolol
MECmax
Carazolol
MECmax
Carbamazepine
MECmax
MECmax
MEC90th-NA
PEC90th-NA
MECDW-average
MECDW-max
MECDW-max
MECSW-max
MECDW-max
Celiprolol
MECmax
Chloramphenicol
MECmax
Chlorotetracycline
MECmax
Cimetidine
MECSW-max
PECDW-max
PECSW-max

1.2
0.3
<0.5
10,000
220,000
470,000
0.12
0.065
340
<10
35
120
250
<5
10.72
19.99
0.05
0.09
2.02
3.76
0.02
<5
0.02
0.12
<0.00025
<50
<5
27
<5
<5
30
<50
150
333
2.8
5.7
0.03
0.227
0.018
<5
<20
<20
580
4,400
9,300

ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
mg/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L

[26]
[26]
[32]
[24]
[24]
[24]
[29]
[29]
[21]
[32]
[24]
[24]
[24]
[32]
[28]
[28]
[28]
[28]
[28]
[28]
[30]
[32]
[18]
[18]
[30]
[32]
[32]
[32]
[32]
[32]
[32]
[32]
[23]
[23]
[27]
[27]
[29]
[29]
[30]
[32]
[32]
[32]
[24]
[24]
[24]
(continued)

Human Health Risk Assessment for Pharmaceuticals in the Environment

Table 1 (continued)
Pharmaceutical substance
Ciprofloxacin

Clarithromycin
Clenbuterol
Clofibrate
Clofibric acidd

Cloxacillin
Codeine

Cyclophosphamide

Dehydrato-erythromycin
Dehydronifedipined

Diazepam
Diclofenac
Dicloxacillin
Digoxin

Digoxigenind

Diltiazem

Doxycycline

Duloxetine
Enalapril

181

Concentration type

Concentrations

Unit

References

MECSW-max
PECDW-max
PECSW-max
MECmax
MECmax
MECDW-max
MECmax
MECDW-max
MECSW-max
MECmax
MECSW-max
MECmax
MECSW-max
PECDW-max
PECSW-max
PECDW
PEClocal-average
MECmax
MECSW-max
MECmax
MECmax
MECmax
MECSW-max
PECDW-max
PECSW-max
MECDW-max
MECmax
MECDW-max
MECmax
MECmax
MECSW-max
PECDW-max
PECSW-max
MECSW-max
PECDW-max
PECSW-max
MECSW-max
PECDW-max
PECSW-max
MECSW-max
PECDW-max
PECSW-max
MECmax
PECusage
PhATE 99th PEC
MECDW-max

30
3,600
7,600
<20
<10
270
<20
0.14
0.091
70
1,750
<50
1,000
1,100
2,400
2.58
5.6
4
10.1
<10
<50
<20
30
1,100
2,300
<0.00025
<20
<0.00025
6
<50
130
6.3
13
4
6.3
13
49
5,900
12,000
50
990
2,100
<20
0.05
0.13
<0.00025

ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
mg/L

[24]
[24]
[24]
[32]
[32]
[21]
[32]
[29]
[29]
[32]
[21]
[32]
[24]
[24]
[24]
[26]
[19]
[19]
[21]
[32]
[32]
[32]
[24]
[24]
[24]
[30]
[32]
[30]
[32]
[32]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[32]
[18]
[18]
[30]
(continued)

182

E.S. Williams and B.W. Brooks

Table 1 (continued)
Pharmaceutical substance
Enalaprilat

Erythromycin-H2O

Etofibrate
Fenofibrate
Fenofibric acidd
Fenoprofen
Fenoterol
Fluoxetine

Gemfibrozil

Ibuprofen

Ifosfamide

Indometacine
Ketoprofen
Lincomycin

Meprobamate

Metaprolol
Metformin

Methicillin
Metropolol
Nadolol
Nafcillin

Concentration type

Concentrations

Unit

References

MECSW-max
PECDW-max
PECSW-max
MECSW-max
PECDW-max
PECSW-max
MECmax
MECmax
MECmax
MECmax
MECmax
MECSW-max
PECDW-max
PECSW-max
MECDW-max
MECSW-max
PECDW-max
PECSW-max
MECDW-max
MECmax
MECSW-max
PECDW-max
PECSW-max
MECmax
MECmax
MECmax
PEClocal-average
MECmax
MECmax
MECSW-max
MECmax
MECSW-max
PECDW-max
PECSW-max
MECDW-average
MECDW-max
MECDW-max
MECDW-max
MECSW-max
MECSW-max
PECDW-max
PECSW-max
MECmax
MECmax
MECmax
MECmax

46
30
63
1,700
3,500
7,300
<20
<20
42
<5
<5
46
620
1,300
<0.00050
1,550
8,000
17,000
0.0021
<5
2,700
63,000
130,000
3
<10
<50
10.9
206
<5
200
<5
730
9.8
21
6.1
13
0.043
2.1
0.2
150
47,000
98,000
<50
<5
<5
<50

ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L

[24]
[24]
[24]
[24]
[24]
[24]
[32]
[32]
[32]
[32]
[32]
[24]
[24]
[24]
[30]
[24]
[24]
[24]
[30]
[32]
[24]
[24]
[24]
[32]
[32]
[32]
[19]
[19]
[32]
[21]
[32]
[24]
[24]
[24]
[27]
[27]
[30]
[29]
[29]
[24]
[24]
[24]
[32]
[32]
[32]
[32]
(continued)

Human Health Risk Assessment for Pharmaceuticals in the Environment

Table 1 (continued)
Pharmaceutical substance
Naproxen
Norfloxacin

Olanzapine
Oxacillin
Oxytetracycline

Pentoxifylline
Phenazon

Phenoxymethylpenicillin
Phenytoin

Propranolol
Ranitidine

Risperidone
Roxithromycin
Salicylic acid
Simvastatin
Sotalol
Sulfamethazine
Sulfamethoxazole

Sulfathiozole

Terbutalin
Tetracycline

183

Concentration type

Concentrations

Unit

References

MECDW-max
MECSW-max
PECDW-max
PECSW-max
PECusage
PhATE 99th PEC
MECmax
MECSW-max
PECDW-max
PECSW-max
MECmax
MECmax
MECmax
MECmax
MECDW-max
MECSW-max
MECmax
MECmax
MECDW-median
MECDW-max
MECDW-max
MECmax
MECSW-max
PECDW-max
PECSW-max
MECDW-max
MECmax
MECmax
MECDW-max
MECmax
MECmax
MECDW-max
MECSW-max
MECSW-max
PECDW-max
PECSW-max
MECDW-max
MECmax
MECSW-max
PECDW-max
PECSW-max
MECmax
MECSW-max
PECDW-max
PECSW-max
MECmax

<0.00050
120
74
160
0.01
0.07
<50
1,340
0.92
1.94
<20
<10
<20
<50
0.03
0.11
50
<50
6.2
19
0.015
<5
39
7,800
16,000
0.00034
<20
<10
<0.00025
<5
<20
0.03
0.11
1,900
8,500
18,000
0.003
<20
80
13
28
<10
1,000
3,100
6,500
<20

mg/L
ng/L
ng/L
ng/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
mg/L
ng/L
ng/L
mg/L
mg/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L

[30]
[24]
[24]
[24]
[18]
[18]
[32]
[24]
[24]
[24]
[32]
[32]
[32]
[32]
[29]
[29]
[32]
[32]
[27]
[27]
[30]
[32]
[24]
[24]
[24]
[30]
[32]
[32]
[30]
[32]
[32]
[29]
[29]
[24]
[24]
[24]
[30]
[32]
[24]
[24]
[24]
[32]
[24]
[24]
[24]
[32]
(continued)

184
Table 1 (continued)
Pharmaceutical substance
Timolol
Triclosan
Trimethoprim

Warfarin

E.S. Williams and B.W. Brooks

Concentration type

Concentrations

Unit

References

MECmax
MECDW-max
MECSW-max
PECDW-max
PECSW-max
MECDW-max
MECmax
MECSW-max
PECDW-max
PECSW-max

<5
0.0012
710
1,800
3,700
<0.00025
<20
0.5
120
250

ng/L
mg/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L

[32]
[30]
[24]
[24]
[24]
[30]
[32]
[24]
[24]
[24]

Oxaliplatin, temozolomide, cisplatin, carboplatin, cyclophosphamide

Epirubicin, doxorubicin
c
Gemcitabine, fludarabine, capecitabine
d
Metabolite
a

showed that only 10 of the 64 were detected in surface water at levels above the
limit of quantitation and even within those that were detected the proportion of
samples above the LOQ was 53% or below [31]. A summary of published concentrations for APIs in drinking water was reported by Jones et al. and indicated that of
those detected, all fell in the ng/L range [129]. The detected compounds included
bezafibrate, clofibric acid, CBZ, diazepam, diclofenac, and ibuprofen. A study of
APIs and EDCs in source, finished, and delivered drinking water showed the presence of atenolol, CBZ, gemfibrozil, meprobamate, phenytoin, and several other
APIs in drinking water [29, 130]. In Germany, several studies have indicated the
presence of APIs in drinking water, including acetaminophen, acetylsalicylic acid,
diclofenac, ibuprofen, CBZ, and naproxen [131]. Despite these data, it is clear that
no systematic analysis of APIs in drinking water has been conducted on a national
or global scale. The broad array of potential analytes is simply too broad. Thus, it is
difficult to assess direct exposure through this pathway.
Cunningham et al. noted several published values for CBZ concentrations in
drinking water, but ultimately opted to use surface water measurements that were
significantly more numerous and regarded as a more conservative exposure value [22].
Schriks et al. used maximum concentration data from surface and groundwater, if
primary data from the Rhine and Meuse rivers were not available [28]. For more
information on the occurrence of APIs in the environment, an excellent review [174].

Predicted Exposure Concentrations

Many of the risk assessments performed to date include the use of modeling
approaches for point-of-exposure concentrations. Environmental concentrations can
be crudely estimated using a very simple equation, as described by Coetsier et al. [49].
This equation integrates total consumption, the excretion rate of the parent API, and

Human Health Risk Assessment for Pharmaceuticals in the Environment

185

the fraction that passes through WWTP, and divides this figure by the volume of
wastewater generated per person, the number of persons, and the extent of dilution.
This general approach appears to provide reasonable PECs for WWTP effluent, but
the PECs for surface water are not in agreement with MECs [49] particularly in
effluent-dominated surface waters [172].
More sophisticated computer models have been established to generate PECs for
the USA and the EU. EUSES, a tool used in risk assessment for chemical substances,
has been applied to provide a worst-case scenario of concentrations of APIs in the
environment by Christensen [25]. However, two primary systems have been used to
model the environmental concentrations of APIs in aquatic systems: the Pharmaceutical
Assessment and Transport Evaluation (PhATE) model, and the Geography-Referenced
Regional Exposure Assessment Tool for European Rivers (GREAT-ER). These models appear useful in addressing the possibility of data bias (resulting from sampling of
areas suspected to be contaminated), filling the many gaps in monitoring data, and
developing testable hypotheses for site-specific field studies.
PhATE was developed by the Pharmaceutical Research and Manufacturers of
America (PhRMA) [132]. This model is designed to offer a PEC of pharmaceuticals
discharged into surface waters through POTWs. PhATE was designed around 11
watersheds in the USA, covering approximately 19% of the nations land area.
Watersheds were selected in which drinking water supplies are derived from sources
that can be impacted by WWTP discharges; hence, most metropolitan areas (Los
Angeles, New York, Chicago, Miami, Denver) are not included in the model. PhATE
is essentially a mass balance model which begins with the API use per capita, estimates
the metabolism of the API, and then models the mass of API that enters the surface
water compartment (as well as providing estimates of the amount of API lost from
surface water through degradation and other processes). A preliminary validation
exercise for this model was conducted using caffeine, triclosan, and linear alkylbenzene sulfonates (LAS), and the results suggested a reasonable relationship between
PhATE-modeled PECs and measured environmental concentrations (MECs), i.e.,
within an order of magnitude [132]. However, PECs generated by the PhATE model
deviated by multiple orders of magnitude when compared with MECs generated by
Kolpin et al. [8]. These deviations were attributed to numerous factors, including variability in API removal in WWTP and possible difficulties in understanding accurate
environmental concentrations due to analytical methodology. iSTREEM represents a
similar model to PhATE that was originally developed for cleaning products, but has
been applied to APIs (http://www.aciscience.org/iSTREEM.aspx)
In the EU, a similar model was developed in the late 1990s by a collaborative
group under the auspices of the European Centre for Ecotoxicology and Toxicology
of Chemicals (ECETOC) [133]. GREAT-ER was not necessarily designed to
establish PECs for APIs, but rather was intended to be applied to a broader group
of substances. However, it has been applied in many such exercises [76]. Input
information for GREAT-ER can include excretion data as a source of input into
WWTPs [76].
Several of the HHRAs published to date use some combination of MECs and
PECs, either as validation or as measures of uncertainty. Before PhATE and

186

E.S. Williams and B.W. Brooks

GREAT-ER were available, Christensen used publicly available data on consumption

as a starting point for point-of-exposure concentrations generated by the EU
computer program EUSES; the consumption data were obtained from the Danish
Medicine Agency and pharmacy contacts [25]. Christensen used this program to
model concentrations in drinking water, edible species, foodstuffs, and air. Kummerer
and Al-Ahmad used a combination of the amount of CPA and ifosfamide used in
Germany and measured concentrations of these two substances in WWTP influent
and effluent to calculate regional and local PECs using a relatively simple equation
[18]. The PECs were then calculated for surface water and used as a proxy for drinking water concentrations.
Johnson et al. also used consumption data, provided by the Health and Safety
Executive (HSE) and Department of Health [24]. In the USA, Schwab et al. obtained
information on the quantity of 26 APIs sold by consulting manufacturers and databases [23]. In contrast, Cunningham et al. used an estimate of the amount sold in the
USA and six European countries coupled with the available data on environmental
concentrations of their APIs of interest [21].
In the absence of monitored data, Cunningham et al. used the PhATE and
GREAT-ER models to estimate PECs for 44 APIs marketed by GlaxoSmithKline
[21]. The authors identified data for only nine of these APIs in the peer-reviewed literature, and these surface water data points were not informative for the purposes of
risk assessment for drinking water and fish ingestion exposures. In contrast, Schulman
et al. were able to find measured concentrations of their four APIs of interest (acetylsalicylic acid, clofibrate, CPA, and indomethacin) in multiple aquatic compartments
including sewage effluent, surface water, and drinking water [20]. Schwab et al. used
measured concentrations from Kolpin et al. and other peer-reviewed sources and supplemented their analysis of exposure concentrations using the PhATE model [8, 23].
Bercu et al. performed two exercises to estimate an exposure concentration to the
three neuroactive compounds they studied; the first estimate was generated by dividing the total mass of the API that was sold in the USA by the annual volume of water
discharged from all POTWs in the USA [17]. The second estimate was calculated
using the PhATE model, again using the total mass of API sold as an input parameter.
Johnson et al. in estimating risks for exposure to one chemotherapeutic agent, 5-FU,
began by cataloging the consumption of the drug at all major cancer treatment centers
in England and calculating the concentration of 5-FU in wastewater effluent [24].
These parameters were then incorporated in GREAT-ER to provide a reasonable estimate of surface water concentrations. Rowney et al. used a derivative model of
GREAT-ER called LF2000-WQX and calculated per capita loading estimate that was
modified by approximate removal efficiency by sewage treatment plants [27].

Exposure Through Food

Exposure to contaminant of potential concerns (COPCs) in fish tissue has become
a target of recent research for a variety of chemicals and sites. Several studies

Human Health Risk Assessment for Pharmaceuticals in the Environment

187

indicate that APIs can accumulate in fish and other aquatic organisms [175] and
thus exposure to these APIs through consumption by recreational or subsistence
anglers must be considered. The examined compounds include fluoxetine, sertraline, ibuprofen, naptoxen, diclofenac, ketoprofen, gemfibrozil, diphenhydramine,
diltiazem, CBZ, and paraoxetine [134138]. Ramirez et al. conducted a reconnaissance study to determine levels of 24 APIs and metabolites in fish tissue in six
effluent-dominated streams [138]. Only five of these analytes (norfluoxetine, sertraline, diphenhydramine, diltiazem, and CBZ) were detected in fillets and seven in
liver (fluoxetine, gemfibrozil). Among the APIs not detected were acetaminophen,
ibuprofen, propranolol, and warfarin. Interestingly, the authors noted that there
was no clear association between lipid content and bioaccumulation of the detected
APIs, though the pKa of the detected substances appears to play a role in this
observation.
Several of the published HHRAs have considered this exposure pathway [21, 23,
26]. As they are critical to understanding the potential for dosages through ingestion
of fish tissue, bioconcentration factors (BCFs) have been estimated for several APIs
[21, 23]. It appears critical, however, to account for site-specific pH influences on
bioaccumulation of APIs in fish and other wildlife [176].
As with drinking water, the scarcity of data necessitates the use of conservative
and/or modeling approaches to this pathway and invites further scientific effort in
this area. The available data are also subject to the same limitation as that in surface
and drinking water; fish have been sampled predominantly in effluent-dominated
systems, where contamination would be expected to potentially represent worst-case
scenarios in the developed world [134].
There is also a potential dietary pathway through crop foods. As mentioned
above, APIs can enter the soil compartment through sewage sludge spreading or
through manure from livestock. Experiments conducted by Boxall et al. indicate that
vAPIs can arise in foodstuffs grown on lands which are contaminated with these
substances [139]. Their findings indicate that florfenicol, levamisole, enrofloxacin,
and trimethoprim were detected in lettuce or carrots grown on soil spiked with these
substances.

Hazard and DoseResponse Assessment

In the context of a contaminated site, the process of hazard assessment is to identify chemicals that have the potential to cause adverse health effects in humans.
This process is fairly straightforward when performed for industrial chemicals.
However, pharmaceutical substances are designed to generate specific health
effects at specific doses. Thus, it can be expected that all APIs designed for
humans will have some type of effect at certain doses. Identification of a complete
suite of APIs in an exposure scenario, as with COPCs at a contaminated site, may
be impossible.

188

E.S. Williams and B.W. Brooks

Selection of Substances for Risk Assessment

All of the HHRAs conducted to date began with a preselected set of APIs. One of
the earliest HHRA exercises began with three representatives of API classes which
were expected to cause effects at low exposure concentrations (i.e., E2, phenoxymethylpenicillin, CPA) [25]. Schwab et al. chose to study 26 APIs from 14 classes
that were examined in national reconnaissance performed by Kolpin et al. [8, 23].
Schulman et al. selected four pharmaceutical compounds that have been found
more commonly or at the higher end of the spectrum of measured concentrations
in aqueous compartments [20]. Their selected compounds included acetylsalicylic
acid, clofibrate, CPA, and indomethacin. The authors note CPA as the sole recognized carcinogen of the four, though there is limited animal evidence for carcinogenicity of clofibrate. Cunningham et al. chose to study CBZ and two metabolites,
as CBZ has been frequently detected and has been observed in multiple aquatic
compartments, including drinking water [22]. In Cunningham et al.s prior study,
they chose to study APIs manufactured by GlaxoSmithKline, as they worked for
that manufacturer. These 44 APIs included amoxicillin trihydrate, digoxin, cimetidine, bupropion HCl, albuterol, metformin, and malphalan [21]. Bercu et al. studied three neuroactive drugs (atomoxetine, duloxetine, and olanzapine) for roughly
the same reason [17]. Kummerer and Al-Ahmad focused on CPA and ifosfamide
(IF), as these compounds were both known to be carcinogenic and to persist to a
moderate degree in aquatic environments [18]. 5-FU was chosen by Johnson et al.
for similar reasons [24]. Webb et al. assessed risk from 64 APIs for which monitoring data in Germany had been previously published [31]. Kumar and Xagoraraki
assessed the potential risk from meprobamate, CBZ, and phenytoin, as they have
previously been detected in water in the USA, and may have toxic effects in women
and children [26]. In the USA, Snyder et al. developed a list of APIs to consider via
a process designed to identify analytes which represented broader classes of compounds, while considering toxicity potential and likelihood of occurrence in raw
and finished drinking water [29]. In a report to the Water Inspectorate of the United
Kingdom, Crane et al. analyzed the potential risks arising from 396 APIs and 11
illegal drugs, though the methodology for selection of the substances was not
explicitly stated [30].
vAPIs pose a slightly more complicated question with regard to human hazard.
As mentioned before, some vAPIs are used in both humans and animals. However,
for APIs designed only for veterinary applications, the human implications may be
less clear. It is to be expected that these types of materials will have effects on
humans (as they were developed for mammals), but understanding the doseresponse relationships will be difficult. To date, no HHRA has been conducted for
vAPIs; however, this represents an area of important research need.
It is also clear that some of the APIs that have been detected in the environment
have genotoxic and/or carcinogenic properties. The most prominent of these are
anticancer therapeutics such as CPA. APIs enter the environment at a low but continuous rate, though their levels may vary seasonally as described above. Thus, it is

Human Health Risk Assessment for Pharmaceuticals in the Environment

189

generally believed that acute effects are unlikely [140]. The possibility of long-term,
chronic exposures is much more likely, and for most if not all APIs there is no
toxicological data for this type of exposure duration.
The reality of the hazard assessment process is that it will not be effective in
the absence of complete toxicity information. Such data are generally lacking for
many chemicals, and for API this may be more complicated as the substances
were engineered to be therapeutic for human or veterinary purposes. Certainly,
many of these substances are expected to have side effects, which may be of interest in a toxicological investigation. When available, data on toxicological properties are used to determine whether adverse health outcomes can be expected from
any dose of API.

Setting of Safety Values

The goal of doseresponse assessment is to determine the quantitative relationship
between doses received and health outcomes. The end result of the process is the
development of one or more criteria values, which cover the gamut of likely or possible adverse health consequences, including carcinogenic and noncarcinogenic
effects. Of course, it is highly likely that the primary consideration in development
of criteria values is the toxicological mode of action (MOA) of the API. Many carcinogenic outcomes are expected to follow from a non threshold MOA, meaning
that risk exists even at infinitesimally minute doses. Non carcinogenic health effects
generally are not believed to follow a nonthreshold MOA.
The earliest example of HHRA used several different avenues to establish a
safety value, against which intake values could be compared for risk characterization [25]. For 17a-EE2, a comparison with endogenous 17b-estrodiol (E2) was
used qualitatively to compare to the intake value. With regard to phenoxymethylpenicillin, a value of 10 IU was used, the level below which no allergic responses
would be expected.
Schulman et al. established safety values for acetylsalicylic acid, clofibrate, CPA,
and indomethacin [20]. The point of departure (POD) for acetylsalicylic acid was a
lowest observed effects level (LOEL) for anticoagulant therapy (30 mg/day), and
safety factors included 3 to estimate a no observed effects level (NOEL) and 10 for
interindividual variability. For clofibrate, the authors used the low end of the dose
range (500 mg/day) for lowering of blood triglycerides as a LOEL and safety factors similar to those used for acetylsalicylic acid. In the case of indomethacin, a
subtherapeutic dose of 37.5 mg/day was chosen as the POD, and as previously a
safety factor of 30 was employed. Schulman et al. used these health-based limits to
generate an ambient water quality criteria value using EPA methodology and compared published concentrations in surface and drinking water to that value.
Therapeutic doses have been used as safety values in other risk assessment exercises
as well [30, 31]; in one case, exposures within three orders of magnitude signaled a
need for further study. This approach did not include the use of further uncertainty

190

E.S. Williams and B.W. Brooks

factors (UFs), though future efforts are clearly needed to develop and thus refine
default UFs applied in HHRAs and ERAs.
Several HHRAs have set safety values using the concept of acceptable daily
intake (ADI) (Table 2). Previously, the ADI has primarily been applied to food additives, pesticides, and veterinary drugs that are not genotoxic or carcinogenic [141,
142]. The ADI is very similar to the reference dose (RfD) in that it determines a
POD and divides that value by UFs to impart a margin of safety (MOS). The POD
for the ADI is often the no observed adverse effects level (NOAEL) or NOEL
derived from the study in which toxicity was seen at the lowest dose [141, 142]. In
some applications, the ADI is based on a NOEL as opposed to a NOAEL [31]. UFs
can account for extrapolation from a lowest observed adverse effects level (LOAEL)
to a NOAEL (when a NOAEL is not available), interindividual and interspecies differences, and for the possibility of an incomplete set of data. When the ADI has
been set, it can be regarded as a safe intake level (without an appreciable risk) for
a healthy adult who is exposed to an average daily amount of the substance in question over a lifetime [142]. ADIs have been determined for APIs in the environment
in several HHRA exercises [17, 21, 22, 26, 28, 31, 139].
A procedure for generating ADIs for APIs was well articulated by Schwab et al.
[23]. The authors chose the lowest therapeutic dose level as the POD for 26 APIs,
including acetaminophen, codeine, fluoxetine, and tetracycline. Each of these PODs
was then divided by up to five UFs, to account for extrapolation from a therapeutic
dose to a NOAEL (UF1), differences in exposure duration (UF2), differences in sensitivity among species (UF3), differences in susceptibility among individuals (UF4),
and quality of the data used to derive the POD (UF5). For example, the POD for
acetaminophen was 9.3 mg/kg/day, as the lowest effective daily therapeutic dose.
This POD was divided by 3 to estimate a NOAEL, 3 to account for chronic exposures
since the POD is based on an acute dose, and 3 to allow for differences in sensitivity
among human individuals; thus the authors arrived at an ADI of 340 mg/kg/day for
acetaminophen. These calculated ADIs were then used, along with estimates of
BCFs, to calculate a set of predicted no-effect concentration (PNEC) values for
ingestion of APIs through drinking water, fish tissue, and a combination of the two.
A similar approach was taken by Kumar and Xagoraraki [26]. Kummerer and
Al-Ahmad compared their exposure values to the lowest dose of CPA or ifosfamide
given in anticancer therapy [18].
Cunningham et al. calculated ADIs following the general process employed by
Schwab et al. [2123]. The authors used LOELs, NOELs, or the lowest daily therapeutic dose as their POD and applied UFs to account for duration of exposure,
interspecies variability, intraindividual susceptibility, and data quality. Ultimately,
PNECs were assigned for ingestion of APIs through drinking water and fish consumption, as well as a combined metric. PNECs for three neuropharmaceuticals
were also calculated using the ADI procedure set forth by Schwab [17]. An ADIbased value has also been offered for CPA, a chemotherapeutic agent which is
known to be genotoxic [25]. Schulman et al., instead of employing the approach of
Christensen, began with the cancer slope factor for CPA generated by California

Human Health Risk Assessment for Pharmaceuticals in the Environment

191

Table 2 Adjusted daily intake and therapeutic daily intake values for active pharmaceutical
ingredients in risk assessment exercises published to date
API
ADI/TDI
Value
Unit
References
17a-Ethinylestradiol
Abacavir
Acetominophen
Acetylsalicylate

Acyclovir/valacyclovir
Albendazole
Albuterol

Amoxycillintrihydrate
Amprenavir/fosamprenavir
Atenolol
Atomoxetine
Atorvastatin
Atovaquone
Beclomethasone
Benzylpenicillin
Betamethasone
Betaxolol
Bezafibrate
Bisoprolol
Bupropion
Carazolol
Carbamazepine

Carvedilol
Cefazolin
Ceftazidime
Cefuroxime
Celiprolol
Chloramphenicol
Chlorotetracycline
Cimetidine
Ciprofloxacin
Clarithromycin
Clavulanic acid
Clenbuterol

TDI
ADI
ADI
ADI
HBL
TDI
ADI
ADI
ADI
ADI
TDI
ADI
ADI
ADI
TDI
ADI
ADI
ADI
ADI
TDI
ADI
TDI
TDI
TDI
ADI
TDI
TDI
ADI
ADI, toxicological
ADI, therapeutic (child)
ADI, therapeutic (adult)
ADI
ADI
ADI
ADI
ADI
TDI
TDI
TDI
ADI
ADI
ADI
TDI
ADI
TDI

0.01
57.1
340
7
1
30
190
25.4
20.7
2.8
0.10
21.4
133
2.7
50
1.4
0.54
238
0.19
600
0.24
10
400
2.5
57.1
15
400
15.9
7.5
78
190
0.34
3
10
14.3
29
200
3,000
1,000
28.6
29
1.6
500
90
0.02

mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/day
mg/day
mg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/day

[32]
[22]
[24]
[29]
[21]
[32]
[22]
[22]
[22]
[24]
[32]
[22]
[22]
[30]
[32]
[18]
[30]
[22]
[22]
[32]
[22]
[32]
[32]
[32]
[22]
[32]
[32]
[23]
[27]
[27]
[27]
[29, 30]
[22]
[22]
[22]
[22]
[32]
[32]
[32]
[22]
[24]
[24]
[32]
[22]
[32]
(continued)

192
Table 2 (continued)
API
Clofibrate
Cloxacillin
Codeine
Cyclizine
Cyclophosphamide
Dehydrato-erythromycin
Diazepam
Diazepam
Diclofenac
Dicloxacillin
Digoxin
Diltiazem
Doxycycline
Doxycycline
Duloxetine
Dutasteride
Enalapril
Erythromycin-H2O
Fenofibrate
Fenoprofen
Fenoterol
Fluoxetine
Fluticasone
Gemfibrozil
Halofantrine
Hydrochlorothiazide
Ibuprofen
Ifosfamide
Indomethacin
Ketoprofen
Lamivudine
Lamotrigine
Lincomycin
Melphalan
Meprobamate

E.S. Williams and B.W. Brooks

ADI/TDI

Value

Unit

References

HBL
TDI
TDI
ADI
ADI
RSD
TDI
TDI
ADI
TDI
ADI
TDI
TDI
ADI
ADI
ADI
TDI
ADI
ADI
ADI
ADI
TDI
TDI
TDI
TDI
ADI
ADI
ADI
ADI
TDI
ADI
ADI
ADI
TDI
TDI
TDI
HBL
TDI
ADI
ADI
ADI
ADI
ADI, toxicological
ADI, therapeutic (child)
ADI, therapeutic (adult)
ADI

16.7
500
1,000
2
71.4
0.014
1
1,000
0.16
6
1.6
25
500
0.071
14
30
100
1.8
0.0016
0.23
40
100
100
900
0.50
2.9
1
0.095
0.56
1,200
200
6
110
1,200
2,160
50
1.67
100
15.9
11.9
25
0.0021
58
130
140
6.1

mg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day

[21]
[32]
[32]
[24]
[22]
[21]
[32]
[32]
[30]
[32]
[30]
[32]
[32]
[22]
[24]
[24]
[32]
[18]
[22]
[30]
[24]
[32]
[32]
[32]
[32]
[24]
[30]
[22]
[30]
[32]
[22]
[22]
[24]
[32]
[32]
[32]
[21]
[32]
[22]
[22]
[24]
[22]
[27]
[27]
[27]
[30]
(continued)

Human Health Risk Assessment for Pharmaceuticals in the Environment

Table 2 (continued)
API
Mercaptopurine
Metaprolol
Metformin
Metformin
Methicillin
Metropolol
Nabumetone
Nadolol
Nafcillin
Naproxen
Naratriptan
Norfloxacin
Olanzapine
Ondansetron
Oxacillin
Oxytetracycline
Pentoxifylline
Phenazon
Phenoxymethylpenicillin
Phenytoin

Proguanil
Propranolol
Ranitidine
Risperidone
Ropinirole
Rosiglitazone
Roxithromycin
Salicylic acid
Salmeterol
Simvastatin
Sotalol
Sulfamethazine
Sulfamethoxazole
Sulfamethoxazole
Sulfamethoxazole
Sulfathiozole
Sumatriptan
Terbutalin
Tetracycline

193

ADI/TDI

Value

Unit

References

ADI
ADI
ADI
ADI
TDI
TDI
ADI
TDI
TDI
ADI
ADI
ADI
ADI
ADI
TDI
ADI
TDI
TDI
TDI
ADI
TDI
ADI, toxicological
ADI, therapeutic (child)
ADI, therapeutic (adult)
ADI
ADI
TDI
ADI
ADI
ADI
ADI
ADI
TDI
TDI
ADI
ADI
TDI
TDI
ADI
ADI
TDI
ADI
ADI
TDI
ADI
TDI

0.0021
14
79.4
62
2,000
25
476
40
1,000
46
1.2
190
1.4
2.38
1,000
30
1,000
1,200
150
36
1,000
10
42
100
0.083
95.2
30
10.7
11
0.014
0.12
0.7
150
3,000
0.05
0.54
80
2,000
130
280
800
50
23.8
0.25
30
1,000

mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/day
mg/day
mg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/day
mg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/day

[22]
[29]
[22]
[24]
[32]
[32]
[22]
[32]
[32]
[30]
[22]
[24]
[18]
[22]
[32]
[24]
[32]
[32]
[32]
[29]
[32]
[27]
[27]
[27]
[30]
[22]
[32]
[22]
[24]
[30]
[22]
[22]
[32]
[32]
[22]
[30]
[32]
[32]
[24, 29]
[30]
[32]
[24]
[22]
[32]
[24]
[32]
(continued)

194
Table 2 (continued)
API
Timolol
Topotecan
Triamterene
Triclosan
Trimethoprim

Triprolidine
Vinorelbine
Warfarin
Zanamivir
Zidovudine

E.S. Williams and B.W. Brooks

ADI/TDI

Value

Unit

References

TDI
ADI
ADI
ADI
ADI
ADI
ADI
TDI
ADI
ADI
ADI
ADI
ADI

20
0.0021
71.5
12
9.4
4.2
100
200
4.8
0.0021
0.16
3.14
14.3

mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day
mg/kg/day

[32]
[22]
[22]
[30]
[22]
[24]
[30]
[32]
[22]
[22]
[24]
[22]
[22]

EPA [20]. A dose level of 1 mg/day was established as a safety value, based on an
excess cancer risk level of 1 105. This figure was cited in risk characterization
performed by Webb et al. [31].
As these ADIs accumulate in the literature, they have been and will continue to
be used in further assessments of potential risks from APIs in the environment
[28, 29]. In many cases, these ADIs and TDIs were extrapolated to ambient and
drinking water concentrations that corresponded to safe levels (Table 3). These
values included PNECs, drinking water equivalent levels (DWELs), ambient
water quality guideline values (AWQCs), and proposed provisional guideline
values (PGVs).
Thresholds of toxicologic concern (TTC) were also employed for 5-FU and other
chemotherapeutics [21, 24]. In a rather qualitative approach, a potential range doses
of 5-FU was compared to the general TTC of 1.5 mg/person/day recommended by
Kroes et al. [143]. The TTC has also been also applied more quantitatively in HHRA
practice, as it was used to set an ADI for four oncology drugs, for which too few
data were available to assess the carcinogenic potency [21]. Schriks et al. and
Rowney et al. also compared their predicted exposure values to the TTC, aside from
the comparisons to other substance-specific safety values found in the peer-reviewed
literature and elsewhere [27, 28].

Risk Characterization
In virtually all studies conducted to date, risk of adverse health effects from exposure to an API through drinking water or fish ingestion was judged to be negligible
[1730]. One exception was from Christensens 1998 publication in which the risk

Human Health Risk Assessment for Pharmaceuticals in the Environment

195

Table 3 Safety values for surface and drinking water for active pharmaceutical ingredients in risk
assessment exercises published to date
Environmental criterial
COPCs
value type
Values
Unit
References
Abacavir
Acetominophen
Acetylsalicylate
Acyclovir/valacyclovir
Albendazole
Albuterol
Amoxycillintrihydrate
Amprenavir/fosamprenavir
Atenolol
Atomoxetine
Atorvastatin
Atovaquone
Beclomethasone
Betamethasone
Bupropion
Carbemazepine

Carvedilol
Cefazolin
Ceftazidime
Cefuroxime
Cimetidine
Ciprofloxacin
Clavulanic acid
Clofibrate
Clofibric acida
Codeine
Cyclizine
Cyclophosphamide
Dehydronifedipinea
Diazepam
Diclofenac
Digoxigenina
Digoxin
Digoxin
Diltiazem
Doxycycline
Duloxetine

PNECDW+F (child)
PNECDW+F
PGV
AWQCDW
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F
PNECDW+F (child)
PNECDW+F (child)
DWEL
PNEC (adults)
PNEC (children)
DWEL
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
DWELDW-toxicity
PGV
DWEL
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F
PNECDW+F
PNECDW+F (child)
AWQCDW
PGV
PNECDW+F
PNECDW+F (child)
AWQCDW
PNECDW+F
DWEL
DWEL
PNECDW+F
PNECDW+F (child)
PNECDW+F
PNECDW+F
PNECDW+F
PNEC (adults)
PNEC (children)

8.20E + 05
4.90E + 06
3.00E + 01
4.80E + 05
2.70E + 06
3.60E + 05
3.00E + 05
4.00E + 04
3.10E + 05
1.90E + 06
8.10E + 01
3.43E + 01
2.57E + 01
1.60E + 01
1.10E + 06
2.00E + 03
3.40E + 03
8.20E + 05
2.26E + 05
3.50E + 05
1.00E + 00
1.00E + 01
4.30E + 04
1.40E + 05
2.00E + 05
4.20E + 05
4.10E + 05
4.10E + 05
2.30E + 04
1.30E + 06
2.20E + 05
3.00E + 01
2.90E + 04
1.00E + 06
4.80E + 02
1.10E + 06
4.80E + 00
4.80E + 01
1.00E + 03
1.00E + 03
1.00E + 03
2.00E + 05
4.30E + 05
2.83E + 01
1.91E + 01

ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L
ng/L
ng/L
ng/L
ng/L
mg/L
mg/L

[22]
[24]
[29]
[21]
[22]
[22]
[22]
[24]
[22]
[22]
[30]
[18]
[18]
[30]
[22]
[22]
[22]
[22]
[23]
[27]
[29]
[30]
[22]
[22]
[22]
[22]
[22]
[24]
[24]
[22]
[21]
[29]
[24]
[22]
[21]
[24]
[30]
[30]
[24]
[22]
[24]
[24]
[24]
[18]
[18]
(continued)

196

E.S. Williams and B.W. Brooks

Table 3 (continued)
COPCs
Dutasteride
Enalapril
Enalaprilata
Erythromycin-H2O
Fluoxetine
Fluticasone
Gemfibrozil
Halofantrine
Hydrochlorothiazide
Ibuprofen
Indomethacin
Lamivudine
Lamotrigine
Lincomycin
Melphalan
Meprobamate
Mercaptopurine
Metaprolol
Metformin
Nabumetone
Naproxen
Naratriptan
Norfloxacin
Olanzapine
Ondansetron
Oxytetracycline
Paroxetine metabolitea
Phenazone
Phenytoin
Proguanil
Ranitidine
Risperidone
Ropinirole
Rosiglitazone
Salmeterol
Simvastatin
Sulfamethoxazole

Environmental criterial
value type

Values

Unit

References

PNECDW+F (child)
DWEL
PNECDW+F
PNECDW+F
PNECDW+F
DWEL
PNECDW+F (child)
PNECDW+F
DWEL
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F
AWQCDW
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F
PNECDW+F (child)
DWELDW-toxicity
DWEL
PNECDW+F (child)
PGV
PNECDW+F (child)
PNECDW+F
PNECDW+F (child)
DWEL
PNECDW+F (child)
PNECDW+F
PNEC (adults)
PNEC (children)
PNECDW+F (child)
PNECDW+F
PNECDW+F
PGV
DWELDW-toxicity
DWEL
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F
DWEL
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F (child)
DWEL
PGV
PNECDW+F
DWEL

1.00E + 01
6.90E + 00
1.00E + 06
5.70E + 05
4.10E + 04
3.00E + 01
1.20E + 03
7.90E + 05
3.90E + 01
2.90E + 06
8.70E + 04
1.60E + 06
7.80E + 05
2.30E + 05
1.70E + 05
3.60E + 05
3.00E + 01
2.63E + 05
1.80E + 02
3.00E + 01
5.00E + 01
1.10E + 06
8.90E + 05
5.20E + 06
1.40E + 03
1.70E + 04
2.70E + 06
4.40E + 01
3.59E + 01
3.40E + 04
4.30E + 05
4.10E + 04
1.25E + 02
2.03E + 06
5.80E + 00
1.40E + 06
1.50E + 05
1.60E + 05
4.10E 01
1.70E + 03
1.00E + 04
7.20E + 02
1.60E + 01
4.40E + 02
1.90E + 06
8.40E + 03

ng/L
mg/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
mg/L

[22]
[30]
[24]
[24]
[24]
[30]
[22]
[24]
[30]
[22]
[22]
[24]
[21]
[22]
[22]
[24]
[22]
[27]
[30]
[22]
[29]
[22]
[24]
[22]
[30]
[22]
[24]
[18]
[18]
[22]
[24]
[24]
[29]
[27]
[30]
[22]
[22]
[24]
[30]
[22]
[22]
[22]
[30]
[29]
[24]
[30]
(continued)

ng/L
ng/L
mg/L
ng/L
ng/L
mg/L
mg/L
ng/L
ng/L
ng/L
mg/L
ng/L
mg/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
mg/L
mg/L
ng/L
mg/L

Human Health Risk Assessment for Pharmaceuticals in the Environment

197

Table 3 (continued)
COPCs
Sulfathiozole
Sumatriptan
Tetracycline
Topotecan
Triamterene
Triclosan
Trimethoprim

Triprolidine
Vinorelbine
Warfarin
Zanamivir
Zidovudine

Environmental criterial
value type

Values

Unit

References

PNECDW+F
PNECDW+F (child)
PNECDW+F
PNECDW+F (child)
PNECDW+F (child)
DWEL
PNECDW+F (child)
PNECDW+F
DWEL
PNECDW+F (child)
PNECDW+F (child)
PNECDW+F
PNECDW+F (child)
PNECDW+F (child)

7.20E + 05
3.40E + 05
4.30E + 05
3.00E + 01
1.00E + 06
3.60E + 02
1.30E + 05
6.00E + 04
3.00E + 03
6.90E + 04
3.00E + 01
2.30E + 03
4.50E + 04
2.00E + 05

ng/L
ng/L
ng/L
ng/L
ng/L
mg/L
ng/L
ng/L
mg/L
ng/L
ng/L
ng/L
ng/L
ng/L

[24]
[22]
[24]
[22]
[22]
[30]
[22]
[24]
[30]
[22]
[22]
[24]
[22]
[22]

A metabolite of a parent pharmaceutical compound

of allergic reactions to phenoxymethylpenicillin was judged to be possible for very

sensitive persons [25]. The author notes that assumptions on biodegradability of
the substance play a very important role in the qualitative conclusion and also that
several of the exposure assumptions taken during the exercise were conservative
such that risk is unlikely.
The author also assessed human health risks to a Dutch population arising from
the presence of 17a-EE2 and CPA [25]. Christensen concluded that the reasonable
worst case exposure for EE2 was 0.085 mg/day (i.e., 6.32 107 mg EE2/kg body
weight/day) and suggested that no significant risk exists in this context, as prepubescent boys produce 6 mg/day endogenously and adult males produce 4548 mg.
The author notes that the possibility of carcinogenic outcomes from CPA cannot be
ruled out, as it is thought to proceed through a nonthreshold MOA. These exposures were apportioned among several potential routes of exposure, including
ingestion of fish tissue, crops, drinking water, dairy products, meat, and inhalation
of ambient air.
Another example of potential risk was illustrated by Kummerer and Al-Ahmad
[18]. They concluded that the potential for excess cancers in children could not be
fully excluded as a result of exposures to CPA through drinking water, primarily as
a result of the use of additional UFs. The authors noted that neither Schulman et al.
nor Webb et al., both of whom assessed risk from CPA, accounted for the possibility of increased risk to children [20, 31]. Another cytotoxic chemotherapy drug,
5-FU, was judged as unlikely to pose a risk to human health, as the MOS between
intake values and therapeutic doses was 1081010 and between 300 and 30,000 for
a TTC value.

198

E.S. Williams and B.W. Brooks

The other HHRAs published to date indicate a de minimus level of risk with
regard to most APIs in the environment of developed counties [1730]. The MOS
and HQs are noted in Table 4. As might be expected, MOS were higher for adults
than for children, when such comparisons are made [17, 26].
Schwab et al. performed a complex comparison between three separate PNECs
(DW, fish, combined) and MECs from USGS data and PECs generated via PhATE
modeling [23]. Their hazard quotients (HQs) ranged from 0.33 for ciprofloxacin
(PNECDW+F to PhATE PEC) to 9.1 108 for oxytetracycline (PNECF to PhATE
PEC). Aside from ciprofloxacin, HQs for metformin (0.11), ranitidine (0.1), and
warfarin (0.11) were also within an order of magnitude of a conclusion of potential
human health risks. The authors concluded that the presence of low levels of APIs
in surface waters and drinking water poses no appreciable risk to human health.
Webb et al., using approaches they had previously published, compared daily
intake of APIs in drinking water (based on a monitored value in German surface
water) to recommended daily therapeutic dosages [31]. The MOS between the calculated intake and therapeutic dosage values ranged from 1 104 (salbutamol) to
1 109 (ioxitalamic acid and iothalamic acid, X-ray contrast media).
Cunningham et al. determined that the MOS between their calculated PNECs
and PECs for CBZ and two metabolites ranged between 340 and 6,560, and as such
there was no appreciable risk to human health from environmental exposures from
drinking water and fish consumption [22]. In an earlier study, they calculated MOS
for North America and the EU ranging from 14 (amoxicillin, NA) to 1.79 1010
(halofantrine, EU). Bercu et al. calculated MOS between 147 and 642 for the three
neuroactive APIs in their study [17]. Crane et al. pursued a tiered risk characterization, in which they targeted APIs with an MOS of less than 1,000 for further study;
only nine of the 364 compounds under study fell within that MOS range and three
of these (d-9-THC, cocaine, and LSD) are illegal drugs [30].

Microbial Resistance
A potential indirect risk from environmental APIs (and vAPIs) is the development
of microbial strains that are resistant to antibiotics, as a result of ongoing selection
in WWTP sludge and other compartments [33, 144, 145]. Bacterial strains resistant
to multiple antibiotics have been detected in biofilms formed in surface water and in
drinking water distribution systems [146]. Experiments have also shown that bacteria found in hospital wastewater effluent are resistant to different types of antibiotics
than other effluents [146]. A greater prevalence of Acetinobacter bacteria were
observed downstream from hospital and API manufacturing wastewater outflows
than in the upstream area [147]. Similar results were seen for Escherichia coli in
wastewater and for Enterobacteriaceae and Aeromonas species in surface water
affected by urban effluents [148, 149].
The possibility remains that increased number of resistant organisms observed in
the environment is related to excretion of such microbes from the gut of humans and
animals being treated with antibiotics [21, 140, 150]. Per Kummerer, a strain of

Alkylating chemotherapeutic
agentsa

Albuterol

Albendazole

Acyclovir/valacyclovir

Acetylsalicylate

Acetominophen

5-Fluorouracil
Abacavir

17a-Ethinylestradiol
MOS
MOS
HQ
HQ
HQ
HQ
HQ
HQ

MOS
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
MOS
HQ
MOS
HQ
MOS

Undefined
TD/DWI
TTC/DWI
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
Concmax-DW/PGV

Concmax-SW/PGV

Concmax/AWQC
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
TD/DWI
DWE/TTC
TTC/DWE
DWE/TTC
TTC/DWE

Qualitative
1.50E + 06
3.70E 04
1.90E 04
4.40E 07
6.00E 07
6.10E 05
2.60E 04
3.60E 04
6.20E 03
1.00E + 04
9.0E 031.4E 01
7.0E + 001.1E + 02
1.8E 022.7E 01
3.8E + 005.4E + 01

3.00E 03

Qualitative
1.00E + 04
3.0E + 023.0E + 04
7.60E 05
2.40E 05
2.10E 03
9.70E 02
5.00E 03

Insignificant risk
No risk to human health
Unlikely to pose a risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable concern for human
health
No appreciable concern for human
health
Not a risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low

Table 4 Risk characterization metrics for active pharmaceutical ingredients in risk assessment exercises published to date
Pharmaceutical
RC type
Value
Conclusion
References

(continued)

[21]
[32]
[22]
[22]
[22]
[22]
[22]
[22]
[24]
[24]
[32]
[28]
[28]
[28]
[28]

[29]

[26]
[32]
[25]
[22]
[22]
[24]
[24]
[29]

Human Health Risk Assessment for Pharmaceuticals in the Environment

199

Betaxolol
Bezafibrate
Bisoprolol
Bupropion

Benzylpenicillin
Betamethasone

Beclomethasone

Atorvastatin
Atovaquone

Atomoxetine

Atenolol

NA PEC/PNEC
EU PEC/PNEC
TD/DWI
TD/DWI
TD/DWI
NA PEC/PNEC
EU PEC/PNEC

NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
DWE/TTC
TTC/DWE
DWE/TTC
TTC/DWE
DWE/TTC
TTC/DWE
DWE/TTC
TTC/DWE
Concmax-DW/DWEL
TD/DWI
Adults
Children
Concmax-DW/DWEL
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC

RC type

Antimetabolite chemotherapeutic agentsc

Anthracycline antibioticsb

Amprenavir/fosamprenavir

Amoxycillin trihydrate

Table 4 (continued)
Pharmaceutical
HQ
HQ
HQ
HQ
HQ
MOS
HQ
MOS
HQ
MOS
HQ
MOS
HQ
MOS
MOS
MOS
HQ
HQ
HQ
HQ
HQ
MOS
HQ
HQ
MOS
MOS
MOS
HQ
HQ

Value
6.70E 02
1.40E 02
4.70E 06
1.30E 06
6.7E 044.0E 05
1.5E + 032.5E + 04
1.2E 038.0E 04
8.3E + 021.3E + 04
1.7E 032.7E 02
3.7E + 015.7E 02
3.5E 035.0E 02
2.0E + 012.9E + 02
2.50E 04
5.00E + 06
2.86E + 02
2.14E + 02
1.50E 04
2.10E 06
2.10E 05
4.10E 05
4.80E 04
6.00E + 06
1.20E 03
2.70E 03
1.00E + 06
7.41E + 06
2.50E + 05
2.40E 04
4.60E 06

Conclusion
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low
Risk to healthy adults is low
Not relevant to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health

References
[22]
[22]
[22]
[22]
[28]
[28]
[28]
[28]
[28]
[28]
[28]
[28]
[30]
[32]
[18]
[18]
[30]
[22]
[22]
[22]
[22]
[32]
[22]
[22]
[32]
[32]
[32]
[22]
[22]

200
E.S. Williams and B.W. Brooks

Clarithromycin

Ciprofloxacin

Celiprolol
Chloramphenicol
Chlorotetracycline
Cimetidine

Cefuroxime

Ceftazidime

Cefazolin

Carvedilol

Carazolol
Carbamazepine

MOS
MOS
MOS
MOS
MOS
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
MOS
MOS
MOS
HQ
HQ
HQ
HQ
HQ
HQ
MOS

TD/DWI
TD/DWI
TD/DWI
PNEC/MEC90th
PNEC/PEC90th
Average, DW
Concmax-SW/PGV

Concmax-DW/PGV

Concmax-DW/DWEL
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
TD/DWI
TD/DWI
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
HQMEC-DW+F
HQPEC-DW+F
TD/DWI

1.80E 03
1.00E 04
9.90E 04
6.50E 03
3.50E 03
6.70E 04
3.20E 04
3.60E 04
2.30E 03
2.00E + 07
7.50E + 07
2.50E + 07
7.40E 04
3.10E 04
1.40E 03
2.20E 02
1.30E 03
3.30E 01
1.25E + 07

2.00E 01

1.50E + 06
6.67E + 06
4.00E + 06
1.50E + 03
6.80E + 02
1.30E 05
3.00E 02

No risk to human health

No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No appreciable concern for human
health
No appreciable concern for human
health
Not relevant to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
[30]
[22]
[22]
[22]
[22]
[22]
[22]
[22]
[22]
[32]
[32]
[32]
[22]
[22]
[24]
[24]
[24]
[24]
[32]
(continued)

[29]

[32]
[32]
[32]
[23]
[23]
[27]
[29]

Human Health Risk Assessment for Pharmaceuticals in the Environment

201

Diclofenac

Diazepam

Dehydrato-erythromycin
Dehydronifedipined

Cyclophosphamide

Cyclizine

Cloxacillin
Codeine

Clenbuterol
Clofibrate
Clofibrate
Clofibric acidd

Clavulanic acid

Table 4 (continued)
Pharmaceutical

MOS
MOS
HQ
HQ
HQ
HQ

TD/DWI
TD/DWI
HQMEC-DW+F
HQPEC-DW+F
NA PEC/PNEC
EU PEC/PNEC
Undefined
RR1.5
RR1.5
Concmax/AWQC
TD/DWI
TD/DWI
TD/DWI
HQMEC-DW+F
HQPEC-DW+F
Concmax-DW/DWEL
TD/DWI

TD/DWI

HQ

Concmax-DW/PGV

MOS
MOS
MOS
HQ
HQ
HQ
MOS
HQ
MOS

HQ
HQ

MOS
HQ

HQ
HQ
MOS

NA PEC/PNEC
EU PEC/PNEC
TD/DWI
Concmax/AWQC
TD/DWI
Concmax-SW/PGV

RC type

Value

3.57E + 06
1.00E + 07
3.50E 02
8.40E 02
2.90E 07
1.60E 06
Qualitative
2.90E 05
2.10E 05
Qualitative
5.00E + 04
1.00E + 04
2.50E + 07
2.60E 05
2.00E 03
5.20E 05
1.50E + 05
5.20E 06
2.08E + 06

3.00E 03

2.50E 04
3.20E 04
1.00E + 03
Qualitative
1.25E + 07
5.00E 03

Conclusion
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
Not a risk to human health
No risk to human health
No appreciable concern for human
health
No appreciable concern for human
health
No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
Unlikely to contribute a significant risk
Unlikely to pose a risk to human health
Unlikely to pose a risk to human health
Not a risk to human health
No risk to human health
No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No risk to human health
Not relevant to human health
No risk to human health

References

[32]
[32]
[24]
[24]
[22]
[22]
[26]
[19]
[19]
[21]
[32]
[32]
[32]
[24]
[24]
[30]
[32]
[30]
[32]

[29]

[22]
[22]
[32]
[21]
[32]
[29]

202
E.S. Williams and B.W. Brooks

Fenoprofen
Fenoterol
Fluoxetine

Etofibrate
Fenofibrate

Erythromycin-H2O

Enalapril
Enalaprilatd

Dutasteride

Duloxetine

Doxycycline

Diltiazem

Digoxigenind

Dicloxacillin
Digoxin

TD/DWI
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
HQMEC-DW+F
HQPEC-DW+F
HQMEC-DW+F
HQPEC-DW+F
HQMEC-DW+F
HQPEC-DW+F
TD/DWI
Adults
Children
NA PEC/PNEC
EU PEC/PNEC
Concmax-DW/DWEL
HQMEC-DW+F
HQPEC-DW+F
HQMEC-DW+F
HQPEC-DW+F
TD/DWI
TD/DWI
TD/DWI
TD/DWI
TD/DWI
HQMEC-DW+F
HQPEC-DW+F
Concmax-DW/DWEL

MOS
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
MOS
MOS
MOS
MOS
MOS
HQ
HQ
HQ
HQ
HQ
MOS
MOS
MOS
MOS
MOS
HQ
HQ
HQ

5.00E + 06
2.00E 03
6.40E 04
N/A
1.30E 02
3.90E 03
1.30E 02
2.40E 04
6.00E 02
1.20E 04
4.30E 03
2.50E + 05
2.18E + 02
1.47E + 02
2.90E 03
2.30E 03
3.60E 05
4.60E 05
6.30E 05
3.00E 03
1.30E 02
7.50E + 06
2.50E + 06
1.19E + 06
9.00E + 07
5.00E + 04
2.90E 04
3.10E 02
1.70E 05

No risk to human health

No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
No risk to human health
No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
(continued)

[32]
[22]
[22]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[24]
[32]
[18]
[18]
[22]
[22]
[30]
[24]
[24]
[24]
[24]
[32]
[32]
[32]
[32]
[32]
[24]
[24]
[30]

Human Health Risk Assessment for Pharmaceuticals in the Environment

203

Melphalan

Lincomycin

Lamotrigine

Ketoprofen
Lamivudine

Indometacine

Ifosfamide

Ibuprofen

Hydrochlorothiazide

Halofantrine

Gemfibrozil

Fluticasone

Table 4 (continued)
Pharmaceutical

NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
Concmax-DW/DWEL
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
TD/DWI
TD/DWI
TD/DWI
RR1.5
RR1.5
TD/DWI
Concmax/AWQC
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
NA PEC/PNEC
EU PEC/PNEC

RC type

MOS
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ

HQ
HQ
HQ
HQ
HQ
MOS
MOS
MOS
HQ
HQ
MOS

HQ
HQ
HQ
HQ
HQ

Value
1.70E 03
2.00E 04
1.00E 03
2.20E 02
5.40E 05
1.20E + 08
N/A (no sales)
5.60E 11
2.60E 02
1.70E 02
6.40E 04
8.30E 02
2.00E + 08
1.08E + 08
2.16E + 07
5.60E 05
1.10E 03
5.00E + 06
Qualitative
1.00E + 07
1.80E 03
5.00E 04
4.10E 05
1.70E 04
2.00E 03
5.90E 05
2.60E 04
2.00E 04

Conclusion
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
No risk to human health
Unlikely to pose a risk to human health
Unlikely to pose a risk to human health
No risk to human health
Not a risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health

References
[22]
[22]
[24]
[24]
[30]
[32]
[22]
[22]
[22]
[22]
[24]
[24]
[32]
[32]
[32]
[19]
[19]
[32]
[21]
[32]
[22]
[22]
[22]
[22]
[24]
[24]
[22]
[22]

204
E.S. Williams and B.W. Brooks

Oxacillin

Ondansetron

Olanzapine

Norfloxacin

Nadolol
Nafcillin
Naproxen
Naratriptan

Methicillin
Metropolol
Nabumetone

Metformin

Metaprolol

Mercaptopurine

Meprobamate

HQ
HQ
HQ
HQ
MOS
MOS
HQ
HQ
MOS
MOS
HQ
HQ
HQ
HQ
HQ
MOS
MOS
HQ
HQ
MOS

HQ

Concmax-SW/PGV

NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
TD/DWI
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
TD/DWI
TD/DWI
Concmax-DW/DWEL
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
Adults
Children
NA PEC/PNEC
EU PEC/PNEC
TD/DWI

HQ
HQ
HQ
HQ
HQ

Average, DW
Concmax-DW/DWEL
NA PEC/PNEC
EU PEC/PNEC
Concmax-DW/PGV

1.60E 02
2.00E 02
1.70E 04
1.10E 01
2.00E + 07
2.50E + 06
2.40E 04
4.90E 06
4.00E + 06
1.00E + 07
3.60E 07
8.70E 06
9.00E 06
4.40E 05
5.90E 05
6.42E + 02
5.24E + 02
1.40E 03
7.80E 04
1.00E + 07

4.00E 03

2.48E 05
2.40E 04
2.40E 02
3.80E 02
4.00E 02

No risk to human health

Not relevant to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable concern for human
health
No appreciable concern for human
health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
Not relevant to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
[22]
[22]
[24]
[24]
[32]
[32]
[22]
[22]
[32]
[32]
[30]
[22]
[22]
[24]
[24]
[18]
[18]
[22]
[22]
[32]
(continued)

[29]

[27]
[30]
[22]
[22]
[29]

Human Health Risk Assessment for Pharmaceuticals in the Environment

205

Rosiglitazone

Risperidone
Ropinirole

Propranolol
Ranitidine

Proguanil

Phenytoin

Phenoxymethylpenicillin

Phenazon

Pentoxifylline

Oxytetracycline

Table 4 (continued)
Pharmaceutical

MOS

TD/DWI
Undefined
MOS
HQ
HQ
HQ
HQ
MOS
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ

HQ

Concmax-SW/PGV

TD/DWI
Average, DW
Concmax-DW/DWEL
NA PEC/PNEC
EU PEC/PNEC
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
Concmax-DW/DWEL
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC

MOS
MOS
MOS
HQ

HQ
HQ

HQMEC-DW+F
HQPEC-DW+F
TD/DWI
TD/DWI
TD/DWI
TD/DWI
Concmax-DW/PGV

RC type

Value

1.00E + 07
1.84E 05
2.60E 03
6.60E 06
1.00E 05
3.00E + 06
5.30E 03
1.90E 02
3.10E 03
2.70E 02
8.40E 04
1.20E 04
7.50E 05
5.00E 04
1.60E 04

1.50E + 06
Qualitative

9.00E 04

7.90E 04
4.50E 06
2.50E + 07
6.00E + 07
3.00E + 07
1.50E + 06
2.00E 04

Conclusion
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
No risk to human health
No risk to human health
No appreciable concern for human
health
No appreciable concern for human
health
No risk to human health
Risk is possible for very sensitive
persons
No risk to human health
No risk to human health
Not relevant to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health

References

[32]
[27]
[30]
[22]
[22]
[32]
[22]
[22]
[24]
[24]
[30]
[22]
[22]
[22]
[22]

[32]
[26]

[29]

[24]
[24]
[32]
[32]
[32]
[32]
[29]

206
E.S. Williams and B.W. Brooks

Triclosan

Triamterene

Timolol
Topotecan

Terbutalin
Tetracycline

Sumatriptan

Sulfathiozole

Simvastatin
Sotalol
Sulfamethazine
Sulfamethoxazole

Roxithromycin
Salicylic acid
Salmeterol

1.00E 03
9.70E 03
3.60E 07
2.00E + 07
3.50E 05
3.90E 05
3.80E 06
8.10E 07
1.25E + 04
2.60E 04
1.50E 02
2.50E + 07
2.00E + 06
1.60E 04
3.00E 04
1.50E 04
7.30E 05
3.30E 06

HQ
HQ
HQ

HQMEC-DW+F
HQPEC-DW+F
Concmax-DW/DWEL
TD/DWI
HQMEC-DW+F
HQPEC-DW+F
NA PEC/PNEC
EU PEC/PNEC
TD/DWI
HQMEC-DW+F
HQPEC-DW+F
TD/DWI
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
Concmax-DW/DWEL
MOS
HQ
HQ
HQ
HQ
HQ

HQ
HQ
HQ
HQ
MOS
HQ
HQ

3.00E 04

HQ

Concmax-SW/PGV

3.75E + 06
1.50E + 08
2.80E 06
1.70E 05
1.50E 05
8.00E + 06
5.00E + 07
7.00E 04

MOS
MOS
HQ
HQ
HQ
MOS
MOS
HQ

TD/DWI
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
Concmax-DW/DWEL
TD/DWI
TD/DWI
Concmax-DW/PGV

No risk to human health

No risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No risk to human health
No risk to human health
No appreciable concern for human
health
No appreciable concern for human
health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No risk to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
(continued)

[24]
[24]
[30]
[32]
[24]
[24]
[22]
[22]
[32]
[24]
[24]
[32]
[32]
[22]
[22]
[22]
[22]
[30]

[29]

[32]
[32]
[22]
[22]
[30]
[32]
[32]
[29]

Human Health Risk Assessment for Pharmaceuticals in the Environment

207

NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
Concmax-DW/DWEL
TD/DWI
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC
HQMEC-DW+F
HQPEC-DW+F
NA PEC/PNEC
EU PEC/PNEC
NA PEC/PNEC
EU PEC/PNEC

RC type
HQ
HQ
HQ
HQ
HQ
MOS
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ

Value
4.20E 03
1.30E 03
1.20E 02
6.20E 02
8.30E 08
5.00E + 06
4.20E 05
2.40E 05
6.70E 04
1.30E 03
2.20E 04
1.10E 01
8.90E 07
2.90E 06
1.50E 03
4.30E 04

Conclusion
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
Not relevant to human health
No risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health
No appreciable risk to human health

References
[22]
[22]
[24]
[24]
[30]
[32]
[22]
[22]
[22]
[22]
[24]
[24]
[22]
[22]
[22]
[22]

TD/DWI therapeutic dose/drinking water intake; TTC/DWI threshold of toxicologic concern/drinking water intake; Concmax-SW/PGV max. concentration in surface water/provisional guideline value; Concmax-DW/PGV max. concentration in drinking water/provisional guideline value; Concmax/AWQC max. concentration/
ambient water quality criteria; DWE/TTC drinking water exposure/threshold of toxicologic concern; TTC/DWE threshold of toxicologic concern/drinking water
exposure; Concmax-DW/DWEL max. concentration in drinking water/drinking water equivalent levels
a
Oxaliplatin, temozolomide, cisplatin, carboplatin, cyclophosphamide
b
Epirubicin, doxorubicin
c
Gemcitabine, fludarabine, capecitabine
d
Metabolite

Zidovudine

Zanamivir

Vinorelbine
Vinorelbine
Warfarin

Triprolidine

Trimethoprim

Table 4 (continued)
Pharmaceutical

208
E.S. Williams and B.W. Brooks

Human Health Risk Assessment for Pharmaceuticals in the Environment

209

bacteria known to be resistant to seven different antibiotics was released into a laboratory-simulated WWTP [150]. Within 2 weeks, the resistant bacterium could not be
detected in the system, suggesting that continuous input of resistant microbes may
be more important than selection events occurring inside a WWTP. Indeed, it has been
suggested that levels of antibiotics found in WWTP or aquatic compartments may be
too high to generate resistance [150]. Other studies have indicated that natural environments are more likely to be the reservoir of genetic material that confer resistance [2].
Kim and Aga propose a framework of risk assessment for the development of
antibiotic-resistant microorganisms and their eventual likely impact on human
health [144]. This conceptual model could provide a calculated probability of an
increase in the annual rate of antibiotic-resistant infections through ingestion in
drinking water or through other exposure pathways. Interestingly, Webb et al. note
that the ADI developed for trimethoprim in their HHRA exercise takes into account
the possibility of selection for resistant bacteria in natural gut flora, though it certainly does not consider the indirect events in WWTP [31].

Uncertainty
There are two main sources of uncertainty in HHRA: variability and lack of knowledge. Studies conducted to date on APIs in the environment and the potential for
adverse human health outcomes are fraught with issues related to these two sources
[35]. Several parameters associated with exposure and potential toxicity of APIs
can show high degrees of variability. Ironically, the uncertainties associated with
HHRA for APIs in the environment are slightly fewer than for ecological or environmental risk assessment (ERA), as in most cases a MOA for the COPC is better
understood than it would be for a nonhuman target.
The limited amount of information on the presence of these materials in the environment gives rise to uncertainty regarding exposure scenarios, and selection of
potential hazards for further study. Perhaps more importantly, there is also a dearth
of toxicity and doseresponse information, which induces most risk assessors to use
therapeutic doses as points of departure in assessment of doseresponse relationships. In particular, the ability of risk assessors to evaluate potential risks arising
from chronic, low-dose exposures is hindered by a lack of data. Further, no suitable
mechanism exists at this time for quantification of the importance of drugdrug
interactions as well, though several observations indicate that exposure to mixtures
can cause more complex responses [129].

Uncertainty in Substance Selection/Hazard Assessment

Several rationales were offered for API selection in the HHRAs completed to
date. One suggested procedure was to consider whether the APIs were among the

210

E.S. Williams and B.W. Brooks

top prescribed or to choose substances for which MECs were available [23]. The
work of Benotti et al. demonstrates that prescription information is a poor predictor of exposure in drinking water, as several of the APIs that persisted through
treatment were not in the top 200 prescribed drugs [130]. Though the work of
Crane et al. analyzed risks from 396 compounds, the vast majority of APIs have
not been assessed [30]. It is to be anticipated that future efforts will include prioritization of APIs to be analyzed in the environment [24, 26, 27]. Obviously, it is
sensible to focus on compounds that are expected to exert toxicological effects at
very low concentrations, including EDCs and genotoxic APIs such as cancer
chemotherapeutics.
Mixtures
Experts in the field of APIs in the environment seem to agree that there is a high
level of uncertainty regarding the potential significance of API mixture effects
[7, 129]. Part of this uncertainty stems from a lack of an accepted and validated risk
assessment methodology for exposure events of this type. For antibiotics in particular, if the possibility of additivity or synergism are not taken into account, it is likely
that real-world risk will be underestimated [140]. Some authors have suggested the
use of toxicogenomic or metabolomics approaches to understand mixture toxicology and ecotoxicology, and certainly new effort in this area is expected [151].
Bioassays, including the yeast estrogen screening (YES) assay, have been used to
assess the importance of mixtures of EDCs, and other classes of chemicals with
similar toxicological mechanisms [152154]; this general technique may prove useful for pharmaceuticals in the future. There is also a lack of information on drug
chemical interactions, which may be important [5, 155157]. Indeed, one author
has recommended a UF of 100,000 be applied to all APIs to account for mixtures,
sensitive subpopulations, and the more standard issues of differences in dose duration [158].
One of the published HHRAs to date attempts to deal quantitatively with the
potential significance of mixture effects [30]. The authors chose to combine potential concentrations of non-steroidal anti-inflammatory drugs (NSAIDs) into a
grouped parameter. Total NSAIDs in their exercise thus had the lowest MOS [77].
They adopted a similar approach for statins, though the margin of safety in that
case was over 1,000. Rowney et al. adopted a similar approach to alkylating (oxaliplating, temozolomide, cisplating, carboplatin, CPA) and antimetabolite (gemcitabine, fludarabine, capecitabine) chemotherapeutics, as well as anthracycline
antibiotics (epirubicin, doxorubicin) [27]. Kumar and Xagoraraki considered mixture effects of their analytes of interest and consulted the RxList internet drug
index and HSDB to determine whether any interaction between these compounds
had been observed in previous studies [26]; however, this approach would not consider the presence of other APIs in a real-life exposure scenario for human or
environmental receptors.

Human Health Risk Assessment for Pharmaceuticals in the Environment

211

Metabolites, Conjugates, and Degradation By-Products

The large number of potential targets for HHRA is further complicated by the generation of metabolites and conjugates through human and biotic environmental processes, and the potential generation of degradation by-products [18, 159]. In some
exercises, conjugates and metabolites were treated as parent compounds for the
purposes of risk assessment [17]. Deconjugation is an important possibility to
account for, as it can result in WWTP effluent concentrations being higher than
influent values [91]. Generally, the importance of metabolites is not considered in
risk assessments, probably because it adds another layer of complexity to an already
complex issue. Cunningham et al. did assess potential risks from CBZ-diol and
CBZ-n-glucuronide, the two major metabolites of CBZ, and found them unlikely to
pose a risk to human health [21]. However, the authors did note a lack of toxicology
data for these two substances and were forced to use essentially proxy data. The
same was true for Snyders analysis of hydroxylated metabolites of atorvastatin and
simstatin [29]. Also, conjugates were assumed to be returned to parent compound in
the environment. The lack of data on toxicological properties of metabolites and
conjugates of APIs would be expected to be the rule.

Uncertainty in Toxicity Values

In several of the HHRAs conducted to date, UFs were used to account for the more
obvious sources of variability: interspecies and interindividual differences, database
quality, etc. Rationales for application of various UFs in HHRA was again well
articulated by Schwab et al. [23]. In one instance, the authors mined clinical trial
data to derive chemical-specific adjustment factors (CSAFs) that ranged from 10 to
35 for variability in human responses [17]. Based on this information, it is worthwhile to reexamine the default factor of 10 used in many other studies. In another
exercise, total UFs integrated into an ADI estimate ranged from 9 to 1,000 across a
relatively small set of APIs [23].
Most of the data that are available to serve as a POD for establishment of
ADIs and/or PNECs is derived from acute toxicity endpoints. Safety values
using such values might be underestimating potential risks from chronic exposures [160]. The use of conservative UFs to correct for these concerns may provide safety values that are indeed protective, but the default UFs employed in
several HHRAs have generally been applied to industrial chemicals, as opposed
to APIs that are designed to generate an effect in a human target [2123].
Therapeutic dose values have also been used for PODs, and it is unclear as to
whether this is appropriate [23, 31]. Furthermore, there is a deficit of information on reproductive effects and mechanisms of action for many of the APIs
under examination [6, 15, 128, 161163]. There is also not enough information
regarding the bioaccumulative potential of APIs, as they were often estimated
based on log Kow values [17, 135].

212

E.S. Williams and B.W. Brooks

TTCs are derived from a class of compounds as opposed to a specific compound

of interest and as such as associated with significant uncertainty when used as a toxicity value [24, 25]. Though the value is expected to be protective as it is theoretically
based on the most potent substance in the most sensitive system, the possibility of
incomplete data cannot be discounted, among other potential uncertainties.
Theoretically, the use of a UF to account for interindividual differences should
be protective of sensitive subpopulations. These populations include the elderly
(who may be compromised in terms of their capacity for detoxification), children
(who receive comparatively larger doses), and pregnant women [129]. This factor
may also be adequate to account for individuals who are exposed to additional quantities of medications that they are consuming [20]. However, more study would be
useful in determining whether the default factor is sufficiently protective. Variations
in standard practice among nations could lead to different conclusions from very
similar estimated intakes [18, 20, 31]. It would also be helpful to understand whether
the current UFs are protective of mixture effects; if data suggest that mixture effects
play an important role in the toxicity of a substance, then it is possible an additional
UF may be needed.

Uncertainty in Exposure
Several of the available HHRAs considered potential human exposures to API
through ingestion of drinking water, surface water, and/or fish [18, 2123]. However,
none incorporated the possibility of exposures through other media or other routes.
Though it is likely that these routes do not contribute significantly to overall exposure, their exclusion is a source of uncertainty for the conclusions of the HHRA.
There has also been no exploration of potential exposures through unintentional
water reuse [33].
One of the primary uncertainties in exposure relates to lack of knowledge of how
APIs move through the environment, including degradation and partitioning to various environmental media [34, 160]. Though the literature has expanded greatly in
recent years on the topics of fate and behavior, as well as removal of APIs in WWTP
and drinking water processing, the data exist only for a limited number of substances and thus our ability to accurately model the occurrence of these APIs in the
environment is also limited.
There is of course uncertainty associated with MECs that would be used in
HHRAs, related to sampling bias and/or analytical methodology [56, 120, 132,
155]. The exploration of new techniques capable of analyzing the concentrations of
multiple APIs at once is an additional source, as well as the variable usage of the
available techniques between studies that are used to determine an MEC for exposure assessment. The seasonal and diurnal variability of API concentrations also
imparts uncertainty to any single value MEC used as an exposure value. As noted
by Daughton et al., the expansion of the universe of published MECs has paradoxically increased uncertainty surrounding concentrations of any single API in the

Human Health Risk Assessment for Pharmaceuticals in the Environment

213

environment [81]. Cunningham et al. rightly point out that MECs represent a snapshot,
or at best a series of snapshots, and suggest that the use of models may enable the
assessor to take into account more variables [22].
Modeling of PECs also brings a certain level of uncertainty, as so little data exist
with which to validate the calculations. Some issues have been raised with the
accuracy of the most basic calculation typically used to create PECs, as employed
by Kummerer and Al-Ahmad [18, 49]. Generation of PECs through more complex
models such as PhATE, iSTREEM or GREAT-ER would include uncertainty
through required input parameters, including usage rates based on prescription
numbers [37], variability in excretion of metabolites or conjugates [18, 56], fate
and behavior of API in multiple compartments [164], flow conditions and surface
water usage patterns [23], the removal of APIs in WWTP [76, 165, 166], and relative source contributions [4, 56, 124, 133, 167]. The accuracy of computer models
will always be dependent upon the quality of the data entered and the assumptions
associated with the model [133]. In that context, it is difficult to provide the model
with data on metabolism, excretion, or fate and behavior that can accurately portray the inherent variability of these parameters. For example, it has been shown
that ibuprofen can degrade in the environment at half-lives ranging from less than
1 to 50 days [160].
Efforts to understand detailed usage patterns may be seen as an intrusion on
medical record-keeping, and thus such parameters may be difficult to compile.
Regardless, information on prescription volume may be uninformative as to ultimate exposure concentrations that would be used in an HHRA [130]. Further, the
usage of illicit substances is even less well understood than legal APIs. The quality
of data on sales and production of APIs is somewhat unclear, so this is also a source
of uncertainty in the modeling of PECs [36, 84]. Schwab et al. note that PECs generated by PhATE rely on the per capita consumption of the APIs under study and as
such may underestimate exposure in some areas [23].
Also, it is worth nothing that in North America, PhATE does not cover urban
areas, whose water supplies are not thought to be significantly impacted by WWTP
effluent. It would be informative to have information from many of the populous
areas not covered by the PhATE model, whose water supplies are not derived from
sources impacted by WWTP.
Little effort has been devoted to quantitative or semiquantitative evaluations of
uncertainty. Boeije et al. performed a Monte Carlo analysis to better understand the
uncertainty surrounding the removal of APIs in WWTP and surface water [165].
Also, MCA was used to examine the uncertainty associated with degradation of
ibuprofen and naproxen as a result of ozonation [122]. These techniques have been
applied to risk assessments for ecological receptors [168, 169]. Variables that could
be assessed via probabilistic distributions have been identified above, including
temporal flow conditions [76], mixture effects [157], and obviously environmental
API concentrations [155, 170]. To an extent, distributions for WWTP removal and
degradation are included in modeling tools for exposure concentrations [133, 164].
Recently, a HHRA published by Kumar and Xagoraraki performed Monte Carlo
simulations in an attempt to assess uncertainty in ADI and exposure parameters

214

E.S. Williams and B.W. Brooks

water ingestion rate, API concentration, and body weight. Ultimately, they concluded
that variability in the ADI contributed more than 95% of variability to the risk
estimates in all risk scenarios [26]. This finding may serve to highlight the need
to develop more robust measures of doseresponse relationships for APIs in the
environment.

Conclusion
Current practice in HHRA for APIs in the environment centers on a number of uncertainties. Firstly, the scientific community is ill equipped at this point to generate a
reliable estimate of exposure. Most of the efforts conducted to date employ modeling
of exposure concentrations using PhATE or GREAT-ER, as few MECs are available
for most APIs in drinking water. Also, the relative importance of exposures through
surface water and soil is very poorly understood. In the setting of safety values, there
is inconsistency in the choice of PODs, as some practitioners begin with therapeutic
doses and some with NOELs or LOELs. It is also clear that risk assessors are not well
served by considering pharmaceuticals as a broad class.
Regardless, the assessments conducted to date all indicate a negligible degree of
risk associated with human exposures to APIs through drinking water and/or ingestion of fish tissue. As illustrated above, these conclusions are associated with a
significant degree of uncertainty as to the potential hazards of long-term, low-dose
exposure to APIs, lack of understanding of exposure in most developing countries,
and also with regard to the potential additive or synergistic effects of API mixtures.
Almost no effort has been expended on quantification of the uncertainties surrounding risk conclusions made so far through probabilistic techniques, though undoubtedly this is a next step in the natural progression of HHRA practice in this area. Due
to the conservative nature of most of the assumptions taken in HHRAs performed so
far, it is unlikely we will see any evidence of significant risk from APIs unless further study uncovers contamination of surface and drinking water on a large scale.
The areas in which a critical need exists for further research include:
1. Realistic monitoring of the occurrence of APIs in multiple environmental
compartments
It has been stated that most monitoring of APIs has been undertaken in areas
where contamination is expected. This is somewhat comforting, as it suggests that
the use of such data in HHRA will provide a conservative risk estimate. However,
the overall lack of comprehensive data is a significant area of uncertainty in such
exercises. Also, this shortcoming makes it difficult for regulatory agencies and
members of the public to prioritize the need for further API research and fully
understand potential exposures.
2. Fate and behavior of APIs
The ability of APIs to move through the environment has already been demonstrated for a number of selected compounds. With the vast number of substances at

Human Health Risk Assessment for Pharmaceuticals in the Environment

215

issue, it is understandable that relatively little information is available on the

mechanisms through which APIs partition to different compartments, are degraded
by biotic and abiotic processes, and persist in the environment. Again, more information on more pharmaceutical substances would be useful in understanding potential exposures and in prioritizing APIs for further study or regulatory action.
3. Efficacy of various technologies for the removal of APIs from waste- and drinking water
As explored more in Chap. 9 of this volume, the removal of APIs varies among
technologies employed in WWTP and drinking water treatment. It is also clear that
removal is not complete and that complete information is not available for the suite
of APIs that enter the environment. From an exposure perspective, this information
is invaluable; any computer model designed to predict exposure concentrations
would greatly benefit from further study in this area.
4. Robust exposure assessment, including further validation of current computer
models
Many potential exposure pathways exist for APIs in the environment, as described
in Fig. 2. To date, only a few studies have examined the sources and transport media
involved in these pathways, and thus robust exposure information is generally not
available. A prime example is the use of surface water data for drinking water exposure calculations; this choice would be expected to be conservative, but if we are to
make rational decisions on action or inaction, complete data on real-world exposures should not be optional. The importance of intentional and unintentional water
reuse should also be evaluated in the context of exposure.
5. Assessment of potential for low-dose, chronic adverse human health outcomes
Due to the pseudopersistent nature of APIs in the environment, the possibility
of low-dose, chronic effects cannot be discounted. Most of the toxicological research
that has taken place on these compounds does not test for this possibility, and thus
hazard identification and setting of safety values may be an incomplete process. It
is quite possible that the conservative assumptions used in current practice for HHRA
in this arena are adequate to account for this possibility, but there is ample need for
understanding the potential for nontherapeutic effects of APIs.
6. Mixture effects, including anthropogenic and naturally occurring complex
chemicals
There are some indications that mixtures of APIs may cause changes in the function of
therapeutic API intake in humans. This possibility again highlights the needs for understanding low-dose effects and in the context of the complex mixture of APIs that might
be expected to occur. Virtually no effort has been expended thus far on the interaction
between APIs and nonpharmaceutical complex substances (anthropogenic and naturally occurring) that form the cocktail in which we live our lives. Further, the importance of environmental (i.e., through drinking water, surface water, or ingestion of fish)
exposures to an API which shares a MOA with a compound being taken therapeutically
has not yet been considered.
7. Development of antibiotic resistance in response to constant influx of antibiotics
Currently, the evidence on resistance selection in bacteria is mixed, but relatively
little work has been done in this area. The rise of new strains of multiresistant pathogens

216

E.S. Williams and B.W. Brooks

is certainly of interest to the public health community and thus more study of this
scenario is required.
Though currently there seems to be little cause for concern to human health with
regard to APIs in the environment of developed countries, copious research should
be devoted to deepening our knowledge, especially in the area of EDCs and genotoxic APIs. Further characterization of the environmental prevalence of APIs in
various compartments will also be an area of ongoing study. A more comprehensive
evaluation of human exposure to environmental APIs, perhaps through biomonitoring, would be useful.

References
1. NCHS (2008) Health, united states, 2008 with special feature on the health of young adults.
NCHS, Hyattsville
2. Zuccato E, Castiglioni S, Fanelli R, Reitano G, Bagnati R, Chiabrando C, Pomati F, Rossetti
C, Calamari D (2006) Pharmaceuticals in the environment in Italy: causes, occurrence, effects
and control. Environ Sci Pollut Res Int 13:1521
3. Kumar A, Chang B, Xagoraraki I (2010) Human health risk assessment of pharmaceuticals
in water: issues and challenges ahead. Int J Environ Res Public Health 7:39293953
4. Fent K, Weston AA, Caminada D (2006) Ecotoxicology of human pharmaceuticals. Aquat
Toxicol 76:122159
5. Carlsson C, Johansson AK, Alvan G, Bergman K, Kuhler T (2006) Are pharmaceuticals
potent environmental pollutants? Part I: environmental risk assessments of selected active
pharmaceutical ingredients. Sci Total Environ 364:6787
6. Stuer-Lauridsen F, Birkved M, Hansen LP, Lutzhoft HCH, Halling-Sorensen B (2000)
Environmental risk assessment of human pharmaceuticals in Denmark after normal therapeutic use. Chemosphere 40:783793
7. Doerr-MacEwen NA, Haight ME (2006) Expert stakeholders views on the management of
human pharmaceuticals in the environment. Environ Manage 38:853866
8. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton HT (2002)
Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. Streams,
19992000: a national reconnaissance. Environ Sci Technol 36:12021211
9. Colborn T, vom Saal FS FS, Soto AM (1993) Developmental effects of endocrine-disrupting
chemicals in wildlife and humans. Environ Health Perspect 101:378384
10. Safe S (2004) Endocrine disruptors and human health: is there a problem. Toxicology
205:310
11. Kime DE, Nash JP (1999) Gamete viability as an indicator of reproductive endocrine disruption in fish. Sci Total Environ 233:123129
12. Jobling S, Beresford N, Nolan M, Rodgers-Gray T, Brighty GC, Sumpter JP, Tyler CR (2002)
Altered sexual maturation and gamete production in wild roach (Rutilus rutilus) living in
rivers that receive treated sewage effluents. Biol Reprod 66:272281
13. Oaks JL, Gilbert M, Virani MZ, Watson RT, Meteyer CU, Rideout BA, Shivaprasad HL,
Ahmed S, Chaudhry MJ, Arshad M, Mahmood S, Ali A, Khan AA (2004) Diclofenac residues as the cause of vulture population decline in pakistan. Nature 427:630633
14. Jones OAH, Voulvoulis N, Lester JN (2002) Aquatic environmental assessment of the top 25
English prescription pharmaceuticals. Water Res 36:50135022
15. Ferrari B, Mons R, Vollat B, Fraysse B, Paxeus N, Lo Giudice R, Pollio A, Garric J (2004)
Environmental risk assessment of six human pharmaceuticals: Are the current environmental

Human Health Risk Assessment for Pharmaceuticals in the Environment

16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.

217

risk assessment procedures sufficient for the protection of the aquatic environment? Environ
Toxicol Chem 23:13441354
Workgroup OOEC (2008) White paper: aquatic life criteria for contaminants of emerging
concern, Part I: general challenges and recommendations. United States Environmental
Protection Agency, Washington, DC
Bercu JP, Parke NJ, Fiori JM, Meyerhoff RD (2008) Human health risk assessments for three
neuropharmaceutical compounds in surface waters. Regul Toxicol Pharm 50:420427
Kummerer K, Al-Ahmad A (2010) Estimation of the cancer risk to humans resulting from the
presence of cyclophosphamide and ifosfamide in surface water. Environ Sci Pollut Res Int
17:486496
Emmanuel E, Pierre MG, Perrodin Y (2009) Groundwater contamination by microbiological
and chemical substances released from hospital wastewater: health risk assessment for drinking water consumers. Environ Int 35:718726
Schulman LJ, Sargent EV, Naumann BD, Faria EC, Dolan DG, Wargo JP (2002) A human
health risk assessment of pharmaceuticals in the aquatic environment. Hum Ecol Risk Assess
8:657680
Cunningham VL, Binks SP, Olson MJ (2009) Human health risk assessment from the presence of human pharmaceuticals in the aquatic environment. Regul Toxicol Pharmacol
53:3945
Cunningham VL, Perino C, DAco VJ, Hartmann A, Bechter R (2010) Human health risk
assessment of carbamazepine in surface waters of north America and Europe. Regul Toxicol
Pharmacol 56:343351
Schwab BW, Hayes EP, Fiori JM, Mastrocco FJ, Roden NM, Cragin D, Meyerhoff RD,
DAco VJ, Anderson PD (2005) Human pharmaceuticals in US surface waters: a human
health risk assessment. Regul Toxicol Pharmacol 42:296312
Johnson AC, Jurgens MD, Williams RJ, Kummerer K, Kortenkamp A, Sumpter JP (2008) Do
cytotoxic chemotherapy drugs discharged into rivers pose a risk to the environment and
human health? An overview and UK case study. J Hydrol 348:167175
Christensen FM (1998) Pharmaceuticals in the environmenta human risk? Regul Toxicol
Pharmacol 28:212221
Kumar A, Xagoraraki I (2010) Human health risk assessment of pharmaceuticals in water: an
uncertainty analysis for meprobamate, carbamazepine, and phenytoin. Regul Toxicol
Pharmacol 57:146156
Rowney NC, Johnson AC, Williams RJ (2009) Cytotoxic drugs in drinking water: a prediction and risk assessment exercise for the Thames catchment in the United Kingdom. Environ
Toxicol Chem 28:27332743
Schriks M, Heringa MB, van der Kooi MM, de Voogt P, van Wezel AP (2010) Toxicological
relevance of emerging contaminants for drinking water quality. Water Res 44:461476
Snyder SA (2008) Occurrence, treatment, and toxicological relevance of EDCS and pharmaceuticals in water. Ozone Sci Eng 30:6569
Crane C, Maycock D, Crane D, Fawell J, Goslan E (2007) Desk based review of current knowledge on pharmaceuticals in drinking water and estimation of potential levels. Drinking Water
Inspectorate http://dwi.defra.gov.uk/research/completed-research/reports/dwi70-2-213.pdf
Webb S, Ternes T, Gibert M, Olejniczak K (2003) Indirect human exposure to pharmaceuticals via drinking water. Toxicol Lett 142:157167
Boxall AB, Kolpin DW, Halling-Sorensen B, Tolls J (2003) Are veterinary medicines causing
environmental risks? Environ Sci Technol 37:286A294A
Daughton CG, Ternes TA (1999) Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspect 107(suppl 6):907938
Kummerer K (2009) The presence of pharmaceuticals in the environment due to human
usepresent knowledge and future challenges. J Environ Manage 90:23542366
Hursthouse A, Kowalczyk G (2009) Transport and dynamics of toxic pollutants in the natural
environment and their effect on human health: research gaps and challenge. Environ Geochem
Health 31:165187

218

E.S. Williams and B.W. Brooks

36. Focazio MJ, Kolpin DW, Barnes KK, Furlong ET, Meyer MT, Zaugg SD, Barber LB,
Thurman ME (2008) A national reconnaissance for pharmaceuticals and other organic wastewater contaminants in the united statesII untreated drinking water sources. Sci Total
Environ 402:201216
37. Erickson BE (2002) Analyzing the ignored environmental contaminants. Environ Sci Technol
36:140A145A
38. Loraine GA, Pettigrove ME (2006) Seasonal variations in concentrations of pharmaceuticals
and personal care products in drinking water and reclaimed wastewater in Southern California.
Environ Sci Technol 40:687695
39. Zuccato E, Castiglioni S (2009) Illicit drugs in the environment. Philos Transact A Math Phys
Eng Sci 367:39653978
40. Bartelt-Hunt SL, Snow DD, Damon T, Shockley J, Hoagland K (2009) The occurrence of
illicit and therapeutic pharmaceuticals in wastewater effluent and surface waters in Nebraska.
Environ Pollut 157:786791
41. Petrovic M, de Alda MJ, Diaz-Cruz S, Postigo C, Radjenovic J, Gros M, Barcelo D (2009)
Fate and removal of pharmaceuticals and illicit drugs in conventional and membrane bioreactor wastewater treatment plants and by riverbank filtration. Philos Transact A Math Phys Eng
Sci 367:39794003
42. Zuccato E, Castiglioni S, Fanelli R (2005) Identification of the pharmaceuticals for human
use contaminating the Italian aquatic environment. J Hazard Mater 122:205209
43. Kasprzyk-Hordern B, Dinsdale RM, Guwy AJ (2009) Illicit drugs and pharmaceuticals in the
environmentforensic applications of environmental data. Part 1: estimation of the usage of
drugs in local communities. Environ Pollut 157:17731777
44. Kasprzyk-Hordern B, Dinsdale RM, Guwy AJ (2009) Illicit drugs and pharmaceuticals in the
environmentforensic applications of environmental data, Part 2: pharmaceuticals as chemical markers of faecal water contamination. Environ Pollut 157:17781786
45. Nikolaou A, Meric S, Fatta D (2007) Occurrence patterns of pharmaceuticals in water and
wastewater environments. Anal Bioanal Chem 387:12251234
46. Zuccato E, Calamari D, Natangelo M, Fanelli R (2000) Presence of therapeutic drugs in the
environment. Lancet 355:17891790
47. Ternes TA (1998) Occurrence of drugs in German sewage treatment plants and rivers. Water
Res 32:32453260
48. Hirsch R, Ternes T, Haberer K, Kratz KL (1999) Occurrence of antibiotics in the aquatic
environment. Sci Total Environ 225:109118
49. Coetsier CM, Spinelli S, Lin L, Roig B, Touraud E (2009) Discharge of pharmaceutical
products (PPS) through a conventional biological sewage treatment plant: MECS vs PECS?
Environ Int 35:787792
50. Comoretto L, Chiron S (2005) Comparing pharmaceutical and pesticide loads into a small
Mediterranean river. Sci Total Environ 349:201210
51. Andreozzi R, Raffaele M, Nicklas P (2003) Pharmaceuticals in STP effluents and their solar
photodegradation in aquatic environment. Chemosphere 50:13191330
52. Ferrari B, Paxeus N, Lo Giudice R, Pollio A, Garric J (2003) Ecotoxicological impact of
pharmaceuticals found in treated wastewaters: study of carbamazepine, clofibric acid, and
diclofenac. Ecotoxicol Environ Saf 55:359370
53. Holm JV, Rugge K, Bjerg PL, Christensen TH (1995) Occurrence and distribution of pharmaceutical organic compounds in the groundwater downgradient of a landfill (Grindsted,
Denmark). Environ Sci Technol 29:14151420
54. Braga O, Smythe GA, Schafer AI, Feitz AJ (2005) Fate of steroid estrogens in Australian
inland and coastal wastewater treatment plants. Environ Sci Technol 39:33513358
55. Falconer IR, Chapman HF, Moore MR, Ranmuthugala G (2006) Endocrine-disrupting compounds: a review of their challenge to sustainable and safe water supply and water reuse.
Environ Toxicol 21:181191
56. Ort C, Lawrence MG, Reungoat J, Eaglesham G, Carter S, Keller J (2010) Determining the
fraction of pharmaceutical residues in wastewater originating from a hospital. Water Res
44:605615

Human Health Risk Assessment for Pharmaceuticals in the Environment

219

57. Fick J, Soderstrom H, Lindberg RH, Phan C, Tysklind M, Larsson DG (2009) Contamination
of surface, ground, and drinking water from pharmaceutical production. Environ Toxicol
Chem 28:25222527
58. Larsson DG, de Pedro C, Paxeus N (2007) Effluent from drug manufactures contains
extremely high levels of pharmaceuticals. J Hazard Mater 148:751755
59. Stumpf M, Ternes TA, Wilken RD, Rodrigues SV, Baumann W (1999) Polar drug residues in
sewage and natural waters in the state of Rio de Janeiro, Brazil. Sci Total Environ 225:135141
60. Han GH, Hur HG, Kim SD (2006) Ecotoxicological risk of pharmaceuticals from wastewater
treatment plants in Korea: occurrence and toxicity to daphnia magna. Environ Toxicol Chem
25:265271
61. Takahashi A, Higashitani T, Yakou Y, Saitou M, Tamamoto H, Tanaka H (2003) Evaluating
bioaccumulation of suspected endocrine disruptors into periphytons and benthos in the Tama
river. Water Sci Technol 47:7176
62. Nakada N, Tanishima T, Shinohara H, Kiri K, Takada H (2006) Pharmaceutical chemicals
and endocrine disrupters in municipal wastewater in Tokyo and their removal during activated sludge treatment. Water Res 40:32973303
63. Nakada N, Kiri K, Shinohara H, Harada A, Kuroda K, Takizawa S, Takada H (2008)
Evaluation of pharmaceuticals and personal care products as water-soluble molecular markers of sewage. Environ Sci Technol 42:63476353
64. Cui CW, Ji SL, Ren HY (2006) Determination of steroid estrogens in wastewater treatment
plant of a controceptives producing factory. Environ Monit Assess 121:409419
65. Richardson BJ, Lam PK, Martin M (2005) Emerging chemicals of concern: pharmaceuticals
and personal care products (PPCPs) in Asia, with particular reference to Southern China. Mar
Pollut Bull 50:913920
66. Managaki S, Murata A, Takada H, Tuyen BC, Chiem NH (2007) Distribution of macrolides,
sulfonamides, and trimethoprim in tropical waters: ubiquitous occurrence of veterinary antibiotics in the Mekong delta. Environ Sci Technol 41:80048010
67. Lin AY, Tsai YT (2009) Occurrence of pharmaceuticals in Taiwans surface waters: impact of
waste streams from hospitals and pharmaceutical production facilities. Sci Total Environ
407:37933802
68. Bound JP, Voulvoulis N (2005) Household disposal of pharmaceuticals as a pathway for
aquatic contamination in the United Kingdom. Environ Health Perspect 113:17051711
69. Joss A, Keller E, Alder AC, Gobel A, McArdell CS, Ternes T, Siegrist H (2005) Removal
of pharmaceuticals and fragrances in biological wastewater treatment. Water Res 39:3
1393152
70. Winker M, Faika D, Gulyas H, Otterpohl R (2008) A comparison of human pharmaceutical
concentrations in raw municipal wastewater and yellowwater. Sci Total Environ 399:96104
71. Bendz D, Paxeus NA, Ginn TR, Loge FJ (2005) Occurrence and fate of pharmaceutically
active compounds in the environment, a case study: Hoje river in Sweden. J Hazard Mater
122:195204
72. Jjemba PK (2006) Excretion and ecotoxicity of pharmaceutical and personal care products in
the environment. Ecotoxicol Environ Saf 63:113130
73. Mompelat S, Le Bot B, Thomas O (2009) Occurrence and fate of pharmaceutical products
and by-products, from resource to drinking water. Environ Int 35:803814
74. Claudel JP, Touboul P (1995) Sotalol: From just another beta blocker to the prototype of
Class III antidysrhythmic compound. Pacing Clin Electrophysiol 18:451467
75. Lienert J, Gudel K, Escher BI (2007) Screening method for ecotoxicological hazard assessment of 42 pharmaceuticals considering human metabolism and excretory routes. Environ
Sci Technol 41:44714478
76. Alder AC, Schaffner C, Majewsky M, Klasmeier J, Fenner K (2010) Fate of beta-blocker
human pharmaceuticals in surface water: comparison of measured and simulated concentrations in the Glatt Valley watershed, Switzerland. Water Res 44:936948
77. Kasprzyk-Hordern B, Dinsdale RM, Guwy AJ (2008) The occurrence of pharmaceuticals,
personal care products, endocrine disruptors and illicit drugs in surface water in South Wales,
UK. Water Res 42:34983518

220

E.S. Williams and B.W. Brooks

78. Heberer T (2002) Occurrence, fate, and removal of pharmaceutical residues in the aquatic
environment: a review of recent research data. Toxicol Lett 131:517
79. Park GR (2001) Drug metabolism. Br J Anaesth CEPD Rev 1:185188
80. Daughton CG (2003) Cradle-to-cradle stewardship of drugs for minimizing their environmental disposition while promoting human health. II. Drug disposal, waste reduction, and
future directions. Environ Health Perspect 111:775785
81. Daughton CG (2009) Chemicals from the practice of healthcare: challenges and unknowns
posed by residues in the environment. Environ Toxicol Chem 28:24902494
82. Rouen D, Dolan K, Kimber J (2001) A review of drug detection testing and an examination
of urine, hair, saliva, and sweat. Vol Technical Report No. 120. National Drug and Alcohol
Research Centre, Sydney, Australia
83. Daughton CG, Ruhoy IS (2009) Environmental footprint of pharmaceuticals: the significance
of factors beyond direct excretion to sewers. Environ Toxicol Chem 28:24952521
84. Bound JP, Voulvoulis N (2006) Predicted and measured concentrations for selected pharmaceuticals in UK rivers: implications for risk assessment. Water Res 40:28852892
85. Glassmeyer ST, Hinchey EK, Boehme SE, Daughton CG, Ruhoy IS, Conerly O, Daniels RL,
Lauer L, McCarthy M, Nettesheim TG, Sykes K, Thompson VG (2009) Disposal practices
for unwanted residential medications in the United States. Environ Int 35:566572
86. Kuspis DA, Krenzelok EP (1996) What happens to expired medications? A survey of community medication disposal. Vet Hum Toxicol 38:4849
87. Adler NE, Koschorreck J, Rechenberg B (2008) Environmental impact assessment and control of pharmaceuticals: the role of environmental agencies. Water Sci Technol 57:9197
88. Jones OA, Voulvoulis N, Lester JN (2003) Potential impact of pharmaceuticals on environmental health. Bull World Health Organ 81:768769
89. Joakim Larsson DG, Fick J (2009) Transparency throughout the production chaina way to
reduce pollution from the manufacturing of pharmaceuticals? Regul Toxicol Pharmacol
53:161163
90. Li D, Yang M, Hu J, Ren L, Zhang Y, Li K (2008) Determination and fate of oxytetracycline
and related compounds in oxytetracycline production wastewater and the receiving river.
Environ Toxicol Chem 27:8086
91. Langford KH, Thomas KV (2009) Determination of pharmaceutical compounds in hospital
effluents and their contribution to wastewater treatment works. Environ Int 35:766770
92. Castiglioni S, Bagnati R, Fanelli R, Pomati F, Calamari D, Zuccato E (2006) Removal of
pharmaceuticals in sewage treatment plants in Italy. Environ Sci Technol 40:357363
93. Choi K, Kim Y, Park J, Park CK, Kim M, Kim HS, Kim P (2008) Seasonal variations of
several pharmaceutical residues in surface water and sewage treatment plants of Han River,
Korea. Sci Total Environ 405:120128
94. Brun GL, Bernier M, Losier R, Doe K, Jackman P, Lee HB (2006) Pharmaceutically active
compounds in Atlantic Canadian sewage treatment plant effluents and receiving waters, and
potential for environmental effects as measured by acute and chronic aquatic toxicity. Environ
Toxicol Chem 25:21632176
95. Conley JM, Symes SJ, Schorr MS, Richards SM (2008) Spatial and temporal analysis of
pharmaceutical concentrations in the upper Tennessee River Basin. Chemosphere 73:
11781187
96. Vieno NM, Tuhkanen T, Kronberg L (2005) Seasonal variation in the occurrence of pharmaceuticals in effluents from a sewage treatment plant and in the recipient water. Environ Sci
Technol 39:82208226
97. Gros M, Petrovic M, Barcelo D (2007) Wastewater treatment plants as a pathway for aquatic
contamination by pharmaceuticals in the Ebro river basin (Northeast Spain). Environ Toxicol
Chem 26:15531562
98. Standley LJ, Rudel RA, Swartz CH, Attfield KR, Christian J, Erickson M, Brody JG (2008)
Wastewater-contaminated groundwater as a source of endogenous hormones and pharmaceuticals to surface water ecosystems. Environ Toxicol Chem 27:24572468

Human Health Risk Assessment for Pharmaceuticals in the Environment

221

99. Lapen DR, Topp E, Metcalfe CD, Li H, Edwards M, Gottschall N, Bolton P, Curnoe W, Payne
M, Beck A (2008) Pharmaceutical and personal care products in tile drainage following land
application of municipal biosolids. Sci Total Environ 399:5065
100. Topp E, Monteiro SC, Beck A, Coelho BB, Boxall AB, Duenk PW, Kleywegt S, Lapen DR,
Payne M, Sabourin L, Li H, Metcalfe CD (2008) Runoff of pharmaceuticals and personal
care products following application of biosolids to an agricultural field. Sci Total Environ
396:5259
101. Boxall AB, Blackwell P, Cavallo R, Kay P, Tolls J (2002) The sorption and transport of a
sulphonamide antibiotic in soil systems. Toxicol Lett 131:1928
102. Radjenovic J, Petrovic M, Barcelo D (2009) Fate and distribution of pharmaceuticals in
wastewater and sewage sludge of the conventional activated sludge (CAS) and advanced
membrane bioreactor (MBR) treatment. Water Res 43:831841
103. Spongberg AL, Witter JD (2008) Pharmaceutical compounds in the wastewater process
stream in Northwest Ohio. Sci Total Environ 397:148157
104. Xia K, Bhandari A, Das K, Pillar G (2005) Occurrence and fate of pharmaceuticals and personal care products (PPCPs) in biosolids. J Environ Qual 34:91104
105. Monteiro SC, Boxall AB (2009) Factors affecting the degradation of pharmaceuticals in agricultural soils. Environ Toxicol Chem 28:25462554
106. Slack RJ, Gronow JR, Voulvoulis N (2005) Household hazardous waste in municipal landfills:
contaminants in leachate. Sci Total Environ 337:119137
107. Barnes KK, Christenson SC, Kolpin DW, Focazio M, Furlong ET, Zaugg SD, Meyer MT, Barber
LB (2004) Pharmaceuticals and other organic waste water contaminants within a leachate plume
downgradient of a municipal landfill. Ground Water Monit Remed 24:119126
108. Schwarzbauer J, Heim S, Brinker S, Littke R (2002) Occurrence and alteration of organic
contaminants in seepage and leakage water from a waste deposit landfill. Water Res
36:22752287
109. Breton R, Boxall A (2003) Pharmaceuticals and personal care products in the environment:
regulatory drivers and research needs. QSAR Comb Sci 22:399409
110. Hamscher G, Sczesny S, Hoper H, Nau H (2002) Determination of persistent tetracycline
residues in soil fertilized with liquid manure by high-performance liquid chromatography
with electrospray ionization tandem mass spectrometry. Anal Chem 74:15091518
111. Boxall AB, Fogg LA, Blackwell PA, Kay P, Pemberton EJ, Croxford A (2004) Veterinary
medicines in the environment. Rev Environ Contam Toxicol 180:191
112. Khetan SK, Collins TJ (2007) Human pharmaceuticals in the aquatic environment: a challenge to green chemistry. Chem Rev 107:23192364
113. Loffler D, Rombke J, Meller M, Ternes TA (2005) Environmental fate of pharmaceuticals in
water/sediment systems. Environ Sci Technol 39:52095218
114. Latch DE, Stender BL, Packer JL, Arnold WA, McNeill K (2003) Photochemical fate of
pharmaceuticals in the environment: cimetidine and ranitidine. Environ Sci Technol
37:33423350
115. Lin AY, Reinhard M (2005) Photodegradation of common environmental pharmaceuticals
and estrogens in river water. Environ Toxicol Chem 24:13031309
116. Canonica S, Meunier L, von Gunten U (2008) Phototransformation of selected pharmaceuticals during UV treatment of drinking water. Water Res 42:121128
117. Kim I, Tanaka H (2009) Photodegradation characteristics of PPCPs in water with UV treatment. Environ Int 35:793802
118. Packer JL, Werner JJ, Latch DE, McNeill K, Arnold WA (2003) Photochemical fate of pharmaceuticals in the environment: naproxen, diclofenac, clofibric acid, and ibuprofen. Aquat
Sci 65:342351
119. Benotti MJ, Brownawell BJ (2007) Distributions of pharmaceuticals in an urban estuary during both dry- and wet-weather conditions. Environ Sci Technol 41:57955802
120. Zorita S, Martensson L, Mathiasson L (2009) Occurrence and removal of pharmaceuticals in a
municipal sewage treatment system in the south of Sweden. Sci Total Environ 407:27602770

222

E.S. Williams and B.W. Brooks

121. Baumgarten S, Schroder HF, Charwath C, Lange M, Beier S, Pinnekamp J (2007) Evaluation
of advanced treatment technologies for the elimination of pharmaceutical compounds. Water
Sci Technol 56:18
122. Stackelberg PE, Gibs J, Furlong ET, Meyer MT, Zaugg SD, Lippincott RL (2007) Efficiency
of conventional drinking-water-treatment processes in removal of pharmaceuticals and other
organic compounds. Sci Total Environ 377:255272
123. Vieno NM, Harkki H, Tuhkanen T, Kronberg L (2007) Occurrence of pharmaceuticals in
river water and their elimination in a pilot-scale drinking water treatment plant. Environ Sci
Technol 41:50775084
124. Huber MM, Gobel A, Joss A, Hermann N, Loffler D, McArdell CS, Ried A, Siegrist H,
Ternes TA, von Gunten U (2005) Oxidation of pharmaceuticals during ozonation of municipal wastewater effluents: a pilot study. Environ Sci Technol 39:42904299
125. Ternes TA, Stuber J, Herrmann N, McDowell D, Ried A, Kampmann M, Teiser B (2003)
Ozonation: a tool for removal of pharmaceuticals, contrast media and musk fragrances from
wastewater? Water Res 37:19761982
126. Paraskeva P, Graham NJ (2002) Ozonation of municipal wastewater effluents. Water Environ
Res 74:569581
127. Ternes TA (2001) Pharmaceuticals and metabolites as contaminants of the aquatic environment. In: Daughton CG, Jones-Lepp TL (eds) Pharmaceuticals and personal care products in
the environment - scientific and regulatory issues, ACS Symposium series 791. American
Chemical Society, Washington, DC, pp 3954
128. Ternes TA (2001) Analytical methods for the determination of pharmaceuticals in aqueous
environmental samples. Trac Trend Anal Chem 20:419434
129. Jones OA, Lester JN, Voulvoulis N (2005) Pharmaceuticals: a threat to drinking water?
Trends Biotechnol 23:163167
130. Benotti MJ, Trenholm RA, Vanderford BJ, Holady JC, Stanford BD, Snyder SA (2009)
Pharmaceuticals and endocrine disrupting compounds in U.S. Drinking water. Environ Sci
Technol 43:597603
131. Rabiet M, Togola A, Brissaud F, Seidel JL, Budzinski H, Elbaz-Poulichet F (2006) Consequences
of treated water recycling as regards pharmaceuticals and drugs in surface and ground waters of
a medium-sized Mediterranean catchment. Environ Sci Technol 40:52825288
132. Anderson PD, DAco VJ, Shanahan P, Chapra SC, Buzby ME, Cunningham VL, Duplessie
BM, Hayes EP, Mastrocco FJ, Parke NJ, Rader JC, Samuelian JH, Schwab BW (2004)
Screening analysis of human pharmaceutical compounds in U.S. Surface waters. Environ Sci
Technol 38:838849
133. Feijtel T, Boeije G, Matthies M, Young A, Morris G, Gandolfi C, Hansen B, Fox K, Holt M,
Koch V, Schroder R, Cassani G, Schowanek D, Rosenblom J, Niessen H (1997) Development
of a geography-referenced regional exposure assessment tool for European riversGREATER contribution to GREAT-ER #1. Chemosphere 34:23512373
134. Brooks BW, Chambliss CK, Stanley JK, Ramirez A, Banks KE, Johnson RD, Lewis RJ
(2005) Determination of select antidepressants in fish from an effluent-dominated stream.
Environ Toxicol Chem 24:464469
135. Brown JN, Paxeus N, Forlin L, Larsson DGJ (2007) Variations in bioconcentration of human
pharmaceuticals from sewage effluents into fish blood plasma. Environ Toxicol Pharmacol
24:267274
136. Chu S, Metcalfe CD (2007) Analysis of paroxetine, fluoxetine and norfluoxetine in fish tissues using pressurized liquid extraction, mixed mode solid phase extraction cleanup and liquid chromatography-tandem mass spectrometry. J Chromatogr A 1163:112118
137. Nakamura Y, Yamamoto H, Sekizawa J, Kondo T, Hirai N, Tatarazako N (2008) The effects
of ph on fluoxetine in Japanese Medaka (Oryzias latipes): acute toxicity in fish larvae and
bioaccumulation in juvenile fish. Chemosphere 70:865873
138. Ramirez AJ, Mottaleb MA, Brooks BW, Chambliss CK (2007) Analysis of pharmaceuticals
in fish using liquid chromatography-tandem mass spectrometry. Anal Chem 79:31553163
139. Boxall AB, Johnson P, Smith EJ, Sinclair CJ, Stutt E, Levy LS (2006) Uptake of veterinary
medicines from soils into plants. J Agric Food Chem 54:22882297

Human Health Risk Assessment for Pharmaceuticals in the Environment

223

140. Jones OA, Voulvoulis N, Lester JN (2004) Potential ecological and human health risks associated with the presence of pharmaceutically active compounds in the aquatic environment.
Crit Rev Toxicol 34:335350
141. Herrman JL, Younes M (1999) Background to the adi/tdi/ptwi. Regul Toxicol Pharmacol
30:S109S113
142. Galli CL, Marinovich M, Lotti M (2008) Is the acceptable daily intake as presently used an
axiom or a dogma? Toxicol Lett 180:9399
143. Kroes R, Galli C, Munro I, Schilter B, Tran L, Walker R, Wurtzen G (2000) Threshold of
toxicological concern for chemical substances present in the diet: a practical tool for assessing the need for toxicity testing. Food Chem Toxicol 38:255312
144. Kim S, Aga DS (2007) Potential ecological and human health impacts of antibiotics and
antibiotic-resistant bacteria from wastewater treatment plants. J Toxicol Environ Health B
Crit Rev 10:559573
145. Jorgensen SE, Halling-Sorensen B (2000) Drugs in the environment. Chemosphere 40:691699
146. Schwartz T, Kohnen W, Jansen B, Obst U (2003) Detection of antibiotic-resistant bacteria
and their resistance genes in wastewater, surface water, and drinking water biofilms. FEMS
Microbiol Ecol 43:325335
147. Guardabassi L, Petersen A, Olsen JE, Dalsgaard A (1998) Antibiotic resistance in
Acinetobacter spp. isolated from sewers receiving waste effluent from a hospital and a pharmaceutical plant. Appl Environ Microbiol 64:34993502
148. Reinthaler FF, Posch J, Feierl G, Wust G, Haas D, Ruckenbauer G, Mascher F, Marth E
(2003) Antibiotic resistance of E. coli in sewage and sludge. Water Res 37:16851690
149. Goni-Urriza M, Capdepuy M, Arpin C, Raymond N, Caumette P, Quentin C (2000) Impact
of an urban effluent on antibiotic resistance of riverine Enterobacteriaceae and Aeromonas
spp. Appl Environ Microbiol 66:125132
150. Kummerer K (2009) Antibiotics in the aquatic environmenta reviewpart II. Chemosphere
75:435441
151. Dorne JL, Skinner L, Frampton GK, Spurgeon DJ, Ragas AM (2007) Human and environmental risk assessment of pharmaceuticals: differences, similarities, lessons from toxicology.
Anal Bioanal Chem 387:12591268
152. Zhang Z, Feng Y, Gao P, Wang C, Ren N (2011) Occurrence and removal efficiencies of eight
EDCS and estrogenicity in a STP. J Environ Monit 13:13331373
153. Kusk KO, Kruger T, Long M, Taxvig C, Lykkesfeldt AE, Frederiksen H, Andersson AM,
Andersen HR, Hansen KM, Nellemann C, Bonefeld-Jorgensen EC (2011) Endocrine potency
of wastewater: contents of endocrine disrupting chemicals and effects measured by in vivo
and in vitro assays. Environ Toxicol Chem 30:413426
154. Swart JC, Pool EJ, van Wyk JH (2011) The implementation of a battery of in vivo and in vitro
bioassays to assess river water for estrogenic endocrine disrupting chemicals. Ecotoxicol
Environ Saf 74:138143
155. Lissemore L, Hao C, Yang P, Sibley PK, Mabury S, Solomon KR (2006) An exposure assessment for selected pharmaceuticals within a watershed in Southern Ontario. Chemosphere
64:717729
156. Cleuvers M (2003) Aquatic ecotoxicity of pharmaceuticals including the assessment of
combination effects. Toxicol Lett 142:185194
157. Richards SM, Wilson CJ, Johnson DJ, Castle DM, Lam M, Mabury SA, Sibley PK, Solomon
KR (2004) Effects of pharmaceutical mixtures in aquatic microcosms. Environ Toxicol Chem
23:10351042
158. Pomati F (2007) Pharmaceuticals in drinking water: is the cure worse than the disease?
Environ Sci Technol 41:8204
159. Gurr CJ, Reinhard M (2006) Harnessing natural attenuation of pharmaceuticals and hormones
in rivers. Environ Sci Technol 40:28722876
160. Ashton D, Hilton M, Thomas KV (2004) Investigating the environmental transport of human
pharmaceuticals to streams in the united kingdom. Sci Total Environ 333:167184
161. Henschel KP, Wenzel A, Diedrich M, Fliedner A (1997) Environmental hazard assessment of
pharmaceuticals. Regul Toxicol Pharmacol 25:220225

224

E.S. Williams and B.W. Brooks

162. Halling-Sorensen B, Nors Nielsen S, Lanzky PF, Ingerslev F, Holten Lutzhoft HC, Jorgensen
SE (1998) Occurrence, fate and effects of pharmaceutical substances in the environmenta
review. Chemosphere 36:357393
163. Bound JP, Voulvoulis N (2004) Pharmaceuticals in the aquatic environmenta comparison
of risk assessment strategies. Chemosphere 56:11431155
164. Schowanek D, Webb S (2002) Exposure simulation for pharmaceuticals in European surface
waters with greater. Toxicol Lett 131:3950
165. Boeije GM, Wagner JO, Koormann F, Vanrolleghem PA, Schowanek DR, Feijtel TC (2000)
New PEC definitions for river basins applicable to GIS-based environmental exposure assessment. Chemosphere 40:255265
166. Sanderson H, Johnson DJ, Reitsma T, Brain RA, Wilson CJ, Solomon KR (2004) Ranking
and prioritization of environmental risks of pharmaceuticals in surface waters. Regul Toxicol
Pharmacol 39:158183
167. Ruhoy IS, Daughton CG (2008) Beyond the medicine cabinet: an analysis of where and why
medications accumulate. Environ Int 34:11571169
168. Brain RA, Sanderson H, Sibley PK, Solomon KR (2006) Probabilistic ecological hazard
assessment: evaluating pharmaceutical effects on aquatic higher plants as an example.
Ecotoxicol Environ Saf 64:128135
169. Sanderson H, Johnson DJ, Wilson CJ, Brain RA, Solomon KR (2003) Probabilistic hazard
assessment of environmentally occurring pharmaceuticals toxicity to fish, daphnids and algae
by ecosar screening. Toxicol Lett 144:383395
170. Cunningham VL, Buzby M, Hutchinson T, Mastrocco F, Parke N, Roden N (2006) Effects of
human pharmaceuticals on aquatic life: next steps. Environ Sci Technol 40:34563462
171. Philips PJ, Smith SG, Kolpin DW, Zaugg SD, Buxton HT, Furlong ET, Esposito K, Stinson B
(2010) Pharmaceutical formulation facilities as sources of opioids and other pharmaceuticals
to wastewater treatment plant effluents. Environ Sci Technol 44(13):491016
172. Brooks BW, Riley TM, Taylor RD (2006) Water quality of effluent-dominated ecosystems:
ecotoxicological, hydrological, and management considerations. Hydrobiologica 556:36579
173. Brooks BW, Huggett DB, Boxall AB (2009) Pharmaceuticals and personal care products:
Research needs for the next decade. Environ Toxicol Chem 28(12):246972
174. Monteiro SC, Boxall AB (2010) Occurrence and fate of human pharmaceuticals in the
environment. Rev Environ Contam Toxicol 202:53154
175. Daughton CG, Brooks BW (2011) Active pharmaceutical ingredients and aquatic organisms.
In: Environmental Contaminants in Wildlife: Interpreting Tissue Concentrations, 2nd ed.
Eds: Meador J, Beyer N. Taylor and Francis. p. 281341
176. Valenti TV, Gould GG, Berninger JP, Connors KA, Keele NB, ProsserKN, Brooks BW (2012)
Human therapeutic plasma levels of the selective serotonin reuptake inhibitor (SSRI) sertraline decrease serotonin reuptake transporter binding and shelter seeking behavior in adult
male fathead minnows. Environ Sci Technol 46:242735

Wastewater and Drinking Water Treatment

Technologies
Daniel Gerrity and Shane Snyder

Abbreviations
AOP
BAC
CAS
CDPH
DBP
DOW
EDC
EEO
GAC
IPR
KOW
MBR
MF
MRL
MW

Advanced oxidation process

Biological activated carbon
Conventional activated sludge
California Department of Public Health
Disinfection by-product
Octanol/water distribution coefficient
Endocrine disrupting compound
Electrical energy per order of magnitude destruction
Granular activated carbon
Indirect potable reuse
Octanol/water partitioning coefficient
Membrane bioreactor
Microfiltration
Method reporting limit
Molecular weight

D. Gerrity
Water Quality Research and Development Center, Southern Nevada Water Authority,
River Mountain Water Treatment Facility, 1299 Burkholder Boulevard, Henderson,
NV 89015, USA
e-mail: [email protected]
S. Snyder (*)
Chemical and Environmental Engineering, University of Arizona, 1133 E. James E. Rogers Way,
P.O. Box 210011, Tucson, AZ 85721, USA
e-mail: [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_9, Springer Science+Business Media, LLC 2012

225

226

NF
NPDES
PAC
PPCPs
RO
S
SRT
TOrC
UF

D. Gerrity and S. Snyder

Nanofiltration
National Pollutant Discharge Elimination System
Powder activated carbon
Pharmaceuticals and personal care products
Reverse osmosis
Solubility
Solids retention time
Trace organic contaminant
Ultrafiltration

Introduction
Although pharmaceuticals and personal care products (PPCPs) and endocrine
disrupting compounds (EDCs) are often considered emerging contaminants,
researchers have been aware of their ubiquity in water for decades. As early as the
1940s, scientists were aware that certain chemicals had the ability to mimic endogenous estrogens and androgens [1, 2], and in 1965, Stumm-Zollinger and Fair of
Harvard University published the first known report indicating that steroid hormones were not completely eliminated by wastewater treatment [3]. In 1977,
researchers from the University of Kansas published the first known report of pharmaceutical discharge from a wastewater treatment plant (WWTP) [4].
Despite these early findings, studies related to PPCPs and EDCs in source water,
drinking water, and wastewater did not become a mainstream research topic until
the late 1990s and early 2000s. Potential human health effects, demonstrated impacts
on aquatic ecosystems, and increased media coverage, which ultimately led to
increased public awareness, were primarily responsible for the spike in scientific
studies [5]. This was coupled with the development of extremely sensitive analytical methods that allowed researchers to approach parts-per-quadrillion (sub-ng L1)
detection limits for a variety of trace organic contaminants (TOrCs) [6, 7]. Each of
these factors increased the number and scope of scientific investigations into the
presence, fate, and transport of TOrCs in natural and engineered systems.
Although there are a number of significant sources of PPCPs and EDCs in the
environment, including industrial manufacturing processes and confined animal
feeding operations [8], municipal wastewater is considered the primary source [9].
The occurrence of these compounds, associated by-products, and transformation
products in wastewater results from their release during manufacturing, excretion
after personal use, and disposal of unused quantities [10]. In 1999, Daughton and
Ternes highlighted the ubiquity of pharmaceuticals, of which more than 3,000 are
now available by prescription [11], due to their direct correlation to human presence: pharmaceuticals will be detected in any water supply in proximity to human
populations [10]. In fact, the presence or absence of any chemical in wastewater
effluent is essentially a function of analytical detection capability. In a 2008 review
of TOrC occurrence in municipal wastewater effluent, Snyder et al. [8] identified

Wastewater and Drinking Water Treatment Technologies

227

pharmaceutical residues, antibiotics, steroid hormones, and fragrances as the most

frequently detected compound classes, and Ternes [12] provided one of the first
comprehensive evaluations of TOrC concentrations in municipal wastewater effluent
and receiving waters. Fent et al. [13] also provided a comprehensive review of TOrC
concentrations in wastewater effluent in addition to the modes of action and toxicological implications of those contaminants.
With respect to wastewater treatment, compound removal and transformation is
highly dependent on the unit processes (e.g., secondary treatment, filtration, and disinfection) and operational variables (e.g., solids retention time [SRT] and oxidant
dose) employed at a particular plant [5, 11]. Even at a single WWTP, effluent concentrations can be highly variable as they are influenced by temporal variations in influent
concentrations, temperature, and dry vs. wet weather flows [12]. Once these contaminants are discharged, natural attenuation occurs through microbial degradation, dilution, adsorption to solids, photolysis, or other forms of abiotic transformation.
However, these natural processes are generally insufficient to reduce TOrC concentrations to nondetect levels. Furthermore, some receiving bodies can be comprised of
5090% wastewater effluent during dry weather conditions [10]. This ultimately leads
to contamination of surface water, groundwater (i.e., after aquifer recharge or leaching
from landfilled solids), and even food supplies (i.e., after plant uptake from reclaimed
irrigation water) [10, 14]. Kolpin et al. [15] documented the extent of contamination
(with respect to 95 TOrCs) of 139 predominantly wastewater-impacted streams in the
USA. Although identified as a conservative estimate due to method limitations (i.e.,
method reporting limits [MRLs]), at least one TOrC was detected in 80% of the sample sites, but the concentrations were generally less than 1 mg L1. To highlight immediate impacts on drinking water supplies, Benotti et al. [11] monitored 51 TOrCs in
the source water, finished drinking water, and distribution systems of 19 US utilities.
Although median concentrations of the target pharmaceuticals rarely exceeded
10 ng L1, some TOrCs were detected at maximum concentrations exceeding
100 ng L1. The herbicide atrazine was even detected in systems with no known agricultural applications. Therefore, recalcitrant compounds certainly persist in drinking
water supplies and ultimately contaminate finished drinking water.
Water and wastewater treatment trains are generally not designed for the removal
of TOrCs. However, the interrelatedness of wastewater discharge and drinking water
sources and potential effects on aquatic ecosystems now justify some consideration
of TOrCs in the design process. In fact, expansion and optimization of wastewater
treatment processes may be the most efficient strategy to mitigate the potential
effects of these contaminants. Countless treatment processes have been evaluated
for their ability to remove or destroy TOrCs. These evaluations span the continua of
physicochemical treatment (e.g., media or membrane filtration), conventional oxidation (e.g., chlorine and ozone), and advanced oxidation processes (AOPs) (e.g.,
UV/H2O2) in drinking water and wastewater [1621]. This chapter discusses the
efficacy of the various treatment technologies available to water and WWTPs for
TOrC removal and/or destruction. It is important to note that the TOrCs included in
most studies in the literature satisfy the following four criteria: (1) high likelihood
of occurrence in the environment, (2) potential toxicological relevance and significant

228

D. Gerrity and S. Snyder

Table 1 Physicochemical properties of selected TOrCsa

Compounds
Classes
MW
S (mg L1)

Log KOW

pKa

Acetaminophen
Androstenedione
Atrazine
Caffeine
Carbamazepine
DEETb
Diazepam
Diclofenac
Dilantin
Erythromycin
Estriol
Estradiol
Estrone
Ethynyl estradiol
Fluoxetine
Galaxolide
Gemfibrozil
Hydrocodone
Ibuprofen
Iopromide
Meprobamate
Metolachlor
Musk ketone
Naproxen
Pentoxifylline
Progesterone
Sulfamethoxazole
TCEPc
Testosterone
Triclosan
Trimethoprim

0.46
2.75
2.61
0.07
2.45
2.18
2.82
4.51
2.47
3.06
2.45
4.01
3.13
3.67
4.05
5.9 [26]
4.33 (est)
2.16 (est)
3.97
2.05
0.7
3.13
4.3 [27]
3.18
0.29
3.87
0.89
1.44
3.32
4.76
0.91

9.38
N/A
1.7
10.4
13.9 [23]
0.7 (est)
3.4
4.15
8.33
8.88
9.85 (est)
10.4 [25]
10.4 [25]
10.4 [16]
10.3 (est)
N/A
4.42
8.35 (est)
4.91
10.2 (est)
10.9 (est)
N/A
N/A
4.15
1.49 (est)
N/A
5.5 [29]
N/A
N/A
7.9 [30]
7.12

Analgesic
Hormone
Herbicide
Psychoactive
Anticonvulsant
Insect repellant
Antianxiety
Analgesic
Anticonvulsant
Antibiotic
Hormone
Hormone
Hormone
Hormone
Psychoactive
Fragrance
Antilipidemic
Analgesic
Analgesic
X-ray contrast
Antianxiety
Pesticide
Fragrance
Analgesic
Vasodilator
Hormone
Antibiotic
Flame retardant
Hormone
Antimicrobial
Antibiotic

151.2
286.4
215.7
194.2
236.3
191.3
284.7
296.2
252.3
733.9
288.4
272.4
270.4
296.4
309.3
258.4
250.3
299.4
206.3
791.1
218.3
283.8
294.3
230.3
278.3
314.5
253.3
285.5
288.4
289.5
290.3

1.40E + 4
57.8
34.7
2.16E + 4
18 [22]
9.9 [24]
50
2.37
32
1.44 (est)
441 (est)
3.6
30
11.3
60.3 (est)
1.75 [26]
19 (est)
6,870 (est)
21
23.8 (est)
4,700
530
0.46 [27], 1.9 [28]
15.9
7.70E + 4
8.81
610
7,000
23.4
10
400

Experimental values from Environmental Science Database SRC PhysProp

Chemical name: N,N-diethyl-meta-toluamide
c
Chemical name: Tri(chloroethyl)phosphate
a

public interest, (3) structural diversity resulting in a range of treatability, and (4)
amenability to available analytical methods. The target pharmaceuticals often
encompass several therapeutic classes, including analgesics, antibiotics, anticonvulsants, psychoactive drugs, and cholesterol-lowering medications. A subset of the
target compounds discussed in this chapter in addition to their structural properties
(e.g., molecular weight [MW], solubility (S), and octanol/water partitioning
coefficient [KOW]) are summarized in Table 1.

Wastewater and Drinking Water Treatment Technologies

229

TOrC Occurrence in Water and Wastewater

TOrC concentrations in raw wastewater may routinely exceed 1 mg L1 for a variety of
compounds, particularly analgesics, some antibiotics, flame retardants, and caffeine.
Fortunately, conventional biological wastewater treatment processes are particularly
effective in removing compounds with high detection frequencies and concentrationswith flame retardants (e.g., TCEP), X-ray contrast media (e.g., iopromide),
some psychoactive drugs (e.g., meprobamate), and herbicides (e.g., atrazine) being
notable exceptions. TOrC concentrations in finished effluent are highly site specific
and dependent on the unit processes and operational conditions at a particular plant.
Therefore, it is difficult to identify typical wastewater concentrations, but concentrations from two US WWTPs are provided for context in Table 2.
Snyder et al. [18] and Benotti et al. [11] present a comprehensive reconnaissance
of pharmaceuticals and EDCs in US source and drinking water, though both studies
targeted systems susceptible to wastewater contamination. Snyder et al. [18] surveyed the occurrence of 36 pharmaceuticals and EDCs in the source and finished
drinking water of 20 US drinking water treatment plants (Table 3). The 12 compounds
that were detected in at least half of the source water samples were atrazine, caffeine, carbamazepine, DEET, gemfibrozil, ibuprofen, iopromide, meprobamate,
naproxen, phenytoin, sulfamethoxazole, and TCEP. Median concentrations of
detected pharmaceuticals and EDCs in source water were usually less than 10 ng L1,
except for atrazine (28 ng L1), caffeine (27 ng L1), fluorene (13 ng L1), galaxolide
(28 ng L1), metolachlor (15 ng L1), musk ketone (16 ng L1), and TCEP (13 ng L1),
though values for fluorene, galaxolide, and musk ketone were biased by low frequencies of detection. The eight compounds that were detected in at least half of the
finished drinking water samples were atrazine, caffeine, carbamazepine, DEET,
ibuprofen, iopromide, meprobamate, and phenytoin. Median concentrations of
detected pharmaceuticals and EDCs in finished drinking water were usually less
than 10 ng L1, except for atrazine (29 ng L1), caffeine (23 ng L1), metolachlor
(86 ng L1), and triclosan (43 ng L1), though values for metolachlor and triclosan
were biased by low frequencies of detection.
Benotti et al. [11] surveyed the occurrence of 51 pharmaceuticals and EDCs in 19
source waters, 18 finished drinking waters, and 15 distribution systems from utilities
throughout the USA (Table 4). The 11 compounds that were detected in at least half
of the source water samples were atenolol, atrazine, carbamazepine, estrone,
gemfibrozil, meprobamate, naproxen, phenytoin, sulfamethoxazole, TCEP, and
trimethoprim. As with the previous study, median concentrations of detected pharmaceuticals and EDCs in source water were usually less than 10 ng L1, except for atrazine (32 ng L1), butylbenzyl phthalate (53 ng L1), BHT (49 ng L1), diethylhexyl
phthalate (150 ng L1), DEET (85 ng L1), estradiol (17 ng L1), metolachlor (17 ng L1),
nonylphenol (100 ng L1), sulfamethoxazole (12 ng L1), TCEP (120 ng L1), and
TCPP (180 ng L1). The values for estradiol, BHT, butylbenzyl phthalate, and diethylhexyl phthalate were biased by low frequencies of detection. Only three compounds
(atrazine, meprobamate, and phenytoin) were detected in at least half of the finished
drinking water samples. Median concentrations of detected pharmaceuticals and

230

D. Gerrity and S. Snyder

Table 2 TOrC concentrations (ng L1) at two US wastewater treatment plants

Wastewater treatment plant 1
Wastewater treatment plant 2
TOrC

Primary
effluent

Secondary
effluent

Finished
effluent

Primary
effluent

Secondary
effluent

Finished
effluent

Acetaminophen
Atenolol
Atorvastatin
Atrazine
Benzophenone
BHA
Bisphenol A
Caffeine
Carbamazepine
Cimetidine
DEET
Diazepam
Diclofenac
Diphenhydramine
Estradiol
Estrone
Ethynylestradiol
Fluoxetine
Gemfibrozil
Ibuprofen
Iopromide
Meprobamate
Musk ketone
Naproxen
Octylphenol
Phenytoin
Primidone
Progesterone
Sucralose
Sulfamethoxazole
TCEP
TCPP
Testosterone
Triclocarbon
Triclosan
Trimethoprim

NA
1,600
98
<5
<1,000
170
550
67,000
160
NA
510
<5
120
NA
<1
<1
<1
25
2,900
17,000
<200
1,400
<500
15,000
<500
97
140
34
NA
1,900
220
<2,000
40
NA
1,300
700

NA
730
<10
<5
<1,000
<20
<100
<100
190
NA
72
<5
96
NA
0.52
6.7
<1
33
<5
<20
<200
470
<500
<10
<500
120
140
7.3
NA
1,500
350
2,000
<0.5
NA
48
19

NA
220
16
<5
<1,000
<20
<100
<100
180
NA
190
<5
63
NA
<0.5
<0.2
<1
29
17
<20
<200
340
<500
13
<500
130
140
8.0
NA
1,500
360
2,200
<0.5
NA
48
17

170,000
2,600
NA
NA
1,000
250
430
120,000
260
350
420
NA
NA
1,200
NA
NA
NA
25
290
30,000
32,000
280
<250
12,000
NA
NA
<5.0
NA
28,000
650
360
1,900
NA
550
1,100
440

<500
430
NA
NA
250
16
<5.0
<5.0
340
120
17
NA
NA
61
NA
NA
NA
24
3.6
<10
7,700
62
<25
13
NA
NA
<0.50
NA
51,000
480
440
1,000
NA
200
12
26

<500
560
NA
NA
440
12
<5.0
30
310
86
17
NA
NA
47
NA
NA
NA
10
3.6
<10
2,000
61
<25
27
NA
NA
<0.50
NA
77,000
370
420
900
NA
67
3.7
24

NA not analyzed

EDCs in finished drinking water were generally less than 10 ng L1, except for
atrazine (49 ng L1), bisphenol A (25 ng L1), galaxolide (31 ng L1), nonylphenol
(93 ng L1), BHT (26 ng L1), metolachlor (16 ng L1), DEET (63 ng L1), TCEP
(120 ng L1), and TCPP (210 ng L1). Again, some of these median concentrations
were biased by low frequencies of detection. Finally, the four compounds that were

Wastewater and Drinking Water Treatment Technologies

231

Table 3 TOrC concentrations (ng L1) in US source and finished drinking water
Source (n = 20)
Finished (n = 20)
Contaminant

Max.

Med.

Max.

Med.

Acetaminophen
Androstenedione
Atrazine
Caffeine
Carbamazepine
DEET
Erythromycin
Estrone
Fluorene
Galaxolide
Gemfibrozil
Hydrocodone
Ibuprofen
Iopromide
Meprobamate
Metolachlor
Musk ketone
Naproxen
Oxybenzone
Phenytoin
Progesterone
Sulfamethoxazole
TCEP
Triclosan
Trimethoprim

9.5
1.9
570
87
39
28
3.5
1.4
13
30
11
1.9
24
46
16
170
17
16
7.4
13
1.1
44
66
30
2.3

1.6
1.9
28
27
3.1
6.9
2.2
1.2
13
28
4.8
1.9
4.2
7.6
5.9
15
16
2.2
2.9
3.2
1.1
8.1
13
1.9
2.2

7
1
17
14
18
20
8
2
1
3
13
1
16
14
16
7
3
10
4
18
1
17
15
6
3

<1.0
<1.0
430
83
5.7
30
1.3
2.3
<1.0
<1.0
6.5
<1.0
32
31
13
160
17
8
1.1
6.7
1.1
<1.0
19
43
1.3

<1.0
<1.0
29
23
2.8
5.1
1.3
1.7
<1.0
<1.0
4.2
<1.0
3.8
6.5
3.8
86
17
8
1.1
2.3
1.1
<1.0
5.5
43
1.3

15
12
11
18
1
2

13
13
15
4
1
1
1
14
2

7
1
1

The # sign represents the number of samples with reportable concentrations for
that particular contaminant [31]

detected in at least half of the distribution system samples were atrazine, atenolol,
meprobamate, and phenytoin. Median concentrations of detected pharmaceuticals
and EDCs in the distribution systems were generally less than 10 ng L1, except for
atrazine (50 ng L1), DEET (49 ng L1), metolachlor (18 ng L1), nonylphenol
(97 ng L1), TCEP (150 ng L1), and TCPP (220 ng L1), though values for metolachlor
and nonylphenol were biased by low frequencies of detection.
TOrC concentrations in source waters are generally a direct function of (1) the contribution of wastewater to the source, (2) TOrC occurrence in the wastewater influent,
(3) unit operations and treatment efficacy at the contributing WWTPs, and (4) degree
of natural attenuation after environmental discharge. Accordingly, in both of the studies
presented above, TOrC occurrence in the finished drinking water was governed by (1)
TOrC concentrations in the source waters and (2) removal during drinking water treatment. Due to the importance of treatment efficacy on TOrC concentrations, the following sections provide a summary of the most common technologies for drinking water
and wastewater treatment. Although the treatment processes have been categorized

Max.

17
1.4
36
1.4
870
49
14
54
51
110
0.47
1.2
170
0.90
3.0
48
24
9.3
73
81

Contaminant

Estradiol
Ethynylestradiol
Atenolol
Atorvastatin
Atrazine
BHT
Bisphenol A
Butylbenzyl phthalate
Carbamazepine
DEET
Diazepam
Diclofenac
Diethylhexyl phthalate
Estrone
Fluoxetine
Galaxolide
Gemfibrozil
Linuron
Meprobamate
Metolachlor

17
1.4
2.3
0.80
32
49
6.1
53
4.1
85
0.43
1.1
150
0.30
0.80
3
2.2
4.1
8.2
17

Med.
1
1
12
3
15
1
3
2
15
6
2
4
2
15
3
4
11
5
16
7

#
<0.50
<1.0
18
<0.25
870
26
25
<50
18
93
0.33
<0.25
<120
<0.20
0.82
33
2.1
6.2
42
27

Max.
<0.50
<1.0
1.2
<0.25
49
26
25
<50
6.0
63
0.33
<0.25
<120
<0.20
0.71
31
0.48
6.1
5.7
16

Med.

15
1
1

8
6
1

2
2
7
2
14
6

#
<0.50
<1.0
0.84
<0.25
930
<25
<5.0
<50
10
63
<0.25
<0.25
<120
<0.20
0.64
<25
1.2
<0.50
40
22

Max.
<0.50
<1.0
0.47
<0.25
50
<25
<5.0
<50
6.8
49
<0.25
<0.25
<120
<0.20
0.64
<25
0.43
<0.50
5.2
18

Med.

Table 4 TOrC concentrations (ng L1) in US source water, finished drinking water, and distribution systems
Source (n = 19)
Finished (n = 18)
Distribution (n = 15)

12

6
4

11
3

232
D. Gerrity and S. Snyder

32
130
<0.50
1.2
29
2.0
3.1
<2.5
110
530
720
1.2
6.4
11

0.90
100
<0.50
0.70
5.1
1.0
2.2
<2.5
12
120
180
1.1
3.0
0.80

11
8

3
14
3
4

17
10
8
2
6
11

<0.50
100
<0.50
<0.50
19
<0.50
0.57
<2.5
3.0
470
510
<0.50
1.2
<0.25

<0.50
93
<0.50
<0.50
6.2
<0.50
0.57
<2.5
0.39
120
210
<0.50
1.2
<0.25

10

4
7
5

<0.50
110
0.77
<0.50
16
<0.50
<0.50
2.9
0.32
200
240
<0.50
<1.0
<0.25

The # sign represents the number of samples with reportable concentrations for that particular contaminant [11]

Naproxen
Nonylphenol
Norfluoxetine
o-Hydroxy atorvastatin
Phenytoin
p-Hydroxy atorvastatin
Progesterone
Risperidone
Sulfamethoxazole
TCEP
TCPP
Testosterone
Triclosan
Trimethoprim

<0.50
97
0.77
<0.50
3.6
<0.50
<0.50
2.9
0.32
150
220
<0.50
<1.0
<0.25

2
1

10

1
1
6
6

Wastewater and Drinking Water Treatment Technologies

233

234

D. Gerrity and S. Snyder

based on their most common applications, there is certainly technology overlap between
drinking water and wastewater treatment. Natural attenuation is not discussed since it
is highly site specific. However, many of the treatment processes (e.g., photolysis, biological wastewater processes, and filtration) mimic natural attenuation mechanisms so
there is some degree of overlap between the two concepts.

TOrC Removal During Drinking Water Treatment

Coagulation/Flocculation/Sedimentation
Coagulation involves the addition of treatment chemicals such as aluminum sulfate
(alum, Al2(SO4)3), ferric chloride (FeCl3), and ferric sulfate (Fe2(SO4)3) to promote
the destabilization of small suspended particles and colloidal material [32]. During
the rapid mix phase, the metal salts hydrolyze, form complexes with organic solutes, and ultimately precipitate as amorphous metal hydroxides. After the rapid mix
phase, a period of slow mixing (flocculation) is often used to promote the aggregation of smaller particulates and organic matter into larger settleable flocs. These can
be removed by granular media filtration either with or without prior gravity settling
or dissolved air flotation [33, 34]. Conventional coagulation is generally intended
for turbidity removal via destabilization of existing particles. Enhanced coagulation, which employs higher coagulant doses and/or pH reduction, is now used for
the removal of dissolved organic compounds.
Westerhoff et al. [35] evaluated the efficacy of alum and ferric chloride coagulation for PPCP and EDC removal in bench-scale experiments. In four different
surface waters, 34 of the 36 PPCPs and EDCs were removed by less than 15%. The
two remaining compounds (DDT and benzo(a)pyrene) were removed by 31 and
70%, respectively, due to their greater hydrophobicity (log KOW > 6.0). Results from
the bench-scale tests suggested that (1) removal efficiency and KOW were linearly
correlated, (2) there was no added benefit with enhanced coagulation, and (3)
removal efficiencies were similar between the two coagulants. Coagulation was also
deemed ineffective for PPCP and EDC removal in another study [36], which noted
no significant difference in pharmaceutical concentrations before and after coagulation. A summary of these results is provided in Table 5.

Activated Carbon Adsorption

Activated carbon is a highly porous material that has typically been used for control
of taste and odor problems, though its applicability is expanding due to changes in
disinfection by-product (DBP) regulation and the ability for activated carbon to remove
DBP precursors [37]. The two main forms of activated carbon are utilized in different
ways. Powder activated carbon (PAC) is applied similar to a coagulation process and

Wastewater and Drinking Water Treatment Technologies

235

Table 5 TOrC removal by coagulation/flocculation/sedimentation [18]

<20% removal
2050% removal
5080% removal
>80% removal
Acetaminophen
Androstenedione
Atrazine
Caffeine
Carbamazepine
DEET
Diazepam
Diclofenac
Erythromycin
Estradiol
Estriol
Estrone
Ethinyl estradiol
Fluorene
Fluoxetine
Galaxolide
Gemfibrozil
Hydrocodone
Ibuprofen
Iopromide
Lindane
Meprobamate
Metolachlor
Musk ketone
Naproxen
Oxybenzone
Pentoxifylline
Phenytoin
Progesterone
Sulfamethoxazole
TCEP
Testosterone
Triclosan
Trimethoprim

DDT

Benzo(a)pyrene

10 mg alum per mg total organic carbon or equivalent dose ([Fe3+]/[Al3+] = 1) of FeCl3

can be used as an additional coagulant during seasonal contaminant spikes. Granular

activated carbon (GAC) requires permanent contactors configured as media filters or
fixed-bed adsorbers, which can also allow for microbial growth and significant biodegradation. As with coagulation or any adsorption process, the efficacy of PAC and GAC
is highly dependent on the hydrophobicity and size of the target compounds.
Many researchers have reported on the efficacy of GAC and PAC for the removal
of PPCPs and EDCs [18, 35, 36, 38]. GAC and PAC treatment for trace contaminants
can be hindered by the presence of high concentrations of NOM, as they compete for
the same adsorption sites on the substrate. Thus, the effectiveness and life span of

236

D. Gerrity and S. Snyder

Table 6 Freundlich parameters for four pharmaceuticals [36]
Deionized water
Groundwater
Contaminant

KA

KA

Bezafibrate
Carbamazepine
Clofibric acid
Diclofenac

0.19
0.38
0.25
0.19

141
430
71
141

0.22
0.22
0.54
0.21

77
90
63
36

activated carbon is highly dependent on the characteristics of the target water matrix
[18]. In contrast to coagulation, the octanolwater distribution coefficient (DOW),
rather than KOW, is a better indicator of performance for many of the compounds [18].
Removal can be correlated with KOW for neutral compounds, however. In general,
higher PAC concentrations lead to higher removal of most PPCPs and EDCs.
Protonated bases are very susceptible to removal by PAC, because they are electrostatically attracted to negatively charged moieties on the substrates surface.
Conversely, deprotonated acids are electrostatically repelled from the surface-bound
negatively charged moieties and do not adsorb. The removal of neutrally charged
molecules is controlled by the hydrophobicity of a particular compound given that
the mechanism for adsorption is hydrophobic exclusion from the aqueous phase:
compounds with low KOW values are less likely to adsorb to activated carbon [35].
Adsorption of target contaminants is often modeled with batch isotherm testing
and the Freundlich isotherm model, as described below [37]:
qA = K A CA1/ n

where
qA = equilibrium adsorbent-phase concentration of contaminant (mg contaminant
g1 adsorbent)
KA = Freundlich adsorption capacity parameter ((mg g1)(L mg1)1/n)
CA = equilibrium concentration of contaminant in solution (mg L1)
n = Freundlich adsorption intensity parameter (unitless)
Empirical determination of the KA and n parameters allows engineers to calculate
the expected removals of certain compounds in addition to design criteria specific to
the activated carbon reactor. Ternes et al. [36] published KA and n values for four
pharmaceuticals in deionized water and groundwater (Table 6).
With respect to general removal trends, Table 7 categorizes TOrC removal for
5 mg L1 and 45 h of contact time with PAC [18]. In contrast to coagulation, only
two of the target compounds are removed by less than 20% with PAC, and a majority of the compounds are removed by more than 50%. Activated carbon is generally
superior to coagulation, but there are still compounds that are resistant to removal.
Again, increased removal of target contaminants must be balanced with the additional operational costs (i.e., infrastructure, virgin material, regeneration, disposal,
etc.) associated with PAC and GAC.

Wastewater and Drinking Water Treatment Technologies

Table 7 TOrC removal by PAC [18]
<20% removal
2050% removal
Ibuprofen
Iopromide

DEET
Diclofenac
Erythromycin
Estriol
Gemfibrozil
Meprobamate
Metolachlor
Naproxen
Phenytoin
Sulfamethoxazole
TCEP

237

5080% removal

>80% removal

Acetaminophen
Androstenedione
Atrazine
Caffeine
Carbamazepine
DDT
Diazepam
Estradiol
Estrone
Ethinyl estradiol
Galaxolide
Hydrocodone
Lindane
Musk ketone
Pentoxifylline
Testosterone
Trimethoprim

Benzo(a)pyrene
Fluorene
Fluoxetine
Oxybenzone
Progesterone
Triclosan

PAC dose = 5 mg L1 and contact time = 45 h

Ultraviolet Light (Photolysis)

Ultraviolet (UV) light has become more common in water treatment since the
discovery in the late 1990s that it is highly effective for Cryptosporidium oocyst
inactivation. Although typical disinfection doses are in the range of 20100 mJ cm2,
much higher doses (e.g., 5001,000 mJ cm2) are usually employed for contaminant
oxidation. Most UV reactors can be divided into two categories based on lamp
characteristics and resulting output: (1) monochromatic/low pressure and (2) polychromatic/medium pressure. Both types of lamps contain mercury gas that emits
ultraviolet light when excited by electrons. Low-pressure lamps produce a monochromatic output at 254 nm, which is extremely effective for UV disinfection, and
medium-pressure bulbs produce a polychromatic output at a higher intensity that
induces reactions in a broader range of contaminants. Both types of reactors are
susceptible to fouling due to the lower solubility of many natural constituents (e.g.,
CaCO3) at higher temperatures found at the surface of the bulb. High turbidity and
high levels of organic matter also reduce the effectiveness of photolysis.
Photolysis modifies and destroys organic contaminants by direct bond cleavage
and through reactions with inorganic constituents to form highly reactive intermediates, such as OH. However, the extent of photolysis at typical UV disinfection
doses is quite small so TOrC mitigation is not considered a synergistic benefit of
UV disinfection [18]. In bench- and pilot-scale experiments, only four of 29 detected
compounds were degraded by more than 20% with medium-pressure photolysis at
a UV dose of 40 mJ cm2 (Table 8). Medium-pressure photolysis at a UV dose of

238

D. Gerrity and S. Snyder

Table 8 TOrC degradation by low-dose UV photolysis [18]

<20% degradation
2050% degradation
5080% degradation
Androstenedione
Atrazine
Caffeine
Carbamazepine
DEET
Diazepam
Dilantin
Erythromycin
Estradiol
Estriol
Estrone
Ethinyl estradiol
Fluoxetine
Gemfibrozil
Hydrocodone
Ibuprofen
Iopromide
Meprobamate
Naproxen
Oxybenzone
Pentoxifylline
Progesterone
TCEP
Testosterone
Trimethoprim

Acetaminophen

>80% degradation

Diclofenac
Sulfamethoxazole
Triclosan

UV dose = 40 mJ cm2

450 mJ cm2 achieved significantly increased removals (Table 9), and the addition
of hydrogen peroxide (H2O2) provided further improvements to the process [18].
The use of H2O2 to improve UV-based oxidation will be discussed in greater detail
in relation to advanced wastewater treatment.
Structural properties of individual compounds play a role in how effectively a
compound may be destroyed by photolysis. For example, aromatic compounds
absorb light in the UV spectrum so compounds with aromatic centers are more
susceptible to photolysis. Of the pharmaceuticals and EDCs investigated, diclofenac,
sulfamethoxazole, and triclosan were most susceptible to removal by photolysis,
and all have absorption spectra that overlap with the wavelength-specific peaks
generated by medium-pressure lamps. Conversely, aliphatic compounds that lack
conjugated double bonds and the appropriate absorption bands are very resistant to
UV photolysis. Although UV photolysis may be effective at removing some pharmaceuticals and EDCs, it is generally not viable as a stand-alone treatment process
as many compounds have structures that are not amenable to UV photolysis.

Wastewater and Drinking Water Treatment Technologies

Table 9 TOrC destruction by high-dose UV photolysis [18]
<20% degradation
2050% degradation
5080% degradation
Androstenedione
Caffeine
DEET
Diazepam
Meprobamate
TCEP

Carbamazepine
Gemfibrozil
Ibuprofen
Pentoxifylline
Progesterone
Testosterone
Trimethoprim

Atrazine
Dilantin
Erythromycin
Iopromide

239

>80% degradation
Acetaminophen
Diclofenac
Estradiol
Estriol
Estrone
Ethinyl estradiol
Fluoxetine
Hydrocodone
Naproxen
Oxybenzone
Sulfamethoxazole
Triclosan

UV dose = 450 mJ cm2

Free Chlorine and Chloramine

Chlorination is the most common form of disinfection due to its effectiveness
against a variety of pathogens (with the exception of protozoan parasites) and the
ease with which a residual can be maintained throughout a distribution system.
However, many utilities are currently turning toward chloramination for residual
disinfection [39] due to its greater stability in distribution systems and lower potential
to form halogenated DBPs [40]. The amount of chlorine or chloramine utilized in
drinking water applications is usually reported as units of concentration time (CT).
Chlorine and chloramine doses of 3 mg L1 for 24 h (CT = 4,320 mg min L1) were
evaluated for PPCP and EDC oxidation [18]. These results are illustrated in Tables 10
and 11, respectively.
Compounds most susceptible to removal by chlorine or chloramine often contain
aromatic structures with electron-donating functional groups (e.g., hydroxyl, amine,
and methoxy groups) [41, 42]. For example, steroid hormones containing phenolic
groups were removed by more than 95%. Other compounds susceptible to chlorine
or chloramine oxidation may contain primary amines attached to conjugated
rings (e.g., trimethoprim and sulfamethoxazole), highly alkylated benzenes (e.g.,
gemfibrozil and hydrocodone), and polycyclic aromatic hydrocarbons (e.g., carbamazepine, benzo(a)pyrene, diclofenac, and naproxen). The most resistant compounds
often lack carboncarbon double bonds and contain carboxyl groups, ketones, heterocyclic nitrogen, or primary amides (e.g., iopromide and meprobamate) [18].
Given that some compounds are resistant to chlorine or chloramine oxidation, complete mineralization is not possible. As with any treatment technology, the potential
effects of molecular (e.g., chlorinated triclosan [43]) or bulk (e.g., total organic
halogens [40]) transformation products must be considered.

240

D. Gerrity and S. Snyder

Table 10 TOrC oxidation by chlorination [18]

<20% degradation
2050% degradation
Androstenedione
Atrazine
Caffeine
Carbamazepine
DDT
DEET
Dilantin
Fluorene
Fluoxetine
Ibuprofen
Iopromide
Lindane
Meprobamate
Metolachlor
Progesterone
TCEP
Testosterone

Diazepam
Galaxolide
Pentoxifylline

5080% degradation

>80% degradation

Gemfibrozil

Acetaminophen
Benzo(a)pyrene
Diclofenac
Erythromycin
Estradiol
Estriol
Estrone
Ethinyl estradiol
Hydrocodone
Musk ketone
Naproxen
Oxybenzone
Sulfamethoxazole
Triclosan
Trimethoprim

Chlorine concentration = 3 mg L1, contact time = 24 h, pH = 7.98.5

Table 11 TOrC oxidation by chloramination [18]

<20% degradation
2050% degradation
5080% degradation
Androstenedione
Atrazine
Caffeine
Carbamazepine
DDT
DEET
Diazepam
Dilantin
Erythromycin
Fluorene
Fluoxetine
Gemfibrozil
Ibuprofen
Iopromide
Lindane
Meprobamate
Metolachlor
Musk ketone
Naproxen
Pentoxifylline
Progesterone
Sulfamethoxazole
TCEP
Testosterone
Trimethoprim

Hydrocodone
Galaxolide

Benzo(a)pyrene
Diclofenac
Oxybenzone

Chloramine concentration = 3 mg L1, contact time = 24 h, pH = 8.0

>80% degradation
Acetaminophen
Estradiol
Estriol
Estrone
Ethinyl estradiol
Triclosan

Wastewater and Drinking Water Treatment Technologies

Table 12 TOrC oxidation by ozonation [18]
<20% degradation
2050% degradation
TCEP

Atrazine
Iopromide
Meprobamate

241

5080% degradation

>80% degradation

DEET
Diazepam
Dilantin
Ibuprofen

Acetaminophen
Androstenedione
Caffeine
Carbamazepine
Diclofenac
Erythromycin
Estradiol
Estriol
Estrone
Ethinyl estradiol
Fluoxetine
Gemfibrozil
Hydrocodone
Naproxen
Oxybenzone
Pentoxifylline
Progesterone
Sulfamethoxazole
Testosterone
Triclosan
Trimethoprim

Ozone concentration = 2.5 mg L1 and contact time = 24 min

Ozone
Although relatively energy intensive, ozone is highly effective for both chemical
oxidation and microbial inactivation (including Giardia cysts and Cryptosporidium
oocysts). Ozone either reacts directly with organic molecules or indirectly through
the formation of radical species [44]. Ozone is relatively unstable in water and
wastewater (i.e., decays in minutes) so it is not possible to maintain a long-term
residual. The natural decomposition of ozone into OH is particularly relevant for
wastewater applications, but H2O2 can also be used to drive the formation of OH in
drinking water and wastewater. For direct reactions, ozone reacts rapidly with
amines, phenols, and double bonds in aliphatic compounds.
In contrast to photolysis, many PPCPs and EDCs are degraded rapidly with
ozone CTs commonly used for disinfection applications (less than 20 mg min L1)
[19, 45, 46]. Since molecular ozone is very effective for pharmaceutical and EDC
treatment, modifying the process with H2O2 is not always necessary, although it
does increase the reaction rate [18]. However, some recalcitrant compounds (e.g.,
clofibric acid and ibuprofen) may necessitate augmentation with H2O2 to achieve
higher levels of treatment, particularly in drinking water applications where the
natural ozone decomposition pathway is not as prevalent [18]. Table 12 describes
the relative removals of a suite of PPCPs and EDCs by ozonation.

242

D. Gerrity and S. Snyder

Table 13 Second-order ozone and OH rate constants
for select TOrCs [16, 25, 29, 4755]
Compound

kO3 (M1 s1)

kOH
(M1 s1)

Meprobamate
Sulfamethoxazole
Trimethoprim
Carbamazepine
Phenytoin
Primidone
Triclosan
Atenolol
TCEP
Musk ketone
Atrazine
Gemfibrozil
Diclofenac
Ibuprofen
Naproxen
DEET
Bisphenol A

<10
2.5 106
2.7 105
3 105
~10
~10
5.1 108
6.3 105
<10
<10
6
~500
1 106
9.6
~1 105
~10
~1 109

(15) 109
5.5 109
6.9 109
8.8 109
(510) 109
(510) 109
(510) 109
8.0 109
7.4 108
(15) 109
3 109
(510) 109
7.5 109
7.4 109
9.6 109
(510) 109
1 1010

Numerous studies have developed second-order rate constants for the ozonation of
PPCPs and EDCs; a subset of these rate constants is presented in Table 13. For compounds with unknown rate constants, quantitative structure activity relationships
(QSARs) can be used to estimate their susceptibility to ozonation. For example,
Huber et al. [16] noted that the aromatic and tertiary amine moieties found in sulfonamide and macrolide antibiotics are reactive with ozone, and all compounds within
these classes should have similar reaction rates. Furthermore, the authors indicated
that ketone-containing steroid hormones are likely to have rate constants that are
approximately one order of magnitude less than the phenolic steroid hormones.
The compounds experiencing the least amount of degradation are generally characterized
by extensive branching (e.g., meprobamate and iopromide) and are sometimes
designed specifically to resist oxidation (e.g., the flame retardant TCEP). As with
chlorine and other oxidation processes, complete mineralization with ozone is impractical given the energy requirement and the potential to form DBPs (e.g., bromate).
Thus, the potential effects of ozone transformation products must be considered.

TOrC Removal During Wastewater Treatment

Raw wastewater quality varies tremendously depending on the contributing sources
(i.e., small residential communities, large urban areas, industrial discharge, etc.),
and the extent of treatment ultimately depends on the intended use or effluent

Wastewater and Drinking Water Treatment Technologies

243

discharge location. For example, wastewater permitted for ocean discharge does not
have the same water quality requirements as that permitted for indirect potable reuse
(IPR). Conventional wastewater treatment has evolved over time but generally
includes the following unit operations and processes: preliminary solids removal,
primary clarification, secondary biological treatment, filtration, and disinfection
[56]. Depending on the specific requirements of the discharge permit, conventional
treatment may be supplemented with nutrient removal (i.e., for nitrogen or phosphorus removal) or other advanced processes to achieve a higher quality effluent.
This may be required for discharge to a sensitive ecosystem (e.g., areas susceptible
to algal blooms and eutrophication) or for IPR applications. Advanced treatment
may include membrane treatment or AOPs, such as UV/H2O2 and ozone/H2O2.
Traditionally, wastewater treatment trains have not been designed for TOrC
removal. However, the growing body of occurrence data for wastewater-derived
contaminants (including PPCPs and EDCs) in surface waters [15, 57], the recognition that wastewater effluents are impacting natural waters, and the potential adverse
effects on aquatic ecosystems [58] have brought these issues to the forefront. Since
wastewater discharge is the primary source of PPCPs and EDCs in the environment
[59], optimization of wastewater treatment processes may be the most efficient
strategy to mitigate the adverse effects of these compounds. The following sections
describe the general efficacy of both conventional and advanced wastewater treatment processes for PPCP and EDC mitigation.

Conventional Wastewater Treatment

Conventional wastewater treatment processes relying on physical separation, including preliminary solids removal, primary clarification, grit removal, and media
filtration, provide limited reductions in TOrC concentrations. On the other hand,
secondary treatment, which involves both adsorption and biological processes, can
be highly effective depending on the target contaminant and operational conditions
[17]. Activated sludge processes, whether in conventional activated sludge (CAS)
configurations or membrane bioreactors (MBRs), may achieve high removals (up to
99%) of hormones and certain pharmaceuticals (e.g., the analgesics acetaminophen
and ibuprofen), but biological treatment may be insufficient to remove the more
recalcitrant compounds (e.g., the anticonvulsants phenytoin and carbamazepine)
[17, 57, 60]. Limited removal efficiencies have been observed for certain antibiotics
and antimicrobial compounds, such as erythromycin (10%), sulfamethoxazole
(64%), and triclosan (68%) [17]. It is important to note that MBR systems contain
microfiltration (MF) or ultrafiltration (UF) membranes, but it is generally the biological processes that are responsible for PPCP and EDC removal. Joss et al. [61]
did not observe any relationships between structural characteristics of the compounds and efficacy of secondary treatment, but the study did identify microbial
transformation, rather than sludge partitioning, as the dominant mechanism for all
of the target compounds.

244

D. Gerrity and S. Snyder

For the most susceptible compounds, CAS and MBRs achieve comparable
removals, but some studies indicate that the longer SRTs associated with MBRs
provide significant benefits with respect to the removal of recalcitrant compounds
[60]. MBRs can be operated with longer SRTs due to their high microbial loads
and more concentrated return activated sludge. CAS would require excessive
return flows to achieve comparable SRTs. Clara et al. [62] observed a positive
correlation between PPCP and EDC removal and longer SRTs. For most of the
target compounds, a critical SRT of 10 days was observed, but for a small number
of compounds (e.g., anticonvulsants), PPCP and EDC removal was poor regardless of SRT.
Geographic factors, such as climate, can also influence the efficacy of secondary treatment. For example, Ternes et al. [63] observed 80% to greater than
99% removals of estrogenic hormones in a Brazilian WWTP, but the removal
efficiencies of those same compounds were lower (070%) in a German WWTP.
This difference was primarily attributed to the higher water temperature of the
Brazilian WWTP. Therefore, the efficacy of biological treatment is dependent
on a variety of factors, including the compound of interest, process configuration,
operational parameters, and geographical location. Regardless, PPCPs and
EDCs are never completely removed, and they are typically detected in secondary effluent at ng L1 to mg L1 concentrations [64], as described earlier in
Table 2.

Advanced Wastewater Treatment: Membranes

The efficacy of membranes for PPCP and EDC removal varies with membrane
pore size. Low-pressure MF and UF processes are generally ineffective alternatives for TOrC removal [18] due to the fact that their pore sizes are relatively large
and the MW cutoff is approximately 100,000 and 2,000 Da, respectively. Thus,
PPCPs and EDCs, which are usually less than 500 Da, have the potential to easily
pass through the pores. Indirect PPCP and EDC removal by MF and UF membranes is affected by physiochemical parameters. Hydrophobic compounds
adsorbed onto particulates or colloids that will not pass through the membrane
pores are readily rejected. High-pressure nanofiltration (NF) and reverse osmosis
(RO) membranes have much tighter pores (the MW cutoff for these membranes is
approximately 250 and 100 Da, respectively) so PPCPs and EDCs are generally
rejected by these membranes. In fact, the concentrations of these target contaminants are generally below the MRL (often 0.2525 ng L1) after RO and NF treatment [17, 65].
Snyder et al. [18] studied a variety of pilot and full-scale membrane processes and
reported similar results, which are summarized in Table 14. They concluded that hydrophobic compounds with aliphatic substituted aromatic ring structures and high pKa
values were removed by low-pressure MF and UF membranes. This can be attributed

Wastewater and Drinking Water Treatment Technologies

245

Table 14 TOrC removal by membrane and MBR processes [18]

Percent removal
Membrane size
Number of systems tested

MF
3

UF
5

UF/MBR
4

NF
3

RO
9

Acetaminophen
Androstenedione
Atrazine
Benzo(a)pyrene
Caffeine
Carbamazepine
DDT
DEET
Diazepam
Diclofenac
Erythromycin
Estradiol
Estriol
Estrone
Ethinyl estradiol
Fluorene
Fluoxetine
Galaxolide
Gemfibrozil
Hydrocodone
Ibuprofen
Iopromide
Lindane
Meprobamate
Metolachlor
Musk ketone
Naproxen
Oxybenzone
Pentoxifylline
Phenytoin
Progesterone
Sulfamethoxazole
TCEP
Testosterone
Triclosan
Trimethoprim

<20
<20
ND
ND
<20
<20
ND
<20
ND
<20
<20
<20
ND
<20
ND
ND
2050
<20
<20
<20
<20
<20
ND
<20
ND
<20
<20
<20
<20
<20
ND
<20
<20
ND
2050
<20

<20
2050
<20
>80
<20
<20
>80
<20
2050
<20
2050
2050
<20
2050
2050
>80
>80
2050
<20
<20
<20
<20
2050
<20
2050
2050
<20
5080
<20
<20
5080
2050
<20
2050
>80
<20

>80
>80
ND
ND
>80
2050
5080
5080
<20
<20
2050
5080
>80
>80
>80
ND
2050
ND
2050
2050
5080
<20
ND
<20
ND
ND
>80
>80
>80
<20
>80
2050
<20
>80
5080
2050

2050
5080
5080
>80
5080
5080
>80
5080
5080
5080
>80
5080
5080
5080
5080
>80
>80
5080
5080
5080
5080
>80
5080
5080
5080
>80
2050
>80
5080
5080
5080
5080
5080
5080
>80
5080

>80
>80
ND
ND
>80
>80
ND
>80
>80
>80
>80
>80
>80
>80
>80
ND
>80
>80
>80
>80
>80
>80
ND
>80
ND
>80
>80
>80
>80
>80
>80
>80
>80
ND
>80
>80

ND not detected

to adsorption onto larger material that is readily rejected by the membrane or to electrostatic repulsion from the membrane surface. Neutrally charged or hydrophilic
compounds were not removed by MF or UF membranes. Effective removal of all
PPCPs and EDCs was observed following treatment with NF or RO membranes.

246

D. Gerrity and S. Snyder

Advanced Wastewater Treatment: Advanced Oxidation Processes

AOPs utilize highly reactive chemical species such as free radicals to oxidize
chemical contaminants in water [66]. The most common AOPs include UV/H2O2
and ozone/H2O2, but other AOP technologies, such as UV/TiO2 (titanium dioxide)
photocatalysis and nonthermal plasma (NTP), may be viable alternatives in the
future [67, 68]. Although AOPs provide some level of treatment with their base
mechanisms (e.g., direct photolysis of chemical contaminants from UV/H2O2),
the dominant treatment pathway generally involves oxidation by highly reactive,
nonspecific OH [18, 19].
In general, AOPs are very effective treatment technologies for removing PPCPs
and EDCs from water, though the processes are usually energy intensive. When
optimized, the processes can also be very fast, given the short-lived and highly
reactive nature of OH. Huber et al. [16] reported second-order OH rate constants
for a suite of PPCPs and EDCs ranging from 3.3 109 to 9.8 109 M1 s1 (also
refer to Table 13). Snyder et al. [18] reported that treatment with ozone vs. ozone/
H2O2 was similar in terms of overall PPCP and EDC degradation, but the AOP
process yielded faster reaction rates (i.e., nearly instantaneous). In the same study,
a limited number of compounds (e.g., clofibric acid and ibuprofen) were not
removed by ozone alone (less than 10% removal), but were effectively removed
by ozone/H2O2 (greater than 90% removal). It is important to note that the efficacy
of ozone vs. ozone/H2O2 is highly dependent on the water matrix. Drinking water
applications provide much greater dissolved ozone exposure, whereas ozone
decomposes rapidly into OH in wastewater applications. Therefore, in the same
example presented above, the oxidation of clofibric acid and ibuprofen with
molecular ozone (relative to ozone/H2O2) may have improved in a wastewater
matrix due to rapid ozone decomposition into OH.
For UV/H2O2, pharmaceutical and EDC removal was generally not attributed
to direct photolysis. The addition of H2O2 was necessary to generate OH, which
was responsible for the oxidation of trace contaminants. Rosenfeldt and Linden
[69] reported small reductions in EDC concentrations with a UV dose of
1,000 mJ cm2, but those EDCs were removed by more than 90% with the same
UV dose and 15 mg L1 hydrogen peroxide. The authors also calculated secondorder rate constants on the order of 1010 M1 s1. Snyder et al. [18] reported greater
than 80% removal for 19 of 29 pharmaceuticals and EDCs following UV/H2O2
treatment (~375 mJ cm2 and 5 mg L1 hydrogen peroxide). Eight of the remaining ten compounds were between 50 and 80% removed, and only meprobamate
and TCEP, which are both highly resistant to oxidation, were less than 50%
removed.
Due to the highly reactive nature of OH, scavengers such as organic matter and
alkalinity reduce the efficacy of AOPs [46, 70, 71]. UV/H2O2 is also affected by
water with high turbidity and high levels of UV absorbance, both of which reduce
UV transmissivity. Therefore, it is important to understand the target water matrix

Wastewater and Drinking Water Treatment Technologies

247

Table 15 Summary of AOP EEO values (kWh m3 per log contaminant removal) for seven pharmaceuticals and EDCs
Contaminant
UVa
UV/H2O2a,b
UV/TiO2a,c
NTPd
Atenolol
Atrazine
Carbamazepine
Meprobamate
Phenytoin
Primidone
Trimethoprim

1.4
3.3
2.3
6.6
2.1
3.7
0.8

0.5
1.2
0.4
1.0
1.0
0.6
0.4

2.0
4.7
2.1
6.8
2.2
3.9
1.5

1.0
3.7
<0.7
3.5
2.0
2.2
<0.7

Benotti et al. [67]

10 mg L1 of H2O2
c
500 mg L1 of TiO2
d
Gerrity et al. [68]
a

when selecting the most appropriate AOP. The UV/H2O2 and ozone/H2O2 AOPs also
require peroxide addition and subsequent quenching, which is a significant cost
over the life of the system.
UV/TiO2 photocatalysis, which generates OH by irradiating a TiO2 slurry or
fixed film with UV light, and NTP, which generates UV light, ozone, and OH with
high-voltage pulses across two electrodes, are not limited by light-attenuating
matrices. Additionally, these processes do not require H2O2 so chemical addition
and quenching are not necessary; H2O2 may increase reaction rates, however.
Benotti et al. [67] and Gerrity et al. [68] evaluated the degradation of a suite of
pharmaceuticals and EDCs in surface waters with direct UV photolysis, UV/H2O2,
UV/TiO2 photocatalysis, and NTP. Table 15 provides a summary of the electrical
energy per order (EEO) of magnitude destruction values for each of the processes.
EEO values are a basis of comparison for many treatment options as they standardize energy consumption to the volume of water treated and the extent of treatment
(i.e., kWh m3 per log contaminant removal). Results from these studies indicate
that of these four AOP technologies that do not use ozone, UV/H2O2 is the most
efficient process, though UV/TiO2 photocatalysis and NTP provide viable, chemicalfree alternatives.

Advanced Wastewater Treatment: Indirect Potable Reuse

Treatment Trains
There is an increasing global trend toward more efficient use of water resources in
both urban and rural communities. In addition to innovative water management and
acquisition strategies (e.g., water transfers, banking, and trading), numerous municipalities are turning to water reuse in a variety of contexts to bolster their water
portfolios. Reclaimed water has the advantage of being a constant and reliable water

248

D. Gerrity and S. Snyder

source, and it is the only source that increases in supply relative to demand.
Historically, the use of reclaimed wastewater for municipal and agricultural irrigation has been the most common and accepted application, but diminishing water
suppliesprimarily the result of dramatic population growth and historic drought
conditions in many areasand a greater acceptance of water reuse have led to more
varied applications, including IPR.
Unplanned IPR can be captured colloquially in that everyone cannot live
upstream. In a more formal sense, unplanned IPR involves the environmental
discharge of conventionally treated wastewater effluent, which is subsequently used
as a drinking water source by another municipality. With respect to water quality,
the discharge of wastewater effluent generally conforms to the requirements of
National Pollutant Discharge Elimination System (NPDES) permits, and additional
requirements are sometimes established by local entities (e.g., the California
Department of Public Health [CDPH] Title 22 requirements). Many utilities are taking a proactive approach to environmental stewardship and public health by employing advanced wastewater treatment technologies (e.g., membrane filtration,
biological activated carbon, and soil aquifer treatment). These additional treatment
processes are common measures in many planned IPR systems. In a planned
IPR system, the discharge of wastewater effluent involves some form of environmental buffer, such as soil aquifer treatment and extended storage in a reservoir, and
is eventually integrated into the local potable water supply. However, planned IPR
systems vary considerably with respect to a number of variables, including level of
treatment in the WWTP, discharge mechanism, storage time in the environment, and
level of treatment in the drinking water treatment plant.
The standard treatment train for planned IPR is generally comprised of MF or
UF, RO, UV/H2O2, and aquifer injection (i.e., the Orange County Groundwater
Replenishment District). MF and UF are included primarily as a pretreatment strategy to reduce RO fouling. As discussed earlier, the use of RO is sufficient to approach
the detection limits of many TOrCs, but UV/H2O2 is included as an additional barrier against N-nitrosodimethylamine (NDMA), which is susceptible to UV light,
and 1,4-dioxane, which is susceptible to OH. The CDPH Title 22 requirements for
recycled water require the UV/H2O2 process to achieve 1.2-log destruction of
NDMA and 0.5-log destruction of 1,4-dioxane. The actual operational conditions
are site specific but generally require UV doses greater than 500 mJ cm2 and H2O2
concentrations exceeding 5 mg L1. Finally, aquifer injection, which must be preceded by mineral stabilization of the RO permeate, is included as an environmental
barrier primarily to increase public acceptance of the concept. Table 16 provides an
example of TOrC concentrations in this type of IPR system.
Although the standard IPR treatment train is extremely effective for TOrC mitigation, the production of concentrated brines, high energy costs associated with UV
oxidation and RO, and significant chemical requirements for operation and maintenance have prompted the development of alternative IPR treatment strategies. One
of the most promising alternatives is comprised of filtration (i.e., media, micro-, or
ultra-), ozone-based oxidation, biological activated carbon (BAC), and aquifer
injection. This type of treatment train, which has already demonstrated promise in

Wastewater and Drinking Water Treatment Technologies

249

Table 16 TOrC concentrations (ng L1) in a standard IPR treatment train

Microfiltration
Contaminant
Secondary effluent
permeate
RO permeate

UV/H2O2
effluent

Atenolol
Atorvastatin
Carbamazepine
Diazepam
Diclofenac
Enalapril
Fluoxetine
Gemfibrozil
Meprobamate
Naproxen
Phenytoin
Risperidone
Sulfamethoxazole
Trimethoprim

1.7
<0.25
<0.5
<0.25
<0.25
<0.25
<0.50
0.65
0.63
<0.50
<1.0
<0.25
<0.25
0.46

2,460
67
304
3.8
134
2.8
38
2,420
339
235
283
3.3
1,300
601

1,970
142
295
3.4
174
16
32
2,510
316
245
258
0.38
719
604

20
<0.25
1.5
<0.25
0.58
<0.25
<0.50
7.8
1.6
1.0
1.3
<0.25
2.6
4.3

pilot- and full-scale installations in Europe and Australia [72, 73], is particularly
promising for inland applications where brine disposal is an issue. Similar to the
standard configuration, filtration is provided as a pretreatment step to improve the
efficacy of ozonation and to reduce solids loadings on the subsequent BAC process.
Ozonation is incorporated as the primary treatment mechanism for TOrC mitigation, and the BAC process is provided for the removal of oxidation by-products and
recalcitrant TOrCs. Again, aquifer recharge is provided to increase public acceptance of the IPR concept. Although this alternative provides significant benefits over
the standard configuration, there are certainly issues that must be considered prior
to implementation, including bromate formation and pathogen regrowth in the BAC
process. Bromate formation can be mitigated with the addition of H2O2 during the
ozone process, but microbial regrowth may necessitate downstream disinfection
prior to discharge. Table 17 provides an example of TOrC concentrations in a pilotscale demonstration of UF, ozone/H2O2, and BAC. Ozone was dosed at 5 mg L1
(mass-based ozone:total organic carbon ratio of ~1.0), and H2O2 was dosed at
3.5 mg L1 (molar H2O2:ozone ratio of ~1.0) for bromate mitigation.

Advanced Wastewater Treatment: Residual Management

Advanced water and wastewater treatment technologies, such as AOPs and NF/RO
membranes, are particularly effective for removing PPCPs and EDCs. However, the
viability of these processes is tempered by residual management issues, including
transformation products and the discharge of concentrated brine streams. With any
type of oxidation process, including more conventional forms such as chlorination

250

D. Gerrity and S. Snyder

Table 17 TOrC concentrations (ng L1) in an alternative IPR treatment train

Secondary
Ultrafiltration
Ozone/H2O2
Contaminant
effluent
permeate
permeate

BAC effluent

Atenolol
Atorvastatin
Atrazine
Benzophenone
Carbamazepine
DEET
Diazepam
Diclofenac
Dilantin
Fluoxetine
Gemfibrozil
Meprobamate
Musk ketone
Naproxen
Phenytoin
Primidone
Sulfamethoxazole
TCEP
TCPP
Trimethoprim

<1.0
<0.5
<0.25
<50
<0.5
<1
<0.25
<0.5
<1.0
<0.5
<0.25
8.0
<25
<0.5
<1.0
0.66
<0.25
<10
<100
<0.25

860
20
0.83
160
300
860
3.0
98
310
72
65
830
50
13
310
230
1,100
480
2,200
460

790
8.1
1.1
130
310
920
3.0
79
110
46
60
840
<25
12
110
270
900
480
2,400
240

9.2
<0.5
0.39
<50
<0.5
14
<0.25
<0.5
3.0
<0.5
<0.25
97
<25
<0.5
3.0
11
5.7
370
1,100
<0.25

and ozonation, it is impractical to achieve complete mineralization (i.e., conversion

of organic molecules to water, mineral acids, and carbon dioxide). Short of complete
mineralization, oxidation processes will convert target compounds into transformation products that may or may not bear toxicological significance. For example,
Vanderford et al. [43] studied the chlorination of the antimicrobial compound
triclosan and noted the formation of mono- and dichlorinated by-products within
minutes of chlorine addition. Furthermore, Canosa et al. [74] noted that the chlorinated by-products of triclosan are more toxic than the parent compound.
Recently, researchers have begun to develop an understanding of the reaction
pathways and/or transformation products that are produced following advanced
treatment of waters containing PPCPs and EDCs. For example, the OH-induced
destruction of several compounds or classes of compounds, including DEET [53],
fibrate pharmaceuticals [51], fluoroquinolone antibiotics [75], and beta-lactam
antibiotics [76], has been documented. Research is currently underway to develop
models that can predict these transformation products, their reactivity, and toxicity
[77]. There is a balance that must be achieved between TOrC oxidation and the
formation of transformation products. It is possible that some transformation products may carry toxicological significance, thereby requiring utilities to (1) avoid
their formation or (2) implement additional treatment to remove them (e.g., BAC)
or convert them into a benign form.

Wastewater and Drinking Water Treatment Technologies

251

Membrane treatment is also affected by residual management issues.

Considering that RO typically requires initial feed pressures exceeding 100 psi, it
is evident that a substantial amount of energy is required to drive these processes.
As the membranes foul, additional energy is required to maintain sufficient water
production, and periodic chemical treatments may be required to clean the membranes.
Assuming a process flow rate of 10 MGD with 90% recovery [37], the RO system
would produce 9 MGD of high-quality permeate, but it would also generate 1 MGD
of concentrated brine containing five to tenfold greater concentrations of PPCPs and
EDCs, salts, organic matter, and other contaminants. Capital costs, operational
costs, and responsible disposal of the brine streamin addition to the loss of this
precious resourceare the major limitations facing widespread use of RO membranes for wastewater treatment and water reuse. At present, disposal of brine
streams is often restricted to coastal environments or to other WWTPs, limiting
areas in which this technology can be implemented.

Conclusions
PPCPs and EDCs are not truly emerging contaminants because the water and
wastewater communities have been aware of their presence in water supplies for
decades. However, recent advancements in scientific and analytical methodologies
in addition to increased media exposure have generated tremendous interest in this
field. Recent studies have monitored the concentrations of numerous TOrCs in
source water, drinking water, and wastewater to characterize the extent of contamination. Although TOrC concentrations in raw wastewater vary greatly and can often
exceed 1 mg L1, TOrCs are often present at very low concentrations (generally less
than 10 ng L1) in source water and finished drinking water. These low concentrations can be attributed to a combination of water and wastewater treatment efficacy
and natural attenuation in the environment.
As mentioned earlier, the presence or absence of any chemical in water is essentially a function of analytical detection capability. Therefore, the research communities must continue to study the potential impacts of TOrCs on human health and
aquatic environments. Until the scientific and regulatory communities reach consensus on the implications of PPCPs and EDCs in water, utilities will likely take a
proactive approach to (1) understand the extent of contamination in their systems,
(2) evaluate the efficacy of their current treatment strategy, and (3) determine
whether additional measures are necessary for TOrC mitigation. As discussed in
this chapter, some conventional water and wastewater treatment technologies are
quite effective for the removal and/or destruction of TOrCs, and there are a number
of advanced technologies that can be implemented for further TOrC reductions.
However, no single treatment process is capable of 100% TOrC removal so it is
important to balance the advantages and disadvantages of the various alternatives
while developing a multibarrier approach to TOrC mitigation.

252

D. Gerrity and S. Snyder

Acknowledgments These data were primarily collected during studies sponsored by the Water
Research Foundation (formerly American Water Works Association Research Foundation
(AwwaRF)) and the WateReuse Research Foundation (formerly WateReuse Foundation). The
Water Research Foundation sponsored Project #2758 entitled Evaluation of Conventional and
Advanced Water Treatment Processes to Remove Endocrine Disruptors and PharmaceuticallyActive Compounds and Project #3085 entitled Toxicological Relevance of EDCs and
Pharmaceuticals in Drinking Water. The WateReuse Research Foundation sponsored WRF-08-05
entitled Use of Ozone in Water Reclamation for Contaminant Oxidation.

References
1. Schueler FW (1946) Sex-hormonal action and chemical constitution. Science 103:221223
2. Sluczewski A, Roth P (1948) Effects of androgenic and estrogenic compounds on the experimental metamorphoses of amphibians. Gynecol Obstet 47:164176
3. Stumm-Zollinger E, Fair GM (1965) Biodegradation of steroid hormones. J Water Pollut
Control Fed 37:15061510
4. Hignite C, Azarnoff DL (1977) Drugs and drug metabolites as environmental contaminants:
chlorophenoxyisobutyrate and salicylic acid in sewage water effluent. Life Sci 20:337341
5. Snyder SA, Westerhoff P, Yoon Y, Sedlak DL (2003) Pharmaceuticals, personal care products, and
endocrine disruptors in water: implications for the water industry. Environ Eng Sci 20:449469
6. Snyder S, Vanderford B, Pearson R, Quinones O, Yoon Y (2003) Analytical methods used to
measure endocrine disrupting compounds in water. Pract Period Hazard Toxic Radioact Waste
Manage 7:224234
7. Vanderford BJ, Snyder SA (2006) Analysis of pharmaceuticals in water by isotope dilution
liquid chromatography/tandem mass spectrometry. Environ Sci Technol 40:73127320
8. Snyder SA, Vanderford BJ, Drewes J, Dickenson E, Snyder EM, Bruce GM, Pleus RC (2008)
State of knowledge of endocrine disruptors and pharmaceuticals in drinking water. AWWA
Research Foundation, IWA Publishing, London
9. Hollender J, Zimmermann SG, Koepke S, Krauss M, McArdell CS, Ort C, Singer H, Von
Gunten U, Siegrist H (2009) Elimination of organic micropollutants in a municipal wastewater
treatment plant upgraded with a full-scale post-ozonation followed by sand filtration. Environ
Sci Technol 43:78627869
10. Daughton CG, Ternes TA (1999) Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspect 107:907938
11. Benotti MJ, Trenholm RA, Vanderford BJ, Holady JC, Stanford BD, Snyder SA (2009)
Pharmaceuticals and endocrine disrupting compounds in U.S. drinking water. Environ Sci
Technol 43:597603
12. Ternes TA (1998) Occurrence of drugs in German sewage treatment plants and rivers. Water
Res 32:32453260
13. Fent K, Weston AA, Caminada D (2006) Ecotoxicology of human pharmaceuticals. Aquat
Toxicol 76:122159
14. Boxall ABA, Johnson P, Smith EJ, Sinclair CJ, Stutt E, Levy LS (2006) Uptake of veterinary
medicines from soils into plants. J Agric Food Chem 54:22882297
15. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton HT (2002)
Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams,
19992000: a national reconnaissance. Environ Sci Technol 36:12021211
16. Huber MM, Canonica S, Park GY, Von Gunten U (2003) Oxidation of pharmaceuticals during
ozonation and advanced oxidation processes. Environ Sci Technol 37:10161024
17. Kim SD, Cho J, Kim IS, Vanderford BJ, Snyder SA (2007) Occurrence and removal of pharmaceuticals and endocrine disruptors in South Korean surface, drinking, and waste waters.
Water Res 41:10131021

Wastewater and Drinking Water Treatment Technologies

253

18. Snyder SA, Wert EC, Lei H, Westerhoff P, Yoon Y (2007) Removal of EDCs and pharmaceuticals in drinking and reuse treatment processes. AWWA Research Foundation, IWA Publishing,
London
19. Snyder SA, Wert EC, Rexing DJ, Zegers RE, Drury DD (2006) Ozone oxidation of endocrine
disruptors and pharmaceuticals in surface water and wastewater. Ozone Sci Eng 28:445460
20. Ternes TA, Meisenheimer M, McDowell D, Sacher F, Brauch HJ, Haist-Gulde B, Preuss G,
Wilme U, Zulei-Seibert N (2002) Removal of pharmaceuticals during drinking water treatment. Environ Sci Technol 36:38553863
21. Westerhoff P, Yoon Y, Snyder S, Wert E (2005) Fate of endocrine-disruptor, pharmaceutical,
and personal care product chemicals during simulated drinking water treatment processes.
Environ Sci Technol 39:66496663
22. Carballa M, Omil F, Lema JM (2005) Removal of cosmetic ingredients and pharmaceuticals in
sewage primary treatment. Water Res 39:47904796
23. Jones OAH, Voulvoulis N, Lester JN (2002) Aquatic environmental assessment of the top 25
English prescription pharmaceuticals. Water Res 36:50135022
24. Qiu HC, Jun HW, McCall JW (1998) Pharmacokinetics, formulation, and safety of insect
repellent N, N-diethyl-3-methylbenzamide (DEET): a review. J Am Mosq Control Assoc
14:1227
25. Deborde M, Rabouan S, Duguet J-P, Legube B (2005) Kinetics of aqueous ozone-induced
oxidation of some endocrine disruptors. Environ Sci Technol 39:60866092
26. Noaksson E, Gustavsson B, Linderoth M, Zebuhr Y, Broman D, Balk L (2004) Gonad development and plasma steroid profiles by HRGC/HRMS during one reproductive cycle in reference
and leachate-exposed female perch (Perca fluviatilis). Toxicol Appl Pharmacol 195:247261
27. EC (2003) Risk assessment: Musk ketone. European Union risk assessment report. Final Draft
June 2003
28. Simonich SL, Begley WM, Debaere G, Eckhoff WS (2000) Trace analysis of fragrance materials in wastewater and treated wastewater. Environ Sci Technol 34:959965
29. Dodd MC, Buffle MO, Von Gunten U (2006) Oxidation of antibacterial molecules by aqueous
ozone: moiety-specific reaction kinetics and application to ozone-based wastewater treatment.
Environ Sci Technol 40:19691977
30. Loftsson T, Hreinsdottir D (2006) Determination of aqueous solubility by heating and equilibration: a technical note. AAPS PharmSciTech 7:E1E4
31. Snyder SA, Trenholm RA, Snyder EM, Bruce GM, Pleus RC, Hemming JDC (2008)
Toxicological relevance of EDCs and pharmaceuticals in drinking water. AWWA Research
Foundation, IWA Publishing, London
32. Letterman RD, Amirtharajah A (1999) Coagulation and flocculation. In: Letterman RD
(ed) Water quality and treatment: a handbook of community water supplies. McGraw-Hill,
New York
33. Cleasby JL, Logsdon GS (1999) Granular bed and precoat filtration. In: Letterman RD (ed) Water
quality and treatment: a handbook of community water supplies. McGraw-Hill, New York
34. Gregory R, Zebel TF (1999) Sedimentation and flotation. In: Letterman RD (ed) Water quality
and treatment: a handbook of community water supplies. McGraw-Hill, New York
35. Westerhoff P, Mezyk SP, Cooper WJ, Minakata D (2007) Electron pulse radiolysis determination of hydroxyl radical rate constants with Suwannee river fulvic acid and other dissolved
organic matter isolates. Environ Sci Technol 41:46404646
36. Ternes TA, Meisenheimer M, McDowell D, Sacher F, Brauch HJ, Gulde BH, Preuss G, Wilme
U, Seibert NZ (2002) Removal of pharmaceuticals during drinking water treatment. Environ
Sci Technol 36:38553863
37. Crittenden JC, Trussell RR, Hand DW, Howe KJ, Tchobanoglous G (2005) Water treatment:
principles and design, 2nd edn. Wiley, New Jersey
38. Snyder SA, Adham S, Redding AM, Cannon FS, DeCarolis J, Oppenheimer J, Wert EC, Yoon
Y (2007) Role of membranes and activated carbon in the removal of endocrine disruptors and
pharmaceuticals. Desalination 202:156181

254

D. Gerrity and S. Snyder

39. Routt JC, Mackey E, Whitby E, Connell G, Passantine L, Noak R (2007) Tracking utility disinfection over four decades of change: disinfection survey 2007preliminary summary. Water
Quality Technology Conference, American Water Works Associon, North Carolina.
40. Hua G, Reckhow DA (2008) DBP formation during chlorination and chloramination: effect of
reaction time, pH, dosage, and temperature. J Am Water Works Assoc 100:8295
41. Reckhow DA, Singer PC, Malcolm RL (1990) Chlorination of humic materialsby-product
formation and chemical interpretations. Environ Sci Technol 24:16551664
42. Deborde M, von Gunten U (2008) Reactions of chlorine with inorganic and organic compounds
during water treatmentkinetics and mechanisms: a critical review. Water Res 42:1351
43. Vanderford BJ, Mawhinney DB, Rosario-Ortiz FL, Snyder SA (2008) Real-time detection and
identification of aqueous chlorine transformation products using QTOF MS. Anal Chem
80:41934199
44. Langlais RT, Reckhow DA, Brink DR (1991) Ozone in water treatment: application and
engineering. Lewis Publishers, Chelsea
45. Wert EC, Rosario-Ortiz FL, Drury DD, Snyder SA (2007) Formation of oxidation byproducts
from ozonation of wastewater. Water Res 41:14811490
46. Wert EC, Rosario-Ortiz FL, Snyder SA (2009) Effect of ozone exposure on the oxidation of
trace organic contaminants in wastewater. Water Res 43:10051014
47. Benner J, Salhi E, Ternes T, von Gunten U (2008) Ozonation of reverse osmosis concentrate:
kinetics and efficiency of beta blocker oxidation. Water Res 42:30033012
48. Huber MM, Goebel A, Joss A, Hermann N, Loeffler D, McArdell CS, Ried A, Siegrist H,
Ternes TA, Von Gunten U (2005) Oxidation of pharmaceuticals during ozonation of municipal
wastewater effluents: a pilot study. Environ Sci Technol 39:42904299
49. Latch DE, Packer JL, Stender BL, VanOverbeke J, Arnold WA, McNeill K (2005) Aqueous
photochemistry of triclosan: formation of 2,4-dichlorophenol, 2,8-dichlorodibenzo-p-dioxin,
and oligomerization products. Environ Toxicol Chem 24:517525
50. Packer JL, Werner JJ, Latch DE, McNeill K, Arnold WA (2003) Photochemical fate of pharmaceuticals in the environment. Aquat Sci 65:342351
51. Razavi B, Song W, Cooper WJ, Greaves J, Jeong J (2009) Free-radical-induced oxidative and
reductive degradation of fibrate pharmaceuticals: kinetic studies and degradation mechanisms.
J Phys Chem A 113:12871294
52. Rosenfeldt EJ, Linden KG, Canonica S, von Gunten U (2006) Comparison of the efficiency of
{radical dot}OH radical formation during ozonation and the advanced oxidation processes
O3/H2O2 and UV/H2O2. Water Res 40:36953704
53. Song W, Cooper WJ, Peake BM, Mezyk SP, Nickelsen MG, OShea KE (2009) Free-radical
induced oxidative and reductive degradation of N, N-diethyl-meta-toluamide (DEET): kinetic
studies and degradation pathways. Water Res 43:635642
54. Suarez S, Dodd MC, Omil F, von Gunten U (2007) Kinetics of triclosan oxidation by aqueous
ozone and consequent loss of antibacterial activity: relevance to municipal wastewater ozonation. Water Res 41:24812490
55. Watts MJ, Linden KG (2009) Advanced oxidation kinetics of aqueous trialkyl phosphate flame
retardants and plasticizers. Environ Sci Technol 43:29372942
56. Tchobanoglous G, Burton FL, Stensel HD (2004) Wastewater engineering: treatment and
reuse, 4th edn. McGraw Hill, Boston
57. Snyder SA, Lei HX, Wert EC (2008) Removal of endocrine disruptors and pharmaceuticals
during water treatment. In: Aga DS (ed) Fate of pharmaceuticals in the environment and in
water treatment systems. CRC Press, New York
58. Snyder SA, Villeneuve DL, Snyder EM, Giesy JP (2001) Identification and quantification of
estrogen receptor agonists in wastewater effluents. Environ Sci Technol 35:36203625
59. Glassmeyer ST, Furlong ET, Kolpin DW, Cahill JD, Zaugg SD, Werner SL, Meyer MT, Kryak
DD (2005) Transport of chemical and microbial compounds from known wastewater discharges: potential for use as indicators of human fecal contamination. Environ Sci Technol
39:51575169

Wastewater and Drinking Water Treatment Technologies

255

60. Radjenovic J, Petrovic M, Barcelo D (2009) Fate and distribution of pharmaceuticals in wastewater and sewage sludge of the conventional activated sludge (CAS) and advanced membrane
bioreactor (MBR) treatment. Water Res 43:831841
61. Joss A, Keller E, Alder AC, Gobel A, McArdell CS, Ternes T, Siegrist H (2005) Removal of
pharmaceuticals and fragrances in biological wastewater treatment. Water Res 39:31373152
62. Clara M, Kreuzinger N, Strenn B, Gans O, Kroiss H (2005) The solids retention timea suitable design parameter to evaluate the capacity of wastewater treatment plants to remove
micropollutants. Water Res 39:97106
63. Ternes TA, Stumpf M, Mueller J, Haberer K, Wilken RD, Servos M (1999) Behavior and
occurrence of estrogens in municipal sewage treatment plantsI. Investigations in Germany,
Canada and Brazil. Sci Total Environ 225:8190
64. Heberer T (2002) Occurrence, fate, and removal of pharmaceutical residues in the aquatic
environment: a review of recent research data. Toxicol Lett 131:517
65. Bellona C, Drewes JE, Oelker G, Luna J, Filteau G, Amy G (2008) Comparing nanofiltration
and reverse osmosis for drinking water augmentation. J Am Water Works Assoc 100:102116
66. Singer PC, Reckhow DA (1999) Chemical oxidation. In: Letterman RD (ed) Water quality and
treatment: a handbook of community water supplies. McGraw-Hill, New York
67. Benotti MJ, Standford BD, Wert EC, Snyder SA (2009) Evaluation of a photocatalytic reactor
membrane pilot system for the removal of pharmaceuticals and endocrine disrupting compounds from water. Water Res 43(6):15131522
68. Gerrity DW, Stanford BD, Trenholm RA, Snyder SA (2010) An evaluation of a pilot-scale
nonthermal plasma advanced oxidation process for trace contaminant degradation. Water Res
44(2):493504
69. Rosenfeldt EJ, Linden KG (2004) Degradation of endocrine disrupting chemicals bisphenol A,
ethynyl estradiol, and estradiol during UV photolysis and advanced oxidation processes.
Environ Sci Technol 38:54765483
70. Rosario-Ortiz FL, Mezyk SP, Wert EC, Doud DFR, Singh MK, Xin M, Baik S, Snyder SA
(2008) Effect of ozone oxidation on the molecular and kinetic properties of effluent organic
matter. J Adv Oxid Technol 11:529535
71. Beltran FJ (2004) Ozone reaction kinetics for water and wastewater systems. CRC Press, Boca
Raton
72. Reungoat J, Macova M, Escher BI, Carswell S, Mueller JF, Keller J (2010) Removal of
micropollutants and reduction of biological activity in a full scale reclamation plant using
ozonation and activated carbon filtration. Water Res 44:625637
73. Stalter D, Magdeburg A, Weil M, Knacker T, Oehlmann J (2010) Toxication or detoxication?
In vivo toxicity assessment of ozonation as advanced wastewater treatment with the rainbow
trout. Water Res 44:439448
74. Canosa P, Morales S, Rodriguez I, Rubi E, Cela R, Gomez M (2005) Aquatic degradation of
triclosan and formation of toxic chlorophenols in presence of low concentrations of free chlorine. Anal Bioanal Chem 383:11191126
75. Santoke H, Song WH, Cooper WJ, Greaves J, Miller GE (2009) Free-radical-induced oxidative and reductive degradation of fluoroquinolone pharmaceuticals: kinetic studies and degradation mechanism. J Phys Chem A 113:78467851
76. Song WH, Chen WS, Cooper WJ, Greaves J, Miller GE (2008) Free-radical destruction of
beta-lactam antibiotics in aqueous solution. J Phys Chem A 112:74117417
77. Lei HX, Snyder SA (2007) 3D QSPR models for the removal of trace organic contaminants by
ozone and free chlorine. Water Res 41:40514060

Pharmaceutical Take Back Programs

Kati I. Stoddard and Duane B. Huggett

Introduction
Prior to September 2010 the US national policy for the proper disposal of
pharmaceuticals was limited to published guidance provided by several federal agencies; however, on September 25, 2010 the US Drug Enforcement Administration
(DEA) established the National Take Back Initiative. This program is designed to
provide citizens an opportunity to safely dispose of medications they no longer need
or want. In 2010 two of these DEA events were held during which 309 tons of medications were collected at thousands of take back sites across the country. Additionally,
the Safe and Secure Drug Disposal Act of 2010 signed by President Obama on October
12, 2010 provided the means for the Controlled Substance Act (CSA) to be amended
to allow the DEA to develop a procedure for individuals to safely dispose of their
unwanted medications, including medications considered controlled under the CSA
[1]. This legislation will ensure future DEA events and other take back events are
legally able to dispose of controlled medications, which is critical as these medications have a high potential for diversion or abuse.
For individuals unable to participate in the national DEA events the DEA and
other national agencies advise individuals to dispose of their unused, unneeded, or
expired pharmaceuticals by removing them from their original containers, mixing
the medications with a deterring substance, such as coffee grounds or used cat litter,
placing the mixture in nondescript containers like sealable bags, and then placing
the bags in the household garbage. Flushing is only recommended when the label
or patient information for the prescription drug specifically calls for such a disposal method [2]. Beyond this basic medication disposal information, several

K.I. Stoddard D.B. Huggett (*)

Department of Biological Sciences, University of North Texas,
1155 Union Circle #305220, Denton, TX 76203, USA
e-mail: [email protected]; [email protected]
B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3_10, Springer Science+Business Media, LLC 2012

257

258

K.I. Stoddard and D.B. Huggett

environmental stewardship organizations and US federal agencies, such as the

Environmental Protection Agency (EPA) and the Fish and Wildlife Service (FWS),
provide information to the general public on the impacts pharmaceuticals can have
on the environment and provide suggestions of actions individual can take to minimize this growing environmental threat.
Despite the concerted efforts of many agencies and organizations, the message
on proper disposal methods for pharmaceuticals appears to not be entirely effective
at reaching target audiences. Studies have found that the most frequent methods of
disposal of pharmaceuticals are the sink, toilet, and trash [37]. The preference
among these methods appears to be significantly influenced by age with older
respondents relying on the sink or toilet as a disposal method and younger respondents relying on household trash [7]. Even with strict observance to the DEA guidance for medication in household trash disposal, residues from pharmaceuticals
disposed of in sanitary landfills can accumulate in leachate and escape the facilities
to nearby receiving waters [811]. Additionally, residuals of medicines are also
reaching surface waters through postconsumer excretion of pharmaceuticals and
their metabolites in human and animal urine. Studies have found detectable concentrations of pharmaceuticals ranging from parts per trillion (ppt) to low end parts
per billion (ppb) in both wastewater effluent [12] and drinking water [1316].
Though there are concerted research efforts aimed at investigating the impact of
pharmaceuticals on the environment, these efforts are greatly complicated by the
conglomeration of pharmaceuticals and other man-made chemical products being
released into the environment. Beyond considering the impact of mixtures of pharmaceuticals with other man-made chemicals and products on the environment,
research is further complicated by the concept of low-dose toxicity, which stresses
that levels of pharmaceuticals below what are traditionally considered harmless
may have subtle or even pronounced acute and chronic toxic effects on aquatic
organisms that are living in ecosystems that receive a continuous stream of these
chemicals via wastewater effluent or industrial outfalls [17].
Through the dedicated efforts of toxicological and environmental research, an
ever increasing volume of scientific evidence is building support for the concept that
pharmaceutical residues in the environment are having a deleterious impact on the
quality of the aquatic environment. Though knowledge and complete understanding
of this emerging environmental threat may not be widespread among the general lay
population, awareness of this issue is increasing steadily in the American and global
population through news and media coverage and public health outreach campaigns.
As public consciousness of the environmental implications of pharmaceuticals in
water resources increases, so too has the public increased their expressed interest in
policies and programs to mitigate some of the documented negative environmental
impacts caused by pharmaceuticals in the environment.
Events such as the newly developed DEA Initiative are one promising type of
program that have been implemented as a means to mitigate or minimize the impacts
of pharmaceuticals on natural resources. Within the USA, pharmaceutical take back
programs have been implemented at local, state, and, with the advent of the DEA
Initiative, national levels. There is a vast array of resources available now that

Pharmaceutical Take Back Programs

259

provide information on the various different local, national, and international

programs that have been instituted with success and community support [18].
The purpose of this chapter is provide a comprehensive overview of the concept of
pharmaceutical take back programs by an examining integral components of the
programs such as typical objectives, methods for evaluating success, gaps or weaknesses of many take back programs, and potential or realized obstacles facing take
back programs. These concepts will be further explored through the examination of
a variety of successful take back programs and their results.

Objectives of Take Back Programs

Take back programs are gaining considerable popularity within communities in the
USA and European nations as the public becomes increasingly aware of their individual and collective impact on the environment. However, these programs rarely are
single minded in their focus on environmental protection and most often these
programs strive to address several other health issues resulting from excess pharmaceuticals. The following section will present a discussion on how some take back
programs have expanded their objectives beyond that of protecting the environment
from contamination associated with pharmaceutical products to include objectives
that address individual and public health issues.
One public health issue that many take back programs have attempted to address is
the opportunity for accidental poisoning due to excess and unused pharmaceuticals
being stored in the home. The following discussion will focus on the data available
on poisonings and how they pertain to take back programs.
The American Association of Poison Control Centers (AAPCC) releases annual
reports on the data submitted by local Poison Control Centers (PCC) to the National
Poison Data System (NPDS). The NPDS is a valuable tool for researchers, policy
makers, and many others as it is the only comprehensive poison surveillance system in
the USA and because it collects real-time data from the 61 PCCs across the nation.
According to AAPCC the total number of human exposures to poisons in 2009
was 2,479,355 with the majority of these exposures (82.4%) being unintentional.
A particularly revealing trend disclosed in the 2010 report was that the majority of
fatalities reported in children 5 years old and younger were unintentional, whereas
most fatalities in adults (20 years or older) were intentional. Additionally, of all the
human exposures reported in 2009, 93.8% were exposures occurring at a residence.
Unfortunately the data are not partitioned to provide a frequency of the types of
poisons causing death in these age groups or by the location of incidence (residence
or other); however, the report does list the top 25 substance categories associated
with the largest number of fatalities. This listing indicates that sedatives, hypnotics,
and antipsychotics rank as the number one substance category leading to poisonrelated fatalities, with cardiovascular drugs, opioids, acetaminophen combinations, and
acetaminophen alone following sequentially [19]. As these are all medications, it
seems reasonable to presume the fatalities caused by these substances likely

260

K.I. Stoddard and D.B. Huggett

occurred in the home. Clearly there is strong evidence to show medication home
storage poses a significant poison risk for both children and adults.
The 2010 AAPCC report also revealed that among the 2,043,155 unintentional
poisoning in 2009, 276,694 (11.2%) were attributed to therapeutic error and 125,742
(5.1%) were attributed to misuse. Specific therapeutic errors resulting in poisonings
included unintentional double-dosing (31.4%), taking or being administered the
incorrect medication (14.7%), taking or being given multiple doses within a shorter
time period than recommended (9.6%), and accidental exposure to a medication
belonging to someone else (9.0%). The AAPCC reports the number of poisonrelated fatalities in 2009 as 1,158, and although children ranging in age from
newborn to under 6 years old were involved in the majority of 2009 exposures, this
age group constituted only 1.8% of the poison-related fatalities. The majority of
individuals reported as dying as a result of poisoning were between the ages of 20
and 59, with this age group comprising 70.9% of all poison-related fatalities [19].
The summary statistics provided in the 2010 AAPCC report are significant in our
review of take back programs for several reasons. To begin with, one may speculate
from the therapeutic error statistics presented in the 2010 AAPCC report that the
presence of excess or expired medications in the home may be one of the leading
causes of misuse of medication, ultimately resulting in avoidable deaths caused by
accidental poisoning. Though the statistics presented by the 2010 AAPCC report on
the frequency of death due to poisoning for age group do not indicate the type of
poison, an argument can still be rationally made that the presence of excess and
expired medicines in the home is a threat to all members of the household. This may
be especially true for adults between 20 and 59, who may have a tendency to selfdiagnose, and for young children, whose curiosity and playfulness can quickly lead
to danger if medications in bottles or containers they can open are left in places
accessible to them. Additionally, although the fatalities in people aged 60 to over 90
years old only account for 20.4% of the poison-related fatalities [19], concern can
still be voiced for this age group as they tend to have more medicines prescribed to
them, age-related memory loss is common among this group, and diminishing
eyesight may hinder reading small labels on medication bottles.
In addition to reducing the occurrence and opportunity for accidental poisoning
and misuse of pharmaceuticals, some take back programs have also been launched
to address another public health problem resulting from the presence of excess pharmaceuticals in the home. An alarming trend that is gaining attention in the US media
and which is the subject of numerous national studies and programs is the growing
popularity of prescription and over-the-counter (OTC) drug abuse among teenagers
and young adults. The common term for this dangerous behavior is pharming.
The nonprofit organization The Partnership for a Drug-Free America (Partnership)
details in a tracking study that one in five teens (19% or 4.5 million) reports abusing
prescription medication to get high and one in ten teens (10% or 2.4 million) reports
abusing cough medicine for the same purpose. The abuse of OTC medications and
prescription drugs is so prevalent now that the Partnership refers to the current
generation of teenagers as Generation Rx [20].
The Partnerships report also reveals that the abuse of prescription and OTC
medications is now as prevalent as or more prevalent than illegal drugs such as

Pharmaceutical Take Back Programs

261

Ecstasy, cocaine/crack, methamphetamine, and heroin [20]. This trend is also

supported by the 2010 National Survey on Drug Use and Health (NSDUH) which
reported nonmedical use of prescription drugs was second in US drug abuse cases
only to marijuana. Comparing data from other NSDUH studies dating back to 2002,
the current NSDUH highlights that although there has not been a significant overall
increase in abuse of prescription pain relievers, there are other troubling signs that
indicate there is a significant prescription drug abuse problem in the US. These
indicators include increases in individuals dependent on pain relievers, increases in
the number of people seeking substance abuse treatment, and increases in emergency room visits attributed to prescription drug abuse [21]. One possible explanation for this new social dilemma is a false perception Americas teenagers and other
medicine abusers have for the safety of prescribed drugs. The Partnerships study
found that two out of five teens believe that prescription drugs are safer than illegal drugs even if they are not prescribed to them. Other prevalent misconceptions
about prescription drugs held by teenagers surveyed were that there was no harm in
occasionally using prescription medication without a prescription and that prescription pain medications, even those not prescribed by a doctor, where not addictive.
The Partnerships study also found that teenagers believed that widespread availability
and easy access to medications are a leading cause of the pharming problem. Three out
of five teens reported that they could easily steal prescriptions from their parents medicine cabinets. Additional views held by teenagers surveyed were that it was easy to
acquire other peoples medicine or that prescribed pain medicine was widely available,
and that prescribed medicines, when purchased illegally, were cheap [20].
Though not specifically reported on by the Partnership, it is reasonable to presume
from the youths survey responses to questions on the accessibility of prescription
and OTC medication that there is some degree of black market buying and selling
of these medications. Drug abuse, whether it be with pharmaceuticals or illegal
drugs, is a trend that has the potential to bring about disastrous future effects for
those involved as individuals and our nation as a whole. Drug use and drug peddling
are far from the foundations of a productive member of society, and continued
participation in these activities threatens much more than the environment or water
resources; drug dealing threatens lives. Additionally, beyond the public health issue
of American teenagers abusing and potential dealing prescription medications, there
is clearly an economic cost to this public health problem in the form of black market
sales. Developing an effective method for estimating the economic costs of black
market transactions of prescribed and OTC medications could provide very useful
information necessary for estimating the additional costs and benefits of take back
programs beyond those aimed at reducing environmental risks.
The presence of stored prescription and OTC medications in the home clearly
has serious implications beyond the accidental misuse by adults or accidental
poisoning of children as discussed previously. In addition to increased efforts to
educate both parents and teenagers on the dangers of abusing prescription medications,
reducing and/or eliminating teenagers access to OTC and prescription medications,
the source commonly relied upon by teenagers for their drug abuse, could clearly
help reduce the problem. Though not a commonly stated objective of take back
programs, pharmaceutical take back programs have the opportunity to play a critical

262

K.I. Stoddard and D.B. Huggett

role in solving this troubling social drug use trend by providing a safe and effective
means for adults to dispose of their unused or unwanted pharmaceuticals that are
usually stored in their homes. Though the campaigns and organizations designed for
addressing drug abuse among the youth of this nation are not usually thought of in
conjunction with pharmaceutical take back programs and environmental awareness
and action programs, the potential for these philanthropic and environmental missions
to unite to improve the health of our nations population and environment is promising and could prove to be a very productive partnership.
Beyond providing a possible solution to the public health issues that arise from
excess pharmaceuticals in the home, another objective take back program may also
address, either purposefully or inadvertently, is improving medicine management
strategies through improved knowledge and data necessary to investigate the costs
and risks individuals and society are incurring from mismanagement of OTC and
prescription medications. The following discussion will highlight how various elements of take back programs can address potential costs or losses in benefits society
may be experiencing. Some of these losses may be hidden or unrealized by most
people as they may not impact consumers directly or they may not be self-evident.
Despite the seeming transparency of these costs and losses of benefits, it is possible
that they may be a key factor in estimating the benefits, costs, and risks associated
with pharmaceuticals in the environment and water.
One very effective means of investigating hidden costs or losses of benefits
resulting from excess pharmaceuticals is to collect and analyze data gathered from
pharmaceutical take back events. This has been accomplished in many individual
take back programs through participation surveys. These surveys are designed to
obtain data necessary to evaluate the overall effectiveness of the take back program
and, in some cases, to estimate the various costs or losses of benefits associated with
the accumulation of excess pharmaceuticals in homes and/or releases of these products into the environment. Survey questions usually will inquire as to demographic
information, name or type (drug category or class) of medication being returned,
reason for return of medication, participant satisfaction or perceptions of the take
back program, and other event specific information. This type of information, along
participation rates for a particular take back event, can then be analyzed to reveal
significant statistics and trends in pharmaceutical use and disposal patterns and consumer behavior as it relates to pharmaceuticals. For example, the value of returned
medication can be determined by researching the market value of the medication
and multiplying it by the unused portion returned. Singularly, this value may only
be significant to the person returning the medicine; however, when this method is
used to estimate the value of all medications returned at a take back event, it provides valuable insight into the costs incurred by society by wasted pharmaceutical
resources. The market value of wasted medications as well as other information that
may be available from participation surveys is essential in meeting the objective of
investigating costs or losses of benefits that may be burdening society as a result of
poor medicine management strategies.
An additional objective that has been included in a handful of take back program
is protection of patient privacy. As the majority of prescribed medicines are labeled
with information specific to the patient that may be sensitive, simply throwing

Pharmaceutical Take Back Programs

263

expired or unused medications in the trash may compromise the personal security of
the patient. Just as privacy information can be stolen from mail in household trash,
so too can private health information be pilfered from thrown-out medicine bottles.
Pharmaceutical take back programs offer security of identity and health information
for individuals because medication packaging and bottles are collected and secured
from general public access at take back events.
Take back programs are gaining attention in the USA and, as has been discussed,
they have far reaching objectives that cover a wide range of public health issues in
addition to the goal of environmental protection through proper disposal of pharmaceuticals. However, though the objectives of preventing accidental poisoning, combating teenage medication abuse, improving medication management strategies,
and protecting patient privacy are admirable and worthwhile goals, the number of
programs that realize the potential to address these issues is still very limited. Yet
take back programs as a whole are still in their youth and as more communities,
regulators, and social and environmental activists become aware of these programs,
it is very likely these individuals and groups will also come to realize that these
programs cannot only provide an effective means of achieving safe disposal of medications, but they can also address a variety of public health issues.

Measuring Success
Of the many examples of current and past pharmaceutical take back programs
reviewed for this work, relatively few included quantifiable and defined means of
measuring the success of the program beyond calculating the amount of prescription and OTC medications collected and possibly participation at a take back event.
Another useful but inconsistently reported data set is the monetary value of returned
medications. Due to widely varying take back programs, unique in terms of their
target audience, participation levels, frequency, duration, and several other factors,
it is difficult to compare and evaluate individual programs based on data relating to
the volume and value of returned medications and overall participation. However,
when this information is available it can be very useful for understanding and assessing, to a limited degree, the success and the outcomes of a particular program.
As an example, Washington state operates its Unwanted Medicine Return
Program Pharmaceuticals from Households: A Return Mechanism (PH:ARM) yearround and reports having collected and disposed of 35,000 pounds of pharmaceuticals since 2009 and when it began operating in October of 2006 [22] while the Bay
Area Pollution Prevention Group (BAPPG) in Chicago, Illinois, reports collecting
over 3 tons of expired and unused medicines between 2004 and 2007 during annual
single day events [18]. Though both programs provide the duration and volume of
medications collected by their programs, the subtle details such as exactly what they
accepted, how many locations they established for collection, and many other
unique factors of each program limit the ability to determine if one program was
more successful than another and limits the ability of evaluating how successful
each individual program was in its own sphere.

264

K.I. Stoddard and D.B. Huggett

Though available information on the success of individual take back programs

conducted throughout the USA and abroad is highly variable, a list of several take back
programs that have made the details and outcomes of their programs widely available
to the public has been compiled in Appendix A. This appendix provides a review of
many current and past take back programs within the USA and other nations and,
when available, information on the outcome of the project (e.g., volume collected,
value of medications collected, and cost of program). Perhaps as the pressing need for
take back programs becomes more apparent and the concept gains national momentum as a means for managing excess and unused pharmaceuticals, more programs will
begin to realize the importance of identifying and quantifying measures of success for
their programs. Examples of some questions take back program operators may ask
themselves as they expand the scope of evaluating the success of their programs
include: Are my programs marketing and advertising methods reaching my target
audience?; Is there an especially needful subpopulation within our area that is not
participating due to lack of knowledge of the program?; and Are participants of our
program better educated about the issues surrounding excess and unused medications
following their participation in the program? Improved measures of success would
be very helpful to take back programs because if definitive measures of success are not
identified it is difficult to assess where improvement can be made so that a program
can be modified or expanded to be more effective and worthwhile.

Identifying Gaps or Weaknesses in Take Back Programs

Although quantifying the volume and value of medications collected does provide
some indication as to the success of a program, the key motivating factors driving
many take back programs can include various other reasons beyond collecting
unused and excess medications. These motivating factors include issues discussed
previously such as avoiding accidental poisonings and abuse of prescription drugs.
While these endeavors are assuredly beneficial, the authors believe the true potential
of take back programs has yet to be fully realized due to several gaps or weaknesses
in the basic model of take back programs. This basic model consists of organized take
back events that simply collect and dispose of medications. Beyond this, some take back
programs survey participants, however it does not appear, based on an intensive literature review, that many take back programs critically analyze and report their findings
from these surveys. As will be discussed, take back programs could be enhanced by
expanding their program scope to address some of the gaps in the program model.

Scientific Justification
One significant gap of take back programs is that of scientific justification. The
question that still remains after innumerable take back programs have been conducted is do these programs improve or mitigate the current state of the water

Pharmaceutical Take Back Programs

265

resources and the environment in regard to pharmaceutical wastes and residues?

Furthermore, are there certain classes of medications that are more harmful to the
environment than others and therefore should be targeted as key classes of medications during take back events?
One promising method that may be used to estimate potential harmful impacts of
pharmaceuticals to fish was developed by a team of researchers at Pfizer Global
Research and Development. Their model is based on the fact that there are similar
enzyme and receptor systems in both fish and mammals. Based on these similarities, the model can use existing data from toxicological and pharmacological studies
on mammals to predict pharmacological responses in fish. The model compares the
measured human therapeutic plasma concentration of a medication (HTPC) to the
predicted steady state plasma concentration (FSSPC) in fish with the result being an
effect ratio (ER). In this model there is an inverse relationship between the ER and
the potential for a pharmacological response in fish, meaning the lower the ER the
greater the likelihood there will be a pharmacological response in fish and thus the
more likely that additional testing needs to be conducted to determine if the medication poses a threat to fish [23, 24]. As it can be time consuming and very difficult to
obtain data on environmental responses to all of the human pharmaceuticals on the
market, this model overcomes that hurdle by capitalizing on the vast amount of
mammalian pharmacological data available and applying it in a new and innovative
way. Using this model, it may be possible to determine if a certain class of pharmaceuticals poses a greater threat to fish than another class of pharmaceuticals. The
information gained from application of this model could make take back programs
more efficient and cost effective because it could educate take back program organizers as to which medications pose the greatest environmental risks and therefore
should be the ones they target for return and proper disposal.
The presence, potential impacts, and known detrimental impacts of pharmaceuticals in aquatic environments have been researched and publicized by the professional
science community as well as the mainstream media. However, despite the widespread coverage of this, to date, the authors are unaware of any research projects or
other efforts designed to investigate if water quality or aquatic life in areas instituting
take back programs has improved as a result of the program. One worthwhile option
for filling this gap of information would be to conduct biological and chemical monitoring of streams and other receiving water bodies in areas where take back programs
are in place. Biological monitoring, or biomonitoring as it is often referred to as, is a
well-established environmental monitoring method which relies on the use of living
organisms as sensors for water quality surveillance. Chemical monitoring utilizes
proven water quality techniques and instruments to measure the chemical characteristics of the water which may be altered by pharmaceutical loads. Biological and chemical monitoring can provide the critical data necessary to determine if water quality and
aquatic life have improved due to a presumable decrease in the pharmaceutical load
released into the environment as a result of a take back program. Unfortunately to the
knowledge of the authors, the incorporation of this type of analytical investigation
before, during, and after a take back event to provide scientific justification for these
programs has not been attempted or even suggested by any take back program.

266

K.I. Stoddard and D.B. Huggett

Risk Perception
Risk perception is generally regarded as the intuitive assessments of risks people
face based on a variety of information sources, personal experiences, and an assortment of other contributing factors. Risk perceptions are naturally developed by
individuals; however, collectively groups may also cohesively form their own distinct risk perceptions, giving rise to the concept of public risk perception. Though
the relationship between risk perception and behavior is very complicated, it has
been shown that subjective risk perceptions may influence the actions of individuals.
However despite the seemingly obvious impact risk perception is likely to have a
pharmaceutical disposal behavior, the authors are unaware of any take back program incorporating risk perception into their program. To address this weakness in
take back programs, this section will present some of the key concepts and areas of
interest of risk perception, discuss significant research findings in this field, and
discuss the applicability of these concepts and findings to pharmaceutical take
back programs.
One reoccurring criticism of industry, government, scientists, and other experts
with professional knowledge of risks is the wide discrepancy that exists between
their objective assessments of risk and the general publics perception of risk. This
disparity may be attributable in part to the complexity of factors and inputs that
individuals process as they develop their personal risk perceptions. Influential factors
may include such characteristics as the control individuals have over the risk, their
willingness to engage or be subject to the risk, and their knowledge or understanding
of the risk [25, 26]. Other inputs that influence risk perceptions include, but are not
limited to, probabilities, biased news reports, confusing personal experiences, and
apprehension over gambles encountered in daily life. The complex nature of these
factors and inputs can confuse individuals and cause them to deny uncertainty, to
over- or underestimate risks, and to assertively hold opinions relating to risks that
they are unable to defend [27, 28].
Another possible cause for the disparity between public risk perception and
experts objective risk assessments is that there seem to be varying definitions of
risk. Studies have shown that lay individuals vary in how they define risks. When
defining risks, lay individuals may include the rich array of risk characteristics that
they relied upon for the formation of their risk perceptions. In addition to those
previously mentioned, these risk characteristics may include concepts such as the
potential for catastrophe and the potential for impacts to future generations. This is
a dramatic departure from experts definition of risk which typically relies on a
single, defined, quantifiable measure of risk such as annual fatalities [25]. Due to
the complex nature of risk perception it is unlikely that there is a single explanation
for the cause of the divergence in lay individuals subjective risk perception and
experts objective assessments of risk; however, given the current knowledge of
risk perception formation, it is likely that the inconsistent definitions of risk and the
complexity of variables which influence risk perceptions are contributing factors to
this anomaly.

Pharmaceutical Take Back Programs

267

Another characteristic of risk perception that has been consistently observed in

risk research studies is that the lay public generally believes the current level of risk
for most activities is objectionably high; indicating the perceived level of risk is
beyond that of the desired level of risk and that regulatory efforts to maintain these
risks are not producing acceptable results. Despite the public being dissatisfied with
the level of risk management provided by regulatory mechanisms, research shows
that people are willing to accept higher levels of risk for activities they may consider
beneficial. Some of the characteristics shown to influence this trade-off of high risk
for perceived benefit include voluntariness, familiarity, control, potential for catastrophe, knowledge of the risk, and fairness; however, no single characteristic has
proved to be the sole determining factor [29, 30]. These findings are significant
because government officials and other professionals involved in managing hazards
could very likely increase the overall effectiveness of their risk management strategies if they addressed some of these factors in their public policies and educational
campaigns.
In addition to investigating factors influencing risk perception and possible
explanations for the divergence between subjective and objective risk assessments,
risk research has also focused on how risk perceptions influence behavior. However
despite the seemingly obvious connection between risk perception and behavior and
the wide array of studies that have investigated the relationship between these two
elements, for many reasons results from these studies vary widely and are difficult
to compare. One possible explanation for this nonconformity of results is that most
studies focus their efforts on a single risk and due to the unique characteristic of
individual risks, the risk perceptions and subsequent behavioral responses to specific
risks are difficult to compare across a wide spectrum of different risks. Investigation
methods also vary widely which can make it difficult to compare studies and results.
Another possible explanation for variation in results of these studies is that individuals usually have multiple motivations for engaging in certain actions, with risk
perception being only one of many contributing factors. Regardless of the reason
behind the variation in results, there are several recent risk perception studies that
provide valuable insight into the complex relationship between risk perception and
behavior and which may be very useful when considering the potential impact risk
perception may have on pharmaceutical disposal behavior and participation in pharmaceutical take back programs.
For an example of a study investigating risk perception and behavior consider a
United Kingdom (UK) survey of students which found that knowledge and risk
perception had very little influence over behavior [31]. It is however important to
note that this study was limited to a very selective subpopulation of university students with health or environmental backgrounds at two universities in the UK and
the majority of risks respondents were questioned on were voluntary risks that have
specific and direct consequences to the person engaging in the behavior (e.g., smoking, alcohol use, and illegal drugs). Similarly, a study investigating risk perception
and smoking behavior in Swedish teenagers found that risk perception of lung
cancer did not affect the number of cigarettes smoked [32]. This finding entirely
contradicts another smoking study that found risk perception of lung cancer to be a

268

K.I. Stoddard and D.B. Huggett

significant factor in the number of cigarettes smoked [33]. The Swedish study speculates that limiting the study to lung cancer risks may have introduced a bias that
would explain this divergence [32]. The lack of evidence to support a link between
risk perception and behavior in the UK and Swedish studies may also be because
these studies examined risks where the individuals choices had a direct impact on
their health and it is quite possible that human and environmental health risks
resulting from excess and unused pharmaceuticals may be very different due to the
greater separation between the behavior and consequence and due to the appeal for
social and environmental responsibility on the part of the individual [34].
To demonstrate the vastly different conclusions of some risk research studies
consider also the far different conclusion of Jakus et al. [2009] compared to the
previously mentioned survey of UK students and the Swedish smoking survey.
Jakus et al. [2009] found that for respondents living in areas with arsenic-contaminated drinking water, perceived risk was a statistically significant factor in the
decision of how much bottled water to purchase [35]. Another significant factor in
risk decision making may be an individuals knowledge of a particular problem.
A Swedish study investigating the concept of environmental awareness as a factor
in decision making found that individuals accounted for environmental factors
such as air pollution when making decisions about personal car use [36]. Though
the risks of air pollution and excess and unused pharmaceuticals may be different in
many ways, similar characteristics between these two risks such as delayed effects,
noncatastrophic effects, and potential impacts to future generations may make the
findings of these studies very applicable to the public health and environmental
problem of improper pharmaceutical disposal behavior. Reviewing the studies
highlighted here, it seems that one significant factor affecting the strength of the
relationship between risk perception and behavior may be the specific risk itself. As
such, take back programs aspiring to address the weakness of including risk perception in their program structure would be wise to consider studies addressing the
risks associated with the improper disposal or access to excess and unused pharmaceuticals or studies of risks with similar characteristics.
One study is particularly applicable to addressing the risk perception weakness
in the majority of take back programs. Bound et al. [2006] investigated the relationship between choice of disposal method for pharmaceuticals and risk perception
and found that there was no definite correlation between these two factors. The
authors of this study speculated that respondents may not feel that the risks posed
by pharmaceuticals in the environment are great enough or their choice of disposal
method was significant enough to make a difference, thus the individuals surveyed
did not have enough incentive to change their disposal behavior. Despite the fact
that a direct link between disposal behavior and risk perception was not found, this
survey did reveal very interesting risk perceptions about the impact of pharmaceuticals on both personal health and the environment which may provide valuable
insight for future take back programs. For example, the majority of survey participants indicated they strongly agreed or simply agreed that pharmaceuticals used
inappropriately could be potentially harmful to their personal health and more than
half agreed that improper disposal of pharmaceuticals could threaten fish or plants.

Pharmaceutical Take Back Programs

269

There was also a high degree of uncertainty associated with the impacts to the environment. This was speculated to be because survey participants were likely less
knowledgeable about toxicology and environmental processes. Respondents also
indicated they perceived a lower threat from medications that were more familiar to
them, such as pain killers and antihistamines, compared to less well-known medications such as antiepileptic medication and lipid regulators. OTC medications and
commonly prescribed drugs are often viewed as less potent and therefore less of a
risk to the environment. This is very likely because familiarity tends to make risks
either more acceptable to individuals and/or causes individuals to underestimate
risks [34]. The perception that nonprescription drugs are less potent and therefore
less harmful to the environment was also evident in a Canadian study that found that
the percentage of respondents who believed OTC pharmaceuticals needed an appropriate disposal method was 81% while the percentage believing that unused and
expired prescribed pharmaceuticals needed to be disposed of in an appropriate way
was 90% [34, 37].
Though it may be difficult to draw a definitive link between risk perception in
regard to excess and unused pharmaceuticals and choice of disposal method, it is
important to consider what other motivations may influence disposal behavior.
As there have been innumerable take back programs launched, it is clear that there
are effective motivations for individuals to participate. In many cases, individuals
often participate in environmental stewardship events, and other activities that are
altruistic in nature, because they can see and understand the benefit of their individual efforts [34, 38]. Individuals may derive personal satisfaction from participating in such environmentally conscious programs like take back programs or they
may have a desire make a small personal change in their behavior for the health of
the public, themselves, their children, or the environment. Some experts suggest
that perhaps the risk to the environment is not well understood by the general public
or compared to other risks such as air pollution and financial instability, the risk
posed by improper disposal of the pharmaceuticals is not great enough to merit a
change in disposal behavior [34]. Additionally, when contemplating participating in
programs such as pharmaceutical take back events, concerns about issues people
recognize and understand better, such as their own health or public health, may be
stronger motivators than environmental problems, which they may not understand
well or acknowledge as a significant problem.
As has been illustrated here, risk perception is a complicated concept, shifting and
changing with specific risks and not entirely understood yet. However the complexity of risk perception is not a viable reason for excluding it as a key element in take
back programs. Due to the need to motivate local citizens to participate in a take back
program, the inclusion of risk perception in these programs has the potential to make
significant improvements in the success of an individual program. To bridge this gap
therefore it would be advisable for take back programs to evaluate motivations of
individuals participating in the program, investigate risk perceptions of their target
audience, and then design their programs to account for the specific risks their target
audience perceives, whether these risks be environmental, social, or a combination of
many factors.

270

K.I. Stoddard and D.B. Huggett

Risk Communication and Education

The well-understood purpose of risk communication is to provide individuals with
the information necessary for them to reach informed decisions about risks they
may encounter relating to their health, safety, or the environment [3945]. Though
a seemingly simple prospect, risk communication and education is complicated by
the fact that, in addition to requiring a comprehensive knowledge of the risks the
public faces, officials responsible for protecting public and environmental health
and safety must also effectively communicate and educate the public while respecting the delicate relationship between risk perception, objective assessments, and the
publics response to risks. While some take back programs have made efforts to
educate and communicate with their target audiences, it seems the majority of programs do not fully appreciate the critical role risk communication and education can
play in the success of the program. To address this common gap in take back programs and in an effort to stress how take back programs could greatly benefit from
more fully incorporating risk communication and education into their programs this
section will highlight some of the leading strategies and intrinsic challenges of risk
communication and education.
Effectively communicating a complex risk message requires advanced planning
and a well-considered strategy. One risk communication strategy recommended by a
group of experts is a straightforward four-step method largely based on what Morgan
et al. [1992] described as the mental model approach, which is centered around
the concept that individuals assess new information based on their existing knowledge and/or beliefs [46]. For example, information provided for a new topic for
which individuals have no prior experience or knowledge will likely be difficult for
them to understand and apply context to. Alternatively, individuals with preexisting
misconceptions or erroneous information may misinterpret correct risk communication and education messages. A significant amount of research has been devoted to
mental models and has revealed the important influence they have on how individuals develop skills, follow directions, and use equipment [4753]. This concept bares
striking familiarity to the risk perception concept that behaviors and actions are
influenced by individuals personal experience and prior understanding of risks.
With such a wide base of research indicating the critical role individuals understanding and beliefs play in determining their actions, the first step for an effective risk
communication and education program would logically be to determine the existing
knowledge and beliefs of the target audience. The four-step method recommends
accomplishing this important task by providing open-ended opportunities for individuals to express their knowledge and beliefs in regard to a particular topic. Additionally,
experts stress that it is equally as important to extract both accurate and inaccurate or
misguided beliefs and knowledge from their audience [46]. Focus groups, interviews,
city council meetings, and a variety of other venues or methods may be employed by
risk communicators and educators as they seek to understand their audience and satisfy
this primary objective of risk communication and education.
Once a model of the audiences beliefs and understanding has been constructed,
the next recommended step is to incorporate this information into structured surveys

Pharmaceutical Take Back Programs

271

to evaluate how prevalent the beliefs and knowledge are within the audience.
Building upon the well-structured model of the target audiences knowledge and
beliefs, the final two steps in the recommend risk communication strategy are to
develop and continually evaluate communication methods [46]. Methods may
include distribution of educational material, broadcasting messages, and establishing an informational website. Communication methods utilized should be carefully
considered in light of the objectives of the communication campaign and the characteristics of the audience. Effectively designed informational material that accounts
for these important variables may be able to clarify skewed or inaccurate beliefs of
lay individuals by providing the additional accurate facts necessary for them to
improve and/or refine the knowledge and beliefs they currently maintain concerning
the risk [46, 54].
Another model that can be applied for developing effective risk communication
and education strategies is the physicianpatient model of communication. In both
environmental risk communication and physicianpatient relationships experts
provide objective information regarding facts on health, safety, uncertainties, and
alternatives. Additionally, experts also are often required to convey their knowledge
concerning the severity of the issue, possible alternative options, resolutions, and
their advice on how to proceed forward with managing the issue. With these similarities between the two communication processes, it seems logical that some
recommendations and approaches used in the physicianpatient model would be
appropriate for environmental risk communication [55]. One source of such recommendations is a presidential commission report on ethical problems in medicine and
biomedical research issued in 1982. Though initially intended for improving health
care decision making, this report provides three practical recommendations that are
quite applicable for developing and improving environmental risk communication
strategies. These recommendations are: approaching risk communication as a
dialogue rather than an isolated occurrence, providing additional sources of information, and providing clarification on types of uncertainties associated with the risk
information [55, 56].
Due to the complex nature of risks and the complications involved in educating
the public about them, it should not be unexpected that experts in risk communication and education must contend with a suite of challenges as they work to provide
accurate and appropriate amounts of information to the public. One ever-present
challenge risk communicators and educators face is establishing and maintaining
trust and credibility with the public [57]. Another challenge is accounting for the
community structure and diversity of the audience. Specifically, concepts generally
associated with environmental justice such as culture, economics, and life experiences of a community have been identified by experts as factors that should be
kept in mind during communication efforts as these factors can lead to health and
opportunity disparities between different communities [58]. Another communitybased challenge of risk communication and education is overcoming the difficulty
of rallying individuals to support public interests as equally as they support
their individual interests. In response to this challenge, experts note that communication efforts are most effective when individuals are united as a community [58].

272

K.I. Stoddard and D.B. Huggett

However, what may be the most challenging task associated with addressing
community components and dynamics may be the fact that these components and
dynamics are unique to each community, requiring risk communicators and educators to constantly reevaluate their message for applicability and appropriateness for
the audience.
Despite the challenges associated with developing an effective risk communication and education strategy, there are a wealth of public health campaigns and environmental risk communication messages that have been launched in the past that
can testify as to the effectiveness of properly planned and executed risk messages.
As an example, consider the current pervasiveness of knowledge among the general
public for concepts such as the importance of wearing a safety belt, the existence of
global climate change, or the risks of drinking alcoholic beverages while pregnant.
These risk communication successes indicate that while the task of providing critical
and complex scientific information to the public concerning excess and unused
pharmaceuticals in the home and the environment may initially appear to be daunting,
incorporation of proven strategies and models of risk communication and education,
and recognition and planning for the challenges that will inevitably confront the
campaign, will very likely significantly improve the overall effectiveness of the
campaign in educating the public and providing them with the information they need
to make informed decisions regarding their health, safety, and the environment.

Improving Medication Management Strategies

A final powerful idea that take back programs as a whole have generally overlooked
is the potential to impact healthcare costs over the long term through improving
medication management strategies. As previously mentioned, data gathered from
surveys completed during take back programs can yield information needed to
determine the value of unused medication and the amounts and types of collected
unused medications. What remains to be seen is if this information, which may not
be currently available by any other means, can lead to a reduction in healthcare costs
through improved use of the medical resource of pharmaceuticals.
Though the economic value of wasted pharmaceuticals is important information,
unless this information is communicated to doctors, pharmacists, healthcare
officials, and regulators, the value of this information is not achieving its full potential. If communicated and acted upon properly, this information could have powerful implications to improve current practices of healthcare providers and thereby
possibly improve healthcare as a whole. For example, by understanding which types
of medications are being wasted, it may be possible to determine if certain medications are being overprescribed, and if so, the value of these wasted resources may be
used as a factor to help determine more efficient prescribing practices in an effort to
reduce the pharmaceutical waste from the beginning of the chain of consumption.
Improving prescribing practices has the potential to reduce the amount of a prescription
drug wasted as valuable healthcare resource, which may ultimately be improperly

Pharmaceutical Take Back Programs

273

disposed of, reduce the time a doctor spends prescribing medicine that goes unused,
and/or reduce the time a patient spends visiting a doctor to receive a prescription for
which a portion of the drug may go unused. As we consider the further reaching
implications of reducing pharmaceuticals in the environment by implementing
pharmaceutical take back programs, it is possible to also anticipate addition benefits
society may gain in the form of improved medicine management strategies, which
may also lead to a reduction in the time and money spent dealing with the consequences of non-use and misuse of prescription and OTC medications.
A scientific investigation into the value of wasted pharmaceutical resources and
the economic and social implications resulting from this wasted resource was
recently conducted in Barcelona, Spain. In this study 38 randomly selected pharmacies were surveyed as they accepted returned medicines, medical care equipment,
and other items available at a pharmacy (e.g., personal care products and nutrition
products). As background information, in 2002 communities and the pharmaceutical industry in Spain collaborated to develop an industry-funded program called
SIGRE to facilitate the collection and disposal of unused and expired medications
and medication packaging. In the Barcelona study, selected pharmacies were surveyed for seven consecutive working days with the survey period beginning on the
first day a return was made. During the survey period, which spanned from February
to April of 2005, 1,176 packages of medicine were returned. Due to missing information and other complications associated with determining the volume or amount
of a medication remaining in a returned package, the value of the returned drugs
was based on 1,119 packages and came to 8,539.90, which at the exchange rate in
April of 2012 of 1 to 1.32 US dollars equaled $11,303.80. Researchers determined
that 75% of this cost or 6,463.90 ($8,549.86) was covered by the public health
care system [59]. The researchers from this study note that the returns they valued
during the event represent a significant unnecessary expense to the healthcare system
of Spain which may be addressed in the future by improvements in prescribing,
dispensing, and use of medicines in Spain [60]. Although Spain has a different
national healthcare policy than the USA, which would account for differences in
the value of returned medicines that would be covered under the US healthcare
system, this study is still applicable to take back programs in the USA because it
illustrates that medications returned equate to wasted medical resources, which
ultimately are unaccounted losses in the healthcare systems budget. Additionally,
the solutions of improving prescription, dispensing, and consumption practices this
study puts forth to address the problem of wasting medicine resources could also be
applied in the USA.
Though the concept of improving medicine management strategies through
pharmaceutical take back programs is not a widely circulating concept as of yet,
there is one broad-based initiative in the USA gaining support within the healthcare
provider community. Practice Greenhealth was originally established in 2004 by
an EPA grant as Hospitals for a Healthy Environment (H2E) and now is a vast network of member healthcare institutions and organizations that are committed to
environmentally sustainable healthcare practices. The original purpose of H2E was
to establish a national program devoted to promoting and developing environmental

274

K.I. Stoddard and D.B. Huggett

sustainable practices for the healthcare system which would serve to improve
efficiency, health, and regulatory compliance within individual communities.
As part of this program Practice Greenhealth has developed a comprehensive
management plan to reduce pharmaceutical waste on a national level. Though not a
take back program in itself, Practice Greenhealth recognizes that significant costs
and risks are associated with disposing of pharmaceutical wastes, and as a part of
the solution to this problem, Practice Greenhealth provides healthcare professionals with education, information, and resources on environmental management
strategies that address this concern as well as other environmental concerns related
to healthcare [18, 61]. With the resources, tools, and contacts within the Practice
Greenhealth network, there seems to be the potential for a mutually beneficial
relationship between local take back programs and Practice Greenhealth that
could truly help to promote the objective of improving healthcare costs and medication management strategies.
Another example of a North American program that acknowledges the fundamental connection between take back programs and improving medicine management strategies is the Canadian-British Columbia Medication Return Program
(MRP). MRP, originally named British Columbia EnviRx, was established in 1996
as a voluntary program; however, it was the pharmaceutical industry itself that eventually lobbied for mandatory product stewardship for pharmaceuticals through the
establishment of the Post-Consumer Residuals Stewardship Regulation. MRPs primary objective is to protect and improve the health of the environment, economy,
and consumers. MRP is supported by Canadas National Association of Pharmacy
Regulatory Authorities (NAPRA) due to the programs commitment to improving
consumer and child safety, reducing costs, improving therapy treatment results, and
preventing detrimental impacts to the environment. One way MRP advances the
cause of reducing pharmaceutical costs is through promoting the prescription and
distribution of manageable medication amounts that can be completed by the patient
[18]. This strategy could just as easily be incorporated into the US healthcare system to reduce pharmaceutical costs, wastes, and other problems associated with
excess pharmaceuticals.
While take back programs themselves are a relatively new concept, just emerging within the last decade or so, the idea that these programs could be used to
improve medicine management strategies through reducing wasted medications is
an even more novel concept with few if any US take back programs mentioning this
as an objective. However with the expanding interest and popularity in promoting
environmentally and economically sustainable practices into more business practices and industry standards, it seems logical that the cost savings potentially available through improving medicine management strategies and medical practices will
become increasingly obvious to healthcare providers, industry leaders, and regulatory leaders. Unfortunately, the degree to which healthcare costs could be reduced
due to take back programs would probably remain unanswered for some time, as
there will likely be considerable lag time before improved efficiency with medication management is translated into actual cost savings in the healthcare industry.
Additionally, in order for these improvements in healthcare and prescription practices

Pharmaceutical Take Back Programs

275

to be realized, an effective communication strategy must be established to educate

doctors, pharmacists, and others involved in medicine on the implications of wasted
pharmaceuticals and the data that are revealed through take back program surveys.

Potential Roadblocks
There is yet another facet of pharmaceutical take back programs that should be
considered to complete the overview of these programs. Though these programs
may provide valuable services to the community and the environment, organizers of
these programs may struggle against a variety of barriers as they proceed with the
planning and implementation process. The following provides brief examples of
some obstacles take back programs have encountered in the past and potential solutions to these problems.
The first and possibly most obvious roadblock for these programs is lack of adequate funding. Cost is not featured as one of the program attributes in the table in
Appendix A, which features a comprehensive list of example take back programs,
because this information is not reported on a consistent basis and when it is reported
it is usually not provided in a format that accommodates comparison between programs. For those programs that do report their costs, it seems most appropriate to
consider each on a case-by-case basis due to the wide variances in program characteristics. For example, the Washington State PH:ARM state-wide continuous dropoff program reports the estimated cost of the project as a lump sum of $3.3 million,
whereas the La Crosse, Wisconsin continuous drop-off program, like many other
programs, reports the costs of select components of the program without an annual
estimate for total program operation. Examples of program components reported
include: cost of disposal of medicine waste, advertising, general staffing, security,
and time and services of pharmacists and law enforcement officials. In many of the
example take back programs featured in Appendix A, local pharmacies, public works
departments, and businesses donated their time and services for take back programs
operating within their communities. Many programs also received direct funding
from federal, state, or local government programs and community organizations.
Even when programs secure adequate funding, the best of efforts can be thwarted
unexpectedly by local, state, and federal laws, regulations, and ordinances. The
most common legal obstacle encountered by take back programs is the Controlled
Substances Act (CSA), which is administered by the DEA; however, with the enactment of the Safe and Secure Drug Disposal Act of 2010 individuals who have legal
possession of controlled medications will likely encounter less hurdles as they seek
to properly dispose of these medications. Medications considered to be controlled
substances by the CSA include narcotics and other medications like Valium, amphetamines, Ritalin, morphine, methadone, and oxycodone. Prior to the Safe and Secure
Drug Disposal Act federal law mandated that once controlled substances were prescribed to the patient the only individuals who could maintain possession of them
were the patient and law enforcement officers. This restriction required take back

276

K.I. Stoddard and D.B. Huggett

programs wanting to include collection of controlled substances at their events to

either have law enforcement officials present at collection events to take possession
of controlled substances or coordinate with police stations and sheriffs offices to
allow citizens to bring controlled substances to these facilities for proper disposal.
Now the law states that individuals who legally obtain controlled medications can
dispose of them through agents authorized to collect and dispose of medication, so
long as the disposal process is in accordance with regulations established by the US
Attorney General [62]. However at the time of this writing the regulations required
to be set forth by the US Attorney General had yet to be established or announced,
so there is still some uncertainty as to the procedures future take back programs will
need to follow to be in compliance with the CSA and the Safe and Secure Drug
Disposal Act. Given the security and safety issues surrounding controlled medications, it is anticipated these rules will include some stringent requirements to prevent
diversion of controlled medications being collected at take back events.
Finally, though public awareness and support may not be an initial hurdle to
overcome for organizing and launching a take back program, community involvement with both individual citizens and community groups, such as businesses and
public works programs, will greatly determine the success of the program. Education
and risk communication play key roles in take back programs, because if citizens
are either unaware of the risks or unaware of the opportunity to dispose of their
excess medicines, the program will not fulfill its basic goal of ensuring proper disposal of pharmaceuticals and protecting the health of individuals and the environment. Though there are many factors which may influence the success of a particular
take back program, a well-prepared and implemented education campaign which
informs citizens about the risks associated with excess pharmaceuticals in the home,
the impact of pharmaceuticals on the natural environment, and the details of how to
participate in their local take back program will likely prove invaluable in increasing citizen participation and mustering community support.

Pharmaceutical Take Back Program Case Studies

Specific details of individual take back programs vary depending on such factors as
the resources, goals, and policies of the entity organizing the program or the area in
which the program will operate; however, the overall structure and operation of
most take back programs are strikingly similar. In general, citizens are given the
opportunity to return unused or unwanted medications to a collection center. The
most common location for collection centers is a local participating pharmacy; however, hospitals, general practitioners offices, and local police stations have also
served as host locations for take back programs. Medications collected are often
separated from their bottles and the pills, tablets, or other forms of the medication
are placed of in a clearly marked container. The collection container may contain a
deactivating liquid which renders the medications useless. This serves as a precautionary measure to ensure that in the event of a breach of security no medicinally

Pharmaceutical Take Back Programs

277

active medications would be retrievable from the container. Filled collection

containers are released to a contracted certified hazardous waste manager for
disposal, which is usually accomplished via incineration. Pharmaceutical take back
programs may be stand alone collections operated as 1-day events or they can be
operated on a more frequent basis such as seasonally or year-round. Pharmaceutical
take back programs have also been launched in conjunction with household hazardous waste collection events sponsored by local agencies or government service
departments such as sanitation or environmental services.
Though there are many examples of successful take back programs, this work is
not intended to provide an exhaustive review of all take back programs, but rather to
highlight unique features and successes of a variety of programs. Here we provide a
review of one widely publicized and documented program. This program was
selected for this review due to the wide availability of information on the program
and the innovative ideas integrated into the program in its efforts to promote the
concept of sustainable medicine. A brief summary of additional programs in the
form of a table is featured in Appendix A.
The Teleosis Institute Green Pharmacy Program (Teleosis) is one example of a
pharmaceutical take back program that has experienced dramatic success. The
Teleosis Institute is an organization dedicated to promoting sustainability and environmental stewardship within the healthcare industry. As a continuation of these
ideals, Teleosis operated their take back program as a pilot program in Berkeley, CA,
for 1 year between June 1, 2007 and June 1, 2008, partnering with a local pharmacyElephant Pharm. Drop-off locations for the Green Pharmacy Program were
located at participating pharmacies, dentist offices, an animal hospital, the Teleosis
Institute, and a healthcare facility. The Green Pharmacy Program was designed as a
product stewardship model take back program in which all participants involved in
pharmaceutical products, from the manufacturers, to healthcare providers, retailers,
patients, and finally those who dispose of the products, were brought together as
partners, equally responsible for ensuring the products were safely disposed of to
reduce the impacts of pharmaceuticals on the environment [63].
With the ultimate goal of promoting and encouraging the adoption of the tenants
of sustainable medicine, this program developed some very unique features.
Incorporation of some of these sustainability guided program features has the potential to enhance other take back programs aimed at reducing environmental and
human health risks associated with excess pharmaceuticals in the home. For example, pills collected by the Green Pharmacy Program were removed from their bottles
and incinerated, while the bottles were shredded and recycled to protect patient
information. The program attested that the separate disposal method had the advantages of significantly reducing both the environmental impact and cost of incineration, as fewer materials need to be incinerated. An additional advantage the program
touted was that their separate disposal method was more attractive to participants
because it protected their sensitive medical information [64]. The Green Pharmacy
Program also incorporated a public education campaign that provided information
on the take back program, proper disposal methods of pharmaceuticals, and the
impact of pharmaceuticals on the environment [63].

278

K.I. Stoddard and D.B. Huggett

Table 1 Top 10 Categories of

pharmaceuticals returned in
the Green Pharmacy Program

Table 2 Top 10 brand name

or generic products returned
in the Green Pharmacy
Program

Category of pharmaceutical

Percent

Central nervous system (CSN)

Nutritional products
Psychotherapeutic
Gastrointestinal
Cardiovascular
Respiratory
Anti-infectives
Alternative medicines
Hormones
Immunologic

22.62
14.29
12.51
8.99
8.77
6.00
6.00
5.69
4.60
2.85

Name

Category of pharmaceutical

Acetaminophen
Aspirin
Tylenol
Vitamin E
Prednisone
Ibuprofen
Warfarin
Topamax
Etodolac
Gabapentin

Analgesic, antipyretic
Analgesic
Analgesic, antipyretic
Supplement
Corticosteroid/steroid
Nonsteroidal anti-inflammatory (NSAID)
Anticoagulant
Anticonvulsant
NSAID
Anticonvulsant

To facilitate future analysis of the program, data on all medicines collected were
recorded in a national registry. This registry is known as the Unused and Expired
Medicine Registry and was developed by The Community Medical Foundation for
Patient Safety. Data collected was used to determine which category of medications were the most overprescribed or unused medicines, which medications were
most often returned, the monetary value of the returned medicines, and to estimate
the environmental impacts of the returned pharmaceuticals. Preliminary results of
the Green Pharmacy Program from its inception on June 1, 2007 to December 31, 2007
indicated that a total of 690 pounds of medicines were returned through the program
with an estimate of 101,359 returned pills, capsules, and tablets. The total wholesale
value of these medicines was estimated to be $159,778 and total retail value was
estimated to be between $228,254 and $399,445. The majority of returned medicines (60.43%) were prescriptions as opposed to OTC medicines (39.14%). Tables 1
and 2 provide the ten most frequently returned class or category of pharmaceuticals
and the brand or generic name of medicines returned, respectively [63].
Although summary statistics were provided in the Teleosis Preliminary Data
Report and their Green Pharmacy Final Report, published in fall of 2008, no data,
summary statistics, or other results are readily available from Teleosis to explain their

Pharmaceutical Take Back Programs

279

results or findings aimed at achieving their final goal of the estimating environmental
impact of returned medicines. Additionally, the Unused and Expired Medicine
Registry website is still in progress and as of this writing does not have publicly available data to share with the community [65].
The Green Pharmacy Program is just one example of a successful take back
program and it should be noted that the concept of pharmaceutical take back programs
has been embraced by many communities and there are many exceptional and
successful programs both here in the USA and abroad. Although complete summaries
of all the existing and past take back programs would be informative, that extensive of
a review is not necessary for understanding significant role these programs can play in
potentially reducing the amount of pharmaceuticals in the environment. Rather a more
efficient method of providing a comprehensive overview of the breadth of take back
programs would be to examine a selection of past and existing programs instituted
across the USA and in foreign nations and to consider a selection of key elements
comparable between the programs. Appendix A provides a list of example programs
and their comparable or distinguishable features. This appendix includes information
from the previously discussed program as well as information adapted from the IllinoisIndian Sea Grant (IISG) resource guide designed to assist communities in managing
programs aimed at the proper disposal of their unused pharmaceutical [18].
Though the Internet provides a vast array of information on pharmaceutical take
back programs both in the USA and worldwide, this information can be regarded
mainly as general in character and lacking the scientific analysis that is necessary
for a deeper understanding of these programs and the possible implications they
may have on the environment, society, government, the pharmaceutical industry,
and other groups involved either directly or indirectly with pharmaceutical manufacturing, distribution, consumption, and disposal. One peer-reviewed article that
touches on the environmental and social problems that prompted the development
of take back programs and that conducts a detailed analysis of the outcomes of a
pharmaceutical take back program launched in the UK is An analysis of returned
medicines in primary care, published in Pharmacy World and Science in 2005 by
Langley et al. This study is unique in that it touches on many of the subjects previously
mentioned in the Objectives and Idendifying Gaps sections of this chapter.
As a scholarly introduction into the concept of take back programs, the authors
of this article provide a discussion of some of the potential impacts of unused and
surplus medications, many of which were discussed at length previously in this
chapter. These potential impacts include such things as the minimization for accidental poisoning and medicine misuse, prevention of detrimental effects on the
environment, and accounting for costs associated with wasting a resource. The
authors of this study also point out that although take back programs provide a
potentially effective method for reducing or eliminating the risk of accidental poisoning or misuse and detrimental environmental impacts, these programs do not
directly address the need to reduce the amount of pharmaceutical resources that are
being wasted. As a budding solution to this problem, this study was launched to
investigate the type and quantity of returned pharmaceuticals and the reasons given
for their return.

280

K.I. Stoddard and D.B. Huggett

To collect the necessary data on pharmaceutical returns and reasons for returns,
two 4-week long take back programs were arranged in East Birmingham in the UK;
one during August 2001 for returns to pharmacies and one in March 2002 for returns
to general practitioners (GP) offices. Returned medicines were cataloged by therapeutic category and the number of doses remaining and information concerning the
person returning the medicine including age, gender, and reason for the return were
collected by a pharmacist or GP. Where appropriate, additional information from
patient notes or pharmacy records was acquired. The study made no efforts to advertise the take back program within the community. During the two 4-week collection
events there were 114 returns totaling 340 items. The majority of returns both in
number of returns (90 of 114 or 78.9%) and total items returned (298 of 340 or
87.6%) came from the pharmacy collection event as opposed to the GP collection
event. A change in doctors prescription orders was the reason given for return of
the majority of medicines. Other reasons cited included, in order of frequency,
clean-out of excess home medicine supplies, clean-out following a patients death,
and because the medication was expired. The value of returned medicines was estimated at 3,986 which at the exchange rate in April of 2012 of 1 GBP to 1.32 US
dollars equaled $5,272.40 [59, 66].
In their discussion, the authors recognized that judging by the responses given for
the return of medicines, in many cases there is unnecessary waste of medicine. The
authors provided several suggestions to minimizing or eliminating this waste such as
modifying prescription practices by reducing the supply of medicines provided
throughout therapy, providing limited test supplies during initiation or during a
change of therapy treatment, and judicious review of a patients reaction and preference for a medicine. They also suggested establishment of more efficient use of
electronic prescribing systems that are capable of tracking patterns in patients medicine use. These systems could also be utilized to track a patients medicine supply
and prevent at-home stockpiling. The authors also expressed concern over the
quantity and monetary value of the unused medicines returned during their study. In
addition to the limited size and scope, this study also excluded healthcare facilities,
which could very likely make a significant impact on medicine use and wastage data.
These excluded facilities included such places as elderly care centers, elderly assisted
living centers, and hospitals. Langley et al. [2005] advise that this data should not be
used to extrapolate costs to the entire nation due to these limitations; however, they
did recognize the significant results of this study in terms of the quantity of unused
medicines and the considerable financial burden this waste placed on the national
health care system of the UK [66].
The Green Pharmacy Program and the Langley et al. [2005] study are just two
examples of take back programs that have achieved success in, not only the fundamental goal of providing a safe and proper pharmaceutical disposal option for
individuals, but also in other areas such as estimating the value of returned
medicines and promoting the concept of product stewardship. The inclusion of
these two examples however should not distract from the fact that there are a
wealth of other noteworthy take back programs in the USA and abroad, a selection of which are featured in Appendix A. The intention of this section rather was

Pharmaceutical Take Back Programs

281

to demonstrate that the highlighted programs incorporated some innovative ideas

and promoted the expansion of take back programs to address many of the social
problems discussed previously in this chapter. By examining The Green Pharmacy
Program and other programs featured in this section and the Appendix it should be
evident that take back programs themselves have a great deal of diversity. If applied
and utilized, this diversity has the potential to enrich future take back programs and
provide them with information and ideas needed to expand their causes to address
both public and environmental health problems associated with excess and unused
pharmaceuticals.

Conclusion
Though the pervasiveness and consequences of pharmaceuticals in the natural environment is not entirely understood at this point, there is increasing evidence that
there are significant impacts to the environment and public health resulting from
excess and unused pharmaceuticals. One commonly understood principle in toxicology and science in general is the precautionary principle, which advocates that
when in doubt one should proceed conservatively. In regard to pharmaceuticals in
the environment, few would argue that adopting such an ideology would be anything but beneficial to society and the environment. Take back programs appear to
be a very logical means of applying the precautionary principle to address the social
and environmental problems individuals and communities are now facing as a result
of excess and unused pharmaceuticals.
Provided they are properly planned, coordinated, and they can overcome potential road blocks, take back programs provide a wealth of opportunities for combating the many consequences of excess and unwanted pharmaceuticals. These
consequences include not only detrimental environmental impacts, but also public
health and other social issues such as accidental poisoning, abuse of pharmaceuticals, patient privacy issues, and inefficiencies in the healthcare system due to
wasteful management of pharmaceutical resources. However, despite the far reaching impacts of take back programs, few programs have developed methods to evaluate the success of the program beyond tallying the amount of medications
collected or the number of people participating in a program event. Additionally
few programs have realized the full potential that take back programs have to offer,
leaving substantial gaps in the take back program framework. In this regard, there
are several elements that would greatly enhance the take back program framework
including:
Addressing the need for scientific justification through biological and chemical
monitoring before, during, and after take back program events.
Accounting for public awareness and risk perception of pharmaceuticals in the
environment and in the home and using this information to promote a change in
disposal behavior through risk communication and education strategies.

282

K.I. Stoddard and D.B. Huggett

Improving medication management strategies by developing a communication

strategy to educate doctors and pharmacists as to the types and quantities of
medications going unused and being returned to take back programs.
It is unfortunate that for many of the environmental challenges we are currently
battling, such as climate change, urban sprawl, and dwindling biodiversity, we as a
society only began to be aware of the issues and take decisive action after the problems were widespread and precariously endangering the balance of the environment. Rather than repeat this pattern with pharmaceuticals, why not take proactive
action to prevent what is a looming environmental and social problem by promoting
and establishing programs such as take back programs that can help reduce the
amount of unused and excess pharmaceuticals that pose both a threat to public
health and a threat to the health of the environmental. We cannot afford to continue
to knowingly engage in irresponsible management of our medical resources because
the impacts of such actions are bound to have serious consequences not only for the
environment but also for the society and individuals.

References
1. Drug Enforcement Administration (DEA) (2011) National Take Back Initiative. http:www.
deadiversion.usdoj.gov/drug_disposal/takeback/takeback.html. Accessed 20 Mar 2012
2. U.S. Food and Drug Administration (FDA) (2012) Proper disposal of prescription drugs.
http://www.fda.gov/ForConsumers/ConsumerUpdates/ucm101653.htm. Accessed 28 Mar
2012
3. Bay Area Pollution Prevention Group (BAPPG) (2006) Report on the San Francisco bay
areas safe medicine disposal days. http://oracwa.org/files/news/168/SFBAYSafeMedsReport-August2006.pdf. Accessed 22 Mar 2009
4. Boivin M (1997) The cost of medication waste. Can Pharm J/Revue Pharm Can
130(4):3239
5. Kuspis DA, Krenzelok EP (1996) What happens to expired medication? A survey of community
medication disposal. Vet Hum Toxicol 38(1):4849
6. Seehusen DA, Edwards J (2006) Patient practices and beliefs concerning disposal of medications. J Am Board Fam Med 19(6):542547
7. Zero Waste Washington (2006) Washington Citizens for Resource Conservation (WCRC): a
soundstats report. Prepared by Northwest Research Group Inc., http://www.zerowastewashington.org/images/pdfs/med-survey.pdf. Accessed 22 Mar 2012
8. Eckel WP, Ross B, Isensee R (1993) Pentobarbital found in groundwater. Groundwater 31(5):
801804
9. Holm JV, Rugge K, Bjerg PL, Christensen TH (1995) Occurrence and distribution of pharmaceutical organic compounds in groundwater down gradient of a landfill (Grinsted Denmark).
Environ Sci Technol 29(5):14151420
10. Ahel M, Jeliic I (2001) Phenazone analgesics in soil and groundwater below a municipal solidwaste landfill. In: Pharmaceuticals and personal care products in the environment: scientific
and regulatory issues. ACS Symposium Series 791, Washington, DC, pp 100115
11. Metzger JW (2004) Drugs in Municipal landfills and landfill teachates. In: Kmmerer K (ed),
Pharmaceuticals in the environment: sources, fate, effects, and risks, 2nd edn. Springer, Berlin,
pp 133138
12. Fent K, Weston AA et al (2006) Ecotoxicology of human pharmaceuticals. Aquat Toxicol
76(2):122159

Pharmaceutical Take Back Programs

283

13. Kotchen M, Kallaos J et al (2008) Pharmaceuticals in wastewater: behavior, preferences, and

willingness to pay for a disposal program. J Environ Manage 90:14761482
14. Stackelberg PE, Furlong ET et al (2004) Persistence of pharmaceutical compounds and other
organic wastewater contaminants in a conventional drinking-water-treatment plant. Sci Total
Environ 329:99113
15. Webb SF (2004) A data based perspective on the environmental risk assessment of human
pharmaceuticals III-indirect human exposure. In: Kmmerer K (ed) Pharmaceuticals in the
environment. Sources, fate, effects, and risk. Springer, Berlin
16. Webb SF, Ternes T et al (2003) Indirect human exposure to pharmaceuticals via drinking
water. Toxicol Lett 142:157167
17. Daughton CG, Ternes T (1999) Pharmaceuticals and Personal Care Products in the
Environment: Agents of Subtle Change? Environmental Health Perspectives 107(6) 907938
18. Boehme SE, Hinchey EK et al (2007) Disposal of unwanted medicines: a resource for action
in your community. Habitats and Ecosystems. Chicago, Illinois Illinois-Indiana Sea Grant
(IISG). http://www.iisgcp.org/unwantedmeds. Accessed 12 Mar 2009
19. Bronstein AC, Spyker DA et al (2010) 2009 Annual Report of the American Association of
Poison Control Centers National Poison Data System (NPDS): 27th Annual Report. Clin
Toxicol 48(10):9791179
20. The Partnership for a Drug-Free America (2006) Generation Rx: National Study Confirms
Abuse of Prescription and Over-the-Counter Drugs. http://www.drugfree.org/newsroom/generation-rx-national-study-confirms-abuse-of-prescription-and-over-the-counterdrugs-%E2%80%9Cnormalized%E2%80%9D-among-teens. Accessed 21 Mar 2009
21. Substance Abuse and Mental Health Services Administration (2011) Results from the 2010
National Survey on Drug Use and Health: Summary of National Findings, NSDUH Series
H-41, HHS Publication No. (SMA) 11-4658. Rockville, MD: Substance Abuse and Mental
Health Services Administration
22. Grasso C (2009) The PH:ARM Pilot: pharmaceuticals from households. A return mechanism.
www.lhwmp.org/home/HHW/documents/PHARM_2009_Exec_
Executive
summary.
Summary_Web_Version.pdf. Accessed 21 Mar 2009
23. Huggett DB, Cook JC et al (2003) A theoretical model for utilizing mammalian pharmacology
and safety data to prioritize potential impacts of human pharmaceuticals in fish. Hum Ecol
Risk Assess 9:17891799
24. Huggett DB, Ericson JF et al (2004) Plasma concentrations of human pharmaceuticals as predictors of pharmacological responses in fish. In: Kummerer K (ed) Pharmaceuticals in the
environment II. Springer, Heidelberg, pp 373386
25. Slovic P (1987) Perceptions of risk. Science 236(4799):280285
26. Slovic P, Fischhoff B et al (1985) Characterizing perceived risk. In: Kates RW, Hohenemser C,
Kasperson JX (eds) Perilous progress: managing the hazards of technology. Westview, Boulder,
pp 91125
27. Henrion M, Fischhoff B (1986) Assessing uncertainty in physical constants. Am J Phys
54(9):791798
28. Kahneman D, Slovic P et al (1982) Judgement under uncertainty: heuristics and biases.
Cambridge University Press, New York
29. Fischhoff B, Slovic P et al (1978) How safe is safe enough? A psychometric study of attitudes
towards technological risks and benefits. Policy Sci 9(2):127152
30. Slovic P, Fischhoff B et al (1980) Facts and fears: understanding perceived risk. In: Schwing
RC, Albers WA Jr (eds) Societal risk assessment: how safe is safe enough? Plenum Press, New
York, pp 181214
31. Cook PA, Bellis MA (2001) Knowing the risk: relationships between risk behaviour and health
knowledge. Public Health 115:5461
32. Lundborg P, Lindgren B (2004) Do they know what they are doing? Risk perceptions and
smoking behaviour among Swedish teenagers. J Risk Uncertainty 28(3):261286
33. Viscusi WK, Carvalho I et al (2000) Smoking risks in Spain: part III-determinants of smoking
behavior. J Risk Uncertainty 21(23):213234

284

K.I. Stoddard and D.B. Huggett

34. Bound JP, Kitsou K et al (2006) Household disposal of pharmaceuticals and perception of risk
to the environment. Environ Toxicol Pharmacol 21:301307
35. Jakus PM, Shaw DW et al (2009) Risk perceptions of arsenic in tap water and consumption of
bottled water. Water Resour Res 45:W05405
36. Nordlund AM, Garvill J (2003) Effects of values, problem awareness, and personal norm on
willingness to reduce personal car use. J Environ Psychol 23:339347
37. COMPAS (2002) F&DA Environmental Assessment Regulations Project Benchmark Survey.
A Report to Health Canada (POR-02-13). Office of Regulatory and International Affairs
http://www.hc-sc.gc.ca/ewh-semt/contaminants/person/impact/por-02-13-eng.php. Accessed
30 Mar 2012.
38. Harrison CM, Burgess J et al (1996) Rationalizing environmental responsibilities: a comparison
of lay publics in the UK and the Netherlands. J Environ Change 6(3):215234
39. Evans JS, Hawkins NC et al (1988) The value of monitoring for radon in the home: a decision
analysis. J Air Pollut Control Assoc 38(11):13801385
40. Fischhoff B (1990) Psychology and public policy: tool or toolmaker? Am Psychol
45(5):647653
41. Gibson M (1985) To breathe freely: risk, consent, and air. Rowman & Littlefield, Totowa
42. Gow HBF, Otway H (1990) Commuincating with the public about major accident hazards.
European conference on communicating with the public about major accident hazards.
Routledge, Italy
43. Morgan MG, Lave L (1990) Ethical considerations in risk communication practice and
research. Risk Anal 10(3):355358
44. Nazaroff N, Teichman K (1990) Indoor radon: exploring U.S. federal policy for controlling
human exposures. Environ Sci Technol 24(6):774782
45. Svenson O, Fischhoff B (1985) Levels of environemntal decisions. J Environ Psychol
5(1):5567
46. Morgan MG, Fischhoff B et al (1992) Communicating risk to the public: first learn what people
know and believe. Environ Sci Technol 26(11):20482056
47. Carroll JM, Olson JR (1987) Mental models in human-computer interaction: research issues
about what the user of software knows. National Research Council, Washington, DC
48. Craik K (1943) The nature of explanation. Cambridge University Press, Cambridge
49. Gentner D, Stevens AL (1983) Mental models. Lawrence Erlbaum Associates, Hillsdale
50. Johnson-Laird P (1983) Mental models. Harvard University Press, Cambridge
51. Murphy GL, Wright JC (1984) Changes in conceptual structure with expertise: differences
between real-world experts and novices. J Exp Psychol Learn Mem Cogn 10(1):144155
52. Norman DA (1983) Some observations on mental models. In: Gentner D, Stevens A (eds)
Mental models. Erlbaum, Hillsdale
53. Rouse WB, Morris NM (1986) On looking into the black box: prospects and limits in the
search for mental models. Psychol Bull 100(3):349363
54. Krug D, George B et al (1989) The effect of outlines and headings on readers recall of text.
Contemporary Educ Psychol 14(2):111123
55. Boyle M, Holtgrove D (1989) Communicating environmental health risks using the doctorpatient model. Environ Sci Technol 23(11):13351337
56. Presidents Commission for the Study of Ethical Problems in Medicine and Biomedical and
Behavioral Research (1982) Making health care decisions: report on the ethical and legal
implications of informed consent in the patient-practitioner relationship. U.S. Government
Printing Office, Washington, DC, pp 68128
57. Peters RG, Covello VT, MacCallum DB (1997) The determinants of trust and credibility in
environmental risk communication. An Emperical Study. Risk Anal 17(1):4354
58. Institute of Medicine (2001) Science and risk communication: a mini-symposium sponsored
by the roundtable on environmental health sciences, research, and medicine. National Academy
Press, Washington, DC
59. XE (2011) Universal currency calculator. http://www.xe.com/ucc/. Accessed 03 Apr 2012

Pharmaceutical Take Back Programs

285

60. Coma A, Modamio P et al (2008) Returned medicines in community pharmacies in Barcelona,

Spain. Pharm World Sci 30:272277
61. Practice Greenhealth (2008) Pharmaceutical waste minimization. http://cms.h2e-online.org/
ee/waste-reduction/waste-minimization/pharma/. Accessed 7 Aug 2009
62. OpenCongress (2011) Open Congress for the 112th United States Congress. http://www.opencongress.org/bill/111-s3397/show. Accessed 20 July 2011
63. Teleosis Institute (2007) Green Pharmacy Program: helping communities safely dispose of
unused medicines: preliminary data report. Berkeley, CA, Teleosis Institue: 16. http://www.
teleosis.org/pdf/GreenPharmacy_FullPreliminaryReport.pdf. Accessed 7 Aug 2009
64. Elephant Pharm (2008) Elephant pharm unveils groundbreaking green pharmacy program.
http://www.elephantpharm.com/content/view/602/132/. Accessed 7 Aug 2009
65. CRG Medical Foundation (2006) Community Medical Foundation for Patient Safety. The
Community of Competence and Foundation for Life. Patient Safety Registry. Available:
http://www.comofcom.com/unused.html. Accessed 03 Apr 2012
66. Langley C, Marriot J et al (2005) An analysis of returned medicines in primary care. Pharm
World Sci 27(4):296299

Appendix A. Take Back Program Case Studies

B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3, Springer Science+Business Media, LLC 2012

287

1 year

1998 to present

Unknown

Continuous
drop-off

Education
Campaign

Education
Campaign

Green Pharmacy
[15]

Northwest Product
Stewardship
Council [11]

Smarxt Disposal
[19]

Operation length

Category of
programa

Program name or
location
RegionalNorthern
California

Level of
implementation

U.S. Fish and Wildlife

Service; American
Pharmacists
Association;
Pharmaceutical
Research and
Manufactures of
America
Nationwide

Coalition of government State level

organizations in
Oregon and
Washington

Teleosis Institute

Organizing body

Outcome

Website [19]

Over 2,000 lbs unwanted

medicines collected;
educational material
developed; 12 collection
sites and 4 single-day
events
Website [11]

Long-term plans
No pharmaceutical
waste in environment; sustainable
medicine; develop
support for TBPs
nationwide
Foster cooperation
among governments, businesses,
and nonprofit
groups to facilitate
the incorporation
of product
stewardship
principals into
public policy
Raise awareness and
provide practical
guidance

Category of
programa

Clark County,
Washington
Unwanted
Medications
Take Back
Program [2]

Continuous
drop-off

Continuous
Washington State
drop-off
Unwanted
Medicine Return
Program:
Pharmaceuticals
from
Households: A
Return
Mechanism
(PH:ARM) [2, 8]
Regional Excess
Continuous
drop-off
Medication
Disposal Service
(RxMEDS)
[9, 17]

Program name or
location

20032010

Collaboration between
local pharmacies,
state board of
pharmacy, the DEA,
and the county
sheriffs office

Countywide, but
required
statewide
cooperation

St. Louis Metro

Region

EPA and RxMEDS

Partnersc

18 months

Level of
implementation
Statewide in
Washington

Organizing body

Pilot program
PH:ARM Coalitionb
operating from
October 2006
to October
2008

Operation length

Outcome

Long-term plans

Collected and disposed of

296,650 doses of returned
medicines from over 892
participants over a 12
month period, consisting
9220 collection days
Program cost of $137,849
covered by an EPA grant
[10]

Exact quantity of collected

medications unknown but
23 lbs of controlled
medications were collected
in 2006

(continued)

Increase public
awareness
Reduce threat posed
by improperly
disposed of
medicines to
environment and
public health
Improve pharmaceutical advertising for
proper disposal of
unused
medications
Continue program at
pharmacies and
police and sheriff
offices year round

Collected and disposed of

Long-term sustain35,000 lbs as of March
ability through
11, 2009
establishing
partnerships with
Statewide program projected
pharmaceutical
cost $3.3 million or ~$5.60/
manufacturers
lbs medicine collected
[2, 8]
Website [12]

Continuous
drop-off

La Crosse,
Wisconsin [2, 5]

Single-event
(California) Bay
collection
Area Pollution
Prevention
Group (BAPPG)
Safe Medicines
Disposal Week
[1, 2]

Category of
programa

Program name or
location

1 week in May
2006

2007 to present

Operation length

BAPPG

La Crosse County Solid

Waste Department

Organizing body

Outcome

Long-term plans

Multiple counties Estimated annual cost of

Information not
$12,000$15,000
available;
presumed program
No data available on quantities;
will continue
however, participation
reported as increasing with
awareness of the program
and very small quantity
generators of hazardous
waste expressing interest in
utilizing the program as
well
Collections held 3,634 lbs pharmaceutical waste Collaborate with DEA
at 39 Waste
collected from 1,500
to develop a
Treatment
participants
cost-effective and
Plants (WTP) Collection method determined
legal collection
from
not cost-effective
system
surrounding
Develop long-term
cities but
and sustainable
regional
solution for
coordination
disposal of unused
was required
pharmaceuticals
such as developing
a continuous
drop-off program
at pharmacies
Also considering
prepaid mailers as
a return method
for unused
medications

Level of
implementation

Single-event
collection

Single-event
collection

Unwanted
Medication
Disposal Drive
[2]

Earth Keeper
Initiative [2]

Mail-back
Safe Medicine
Disposal for ME
[2, 6, 14, 18]

Category of
programa

Program name or
location
Event sponsored by a
variety of local
public service
agenciesd

Organizing body
1-day event
consisting of
25 collection
sites in areas
surrounding
Chicago

Level of
implementation
Outcome

Long-term plans

(continued)

Approximately 6,000 lbs OTC Information not

and prescription medicaavailable;
tions collected in 4 years;
presumed program
also installed permanent
will continue
collection containers at 5
police stations and
collected ~1,000 lbs in 1
year
1 day- Earth Day Earth Keeper Initiative
19 collections
Over 1 ton of medicines collected Information not
2007
(an affiliation of a
sites scattered
from 2,000 participants
available;
variety of religious
across 15
presumed program
Estimated street value of
groups), Cedar Tree
counties in
will continue
controlled substances:
Institute and
Michigans
$500,000
Year-round
Coordination and
Statewide in
See Tables A.1 and A.2 for
Continue program
operating since
administration of
Maine
preliminary findings
through funding
2007
program provided by
(before program went
provided by EPAs
University of Maine
statewide)
Aging Initiative
Center on Aging
One envelope of returned
narcotics estimated to have
a $7,000 street value
Many envelopes contained full
bottles of unused medications from mail-order or
VA pharmacy services
Received full bottles of
antiretroviral (HIV/AIDS)
drugs, which have a very
high value

Annually 1-day
collection
event

Operation length

Category of
programa

Single-event
collection

Foreign
British Columbia
continuous
Medications
drop-off
Return Program
(formerly British
Columbia
EnvirRx) [3, 13]

National Take Back

Initiative [7]

Single-event
Denton Drug
collection
Disposal Day [4]
and
continuous
drop-off

Program name or
location

Since 1996

Since September
2010

Semi-annually
since April
2010

Operation length

Level of
implementation
Outcome

Long-term plans

1-day event held Over 2500 lbs collected during Continue hosting
at area
4 separate events; has
semi-annual
hospital in
earned regional, state and
single-day events;
Denton, Texas
national awards; established
possible expansion
a permanent drop-off
to multiple
Kiosk; 1st take back event
collection sites
in Texas approved by the
DEA and the Texas
Commission on
Environmental Quality
Website [4]
US DEA
Nationwide
During 3 events 499 tons of
Continue operating
(USA)
medications were collected
single-day national
collections events
one or more times
a year
Administered by
Nationwide
From January 2007 to
Increase public
Post-Consumer
(British
December 2007 52,635 lbs
awareness and
Pharmaceutical
Columbia)
medicines returned
motivate citizens
Stewardship
Participation rate: 93%
to dispose of
Association (PCPSA)
Access at 913 pharmacies
excess and unused
medications
Regulated by Postthrough the
Consumer Residual
program
Stewardship Program
Regulation
City of Denton, Texas
and University of
North Texas; event
sponsored by a
variety of local
public and private
entitiese

Organizing body

Foreign
continuous
drop-off

Australias Return
Unwanted
Medicines
(RUM) Project
[2, 16]

Since 1998

Operation length
The Commonwealth
Department of
Health

Organizing body
Nationwide
(Australia)

Level of
implementation
Outcome

Long-term plans

From 1998 to 2002

Information not
1,675,513 lbs of pharmaavailable;
ceuticals were collected and
presumed program
destroyed; during 2005 the
will continue
RUM project served 21
million citizens and
collected and destroyed
696,241 lbs of
pharmaceuticals
a
Program categories include Continuous drop-off, Single-event collection, Mail-back, Education campaign, and Foreign continuous drop-off
b
PH:ARM (Pharmaceuticals from Households: A Return Mechanisms) Coalition includes: Interagency Resource for Achieving Cooperation (IRAC), King County
Local Hazardous Waste Management Program, Northwest Product Stewardship Council, Pacific Northwest Pollution Prevention Resource Center, Public Health
Agencies in King and Seattle County, Snohomish County Solid Waste Division, Washington Citizens for Resource Conservation, and Washington State Department
of Ecology
c
Regional Excess Medication Disposal Service (RxMEDS) Partners include: Areas Resource for Community and Human Services (ARCHS), Schnuck Markets, Inc.,
Cintas Corporation, St. Louis College of Pharmacy, Missouri AARP, Mid-East Area Agency on Aging, St. Louis OASIS, Senior Services Plus, WK Health, Stericycle,
St. Louis University, St. Louis City on Aging, MO Environmental Water Association, Metropolitan Sewer District, Living Lands and Waters, PhRMA, STL County
Waste Management Program, and American Water Company
d
Partners in Unwanted Medication Disposal Drive in Chicago, Illinois include: Cook County Sheriffs Police, Chicago Police Department, Chicago Department on
Aging, Chicago Department of Public Heath, Illinois Attorney Generals Office, Illinois TRIAD, and the Metropolitan Water Reclamation District of Greater
Chicago
e
Partners in Denton Drug Disposal Day Program: University of North Texas, the City of Denton Texas, Denton Independent School District, Denton Police Department,
Denton County Sherriffs Dept; and Denton Regional Medical Center

Category of
programa

Program name or
location

294
Table A.1 Safe medicine disposal
for ME: returned medicines data [6]

Table A.2 Safe medicine disposal for

ME: top 4 category of medicines
returned [6]

Appendix A. Take Back Program Case Studies

Category of medicine

Percent

Prescription
Controlled medicines
Over the counter
Total

90
10
10
73,000 pills, creams,
patches, etc.

Medication category

Percent

Pain/anti-inflammatory
Heart, blood, or cholesterol medicine
Sleep or anti-anxiety medicine
Antibiotics

35
34
19
18

References
1. Bay Area Pollution Prevention Group (2006) Report on the San Francisco Bay Areas Safe
Medicine Disposal Days. Bay Area Pollution Prevention Group, San Francisco, CA. Available:
http://oracwa.org/files/news/168/SFBAYSafeMeds-Report-August2006.pdf
2. Boehme SE, Hinchey EK et al (2007) Disposal of unwanted medicines: a resource for action
in your community. Habitats and Ecosystems. Chicago, Illinois Illinois-Indiana Sea Grant
(IISG). http://www.iisgcp.org/unwantedmeds. Accessed 12 Mar 2009
3. Post Consumer Pharmaceutical Stewardship Association (2012) Help protect the environment:
return expired medications. http://www.medicationsreturn.ca/. Accessed Jan 2009
4. City of Denton Texas (2011) Denton drug disposal day. http://www.dentondrugdisposal.com.
Accessed 21 July 2011
5. City of La Crosse Wisconsin (2009) Household hazardous materials. http://www.cityoflacrosse.
org/index.asp?NID=1167. Accessed Feb 2009
6. Crittenden JA, Kaye LW et al (2008) Implementing a consumer pharmaceutical mailback program: an analysis of the first year of the Safe Medicine Disposal for ME Program. 61st Annual
Scientific Meeting of the Gerontological Society of America. Available: http://www.safemeddisposal.com/documents/GSAsymposiumPPT12-1-08.pdf
7. Drug Enforcement Administration (DEA) (2011) National take back initiative. http://www.
deadiversion.usdoj.gov/drug_disposal/takeback/takeback_102911.html. Accessed November
2011
8. Grasso C (2009) The PH:ARM Pilot: pharmaceuticals from households. A return mechanism.
Executive summary. http://www.lhwmp.org/home/HHW/documents/PHARM-2009-ExecSummary-Web-Verson.pdf. Accessed 21 Mar 2009
9. Hayden SW, Gattas NM (2007) Regional excess medication disposal service (RXMEDS) presentation. Available: http://www.slidefinder.net/r/rxmeds2/rxmeds2/26870240
10. Kimbrough, W (2009) Prudent disposal of unwanted medications (RxMEDS) Final Report.
Available: http://www.epa.gov/grants/winners/rx-meds-technical-reports508.pdf. Accessed
Jan 2009
11. Northwest Product Stewardship Council (2009) http://www.productstewardship.net. Accessed
Feb 2009
12. Pharmaceuticals from households: A return mechanism (PH:ARM) (2009) Washington State
Unwanted Medicine Return Program. http://www.medicinereturn.com. Accessed Feb 2009

Appendix A. Take Back Program Case Studies

295

13. Vanasse, G. (2008) Medications Return Program Annual Report: January 2007 to December
2007. Ottawa, Ontario. Available: http://www.medicationsreturn.ca/ar2007.pdf.
14. University of Maine Center on Aging (2012) Safe Medicine Disposal for ME program http://
www.safemeddisposal.com/index.php. Accessed Jan 2009
15. Teleosis Institute (2007) Green Pharmacy Program: helping communities safely dispose of
unused medicines: preliminary data report. Berkeley, CA, Teleosis Institue: 16. http://www.
teleosis.org/pdf/GreenPharmacy_FullPreliminaryReport.pdf. Accessed 7 Aug 2009
16. The National Return & Disposal of Unwanted Medicines Limited (2011) Returning your
unwanted medicines (RUM). http://www.returnmed.com.au/. Accessed Jan 2009
17. U.S. EPA (2009) Area Resources for Community and Human Services (ARCHS). http://www.
epa.gov/aging/grants/winners/archs.html. Accessed Jan 2009
18. U.S. EPA (2009) Winners of prudent disposal of unwanted medications: University of Maine
Center on Aging. http://www.epa.gov/aging/grants/winners/umca.htm. Accessed Jan 2009
19. U.S. Fish and Wildlife Service, American Pharmacists Association et al (2009) Smarxt disposal: a prescription for a healthy planet. http://www.smarxtdisposal.net/index.html. Accessed
Feb 2009

Index

A
Active pharmaceutical ingredient (API),
168, 169
adjusted daily intake and therapeutic daily
intake values, 190194
adverse effect levels, 190
bioconcentration factors, 187
biodegradability, 197
biosolids, 175176
CCL3, 170
detection, 170171
dose-response assessment, 189
drinking water, 190, 198
environmental impacts, 168
fate and behavior, 176177
fish, 186, 187
human exposure, pathway, 178, 179
lipid content and bioaccumulation, 187
measured and predicted environmental
concentrations, 179186
microbial resistance, 198209
MSW/landfills, 176
physicalchemical properties, 65
point of departure, 190
predicted no-effect concentration, 190
public databases, 87
recreational/subsistence anglers, 187
removal in WWTP and drinking water
treatment, 178
risk characterization metrics,
198208
substances selection, 188189
surface and drinking water,
194197
therapeutic doses, 189
thresholds of toxicologic concern, 194

uptake of, 92
vapor pressure and density, 66
veterinary pharmaceuticals, 176
Advanced oxidation processes,
246247
American Association of Poison Control
Centers (AAPCC), 259, 260
B
Bay Area Pollution Prevention Group
(BAPPG), 263
Bioaccumulation and effects
antidepressants fluoxetine and sertraline,
67
aqueous effect threshold, 10, 11
gill uptake model, 9
hazard assessment, 9
plasma model, 9
toxicity, aquatic organisms, 7
veterinary medicines, 6
volume of distribution, 9, 10
C
Center for Biologics Evaluation and Research
(CBER), 53, 56
Center for Drug Evaluation and Research
(CDER), 20, 53, 56
Chlorination, 239241
Coagulation, 234
Contaminant candidate list (CCL3), 170
Controlled Substance Act (CSA),
257
Conventional wastewater treatment,
243244

B.W. Brooks and D.B. Huggett (eds.), Human Pharmaceuticals in the Environment:
Current and Future Perspectives, Emerging Topics in Ecotoxicology 4,
DOI 10.1007/978-1-4614-3473-3, Springer Science+Business Media, LLC 2012

297

298
D
Drug regulation
EA (see Environmental assessment)
in European Union, 57
NEPA process, 5153
risk assessment, 58
in USA, 5051
E
Ecological comparative pharmacology
absorption and distribution, 9293
in aquatic plants, 86
carrier-mediated uptake, 9395
drug targets
definition, 98
enzymes, 98, 100
ortholog prediction, 98, 99
protein family (Pfam) domains, 99
receptors, 98
type of, 99, 100
homologous proteins, 8990
microarray analysis, 101102
mode-of-action, 100
nontarget species, 88
orthologous proteins, 8990
passive diffusion, 9395
pathway prediction, 101
pharmacokinetics and pharmacodynamics,
8788
phylogenetics
one-way comparison approach, 90
OrthoMCL, 91
RBH-approach, 91
plasma-protein binding
CYP1A1 and CYP1A2, 96
phase II metabolizing enzymes, 97
phase I metabolizing enzymes, 96
P450 mediated drug metabolism, 9697
serum albumin and orosomucoid, 95
sex hormone-binding globulin, 95
serotonin reuptake inhibitors (SSRIs), 8687
transcription-factor proteins, 102103
Emerging contaminants, 226
Endocrine-disrupting compounds (EDC)
AOP, 246
biomarker exposure, 112
chlorine and chloramine, 239
copious effort, 168
definition, 169
membrane pore size, 244
ozone, 241
vitellogenin transcription, 126
wastewater discharge, 243

Index
Endocrine Disruptor Screening Program
(EDSP), 142
Environmental analysis and exposure
GCMS, 5
LCMS, 56
UPLC, 5
Environmental assessment (EA)
acute ecotoxicity testing, 55
CBER, 53
CDER, 53
CDER/CBER tiered approach, 55, 56
chronic toxicity testing, 5556
depletion mechanisms, 55
expected introductory concentration
(EIC), 54
extraordinary circumstances provision, 54
IND, 54
substance concentration, 54
Environmental Protection Agency (EPA), 22,
50, 123, 141, 258
Environmental risk assessment (ERA)
exposure reconstruction (see Exposure
reconstruction)
regulation, 58
sub-lethal effects (see Sub-lethal effects)
Environment and its fate
bioconcentration and transport, 79
climate change, 80
exposure models, 80
organisms uptake, 75
partitioning and persistence
activated sludge, 7071
biotransformation, 73
degradation rates, 73
elimination rates, 6970
in environmental matrices, 69
in sediments, 71
water and air, 67
water and sludge, 6869
xenobiotics, 74
physical-chemical properties
equilibrium process, 6465
ionization, 64
nonequilibrium process, 65
octanol-water partitioning, 6667
solubility and melting point, 65
vapor pressure and density, 66
sorption mechanisms, 79
transformation product exposure, 79
transport of pharmaceuticals
aquatic exposure, 7778
FOCUS modeling framework, 77
hydrology, 75
soil contaminants, 76

Index
surface water exposure, 7677
terrestrial risk, 76
treatment processes, 79
Epigenetics
CYP1A phenotype, 132
DNA methylation, 129130
noncoding RNA molecules, 131
plasticity, 130
European Centre for Ecotoxicology and
Toxicology of Chemicals
(ECETOC), 185
Exposure reconstruction
in aquatic systems, 113
batteries of molecular assays, 118
biomarkers of exposure, 112
biomonitoring data, 110
Canyon river, 114116
chloroform, 111112
definition, 111
epigenetics, 129132
fractionation/biomolecular readout
approach, 116
molecular indicators, 113115
otolith geochemistry, 132133
PBTs, 111
phenotypic anchoring, 116
primrose pathway, 126128
vitellogenin (see Vitellogenin)
WWTPs, 116117
F
Fish and Wildlife Service (FWS), 258
Flocculation, 234
G
Geography-Referenced Regional Exposure
Assessment Tool for European
Rivers (GREAT-ER), 185
Green Pharmacy Final Report, 278
Green Pharmacy Program, 277281
H
Henrys Law Constant, 67
Human health risk assessment (HHRA)
API
adjusted daily intake and therapeutic
daily intake values, 190194
adverse effect levels, 190
bioconcentration factors, 187
biodegradability, 197
biosolids, 175176

299
CCL3, 170
detection, 170171
dose-response assessment, 189
drinking water, 190, 198
environmental impacts, 168
fate and behavior, 176177
fish, 186, 187
human exposure, pathway, 178, 179
lipid content and bioaccumulation, 187
measured and predicted environmental
concentrations, 179186
microbial resistance, 198209
MSW/landfills, 176
point of departure, 190
predicted no-effect concentration, 190
recreational/subsistence anglers, 187
removal in WWTP and drinking water
treatment, 178
risk characterization metrics, 198208
substances selection, 188189
surface and drinking water, 194197
therapeutic doses, 189
thresholds of toxicologic concern, 194
veterinary pharmaceuticals, 176
environment, pharmaceutical introduction
routes
excretion, 172173
human active pharmaceutical
ingredients, 171, 172
improper disposal, 173174
waste stream manufacturing, 174175
seasonal variability, 175
uncertainty
exposure, 212214
main sources, 209
substance selection/hazard assessment,
209211
toxicity values, 211212
I
Ibuprofen and metaprolol, 171
Indirect potable reuse (IPR) treatment trains,
247249
Investigational New Drug (IND) applications,
54
M
Maximum expected environmental
concentration (MEEC), 143
Medication management strategies, 272275
Membrane treatment, 251
Methamphetamine, 171

300
N
National Association of Chain Drug Stores,
171
National Association of Pharmacy Regulatory
Authorities (NAPRA), 274
National Center for Health Statistics, 168
National Environmental Policy Act of 1969
(NEPA)
EIS process flowchart, 51, 52
environmental assessment, 51
history of, 5253
US FDA, 5253
National Poison Data System (NPDS), 259
National Survey on Drug Use and Health
(NSDUH), 261
O
Otolith geochemistry, 132133
Ozonation, 241242
P
Persistent bioaccumulative and toxic
chemicals (PBTs), 111
Pharamaceuticals environmental risk
assessment (PERA)
in Australia, 30
in Canada, 29
in European Union
adsorption and sediment/water fate
study, 27
biodegradability, 2627
decision tree, 2425
ecotoxicity, 27
guideline, 22
Phase 2 Tier A, 2324
Phase 2 Tier B, 2526
Q&ADoc, 28
registrations, 23
existing guidelines vs. EU biocides, 3941
in Japan, 2930
PAS, 17
production, 3435
REACH, 3233
Swedish environmental classification,
3132
in Switzerland, 29
in USA
Citizen Petition, 22
experimental evidence, 20
guidance, 20
water abstraction, human risk, 22
vs.veterinary pharmacueticals (see
Veterinary medicinal products)

Index
Pharmaceutical Research and Manufacturers
of America (PhRMA), 185
Pharmaceutical take back programs
at-home stockpiling prevention, 280
BAPPG, 263
gaps/weaknesses identification
medication management strategies,
272275
motivating factors, 264
risk communication and education,
270272
risk perception, 266269
scientific justification, 264265
Green Pharmacy Final Report, 278
Green Pharmacy Program, 277280
household hazardous waste collection
events, 277
monetary value of returned medications, 263
objectives
AAPCC, 259, 260
accidental poisoning, 259, 260
drug abuse, 261, 262
NPDS, 259
NSDUH, 261
over-the-counter medications, 260261
participation surveys, 262
PCC, 259
pharming, 261
poison-related fatalities, 260
public health issues, 263
therapeutic errors, 260
value of returned medication, 262
PH:ARM return mechanism, 263
potential roadblocks, 275276
public and environmental health problems,
281
returned medicines, 280
sustainable medicine, 277
Teleosis Institute Green Pharmacy
Program, 277279
Unused and Expired Medicine Registry,
278, 279
unwanted medications, collection centers,
276
unwanted medications, collection
container, 276, 277
Pharmacologically active substances (PAS)
analytical detections, 18
European water framework directive,
3334
formal guidelines, 18
population density, 17
Photodegradation, 177
Poison Control Centers (PCC), 259
Practice Greenhealth, 274

Index
S
Safe and Secure Drug Disposal Act, 257
Sedimentation, 234
Serotonin reuptake inhibitors (SSRIs), 86
Sub-lethal effects
ACR, 158
causality and confidence
biomarkers, 146147
general criteria/core elements, 145, 146
hypothetical concentrationresponse
curves, 145, 147
inverted biological cascade pyramid, 145
decision tree flowchart, 155, 156
omics technology, 157
in plants, 153155
receptor homology, 156
risk assessment
endocrine-sensitive endpoint, 144
EU aquatic ecotoxicology guidance, 143
MEEC, 143
US Environmental Protection Agency
(EPA), 141
VTG/testis-ova, 144
in vertebrates (fish) and invertebrates
AChE inhibition, 150151
Daphnia magna, 152153
estrogen antagonists, 149150
Nilaparvata lugens, 153
T
Teleosis Institute Green Pharmacy Program,
277279
Trace organic contaminant (TOrCs)
activated carbon adsorption, 234237
advanced oxidation processes, 246247
chlorination, 239241
coagulation/flocculation/sedimentation,
234
compounds detected, 229, 234
concentration in raw wastewater, 229
concentration in US source and finished
drinking water, 229, 231
concentration in US source water, finished
drinking water, and distribution
systems, 229, 231, 232
concentration in US wastewater treatment
plants, 229, 230
conventional wastewater treatment,
243244
indirect potable reuse treatment trains,
247249
membrane and MBR processes, 244245
optimization, 243
ozonation, 241242

301
physicochemical properties, 228
residual management, 249251
treatment efficacy, 234
UV photolysis, 237238
U
Ultraperformance liquid chromatography
(UPLC), 5
Unused and Expired Medicine Registry,
278, 279
US Drug Enforcement Administration (DEA),
257, 258
V
Veterinary API (vAPI), 176
Veterinary medicinal products (VMP)
aquatic effects testing, 38
endo/ecto-parasiticides, 37
Phase II Tier A, 36, 37
Phase II Tier B, 38
terrestrial ecotoxicity, 37
terrestrial side, 35
Vitellogenin
atrazine, 126
biomarker, 125
fathead minnow, 118
7-day deployments, 123, 124
4-day EE2 exposure, 123, 124
deploying indigenous fathead minnows,
120
experimental Lakes Area (ELA), 119
male Pearl Dace, 121, 122
reproductive failure, 123
summer and fall results, 120, 121
testicular tissues, 122, 123
gene and protein, 128129
stressor, 125
surrogate ecosystems, 125, 126
W
Wastewater and drinking water treatment
technologies
compound removal and transformation,
227
expansion and optimization, 227
herbicide atrazine, 227
TOrCs
activated carbon adsorption, 234237
advanced oxidation processes, 246247
chlorination, 239241
coagulation/flocculation/sedimentation,
234

302
Wastewater and drinking water treatment
technologies (cont.)
compounds detected, 229, 234
concentration in raw wastewater, 229
concentration in US source and finished
drinking water, 229, 231
concentration in US source water,
finished drinking water, and
distribution systems, 229, 231, 232
concentration in US wastewater
treatment plants, 229, 230
conventional wastewater treatment,
243244
IPR treatment trains, 247249
membrane and MBR processes, 244245
optimization, 243

Index
ozonation, 241242
physicochemical properties, 228
residual management, 249251
treatment efficacy, 234
UV photolysis, 237238
Waste water treatment plants (WWTPs)
API, 178
deconjugation, 211
degradation, 175
drinking water treatment, 178
Exposure reconstruction,
116117
landfills, 176
microbial resistance, 198209
semisolid waste, 172
surface and groundwater, 174