Clin Chem Lab Med 2015; 53(2): 299305

Carmen Perich, Joana Minchinela, Carmen Rics*, Pilar Fernndez-Calle, Virtudes Alvarez,
Mara Vicenta Domnech, Margarita Simn, Carmen Biosca, Beatriz Boned, Jos Vicente
Garca-Lario, Fernando Cava, Pilar Fernndez-Fernndez and Callum G. Fraser

Biological variation database: structure and

criteria used for generation and update
DOI 10.1515/cclm-2014-0739
Received July 17, 2014; accepted September 29, 2014; previously
published online November 21, 2014

*Corresponding author: Dra. Carmen Rics, Spanish Society of

Clinical Biochemistry and Molecular Pathology (SEQC), Analytical
Quality Commission, Plaza Gala Placidia 2, Barcelona, Spain,
E-mail: [email protected]
Carmen Perich: Spanish Society of Clinical Biochemistry and
Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Laboratoris Clnics, Hospital Vall dHebron,
Barcelona, Spain
Joana Minchinela: Spanish Society of Clinical Biochemistry and
Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Laboratori Clnic, Barcelons Nord i Valls
Oriental, Badalona, Spain
Pilar Fernndez-Calle: Spanish Society of Clinical Biochemistry
and Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Hospital Universitario La Paz, Madrid,
Spain
Virtudes Alvarez: Spanish Society of Clinical Biochemistry and
Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Laboratori Clnic, LHospitalet, Hospitalet de
Llobregat, Spain
Mara Vicenta Domnech and Pilar Fernndez-Fernndez: Spanish
Society of Clinical Biochemistry and Molecular Pathology (SEQC),
Analytical Quality Commission, Barcelona, Spain
Margarita Simn: Spanish Society of Clinical Biochemistry and
Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Consorci del Laboratori Intercomarcal de lAlt
Peneds, lAnoia i el Garraf, Barcelona, Spain
Carmen Biosca: Spanish Society of Clinical Biochemistry and
Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Hospital GermansTrias i Pujol, Badalona,
Spain
Beatriz Boned: Spanish Society of Clinical Biochemistry and
Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Hospital Royo Villanova, Zaragoza, Spain
Jos Vicente Garca-Lario: Spanish Society of Clinical Biochemistry
and Molecular Pathology (SEQC), Analytical Quality Commission,
Barcelona, Spain; and Hospital Universitario Virgen de las Nieves,
Granada, Spain
Fernando Cava: Laboratorio Clnico BR Salud, Hospital Infanta Sofa.
San Sebastin de los Reyes, Madrid, Spain
Callum G. Fraser: Centre for Research into Cancer Prevention and
Screening, University of Dundee, Dundee, Scotland, UK

Abstract
Background: Numerical data on the components of biological variation (BV) have many uses in laboratory
medicine, including in the setting of analytical quality
specifications, generation of reference change values and
assessment of the utility of conventional reference values.
Methods: Generation of a series of up-to-date comprehensive database of components of BV was initiated in 1997, integrating the more relevant information found in publications
concerning BV. A scoring system was designed to evaluate
the robustness of the data included. The database has been
updated every 2 years, made available on the Internet and
derived analytical quality specifications for imprecision, bias
and total allowable error included in the tabulation of data.
Results and conclusions: Our aim here is to document, in
detail, the methodology we used to evaluate the reliability
of the included data compiled from the published literature. To date, our approach has not been explicitly documented, although the principles have been presented at
many symposia, courses and conferences.
Keywords: biological variation; quality specifications; reference change values; reference intervals.
Abbreviations: BV, biological variation; CV, coefficient
of variation; CVI, within-subject coefficient of variation;
CVG, between-subject coefficient of variation; CVA, analytical coefficient of variation; CVI min and max, maximum
(CVI max) and minimum (CVI min) within-subject biological variation (CVI); EFLM, European Federation of Clinical
Chemistry and Laboratory Medicine; MM, mathematical
model score; PI, performance index score; RCV, reference
change value; SEQC, Spanish Society of Clinical Biochemistry and Molecular Pathology; STARD; standards for the
reporting of diagnostic accuracy studies.

Introduction
Data on the components of biological variation (BV),
namely, within-subject and between-subject BV, have

Brought to you by | Michigan State University

Authenticated
Download Date | 2/8/15 8:49 PM

300Perich etal.: Biological variation database generation

many uses in laboratory medicine, including derivation of
analytical quality specifications for imprecision, bias and
total allowable error, creation of reference change values to
objectively assess changes in serial results from an individual and examination of the utility of conventional population-based reference values [1]. Since derivation of BV is not
without difficulties and requires expenditure of considerable resources of many types, a need for a database of published information found in the literature was perceived
some years ago. As a consequence, compilations of data
were generated by a number of authors [24]. However,
these databases simply provided lists of published data.
It was decided that publication of a database giving
one value for each analyte studied, based upon objective
assessment of the reliability of the data in the literature,
and updated regularly, would be of major benefit.
Generation of a comprehensive database of components of BV was initiated in 1997 by the Analytical Quality
Commission of the Spanish Society of Clinical Chemistry
(SEQC), integrating the more relevant information found
in publications concerning BV. A scoring system was
designed to evaluate the robustness of data included [5].
The database has been regularly updated and derived
analytical quality specifications for imprecision, bias and
total allowable error included in the tabulation. Every 2
years, an update of this table has been made available
on the Internet, and the information documented comprises a brief introduction concerning the aims of provision of the database, the changes that have been made as
compared with the previous edition and three important
appendices, namely:
1) a list of the analytes studied with the within- and
between-subject components of BV expressed as
coefficients of variation (CVI and CVG, respectively),
the derived analytical quality specifications for
imprecision, bias and total allowable error at three
different levels (desirable, minimum and optimum)
and the number of publications examined for each
analyte;
2) a list of the publications examined, by analyte; and
3) a list of the full citation for every publication included.
The acronyms for analytes are listed in the Supplementary
Material, Table 1.
The derivation of analytical quality specifications
was done using only data from publications dealing with
apparently healthy adults.
The first version of this table was presented at the
Consensus Conference on Quality Specifications in Laboratory Medicine held in Stockholm in 1999 [5]; the currently available database is the 8th edition [6, 7].

Our aim in this paper is to describe in detail the methodology used to document the criteria that have been
applied for inclusion or exclusion of the published data in
the database. To date, this has never been published in an
explicit and comprehensive way, although it has been presented at many symposia, courses and conferences over
the last 15 years.

Materials and methods

Examination of the 2014 edition is documented here to detail the generation of the database [6, 7]. Included are descriptions of the above
mentioned table of data on BV together with the derived analytical
quality specifications and other additional information that has not
been published previously. These comprise the criteria used in the
evaluation of a publication to decide whether or not the data therein
are to be included in the BV database.

Information encompassed in the database

For each analyte, the information included is:
1) identification of the group member that recorded the
information;
2) type of analytical imprecision obtained in the research (withinrun, between-run, day to day), expressed in terms of CVA;
3) index of individuality (CVI/CVG or (CVI2+CVA2)1/2/CVG);
4) reference change value (RCV, e.g., for p<0.05, bidirectional,
RCV=2.77 (CVI2+CVA2);
5) analytical quality specification for imprecision (0.5*CVI);
6) analytical quality specification for bias (0.25*CVI2+CVG2)1/2;
7) number of subjects studied (total and partitioned by gender);
8) number of samples obtained per subject;
9) length of the study (expressed in days, hours, weeks, or months);
10) documentation of central tendency (median) and units used for
the analyte examined in the group of subjects;
11) type of population studied (healthy, relevant pathology); and
12) full citation, including authors, journal title, year, volume and
pages.

Criteria for inclusion

The following provides a comprehensive explanation concerning
the criteria used for inclusion of data in the database, which demonstrates the robustness of the BV data made available. These are:
1) The criterion used to include data in a publication: only data
from studies specifically designed to estimate the components
of BV have been included.
2) The principle used to derive the data in each publication: a combination of two criteria has been applied.
a) Performance Index: PI=CVA/0.5*CVI, first defined as the
Index of Fiduciality [8], stratified according to the following reliability decreasing score:
i) Score 2: PI <1;
ii) Score 1: PI between 1 and 2;
iii) Score 0: PI >2 or unknown.

Brought to you by | Michigan State University

Authenticated
Download Date | 2/8/15 8:49 PM

Perich etal.: Biological variation database generation301

In general, data with score 0 were not included in the final calculations, unless the second criterion, below, was fulfilled and no
more publications concerning the analyte were available.
b) Mathematical model (MM) used by the authors to calculate
CVI and CVG:
i) Score 4: ANOVA;
ii) Score 3: model described by Fraser and Harris [1, 9];
iii) Score 2: unclear model;
iv) Score 1: not described model.
Data with score 1 were not included in the final calculations.
3) Separation of data derived from studies on apparently healthy
adults: data derived from studies on subjects with documented
pathology are compiled in a different database.
4) Generation of estimates of CVI and CVG, using the median of all
data compiled.

Results
The database is maintained in a spreadsheet (MS Excel)
containing information on all the analytes studied. An
example of the database content is shown in Table 1. The
analyte selected is serum calcium, because the number of
available publications on BV is not so large as to make the
table difficult to read and understand. Once data generated
from subjects with documented pathology are separated to
a different file (data not shown in Table 1), the medians of

estimates from apparently healthy subjects are considered

to be the best available estimates of CVI and CVG.
In Table 1, each row corresponds to a single subject
group, which, in the case of serum calcium, comprises
42 groups. A cell without data indicates that the required
information was not given in the original publication. Some
rows are crossed out with horizontal bars, indicating that
the data therein were excluded from the final calculations,
either because the PI is higher than 2 (Eckfeldt, Gallagher,
Gowans, Harris, Juan, Pineda, Van Steirteghem, Winkel),
the CVA was not documented, or the MM score was 1. In
consequence, the data in these rows were not included in
for the calculation of the median CVI and CVG.
Data in Table 1 are ordered according the author
and publication; also the reference numbers are shown
[1021].
Other information also included in the database,
but not shown in Table 1 to simplify the current description, are: identification of the member of the Analytical
Quality Commission group that recorded the data in the
database, details of the calculations of the data recorded
in each row, such as desirable analytical quality specifications for imprecision, bias and total allowable error,
Index of Individuality and full citation details of the
publication.

Table 1An example of the raw data tabulation for serum calcium derived from publications on biological variation.
CVI

CVG

CVA

PI

1.80
2.20
2.60
1.60
3.25
4.77
2.39
2.10
1.62
1.20
1.60
2.50
1.60
1.70
2.50
3.10
2.10

2.80

3.78
1.50

3.30
1.50
2.30
2.90
3.70
2.70

0.80
0.80
0.80
1.00
0.90
2.60
0.90
1.20
0.98
1.00
1.00
1.10
0.80
1.00
1.00
1.80
2.00

0.9
0.7
0.6
1.3
0.6
1.1
0.8
1.1
1.2
1.7
1.3
0.9
1.0
1.2
0.8
1.2
1.9

62
148
126
27
5
5
15
15
10
14
10
60
37
71
64
10
1105

Time

S/sub

Mean

1d

6m

6m

20 w
5d

5w

4w

40 w
8w

8w

8w

135 m
22 w
28 d
28 d
5d

8.5 w

4
6
6
10
5
5
4
10
8
8
8
2
22
7
7
5
9

2.5
2.4
2.4
2.36
2.5
2.5
2.4
2.41
2.33
2.29
2.32
2.41
2.4

2.4

Units

MM

Health
status

Year of
publication

mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mmol/L
mg/dL
mmol/L

3
3
3
4
4
4
3
3
3
3
3
4
3
4
4
3
2

Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy
Healthy

1985
1985
1985
1989
1989
1989
1991
1987
1988
1987
1987
1990
1977
2013
2013
1986
1978

First author
Costongs [10]
Costongs [10]
Costongs [10]
Fraser [11]
Gallagher [12]
Gallagher [12]
Gonzalez-Revelderia [13]
Gowans [14]
Holzel [15]
Holzel [16]
Holzel [16]
Morris [17]
Pickup [18]
Pineda [19]
Pineda [19]
Rics [20]
Williams [21]

Cells without data indicate that the required information was not given in the original publication. Rows crossed out with horizontal bars,
indicate that the data therein were excluded from the final calculations, either because the PI is higher than 2 (Eckfeldt, Gallagher, Gowans,
Harris, Juan, Pineda, Van Steirteghem, Winkel), the CVA was not documented, or the MM score was 1. In consequence, the data in these rows
were not included in for the calculation of the median CVI and CVG. CVI, within-subject coefficient of variation; CVA, analytical coefficient of
variation; CVG, between-subject coefficient of variation; MM, mathematical model score; N, number of subjects studied; PI, performance
index score; S/sub, number of samples per subject; Time, duration of the study (h: hours, d: days, w: weeks, m: months, y: years).

Brought to you by | Michigan State University

Authenticated
Download Date | 2/8/15 8:49 PM

302Perich etal.: Biological variation database generation

Discussion

Table 2CVI estimates and range for analytes with more than nine
publications and the corresponding CVI range.

Number of publications compiled for each

analyte
From a total of 358 analytes included in the 8th edition,
data for 202 were found in a single publication, 129 have
data in from 2 to 9 publications and 27 have data in 10 or
more publications, the maximum in the present edition
being 45 publications.
Regarding the 202 analytes with only data in only one
publication, CVI and CVG are simply estimated in a single
study. This could constitute a weakness: however, 55% of
the analytes in this group had maximum scores (5 or 6) for
the sum of the two numerical criteria described in the Materials and methods above, indicating that the estimates of CVI
and CVG can be considered reliable. The remaining analytes
(45%) are considered to be the weakest aspect of the database; however, they have been included in the database to
avoid losing valuable and sometimes difficult to find information and make this easily available. However, we consider that further studies should be performed to improve
the accuracy of these estimates of BV in future editions of our
database.
Regarding the 129 analytes with the number of publications from 2 to 9, 36% had high scores (PI=2 and
MM=3 or 4). Of the remaining 64%, 48% obtained a
lower score for both criteria, and, in consequence, this
group of analytes might be suspected of having less
than ideal estimates of CVI and CVG. In order to examine
the quality of the data documented for this group, a
new criterion was used when preparing this current
exploration and discussion of the generation of the
database: the ratio between the maximum CVI (CVI max)
and the minimum CVI (CVI min) found for each analyte
and subject group.
For the 27 analytes with more than nine publications
included, the total score (sum of PI and MM) was from 5 to
6; thus indicating that the CVI and CVG estimates are valid.
An additional calculation has now been made for the analytes in this group: the CVI range, as shown in Table 2.
None of the publications compiled include the confidence interval of the CII estimated, as has been recently
recommended by Roraas etal. [22].

Distribution of CVI values assembled

For analytes with two to nine publications the ratio CVI max/
CVI min was calculated. Figure 1 shows these ratios and

System Analyte
SSSSSSSSSSSSSBBSSSSSSSSSSSS-

Papers, Groups, CVI

n
n

Alanine

aminotransferase
Albumin

Alkaline phosphatase
Apolipoprotein A1

Aspartate

aminotransferase
Bilirubin, total

Calcium

Carcinoembryonic

Antigen
Chloride

Cholesterol

Creatinine

Glucose

HDL Cholesterol

Hematocrit

Hemoglobin

Iron

Lactate

dehydrogenase
Phosphate

Potassium

Protein, total

Sodium

Thyroxin

Triglyceride

Triiodothyronin

Urate

Urea

-Glutamyltransferase

CVI
range

10

19 17.3 5.358

25
21
11
13

42
32
15
22

3.1
6.5
6.5
12.8

1.511
2.713
3.610
8.224

11
24
10

16 21.8 1228
38 1.8 0.56.5
10 12.7 8.415

19
45
29
16
25
12
13
11
11

29
72
47
23
50
22
30
18
12

1.3
6.1
5.3
5.5
7.3
2.7
2.8
26.5
8.6

0.83.4
1.59.7
2.213
3.713
3.516
1.06.3
1.14.5
7.336
4.215

17
16
18
17
10
32
9
18
11
10

24
23
28
23
11
57
12
33
15
15

8.1
4.6
2.7
0.6
4.9
21.2
6.9
8.6
12.2
13.4

4.810
3.06.3
1.45.7
0.21.4
3.510
1239
5.710
2.517
4.816
3.935

Analytes are shown in alphabetical order. Groups, n, number of CVI

considered.

it is apparent that it is from 1 to 2 for 55% of the analytes included in this group (Figure 1A), from 2 to 4 for
31% (Figure 1B) and from 4 to 7 for the remaining analytes (Figure 1C). This suggests that the distributions of
CVI values for the 86% of the analytes studied are generally quite compact and gives an idea of that differences
among publications are not important. There are four
exceptions, not shown in Figure 1 (urine proteins and
serum C-peptide, copper and ferritin, with ratios from 11
to 44) in which the estimated CVI could be considered as
less than robust.
Moreover, no significant relationships between the
CVI max/CVI min ratio and the number of publications or the
number of subject groups were evident.
For analytes with more than nine publications the
ratio CVI max/CVI min was from 1 to 7 indicating a compact

Brought to you by | Michigan State University

Authenticated
Download Date | 2/8/15 8:49 PM

Perich etal.: Biological variation database generation303

2.0

CVmax/CVmin

1.8
1.6
1.4
1.2
SHBG
p-Amy
Ig-l
Ca19,9
Endop
Crea
VLDLc
a2-Glo
DHEAs
Prol
HDL-2
LDH-5
IgM
Fibr
Lym-c
C3
Sod-bi
U-Ure
Aldo
FT3
F-Car
Cys
IgA
a1-Glo
Transf
Neu-c
CKMB-a
MCH
CRP
Zn
U-p-Amy
MCV
U-Pho
b-Glob
Pla-c
U-Alb
U-Ura
a1-Antrip
CK
TSH
Thyrog
Mon-c
MPV
Eos-c
U-Sod
Ery-Mg
b-Caro
F-VW
U-Mg
LDH-4
An T-III
Leu-c
Cy21,1
Ca++
a1-AcGlyc
Lyco
Testo
LDH-2
a-Toco
Cerul
Insu
NpBNP
Choli-a
IGF-1

1.0

4.0

CVmax/CVmin

3.6
3.2
2.8
2.4
2.0

CVmax/CVmin

LH

ApoB

APTT

DeoxPyr

Ca15.3

17OHP

Ery

Placrit

Homocy

Andr-dio

Lp-a

ACP

U-Cal

a-Amy

a2-Mac

Prothr-t

ALP-bone

TBG

cT-clTP

Bili-c

5Nuc

T-Carn

Cort

Estra

VitC

Fruc

Reti

ACP-tr

Lip

FSH

Baso-c

IgG

Oste

NAC

C4

Myog

NTIC

Choli-c

g-Glob

Ig-k

LDLC-m

AFP

1.6

8.0
7.0
6.0
5.0
4.0
PINP

FT4

MCHC

Hapt

TropI

LDH-3 RCDw HDL-3

PICP

s-Mg

PSA

TropT

U-Pot

LDH-1 HbA1c Ca125

F-Tes Pla-wide MCA

Figure 1Ratio between maximum (CVI max) and minimum (CVI min) within-subject biological variation (CVI) documented for each analyte with
number of publications from 2 to 9.
(A) Analytes with CVmax/CVmin from 1 to 2. (B) Analytes with CVmax/CVmin from 2 to 4. (C) Analytes with CVmax/CVmin from 5 to 7.

distribution of CVI values. Exceptions are serum albumin,

calcium, ALT and GGT with higher ratios, as shown in Figure
2. However, for these analytes, when the extreme values
were eliminated, the median CVI did not change and, consequently, such extreme values should be included, exactly
as was done in the previous edition of the database.

Weaknesses of this database

Some limitations of our database are due to the paucity
of studies, e.g., the minimal existence of data on children
[23], the absence of data on some analytes commonly
examined in medical laboratories and some studies
with few subjects included. To increase the availability
of data on BV for analytes with poor or no information
existing at present, the group believes that a standardized method to estimate CVI and CVG using serial results
from patients derived from laboratory information
systems should be developed. This approach has been
demonstrated by Cembrowski etal. [24], studying nine

analytes and obtaining estimates of CVI and CVG of the

same numerical order as those documented in our database at that time.
Recently, a number of criticisms regarding the validity
of using published data on BV have appeared and these
include concerns regarding use of our database [2528]. In
consequence, an international group of experts on laboratory quality met in 2010 with the aim of debating the
validity of continuing to use analytical quality specifications derived from BV, 10years after the Consensus Conference [29]; the contributions revealed that the majority
of the experts present did use analytical quality specifications derived from the BV database. However, the group
pointed out three weak points that required attention of
professionals in laboratory medicine, namely:
the less than ideal estimates of analytical quality
specifications for a number of analytes due to the
paucity of publications available for inclusion in the
database;
data are available only for a somewhat limited number of analytes (the 2014 edition includes 358),

Brought to you by | Michigan State University

Authenticated
Download Date | 2/8/15 8:49 PM

304Perich etal.: Biological variation database generation

BV based on data available on the existing spectrum
of laboratory information systems. Moreover, the criteria used to select the more reliable publications dealing
with estimation of the components of BV should be better
developed on a basis of some international team, as in the
case of the BV working group of the European Federation
of Laboratory Medicine (EFLM) [32, 33]. We are ready and
willing for collaboration with such worthy initiatives.

12
Albumin

10

Calcium
Hematocrit

CVI

8
6
4
2
0
Different publications compiled

60
ALT

50

GGT

CVI

40
30

Author contributions: All the authors have accepted

responsibility for the entire content of this submitted
manuscript and approved submission.
Financial support: None declared.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played
no role in the study design; in the collection, analysis, and
interpretation of data; in the writing of the report; or in the
decision to submit the report for publication.

20

References

10
0
Different publications compiled

Figure 2Distribution of estimates of within-subject biological variation (CVI) for analytes with more than nine publications and with
apparently widely dispersed CVI.

whereas the test repertoires of many medical laboratories include approximately 1000 examinations; and
total allowable error derived from BV seems too
restrictive for some analytes as compared with current technological capability (e.g., serum sodium,
albumin and chloride and blood HbA1c) and too loose
for others with high BV (e.g., serum triglycerides and
urea).
Moreover, Carobene et al. [30] examined, in detail, all
accessible publications on the BV of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and
-glutamyltransferase (GGT). They found some discrepancies among the data in the publications; they suggested the need to develop a checklist such as that used
in the preparation of publications on diagnostic accuracy
studies, namely, the STARD guidelines [31], and assist in
the highest quality generation, compilation and publication of estimates of the components of BV.
We believe that this type of work would be improved
by preparing international guidelines on how to estimate

1. Fraser CG. Biological variation: from principles to practice. Washington DC: AACC Press, 2001.
2. Fraser CG. The application of theoretical goals based on biological variation data in proficiency testing. Arch Pathol Lab Med
1988;112:40415.
3. Fraser CG. Biological variation in clinical chemistry. An update:
collated data, 19881991. Arch Pathol Lab Med 1992;116:91623.
4. Sebastin-Gmbaro MA, Lirn-Hernndez FJ, Fuentes-Arderiu X.
Intra- and inter-individual biological variability data bank. Eur J
Clin Chem Clin Biochem 1997;35:84552.
5. Rics C, lvarez V, Cava F, Garca-Lario JV, Hernndez A, Jimnez
CV, etal. Current databases on biological variation: pros, cons
and progress. Scand J Clin Lab Invest 1999;59:491500.
6. Minchinela J, Rics C, Perich C, Fernndez-Calle P, Alvarez V,
Domnech MV, etal. Biological variation database and quality
specifications for imprecision, bias and total error (desirable and
minimum). The 2014 update. Available from: http://www.westgard.com/biodatabase-2014-update.htm. Accessed 7 July, 2014.
7. Minchinela J, Rics C, Perich C, Fernndez-Calle P, Alvarez V,
Domnech MV, etal. Base de datos de los componentes de Variacin Biolgica, con las especificaciones de la calidad analtica
(deseable, mnima y ptima). Actualizacin del ao 2014.
Available from: http:// www.seqc.es/es/Comisiones/18/9/102 /
Base_de_datos_de_Variacion_biologica_%7C_Comision_de_Calidad_Analitica_%7C_Comite_Cientifico/. Accessed 7 July, 2014.
8. Fraser CG, Browning MC. The index of fiduciality proposed
for use in evaluation and comparison of methods. Clin Chem
1988;34:13567.
9. Fraser CG, Harris EK. Generation and application of data on
biological variation in clinical chemistry. Crit Rev Clin Lab Sci
1989;27:40937.

Brought to you by | Michigan State University

Authenticated
Download Date | 2/8/15 8:49 PM

Perich etal.: Biological variation database generation305

10. Costongs GM, Janson PC, Bas BM, Hermans J, Van Werch JW,
Brombacher PJ. Short-term and long-term intraindividual variations and critical differences of clinical chemistry laboratory
parameters. J Clin Chem Clin Biochem 1985;23:716.
11. Fraser CG, Cummings ST, Wilkinson SP, Neville RG, Knox JD, Ho
O, etal. biological variability of 26 clinical chemistry analytes in
elderly people. Clin Chem 1989,35:7836.
12. Gallagher SK, Johnson LK, Milne DB. Short-term and long-term
variability of indices related to nutritional status. I: Ca, Cu, Fe,
Mg and Zn. Clin Chem 1989;35:36973.
13. Gonzlez-Revalderia J, Garca-Bermejo S, Menchn-Herreros A,
Fernandez-Rodriguez E. Towards narrower analytical goals in
routine laboratory work. Clin Chem 1991;37:596.
14. Gowans EM, Fraser CG. Long-term biological variability of commonly analyzed serum constituents. Clin Chem 1987;33:717.
15. Hlzel WG. Influence of hypertension and antihypertensive
drugs on the biological intra-individual variation of electrolytes
and lipids in serum. Clin Chem 1988;34:14858.
16. Hlzel WG. Intra-individual biological variation of some analytes
in serum of patients with chronic renal failure. Clin Chem
1987;33:6703.
17. Morris HA, Wishart JM, Horowitz M, Need AG, Nordin BE. The
reproducibility of bone-related biochemistry variables in postmenopausal women. Ann Clin Biochem 1990;27:5628.
18. Pickup JF, Harris EK, Kearns M, Brown SS. Intra-individual variation of some serum constituents and its relevance to population-based reference ranges. Clin Chem 1977;23:84250.
19. Pineda-Tenor D, Laserna-Mendieta EJ, Timn-Zapata J, RodelgoJimnez L, Ramos-Corral L, Recio-Montealegre A, etal. Biological variation and reference change value of common chemistry
and haematologic laboratory analytes in the elderly population.
Clin Chem Lab Med 2013;51:85162.
20. Rics C, Garca-Arum E, Rodriguez-Rubio R, Schwartz S. Eficacia de un programa interno de control de calidad. Quim Clin
1986;5:15965.
21. Williams GZ, Widdowson GM, Penton J. individual character variation in time-series studies of healthy people. II. Differences in
values for clinical chemistry analytes in serum among demographic groups, by age and sex. Clin Chem 1978;24:31320.
22. Roraas R, Petersen HP, Sandberg S. Confidence intervals and
power calculations for within-subject biological variation: effect
of analytical variation, number of replicates, number of samples
and number of individuals. Clin Chem 2012;58:130613.
23. Bailey D, Bevilcqua V, Colantonio DA, Pasic MD, Perumal N,
Chan M, etal. Pediatric within-day biological variation and

quality specifications for 38 biochemical markers in the CALIPER

cohort. Clin Chem 2014;60:51829.
24. Cembrowski GS, Tran DV, Higgins TN. The use of serial patient
blood gas, electrolyte and glucose results to derive biological
variation: a new tool to assess the acceptability of intensive
care unit testing. Clin Chem Lab Med 2010;48:144754.
25. Braga F, Dolci A, Mosca A, Panteghini M. Biological variation of
glycated hemoglobin. Clin Chim Acta 2010;411:100610.
26. Carlsten S, Hyltoft Petersen P, Skeie S, Skadberg O, Sandberg
S. Within subject biological variation of glucose and HbA1c in
healthy persons and type I diabetes patients. Clin Chem Lab
Med 2011;49:15017.
27. Carobene A, Graziani MS, Lo Cascio C, Tretti L, Cremonese E, Yabarek T, etal. Age dependence of within-subject BV of nine common
clinical chemistry analytes. Clin Chem Lab Med 2012;50:8414.
28. Braga F, Panteghini M. Standardization and analytical goals
for glycated hemoglobin measurement. Clin Chem Lab Med
2013;51:171926.
29. Cooper G, DeJonge N, Ehrmeyer S, Yundt-Pacheco J, Jansen R,
Rics C, etal. Collective opinion paper on findings of the 2010
convocation of experts on laboratory quality. Clin Chem Lab
Med 2011;49:793802.
30. Carobene A, Braga F, Roraas T, Sandberg S, Bartlett WA. A
systematic review of data on biological variation for alanine aminotransferase, aspartate aminotransferase and
-glutamyltransferase. Clin Chem Lab Med 2013;51:19972007.
31. STARD guidelines. Available from: http://www.stard-statement.
org/. Accessed 7 July, 2014.
32. Bartlett W, Braga F, Carobene A, Coskun A, Prusa R, FernandezCalle P, etal. Identification of key meta data to enable safe
accurate and effective transferability of biological variation
data. 2014 American Association of Clinical Chemistry Annual
Meeting AACC, Chicago, Illinois. Available from: http://www.
aacc.org/events/2014_annual_meeting/abstracts/Documents/
AACC_14_AM_A20.pdf. Accessed 8 July, 2014.
33. Bartlett W, Braga F, Carobene A, Coskun A, Prusa R, FernandezCalle P, etal. Definition of a minimum data set to accompany
indices of biological variation. 2014 International Federation of
Clinical Chemistry Poster Abstract -IFCC WorldLab Istanbul 2014.
Clin Chem Lab Med 2014;52(Suppl):S315.

Supplemental Material: The online version of this article (DOI:

10.1515/cclm-2014-0739) offers supplementary material, available
to authorized users.

Brought to you by | Michigan State University

Authenticated
Download Date | 2/8/15 8:49 PM