Experimental Studies: Randomized Controlled Trials, Field Trials, Community Trials
Experimental Studies: Randomized Controlled Trials, Field Trials, Community Trials
CONTROLLED TRIALS, FIELD TRIALS,
COMMUNITY TRIALS
Abdullatif Husseini, MSc, MPH, PhD
Learning Objectives
• To explore various experimental studies
• To describe the important elements of
randomized trials.
• To define the purpose of randomization and
of masking.
• To introduce design issues related to
experimental design.
Gordis, 2014
Epidemiological study designs
Study type
Experimental Observational
Case report, case series, Case control,
cross sectional cohort, ecological
3
Lind conducted an experiment, which he described as follows:
“I took 12 patients in the scurvy on board the Salisbury at sea. The
cases were as similar as I could have them…they lay together in one
place and had one diet common to them all. Two of these were
ordered a quart of cider per day…Two others took 25 gutts of elixir
vitriol…Two others took two spoonfuls of vinegar…Two were put
under a course of sea water…Two others had two oranges and one
lemon given them each day… Two others took the bigness of
nutmeg. The most sudden and visible good effects were perceived
from the use of oranges and lemons, one of those who had taken
them being at the end of 6 days fit for duty…The other…was
appointed nurse to the rest of the sick.”
Gordis, 2014
Experimental studies
“ A study in which conditions are under the direct control of the
investigator. In epidemiology, a study in which a population is
selected for a planned trial of a regimen whose effects are
measured by comparing the effects of the regimen in the
experimental group with the outcome of another regimen in a
control group. To avoid bias members of the experimental and
control groups should be comparable except in the regimen that is
offered (to) them. Allocation of individuals to experimental or
control groups is ideally by randomization ”
(Last 1995)
5
A simpler definition
“ An experimental study assigns participants to
intervention and control groups in order to examine
whether an intervention causes an intended outcome”
(Jacobsen 2012)
• Which one is a better definition in your opinion? and
why?
6
Characteristics of experimental studies
• Objective: comparing outcomes between intervention
and control groups
• Study question: Is there a causal relation between the
outcome and the exposure
• Population: Individuals with similar characteristics
• Used for measuring relations and effects
• Beware of non compliance with the interventions
(Source: Jacobsen 2012) 7
Experimental/ intervention studies
8
Prophylactic vs. Therapeutic trial
•Prophylactic trial: designed to evaluate the effectiveness of a
substance (vaccine) or program (patient education) to prevent
a disease
•Therapeutic: study of curative drugs or a new medical
procedure to assess whether they can improve human health
Types of ES
• Clinical trials (most common type, e.g. aspirin in
reducing mortality after heart attacks)
• Field studies (e.g. testing of vaccines in military
conscripts)
• Community studies (e.g. effectiveness of health
education campaigns in smoking cessation and
prevention in different cities)
What ES can involve…
Targets Interventions
• Individuals • Screening/diagnosis
• Communities • Treatment
• Cities • Prevention
Fields Outcomes
• Medicine • Life
• Public health • Death
• Related fields • Anything in between
(Jabbour, 2004)
Experimental (randomized) Studies
•Experimental designs help us overcome many of the problems
of observational studies related to confounding and bias
•Because of that, experimental designs can more confidently
attribute cause and effect to associations than non‐
experimental designs
•Although it is a very strong design, it is not always feasible or
ethical (give examples)
Randomization
•Assigning subjects to the treatment or control
conditions of a clinical trial randomly
•Well‐implemented randomization avoids the effects
of confounding by balancing the control and
treatment groups in terms of characteristics that
could affect the outcome.
Randomized controlled trial
“A randomized controlled trial is an epidemiological
experiment designed to study the effects of a particular
intervention, usually a treatment for a specific disease
(clinical trial). Subjects in the study population are
randomly allocated to intervention and control groups, and
the results are assessed by comparing outcomes. To
ensure that the groups being compared are equivalent,
patients are allocated to them randomly, i.e. by chance.”
(Bonita et al 2006)
14
For example an RCT to measure the effect of a
new blood pressure treatment
Bonita el al 1993
Randomized controlled trial (RCT)
• Considered the strongest study design or the golden standard of
epidemiology
• Example: We have a new drug for lowering blood pressure (BP)
(A).
• We divide people with hypertension randomly into two groups:
the intervention group we give then the new drug A and the
control group we give them the old drug B.
• Then we observe the mean reduction in BP (for A= 5 mm Hg ;
for B= 2 mm Hg)
16
Efficacy of propranolol in treatment of heart attacks
among 91 males (45‐60 years) with MI
Outcome (alive)
(28 days after admission)
(Jabbour, 2004)
Field trial
• “Field trials, in contrast to clinical trials, involve people
who are healthy but presumed to be at risk; data
collection takes place “in the field”, usually among non‐
institutionalized people in the general population.
• Field trials can be used to evaluate interventions aimed at
reducing exposure without necessarily measuring the
occurrence of health effects.”
(Bonita et al 2006)
18
Field trial
Bonita et al, 2006
An example of a field trial
• We have famers living in an endemic area of malaria
• We divide them into two groups, one group we will give
the new vaccine and the other a placebo
• After one year we measure the incidence of malaria.
• Vaccine group 5%, Placebo group 10%
• This means that the vaccine reduced malaria incidence
by 50%.
20
Community trial
• “ Experiment in which the unit of allocation to receive a
preventive or therapeutic regimen is an entire
community or political subdivision” (Last 1995)
• Example: community trial of a tuberculosis outreach
programme in rural Ethiopia. (Bonita 2006)
21
Another example of a community trial
• We have two adjacent villages
• In village A we fluoridated their drinking water
• In village B we left the water as is
• After a year we looked at the % of caries in village A
children (2%) compare to (6%) in village B children
• So fluoridation reduced the caries among children by ?
22
Phases in testing new drugs
• Animal studies: Assess efficacy and tolerability
• Phase I: Clinical pharmacological studies
• Phase II: Initial clinical investigation for treatment effects (preliminary
studies)
• Phase III: Full‐scale evaluation of treatment effects (trials)
• Phase IV: Post‐marketing surveillance
Phases of a Clinical Trial
Source: http://www.nlm.nih.gov/services/ctphases.html
II Larger group of people Test effectiveness of drug or treatment and
further evaluate its safety
•Operated by the US National Institutes of Health
•“ a registry and results database of publicly and privately
supported clinical studies of human participants conducted
around the world.”
•Clinicaltrials.gov
http://clinicaltrials.gov/
BIOM 530
Design issues
ES can (be):
• randomized or non‐randomized (then called quasi‐
experimental),
• controlled or un‐controlled,
• use placebo or active treatment,
• blind subjects, investigator, statisticians or none.
Statement of research questions
• Is intervention A 'superior‘ to intervention B in
delivery the desired outcome?
• Define superior: which endpoints are most clinically
significant, and which of these can be measured?
• Two ways of assessing outcomes
• Risk (proportion)
• Rate (person‐time)
Randomization
• Randomization is the random assignment of
interventions to study subjects.
• It eliminates the possibility of knowing what the
assignment of the next subject will be.
• The critical element of randomization is the
unpredictability of the next assignment.
Subjects should register for the trial BEFORE randomization
Otherwise, we may loose them before we start!
There are many ways of randomization:
Two most common: advance (assignment generated beforehand) or
dynamic (assignment generated on the spot)
To assess whether randomization was successful in having
comparable groups:
Assess the equality of intervention groups with regards to baseline
characteristics e.g. demographic, clinical and prognostic factors
Cross‐over trials
• During the trial, a subject crosses over from one group to another (i.e.
from group A to group B and the reverse)
• Each subject receives both A and B treatments (or non‐treatment)
• In this design, subjects are their own control
• Data analysis is matched, but there might by “carry‐over” or residual
effects from the first treatment received
Non‐randomized (Quasi‐experimental) Studies
• Sometimes, it is not possible to allocate individuals or groups to
treatment and control groups randomly. It may not be practical, or it
may be unethical
• The absence of randomization means that we have to look more
carefully at the effects of possible confounding. This can be done in the
design phase and/or in the analysis
Blinding (masking)
• Blinding means that certain persons do not know the treatment
assignment.
• Commonly both the investigators and subjects are blinded (double blinding)
• The aim is to reduce subject and investigator bias. This is of particular
importance when the outcome is a subjective measure, such as headache or
low back pain.
• Who could be blinded?
• Subjects
• Treating clinicians
• Investigators (especially those who assess the outcomes)
• Statisticians
Problems Addressed with Blinding
• Subjects in a trial try to get well or may feel differently if they know what
they are taking (Placebo Effect)
Blind subjects
• Placebo effect is especially important when using subjective outcomes.
• If we are comparing a drug to no drug, a placebo is commonly used. This
improves comparability of treatment groups in terms of compliance & side
effects.
• In non‐drug trials, may be impossible / unethical to have a placebo.
• One treatment may have a characteristic side effect which may make blinding
ineffective.
Problems Addressed with Blinding
• Assessing investigators may try hard to look for outcomes if they know
assignment (Unconscious bias)
Blind assessing investigators
• Decisions about ancillary treatment, follow‐up, severity of
toxicity…often made by those who know treatment assignment
( Unconscious bias)
Blind treating physicians
• Statisticians try to please investigators by “looking hard” for results
Blind the statistician
Key issues in analysis
Problem Way to address
Comparability of treatment groups •Good randomization scheme
•Stratified design
Control for confounders •Stratified analysis
•Multivariate analysis
Potential loss to follow‐up •Intention to treat analysis “Once randomized,
always analyzed”
(Jabbour, 2004)
Analysis method is dictated by study endpoint
Study endpoint Type of analysis
•Binary (yes/no, categorical) Comparison of risks (Relative risk)
•time to event NOT relevant
•Binary (yes/no, categorical) Comparison of rates (Rate ratio),
•time to event is relevant survival analysis (Kaplan‐Meier
curve)
•Continuous comparison of two means (t‐test, CI)
(Jabbour, 2004)
Eligibility Criteria
• The criteria that determine whether an individual can be included in
the sample or not
• Eligibility criteria must be carefully defined, explicitly described, and
enforced
• Inclusion criteria vary according to the intervention being proposed
• Consult the literature and the records of previous clinical trials to see
how your inclusion/exclusion criteria compare
Ethical issues
• Equipoise:
We must NOT withhold a potentially effective interventions from subjects,
particularly those with serious and life‐threatening diseases. Therefore, RCTs are
ethical only when we do NOT know that drug A is better than drug B
• Informed consent:
should be obtained from the well‐informed patient or his family and from the
attending physician.
• Stoppage rules:
There is an ethical obligation to monitor and analyze data regularly (interim
analyses) to check for evidence of adverse or beneficial effect of one of the
interventions.
If one arms proved unequivocally better than the other, the study should be
stopped and subjects informed (Data Safety & Monitoring Committee).
In summary… ES
ADVANTAGES DISADVANTAGES
• Randomization tends to balance • Subject exclusions may limit ability to
prognostic factors across study groups generalize findings to other patients
• Detailed information can be collected • A long period of follow‐up may be
on baseline and subsequent required to reach a conclusion
characteristics of participants • A large number of participants may
• Dose level can be determined by the be required
investigator • Financial costs are typically high
• Blinding can reduce distortion in • Ethical concerns may arise
assessment of outcome • Subjects may not comply with
• Assumptions of statistical tests can be treatment assignment
met
(Jabbour, 2004)
• “STUDY DESIGN: We enrolled 126 pregnant women at risk for depression (Edinburgh
Postnatal Depression Scale score 9–19 or a history of depression) in early pregnancy and
randomly assigned them to receive EPA‐rich fish oil (1060 mg EPA plus 274 mg DHA),
DHA‐rich fish oil (900 mg DHA plus 180 mg EPA), or soy oil placebo. Subjects completed
the Beck Depression Inventory (BDI) and Mini‐International Neuropsychiatric Interview at
enrollment, 26–28 weeks, 34–36 weeks, and at 6–8 weeks’ postpartum. Serum fatty acids
were analyzed at entry and at 34–36 weeks’ gestation.
• CONCLUSION: EPA‐rich fish oil and DHA‐rich fish oil supplementation did not prevent
depressive symptoms during pregnancy or postpartum.”
40
Questions to consider in reviewing the design and analysis
of ES
• Have sufficient data been generated from observational studies to justify
conducting an RCT?
• Have the investigators sufficiently described the trial’s inclusion and
exclusion criteria and how theses criteria might affect the generalizability of
the trial’s findings?
• Was the randomization of study subjects to treatment or exposure groups
done appropriately? Have potentially confounding factors been adequately
measured and are they equally distributed between the experimental and
control groups?
• Were the treatment regimens adequately described?
Questions to consider in reviewing the design and
analysis of ES
• Does an adequate placebo exist to the active therapy being tested or is
the new therapeutic approach being compared with standard medical
practice?
• If competing interventions are being used as either adjunctive or
standard therapy, are these interventions similarly distributed between
the respective comparison groups and/or have they been controlled for
analytically?
• Were the data analyzed according to the principles of intention to
treat? Was the sample size large enough to adequately examine
differences in the principal study outcome(s) in intervention and
comparison groups?
Bonita et al, 2006
43
Ability to prove causation
Bonita et al, 2006
Thalib,
2015
45
References:
• Gordis L, (2009). Epidemiology, 5th ed. Philadelphia: Elsevier Saunders.
• Bonita R, Beaglehole R, Kjellstrom T, 2006. Basic Epidemiology. Geneva: WHO.
• Friis R, (2010) . Epidemiology 101. Sudbury, MA: Jones and Bartlett Publishers.
• Coggon, D., Rose, G. A., & Barker, D. J. P. (2003). Epidemiology for the uninitiated. London: BMJ
Books.
• Jacobsen K, 2012. Introduction to health research methods: a practical guide. Jones and
Bartlett Learning.
• Last J,1995. Dictionary of epidemiology. Oxford: OUP.
• Intervention Studies‐ Chapter 7. In Bruce N, Pope D, and Stanistreet D. Quantitative methods
for health research: a practical interactive guide.
• Mills EJ, Chan A‐W, Wu P, Vail A, Guyatt GH, Altman DG. Design, analysis, and presentation of
crossover trials. Trials. 2009;10(1):27. https://doi.org/10. 1186/1745‐6215‐10‐27.
• Mozurkewich E, et al. Am J Obstet Gynecol. Author manuscript; available in PMC 2014 May 05.
• Some slides were extracted with permission from presentations prepared by:
– Jabbour Samer. Experimental studies . A. Cohort Studies (CS). AUB 2004
– Thalib L. Experimental studies. Qatar University 2015
– Abdul Rahim H. Experimental studies Presentation. Qatar University 2015