2.

ANALYTIC STUDIES
 focuses on the determinants of a disease
 testing the hypothesis formulated from
descriptive studies
 ultimate goal of judging whether a
particular exposure causes or prevents
disease.
 analytic studies are broadly classified into
 observational
 interventional studies.
 both types use "control group”
Observational studies
 information are obtained by observation
of events.
 no intervention is done
 no deliberate interference with natural
course of disease.
 Individuals can be observed,
prospectively, retrospectively or
currently
 Cohort and case-control (and cross-
sectional) are in this category.
Interventional / Experimental
 The researcher does something about the
disease or exposure and observes the
changes
 Investigator has control over who gets
exposure and who don't.
 The key is that the investigator assign into
either group, whether it is done randomly
or not.
 Always prospective.
Cohort Studies
 An epidemiologic design that identifies
comparison groups according to their
exposure status
„A cohort‟ Vs „Cohort study‟

 A „cohort‟ is simply a group of persons

with common characteristics who are
followed or traced over a period of time.
Prospective Vs Retrospective Cohort Study

 This refers to the timing of the study in

relation to the occurrence of the
outcome of interest

Exposure Out come

Time

Prospective Retrospective
 Cohort Studies – Timing

Exposure Disease

Exposed ?
Unexposed ?

Prospective Retrospective
Collect concurrent Collect historical
exposure and illness exposure and illness
data data

8
Prospective Study/Concurrent study
 Cohort constructed in present time
 exposures documented in present time
and possibly in future
 cohort followed prospectively in future
calendar time.
Historical Prospective/Non-concurrent
Study.
 Cohort constructed in past time
 exposures documented in past
 follow-up can extend into future.
 Mixed: Can include both concurrent
and non-concurrent exposures
Factors in Selection of Exposed Group

 Frequency of the exposure of interest:

ability of obtaining sufficient exposed
individuals in a reasonable period of time
 Ability to obtain complete and accurate
exposure and outcome information on all
study subjects
Cohort…
 Prospective cohort study is the “gold standard” of
observational studies because events can be recorded
as they occur, as opposed to obtaining
information retrospectively
 Cohort is followed over time, and outcomes
ascertained (i.e. disease incidence, death,
remission, etc)
Selection of Non-Exposed (controls)

 Need to be comparable to the

characteristics of the exposed population.
Selection of Non-Exposed…

Exposed groups Non-Exposed groups

Characteristics Characteristics
 Smokers  Not-Smokers
 Largely Educated  Largely Educated
 Urban residents  Urban residents
 Male  Male
Exposure Ascertainment
Pre-existing records:
 From hospital, employers record..
 can make available information for large number
of individuals!
 relatively inexpensive
 allow objective and unbiased determination of
exposure status
But…
 information on exposure level may be insufficient
 may not contain adequate information on
potential confounders.
Exposure Ascertainment..
Survey:
 enables to record exposure information
that are not routinely recorded,
particularly lifestyle factors
 potential for information bias, particularly
recall
Outcome Ascertainment
 Obtain complete, comparable and unbiased
information on the subsequent health
experience of every study subject via:
routine surveillance
periodic health examination
hospital records
death certificate. etc.
 Using a firm outcome criteria and standard
diagnostic procedure which are equally
applied for exposed and non-exposed
individuals
Follow-up: challenges…

 Cost in terms of time and resources

 Obtaining complete information for all
comparison groups
 Loss to follow-up
Effects of loss to follow-up

 Effect is significant if loss is related to

exposure, outcome or to both, or if the
proportion is large in one than the other
group
 If losses to follow-up are large, the study
losses it power and its usefulness limited
Information of loss to follow-up

For losses:
 Obtain information from other sources
 Examine previously collected data to determine
whether there are systematic differences between
the losses and follow-ups
 Indirectly calculate exposure-disease association,
assuming extreme outcomes
Non-participation
 Affect validity when non-participation is related
to the exposure and/or other risk factors for the
outcome under study
 When not related to exposure the effect is
mainly on generalizability of the study results.
 The possible effect can be assessed by comparing
basic characteristics of those participated with
non-participants
Analysis of Cohort Studies

 Compare groups to check similarity at

baseline
 Calculate incidence of the outcome for
exposed and non-exposed
 Control of confounding
Advantage of Cohort Design

 Valuable when the exposure is rare

 Allows direct measurement of risk

 Can elucidate temporal relationship

 Minimize bias in ascertainment of exposure

 Can examine multiple effects of a single

exposure
Disadvantage / weakness of Cohort Design

 Costly and time consuming if disease is

rare and/or slow to develop.
 Loss to follow-up (attrition) may lead to
selection bias.
 Relatively statistically inefficient unless
disease is common.
CASE-CONTROL STUDIES
 A case-control study is one in which
persons with a condition ("cases") and
suitable comparison subjects ("controls")
are identified, and then the two groups
are compared with respect to prior
exposure.
 – subjects are sampled by their outcome
status
 Case-Control Studies – Timing

Exposure Disease
Exposed ? Yes (case)
Unexposed ? No (control)

Investigator
STEPS TO CONDUCT A CASE CONTROL
STUDY
 Select cases
 Select controls
 Obtain exposure information
 Analyze data (odds ratios)
Selection of cases
 Setting clear definition of cases
 Incident Vs Prevalent cases
 Representing spectrum of disease:
 mild, moderate and severe groups
Hospital-based Vs population-
based cases
Hospital-based: Population-based:
 easy and inexpensive  avoids selection bias
to conduct  allows the
 prone to selection description of a
bias. disease in the entire
population
 direct calculation of
rates possible
Prevalent Cases Vs Incident cases
Prevalent Cases Incident cases
 Increase sample size  Helpful to establish
available for rare temporal relationship
diseases between exposure
 Difficult to establish and outcome
temporal sequence  Records are easily
between exposure obtainable and recall
and outcome – bias minimized
reverse causation.
Selection of controls
 Avoiding selection bias.
 Avoiding information („recall‟) bias.
Hospital Controls
Advantages:
 Easily identified and readily available in sufficient
number with reduced cost.
 More likely than healthy individuals to be aware of
antecedent exposures or events
minimize recall bias
• Controls are also likely to have been subject to the
same intangible selection factors
minimize selection bias
 More likely to be cooperative ->reduce bias due to
non response.
Hospital Controls…
Disadvantages
 Ill individuals are different from healthy
 Danger of altering the direction of
association or masking a true association
between exposure and outcome of
interest.
WHY DO CASE CONTROL STUDIES?

ADVANTAGES
 Efficient for rare diseases
 Less costly to conduct a case control study
(requires less subjects)
 Relatively quick, disease and exposure
measurements can be made at the same
point in time
 Useful for studying several potential
exposures
 Efficient in resources and time
DISADVANTAGES
 Cannot yield a population-level measure
of disease incidence or prevalence
 May be difficult to establish that "cause“
preceded "effect".
 Not suitable for rare exposures.
 Prone to selection bias and recall bias.
 May be difficult to ascertain cases.
Cohort Case- control
 Optimal for RARE  Optimal for RARE
exposure outcome
 Can examine multiple  Can examine multiple
effects etiologic factors
 Can elucidate  Can be quick and
temporal relationship inexpensive
 Allows direct  Relatively simple to
measurement of risk carry out
 Minimize bias in  Guarantee the number
ascertainment of of persons with cases
exposure
 An experiment is a set of
observations, conducted under
controlled circumstances, in which the
scientist manipulates the conditions
to ascertain what effect such manipulation
has on the observations.
Key Features of Experimental Design

1. Investigator manipulates the

condition under study
2. Always prospective
 With
outcome
Received
Intervention
Without
outcome
Eligible R B P
individuals
With
outcome

Not-received
Intervention
Without
outcome
 Prevention Trial
With
outcome

Vaccinated
Without
outcome
Healthy R B P
individuals
With
outcome

Not
vaccinated
Without
outcome
 Therapeutic Trial
No
improvement
New
Treatment
Improved
outcome
Sick R B P
individuals
No
improvement
Exciting
Treatment
Improved
outcome
Classification of Intervention
Studies: Based on population
 Clinical trial - usually performed in
clinical setting and the subjects are
patients.
 Field trial- used in testing medicine for
preventive purpose and the subjects are
healthy people.
 Community trial - a field trial in which
the unit of the study is group of people/
community.
Classification of Intervention
Studies: Based on design
 Uncontrolled trial - no control group.
control will be past experience (history).
 Non-randomized controlled - there is
control group but allocation into either
group is not randomized.
 Randomized controlled - there is
control group and allocation into either
group is randomized.
Classification of Intervention
Studies: Based on Trial Objective
 Phase I - trial on small subjects to test a
new drug with small dosage to determine
the toxic effect.
 Phase II - trial on small group to
determine the therapeutic effect.
 Phase III - study on large population to
test effectiveness.
Selection of Study Population
 Reference population: the general
group to whom investigators expect the
results of the particular trial to be
applicable.
 Experimental population: The actual
group in which the trial is conducted,
should not be different if possible to
achieve generalizability.
Considerations for Selection of
study population
 Must be sufficiently large to achieve the
required sample size for the trial
 Must produce sufficient number of
endpoints
 Likelihood of obtaining complete and
accurate follow-up information for the
period of trial.
Enhancing Compliance during
Follow-up
 Selection of study population that are
both interested and reliable.
 Frequent contacts with individuals
 Use incentives, such as providing medical
information
Ascertainment of outcome
 Use uniform ascertainment of outcome
for complete follow-up period for all
study subjects
 Maintain a high level of follow-up
Reduce the proportion of outcomes that are
not ascertained to the minimum and
comparable between the two groups
 Use of placebo and blinding to prevent
bias in identification and reporting of
event of interest.
The Quality of “Gold Standard"

 Randomization
 Blinding
 Use of Placebo
Randomization

 Maximize the chance that both groups are

the same at baseline.
 Ideally, both groups are at the same stage of
the natural history of disease.
 Ideally, only one factor affecting the outcome
of interest would vary between the study
groups.
Double -blinding
 Blinding prevents biases related to
assignment, assessment, or compliance.
 Eliminates the potential for observation
bias:
in reporting of side effects
in assessing outcome
 Eliminates differential compliance or loss
to follow-up.
Placebo
 Placebos are inert treatments intended to have no
effect other than the psychological benefit of offering
treatment.
 Can only be used if there is no accepted treatment
for the condition under study.
 Use of placebo minimizes the bias in the
ascertainment of both subjective disease outcomes
and side effects.
 It facilitates that both groups in the study gain equal
attention.
 Placebo effect: tendency for individuals to report
favourable response to any therapy regardless of the
physiologic efficacy of what they received.
Major Problems Related to
Intervention Studies
 Feasibility/practical problems related to
compliance
 Cost
 Ethics
Feasibility/ Practical Issues
 Subject recruitment
getting adequate individuals to enrol into a difficult
study is not easy.
 Field trials particularly require greater number of
subjects since the risk of contracting a given disease
for the first time is small.
 Loss to follow-up
Select population who are both interested and
reliable.
Arrange frequent contacts with individuals
Use incentives, such as providing medical information
Issues in Follow-Up

 Ascertainment of Compliance

 Outcome Ascertainment
"Run-in or Wash out" period
 Strategy to increase compliance
 All participants receive either the active
treatment or the placebo for a number of
weeks or months before formal
randomization to a treatment group.
 Subjects failing to comply with the study
procedures are eliminated before actual
randomization
Exclusions

 Subjects whose illness is too mild or too

severe to permit the form of treatment
being studied must be excluded.
Cost

 Large number of subjects

 Expenses for visits by study subjects to

designated sites (especially if they do not
come to clinics)

 Preventive trials are limited either extremely

common or extremely serious diseases.
Ethical Issues
 Ethical regulations are more stringent for
intervention studies
 End benefit assurance: chances to offer the
benefit of the trial for study subjects/community.
 No participating subject could be treated better
than the alternative treatment possibilities that
the study protocol provides.
 This implies that any exposure or treatments
given to subjects should be limited to potential
preventives of disease or disease consequences-
serious limitation to etiologic research.