Community Medicine

Dr: Nasih Lec: -5-

11-Oct-06

Cohort studies
(Longitudinal studies)

Design
 The starting point for cohort studies is exposure to a risk factor. We select
a group of subjects who have experienced an exposure of interest, and
follow them over a period of time to determine the incidence of one or
more outcomes. We can test the association between the exposure and an
outcome. Study subjects are classified as exposed or unexposed to the risk
factor of interest…

 Example
A cohort study can follow up for several years, the workers of a factory who
are exposed to asbestos and determine the incidence of lung cancer among the
cancer. The study can simultaneously follow up a similar group of other workers
who are not exposed to asbestose and compare the incidence of lung cancer
between the 2 groups.

What do we measure in a cohort study?

 We measure incidence of the outcome among the exposed group and may
compare it with the incidence among the unexposed group.

 Incidence= # of cases/ Total at risk

 Incidence ratio= incidence in exposed/ incidence in unexposed

Why Conduct cohort studies?

 To study natural history of disease: we can follow a population to see how
a disease evolves.

 To study association between exposures and outcomes. Smoking and lung

cancer

 To study rare exposures: radiation is a rare exposure. We can study its

association with leukemia among workers of a power plant.

1
Steps in undertaking a cohort study
1. Defining the study question

 What question we want to answer?

 What is the incidence of death?

2. Selecting the Study Population

 Incidence of death among whom?

 If the exposure of interest is common such as smoking, we can select a

population of people and then classify each individual, at the start of the
study, as exposed or unexposed.

 If the exposure of interest is rare, then we need to select a group of people

who are known to have been exposed, and then choose a suitable
comparison group. Examples asbestos, ionising radiation

3. Collecting exposure data: e.g. Collect data on the smoking habits of

the study group.

 Design and administer data collection methods.

 We must use standard procedures

 Data collection tools:

 Personal interviews
 Questionnaires
 Medical records
 Physical examinations
 Diagnostic tests

4. Follow up for ascertainment of the outcome

 Follow up the study group to record the occurrence of the outcome. We

follow up the individuals to see whether and when do they develop the
outcome under study e.g. death.

 The follow up must be complete for all individuals. Incomplete follow up

mean we miss cases which we don’t know whether they have developed
the outcome or not.

2
5. Analysis of data

 We measure incidence risk (and incidence rate) in a cohort study.

 Incidence risk is the proportion of individuals developing the outcome

after a specified time period

 We can compare the risk among exposed and exposed groups(risk ratio).

Risk of cancer in the exposed group= number of cases of cancer among the
smokers during the study period/ total smokers in study

Risk of cancer in the unexposed group= number of cases of cancer among the non-
smokers during the study period/ total nonsmokers in the study

Risk ratio is the comparison of the above two risks.

Sources of bias
Bias is any error in the design and conduct of a study that makes the results
different from the truth

 Information bias: any error in the measurement of exposure or outcome

that results in inaccuracy of information collected between comparison
groups. has the observer (data collector) made mistakes in obtaining/
recording data on outcome? If he/she has done plenty of mistakes then
our results will be biased.

 Selection bias: if the exposed and unexposed groups are not comparable.
Were the two comparison groups similar? If the exposed and non-exposed
groups are not similar in basic relevant characteristics, then our results
will be biased.

 Loss to follow up: Did we follow up the study groups completely? If we

have failed to follow up all individuals completely to ascertain
development of the outcome then our results will be biased.

3
Strengths and Weaknesses of Cohort studies
Strengths
 Less chance for bias: the exposure is measured at the start of the study,
before the outcome occurs, and so measurement of exposure is not biased
by the presence or absence of the outcome

 Cohort studies can provide data on the time course of the development of
the outcome.

 More than one outcome can be examined in one study

 Rare exposures can be investigated using appropriately selected

populations.

Weaknesses
 Time consuming and expensive

 Not good for rare diseases

 Incomplete follow up makes the results weak

Prepared by:
Rand Aras Najeeb