CROSS SECTIONAL

ADVANTAGES

DISADVANTAGES

Relatively short duration

Does not establish sequence of

events

A good first step for a cohort

study or clinical trial

Not feasible for rare predictors or

rare outcomes

Yields prevalence of multiple

predictors and outcomes

Does not yield incidence

CASE CONTROL
ADVANTAGES

DISADVANTAGES

Useful for rare outcomes

Bias and confounding from

sampling two populations

Short duration

Differential measurement bias

Small sample size

Limited to one outcome variable

Relatively inexpensive

Sequence of events unclear

Yields odds ratio (resembles

relative risk for uncommon
outcomes)

Does not yield prevalence,

incidence, or excess risk

Convenient for studying many

exposures

COHORT
DESIGN
Prospective

Retrospective

ADVANTAGES

DISADVANTAGES

More control over subject selection

and measurements

Follow up can be lengthy

Avoid bias in measuring predictors

Often expensive (because of

resources necessary to study
many people over time)

Follow up is in the past

Less control over subject

selection and measurements

Relatively inexpensive
All

Establishes sequence of events

Often requires large sample

sizes

Multiple predictors and outcomes

Less feasible for rare outcomes

Number of oucome events grows over

time

Results not available for a long

time

Yields incidence, relative risks, excess

risk

Assesses the relationship

between disease and exposure
to only relatively few factors
(i.e., those recorded at the
outset of the study)

COHORT
DESIGN
All

ADVANTAGES
Follows the same logic as the
clinical question : If persons are
exposed, do they get the
disease?
Convenient for studying many
diseases

DISADVANTAGES

CLINICAL TRIAL
ADVANTAGES

DISADVANTAGES

Demonstrate causality

Expensive

Randomization: eliminate the

influence of confounding
variables

Compliance

Blinding : decreases bias

Selective population
Ethical issue

Characteristics of cross sectional, case control

and cohort
Cross sectional

Case control

Cohort

Begins with a defined

population

Population at risk may

be undefined

Begins with a defined

population at risk

Cases not selected but

ascertained by a single
examination of the
population

Cases selected by
Cases not selected but
investigator from an
ascertained by
available pool of patients surveillance

Non cases include those

free of disease at the
single examination

Controls selected by
investigator to resemble
cases

Controls, the comparison

group (i.e non cases),
not selected-they evolve
naturally

Exposure measured at
the same time as
disease

Exposure measured,
reconstructed, or
recollected after
development of disease

Exposure measured
before the development
of disease

Risk or incidence of
disease cannot be
measured directly :
relative risk of exposure

Risk or incidence of
disease cannot be
measured directly:
relative risk of exposure

Risk or incidence of
disease and relative risk
measured directly