Study Design

Question 1

Cross Sectional study

Answer 1

In study population, concurrently measure outcome (disease) and risk factor
Compare proportion of diseased group with risk factor to proportion of nondiseased group with risk factor

Advantages: Defines prevalence, Short time to complete
Inexpensive

Disadvantages: Selection bias, Weak evidence for causality

Question 2

Case-Control (Retrospective

Answer 2

Define cases (with outcome of interest) and controls (without outcome)
Compare proportion of cases with exposure (risk factor) to proportion of controls with exposure (risk factor)

Advantages: Good for rare diseases/outcomes
Small sample size
Shorter study times
Less expensive
Can study association of multiple exposures with outcome

Disadvantages: Highest potential for biases
Weak evidence for causality
Unable to determine prevalence, incidence

Question 3

Cohort (Usually Prospective, Occasional Retrospective)

Answer 3

In study population, define exposed group (with risk factor) and nonexposed group (without risk factor)
Over time, compare proportion of exposed group with outcome (disease) to proportion of nonexposed group with outcome (disease)

Advantages:Defines incidence
Stronger evidence for causality
Decreases biases (sampling, measurement, reporting)
Can study association of exposure with multiple outcomes

Disadvantages Expensive
Long study times
May not be feasible for rare diseases/outcomes
Factors related to exposure and outcome may falsely alter effect of exposure on outcome (confounding)

Question 4

Clinical Trial (Experimental)

Answer 4

In study population, randomly assign subjects to receive intervention or receive no intervention
Compare rate of outcomes between intervention and control groups

Advantages: Randomized controlled trial is gold standard
Randomization reduces confounding
Best evidence for causality

Disadvantages: Expensive
Risks of experimental treatments in humans
Longer study time
Not suitable for rare diseases/outcomes

Question 5

Systematic Reviews/Meta-Analysis

Answer 5

Combine data from multiple independent studies to maximize precision and power in testing for statistical significance

Advantages: Higher statistical power
Can control for interstudy variation

Disadvantages: Publication bias. Only as good as the studies included in the analysis.