Table 1.

Observational Population Research Study Designs

	Cohort Studies
	Prospective	Retrospective	Case-Control Studies
Study Group	Exposed persons $\left(\text{a}+\text{b}\right)$	Exposed persons $\left(\text{a}+\text{b}\right)$	Persons with the disease (cases): $\left(\text{a}+\text{c}\right)$
Comparison Group	Nonexposed persons $\left(\text{c}+\text{d}\right)$	Nonexposed persons $\left(\text{c}+\text{d}\right)$	Persons without disease (controls): $\left(\text{b}+\text{d}\right)$
Outcome Measurements	Incidence in the exposed $\frac{a}{\left(a+b\right)}$ And incidence in the nonexposed $\frac{c}{\left(c+d\right)}$	Incidence in the exposed $\frac{a}{\left(a+b\right)}$ And incidence in the nonexposed $\frac{c}{\left(c+d\right)}$	Proportion of cases exposed $\frac{a}{\left(a+c\right)}$ And proportion of controls exposed $\frac{b}{\left(b+d\right)}$
Measures of Risk	Absolute Risk, Relative Risk, Odds Ratio, Attributable Risk	Absolute Risk, Relative Risk, Odds Ratio, Attributable Risk	Odds ratio and attributable Risk.
Temporal Relationship between exposure and disease	Easy to establish	Sometimes hard to establish	Sometimes hard to establish
Multiple Associations	Possible to study association of an exposure with several diseases	Possible to study association of an exposure with several diseases	Possible to study association of an exposure with several diseases
Time required for the study	Generally, long because of need to follow-up the subjects	May be short	Relatively short
Cost of study	Expensive	Generally, less expensive than prospective studies	Relatively inexpensive
Population size needed	Relatively large	Relatively large	Relatively small
Potential Bias	Assessment of outcome	Susceptible to bias both in assessment of exposure and assessment of outcome	Assessment of exposure
Best when	Exposure is rare, disease is frequent among exposed	Exposure is rare, disease is frequent among exposed	Disease is rare and exposure is frequent among persons with disease
Problems	Selection of nonexposed comparison group often difficult. Changes over time in criteria and methods.	Selection of nonexposed comparison group often difficult. Changes over time in criteria and methods.	Selection of appropriate controls often difficult. Incomplete information on exposure.