Selecting the appropriate study design: Case–control and cohort study designs

Abstract

This article discusses the observational analytic study designs, i.e., case–control and cohort studies. These two study designs are useful for testing a hypothesis to determine the association between a risk factor and a disease. The analysis for both the studies is based on the conventional 2 × 2 table with the disease status in columns and the risk factor status in rows. The case–control studies start from the disease status and compare the exposure to the risk factor(s) between the diseased (cases) and the not diseased (controls) groups. The odds ratio is determined to compare the proportion of exposed persons in the two groups. The cohort studies start from the exposure to the risk factor status and compare the incidence of the disease in the exposed and not exposed groups. The relative risk compares the incidence between the two groups. The 95% confidence interval is estimated for both studies to determine an actual association between the risk factor and the disease. The strengths and limitations of the two study designs differ based on the direction of the two designs. The case–control study goes backward from the disease status so is more useful for rare diseases and for evaluating multiple risk factors, but it cannot determine causality, and there are chances of recall bias affecting the results of the study. The cohort studies are generally prospective in design from the exposure status and can determine the causal association between the risk factor and the disease. However, the cohort studies are more expensive and require a longer time as well as a larger sample size; the loss to follow-up and misclassification biases can affect the results of the cohort studies.