Which measure is used in case-control studies, and when does it approximate relative risk?

Which measure is used in case-control studies, and when does it approximate relative risk?

• A. Relative risk is used in case-control studies.
• B. Incidence rate is used in case-control studies.
• C. Prevalence ratio is used in case-control studies.
• D. Odds ratio is used in case-control studies and can approximate RR when the outcome is rare.

Answer: (D)

In case-control studies, you start by selecting people based on disease status and then look back to see their exposure. Because the design doesn't follow people over time to observe who develops the disease, you can’t directly calculate incident risk or incidence rates in exposure groups. The measure that naturally comes from this setup is the odds ratio: it compares how common exposure is among cases (those with the disease) to how common exposure is among controls (those without the disease). This odds ratio is the best way to quantify association in this design. The odds ratio also has a useful property: it approximates the relative risk when the outcome is rare in the population. When the disease is uncommon, the odds of disease among the exposed is close to the probability (risk) of disease, and the odds ratio then closely mirrors the relative risk. That’s why, in case-control studies, the odds ratio is preferred, and its interpretation as an approximation of RR holds mainly when the outcome is rare. Why other measures aren’t appropriate here: incidence-related measures require cohort or longitudinal data to track new cases over time, which isn’t available in a case-control setup. Prevalence ratios come from cross-sectional data that capture how widespread disease is at a single time point, not from studying the association between exposure and disease outcome through case-control sampling.

In case-control studies, you start by selecting people based on disease status and then look back to see their exposure. Because the design doesn't follow people over time to observe who develops the disease, you can’t directly calculate incident risk or incidence rates in exposure groups. The measure that naturally comes from this setup is the odds ratio: it compares how common exposure is among cases (those with the disease) to how common exposure is among controls (those without the disease). This odds ratio is the best way to quantify association in this design.

The odds ratio also has a useful property: it approximates the relative risk when the outcome is rare in the population. When the disease is uncommon, the odds of disease among the exposed is close to the probability (risk) of disease, and the odds ratio then closely mirrors the relative risk. That’s why, in case-control studies, the odds ratio is preferred, and its interpretation as an approximation of RR holds mainly when the outcome is rare.

Why other measures aren’t appropriate here: incidence-related measures require cohort or longitudinal data to track new cases over time, which isn’t available in a case-control setup. Prevalence ratios come from cross-sectional data that capture how widespread disease is at a single time point, not from studying the association between exposure and disease outcome through case-control sampling.