In a cohort study, how do you compute and interpret relative risk?

• A. RR = incidence in exposed / incidence in unexposed; RR > 1 indicates increased risk; RR = 1 indicates no association; RR < 1 indicates protective effect.
• B. RR is calculated as (a+d)/(b+c) and always equals 1.
• C. RR uses odds ratios for interpretation.
• D. RR compares prevalence rather than incidence.

Answer: (A)

Relative risk in a cohort study compares how often the outcome happens in those exposed to a factor versus those not exposed. You compute it by looking at the two risks: the risk in the exposed group (how many exposed develop the outcome divided by the total exposed) and the risk in the unexposed group (how many unexposed develop the outcome divided by the total unexposed). Then you take the ratio of these two risks, RR = [risk in exposed] / [risk in unexposed].

Interpreting it is straightforward: if RR is greater than 1, exposure is associated with a higher risk of the outcome; if RR equals 1, there’s no association; if RR is less than 1, the exposure may be protective or linked to a lower risk. For example, if 20 out of 100 exposed people develop the disease and 10 out of 100 unexposed do, the RR is (0.20)/(0.10) = 2, meaning the exposed group has twice the risk.

The other statements don’t fit because they mix up the calculation or the type of measure: using (a+d)/(b+c) isn’t how relative risk is computed, and RR is not forced to be 1. Relative risk relies on incidence (risk over time) rather than odds, which is a different measure, and in cohort studies it reflects incidence, not prevalence.