Which statement correctly differentiates relative risk from hazard ratio?

• A. Relative risk compares cumulative incidence over a fixed period; hazard ratio compares instantaneous risk over time.
• B. RR and HR are identical in all studies.
• C. RR accounts for censoring, HR does not.
• D. RR is used in time-to-event data; HR is used for cross-sectional data.

Answer: (A)

Relative risk and hazard ratio describe risk in different ways in longitudinal data. Relative risk compares the probability of the event over a defined, fixed period—it's based on cumulative incidence within that window. Hazard ratio, by contrast, comes from time-to-event analysis and compares the instantaneous risk (hazard) at any moment in time, properly handling censoring and varying follow-up through models like the Cox proportional hazards model. Because hazard reflects the risk at each point in time and can account for censoring, while relative risk uses a single fixed window, they respond differently to how follow-up and timing are handled. This is why the statement that RR uses cumulative incidence over a fixed period and HR compares instantaneous risk over time is correct. The other options misstate these distinctions: HR does account for censoring, RR does not; and RR is not used for time-to-event data in the same way HR is, while HR is specifically suited to time-to-event analysis rather than cross-sectional data.