Which formula defines attributable risk?

• A. Incidence among unexposed minus incidence among exposed.
• B. Incidence among exposed minus incidence among unexposed.
• C. Incidence in population divided by prevalence.
• D. Relative risk minus 1.

Answer: (B)

Attributable risk measures the extra risk of disease in the exposed group that can be attributed to the exposure. It is the absolute difference in incidence between those who are exposed and those who are not exposed.

The correct expression is incidence among exposed minus incidence among unexposed. This difference tells you how much of the disease rate in the exposed group is due to the exposure itself and how much could be prevented if the exposure were eliminated. For example, if the exposed group has 20 cases per 1,000 per year and the unexposed group has 8 per 1,000 per year, the attributable risk is 12 per 1,000 per year.

The other ideas don’t reflect this absolute difference. Subtracting the unexposed from the exposed would flip sign if the exposed rate were lower, which isn’t the intended measure. Dividing population incidence by prevalence isn’t a valid way to express attributable risk. Taking relative risk minus one gives the excess relative risk, a relative (not absolute) measure; to get an AR% you’d use a different formula involving RR in a proportion, not RR minus 1 alone.