Measures of Risk: Relative and Attributable Risk

Key points

• • Measures of risk compare the chance of disease in exposed people with that in unexposed people, to show how strongly an exposure is linked to a disease.
• • Relative risk is the ratio of the disease rate in the exposed to the rate in the unexposed; a value of one means no association, and higher values mean a stronger association (though not on its own proof of causation).
• • Relative risk answers how many times more likely, so it shows the strength of an association and is used in cohort studies.
• • The odds ratio is a similar measure used in case-control studies, and it approximates the relative risk when the disease is uncommon.
• • Attributable risk is the extra disease rate in the exposed compared with the unexposed (the exposed rate minus the unexposed rate), taken to be due to the exposure when the association is causal.
• • Attributable risk shows how much disease could be prevented by removing the exposure, so it is more useful for public-health planning than relative risk.