Which of the following is NOT typically used as a criterion for causal inference in epidemiology?

• A. Kappa coefficient
• B. Temporal order of cause
• C. Dose-response relationship
• D. Strength of association

Answer: (A)

In epidemiology, deciding whether an exposure truly causes an outcome relies on criteria that link the exposure to the disease in ways that go beyond a simple association. Temporality matters because the cause must come before the effect. A dose–response relationship strengthens the case: as exposure increases, the risk of the outcome tends to rise. The strength of the association also influences judgment; a strong, consistent association is more supportive of causality, though it can still be affected by confounding.

Kappa coefficient, however, measures agreement between raters or observers on classifications or measurements. It tells you how reliably data are collected, not whether an exposure causes an outcome. It addresses measurement quality rather than causal relationships. Therefore, it is not typically used as a criterion for causal inference.