A case-only study is appropriate when which condition is true?

• A. A full control group is available
• B. The exposure of controls can be predicted without a control group
• C. The outcome is continuous
• D. There is no exposure variable

Answer: (B)

A case-only design tests for interactions by looking at how two factors relate within cases alone, and it relies on the assumption that those factors are independent in the population outside of disease. Because you don’t include controls, you don’t need to observe exposure in controls inside the study; you can predict how exposure would be distributed in the non-diseased group from external information or prior knowledge. That’s why this approach is appropriate when the exposure distribution among controls can be anticipated without actually enrolling a control group.

The other scenarios don’t fit this design. Having a full control group is not required and is, in fact, the design’s typical exclusion. The outcome being continuous isn’t what enables a case-only study, since this approach focuses on exposure-factor interaction rather than the nature of the outcome. And there must be an exposure variable to study its interaction with another factor; having no exposure variable at all would preclude the analysis.