Which of the following is a mitigation for selection bias in outbreak investigations?

• A. Use representative case and control selection and population-based sampling
• B. Use convenience sampling
• C. Ignore nonresponse
• D. Enroll only severely ill patients

Answer: (A)

Selection bias in outbreak investigations happens when the people included in a study don’t represent the population at risk, so the associations you observe between exposures and illness can be distorted. The best way to mitigate this is to use representative case and control selection with population-based sampling. This means drawing cases from a defined, representative population and choosing controls from the same source population, ideally through a sampling frame that mirrors the actual population at risk. When the cases and controls come from the same well-defined population, comparisons are more valid, and observed associations better reflect reality rather than the quirks of who happened to be included.

Using representative sampling reduces bias because it avoids over- or under-representing subgroups that might have different exposure patterns. In contrast, convenience sampling relies on easy-to-reach individuals and can skew results, ignoring segments of the population. Ignoring nonresponse can introduce nonresponse bias if those who don’t participate differ in exposure or disease status. Enrolling only severely ill patients concentrates on a particular end of the disease spectrum and can distort the relationship between exposure and illness, missing milder cases or different exposure profiles.