In cohort studies, what type of bias results from misclassification of exposure or outcome data?

• A. Information bias.
• B. Selection bias.
• C. Confounding bias.
• D. Detection bias.

Answer: (A)

Misclassification of exposure or outcome data leads to information bias. When exposure status (like smoker vs non-smoker) or the outcome (such as developing a disease) is measured imperfectly, some participants are wrongly labeled. This misclassification distorts the observed association between exposure and outcome in a cohort study.

In practice, if exposure is self-reported or outcome is determined from imperfect records, the resulting bias can pull the effect estimate toward or away from the null depending on whether misclassification is non-differential (similar across groups) or differential (different by exposure status). Non-differential misclassification usually attenuates true associations, while differential misclassification can either exaggerate or mask associations.

This bias is distinct from selection bias, which comes from how participants are chosen or retained; confounding bias, which involves a third variable related to both exposure and outcome; and detection bias, a type of information bias where outcome ascertainment differs by exposure due to observation intensity.