Attrition bias in cohort studies refers to bias from differential loss to follow-up. Which mitigation strategies are commonly employed?

• A. Attrition bias occurs when differential loss to follow-up biases results; mitigate with robust follow-up, tracking, multiple imputation, and sensitivity analyses.
• B. Attrition bias is due to misclassification of exposure; mitigate with blinding.
• C. Attrition bias occurs when there is selection of controls; mitigate with randomization.
• D. Attrition bias is the same as confounding.

Answer: (A)

Attrition bias happens when people drop out of a study in a way that is related to the exposure or outcome, causing the results to be biased. To lessen this risk, researchers focus on keeping participants in the study and collecting follow-up data as complete as possible, which is robust follow-up and tracking. When data are missing, multiple imputation uses information from the observed data to estimate plausible values for the missing outcomes, reducing bias from incomplete data. Sensitivity analyses then test how the results would look under different assumptions about the missing data, helping determine whether the conclusions are robust to potential attrition bias.

Other options describe biases not related to loss to follow-up: misclassification of exposure is addressed by better measurement and sometimes blinding; selection of controls is addressed by randomization; and confounding is addressed by design or analytic adjustment.