Which approach is cited for handling missing data in cohort studies?

• A. Multiple imputation to account for uncertainty
• B. Complete-case analysis only
• C. Ignore missing data
• D. Delete records with any missing values

Answer: (A)

Missing data in cohort studies can bias results if not handled properly. The approach that is commonly cited as best practice is multiple imputation, which fills in missing values more than once to create several complete datasets. Each dataset is analyzed with the same method, and the results are then combined to produce overall estimates that reflect the uncertainty introduced by the missing data. This preserves sample size and typically reduces bias and increases efficiency, especially when data are missing at random. Other approaches, like using only complete cases, ignoring missing data, or deleting all records with missing values, can bias results and waste information, unless the missingness is completely unrelated to the study variables, which is rarely the case.