In evaluating prognostic studies, which question relates to completeness of follow-up data?

In evaluating prognostic studies, which question relates to completeness of follow-up data?

• A. Were outcome data collected for all enrolle subjects?
• B. Was randomization performed?
• C. Was blinding used?
• D. Was sample size justified?

Answer: (A)

Completeness of follow-up data is about attrition bias in prognostic studies. When outcome information is collected for all enrolled participants, the study can more reliably reflect the true prognosis because there’s no systematic difference between those followed and those lost to follow-up. If some participants lack outcome data, the results may be biased if their outcomes differ from those who were followed, distorting estimates of prognosis. Therefore, asking whether outcome data were collected for all enrolled subjects directly assesses whether follow-up was complete and the findings are less prone to bias. The other options relate to study design rather than follow-up completeness: randomization is about balancing groups in experimental studies, blinding reduces measurement bias, and sample size justification concerns statistical power and precision.

Completeness of follow-up data is about attrition bias in prognostic studies. When outcome information is collected for all enrolled participants, the study can more reliably reflect the true prognosis because there’s no systematic difference between those followed and those lost to follow-up. If some participants lack outcome data, the results may be biased if their outcomes differ from those who were followed, distorting estimates of prognosis. Therefore, asking whether outcome data were collected for all enrolled subjects directly assesses whether follow-up was complete and the findings are less prone to bias.

The other options relate to study design rather than follow-up completeness: randomization is about balancing groups in experimental studies, blinding reduces measurement bias, and sample size justification concerns statistical power and precision.