In a randomised controlled trial evaluating a new antihypertensive, patients who feel better from placebo effects are more likely to continue attending follow-up visits. Compared to the treatment arm, higher dropout occurs in truly uncontrolled patients in the control group. The resulting estimate of treatment efficacy is most likely to be:
- A Underestimated due to attrition bias
- B Unaffected because intention-to-treat analysis was used
- C Overestimated due to attrition bias ✓
- D Underestimated due to performance bias
Correct answer: C. Overestimated due to attrition bias
Explanation
Differential dropout (attrition bias) where sicker control patients drop out leaves a healthier residual control group; the treatment group therefore looks better than it actually is, overestimating efficacy. ITT analysis preserves allocation regardless of dropout only when outcome data are available; if data are missing (as with dropouts), ITT does not fully prevent bias without imputation. Performance bias relates to differences in care received, not dropout.
Reference: Park's Textbook of Preventive and Social Medicine, 27th ed.
High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP
Written and medically reviewed by the StethoPrep medical team.