In a trial, treatment A shows lower overall mortality than treatment B. When the data are stratified by severity of illness, treatment B actually shows lower mortality in both mild and severe subgroups. The reversal caused by unequal distribution of severity between groups is termed:

• A Effect modification
• B Overmatching
• C Regression to the mean
• D Simpson's paradox ✓

Explanation

Simpson's paradox occurs when an association present in combined data reverses direction within every stratum, typically because a confounder such as disease severity is distributed unequally between groups; sicker patients may preferentially receive the apparently superior treatment. Effect modification means the effect genuinely differs by stratum, which is not the case here since the direction reverses uniformly. Regression to the mean and overmatching are unrelated phenomena.

Reference: Modern Epidemiology (Rothman), 4th ed.

High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP

Written and medically reviewed by the StethoPrep medical team.