A study across 30 districts finds that districts with higher average per capita tea consumption have higher rates of oesophageal cancer. An investigator concludes that drinking tea causes oesophageal cancer. The logical error committed is:

• A Berkson's bias
• B Regression to the mean
• C Hawthorne effect
• D Ecological fallacy ✓

Explanation

An association observed between group-level (aggregate) variables cannot be validly applied to individuals because an unmeasured confounder concentrated in those groups may explain the pattern; drawing individual-level conclusions from group-level data is the ecological fallacy. Berkson's bias is selection bias from hospital-based sampling, and regression to the mean concerns extreme values on repeat measurement. Individual-level analytic studies are needed to confirm such associations.

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.