A study finds a crude relative risk of 3.0 for coffee consumption and pancreatic cancer. When the data are stratified by smoking status, the stratum-specific relative risks are 1.1 among smokers and 1.1 among non-smokers. The most appropriate conclusion is:
- A The crude association is explained by confounding by smoking ✓
- B Smoking modifies the effect of coffee on pancreatic cancer
- C Coffee is truly associated with pancreatic cancer, and stratification introduced random error
- D There is evidence of interaction on the additive scale
Correct answer: A. The crude association is explained by confounding by smoking
Explanation
When stratum-specific estimates are similar to each other but differ markedly from the crude estimate, the crude association is attributable to confounding. Smokers drink more coffee and also have higher pancreatic cancer risk, so smoking distorts the crude estimate. Effect modification requires the stratum-specific estimates to differ meaningfully from one another, which rules out options B and D. Since both strata show a similar weak association, the true effect is close to null.
Reference: Modern Epidemiology, Rothman, as cited in Park 27th ed. ed.
High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP
Written and medically reviewed by the StethoPrep medical team.