Mendelian Randomization Meta-Analysis Links Predicted ETS Exposure to Lung Cancer and COPD

An abstract reporting a meta-analytic Mendelian randomization (MR) analysis by Woolf and colleagues examines whether genetically predicted exposure to environmental tobacco smoke (ETS) is associated with selected health outcomes. ETS (second-hand or passive smoking) is framed as a difficult exposure to study using conventional observational designs, with respiratory endpoints highlighted as a central focus.

In that context, the abstract’s results emphasize lung cancer and chronic obstructive pulmonary disease (COPD) alongside several non-respiratory outcomes.

The abstract describes establishing ETS as a cause of disease as challenging in part because of confounding and reverse causation, motivating the use of MR for causal inference. As summarized in the methods, ETS was proxied using parents’ genetically predicted smoking in four MR approaches: two approaches used an index individual’s parent’s genetically predicted smoking instrument, specified as independent of the index individual’s genetically predicted smoking, to assess effects of that parent’s smoking on the index individual.

Two additional approaches used one parent’s genetically predicted smoking, specified as independent of the other parent’s genetically predicted smoking, to assess effects of the first parent’s smoking on the second parent. Across the four approaches, the abstract reports that results were meta-analyzed to provide a combined summary.

For respiratory outcomes, the abstract reports that the findings suggest a causal effect of genetically predicted ETS exposure on lung cancer and COPD, with pFDR < 0.001 for both outcomes. The abstract characterizes these results as suggesting a causal effect for these endpoints within the MR framework applied. For non-respiratory outcomes, the abstract reports no evidence supporting effects of genetically predicted ETS exposure on hypertension, depression, coronary heart disease, or stroke, with pFDR = 1.000 for each.

The abstract concludes that these results support existing public health measures to limit exposure to ETS.

Key Takeaways:

• The abstract describes genetically predicted ETS exposure instrumented via parental genetically predicted smoking, using four MR approaches whose results were meta-analysed.
• The abstract reports that the findings suggest a causal effect of genetically predicted ETS exposure on lung cancer and COPD (pFDR < 0.001 for both).
• The abstract reports no evidence for effects on hypertension, depression, coronary heart disease, or stroke (pFDR = 1.000 for each), and the authors conclude that these results support existing public health measures to limit ETS exposure.