Genetically predicted tea intake increases the risk of osteoarthritis: A Mendelian randomization study

Li G, Zhang Z, Liu Y · Frontiers in Genetics

Points the opposite way from the observational tea associations. Inverse-variance weighted OR 1.19 (95% CI 1.08-1.30) and weighted median OR 1.22 (95% CI 1.07-1.40) for osteoarthritis per unit of genetically predicted tea intake, with no evidence of directional pleiotropy on MR-Egger (intercept −0.002, p = 0.73) and no heterogeneity on Cochran's Q.

Population

50,508 participants of European ancestry, 10,083 with osteoarthritis

Intervention

Genetically predicted higher daily tea intake

Comparator

Lower genetically predicted intake

Limitations

Genetic instruments for tea intake capture a drinking habit, not a dose of catechins, and in a European cohort that habit is overwhelmingly black tea with milk and sugar. The MR-Egger estimate did not reach significance (OR 1.22, 95% CI 0.98-1.53), and exposure and outcome both come from UK Biobank, an overlap the authors say can inflate the error. Mendelian randomization on dietary phenotypes is prone to instrument overlap with socioeconomic position. It is evidence against the observational protective signal, not evidence that tea harms cartilage.