Dietary fatty acid composition and incident osteoarthritis: a parallel prospective cohort study of the UK biobank, HRS, and ELSA

Objectives Prospective evidence linking dietary fatty-acid composition with incident osteoarthritis (OA) is limited. We examined the dietary polyunsaturated-fatty-acid (PUFA)-to-saturated-fatty-acid (SFA) ratio and the energy contributions of PUFA and SFA in three cohorts. Methods Parallel analyses were conducted in the UK Biobank (UKB), Health and Retirement Study (HRS), and English Longitudinal Study of Ageing (ELSA). The prespecified primary exposure was cohort-standardized ln(PUFA/SFA). Cohort-specific estimates were combined using common-effect meta-analysis; modified Hartung–Knapp inference was a sensitivity analysis. PUFA and SFA energy contributions were prespecified secondary exposures. Results Outcome follow-up was available for 89,877 UKB, 1,118 HRS, and 2,134 ELSA participants. Per 1-SD higher ln(PUFA/SFA), the common-effect pooled hazard ratio (HR) was 0.981 (95% CI, 0.963–0.999; p = 0.040), but the modified Hartung–Knapp sensitivity estimate was not statistically significant (HR, 0.981; 95% CI, 0.942–1.022; p = 0.177). Per 1-percentage-point higher PUFA energy contribution, the pooled HR was 0.967 (95% CI, 0.956–0.979); UKB contributed 92.9% of the inverse-variance weight. Conclusion The PUFA/SFA ratio showed a very small, non-robust inverse association with recorded incident OA. The stronger secondary PUFA association was informed predominantly by UKB. These observational results do not establish causality or justify a dietary target for OA prevention.