Non-linear Association Between Social Determinants of Health and Osteoarthritis: A Cross-sectional Analysis of NHANES 2003-2018

Background Despite the well-established role of biomechanical and inflammatory factors in osteoarthritis pathogenesis, the impact of social determinants of health (SDoH) remains understudied. We investigated the association between SDoH burden and osteoarthritis prevalence in a nationally representative U.S. adult population. Methods This cross-sectional study analyzed data from 27,669 participants in the National Health and Nutrition Examination Survey (2003–2018). A composite SDoH score (0–8) was constructed from eight domains: employment status, poverty-to-income ratio, food security, educational attainment, healthcare access, health insurance, housing status, and marital status. Survey-weighted logistic regression models examined associations between SDoH tertiles and self-reported physician-diagnosed osteoarthritis, adjusting for demographic and clinical covariates. Restricted cubic spline analysis explored non-linearity, and formal mediation analysis assessed serum albumin as a potential mediator. Results After comprehensive adjustment, individuals with the highest SDoH burden (Tertile 3) had 26% higher odds of osteoarthritis compared to those with the lowest burden (Tertile 1) (OR 1.26, 95% CI: 1.07–1.47, p = 0.005). Restricted cubic spline analysis revealed a significant non-linear U-shaped relationship (P-non-linear < 0.001). Employment status, food security, healthcare access, health insurance, and housing conditions were independently associated with osteoarthritis, while education, income ratio, and marital status were not. Serum albumin mediated only 3.6% of the total effect. Significant effect modification was observed by ethnicity and smoking status. Conclusion Social determinants of health demonstrate a significant non-linear relationship with osteoarthritis risk, independent of traditional risk factors. This association is minimally mediated by nutritional status and varies across population subgroups. These findings support integrating social risk assessment into osteoarthritis care and developing targeted interventions addressing modifiable social factors to reduce disease burden.