Association between serum 25-hydroxyvitamin D and osteoarthritis prevalence in US adults: a cross-sectional study using NHANES 2007–2023

Abstract

Osteoarthritis (OA) is a leading cause of chronic disability globally, and the role of vitamin D in OA pathogenesis remains controversial. This cross-sectional study analyzed data from 35 887 US adults aged ≥20 years across seven NHANES cycles (2007–2023), all using standardized LC-MS/MS for serum 25-hydroxyvitamin D [25(OH)D] measurement. Survey-weighted logistic regression, restricted cubic splines (RCS), subgroup analyses with interaction testing, E-value assessment, Baron–Kenny CRP mediation analysis, and temporal trend evaluation were performed. The survey-weighted OA prevalence was 26.2%. In the fully adjusted model, each 10 nmol/L increase in 25(OH)D was associated with 5.3% higher odds of OA (OR = 1.053; 95% CI: 1.037–1.069; P < 0.001). RCS analysis revealed a threshold effect above 50 nmol/L, plateauing at 100–125 nmol/L. Significant effect modification was observed by age (P < 0.001) and BMI (P = 0.004), with the strongest association among obese adults (OR = 1.058). CRP did not mediate the association (−4.7%). The E-value was 1.29. Results were robust across seven sensitivity analyses and consistent across all survey cycles (ORs: 1.032–1.088). Higher serum 25(OH)D was positively associated with OA prevalence in a non-linear fashion, likely reflecting reverse causation or health-seeking behavior (i.e., patients with OA proactively taking vitamin D supplements) rather than a causal harmful effect, consistent with null results from vitamin D supplementation trials for OA.