Background/Purpose: Vitamin D has demonstrated immunomodulatory properties with potential etiologic implications for autoimmune diseases including rheumatoid arthritis (RA). However, a causal association between decreased vitamin D levels and increased RA risk has yet to be definitively demonstrated. Cross-sectional studies are not able to rule-out reverse causation. We examined the relationship between circulating 25-hydroxyvitamin D (25(OH)D) and incident RA in 2 nested case-control studies, in the Nurses’ Health Study (NHS) and Nurses’ Health Study II (NHSII) cohorts.

Methods: We conducted a nested case-control investigation of incident RA in prospective cohorts the NHS and NHSII. We included 170 cases with blood samples collected from at least 3 months to 14 years prior to RA diagnosis, each matched to 3 controls on age, menopausal status, postmenopausal hormone use, date, time and fasting status of blood draw. 25(OH)D levels measured by chemiluminescence immunoassay. We used conditional logistic regression to calculate the odds ratio (OR) and 95% confidence intervals for incident RA for continuous 25(OH)D, dichotomous levels categorized as insufficient(<=20ng/mL) vs. sufficient 25(OH)D and quartile cutoffs of 25(OH)D, using age-adjusted and multivariable adjusted models. We repeated the analyses stratified by categories of time between the blood draw and RA diagnosis (3 months to < 4 years, 4 years to < 8 years and >8 years). Random effects models were used to meta-analyze estimates of association from the two cohorts.

Results: Incident RA was confirmed in 123 NHS and 47 NHSII participants. Mean age at RA diagnosis was 64.3 ± 7.9 years for NHS and 49.4 ± 5.2 years for NHSII, 60% were rheumatoid factor positive. Mean time from blood draw to RA diagnosis was 8.3 ± 4.8 years for NHS and 5.0 ± 2.7 years for NHSII. Meta-analysis of crude and multivariable-adjusted conditional logistic models did not show significant associations between circulating 25(OH)D levels (continuously, dichotomously or in quartiles) and odds of RA. However, there was a 25% increased odds of developing RA with every 1ng/ml decrease in circulating 25(OH)D [OR 1.25 (95% CI 1.01, 1.54)] among those NHSII women who had blood analyzed between 3 months and <4 years prior to RA diagnosis (Table). There was no association between 25(OD)D levels and RA in longer time intervals to diagnosis.

Conclusion: We did not find a significant overall association between circulating 25(OH)D levels and odds of developing RA among women in NHS and NHSII.  However, we did detect an increased odds of RA associated with low 25(OH)D among women in the NHSII cohort only, with levels measured closest to RA diagnosis (>3months to <4 years). These results suggest that vitamin D levels fall in the time windows closest to RA diagnosis and may represent reverse causation.

Table: Odds ratios and 95% confidence intervals for rheumatoid arthritis associated with a 1ng/mL decrease in 25-hydroxyvitamin D