Background/Purpose: Vitamin D is thought to be an immunomodulator able to suppress autoimmunity in rheumatoid arthritis (RA).  Some reports describe an association between low vitamin D and RA, but they had significant limitations. The objective of this study was to evaluate the association of vitamin D level with incident RA in a well characterized patient population using electronic health records (EHR).

Methods: Case-control study with data extracted from EHR in a tertiary health system, from 2001 to 2012. RA was defined as ICD-9 code 714.0 twice by a rheumatologist. Only patients with a primary care physician in the system were included. Incident RA patients (cases) were identified at one point in time and followed retrospectively with regard to vitamin D levels prior to the development of RA.  Cases were matched for age and gender with 5 controls from the general non-RA population.  An index date was created such that for cases it was the date of the RA diagnosis and for the controls it was the date of the RA diagnoses of the matched case.  The most recent value of vitamin D to the index date was used in the analysis. Vitamin D levels were treated both as continuous and categorical; we used <30 IU as the primary cutoff, but also used <20 IU to investigate the RA association with severe vitamin D deficiency. This association was presented as the odds ratio (OR) from a conditional logistic regression model adjusting for obesity and smoking status. The study had a power of 82% to detect an OR of 1.5 between cases and controls.

Results: 270 patients with incident RA and 1341 matched controls were included in the analysis with median age 62.4 years, BMI 28.8 kg/m²and vitamin D level 31 IU. There were no significant differences in median BMI and vitamin D levels between cases and controls. Smoking was significantly more prevalent in the cases than the controls. Of the RA patients, 141 (52.2%) were rheumatoid factor (RF) positive, 97 (35.9%) were RF negative, and 32 (11.9%) had unavailable RF status. The OR for the association between RA and vitamin D overall and stratified by gender is shown in Table 1. Subgroup analysis according to RF positivity is shown in Table 2.

Conclusion: Low vitamin D level was not associated with increased odds for incident RA in this patient population. Our study was powered at 80% to detect a minimum of 50% increase in the odds of low vitamin D levels.

Table 1. Association between low vitamin D level and Rheumatoid arthritis according to gender*

Vitamin D level	All		Female		Male
	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value
Vitamin D (continuous)	0.996  (0.987, 1.005)	0.3652	0.996 (0.985, 1.006)	0.4014	0.997 (0.976, 1.018)	0.7693
Vitamin D <30 IU vs.  ≥ 30 IU	0.980 (0.745, 1.288)	0.8825	0.924 (0.682, 1.254)	0.6134	1.233 (0.649, 2.343)	0.5231
Vitamin D <20 IU vs. ≥ 20 IU	1.121 (0.801, 1.570)	0.5048	1.192 (0.825, 1.724)	0.3493	0.860 (0.362, 2. 045)	0.7336

*adjusted for smoking, BMI (kg/m²)

Table 2. Association between low vitamin D and Rheumatoid Arthritis according to Rheumatoid Factor positivity*

Vitamin D level	RF-positive vs. controls		RF-negative vs. controls
	OR (95% CI)	p-value	OR (95% CI)	p-value
Vitamin D (continuous)	0.991  (0.978, 1.005)	0.1945	0.997 (0.981, 1.013)	0.6974
Vitamin D <30 IU vs. ≥30 IU	1.062 (0.723, 1.561)	0.7594	0.810 (0.503, 1.304)	0.3854
Vitamin D <20 IU vs. ≥20 IU	1.244 (0.790, 1.961)	0.3464	0.789  (0.415, 1.502)	0.4710

*adjusted for smoking, BMI (kg/m²); RF=rheumatoid factor

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/low-vitamin-d-level-is-not-associated-with-increased-incidence-of-rheumatoid-arthritis/