Objective A protective function of vitamin D in metabolic syndrome (MetS) has been described. The objective of the present study was to examine the relationship between serum 25‐hydroxyvitamin D (25(OH)D) concentrations and MetS in non‐diabetic systemic lupus erythematosus (SLE) women. Methods Cross‐sectional analyses of the relationship between concentrations of 25(OH)D, MetS, and its components were made in 160 non‐diabetic SLE women. MetS was defined according to National Cholesterol Education Program Adult Treatment Panel III criteria. Serum 25(OH)D was measured by chemiluminescent immunoassay. Serum 25(OH)D concentrations were categorized into quartiles (<16.6, 16.6‐21.1, 21.2‐26.3, ≥26.4 ng/mL). Results A total of 79 (49.3%) SLE women had MetS. Without adjusting for body mass index (BMI) or smoking, the odds of having MetS decreased according to increasing quartiles of 25(OH)D concentrations (P for trend = .03). The odds ratio (OR) of having MetS was 0.4 (95% confidence interval: 0.2‐0.9, P = .04) for the highest vs the lowest quartile of 25(OH)D concentrations when adjusted by age. The crude OR of having elevated hypertriglyceridemia decreased according to increasing quartiles of 25(OH)D concentrations (P for trend = .036). However, further adjustments for BMI and smoking removed the inverse association between 25(OH)D concentrations and MetS and its individual components. Conclusion In non‐diabetic SLE women with mild activity, 25(OH)D concentrations are not associated with MetS and its components.