BACKGROUND: Vitamin D is a key substance that promotes calcium absorption and maintains the health of skeletal system, but its relationship with bone mineral density remains controversial. OBJECTIVE: To determine the relationship between serum level of vitamin D and bone mineral density in the middle-aged and elderly patients in southern China. METHODS: The patients aged over 50 years old admitted in the clinic and in-patient department of a large-scale 3A hospital from July 2014 to December 2016 were enrolled and the clinical data were collected. Then, the association between the serum level of 25-hydroxy vitamin D (25(OH)D) and bone mineral density (at the lumbar spine, femoral neck and total hip) was analyzed. RESULTS AND CONCLUSION: A total of 1 154 patients (77.2% female) were enrolled with a mean age of (64.62±10.48) years, and the mean serum level of 25(OH)D was (62.07±37.40) nmol/L. The serum level of 25(OH)D decreased with age increasing, and the proportion of vitamin D insufficiency/deficiency in patients aged 50-59, 60-69 and ≥ 70 years old was 71.3%, 77.0% and 80.1%, respectively. Correlation analysis showed that the serum level of 25(OH)D was not associated with bone mineral density at the lumbar spine, but had a weakly positive correlation with bone mineral density at the femoral neck and total hip (r=0.09 and 0.08, P < 0.01). For patients with vitamin deficiency (25(OH)D ≤ 50 nmol/L), there was a significantly positive correlation between serum level of vitamin D and bone minerat density at the lumbar spine, femoral neck and total hip (r =0.117, 0.120 and 0.146, P < 0.01). After subgroup analysis, vitamin D level was not associated with bone mineral density in male, but was significantly associated with that in female (r=0.105, 0.135 and 0.171 for the lumbar spine, femoral neck and total hip, respectively, P < 0.01). Therefore, the prevalence of vitamin D deficiency is common in middle-aged and elderly patients in southern China. The serum level of vitamin D is not related to bone mineral density in male, but was positively correlated with that in female. [ABSTRACT FROM AUTHOR]