Cut-points for associations between vitamin D status and multiple musculoskeletal outcomes in middle-aged women.

This was the first study examining optimal vitamin D status for musculoskeletal health in middle-aged women. A 25-hydroxyvitamin D level of at least 29 to 33 nmol/L appears required for optimal musculoskeletal health, but the current cut-off of 50 nmol/L may be warranted.

INTRODUCTION: This study aimed to determine whether cut-points exist for associations between serum 25-hydroxyvitamin D (25OHD) and musculoskeletal health outcomes in middle-aged women, below which greater 25OHD levels are associated with musculoskeletal health benefits and above which no such associations exist.

METHODS: This is a cross-sectional study of 344 women aged 36-57 years. Cut-points for associations of serum 25OHD with lumbar spine (LS) and femoral neck (FN) bone mineral density (BMD), lower limb muscle strength (LMS), timed up and go test (TUG), functional reach test (FRT), lateral reach test (LRT), and step test (ST) were explored using locally weighted regression smoothing and nonlinear least-squares estimation, and associations above and below the identified cut-points were estimated using segmented regression.

RESULTS: The prevalence of low 25OHD was 28 % (<50 nmol/L). Significant cut-points (nmol/L) were identified for FN BMD 31 (95 % confidence interval (CI): 18, 43), LS BMD 31 (17, 45), TUG 30 (24, 36), ST 33 (24, 31), FRT 31 (18, 43), and LMS 29 (8, 49) but not LRT (42 (-8, 93). Below these cut-points, there were beneficial associations between higher 25OHD level and each outcome, while above the cut-points, there were no beneficial associations.

CONCLUSIONS: In middle-aged women, there are thresholds for associations between serum 25OHD concentrations and bone density and most balance measures, suggesting that 25OHD levels of at least 29 to 33 nmol/L are required for optimal musculoskeletal health in this population. The current cut-off of 50 nmol/L may be higher than needed for some outcomes but appears warranted overall.