Hip fracture risk in relation to vitamin D : quantifying the different types of evidence

Abstract

Background: Hip fractures cause significant morbidity and mortality and are often associated with bone demineralisation. Vitamin D plays an essential role in bone mineralisation. The evidence of a benefit of vitamin D supplementation for the prevention of hip fractures is apparently conflicting. While observational studies generally show a strong association between low vitamin D status (usually measured as the serum concentration of 25-hydroxyvitamin D, 25(OH)D) and increased risk of hip fractures, evidence from randomised controlled trials (RCTs) is less consistent. Despite this, widespread and costly vitamin D testing and supplementation continues. Objective: To assess quantitatively the evidence on vitamin D and hip fracture and investigate the implications of key limitations in current studies, in particular, variability in serum 25(OH)D measurement and confounding, for the interpretation of observational studies. Method and Results: Five published studies are presented. First, a systematic review and meta-analysis of current evidence showed an overall null effect of RCTs of vitamin D supplementation vs placebo/control (RR=1.13[95%CI 0.98-1.29]), both at high and low dose, on the risk of hip fracture. In apparent contrast, meta-analysis of seventeen case-control studies showed significantly lower serum 25(OH)D levels in hip fracture cases versus controls, with greater differences in 25(OH)D levels in studies with population- compared with hospital-based controls. The second paper reviews issues in vitamin D measurement, including descriptions, comparative performance and limitations of current assays. Third, variability in measurement of 25(OH)D concentration was investigated. Serum 25(OH)D concentration measured by two assays (DiaSorin Liaison and a liquid chromatography tandem mass spectrometry assay, LC-MS/MS) in three laboratories for 813 participants of the Ausimmune Study revealed misclassification of vitamin D status in up to 1-in-3 samples, with significant intra- and inter-assay measurement variability, sufficient to materially affect research and clinical decision making. Fourth, a paper published in the Osteoporosis Australia Building healthy bones throughout life White Paper summarises the challenges and implications of current vitamin D measurement. The final paper examines empirically potential confounding in observational studies of vitamin D and hip fracture risk. Linked questionnaire and hospital data from 158 057 participants in the prospective 45 and Up Study yielded 293 incident hip fractures. Decreased physical functioning (increasing disability), decreased physical activity and decreased time outdoors were each, individually, associated with increased hip fracture risk. When all three factors were considered together, only physical functioning/disability retained a statistically significant association with hip fracture risk (RR=5.61[95%CI 3.33-9.42] for lowest versus highest functioning groups), suggesting that physical activity and time outdoors, markers of vitamin D status, are not independently associated with hip fracture. Conclusion: RCTs demonstrate a lack of benefit of vitamin D supplementation for hip fracture prevention. Confounding between frailty and exposures such as time outdoors and serum 25(OH)D levels may account for the apparent protective association between serum 25(OH)D levels and hip fracture risk in observational studies. Substantial measurement variability of current 25(OH)D assays impairs the interpretation of observational study results and clinical testing. Neither vitamin D supplementation nor widespread 25(OH)D testing for prevention of hip fracture are supported by the current evidence.