The effect of monthly vitamin D supplementation on fractures: a tertiary outcome from the population-based, double-blind, randomised, placebo-controlled D-Health trial

Background Low serum 25-hydroxy vitamin D concentration is associated with increased fracture risk. It is uncertain whether vitamin D supplementation reduces fractures, or whether intermittent doses are harmful. We aimed to investigate if supplementing adults living in Australia with monthly doses of 60 000 international units (IU) vitamin D3 for 5 years or less altered the rate of fractures. Methods We did a population-based, double-blind, randomised, placebo-controlled trial of oral vitamin D3 supplementation (60 000 IU per month) for up to 5 years in adults aged 60-84 years living in Australia. We randomly assigned (1:1) 21 315 participants to either vitamin D or placebo. We ascertained fractures through linkage with administrative datasets. The main outcome was total fractures. Additional outcomes were non-vertebral, major osteoporotic (hip, wrist, proximal humerus, and spine), and hip fractures. We excluded participants (989 [4 center dot 6%]) without linked data, and estimated hazard ratios (HRs) and 95% CIs using flexible parametric survival models. The trial is registered with the Australian New Zealand Clinical Trials Registry, ACTRN12613000743763, and the trial intervention ended in February, 2020. Findings Between Feb 14, 2014, and June 17, 2015, we recruited 21 315 participants. For the current analysis, we included 20 326 participants (vitamin D 10 154 [50 center dot 0%]; placebo 10 172 [50 center dot 0%]). 9295 (45 center dot 7%) of 20 326 participants were women and the mean age was 69 center dot 3 years (SD 5 center dot 5). Over a median follow-up of 5 center dot 1 years (IQR 5 center dot 1-5 center dot 1), 568 (5 center dot 6%) participants in the vitamin D group and 603 (5 center dot 9%) in the placebo group had one or more fractures. There was no effect on fracture risk overall (HR 0 center dot 94 [95% CI 0 center dot 84-1 center dot 06]), and the interaction between randomisation group and time was not significant (p=0 center dot 14). However, the HR for total fractures appeared to decrease with increasing follow-up time. The overall HRs for non-vertebral, major osteoporotic, and hip fractures were 0 center dot 96 (95% CI 0 center dot 85-1 center dot 08), 1 center dot 00 (0 center dot 85-1 center dot 18), and 1 center dot 11 (0 center dot 86-1 center dot 45), respectively. Interpretation These findings do not support concerns that bolus doses of vitamin D administered monthly increase fracture risk. Long-term supplementation might reduce the incidence of total fractures, but additional research is needed to clarify this effect. Copyright (c) 2023 Published by Elsevier Ltd. All rights reserved.