The D-Health Trial: a randomised controlled trial of the effect of vitamin D on mortality
RCT (n=21,315, Australia) found administering one oral gel capsule of 60,000 IU vitamin D3 monthly for 5 years to unscreened older, largely vitamin D replete population, did not reduce all-cause mortality (deaths in 5.3% vs 5.1% placebo arm; HR 1.04; 95% CI 0.93-1.18,p=0.47).
Source:
The Lancet Diabetes & Endocrinology
SPS commentary:
The researchers note that these findings give pause to earlier reports that vitamin D supplements might reduce cancer mortality and they would invoke the precautionary principle and suggest avoiding the use of this dosing regimen, despite the attraction of higher compliance of once a month dosing than daily dosing, until the question is resolved.
According to a commentary, there is no international consensus about the 25(OH)D threshold for vitamin D sufficiency for non-skeletal outcomes, but most advisory bodies consider a 25(OH)D concentration of at least 50 nmol/L sufficient and many intervention studies aimed to increase 25(OH)D above this concentration. It notes that with monthly doses of 60 000 IU per month (corresponding to 2000 IU per day), the intervention exceeded Australia's recommendations and those of most other countries. It adds that the findings of the D-health study are largely consistent with previous intervention trials and two other megatrials with vitamin D supplementation as mono or co-intervention reporting mortality and cancer outcomes. It suggest that taken together, these findings do not suggest that vitamin D supplementation influences all-cause mortality in mostly vitamin D replete older populations, but the most important caveat in these intervention studies is the scarcity of data from participants with a 25(OH)D concentration below 50 nmol/L, who are most likely to benefit. In addition, for population-based policy, potential adverse effects also need to be considered, and also longer follow-up might be required to adequately investigate the effects on conditions that develop over decades, rather than years. It is anticipated that 10-year mortality rates from D-Health will be reported in future.