Vitamin D supplementation trials aimed at reducing mortality have much higher power when focusing on people with low serum 25-hydroxyvitamin D concentrations

Abstract

Background: Evidence of an inverse association between serum vitamin D and mortality from epidemiological studies has prompted efforts to reduce mortality by vitamin D supplementation, either in targeted interventions for people with vitamin D insufficiency or deficiency, or in untargeted interventions regardless of baseline vitamin D status. Objective: We aimed to assess the expected impact of the 2 different approaches on effect sizes and power of intervention studies. Methods: Serum concentrations of 25-hydroxyvitamin D [25(OH)D] were measured in 9579 participants aged 50-75 y in the German Epidemiologische Studie zu Chancen der Verhütung, Früherkennung und optimierten Therapie chronischer Erkrankungen in der älteren Bevölkerung (ESTHER) study who were followed for mortality for a median of 12.4 y. We estimated adjusted HRs of mortality from all causes, cardiovascular disease, and cancer for defined increases in serum 25(OH)D by Cox regression, both across the full range of 25(OH)D concentrations and for those with vitamin D insufficiency [30 nmol/L ≤ 25(OH)D < 50 nmol/L] or deficiency [25(OH)D < 30 nmol/L] only, and we calculated the power of intervention studies achieving those effect sizes. Results: An inverse association between serum 25(OH)D and mortality was observed only for participants with vitamin D insufficiency or deficiency and was strongest for the latter. Accordingly, the expected effects were much stronger and the expected power was much higher for interventions that targeted these groups than for untargeted interventions. For example, a targeted intervention study with 10,000 older adults (age 50-75 y) with serum 25(OH)D < 50 nmol/L that increases serum 25(OH)D concentrations by 20 nmol/L in the intervention group (n = 5000) would be expected to yield a 26% reduction of all-cause mortality that could be detected with 89% power within 5 y of follow-up compared with a 10% mortality reduction and 20% power in an untargeted intervention study of the same size. Conclusions: Vitamin D supplementation trials aimed at reducing mortality in older adults have much higher power when focused on those with low serum 25(OH)D concentrations.