Prevalence and novel risk factors for vitamin D insufficiency in elite athletes: systematic review and meta-analysis

Background and purpose Vitamin D insufficiency may be common among elite athletes, but prevalence is unclear, and some potentially important risk factors are uncertain. The present study aimed to (a) estimate the prevalence of vitamin D insufficiency in elite athletes, and (b) examine differences in prevalence between the sexes, and between adults and adolescents, from recent studies which used a contemporary definition of insufficiency. Methods Four databases (Web of Science, SPORTDiscus, PubMed, and Sports Medicine and Education Index) were searched for studies in elite athletes. Literature selection, data extraction, and risk of bias assessment were conducted independently by two researchers. Vitamin D insufficiency was defined as 25(OH)D < 50 nmol/L. Meta-analysis was conducted, using R software x64 4.0.2, to provide estimates of prevalence of insufficiency for adults and adolescents, and to examine between-sex differences in risk of insufficiency. Results From the initial 943 literature search hits, 51 studies were eligible with 5456 participants, 33 studies in adults (12/33 in winter and spring), 15 studies in adolescents (6/15 in winter and spring) and 3 studies with age of study participants not given. Prevalence of vitamin D insufficiency from meta-analysis was 30% (95% CI 22-39%) in adults and prevalence was higher, though not significantly so, at 39% (95% CI 25-55%) in adolescents. Differences in the prevalence of insufficiency between the sexes for the eight studies which provided within-study comparisons was not significant (RR = 1.0; 95% CI 0.79-1.26). Evidence quality was moderate. Conclusions Prevalence of vitamin D insufficiency (<= 50 nmol/L) in elite athletes is high, suggesting a need for greater attention to prevention and treatment. Prevalence estimates in the present study are conservative due to a relative lack of studies in winter. While there was no evidence of higher risk among women than men in the present study, there was less evidence on women.