Association between 25-hydroxy vitamin D levels and risk of frailty and pre-frailty in elderly adults: a systematic review and dose-response meta-analysis of epidemiologic studies with GRADE assessment.

BACKGROUND: A growing number of studies investigated vitamin D and frailty have led to conflicting results. This systematic review and meta-analysis aimed to summarize and update the current evidence surrounding the association between vitamin D and risk of frailty and pre-frailty in elderly adults. METHODS: MEDLINE (PubMed), Scopus, Institute for Scientific Information (ISI) Web of Science databases, and Google Scholar were searched for all published articles up to January 2025. All observational studies that reported odds ratios (ORs) or relative risks (RRs) with 95% confidence intervals (CIs) for the association between 25-hydroxy vitamin D (as independent variable) and frailty (as dependent variable) were included. Maximally adjusted effect sizes were pooled. Certainty of evidence was assessed by the GRADE approach. RESULTS: A total of 41 studies (6 cohorts, 31 cross-sectionals and 4 studies with information of baseline and longitudinal) were included. Highest (vs. lowest) level of vitamin D in prospective cohort studies (with 14,153 participants and 2,137 cases with frailty) was significantly related to a 29% reduction in risk of frailty (pooled RR: 0.71; 95% CI: 0.60, 0.84). Also, a significant inverse association was found in cross-sectional studies with 42,936 cases out of 500,826 participants (OR: 0.50; 95% CI: 0.44, 0.56). Based on linear dose-response analysis, each 10 ng/mL (or 25 nmol/L) increase in serum vitamin D level was inversely related to risk of frailty in cohort (RR: 0.83; 95% CI: 0.77, 0.91) and cross-sectional (OR: 0.73; 95%CI: 0.66, 0.81) studies. Also, non-linear dose-response analysis on eligible studies was significant in both cross-sectional and cohort studies (Pnon-linearity < 0.001). Combining effect sizes of 15 studies (463,103 participants and 259,673 cases) indicated that the highest level of vitamin D compared to the lowest level was significantly related to a 40% reduction in risk of pre-frailty (OR/RR: 0.60; 95% CI: 0.49, 0.73). CONCLUSION: This systematic review and meta-analysis documented inverse associations between vitamin D and frailty, as well as prefrailty, and supported vitamin D as a potential cost-effective screening tool for identifying individuals at risk of frailty.