Relationship between vitamin D and hyperglycemia in older people from a nationally representative population survey.

OBJECTIVES: To investigate the association between vitamin D levels and hyperglycemia in a national community sample of older people. DESIGN: Cross-sectional, nationally representative sample. SETTING: Community. PARTICIPANTS: Two thousand thirty-eight noninstitutitionalized adults aged 65 and older taking part in the Health Survey for England 2005. MEASUREMENTS: Serum 25-hydroxy vitamin D (25(OH) D) levels, glycosylated hemoglobin (HbA1c), and covariates: age, sex, social class, season of examination, use of vitamin supplements, and physical health status. RESULTS: Hyperglycemia was independently associated with low vitamin D levels (odds ratio (OR) = 2.30, 95% confidence interval (CI) = 1.20-4.42 for 25(OH)D <25.0 nmol/L and OR = 2.09, 95% CI = 1.22-3.58 for 25(OH)D 25.0-49.9 nmol/L) but not for 25(OH)D between 50.0 and 74.9 nmol/L (OR = 1.49, 95% CI = 0.85-2.62). CONCLUSION: The higher-than-expected co-occurrence of low vitamin D levels and hyperglycemia (HbA1c ≥ 6.5%) are important public health concerns for older populations living in northern latitudes because both are common, and both have substantial adverse health consequences. Ensuring adequate vitamin D levels may help reverse the increasing trend in the development of diabetes mellitus and related complications in older people.