OBJECTIVES: To investigate the association between low vitamin D levels (<50 nmol/L) and 10-year mortality in women aged 75 and older.
DESIGN: Prospective with 15 years of follow-up.
SETTING: Malmö, Sweden.
PARTICIPANTS: Population-based cohort of 75-year-old women (N = 1,044).
MEASUREMENTS: Serum 25-hydroxyvitamin D (25(OH)D) levels at age 75 (n = 1,011), 80 (n = 642), and 85 (n = 348) were categorized as low (<50 nmol/L), intermediate (50-75 nmol/L) and high (>75 nmol/L) at all ages. Hazard ratios (HRs) for all-cause mortality between ages 75 and 90 were calculated according to 25(OH)D category.
RESULTS: Between ages 80 and 90, all-cause mortality (HR = 1.8, 95% confidence interval (CI) = 1.3-2.4, P < .001; adjusted for comorbidities (aHR) = 1.9, 95% CI = 1.4-2.6, P < .001) was significantly higher in women with low 25(OH)D levels than in those with high levels. Osteoporosis had the greatest effect on mortality, but even after excluding women with osteoporotic fracture during the risk of dying associated with low 25(OH)D remained greater (HR = 1.8, 95% CI = 1.2-2.7, P = .002; aHR = 1.7, 95% CI = 1.2-2.5, P = .006).
CONCLUSION: In this observational study of women aged 75 and older, 25(OH)D levels of less than 50 nmol/L were associated with greater all-cause mortality for up to 10 years. This difference was at least partially independent of comorbidities and fracture, indicating that low 25(OH)D not only is an indicator of impaired health, but also plays a role in disease outcome.