Seasonal variation of serum vitamin D and the effect of vitamin D supplementation in Irish community-dwelling older people.

BACKGROUND Ireland is at 53°N, and its population risk of vitamin D deficiency is high. Previous Irish studies suggested a significant seasonality of serum 25-hydroxyvitamin D [25(OH)D] and a beneficial effect of supplementation in raising 25(OH)D levels. However, in Irish older people, little is known about the magnitude of the supplementation effect and whether supplementation affects 25(OH)D seasonality. DESIGN cross-sectional observational. SETTING outpatient clinic. SUBJECTS five hundred and forty-six community-dwelling subjects (mean age 73.0, SD 7.4; 68.5% females) were assessed between September 2007 and May 2009. METHODS for supplemented and non-supplemented: 'cosinor' analysis (Pulse_XP®) of monthly 25(OH)D. Period global solar radiation (GSR) and solar elevation angle (SEA) data were collected as proxy markers of ultraviolet-B radiation exposure. Multivariate linear regression was conducted to investigate the independent effect of GSR and SEA on 25(OH)D, controlling for confounders. RESULTS supplemented group (N = 183): 89.1% were on cholecalciferol 800 IU/day. Mean 25(OH)D = 64.1 (95% confidence interval: 52.2-75.8) nmol/l, with no significant seasonality; regression: neither GSR nor SEA predicted 25(OH)D. Non-supplemented group (N = 363): mean 25(OH)D = 40.3 (35.5-45.0) nmol/l, with significant seasonality (55.5% variance remaining), peak in August, amplitude = 6.0 (3.1-8.8) nmol/l; regression: both GSR (P = 0.002) and the interaction GSR * SEA (P = 0.018) predicted 25(OH)D. CONCLUSIONS vitamin D supplementation was associated with a mean serum 25(OH)D increase of 23.8 nmol/l. Interestingly, supplementation seemed to blunt seasonality. In the supplemented group, 72.1% had individual 25(OH)D levels below the recommended 75 nmol/l. There is a case for universal supplementation in Irish older people, probably at a higher dose. Further research is needed to establish the optimum dose.