Reliability of self-reported sun exposure in Canadian women and estimation of lifetime exposure to vitamin D from sun and diet.

OBJECTIVE To assess the inter-method reliability of the Ovarian Cancer in Alberta (OVAL) survey developed to estimate adult vitamin D exposure from sun and diet for every tenth year, against the longer Geraldton Skin Cancer Prevention Survey (the assumed 'gold standard'). We also estimated total vitamin D exposure using the OVAL survey. DESIGN A randomized crossover design to assess the inter-method reliability of sun exposure (OVAL v. Geraldton survey), using intra-class correlation and estimated total vitamin D exposure from sun and diet. SETTING Calgary, Alberta, Canada. SUBJECTS Randomly selected women (n 90) aged 40-79 years. RESULTS The average lifetime sun exposure of 13,913 h (average 411 h/year) from the Geraldton survey was not significantly different from the 13,034 h (average 385 h/year) from the OVAL survey for periods with sufficient UV radiation to stimulate vitamin D production. The intra-class correlation coefficient for average lifetime sun exposure was 0.77 (95% CI 0.69, 0.86); the annual average was 0.60 (95% CI 0.47, 0.74). Estimated vitamin D from diet and supplements increased with age. CONCLUSIONS Our OVAL survey reliably estimated adult sun exposure relative to the Geraldton survey, suggesting that assessing sun exposure every tenth year is a reliable and efficient method for estimating sun contributions to lifetime vitamin D exposure.