Abstract Vitamin D may be associated with reduced risks of several types of cancer, including colon, prostate, and breast. We examined the relationship between vitamin D–related questions administered in a telephone interview and serum 25-hydroxyvitamin D [25(OH)D]. Three hundred and eight eligible women were randomly selected from controls in a breast cancer case-control study. Questions pertaining to sun exposure and dietary sources of vitamin D over the previous 4 weeks were asked in both summer and winter. We assessed the association between questionnaire-derived items and 25(OH)D using multiple linear regression. There were 217 participating women, 203 in summer and 213 in winter. Models were adjusted for age, body mass index, and skin color. Number of days when more than 0.5 hour was spent outdoors per week (“7” versus “<7” β = 11.12; P = 0.01), limb coverage (“no” β = 24.90 and “partial” β = 8.15 versus “yes”; P = 0.0001), and milk intake (glasses/wk; “>10” β = 18.94, “>5-10” β = 9.16, and “1-5” β = 7.90 versus “<1”; P = 0.02) best predicted 25(OH)D in the summer. The best predictors in the winter were sunlamp use (“yes” β = 27.97 versus “no”; P = 0.01), milk intake (glasses/wk; “>10” β = 14.54, “>5-10” β = 11.54, and “1-5” β = 2.15 versus “<1”; P = 0.01), and vitamin D–containing supplements (“high” β = 17.30 and “moderate” β = 13.82 versus “none”; P = 0.0006). The R2 was 0.29 for the summer model and was 0.21 for the winter model. Overall, there was evidence to suggest that questions designed to assess vitamin D exposure were in fact related to serum 25(OH)D. (Cancer Epidemiol Biomarkers Prev 2008;17(1):232–8)