Sequestration of vitamin D in adipose tissue is the main cause of the lower serum 25(OH)D levels in obese subjects. However, it remains unknown whether the adipose tissue stores of vitamin D are readily mobilized for meeting body needs. We aimed to examine whether the association between serum 25(OH)D and all-cause mortality differs by body weight. Using the computerized database of the largest health care provider in Israel, we identified a cohort of subjects ≥20-years old with serum 25(OH)D levels measured between January 2008 and December 2009. Mortality was ascertained through April-2013. Cox regression with restricted cubic-spline function was used to assess the association between serum 25(OH)D and mortality. Median follow-up was 48 months (IQR 43-53 months); 12,337 of 175,781 participants (7.0 %) died. The association between 25(OH)D and mortality was not linear (P < 0.001), and differed between BMI categories (P = 0.019). The lowest adjusted risk for mortality was observed at serum 25(OH)D levels of 73.0, 68.0, and 66.5 nmol/L among subjects with BMI <25, 25-29.9, and ≥30 kg/m(2), respectively. Compared to reference level of 75 nmol/L, increased adjusted HR as evident by 95 % confidence interval lower bound >1.0 was observed at serum 25(OH)D levels less than; 61.0, 48.0, and 40.0 nmol/L among subjects with BMI <25, 25-29.9, and ≥30 kg/m(2), respectively. Serum 25(OH)D levels associated with increased all-cause mortality were inversely related to BMI. However, this observational cohort study may still suffer from residual confounding; therefore, cause and effect relationship cannot be established from this study.