Robert P. Heaney, Laura A.G. Armas, Christine French

The magnitude of vitamin D inputs in individuals not taking supplements is unknown; however, there is a great deal of information on quantitative response to varying supplement doses. We reanalyzed individual 25-hydroxyvitamin D [25(OH)D] concentration data from 8 studies involving cholecalciferol supplementation (total sample size = 3000). We extrapolated individual study dose-response curves to zero concentration values for serum 25(OH)D by using both linear and curvilinear approaches and measured seasonal oscillation in the serum 25(OH)D concentration. The total basal input (food plus solar) was calculated to range from a low of 778 iu/d in patients with end-stage renal disease to a high of 2667 iu/d in healthy Caucasian adults. Consistent with expectations, obese individuals had lower baseline, unsupplemented 25(OH)D concentrations and a smaller response to supplements. Similarly, African Americans had both lower baseline concentrations and lower calculated basal, all-source inputs. Seasonal oscillation in 4 studies ranged from 5.20 to 11.4 nmol/L, reflecting a mean cutaneous synthesis of cholecalciferol ranging from 209 to 651 iu/d at the summer peak. We conclude that: 1) all-source, basal vitamin D inputs are approximately an order of magnitude higher than can be explained by traditional food sources; 2) cutaneous, solar input in these cohorts accounts for only 10-25% of unsupplemented input at the summer peak; and 3) the remainder must come from undocumented food sources, possibly in part as preformed 25(OH)D.