Heaney · The Journal of nutrition 2013 · pooled dose-response analysis of clinical trials · n=3000 across 8 studies

All-source basal vitamin D inputs are greater than previously thought and cutaneous inputs are smaller.

Cited 56 times in the scientific literature.

Level 1 - systematic review of randomized trials

Meta-analysis/pooled individual participant data analysis of 8 supplementation trials

PubMed 23514768 · doi:10.3945/jn.112.168641 · record verified 2026-08-30

What was done

Researchers reanalyzed individual serum 25-hydroxyvitamin D [25(OH)D] concentration data from 8 cholecalciferol supplementation studies (total n = 3000). Linear and curvilinear models were used to extrapolate dose-response curves to zero concentration values to estimate basal input, and seasonal oscillation in serum 25(OH)D was measured across 4 studies to quantify peak cutaneous synthesis.

What was found

Calculated total basal input (food plus solar) ranged from 778 IU/d in end-stage renal disease patients to 2667 IU/d in healthy Caucasian adults. Obese individuals had lower baseline 25(OH)D concentrations and smaller responses to supplementation; African Americans also had lower baseline concentrations and lower calculated basal inputs. Seasonal oscillation ranged from 5.20 to 11.4 nmol/L, translating to mean peak summer cutaneous synthesis of 209 to 651 IU/d. Cutaneous synthesis accounted for only 10% to 25% of unsupplemented input at summer peak.

Why it matters

This analysis suggests that basal, unsupplemented vitamin D input is an order of magnitude higher than expected from recognized food sources, while cutaneous synthesis contributes a smaller fraction than previously assumed.

Limits

Values are derived from mathematical extrapolations of dose-response curves rather than direct measurements of dietary intake or sun exposure. The specific sources accounting for the remaining basal input (such as preformed 25(OH)D in foods) were hypothesized rather than directly quantified.

Cited by