BACKGROUND Few studies in subjects over 50 yr of age have evaluated the influence of variable doses of vitamin D supplementation on serum 25-hydroxyvitamin D levels. OBJECTIVE We performed a meta-analysis of changes in circulating 25-hydroxyvitamin D level associated with vitamin D supplementation in Caucasian subjects over 50 yr old. METHODS We conducted a systematic search in literature databases and in references of past reviews. METHODS Randomized placebo or open-label trials that evaluated the influence of vitamin D supplementation on clinical outcomes were included in the study. METHODS We reviewed trial characteristics and serum 25-hydroxyvitamin D concentrations at baseline and during the trial. RESULTS Seventy-six trials published from 1984 to March 2011 included 6207 subjects allocated to 101 intervention groups that tested supplement doses ranging from 5 to 250 μg/d (median, 20 μg/d). For similar doses, trials could obtain increases in 25-hydroxyvitamin D three to four times lower than other trials. A meta-regression showed that in the absence of concomitant use of calcium supplements, the average increase in serum 25-hydroxyvitamin D concentrations was 0.78 ng/ml (1.95 nmol/liter) per microgram of vitamin D3 supplement per day. Compared to the vitamin D3, the vitamin D2 was associated with significantly lower increases (P = 0.03). Concomitant use of calcium supplementation and high 25-hydroxyvitamin D concentration at baseline was nonsignificantly associated with lower increases in 25-hydroxyvitamin D concentrations. CONCLUSIONS Dietary recommendations and randomized trials on vitamin D supplementation should evaluate whether increases in circulating 25-hydroxyvitamin D levels match expectations--for instance, the average increases obtained by trials on vitamin D3 without concomitant calcium supplements.