If dietary calcium intake is low and serum calcium concentrations decrease, the compensatory metabolic response is the accelerated conversion of 25-hydroxyvitamin D [25(OH)D] to 1,25-dihydroxyvitamin D (via parathyroid hormone), so as to normalize serum calcium levels ^{( 1 )}. These interactions between vitamin D and calcium may have implications for the regulation of 25(OH)D production and its catabolism, and consequently for the dietary vitamin D requirement ⁽ Reference Clements, Johnson and Fraser ^{2 )}. While highlighted as a significant knowledge gap by the North American Institute of Medicine in their recent DRI report for vitamin D and calcium ^{( 1 )}, to date this hypothesis has only received limited attention in human studies. The aim of this study was to test this hypothesis using relevant data from our previously published vitamin D intervention studies in 20–40 y olds ⁽ Reference Cashman, Hill and Lucey ^{3 )} and 64+ y olds ⁽ Reference Cashman, Wallace and Horigan ^{4 )} as well as an updated version of our recent meta-regression analysis ⁽ Reference Cashman, Fitzgerald and Kiely ^{5 )}, all of which up to now have not included dietary calcium intake as a possible additional determinant of vitamin D requirements.

Within the datasets from our two vitamin D intervention studies, baseline serum 25(OH)D and response of serum 25(OH)D to vitamin D₃ intervention (0, 5, 10 or 15 μ/d) over winter months were stratified by<or ⩾three different cut-offs for dietary calcium intake (550, 700 and 800 mg/d), representing EU estimated average requirement (EAR), UK reference nutrient intake and US EAR for calcium, respectively). Students' paired and unpaired t-tests were used to test differences at baseline and response of serum 25(OH)D to intervention in groups stratified according to each cut-off for calcium intake, respectively. Regression models were also run to account for possible confounding effects of sex, age, BMI, habitual vitamin D intake and study centre. In the meta-regression analysis, two new studies published since our original meta-regression in 2011⁽ Reference Cashman, Fitzgerald and Kiely ^{5 )} were included and habitual calcium intake was reported in each of the 13 studies.

In the meta-regression approach, the relationship between serum 25(OH)D and total vitamin D intake were not significantly different (P>0.05) when studies were stratified on the basis of habitual group mean calcium intakes<or ⩾∼1000 mg/d. Data from our own two intervention studies suggest that while baseline serum 25(OH)D concentrations were significantly lower (P<0.05) in the<700 mg/d calcium intakes groups compared to the>700 mg/d in the 64+ y olds, these differences disappeared when data was adjusted for the aforementioned confounding factors. There were no other significant differences (P>0.05) in baseline serum 25(OH)D in 20–40 y olds or 64+ y olds when stratified by habitual calcium intake. Upon stratification on the basis of <or ⩾three selected dietary calcium intake cut-offs, there was no significant difference in the response of serum 25(OH)D to vitamin D intervention over winter in the 20–40 y olds. Likewise, there was no significant difference in response of serum 25(OH)D to vitamin D intervention in 64 y olds stratified by</>550 mg/d or</>800 mg/d but there were lower responses to 5 μ/d (12.2 nmol/L; P=0.007), 10 μ/d (14.6 nmol/L; P=0.008), and 15 μ/d (9.4 nmol/L; P=0.097) in those 64+ y olds with calcium intakes<compared to ⩾700 mg/d. Calcium intake</>700 mg/d remained a significant predictor (P=0.041) of the change in serum 25(OH)D over winter in the regression models that accounted for treatment but also the aforementioned confounding factors.

In conclusion, post-hoc analysis seems to suggest that a habitual calcium intake less than 700 mg/d may lead to a more blunted response of winter serum 25(OH)D to increased vitamin D intake in older adults. This may have implications for the dietary vitamin D requirement. *The two vitamin D intervention studies were collaborative studies with the University of Ulster, Coleraine. **Current analysis supported by the Department of Health, England.