The vitamin D RDA for African American adults: higher than that for white persons?

Abstract

The Recommended Dietary Allowance (RDA) values for vitamin D (600 IU/d up to the age of 70 y and 800 IU/d thereafter) established by the North American Institute of Medicine (IOM) in 2011 are the intakes that are likely to meet or exceed the needs of ;97.5% of the population (1). These were based on indicators of bone health and a serum 25-hydroxyvitamin D [25(OH)D] threshold of 50 nmol/L during winter (1). These RDA values are of enormous importance from a public health perspective in terms of preventing vitamin D deficiency and promoting adequate vitamin D status in the population. They were established for the entire population and, as such, cover dark-skinned population groups; however, they are based on an assumption that the requirements between white and other ethnic groups do not differ, largely due to the absence of data. This knowledge gap, encountered by the most recent Dietary Reference Intakes (DRI) committee, persists despite the fact that research recommendations in the previous DRI review of vitamin D (2), as well as a roundtable discussion on DRI research needs (3), called for studies to evaluate the intake requirements for vitamin D as related to optimal circulating 25(OH)D concentrations across life-stage and race-ethnicity groups of US and Canadian populations (1). In the United States, non-Hispanic blacks and Mexican Americans (representing ;13% and 17% of the adult population, respectively, based on 2012 US Census data) have been shown to have a higher risk of vitamin D inadequacy [serum 25(OH)D ,50 nmol/L] than non-Hispanic whites (73%, 42%, and 21%, respectively), after adjusting for age and season (4). Thus, it is of note that new data on the dietary vitamin D requirements of African American adults have been published in the past 9 mo. In the present issue of the Journal, Ng et al (5) report their findings from a 4-arm (placebo and 1000, 2000, and 4000 IU/d) randomized placebo-controlled trial (RCT) with vitamin D3 supplements daily for 3 mo in African American adult men and women (aged 30–80 y; n 1⁄4 292; conducted in Boston, MA; ;42 N). By using the data on plasma 25(OH)D response after 3 mo of vitamin D3 supplementation, the authors estimated by using a mixed-model regression approach that the vitamin D RDA to maintain circulating 25(OH)D .50 nmol/L in 97.5% of African American adults is 1640 IU/d. This is considerably higher than the age-specific 600 and 800 IU/d established by the IOM, albeit using data from multiple RCTs with predominantly white subjects (1). If the new data from Ng et al are correct then the current RDA for vitamin D (1) may not provide the intended population-protective impact in the non-Hispanic black segment of the US population, a group at high risk of vitamin D deficiency and inadequacy (4). It is tempting to suggest that the difference between the RDA estimates from the IOM (1) and Ng et al (5) may arise from the fact that the meta-regression approach as per the IOM possibly more reflects an average serum 25(OH)D response and thus requirement, whereas use of a 95% lower prediction interval (PI) approach with individual subject data as per Ng et al estimates the requirement of 97.5% of the population (see reference 6 for review). In this regard, it is of note that the RDA estimate from 2 winter-based vitamin D RCTs in white adults (age 20–40 y) and in older adults (age 64 y) at 51–55N (7, 8), based on a 95% lower PI on a combined data set of individual subject data (n 1⁄4 381), was 1216 IU/d and lower than the 1640-IU/d estimate for African American adults. Unfortunately, the RCT by Ng et al did not include a white group, so it is not possible to be sure whether requirements between their African American adults and a corresponding group of white adults would actually differ. Gallagher et al (9) also recently reported RCT data on the doseresponse effects of vitamin D3 supplementation on serum 25(OH)D over 12 mo in older African American women (n 1⁄4 110), which was in parallel to their similarly designed RCT in older white women (10). The RCT was conducted in Indiana (;40N) and in Omaha, Nebraska (;41N). By using combined data from both of their RCTs, Gallagher et al (9) evaluated the potential interaction of race in their mixed-model regression analysis and found no significant interaction. Thus, the authors concluded that, on the basis of their serum 25(OH)D doseresponse data from the RCTs, the vitamin D RDA estimate was similar for white and African American postmenopausal women of a similar age (average age of 67 y). Moreover, despite also applying a 95% lower PI approach to their data so as to generate