Vitamin D Requirements for the Future-Lessons Learned and Charting a Path Forward

Abstract

Estimates of dietary requirements for vitamin D or Dietary Reference Values (DRV) are crucial from a public health perspective in providing a framework for prevention of vitamin D deficiency and optimizing vitamin D status of individuals. While these important public health policy instruments were developed with the evidence-base and data available at the time, there are some issues that need to be clarified or considered in future iterations of DRV for vitamin D. This is important as it will allow for more fine-tuned and truer estimates of the dietary requirements for vitamin D and thus provide for more population protection. The present review will overview some of the confusion that has arisen in relation to the application and/or interpretation of the definitions of the Estimated Average Requirement (EAR) and Recommended Dietary Allowance (RDA). It will also highlight some of the clarifications needed and, in particular, how utilization of a new approach in terms of using individual participant-level data (IPD), over and beyond aggregated data, from randomised controlled trials with vitamin D may have a key role in generating these more fine-tuned and truer estimates, which is of importance as we move towards the next iteration of vitamin D DRVs.

Keywords: Dietary Reference Values; individual participant data; vitamin D requirements.

Figure 1

(a) The placement of (Estimated) Average Requirement [EAR(AR)] and Recommended Dietary Allowance (Population Reference Intakes) [RDA(PRI)] on the distribution of nutrient requirements within a population, as per the Institute of Medicine and European Food Safety Authority, and (b) and normal requirement distribution of hypothetical nutrient including percentile rank (taken from [4]).

Figure 2

Relationship between total vitamin D intake and achieved serum 25(OH)D in winter time using data from seven randomized controlled trials (RCT) used in our recent individual participant data (IPD) level meta-regression [9]. (a) The black dots represent the aggregate RCT group mean data (n = 23 arms from the collection of 7 RCTs) with associated regression line and 95% confidence intervals around that mean response shown in gray shading. (b) The black dots represent the individual data point from each participant from the same 7 RCTs (n = 882 individuals) with associated regression line and 95% prediction intervals around that mean response shown in gray shading.

Figure 3

The placement of a hypothetical individual X (highlighted in red) close to the ‘RDA’ (as derived by use of the lower 95th confidence interval on the regression line), but much lower than the ‘True RDA’ is the estimated derived by use of the lower 95% prediction interval from the regression analysis. UL, Tolerable Upper Intake Level for vitamin D.