The background information provided here is intended to illustrate the major knowledge gaps that have posed challenges for setting DRIs for the micronutrients covered in the DRI Micronutrients Report. The Estimated Average Requirements (EARs) for vitamin A, iron, and zinc were all based on factorial considerations including percent absorption, bioavailability, excretion, turnover, and utilization of the nutrient. The panel had considered using indicators for setting the EARs for these nutrients. Examples of indicators the panel considered when setting the EAR for vitamin A include
conjunctival impression cytology,
relative dose–response and modified relative dose–response,
serum/plasma retinol concentration, and
The first three indicators were not selected either because most of the work was done in animals or because the observations were made in developing countries where the total nutritional profiles were uncertain. Insufficient data were available on isotope dilution studies to obtain a good prediction of body stores of vitamin A. Neither serum nor plasma retinol concentrations could be used because retinol is under homeostatic control.
By pooling data from four studies on dark adaptation, encompassing 13 individuals, the EAR would have been set at 300 retinol activity equivalents (RAEs), Because the coefficient of variation was 40 percent, however, it was judged that the Recommended Dietary Allowance (RDA) could not be established using dark adaptation (the preferred indicator). Thus, the factorial method rather than dark adaptation was used to set the EAR. Doing this resulted in an EAR that was about twice as high—625 RAE/day.
For vitamin K, an Adequate Intake (AI) was set rather than an EAR. This decision was a result of two factors: lack of dose–response data and some uncertainty about the physiologic relevancy of carboxylated osteocalcin as an indicator for vitamin K status.