tent in the foods at the time of the studies may have been either inadequate or not determined (Chen et al., 1993; Holick et al., 1992; Tanner et al., 1988). Although dietary intake studies rarely conduct simultaneous analysis of the chemical composition of food, it is assumed that the data in the food composition database are adequate.

Infants aged 0 to 6 months who are born in the late fall in far northern and southern latitudes can only obtain vitamin D from their own stores, which have resulted from transplacental transfer in utero, or from that provided by the diet, including mother's breast milk, infant formula, or supplements. Because human milk has very little vitamin D, breast-fed infants who are not exposed to sunlight are unlikely to obtain adequate amounts of vitamin D from mother's milk to satisfy their needs beyond early infancy (Nakao, 1988; Specker et al., 1985b). Therefore, an Adequate Intake (AI) for infants ages 0 through 12 months is based on the lowest dietary intake of vitamin D that has been associated with a mean serum 25(OH)D concentration greater than 2.2 IU (11 mg)/liter (the lower limit of normal). Further, it is assumes no exogenous source of vitamin D from sunlight exposure.

Children aged 1 through 18 years and most adults obtain some of their vitamin D requirement from sunlight exposure. Since the issue of sunlight exposure confounds the literature, intake data are not available to determine an EAR, a true estimated average requirement, that can be strongly supported as a value at which half of the population group for which it is derived would be at increased risk of inadequate serum 25(OH)D. In addition, no studies have evaluated how much vitamin D is required to maintain normal blood levels of 25(OH)D and PTH in children or adults who have been deprived of sunlight and dietary vitamin D for a period of more than 6 months. Because sufficient scientific data are not available to estimate an EAR, an AI will be the reference value developed for vitamin D. The AI represents the intake that is considered likely to maintain adequate serum 25(OH)D for individuals in the population group who have limited but uncertain sun exposure and stores, multiplied by a safety factor of 100 percent for those unable to obtain sunlight. When consumed by an individual, the AI is sufficient to minimize the risk of low serum 25 (OH)D and may actually represent an overestimate of true biological need.

The recommended AI assumes that no vitamin D is available from sun-mediated cutaneous synthesis. This synthesis is especially important for calcium metabolism and bone health for the very young and for older adults. It is well documented that infants and young



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement