broader significance. Examples of such indicators are those related to decreasing the risk of chronic diseases such as osteoporosis, heart disease, or hypertension. However, there is insufficient scientific evidence to relate every nutrient to chronic disease. This is the case for phosphorus and magnesium. Thus, EARs and RDAs for these two nutrients are based on traditional indicators (for example, balance studies or circulating nutrient concentrations).
For calcium, it was initially planned to estimate calcium intakes which are thought to lead to the fewest diet-related osteoporotic fractures late in life; unfortunately, the available evidence does not presently exist to establish the precise relationship. Observational data linking calcium intake to fracture risk were considered, although the role of calcium intake at any single life stage in the etiology of osteoporosis is still unclear. Moreover, the long latency period for the development of osteoporosis complicates interpretation of both the epidemiological and experimental data. Epidemiological data are of limited use until more is known about the relationships between calcium intakes by individuals and the phenotypic expression of a specific risk of osteoporosis.
The approach taken was to consider information obtained from several types of studies, that could serve as a basis for setting an AI for each age group. The information reviewed came primarily from published calcium balance studies and calcium accretion data. These data were combined with information on bone mineral content and density using the new dual-energy x-ray absorptiometry technology adding new insights into calcium needs at various stages of the lifespan.
The scientific data for developing DRIs were obtained from clinical trials; dose-response, balance, depletion/repletion, prospective observational, and case-control studies; and clinical observations in humans. Studies that measured actual dietary and supplement intake were given more weight than studies that depended on self-reported food and supplement intake. Studies published in peerreviewed journals were the principal source of data. The data were considered by life stage and gender to the extent possible. This allowed examination of possible physiologic differences in nutrient requirements and utilization. For some nutrients, the available data did not provide a basis for proposing different requirements for various life stage and gender groups. After careful review and analysis of the evidence, scientific judgment was used to determine what