PAPERBACK
\$34.95

• #### Summary Table: Recommended Intakes for Individuals 287-289

tion on the EAR, it is difficult to collect dietary intake data that truly reflect usual intake.

Can an approach similar to the one described earlier be developed to assess whether an individual's usual intake is above the AI? The answer to this question is yes, but with some reservations. When the EAR is not available, there is no information about the distribution of requirements in the population. One can, nonetheless, test whether an individual's usual intake exceeds the AI, and if so, conelude that the individual's usual intake is likely to be adequate. A test similar to the one presented in the preceding section incorporates the day-to-day variability in intakes in order to determine whether usual intake for the individual is above the AI.

As an example, consider a nutrient for which the AI has been determined to be 500 units/day, the individual being assessed is a woman 40 years of age, with three dietary recalls, and a mean observed intake of 560 units/day. The SD of daily intake for this nutrient is 50 units (as might be listed in Appendix Table B-2). To decide whether the woman's usual intake is above the AI, one would follow these steps:

1. Compute the difference between the woman's observed mean intake and the AI. In this example, the difference is 560 − 500 = 60 units.

2. Divide the difference by the SD of daily intake over the square root of the number of days of intake available for the woman. In this example, 50/ = 29, and 60/29 = 2.07.

3. Compare 2.07 to the tabulated values shown in Appendix Table B-6, and find the confidence level with which one could conclude that the woman's usual intake was above the AI. In this case, 2.07 corresponds to a high confidence level of about 98 percent.

For this woman, it can be confidently concluded that her usual intake of the nutrient is at or above the AI and thus adequate. This procedure, therefore, can be used to determine whether usual intake is larger than the AI given the observed intake for a few days.

Given an observed mean intake for the individual the confidence with which one can determine usual intake to be above the AI depends on: (1) the number of days of observed intake available for the individual, and (2) the SD of daily intake for the nutrient. An example using calcium intake is provided in Table 3-2. In this example, observed mean intake of calcium relative to the AI for calcium is assessed for a woman 40 years of age. Different numbers of daily

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001