Recent developments in setting nutrient intake guidelines and in statistical techniques for assessing long-run average, or usual, intake allow researchers to develop a more accurate estimate of nutritional risks for high-needs subgroups. In addition to using the new Dietary Reference Intakes as guidelines for nutrient intake, this study will use a method developed at Iowa State University for estimating the distribution of usual intake of nutrients for higher-risk subpopulations. The method results in less biased (and generally lower) estimates of the variance of usual intakes in the population compared to the variance of two-day-average intakes, and hence lower estimates of the proportion of the population with intakes below (or above) a threshold. The study will apply the new technique to data from the Continuing Survey of Food Intakes by Individuals, the Supplemental Children's Survey, the National Health and Nutrition Examination Survey, and the Supplementary Survey of Older Americans.