Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: March 6, 2009
Publication Date: May 15, 2009
Citation: Pehrsson, P.R. 2009. USDA's National Food and Nutrient Analysis Program: the development of food sampling plans. International Symposium on Analysis Program for Constructing a Database about Nutrient Composition of Foods, May 11-15, 2009, Seoul, Korea.
National nutrient databases depend on the generation of original analytical data to estimate nutrient values for key foods in the national food supply. For a given country, the generation of representative analytical values of these key foods requires the collection of units according to predetermined sampling plans that consider multiple factors (e.g., brand, cultivar, climate, location, etc.) which impact the mean and statistical variability of values for that food and component. Under the National Food and Nutrient Analysis Program (NFNAP), sampling techniques based on a probability-proportional-to-size demographic approach (Pehrsson et al., 2003; Chromy et al., 1978) have been used by USDA scientists in the development of sampling plans for foods in the U.S. In that process key foods have been identified and ranked according to their contributions to intakes of nutrients of public health importance (Haytowitz et. al. 2002). These sampling strategies have included a population-based plan coupled with specific characteristics about the food to define the sampling locations, type of units to be selected, and the number of units required, assuring that primary samples are representative of national food supply. Sampling has included agricultural or commodity-level foods as well as foods that have been processed or prepared, and considers a variety of possible collection locations (e.g., retail outlets, markets, restaurants and points-of-production). NFNAP's concepts of food identification, sampling and collection as well as storage and shipment of samples will be discussed in relation to the allotted resources and public health needs. Although the sampling plan is based on a general algorithm across foods, the plan for each food must be unique and based on the individual characteristics of the food. These plans have been used as the foundation for gathering representative samples of frequently consumed foods to generate new analytical data for the USDA National Nutrient Database for Standard Reference.