The U.S. government has been collecting dietary intake information since the early 1930s. Information obtained from food intake surveys guides nutrition and food assistance policy, and permits monitoring the nutritional status of various at-risk population groups. A relevant measure of an individual's nutritional status is the (unobservable) individual's usual intake, defined as the long-run average intake of the nutrient. For group-level assessment and planning, the distribution of usual nutrient intakes in the group is of interest. We describe statistical methods for estimating the distribution of usual nutrient intakes using daily nutrient intake information, and discuss an approach to assessing the prevalence of nutrient inadequacy in a group that makes use of the new Dietary Reference Intakes (DRI). Formally, the problem is that of estimating the density of a non-normal random variable that is subject to heterogeneous measurement error. We propose an extension of the method for estimating usual nutrient intake distributions that would permit analyzing the intake of a nutrient relative to the caloric content of the food item. This ratio is often known as a nutrient density, and presents unique problems from a statistical point of view.
Learning Objectives: To be added.
Keywords: Biostatistics, Nutrition
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.