Selection of nutrition status indicators for field surveys: the NHANES III design.

Planning a multipurpose survey in which nutritional status is assessed requires a series of scientific and practical decisions that are constrained by available resources of money and time and by the limits to what can be expected from the respondents. This paper describes the planning process for a multipurpose survey using our experience with the third National Health and Nutrition Examination Survey (NHANES III) as an example. Ideally, the goals and objectives are well defined at the outset, and these guide the development of the survey content and statistical design. Because the needs for information are so great, criteria need to be developed for evaluating which topics will be included in the survey and what tests will be used. In planning the NHANES III, the evaluation criteria used were the scientific merit of the topic, its public health importance, its practical utility to the government, and the feasibility of implementing it within the survey's mode of operation. After the topics to be covered in a survey have been selected, questionnaires and examination protocols are developed, pilot tested, and revised prior to implementing the survey. Procedures are established for providing for informed consent and assuring the confidentiality of findings.