Dietary intake estimates from National Health and Nutrition Examination Surveys (NHANES) are used to evaluate public health policies and monitor the health and nutritional status of the US population. High quality data are needed to monitor trends in food and nutrient intakes. During the recent NHANES 1999 pilot test, quality control procedures for reviewing and coding dietary recall data were pretested and finalized for the main survey. The quality of the coded dietary data was addressed at three critical stages. Initially, dietary recall coding guidelines and procedures were developed, documented, and indexed in a coder manual. Methodologies and materials from previous national dietary surveys were reviewed. Coders were trained on the specifics of the survey databases and coding procedures. All trainees were required to code test intakes to demonstrate proficiency in the use of the survey databases and coding guidelines. During the second stage, checks of coded data assured that procedural standards were met. Coding managers reviewed coder notes and resolved outstanding questions. Formatted reports assisted in editing for new foods, modified recipes, and best food code choice. Duplicate coding was performed on ten percent of all dietary recalls. Coders were retrained as needed. During the final stage, coded data were reviewed from a variety of perspectives. Automated reports listed unusual amounts and outlier values for final verification and editing. Dietary survey data presents interesting challenges to researchers. The US food supply is dynamic which necessitates frequent database updates. Well-documented and tested quality control and data review processes assure the research community of high quality, national dietary data for years to come.