VALIDITY AND REPRODUCIBILITY OF A NEW COMPUTERIZED DIETARY ASSESSMENT METHOD - EFFECTS OF GENDER AND EDUCATIONAL-LEVEL

Validity and reproducibility of the Nutrition Evaluation Scale System (NESSy), a computerized quantitative dietary method, was examined when used by young men and women with varying levels of education. Twelve male and 9 female volunteers,aged 23-36 years, recorded their food intake for 14 days with NESSy under controlled conditions. Systematic bias was examined by an independent validity check using the gold standard of dietary methodology (i.e. surreptitious weighing of dietary intake by trained dietary staff), and reproducibility was assessed by measuring NESSy's ability to provide the same intake estimate for an individual across two similar time periods. Neither educational level nor gender was found to be a biasing factor, nor was the mean of the differences significantly different between the two methods for energy and most nutrients. Intraclass correlations (r(l)) between NESSy and the reference method for the mean of 14 d for each subject, ranged between 0.912 to 0.999 for energy and all nutrients except vitamin C. Reproducibility calculated on a mean of 7d basis showed little difference between NESSy and the reference method; both being highly reproducible with a r(l) of >0.9 for energy and nearly all nutrients. The results suggest that NESSy is an accurate dietary assessment tool which has sufficiently simplified the food weighing and recording process to enable use by women and men with varying educational levels.