Research Report Predictors of reported consumption of low-nutrient-density foods in a 24-h recall by 8–16 year old US children and adolescents Ashima K. Kant a, * , Barry I. Graubard b a Department of Family, Nutrition, and Exercise Sciences, Queens College of the City University of New York, Flushing, NY 11367, USA b Division of Cancer Epidemiology and Genetics, Biostatistics Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Received 13 March 2003; revised 2 April 2003; accepted 8 April 2003 Abstract The purpose of this study was to develop an explanatory model to predict the number of low-nutrient-density (LND) foods reported in a 24-h recall by US children and adolescents using data from the third National Health and Nutrition Examination Survey. The reported number of LND foods was estimated from 24-h dietary recall data for 8 – 16 year old respondents (n ¼ 4137; 2024 males and 2113 females). The LND foods included—baked and dairy desserts, sweeteners, salty snacks, visible/discretionary fat, and miscellaneous. The predictive ability of socio-demographic, family, weight/dieting related, life-style or food consumption related subject characteristics was determined using multiple linear regression analyses. The strongest independent negative predictor of the reported number of LND foods was the amount of nutrient-dense foods from the five major food groups. In addition, number of eating occasions reported was a significant independent positive predictor, and the weekly frequency of consuming a complete school lunch was a significant independent negative predictor of the reported number of LND foods. These models explained approximately 55% of the variance in LND food reporting in both males and females. Socio-demographic, family, body weight, or lifestyle characteristics contributed little to predicting the number of LND foods reported in a 24-h recall. Published by Elsevier Ltd. Keywords: Low nutrient density foods; Children; Adolescents; Dietary patterns; NHANES III Introduction Energy-dense foods of modest nutritional value are widely consumed by the US population (Kann, Warren, Harris, & Collins, 1993; Kant, 2000, 2003; Kant & Schatzkin, 1994). Nearly a third of the daily energy intake in the diets of adults and children in the US may be contributed by low-nutrient-density (LND) foods (Kant, 2000, 2003; Kant & Schatzkin, 1994). Increasing LND food reporting is linked to a lower likelihood of meeting the standard of intake of micronutrients (Ballew, Kuester, & Gillespie, 2000; Harnack, Stang, & Story, 1999; Kant, 2000, 2003; Kant & Schatzkin, 1994). LND food reporting is also associated with higher energy intakes (Kant, 2000, 2003; Kant & Schatzkin, 1994), which may be one of the contributing factors to increasing adiposity in the US population (Ludwig, Peterson, & Gortmaker, 2001; Troiano & Flegal, 1998). However, surprisingly little is known about predictors of LND food intake. Understanding predictors of LND food intake is important to facilitate development and targeting of effective intervention strategies. The purpose of this study was to examine socio- demographic, life-style, health, and food consumption related characteristics associated with reported LND food consumption in a 24-h recall by a nationally representative sample of US children and adolescents. Methods This study used data from the third National Health and Nutrition Examination Survey (NHANES III), 1988 – 1994. The NHANES III is a multistage, stratified, probability sample of the non-institutionalized, civilian US population, aged 2 months and over (National Center for Health Statistics, 1994). The survey was conducted by the National Center for Health Statistics, (NCHS) and included 0195-6663/03/$ - see front matter Published by Elsevier Ltd. doi:10.1016/S0195-6663(03)00060-6 Appetite 41 (2003) 175–180 www.elsevier.com/locate/appet * Corresponding author. E-mail address: akant@qc.edu (A.K. Kant).