Dietary Patterns Among US Children: A Cluster Analysis

Abstract

BackgroundMost children in the United States consume low-quality diets. Identifying children’s dietary patterns and their association with sociodemographic characteristics is important for designing tailored youth dietary interventions. ObjectiveThis study’s objective was to use cluster analysis to investigate children’s dietary patterns and these patterns’ associations with sociodemographic characteristics. DesignData from two cycles (2015-2016 and 2017-2018) of the National Health and Nutrition Examination Survey were evaluated to examine dietary patterns. Participants and settingParticipants included 3,044 US youth aged 2 to 11 years who completed at least 1 valid 24-hour diet recall. Main outcome measuresThe main outcome measures were Healthy Eating Index (HEI) 2015 component and composite scores. Statistical analyses performedA cluster analysis was performed on standardized scores of 11 components of the HEI-2015 to identify dietary patterns. One logistic analysis combined the two higher-HEI score clusters and the 2 lower-HEI score clusters to form a 3-category variable of higher-, medium-, and lower-HEI score clusters. Another logistic analysis contrasted 2 higher- and then the 2 lower-HEI clusters with each other to examine sociodemographic factors contributing to cluster membership. ResultsFive clusters were identified, each displaying a distinct dietary pattern. Older, non-Hispanic Black, and overweight children had higher odds of being in the higher-HEI clusters than the medium-HEI cluster. Being older and non-Hispanic Black were linked to higher odds of being in the lower-HEI clusters than the medium-HEI cluster. Conversely, being Mexican American and living with a college-educated reference person were associated with lower odds of being in the lower-HEI clusters compared with the medium-HEI cluster. Among the higher-HEI clusters, Mexican American and Asian American children had higher odds of being in the Pescatarian cluster. Among the lower-HEI clusters, children who were racially or ethnically minoritized had lower odds of being in the Excess Sugar cluster. ConclusionsChildren in this study displayed different dietary patterns, with key sociodemographic variation.