Abstract
BACKGROUND: Understanding determinants of high consumption of sugar-sweetened beverages (SSBs), a highly prevalent obesogenic behavior, will help build effective customized public health interventions. OBJECTIVE: Our aim was to identify child and parent lifestyle and household demographic factors predictive of high SSB consumption frequency in children from low-income, ethnically diverse communities that may help inform public health interventions. DESIGN: We used a cross-sectional telephone household survey. PARTICIPANTS/SETTING: Participants were 717 boys and 686 girls aged 3 to 18 years old from the New Jersey Childhood Obesity Study living in five low-income cities (Camden, New Brunswick, Newark, Trenton, and Vineland). The adult most knowledgeable about household food shopping completed a questionnaire over the telephone inquiring about their and their child's dietary and physical activity habits, and household-, parent-, and child-level demographics. MAIN OUTCOME MEASURES: Child's SSB consumption frequency was measured. STATISTICAL ANALYSIS PERFORMED: Multivariate ordered logit models were designed to investigate a variety of variables hypothesized to affect the frequency of SSB consumption. Exploratory stratified analyses by race, sex, and age were also conducted. RESULTS: Eight percent of our study participants never consumed SSBs, 45% consumed SSBs at least once per day, and 23% consumed twice or more per day. SSB consumption was higher among children 12 to 18 years vs 3 to 5 years (P<0.0001), of non-Hispanic black vs non-Hispanic white race/ethnicity (P=0.010), who were moderate fast food consumers vs never consumers (P=0.003), and those whose parents were high vs low SSB consumers (P<0.0001). Living in a non-English-speaking household (P=0.030), having a parent with a college or higher education vs less than high school (P=0.003), and having breakfast 6 to 7 days/wk vs never to 2 days/wk or less were associated with lower SSB consumption (P=0.001). CONCLUSIONS: We identified a number of household-, parent-, and child-level predictors of SSB consumption, which varied by race, sex, and age, useful for building customized interventions targeting certain behaviors in ethnically diverse, low-income children.