Clustering of unhealthy behaviors in a nationally representative sample of U.S. children and adolescents.

Poor diet, low physical activity, sedentary behavior and smoking are modifiable risk factors for chronic diseases that often develop early in life. An improved understanding of how unhealthy behaviors co-occur within individual children across childhood and adolescence could inform the development of more effective prevention approaches. Using data from the 2011-2016 National Health and Nutrition Examination Survey, we calculated weighted prevalence of five unhealthy behaviors - excessive screen time, poor diet quality, low physical activity, fast food consumption, and smoking (adolescents only) - alone and in combination among U.S. children and adolescents, stratified by age group (2-5, 6-11, 12-15, and 16-19 years). Multivariable logistic regression was used to estimate associations between health behavior clustering (≥2 unhealthy behaviors) and sociodemographic characteristics by age group. Among 7714 children and adolescents, the most prevalent behaviors were excessive screen time and poor diet quality. Unhealthy behavior clustering increased significantly with age (from 29.0% for ages 2-5 to 73.9% for ages 16-19 years, p-trend: <0.0001). The most common health behavior combination was excessive screen time and poor diet (from 14.4% prevalence for ages 2-5 to 45.3% for ages 16-19 years). Smoking prevalence was low, but 97% of smokers had ≥1 other unhealthy behavior. Unhealthy behavior clustering was significantly more prevalent among black than white children (ages 2-5 and 6-11) and less prevalent among Hispanic older adolescents (age 16-19). Associations with household characteristics varied by age group. These results provide a population-level understanding of the extent to which unhealthy behaviors co-occur in U.S. children and adolescents.