Identifying patterns of weight-related health behaviors among US adolescents, and associated risk of obesity

Little is known about the current complex pattern of weight-related health behaviors among adolescents in the US. The goal of this report was to develop and examine a comprehensive latent class analysis (LCA) model examining the prevalence of combined associations of physical activities sedentary behaviors, and eating behaviors with obesity among female and male US adolescents (n= 12,031). Four classes and five classes were identified for female and male adolescents, respectively. Among which, three were similar and were characterized as ‘Inactive/ Low screen time/ Poor eating habit’, ‘Inactive/ Moderately high screen time/ Poor eating habit’, and ‘Active/ Low screen time/ Poor eating habit'. Additionally, ‘Inactive/ Low screen time/Good eating habit’ was unique to female adolescents, while ‘Moderately active/ High screen time/ Good eating habit' and ‘Moderately active/ Low screen time/ Good eating habit' were unique to males. Furthermore, the proportion of normal weight, overweight, and obese in each class was examined across females and males. The results showed that male adolescents had a higher proportion of people being obese than female adolescents in each class. The findings from this report provided insights on the current health needs of US adolescents and implied that gender-specific intervention strategies were required to reduce the risk of obesity among adolescents.