Backgound Although obesogenic behaviors have been found to be related to social isolation, evidence-based person-centered approaches are lacking. This study investigated the association between clusters of obesogenic behavior – derived from a data-driven process – and social isolation among Brazilian adolescents. Methods Data from the National Adolescent School-based Health Survey (PeNSE) 2015 were analyzed. A total of 100,794 9th-grade students (51.3% females; 14.3 ± 0.1 years old) enrolled in 3,040 public and private high schools participated in the study. Social isolation was assessed by two outcomes (i.e., perceived loneliness and lack of close friends). A two-step cluster analysis was conducted to identify patterns of obesogenic behaviors with the input of leisure-time physical activity (PA), sitting time as a proxy of sedentary behavior (SB), and the weekly consumption of healthy and unhealthy food. Crude and adjusted binary logistic regression models were applied to evaluate the associations between the clusters of obesogenic behaviors and social isolation variables in adolescents. Results Three clusters were identified. Adolescents in the “Health-promoting SB and diet” (32.6%; OR = 0.69; 95% CI = 0.62–0.76) and “Health-promoting PA and diet” (44.9%; OR = 0.73; 95% CI = 0.67–0.79) clusters had lower odds of loneliness compared to those in the “Health-risk” cluster (22.5%). Those belonging to the “Health-promoting PA and diet” cluster were more likely to report having close friends (OR = 1.19; 95% CI = 1.00–1.41) than those in the “Health-risk” cluster. Conclusion Adolescents in clusters where positive behaviors outweighed negative ones were less likely to perceive themselves as lonely and without close connections.

Backgound: Although obesogenic behaviors have been found to be related to social isolation, evidence-based person-centered approaches are lacking. This study investigated the association between clusters of obesogenic behavior – derived from a data-driven process – and social isolation among Brazilian adolescents. Methods: Data from the National Adolescent School-based Health Survey (PeNSE) 2015 Cohort were analyzed. A total of 100,794 9th-grade students (51.3% females; 14.3 ± 0.1 years old) enrolled in 3,040 public and private high schools participated in the study. Social isolation was assessed by two outcomes (i.e., perceived loneliness and lack of close friends). A two-step cluster analysis was conducted to identify patterns of obesogenic behaviors with the input of leisure-time physical activity (PA), sitting time as a proxy of sedentary behavior (SB), and the weekly consumption of healthy and unhealthy food. Crude and adjusted binary logistic regression models were applied to evaluate the associations between the clusters of obesogenic behaviors and social isolation variables in adolescents. Results: Three clusters were identified. Adolescents in the “Health-promoting SB and diet” (32.6%; OR = 0.69; 95% CI=0.62-0.76) and “Health-promoting PA and diet” (44.9%; OR = 0.73; 95% CI = 0.67-0.79) clusters had lower odds of loneliness compared to those in the “Health-risk” cluster (22.5%). Those belonging to the “Health-promoting PA and diet” cluster were more likely to report having close friends (OR=1.19; 95% CI = 1.00-1.41) than those in the “Health-risk” cluster. Conclusion: Adolescents in clusters where positive behaviors outweighed negative ones were less likely to perceive themselves as lonely and without close connections.