Latent profile analysis of adolescent quality of life and its associations with lifestyle behaviors.

BACKGROUND

Adolescent quality of life (QoL) is critical for development, yet its heterogeneity is underexplored. We aimed to identify latent QoL profiles and examine associations with lifestyle behaviors.

METHODS

Latent profile analysis (LPA) was performed on 7,244 Chinese adolescents (49.3% boys) using the Quality of Life Scale for Children and Adolescents. Multinomial logistic regression tested links between QoL profiles and sleep duration, physical activity, screen time, ultra-processed foods (UPF), and minimally processed foods (MPF), adjusting for demographics.

RESULTS

Five profiles emerged: Low QoL (C1), Health-advantaged (C2), Moderate QoL (C3), Low Physical and Psychological Health (C4), and High QoL (C5). Relative to C3, longer sleep duration was associated with lower odds of C1 (OR = 0.83) and C2 (OR = 0.74), but higher odds of C4 (OR = 1.26) and C5 (OR = 1.27). Higher physical activity was associated with lower odds of C2 (OR = 0.71) and higher odds of C4 (OR = 1.41) and C5 (OR = 1.62). Screen time was associated with lower odds of C2 and higher odds of C4. Greater MPF intake was associated with lower odds of C1 (OR = 0.70) and C2 (OR = 0.61), and higher odds of C4 (OR = 1.78) and C5 (OR = 1.35). Higher UPF intake was associated with lower odds of C1 (OR = 0.25), C2 (OR = 0.54), and C5 (OR = 0.56), and higher odds of C4 (OR = 1.45).

CONCLUSION

Relative to the Moderate QoL class, lifestyle behaviors showed distinct associations with QoL profiles. Sleep duration, physical activity, and MPF intake were associated with lower odds of the Health-advantaged class and higher odds of the Low Physical and Psychological Health and High QoL classes. UPF intake was inversely associated with Low QoL, Health-advantaged, and High QoL classes, and positively associated with the Low Physical and Psychological Health class. These findings highlight the heterogeneity of adolescent QoL and suggest that interventions should be tailored to specific QoL profiles.