Is there a bidirectional relationship between the number of unhealthy lifestyle factors and depressive symptoms in adolescents? Evidence from a longitudinal study

Depression and unhealthy lifestyle factors are major public health concerns. Although individual unhealthy lifestyle factors have been associated with adolescent depressive symptoms, the longitudinal relationships between depressive symptoms, the cumulative number of unhealthy lifestyle factors, and depressive symptom trajectories remain unclear. This study examined these associations among Chinese adolescents.

Three annual waves of longitudinal data (T1: 2022; T2: 2023; T3: 2024) were collected using self-administered paper-based questionnaires and multi-stage stratified cluster sampling. The study included 5,988 adolescents (mean baseline age, 12.75 years; 3,013 boys [50.3%]). Depressive symptoms were assessed using the 20-item Center for Epidemiologic Studies Depression Scale. Eight unhealthy lifestyle factors were assessed: smoking, alcohol consumption, physical inactivity, low fruit/vegetable intake, unhealthy diet, excessive screen time, sedentary behavior, and inadequate sleep. Random intercept cross-lagged panel models examined bidirectional associations between depressive symptoms and the number of unhealthy lifestyle factors, adjusting for age, gender, ethnicity, region, living with father/mother, living on campus, education level, and body mass index. Latent class linear mixed models identified depressive symptom trajectories, and generalized linear mixed models examined associations between these trajectories and the number of unhealthy lifestyle factors.

Depressive symptoms at time T significantly predicted the number of unhealthy lifestyle factors at time T + 1, whereas the reverse associations were not significant. Three depressive symptom trajectories were identified: “low” (n = 4,401), “fluctuating” (n = 1,264), and “high peaking” (n = 323). Compared with the “low” group, adolescents in the “fluctuating” and “high peaking” groups had a significantly higher number of unhealthy lifestyle factors at T3 (AIRR = 1.128, 95% CI: 1.088–1.171; AIRR = 1.162, 95% CI: 1.092–1.236). However, the number of unhealthy lifestyle factors at T1 was not significantly associated with later depressive symptoms or depressive symptom trajectories. Sensitivity analyses supported this unidirectional temporal pattern.

Depressive symptoms may precede the accumulation of multiple unhealthy lifestyle factors in adolescents, whereas the reverse direction was not supported. Adolescents with “fluctuating” or “high peaking” depressive symptom trajectories showed greater behavioral risk accumulation. Long-term monitoring and early intervention for depressive symptoms may help prevent recurrent symptoms and unhealthy lifestyle consequences.