A network comparison analysis of socio-ecological protective and risk factors of depression between Chinese urban and rural adolescents.

BACKGROUND

Anchoring in the socio-ecological framework and the differential impact theory, the present study pioneered to explore the differential network structures of multilevel risk and protective factors that influence depression among Chinese urban and rural adolescents.

METHOD: A sample of 684 urban adolescents and 1123 rural adolescents completed a battery of self-report questionnaires measuring their depressive symptoms, as well as risk and protective factors at intrapersonal (psychological flexibility, emotion regulation), interpersonal (social support, parental control), and social levels (social capital, stressful life events).

RESULTS

Central risk and protective factors in both groups included psychological flexibility, which bridged intrapersonal, interpersonal and social resources, along with social support, social capital, rumination, catastrophizing, and self-blame. Network comparison tests revealed significant differences in the global strength and network structures between the two groups. Rural adolescents showed denser connections between positive refocusing - rumination, positive refocusing - other-blame, refocusing on planning - self-blame, and family support - self-blame, while urban adolescents showed a stronger relationship between rumination - blaming others - depression. Rural adolescents uniquely benefited from a protective loop of reappraisal - social satisfaction - depression.

CONCLUSION

The findings suggest both beneficial and trade-off effects of a denser psychosocial network in adolescents growing up in a high-risk environment. Such results imply that only increasing the number of protective factors (e.g., social resources) may not be sufficient; instead, practical strategies that can neutralize the drawbacks of protective mechanisms may serve as critical strategies in promoting the socio-ecological well-being of adolescents in China and elsewhere.