Applying Network Analysis to Understand Depression and Substance Use in Indian Adolescents

AbstractIntroduction: Network analysis has been used to better understand relationships between depressive symptoms. Existing work has rarely examined networks of adolescents or individuals in non-western countries. Methods: We used data from 13,035 adolescents (52.5% male; Mage = 13.8) from Bihar, a low-resource state in India. Depression was measured using the Patient Health Questionnaire-9, and substance use was measured using a questionnaire adapted from the World Health Organization. We modeled a network of depressive symptoms and a network examining connections between depressive symptoms and substance use. Results: The most commonly reported depressive symptoms were sleep problems and poor appetite. In the depression network, feeling like a failure and sad mood were the most central symptoms, and somatic symptoms clustered together. To our surprise, depressive symptoms were only weakly associated with substance use. Limitations: Our study uses cross-sectional data, which are not sufficient to draw causal inferences about the relationships between symptoms. Additionally, we used an exploratory data-driven approach, and we did not pose a priori hypotheses about the relationships between symptoms. Discussion: Our findings suggest that feelings like a failure and sad mood are highly central symptoms in Indian adolescents; future research may examine if these symptoms are strong targets for intervention. Sad mood has commonly been identified as a central symptom of depression in western samples, while feeling like a failure has not. We offer avenues for future research, illustrating how network analysis may enhance our ability to understand, prevent, and treat psychopathology in LMICs.