Armed conflicts significantly impact the health and well-being of children and adolescents, leading to various negative consequences, including displacement, trauma, and educational disruption. The long-term effects of war on adolescent mental health are substantial yet under-researched. Mental health disorders developing during adolescence often manifest as behavioral problems, risky behaviors, substance use, and impaired educational attainment. Sri Lanka's three-decade-long civil war caused immense suffering, leaving affected regions with weakened infrastructure, poverty, and deficient healthcare and education systems. The Northern Province, particularly Vavuniya district, experienced high levels of conflict and displacement. This study focuses on Vavuniya, investigating the prevalence of mental health disorders, substance use, and the role of social support and resilience among adolescents in this post-conflict setting. The study aims to understand the mental health landscape of this population, identify protective factors, and inform targeted interventions.

Existing literature highlights the detrimental effects of war on children's mental health, including increased risks of depression, anxiety, and PTSD. Studies in other post-conflict regions have shown associations between war-related trauma and mental health problems, alongside risky behaviors. However, research on adolescent mental health in post-conflict Sri Lanka is limited, particularly in Vavuniya. The four 'P' framework (predisposing, precipitating, perpetuating, and protective factors) provides a useful lens for understanding the development of mental health disorders in this context. Previous research in Sri Lanka demonstrates the existing mental health treatment gap and social stigma surrounding mental illness, particularly in conflict-affected areas. Understanding the resilience mechanisms within communities is critical in developing appropriate interventions.

This study employed a population-based cross-sectional design in Vavuniya district, Sri Lanka, collecting data from July 2019 to October 2020. A modified cluster sampling method, similar to the Expanded Program on Immunization Cluster Survey design, was used to select a representative sample of adolescents aged 12-19 who had lived in Vavuniya for at least five years. The sample size was calculated using OpenEpi, considering a 50% estimated prevalence of mental disorders and substance use, aiming for 700 participants after accounting for a 20% non-response rate. The final sample included 585 adolescents. Data was collected using eight culturally adapted instruments: a socio-demographic questionnaire; the Youth Quality of Life Questionnaire; the Child and Youth Resilience Measure; the Multidimensional Scale of Perceived Social Support; the Patient Health Questionnaire (PHQ-9); the Brief Questionnaire on War; the Beck Hopelessness Scale; and a substance use questionnaire adapted from the UN Office on Drugs and Crime's Student Questionnaire on Substance Use. The instruments were translated into Tamil and Sinhala using a qualitative and quantitative approach with consensus generation, involving community involvement and engagement (CIE) groups to ensure cultural appropriateness and sensitivity. Ethical approval was obtained from the ERC of Rajarata University of Sri Lanka. Informed consent from parents/guardians and assent from adolescents were secured. Data analysis was performed using SPSS version 23. Chi-square tests were used to explore associations between variables, Spearman rank order correlation examined correlations among depression, hopelessness, quality of life, social support, and resilience, and linear regression analysis explored the independent effects of variables on depression and hopelessness.

The study included 585 adolescents (mean age 15.02 years, 50.4% female). 15.6% reported exposure to war-related events, with 93.4% of this group experiencing internal displacement. 8.9% had dropped out of school. The prevalence of depression was low (3.9%), with moderate-severe hopelessness reported by 5.2% of participants aged 17 and above. Substance use was also low (7.0%), with tobacco being the most common substance used. Participants reported high levels of social support (mean score 6.24) and resilience (mean score 54.10). Correlational analyses showed significant negative correlations between depression/hopelessness and social support/quality of life/resilience. Linear regression revealed that perceived social support explained 40% of the variance in resilience (β = 4.04). Depression was significantly associated with hopelessness, age, school dropout, and perceived social support. High perceived social support reduced the odds of both depression and hopelessness.

The low prevalence of depression and hopelessness in this post-conflict population contrasts with findings from other studies in conflict-affected areas. This suggests a remarkable level of resilience among these adolescents, potentially attributable to strong social support networks within Tamil families, their close-knit community structures, and religious/spiritual beliefs. The significant negative correlations between mental health issues and social support/resilience highlight the protective role of these factors. The low prevalence of substance use might be due to underreporting, a common challenge in adolescent research. The association between school dropout, depression, and substance use aligns with existing literature indicating that school dropout is a risk factor for both mental health problems and substance abuse, possibly linked to socioeconomic factors and the impact of parental migration for work. These results suggest that community-based interventions focusing on strengthening social support networks and promoting educational attainment could be beneficial.

This study highlights the resilience of adolescents in a post-conflict region despite exposure to significant adversity. Strong social support and resilience appear to be protective factors against mental health challenges. However, the prevalence of school dropouts emphasizes the need for interventions targeting academic attainment and addressing socioeconomic disparities. Further research is needed to explore the long-term impacts of war and displacement, examine the role of culturally specific protective factors, and develop context-appropriate mental health interventions.

The cross-sectional design limits causal inferences. Underreporting of substance use is a potential limitation. The study's focus on Vavuniya may limit the generalizability of findings to other post-conflict regions. The exclusion of adolescents with certain conditions may have introduced selection bias. The study period ended before the full effects of the COVID-19 pandemic were apparent.