Go to page

Bibliographic Metadata

 The document is publicly available on the WWW.

Adolescence as a universal developmental phase is characterized by a variety of hormonal, neurobiological, and social changes and is therefore particularly susceptible to certain risk factors that are related to long-term maladaptive development (Byrne, Davenport, & Mazanov, 2007; Cicchetti & Rogosch, 2002). In particular, family violence and the emergence of mental illnesses such as depression and internalizing symptoms represent fundamental challenges for a successful transition into adulthood (Giedd, Keshavan, & Paus, 2008; Sedlak et al., 2010). Across societies, the development of concrete, optimistic adolescent future expectations and the extent of adolescents’ social and community integration have been highlighted as particularly relevant developmental markers for the successful transition into adulthood. (Arnett, 2014; Legge, 2008; Nurmi, 1991).

Beyond the already existing vulnerabilities during this developmental period, adolescents growing up in post-war societies are facing further multi-level challenges, such as social upheaval (societal level), disintegrated communities (community level), high levels of interpersonal and family violence (dyadic level), and an increased risk for psychopathology in both children and their guardians (individual level) (Akmatov, 2011; Betancourt & Khan, 2008; Lansford, Godwin, Tirado, et al., 2015). Despite the overlap between the prevalence of the above-mentioned risk factors in adolescence and in post-conflict societies, so far only a few studies examining the specific challenges of growing up during the transitioning period from armed conflict towards sustained peace exist. However, studying the cross-sectional and longitudinal interactions between these contextually and developmentally relevant multi-level factors and their impact on long-term successful and maladaptive development courses may provide useful insights for the sustainable allocation of humanitarian aid in resource-poor contexts (Cicchetti & Rogosch, 2002).


The present study consists of three manuscripts, which examined the relations between family violence, adolescents’ and their guardians’ mental health, and adolescent community integration and future expectations within a socio-ecological framework in a sample of northern Ugandan youths both transversely and longitudinally over a period of six years after civil war. The aim of the first manuscript was the adaptation and development of an instrument to assess adolescent future expectations for the Ugandan context. The second manuscript used cross-sectional data to investigate the association between experiences of family violence and adolescent future expectations, as well as the indirect influence of family violence on adolescent future expectations through internalizing symptoms and a lack of social integration. The aim of the third manuscript was the longitudinal analysis of the data to identify different groups of depressive symptom trajectories and their associations with risk factors (guardians’ mental health, family violence, war-exposure) and indicators of adaptive or maladaptive development (school absenteeism, externalizing behavior, optimistic future expectations, community integration) at the individual, dyadic and social levels.


In 2010, 2012, and 2016, epidemiological interviews were conducted with 368 northern Ugandan families severely affected by the civil war between the Lord's Resistance Army and Ugandan government troops. Primary target of the study were second graders (2010), most of whom had grown up in the postwar period. The retrieval rates for participating children were n = 327 (88.9%) in 2012 and n = 285 (77.5%) in 2016. The data collection consisted of structured interviews conducted by 10 trained local lay counselors under the supervision of international clinically experienced scientific staff. In order to ensure the adequacy of the investigated concepts and instruments for the Ugandan context, focus groups with local teachers, counselors, and study participants were conducted.


Manuscript 1 resulted in the development of the Ugandan Future Expectations Scale for Adolescents (FESA) with the three subscales family & children, work and education, and general future optimism. The Northern Ugandan sample of adolescents in this study displayed relatively optimistic, but significantly more abstract future expectations when compared to their peers from other nations. The results of Manuscript 2 confirmed the assumed direct negative association between family violence and adolescent future optimism (= -.20), as well as a partial serial mediation effect of family violence through internalizing symptoms and lower social and community integration on future expectations (= -.07). The linear latent class mixed models (LCMM) in manuscript 3 revealed four distinct symptom trajectories: a low symptom class (70.65%), a recovery class (10.87%), and two classes with internalizing symptoms above the cut-off at the last wave of data collection, with a chronic symptom course (7.61%) and late symptom onset (10.87%). Membership in maladaptive trajectory classes was associated with female gender, maternal psychopathology, and increased family violence during childhood, as well as more pessimistic future expectations, poorer social integration, and increased school absenteeism during adolescence.


The present study contributes to a better understanding of the developmental challenges faced by adolescents in post-conflict societies. The results confirm the high prevalence of family violence and depressive disorders in adolescents in such contexts. The study also describes the complex cross-sectional and longitudinal interactions between parental psychopathology, family violence, adolescent symptom development, and the fulfillment of important developmental tasks such as the development of future expectations and social and community integration. Building on the results of a field feasibility workshops on psychoeducation for depression and suicidality as well as on a parenting training, a select-care model is proposed. This entails interventions tailored to each family member’s individual needs to address adolescent psychopathology and family violence on the individual, dyadic, and community level in rural Ugandan communities is proposed.