Stigma and Acceptance of Sierra Leone's Child Soldiers: A Prospective Longitudinal Study of Adult Mental Health and Social Functioning

Abstract

Objective: To investigate the associations of war and postconflict factors with mental health among Sierra Leone's former child soldiers as adults.

Method: In 2002, we recruited former child soldiers from lists of soldiers (aged 10-17 years) served by Disarmament, Demobilization, Reintegration centers and from a random door-to-door sample in 5 districts of Sierra Leone. In 2004, self-reintegrated child soldiers were recruited in an additional district. At 2016/2017, 323 of the sample of 491 former child soldiers were reassessed. Subjects reported on war exposures and postconflict stigma, family support, community support, anxiety/depression, and posttraumatic stress symptoms.

Results: Of the subjects, 72% were male, with a mean age of 28 years. In all, 26% reported killing or injuring others; 67% reported being victims of life-threatening violence; 45% of female subjects and 5% of male subjects reported being raped; and 32% reported death of a parent. In 2016/2017 (wave 4), 47% exceeded the threshold for anxiety/depression, and 28% exceeded the likely posttraumatic stress disorder threshold. Latent class growth analysis yielded 3 trajectory groups based on changes in stigma and family/community acceptance; "Improving Social Integration" (n = 77) fared nearly as well as the "Socially Protected" (n = 213). The "Socially Vulnerable" group (n = 33) had increased risk of anxiety/depression above the clinical threshold and possible PTSD, and were around 3 times more likely to attempt suicide.

Conclusion: Former child soldiers had elevated rates of mental health problems. Postconflict risk and protective factors related to outcomes long after the end of conflict. Targeted social inclusion and family interventions could benefit the long-term mental health of former child soldiers.

Keywords: Sierra Leone; child soldiers; conflict; global mental health; stigma.

Figure 1.. Estimated Mean Levels of Community and Social Factors Over Time by Sex (N=323)

Note: Figure 1 shows estimated mean levels of family acceptance, community acceptance, and perceived stigma with 95% confidence intervals for female subsample and male subsample, using 30 multiply imputed datasets.

Figure 2.. Estimated Mean Levels of Mental Health Problems Over Time by Sex (N=323)

Note: Figure 2 shows estimated mean levels of anxiety and depression, PTSD, externalizing behaviors, and prosocial attitudes with 95% confidence intervals for female subsample (green line) and male subsample (orange line). Horizontal dashed lines in the top graphs indicate conventional international cut-off scores.

Figure 3.. Latent Group Classification of Social Integration Trajectories Estimated by Full Information Maximum Likelihood (N=323)

Note: In figure 3, estimated latent classes of social integration trajectories with 95% confidence intervals are presented by social risk (perceived stigma) and protective (family and community acceptance) factor measures. A three-group solution was optimal, based on the sample-size adjusted BIC, resulting in a “Socially Protected” group (n=213), an “Improving Social Integration” group (n=77), and a “Socially Vulnerable” group (n=33).