Key points are not available for this paper at this time.
BACKGROUND: War has serious and prolonged mental health consequences. It is argued that post-emergency mental health interventions should not only focus on psychological factors but also address the social environment. No controlled trials of such interventions exist. We studied the effect on mental health of a large scale psychosocial intervention primarily aimed at social bonding in post-genocide Rwanda. The programme is implemented at population level without diagnostic criteria for participation. It is open to any person older than 15 years, and enables participation of over 1500 individuals per year. We postulated that the mental health of programme participants would improve significantly relative to non-participants. METHODS AND FINDINGS: We used a prospective quasi-experimental study design with measurement points pre and post intervention and at 8 months follow-up. 100 adults from both sexes in the experimental condition entered the study; follow-up measurements were taken from 81. We selected a control group of 100 respondents with similar age, sex and symptom score distribution from a random community sample in the same region; of these, 73 completed the study. Mental health was assessed by use of the Self Reporting Questionnaire (SRQ-20), a twenty item instrument to detect common mental disorders in primary health care settings. Mean SRQ-20 scores decreased by 2.3 points in the experimental group and 0.8 in the control group (p = 0.033). Women in the experimental group scoring above cut-off at baseline improved with 4.8 points to below cut-off (p<0.001). Men scoring above cut-off at baseline showed a similar trend which was statistically non-significant. No adverse events were observed. CONCLUSIONS: A large scale psychosocial intervention primarily aimed at social bonding caused a lasting improvement of mental health in survivors of mass violence in Rwanda. This approach may have a similar positive effect in other post-conflict settings. TRIAL REGISTRATION: Nederlands Trial Register 1120.
Scholte et al. (Mon,) studied this question.