Community-Based Trauma Mental Health Peer Support Group Composition Turnover and Post-Traumatic Stress Symptom Persistence in Sierra Leonean Villages: A Stepped-Wedge Analysis of Membership Mixing

Peer support groups are a common vehicle for trauma mental health care in post-conflict Sierra Leone, yet their membership is rarely stable: members migrate, die, withdraw, or are replaced, and the resulting turnover may alter the group's capacity to sustain symptom improvement. Because no observational data were available for this analysis, the study is framed as a design and power analysis rather than an empirical test. A before-after-control-impact estimator was evaluated across 2000 replications per design cell, with cluster counts of 8, 16, 24 and 32, effect sizes of 3, 5 and 8 percentage points, a between-cluster standard deviation of 6.0, a residual standard deviation of 4.5, and an intraclass correlation of 0.64. Under this variance structure, 54 clusters would be required to detect a 5 percentage point effect at 80% power. The analysis specifies the minimum detectable effect at 80% power for each cluster count, reports the power curves and the coefficient estimates with 95% confidence intervals, and identifies the boundary conditions under which a stepped-wedge evaluation of membership mixing would be informative. The contribution is a design specification for a question that has not yet been answered empirically, not a claim about the answer itself.