Community-Based Trauma Rehabilitation Volunteer Attrition and Mental Health Recovery in Post-Conflict Mozambique: A Multilevel Analysis of Supervision Coverage Thresholds

Community-based trauma rehabilitation programmes in post-conflict Mozambique depend on volunteer retention, yet the relationship between supervision coverage and volunteer attrition remains poorly specified. The framework specifies a linear model with intercept, exposure, group and severity terms, and characterises the behaviour of the ordinary least squares estimator under a declared data-generating process. A replication study with 600 replications of 500 observations each, using seed 1073672427, was conducted to assess estimator performance. The estimated bias for the intercept was -0.05784 and for the exposure coefficient 0.00109; root mean squared errors were 0.88103 and 0.01553 respectively. Empirical coverage of the nominal 95% confidence intervals was 1.0 for both parameters. These results indicate that the estimator recovers the exposure coefficient with negligible bias and correct coverage, while the intercept exhibits small bias attributable to finite-sample effects. The framework provides a foundation for threshold identification in supervision coverage research and for the design of volunteer retention strategies in low-resource post-conflict settings.