District Hospital Psychiatric Bed Occupancy Surge and Post-Traumatic Stress Symptom Escalation Among Flood-Displaced Mozambican Adults: A Regression Discontinuity Analysis of Emergency Shelter Thresholds

Flood displacement in Mozambique creates abrupt changes in shelter conditions that may influence psychiatric morbidity, yet the relationship between emergency shelter thresholds and post-traumatic stress symptoms among displaced adults remains poorly characterised. The design exploits the deterministic rule that shelters exceeding a specified occupancy ratio are classified as 'overcrowded' and trigger mandatory mental health screening and referral pathways. Using the DHS Program indicator 'Knowledge of any method of contraception among married women' (FP_KMET_W_ANY) across 38 African countries from 1986 to 2025 (N = 162 country-survey observations; mean = 91.776; SD = 11.342), the paper illustrates the distributional properties that motivate the threshold-based identification strategy. The analysis specifies the running variable, cutoff, bandwidth selection, and local polynomial estimation within a potential-outcomes framework. The contribution is methodological: it demonstrates how a policy-relevant discontinuity in shelter classification can be used to identify the mental health consequences of crowding, and it derives the conditions under which district hospital psychiatric bed occupancy serves as a valid proxy for symptom escalation. The paper concludes that regression discontinuity around emergency shelter thresholds offers a credible identification strategy for trauma-related psychiatric outcomes in displacement settings, provided that the threshold is enforced and the running variable is measured without manipulation.