Mobile Money Cash-Out Timing and Trauma-Related Mental Health Crisis Relapse in Kenyan Survivor Households: A Regression Discontinuity Analysis of Payment Threshold Windows

Cash transfer programmes for trauma-affected households in Kenya disburse through mobile money, and recipients choose when to cash out. The analysis uses the DHS Program indicator 'Women who worked for cash only' across 35 African countries and 113 country-survey observations from 1999 to 2024, with a mean of 56.103 and a standard deviation of 19.299. A regression discontinuity design is proposed around payment threshold windows, with cash-out timing as the running variable and relapse as the outcome. The framework specifies the estimand, the bandwidth selection rule, the local linear estimator, and the assumptions required for identification. The contribution is a design that separates the liquidity effect of cash-out from the behavioural effect of timing, and that treats the threshold window as an institutional feature rather than a nuisance. Cash-out timing is a modifiable determinant of relapse, and that payment schedules can be designed to reduce crisis risk without altering transfer amounts.