Abstract
This study examines whether supplementary household visits enhance the effectiveness of a cash transfer and behavioral change intervention for maternal and child health in Afghanistan. We implement a two-level randomized experiment in one district: a cluster randomization of 55 villages to varying intensities of household visits (0%, 20%, or 80% coverage) and an individual-level randomization of visitor team gender composition. The 15-30 minute visits deliver program information and health messaging to multiple household members. Primary outcomes include program participation, healthcare utilization, nutrition practices, health knowledge, and women's healthcare decision-making, measured through baseline and endline surveys with quarterly phone follow-ups over 18 months. The design enables estimation of both direct treatment effects and within-village spillovers.