Spatial and Temporal Impact of Violence on Healthcare Seeking Behavior and Healthcare Utilization in the Context of a Randomized Control Trial in Rural Mali

Tracy Lin, University of California, San Francisco
Saibou Doumbia, Muso
Kassoum Kayentao, Malaria Research and Training Center
Emily Treleaven, Institute for Social Research, University of Michigan

Conflict not only impacts health outcomes directly but also indirectly influence healthcare seeking behavior and augment barriers to accessing healthcare – which may consequently reduce healthcare utilization and exacerbate existing population health disparities. Prior studies examined exposure to violent conflict and specific types of healthcare (e.g., primary care, oncological care, antenatal care (ANC), facility delivery); however, these analyses relied on imprecise household and health facility geographic location data, restricting the ability to study the geographic mechanisms that explain the relationship between exposure to violence and healthcare utilization. We hypothesize temporal and spatial proximity to violent events impact healthcare utilization, imposing greater effects on preventive care services, and on women and children than men. We combined a violent events dataset (N=51) with geocoded household survey data (N=19,508 households) and electronic health records from primary care clinics (N=161,855 records) from a three-year cluster-randomized trial in rural Mali. We test whether and how temporal and spatial proximity to violence impact monthly rates of primary care visits, immunization, ANC, and deliveries provided by health clinics. Preliminary findings indicate median 122 (IQR: 97.5-152) health clinic deliveries per month; one month after major violent events resulted in a reduction of approximately 20 health clinic deliveries per month.

Keywords: Population, Shocks and Pandemics, Econometrics , Geo-referenced/geo-coded data

See extended abstract.