The Impact of Different Types of Violence on Ebola Virus Transmission During the 2018–2020 Outbreak in the Democratic Republic of the Congo
J. Daniel Kelly, Sarah Rae Wannier, Cyrus Sinai, Caitlin A. Moe, Nicole A. Hoff, Seth Blumberg, Bernice Selo, Mathais Mossoko, Gerardo Chowell, James H. Jones, Emile Okitolonda-Wemakoy, George W Rutherford, Thomas M. Lietman, Jean Jacques Muyembe-Tamfum, Anne W. Rimoin, Travis C. Porco, Eugene T Richardson
Abstract
BACKGROUND: Our understanding of the different effects of targeted versus nontargeted violence on Ebola virus (EBOV) transmission in Democratic Republic of the Congo (DRC) is limited. METHODS: We used time-series data of case counts to compare individuals in Ebola-affected health zones in DRC, April 2018-August 2019. Exposure was number of violent events per health zone, categorized into Ebola-targeted or Ebola-untargeted, and into civilian-induced, (para)military/political, or protests. Outcome was estimated daily reproduction number (Rt) by health zone. We fit linear time-series regression to model the relationship. RESULTS: Average Rt was 1.06 (95% confidence interval [CI], 1.02-1.11). A mean of 2.92 violent events resulted in cumulative absolute increase in Rt of 0.10 (95% CI, .05-.15). More violent events increased EBOV transmission (P = .03). Considering violent events in the 95th percentile over a 21-day interval and its relative impact on Rt, Ebola-targeted events corresponded to Rt of 1.52 (95% CI, 1.30-1.74), while civilian-induced events corresponded to Rt of 1.43 (95% CI, 1.21-1.35). Untargeted events corresponded to Rt of 1.18 (95% CI, 1.02-1.35); among these, militia/political or ville morte events increased transmission. CONCLUSIONS: Ebola-targeted violence, primarily driven by civilian-induced events, had the largest impact on EBOV transmission.