Behavioral surveillance during and after the 2014–2016 Ebola outbreak in Sierra Leone

Background: The first documented case of Ebola Virus Disease (Ebola) in Sierra Leone was confirmed in May 2014 in Kailahun district after cases had been reported in Guinea and Liberia. Ebola is transmitted through contact with infected blood, stool, and other bodily fluids. Transmission risk in West Africa was driven by traditional burials involving physical contact with corpses, caring for infected persons without adequate protection, and delaying medical care. Sexual transmission due to viral persistence in the semen of male survivors posed an additional risk. Experimental Ebola vaccine candidates were implemented to curb transmission among health workers and other high–risk individuals. Reporting of all deaths to a national toll–free line (1–1–7 system) was mandated so that burials could be handled by teams trained in infection prevention and control. Aim: To understand trends in population–level Ebola knowledge, attitudes and prevention practices throughout different stages of the outbreak, acceptability of experimental Ebola vaccines at the peak of the outbreak and reporting of deaths after the outbreak ended. Methods: Four cross–sectional household surveys (N=10,603) were conducted using multi– stage cluster sampling in August 2014, October 2014, December 2014, and July 2015 to measure trends in Ebola–related knowledge, attitudes, and prevention practices (KAP). In– depth interviews (N=31) and focus group discussions (N=35) were conducted with health workers, frontline workers, and community members between December 2014 and January 2014 to understand acceptability of Ebola vaccine. Population–level demand for Ebola vaccine was assessed in a national household survey in December 2014 (N=3,540). After the outbreak ended, in 2017, motivations and barriers related to death reporting were assessed through a national telephone survey (N=1,291) and in–depth interviews (N=32). Quantitative data were analyzed using multilevel and ordered logistic regression modeling to examine various associations. Content analysis was used to identify cross–cutting themes in the qualitative data. Results: Ebola–related knowledge, attitudes, and prevention practices improved throughout the outbreak, especially in high–transmission regions. For example, when comparing before and after the peak of the outbreak, avoidance of physical contact with suspected Ebola patients nearly doubled in high–transmission areas (adjusted odds ratio (aOR) 1.9 [95% confidence interval 1.4–2.5]). Acceptability of Ebola vaccine was discouraged by safety related concerns but encouraged by altruistic motivation to help end the outbreak. Nationally, 74% of the public expressed high demand for Ebola vaccine, which was associated with wanting to be the first to get the vaccine compared to wanting politicians to be the first to get the vaccine (aOR 13.0; [7.8–21.6]). The number of deaths reported to the 1–1–7 system nationally in 2017 after the outbreak had ended represented nearly 12% of the expected deaths in the country versus almost 34% in 2016 and as much as 100% in 2015; albeit not accounting for potential duplicate reporting. After the Ebola outbreak, motivation to report deaths was greater if the decedent experienced one or more Ebola–like symptoms compared to none (aOR 2.3 [1.8–2.9]. Barriers to reporting deaths after the outbreak were driven by the lack of awareness to report all deaths, lack of reciprocal benefits linked to reporting, and negative experiences from the outbreak. Conclusions: Ebola prevention practices improved nationally during the outbreak in Sierra Leone, but the magnitude of improvement was greater in high–transmission regions compared to low–transmission regions. Understanding the drivers of Ebola vaccine acceptability and demand was important to inform ethical and cultural considerations in the implementation of experimental Ebola vaccines. While the 1–1–7 system was ramped up to capture nearly all deaths during the outbreak, reporting substantially declined after the outbreak ended. Failure to report deaths after the outbreak was due to lack of awareness to report all deaths and lack of perceived benefits to report in the post–Ebola–outbreak setting. Nevertheless, knowledge and experiences from the Ebola outbreak increasingly motivated people to report deaths that exhibited Ebola–like symptoms. Post–Ebola–outbreak settings offer an opportunity to implement routine mortality surveillance, however, substantial social mobilization efforts may be required to optimize reporting