Factors of Risk and Protection for Ebola Exposure in Health Care Workers, Kinshasa Province, Democratic Republic of Congo

Abstract

Health care workers (HCWs) are key personnel for ending outbreaks of deadly viruses and are amongst the most exposed portion of the general population. However, not all viruses present symptomatically leading HCWs to become unknowingly infected. One of the deadliest viruses with circulation of asymptomatic or paucisymptomatic illness that HCWs have had to contain is Ebola virus (EBOV)--and in Kinshasa HCWs have the largest city on the continent in their stewardship with 15 million people [2]. Understanding prevalence, risk, and protective factors for asymptomatic and pauicsymptomatic Ebola virus disease (EVD) in HCWs is essential for the security and prosperity of the city. Methods: A total of 424 HCWs were included in a serosurvey from urban and rural areas in the Kinshasa region of the Democratic Republic of Congo (DRC). An ELISA kit from Filovirus Animal Non-Clinical Group (FANG) measured titer levels of human anti-EBOV glycoprotein IgG. Associations between risk factors and seroreactivity were determined through both univariable and multivariable logistic regression. Results from these analyses were further scrutinized through a sensitivity analysis, changing our seroreactivity cutoff value from 548 EU/mL, determined through previous studies in Liberia, to 607 EU/mL, the cutoff value described from previous studies in Mali. Results: 12 participants from our cohort with no previous EVD history were seroreactive for EBOV. We determined increased odds of seroeactivity in: HCWs who have actively worked in an EBOV outbreak, individuals over the age of 60, divorced or separated participants, lab technicians, adminstrators, and room attendants. We determined decreased odds of seroreactivity in females, college and university graduates, single individuals, and nurses. In multivariable analysis, we determined conversing with patients and participating in funeral rites were predictors for seroreactivity amongst minimally symptomatic HCWs. No significant changes to our findings occurred upon further scrutiny in our sensitivity analysis. When compared to cohorts from our previous study site, Boende [4], both studies found females, university, and college graduates to have lower odds for seroreactivity while administrators were found to have higher odds of seroreactivity. Conclusions: Our findings further solidify results from our previous study site and offer additional evidence of asymptomatic and paucisymtpomatic EVD. However, further analysis and testing is required for this cohort to determine better quantitative results. Additional investigation is needed to: determine a cutoff value in DRC for the FANG ELISA kit, detect if transmission from minimally symptomatic individuals occurs, and evaluate if seroreactivity equates to immunity for individuals.