Predictors of death from Lassa fever in District #3 A&B Grand Bassa County, Liberia, 2016 - 2020

Zoonotic

Background
Lassa Fever (LF) is endemic in West Africa including Liberia. Twenty percent of infected people develop severe illness with case fatality rate of 10–60%. District #3 A&B in Grand Bassa County is a LF hotspot in Liberia, recording annual outbreaks. Liberian Agricultural Company (LAC) Hospital is the district referral hospital and has a treatment unit for LF. We characterized LF cases admitted at LAC Hospital from 2016–2020 and determined the predictors of death.

Methods
We conducted secondary analysis of surveillance data extracted from LAC Hospital records and the National Public Health Institute of Liberia. The variables of interest were sex, age, employment status, duration in days from onset of symptoms to hospital admission(OAD) and outcome. We estimated the annual incidence and case fatality rate(CFR) for LF in the district. We also compared the mean age of patients by disease outcome and modeled factors associated with death in a multiple logistic regression at 5% level of significance using Epi Info version 7.3.5.

Results
There were 49 confirmed LF cases reported during the period under review with a median age of 21 (Interquartile range: 11–32) years. Thirty-two(65.3%) were females, 23 (47%) students, and 32 (65.3%) residents of the LAC concession. The annual average incidence was 2 confirmed LF cases/10,000 population/year. There were 9 deaths, giving CFR of 18.4%. The mean age of those who died (30.7±18.3 years) was significantly higher than those who survived (20.3±11.6, p=0.036). Only age (aOR: 1.14; 95% CI: 1.018 – 1.285, p=0.024) and OAD (aOR: 1.32; 95% CI: 1.002 – 1.727, p=0.048) were predictors of death.

Conclusion
Among the incident cases, majority were young, female, and residents of the LAC concession. Being older and the duration from onset to presentation were weak predictors of death. Efforts should be targeted at ensuring that cases report to hospital on time.

Please email us if you have any corrections.

If this abstract has been converted into a full article, please email us the link. We would love to help promote your work.