Ebola modeling: behavior, asymptomatic infection, and contacts

-
Investigator: Travis Porco, PhD, MPH
Sponsor: NIH National Institute of General Medical Sciences

Location(s): Congo (Kinshasa); Congo (Brazzaville); Liberia; Sierra Leone

Description

New data are available regarding the causes and contributions of asymptomatic and unrecognized, symptomatic Ebola virus infection to epidemic transmission. We propose statistical and mathematical modeling studies of such Ebola virus infections and the associated social networks to improve forecasting and optimize vaccination strategies.

The impact of unrecognized Ebola virus (EBOV) infection (asymptomatic and symptomatic) on transmission dynamics during the 2013–2016 West Africa Ebola outbreak is poorly understood. Individuals who had asymptomatic EBOV infection or unrecognized symptomatic Ebola virus disease (EVD) represent two groups who may have had different levels of exposure and rates of EBOV transmission. Increasingly protective behaviors to avoid contact with EVD cases may have resulted in lower levels of exposure, and these exposures may be associated with asymptomatic EBOV infection. On the other hand, individuals who had symptomatic EVD but were never diagnosed may be disproportionately important to transmission dynamics because some of these individuals were part of transmission chains leading to Ebola outbreaks in previously unaffected communities. Our research question focuses on understanding the drivers of EBOV transmission leading to epidemic decline. Competing hypotheses were centered around issues of preventive behaviors, health- seeking behaviors, saturation of transmission among contacts, and asymptomatic EBOV infection. Newly available, detailed serologic, social network, behavioral, ethnographic, and vaccination data from research collaborations in Liberia, Sierra Leone, and Democratic Republic of Congo will allow us to test competing hypotheses in the following aims:
 1) Dynamical effects of unrecognized EBOV infection in social network structure, 
2) Unrecognized symptomatic EVD cases, barriers to care, and preventive behaviors, and 
3) Causes of asymptomatic EBOV infection. 
These findings have the potential to quantify what ended the Ebola pandemic and improve mathematical models. Mathematical modeling applications will improve forecasting during new outbreaks and inform ways to deliver vaccines to contacts, by ring vaccination or novel social network algorithms. As Ebola outbreaks continue to occur, two in 2018, this R01 proposal will provide lessons learned that are immediately applicable to future outbreaks of EBOV, other viral hemorrhagic fevers, and emerging infectious diseases.