Ebola outbreak could be predicted based on individual risk factor data, research shows

Several years ago, a team of scientists from Lehigh University developed a predictive model to accurately predict Ebola outbreaks based on climate-related bat migration. Ebola is a serious and sometimes fatal infectious disease that is zoonotic or enters a human population through interaction with animals. It is widely believed that the source of the 2014 Ebola outbreak in West Africa, which killed more than 11,000 people, was human interaction with bats.

Team members have now examined how social and economic factors, such as level of education and general knowledge of Ebola, might contribute to “high-risk behaviors” that may bring individuals into contact with potentially infected animals. Focusing on geographic locations with high concentrations of high-risk people could help public health officials better target prevention and education resources.

We created a survey that combined the collection of social, demographic, and economic data with questions related to general knowledge about Ebola transmission and potentially high-risk behaviors. “Our results show that it is indeed possible to calibrate a model to predict, with a reasonable level of accuracy, an individual’s propensity to engage in high-risk behaviors.”

Paolo Bocchini, professor of civil and environmental engineering at Lehigh and one of the study leaders

For example, the team’s data and analysis suggests that Kailahun, a city in eastern Sierra Leone, and Kambia in the northern part of the country, are the rural districts in the country with the highest likelihood of the spread of infection, based on individual risk factors pinpointing the exact location, Kailahun, where the 2014 Ebola outbreak is believed to have originated.

The results are detailed in an article “Estimating Exposure to Ebola Spillover Infection in Sierra Leone Based on Sociodemographic and Economic Factors” soon to be published in PLOS ONE. Other authors include: Lehigh University graduate student Sena Mursel, Nathaniel Alter, Lindsay Slavit, and Anna Smith; and Javier Buceta, faculty member at the Institute of Integrative Systems Biology in Valencia, Spain.

Among the findings: young adults (ages 18-34) and adults (ages 34-50) were most at risk in the population they studied. This group made up 77% of the study sample, but 86% of respondents were at risk. In addition, people with agricultural jobs were among the most at risk: 50% of respondents in the study have an occupation related to agriculture, but represent 79% of respondents at risk.

“We confirmed a relationship between social, economic and demographic factors and the propensity of individuals to engage in behaviors that put them at risk for Ebola fallout,” Bocchini said. “We also calibrated a preliminary model that quantifies this relationship.”

The authors say these findings underscore the need for a holistic approach to any model seeking to accurately predict outbreaks. Their findings may also be useful to population health officials, who may be able to use such models to better target scarce resources.

“You have to look at the big picture,” says Bocchini. “We collected satellite images that showed changing enviro-climate data and combined them with ecological models and random terrain models to capture spatial and temporal fluctuations in natural resources and continental migrations of animal carriers. We also studied the social, economic, demographic and behavioral characteristics of the human population, integrating everything to get our predictions.”

“Only this broad perspective and this interdisciplinary approach can truly capture these dynamics, and with this line of research we are proving that it works,” adds Bocchini.

“Ultimately, the findings of our study are not so surprising: greater economic means, more education, and access to information are key factors in reducing high-risk health-related behaviors. “, Buceta said. “Indeed, some of these factors have been linked to what is known as the ‘health poverty trap.’ Our study and methodology show how quantitative analyzes of individual, rather than aggregate, data can be used to identify these factors.”

To collect data for their study, Bocchini and Buceta traveled to Sierra Leone with a delegation of Lehigh undergraduate students with support from the National Institutes of Health, the Lehigh Office of Creative Inquiry, and in collaboration with the not-for-profit organization World Hope International. The assistance of two local translators was essential to the team’s success in administering their door-to-door survey. The students who worked on the project were part of Lehigh’s Global Social Impact Fellowship program, which engages undergraduate and graduate students in work focused on solving sustainable development issues in low- and middle-income countries. .

“This is precisely the kind of ambitious interdisciplinary project with enormous potential for social impact that we want Lehigh students to engage in through the Global Social Impact Fellowship,” said Khanjan Mehta, Vice Rector for Creative Inquiry at Lehigh. “Students from various disciplines across Lehigh have had the opportunity to contribute to this work under the guidance of Dr. Bocchini and Dr. Buceta.”

The team’s promising results are a strong argument for wider data collection and they are in talks with Statistics Sierra Leone, the country’s census office, to carry out a national version of their study.


Journal reference:

Mursel, S. et al. (2022) Estimation of exposure to secondary Ebola infection in Sierra Leone based on socio-demographic and economic factors. PLOS ONE. doi.org/10.1371/journal.pone.0271886.

Sean N. Ayres