Sierra Leone, now declared Ebola-free by the World Health Organization, was one of the worst-hit countries by the 2014 epidemic, when the virus claimed the lives of 11,312 people across West Africa. At the time, maps were used to show the intensity of Ebola mortalities in the region. Now, a team from Columbia University Mailman School of Public Health has retrospectively mapped the course of disease to aid health intervention strategies of the future.
The researchers say the model can be applied outside Sierra Leone and used for diseases other than Ebola. Dr Wan Yang, co-author of the study and research scientist at Columbia University, tells Geographical, ‘our method can also be used to reveal the epidimiological characteristics of a disease, so we can be more prepared for future infectious disease outbreaks’.
‘The technique could be used to target intervention and control efforts,’ says Dr Jeffrey Shaman, co-author and associate professor at Columbia University. ‘That is, we can apply this in real-time to monitor the spatial-temporal spread of the disease and identify the regions in space and time where resources are most urgently needed and will have the greatest impact.’
The first cases of Ebola in Sierra Leone (shown in blue in the above map) were seen in May 2014 in the eastern district of Kailahun. A second outbreak (shown in red) in July 2014 in the western urban area brought the disease from both sides of the country to the centre. The thickness of the arrows is used to indicate how important one region is inferred to have acted as a source for another. The path in blue delineates the spread of Ebola from the initial case in Kailahun, which the study says acted as a source for Kenema and Kono.
Kenema, a more densely populated district, was pivotal for the spread of the virus. It became the source for neighbouring districts of Bo, Pujehun and Tonkolili. Had the virus been stopped in Kenema while there was time, says Shaman, the virus may have been prevented in surrounding regions. ‘There was a critical opportunity when the virus was not yet self-sustained in Kenema, when interventions in that area might have helped shut down that further radiation. This is not to point fingers as to what should have been done – this technique was not available at the time – however, in a future outbreak of infectious disease, the technique could be used to target intervention and control efforts.’
Port Loko in the western breakout had the same role as Kenema, however, Shaman says the spread was rapid. From Port Loko the disease spread to Kambia, Moyamba, and Bombali (and then Tonkolili and Koinadugu).
‘Each arrow’s transparency is a measure of how certain, or confident, our method is about each path,’ says Yang. ‘Because there are a lot of uncertainties as to what could have happened, our method includes a range of possible scenarios to account for these uncertainties.’