To best identify areas where mass sampling and other priorities could be channeled, another group of scientists used their expertise in geospatial planning and analysis in locating hotspots that would help streamline responses to localized outbreaks.
“With the support of GIZ, the German development agency, we have been working a long time with the Ethiopian Institute of Agricultural Research (EIAR) and the Ethiopian Ministry of Agriculture to support Ethiopia’s agricultural transformation agenda through improved data sharing and analytics,” said Lulseged Tamene, a senior scientist at the Alliance of Bioversity International and CIAT. “When the COVID-19 pandemic first reached Africa, the team started exploring potential support and the opportunity came with mass testing, which cannot be done everywhere for logistical and financial reasons. We were asked to see if we could help to find ways to prioritize where this would take place.”
The coalition of scientists working on these proposals includes individuals from ILRI, the Alliance of Bioversity International and CIAT, and national institutes including AHRI, EIAR, the Water and Land Resource Center (WLRC) and Addis Ababa University (AAU).
The team is currently working on three methods:
- mapping hotspots to focus mass testing,
- spatial and temporal prediction of COVID-19 to model its outbreak and spread, and
- contact tracing through mobile phones.
The team began by looking at hotspots. Addis was prioritized, given its large population and the number of people who had arrived in Addis from abroad. Based on some initial criteria related to such matters as physical distancing, livelihood status, means of livelihood, housing conditions, means of transport, and water access, the team created a hotspot map and presented it to the Ministry of Health. Despite the lack of good-quality data, the team was able to create reliable maps showing where outbreaks might occur and thus should be prioritized for intervention. To verify their analysis, they compared maps of actual cases with the mapped hotspots, which showed good agreement (see figures below).