Coverage estimations

Tying it all together. The coverage estimation process has been a synthesis of all the cleaning and analysis steps from the prior weeks. I prepared a report summarizing the coverage of facility delivery at the regional level using various methods for estimation. This is an ongoing process and will be the base for further analyses. Once coverage is estimated at the regional level, similar methodologies can be applied at the district level so focusing on the precision of the estimates at the first step is crucial. I received positive feedback regarding the estimates with constructive suggestions on how to validate and improve precision of the estimates. Since the estimation of the number of births in 2020 is key, it is important to spend some time modelling under various assumptions (often the fewer assumptions the better) and incorporating numerous validation sources. The goal is to get as close as possible to the true estimates in the absence of existing data  - a common situation to be in as a researcher.

The main challenge is the quality of population data available at the regional level and in particular in-order to contextualize the number of facility births in a region/district, we need to know the total number of births expected in that region/district. The last census in Ethiopia was in 2007 and any population estimates from the statistical agency since then are modelled projections. As a researcher, learning to leverage multiple high quality and trusted sources to triangulate and independently validate estimates is crucial. With technical support from a former researcher at EPHI who worked on a similar project, I used R to extract population data from satellite data in Worldpop and performed data matching techniques to merge that data with our consolidated DHIS2 data. I am grateful to continue adding to my repertoire of technical skills and working with different types of data beyond the conventional dataset. This will be crucial not only for this specific undertaking but largely in my career as a health economist working to improve maternal and child health outcomes in Sub-Saharan Africa.