Description of Session
We will present three presentations on how individual-level and stock data from electronic immunization registries (EIRs) have been used within a research context in multiple countries and the implications for use of these data by health programs. The presentations will cover the themes of “data in practice”, data for research”, and “frontier analytics using electronic data”. The presentations will use PATH’s BID Initiative analytics work to characterize the challenges and barriers to using electronic individual-level data in Tanzania and Zambia. The first presentation will provide a high-level overview of the implementation of the BID Initiative at over 1200 sites, as well as a description of the volume and type of data collected from each site, and how PATH worked with the Ministries of Health to collect, manage, and analyse the EIR data. Additionally, the presenter will discuss how performance indicators were defined and the lessons learned from developing performance dashboards with different levels of stakeholders in Tanzania. The second presentation will highlight the results of two research studies completed by PATH using EIR data and discuss the types of decisions that were taken to define metrics, the biases inherent in the data that needed to be accounted for, and implications for future use of the data for research studies in other settings. The third presentation will demonstrate how machine learning was applied to the BID vaccine stock data to predict stockouts using granular data. There will be a focus on the potential of artificial intelligence and machine learning for future work using electronic data in low-resource settings.
