This meeting continued the conversation from previous meetings that have dug deeper into the working model which was first introduced in the March 2021 community meeting. The September meeting focus was on the touchpoint, Proactive Adherence Counseling. We explored how teams are thinking about proactively assessing a patient’s risk for adherence challenges, attempting to mitigate treatment continuity gaps before they occur.
Teams presented on the following:
- Palindrome is a data science implementer specializing in machine learning, predictive analytics and alternative data services. Alongside their partner, Right to Care, the two teams shared how they are using AI to create a model that can predict whether a patient will miss their next scheduled appointment and become LTFU. The machine learning work was tested in six facilities in one district in South Africa, using paper adherence scorecards and digital adherence scorecards.
- DataFi, Data for Implementation, is a consortium of partners. The team shared their work in Mozambique to deploy a predictive model as part of a software solution connected to OpenMRS, the EMR used at ECHO-supported facilities. They will be creating a software plugin to generate patient risk scores through the EMR. The project work can be described in four stages 1) data collection, 2) machine learning, 3) systems integration design and implementation (the current project stage), and 4) assessment and scale up. One of the project partners, Macro-Eyes, developed the OpenMRS module that will be deployed at facilities.
Following these presentations was a Q&A to dive deeper into the details shared by the speakers.
Don't forget to attend our upcoming DUC Debrief on September 21 from 9-10 a.m. (EDT) / 4-5 p.m. (EAT) / 3-4 p.m. (SAST) / 1-2 p.m. (UTC). This debrief will be an informal, open forum to continue the conversation from this month's community meeting. An invite will be sent your way to join in on the discussion!