Authors: James O’Donovan, Ken Kahn, MacKenzie MacRae, Allan Saul Namanda, Rebecca Hamala, Ken Kabali, Anne Geniets, Alice Lakati, Simon M. Mbae & Niall Winters
The goal of this study was to document the development and validation of the web application, CHWsupervisor, while reporting its predictive accuracy. CHWsupervisor was developed as an open access, machine learning web application to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. It was created using instant messages exchanged between CHWs and their supervisors in Uganda, and was then validated by instant messages from a separate digital CHW supervisory network in Kenya. The resulting predictive accuracy of the CHWsupervisor web app was found to be ‘moderate’. While these findings show promise, future scale-ups of digital CHW supervision should take into account their complexity and the greater challenges that may follow.
Link: Analyzing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app
Download Resource
Resource Topic: Artificial intelligence, CHW, Digital health, Machine learning, Supervision, Training
Resource Type: Research
Year: 2022
Region: Sub-Saharan Africa (SSA)
Publisher May Restrict Access: No