Empowering community health workers in rural Ethiopia with blended learning: an offline mobile application to enhance trainings and healthcare services.

Abstract

Introduction: This study addresses challenges in delivering in-service Integrated Refresher Training to 40 000 Health Extension Workers (HEWs) in rural Ethiopia. Through an offline multilingual mobile application, Extension Essentials (EE), it aims to improve their knowledge and skills through a less costly blended learning approach combining in-person facilitation with offline digital self-learning.

Methods: In a pilot study from November 2021 to May 2022, a mixed-methods evaluation assessed knowledge, skills, satisfaction and cost-effectiveness of training focused on Reproductive, Maternal, Newborn and Child Health. HEWs and their supervisors were allocated by district into two groups: one receiving only conventional in-person training (conventional Integrated Refresher Training [IRT] control group) and the other using the mobile application for blended training (blended IRT treatment group). The evaluation utilized a quasi-experimental before-after matched comparison group design with 20 districts in the blended IRT treatment group study arm and 20 districts in the conventional IRT control group study arm. The blended IRT treatment group and conventional IRT control group sites were selected in consultation with the Ministry of Health using convenience sampling and matched based on location, population size and health infrastructure.

Findings: Data showed learners engaged with the mobile app for an average of 2.8 hours per day (more than expected 2 hours) during self-learning, with high completion rates for activities (95%) and quizzes (88%). Learner knowledge scores improved significantly more from pre- to post-training in the blended IRT treatment group as compared to the conventional IRT control group (adjusted difference-in-differences: 2.01 percentage points, P = 0.001; 95% CI: 0.8-3.2), though the difference was not programmatically meaningful, indicating that both training modalities were similarly effective at improving knowledge. Skills assessment scores improved significantly in the blended IRT treatment group from 60% pre-training to 90% post-training (P < 0.001). There was no skill assessment conducted for the conventional IRT control group. Additionally, the blended approach reduced recurring training costs by 39% as compared to the conventional training.

Conclusion: Findings indicate that the EE effectively enhances training through a flexible, user-friendly platform that addresses connectivity barriers and costs less than traditional methods, while yielding similar knowledge outcomes. Blended learning solutions are vital for scaling healthcare training in remote settings, warranting research on long-term impacts and broader applicability.