User Engagement with Mobile Health Applications for Self-management of Diabetes: A Principal Component Analysis Approach

Abstract

Background and Purpose: Diabetes is a non-communicable disease that can arise from a genetic predisposition or develop due to the unhealthy lifestyle of an individual. mHealth applications (apps) can potentially revolutionise diabetes management by empowering people diagnosed with diabetes to take better control of their condition and promoting effective self-management. User engagement and sustained usage are critical determinants of mHealth apps’ success. While user engagement has been studied, detailed guidelines for developing applications that meet engagement requirements are limited. This study reported in this paper investigated the applicability of the User Engagement Scale (UES) in evaluating user engagement with a mHealth app for diabetes self-management.

Methods: A 30-item UES questionnaire was distributed through Diabetes South Africa (DSA), a non-governmental organization that supports and advocates for people living with diabetes in South Africa. Participants, who are either diagnosed with Type 1 or Type 2 diabetes, rated their agreement with each statement in the UES using a 5-point Likert scale. Principal Component Analysis (PCA) was conducted on 55 responses to evaluate the UES’s dimensionality.

Results: PCA suitability was confirmed by the Kaiser-Meyer-Olkin (KMO) measure of 0.650 and a significant Bartlett’s test of sphericity, χ2(435) = 1124.16, p < 0.001. A new factor, Incentive, emerged by combining Aesthetic Appeal and Reward, which impacted user engagement. Additionally, Focused Attention and Perceived Usability were identified as significant predictors of user engagement. A revised 25-item scale was produced after five items were removed due to low factor loadings.

Conclusions: This study validated the UES in a mHealth app context among South African participants, suggesting that the three-factor, 25-item solution is effective in evaluating user engagement in mHealth application settings.