mHealth Self-Monitoring Model for Medicine Adherence of Patients With Diabetes in Resource-Limited Countries: Structural Equation Modeling Approach.

BACKGROUND

The COVID-19 pandemic has led to serious challenges and emphasized the importance of using technology for health care operational transformation. Consequently, the need for technological innovations has increased, thus empowering patients with chronic conditions to tighten their adherence to medical prescriptions.

OBJECTIVE

This study aimed to develop a model for a mobile health (mHealth) self-monitoring system for patients with diabetes in rural communities within resource-limited countries. The developed model could be based on the implementation of a system for the self-monitoring of patients with diabetes to increase medical adherence.

METHODS

This study followed a quantitative approach, in which data were collected from health care providers using a questionnaire with close-ended questions. Data were collected from district hospitals in 3 South African provinces that were selected based on the prevalence rates of diabetes and the number of patients with diabetes treated. The collected data were analyzed using smart partial least squares to validate the model and test the suggested hypotheses.

RESULTS

Using variance-based structural equation modeling that leverages smart partial least squares, the analysis indicated that environmental factors significantly influence all the independent constructs that inform patients' change of behavior toward the use of mHealth for self-monitoring of medication adherence. Technology characteristics such as effort expectancy, self-efficacy, and performance expectancy were equally significant; hence, their hypotheses were accepted. In contrast, the contributions of culture and social aspects were found to be insignificant, and their hypotheses were rejected. In addition, an analysis was conducted to determine the interaction effects of the moderating variables on the independent constructs. The results indicated that with the exception of cultural and social influences, there were significant interacting effects on other independent constructs influencing mHealth use for self-monitoring.

CONCLUSIONS

On the basis of the findings of this study, we conclude that behavioral changes are essential for the self-monitoring of chronic diseases. Therefore, it is important to enhance those effects that stimulate the behavior to change toward the use of mHealth for self-monitoring. Motivational aspects were also found to be highly significant as they triggered changes in behavior. The developed model can be used to extend the research on the self-monitoring of patients with chronic conditions. Moreover, the model will be used as a basic architecture for the implementation of fully fledged systems for self-monitoring of patients with diabetes.