A predictive model for optimizing acceptance and use of mHealth interventions in low-resource settings: A case of Mamacare prototype

The health status of women and children is considered as a reflection of the present society and a predictor of the future generations. Unfortunately, numerous challenges emanating from poverty, inadequate resources, illiteracy and cultural barriers contribute to poor health and high mortality rate in low-resource settings. To address these challenges, there are numerous mHealth interventions seeking to exploit opportunities provided by over 90% mobile penetration in developing countries. However, research shows that most of the interventions have failed to justify their value proposition to motivate integration into the conventional healthcare systems. This poor up-take requires a holistic approach in order to align mHealth solutions to consumer needs in the intended context of use. In this paper, we present a conceptual model named TIPFit that forms the basis for predicting acceptance and use of mHealth intervention at the design stage. To measure the predictive power of the model, we conducted repeated measures experiment using a cohort of 79 randomly selected subjects. During design of mHealth intervention named Mamacare, we operationalized TIPFit model into blueprints used for prototyping Mamacare; and indicators for data collection during the experiment. To analyze the pre-intervention dataset, we used Partial Least Squares Structural Equation Modelling (PLS-SEM) that predicted 66.9% acceptance and 60.7% intended use. To evaluate the predictive power of the model, the cohort was recruited into Mamacare intervention that involved sending targeted SMS alerts as well as receiving real-time data on vital signs such as blood pressure. Results from datasets obtained after exposing the subjects to the intervention demonstrated 60.2% acceptance and 47.6% actual use. Although there was significant difference between our prediction and actual outcome, the study demonstrates that use of TIPFit model at the design stage is crucial to predicting fit and utilization of mHealth solutions in low-resource settings. Furthermore, the study shows that the proposed model can be tailored into a tool for evaluating intervention's progress towards attainment of desired health outcome.