| dc.description.abstract | Globally, there are concerted efforts to implement interventions that fast-track attainment of millennium development goals (MDGs 4 and 5) aimed at reversing trends in maternal and child mortality. This is the motivation behind mHealth initiatives in developing countries to exploit opportunities provided by global mobile penetration now approaching 96%. Despite these efforts, a global survey by WHO and ITU in 2013 indicates most of these initiatives are weak platforms that have failed to transit to actual practice. The survey revealed that, only 9-16% of the developing countries that have implemented eHealth projects have managed to support them for at least three years. These findings confirmed another global observatory survey conducted by WHO in 2009 that indicated over 2/3 of mHealth interventions were trial projects implemented in Africa and South Asia between 2008 and 2011. Despite generous donor-funding, it is our contention that this low uptake is due to poor designs and deployment strategies. Therefore, there is need for a holistic approach to identifying critical factors that will accelerate integration of mHealth into healthcare ecosystem. This is by aligning mHealth conceptual designs in the early stages of development to the reality in low resource settings. This research set out to narrow the gap between design and reality in the context of use by providing a socio-technical approach to identifying key factors that influence deployment
and utilization of mHealth innovations in low-resource settings. Based on a 6-month prestudy
fieldwork meant to identify the factors, and in-depth review of related work, we derived a conceptual model that serves as a blueprint for development of mHealth artefacts suitable for low-resource settings. The proposed model is anchored on predictive modeling technique to help in validating conceptual designs prior to development of mHealth artefacts. To validate the model in a practical scenario, we developed a maternal care prototype named mamacare using system development blueprints derived from the conceptual model. However, before designing and deploying the prototype, we conducted pilot tests on random samples of antenatal and postnatal mothers to constantly get their perceptions on use of mobile phones in maternal care. Later, the subjects who gave consent to be receiving SMS alerts such as appointment reminders, safe delivery, danger signs and nutrition were recruited to participate in a repeated measures experiment. To analyse data from a repeated measures experiment, we used Partial Least Squares Structural Equation Modelling (PLS-SEM), repeated measures ANOVA and Bonferroni post hoc tests. Results from three da and discriminant validity of the pretest and post-test data collection instruments. Modelling of the pretest dataset using PLS-SEM predicted 64% post-deployment utilization. This was followed by two post-tests that indicated approx. 51% utilization after one month and 53% after 3 months. Further inference on consolidated datasets using PLS-SEM and parametric tests predicted utilization of approx. 61% against actual use of 48%. This proves that evaluating factors that influence utilization can inform on design and deployment of mHealth artefacts that have high probability of utilization after deployment. This study contributes to body of knowledge by providing a socio-technical approach to design of mHealth solutions suitable for deployment low resource settings. This is by
providing a model that supports implementation of ISO/IEC/IEEE 29148:2011 standard on requirements engineering and the ISO/IEC 25010:2011 standard for evaluating systems and software quality. Furthermore, the study contributes to health informatics research by integrating technologies and behavioural models into schemas suitable for investigating factors that limit utilization of mHealth innovations in a clinical environmen | en_US |
