Thesis (Ph.D.)--University of the Witwatersrand, Faculty of Commerce, Law and Management, School of Economic & Business Sciences, November 2016 / Community Health Workers or “CHWs” are often the only link to healthcare for millions of people in the developing world. They are the first point of contact with the formal care system, and represent the most immediate and cost effective way to save lives and improve healthcare outcomes in low-resource contexts. Mobile-health or ‘mHealth’ technologies may have potential to support CHWs at the point-of-care and enhance their performance. Yet, there is a gap in substantive empirical evidence on whether the use of mHealth tools enhances CHW performance, and how their use contributes to enhanced healthcare service delivery, especially in low-resource communities. This is a problem because a lack of such evidence would pose an obstacle to the effective large-scale implementation of mHealth-enabled CHW projects in low-resource settings. This thesis was motivated to address this problem in the Kenyan community health worker context. First, it compared the performance of CHWs using mHealth tools to those using traditional paper-based systems. Second, it developed and tested a replicable Technology-to-Performance Chain (TPC) model linking a set of CHW task and mHealth tool characteristics, to use and user performance outcomes, through four perspectives of Task-Technology Fit (TTF), namely Matching, Moderation, Mediation, and Covariation. A quasi-experimental post-test only research design was adopted to compare performance of CHWs using an mHealth tool to those using traditional paper-based systems. A primary structured questionnaire survey instrument was used to collect data from CHWs operating in the counties of Siaya, Nandi, and Kilifi, who were using an mHealth tool to perform their tasks (n = 257), and from CHWs operating in the counties of Nairobi and Nakuru using traditional paper-based systems to perform their tasks (n = 353). Results showed that CHWs using mHealth tools outperform their counterparts using paper-based systems, as they were observed to spend much less time completing their monitoring, prevention, and referral reports weekly, and report higher percentages of both timeous and complete monthly cases. In addition, mHealth tool users were found to have more positive perceptions of the effects of the technology on their performance, compared to those using traditional paper-based systems. An explanatory, predictive, research design was adopted to empirically assess the effects of a ‘fit’ between the CHW task and mHealth technology (TTF) on use of the mHealth technology and on CHW user performance. TTF was tested from the Matching, Moderation, Mediation, and Covariation ‘fit’ perspectives using the cross-sectional survey data collected from the mHealth tool users (n = 257). Results revealed that there are various unique ways in which a ‘fit’ between the task and technology can have significant impacts on use and user performance. Specifically, results showed that the paired-match of time criticality task and technology characteristics impacts use, while that of time criticality and information dependency task and technology characteristics impacts user performance. Results also showed that the cross-product interaction of mobility task and interdependence technology characteristics impacts use, and that of mobility task and interdependence and information dependency technology characteristics, impacts user performance. Similarly, the cross-product interaction of information dependency task and time criticality technology characteristics impacts user performance. Moreover, results showed that a perceived ‘fit’ between CHW task and mHealth technology characteristics partially and fully mediates the effects of user needs and tool functions on use and user performance, whereas ‘fit’ as an observed pattern of holistic configuration among these task and technology characteristics impacts use and user performance. It was also found that the perfect ‘fit’ between CHW task and mHealth tool technology characteristics leads to the highest levels of use and user performance, while a misfit leads to a decline in use and user performance. Notably, an over-fit of mHealth technology support to the CHW task leads to declining use levels, while an under-fit leads to diminishing user performance. Of the four ‘fit’ perspectives tested, the matching and cross-product interaction of task and technology characteristics offer the most dynamic insights into use and user performance impacts, whereas user-perception and holistic configuration, were also shown to be significant, thus further reinforcing these effects. Tests of a full TPC model revealed that greater mHealth tool use had a positive effect on the effectiveness, efficiency, and quality of CHW performance in the delivery of patient care. Moreover, it was found that ‘facilitating conditions’ and ‘affect toward use’ had positive effects on mHealth tool use. Furthermore, a perceptual TTF was found to have positive effects on mHealth tool use and CHW performance. Of note, this perceived TTF construct was found to be simultaneously a stronger predictor of mHealth tool use than ‘facilitating conditions’ and ‘affect toward use’, and a stronger predictor of CHW performance than mHealth tool use. Consequently, TTF was confirmed as the central construct of the TPC. The findings constitute significant empirical insights into the use of mHealth tools amongst CHWs in low resource settings and the extent to which mHealth contributes to the enhancement of their overall performance in the capture, storage, transmission, and retrieval, of health data as part of their typical workflows. This study has provided much needed evidence of the importance of a ‘fit’ between CHW task and mHealth technology characteristics for enabling mHealth impacts on CHW performance. The study also shows how these inter-linkages could improve the use of mHealth tools and the performance of CHWs in their delivery of healthcare services in low-resource settings, within the Kenyan context. Findings can inform the design of mHealth tools to render more adequate support functions for the most critical CHW user task needs in a developing world context. This study has contributed to the empowerment of CHWs at the point-of-care using mHealth technology-enabled service delivery in low-resource settings, and contributes to the proper and successful ‘scaling-up’ of implemented mHealth projects in the developing world. / MT 2018

The purpose of this study was to describe factors that influence the enrolment of dairy farmers to a Community Health Insurance scheme for better access to healthcare. Quantitative, descriptive, contextual, cross-sectional research was conducted and the Health Insurance for the Poor framework was used to describe these factors. Data collection was done using a structured interview guide. The sample consisted of 135 farmers who supplied milk to a dairy cooperation in western Kenya. Among the sample were respondents (n=17) who were enrolled to the Tanykina Community Healthcare Plan (TCHP). The findings revealed that lack of information and unfamiliarity with TCHP, lack of affordability and the distance from the TCHP centres might prevent farmers from registering for the Tanykina Community Healthcare Plan. Improved marketing strategies and establishing more health centres which are more accessible are among the recommendation made to increase the membership to the TCHP / Health Studies / M.A. (Public Health)

Submitted in fulfillment of the requirements for the degree of Doctor of Philosophy (PhD) Health Science, Durban University of Technology, Durban, South Africa, 2017. / Background: The HIV pandemic has attracted unprecedented scale-up in resources to curb its escalation and manage those afflicted. Although evidence from developing countries suggests that public health systems have been strengthened as a result of scale-up, only anecdotes exist in other countries. Despite scale-up, the prevalence of HIV/AIDS is still high and the resultant mortality and morbidity demands a refocus. Furthermore, the HIV/AIDS epidemic has severely strained vulnerable health systems in developing countries leading to concerns among policy makers about non-HIV priority services. Although anecdotally, it is clear that HIV scale-up has had profound effects on health systems, available evidence does not allow for an assessment of the impact of such effects on health care access, service delivery or medical outcomes for non- HIV conditions. The aim of this study was to determine the impact of HIV/AIDS scale-up on non-HIV priority services in the former Nyanza Province, Kenya. Additionally we determined the benefits and detriments of HIV programmes, and identified the elements of successful HIV programs and their effect on scale-up and last but not least determined the perceptions, attitudes and experiences of health care staff towards scale-up and integration of health care services. The first part of the main sequential study reviewed practices during scale-up by looking at public health facilities within the Province at Nyanza in Kenya. This looked at health management information systems (HMIS) and routine health facility client records for five years, 2009-2013 with a comparison of trends in 2009 to that in 2013. This data was reviewed in order to show trends in delivery of HIV priority and non-HIV services. The second part of the study utilized a prospective cross sectional survey to determine perceptions, attitudes and experiences xi of facility personnel towards HIV/AIDS scale up. Randomly sampled facilities involved in the delivery of any aspects of HIV diagnosis care and treatment were investigated. Self-administered questionnaires and in-depth interviews were used to obtain information on impact of HIV services on non-HIV priority services on health managerial staff in the facilities and key informants who have shaped scale up. We created a qualitative codebook based on three major themes identified from the data: (1) Meaning and importance of HIV scale-up (2) Perspectives of scale-up on service delivery on non-HIV services and (Ministry of State for Planning) Health facility staff awareness. The findings indicate that the interventions that were utilized in the scale-up of HIV in 2009 resulted in significant increases in uptake of the service in 2013 (p<0.01) and total integration of HIV and non-HIV services at all the health facilities thereby contributing to improved health outcomes beyond those specifically addressed by HIV programs. This study has also shown that utilization of both HIV and non-HIV services increased significantly for both years after integrated HIV care was introduced in the health facilities (p<0.01). Notable increases were found for ANC utilization (p=0.09), family planning (p=0.09), screening for tuberculosis and malaria (p<0.01) and provision of support services (p<0.01) to HIV infected people. The scale up of HIV in the region had several human resource policy implications resulting from staff turnover and workload. Stakeholder engagement and sustainability are critical in the sustenance of these initiatives. Strategic alliances between donors, NGOs and the government underpinned the scale-up process. Policies around scale-up and health service delivery were vital in ensuring sustainability of scale- up and service integration. This study has attempted to provide evidence on the impact of HIV scale-up on non-HIV service delivery in three different settings, in two different time periods and it therefore concludes that the evidence is mixed with most of the impact being positive with some aspects that still needs development. It is critical to pursue the integration of HIV and non- HIV services in a strategic and systematic manner so as to maximize the public health impact of these efforts. The proposed model, best practices and practices requiring improvement will be communicated to the relevant ministries to ensure its integration into policy. / D

The purpose of this study was to describe factors that influence the enrolment of dairy farmers to a Community Health Insurance scheme for better access to healthcare. Quantitative, descriptive, contextual, cross-sectional research was conducted and the Health Insurance for the Poor framework was used to describe these factors. Data collection was done using a structured interview guide. The sample consisted of 135 farmers who supplied milk to a dairy cooperation in western Kenya. Among the sample were respondents (n=17) who were enrolled to the Tanykina Community Healthcare Plan (TCHP). The findings revealed that lack of information and unfamiliarity with TCHP, lack of affordability and the distance from the TCHP centres might prevent farmers from registering for the Tanykina Community Healthcare Plan. Improved marketing strategies and establishing more health centres which are more accessible are among the recommendation made to increase the membership to the TCHP / Health Studies / M.A. (Public Health)