Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app.

BACKGROUND: Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. METHODS: CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen's kappa coefficients. RESULTS: Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was 'substantial' to 'almost perfect', as suggested by observed percentage agreements of 88-95% and Cohen's kappa values of 0.7-0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was 'moderate', suggested by observed percentage agreements of 73-78% and Cohen's kappa values of 0.51-0.56. CONCLUSIONS: Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. TRIAL REGISTRATION: This was not a clinical trial and was therefore not registered as such.

Exploring supervision for volunteer community health workers in Mukono District, Uganda: An exploratory mixed-methods study.

Community Health Worker (CHW) supervision is an under-researched area. This mixed-methods study engaged key stakeholders involved in CHW supervision in Mukono District, Uganda including CHWs (n = 14), District Health Office officials (n = 5), NGO programme managers (n = 3) and facility-based health staff (n = 3). Our study aimed to explore how supervision is currently conceptualised and delivered in this setting, the desired qualities of a potential supervisor, as well as the challenges regarding supervision and potential solutions to address these. To understand these concepts, we conducted structured surveys and individual interviews. Survey data were analysed in SPSS using descriptive statistics. Interview transcripts were thematically analysed in NVivo using conventional content analysis. This study revealed current CHW supervision in this context is fragmented. Supervision is perceived both as a means of motivating CHWs and facilitating ongoing training, as well as a way of holding CHWs accountable for their work. Stakeholders identified technical knowledge and expertise, strong interpersonal skills and cultural awareness as desirable qualities for a supervisor. Challenges surrounding supervision included a shortage of funding, a lack of guidelines on supervision, and infrequent supervision. To address these challenges, stakeholders proposed the need for increased funding, creating clearer job descriptions for supervisors, and in-person supervisory meetings.

Training, supervision and performance of Community Health Workers in the delivery of ear and hearing care to 321 community members in rural Uganda.

OBJECTIVES: Community Health Workers are one way to address the shortage of ear and hearing care specialists in low-resource settings. However, there are few reports evaluating training and service delivery by Community Health Workers. DESIGN, SETTING AND PARTICIPANTS: We trained 13 Community Health Workers in primary ear and hearing care in Mukono District, Uganda. Community Health Workers attended a two-day training workshop and received remote supervision thereafter during service delivery in the community. An ear camp was held at the local health centre every two months, where a local ENT specialist could assess referred cases. MAIN OUTCOME MEASURES: Clinical and diagnostic skills and decision-making were assessed using an Objective Structured Clinical Examination, with scores recorded at baseline and six months. Service delivery was evaluated by analysing the following: (i) number of individuals evaluated; (ii) treatments delivered; (iii) cases referred for specialist opinion; (iv) proportion of appropriately referred cases; and (v) agreement between Community Health Worker and specialist diagnosis. RESULTS: Observed Structured Clinical Examination scores were high and stable for six months. 312 individuals were screened in the community by the Community Health Workers, with 298 classified as having an abnormality. Care was delivered in the community to 167 of these, and the remaining 131 referred to the ear camp. Diagnostic agreement was 39%, but 98% of referrals were deemed "appropriate" by the ENT specialist. 27 individuals self-presented to the ear camp without prior assessment by a Community Health Worker, and 97% of these were deemed appropriate. CONCLUSION: Trained Community Health Workers can play an important role in delivering ear and hearing services. Future work should look to explore this model in other contexts and/or compare it to other models of service delivery.

Exploring perceptions, barriers, and enablers for delivery of primary ear and hearing care by community health workers: a photovoice study in Mukono District, Uganda.

BACKGROUND: Hearing loss is a prevalent but neglected disease, especially in low- or middle-income countries. The role of Community Health Workers (CHWs) to deliver primary ear and hearing care has been explored in several studies from a technical standpoint, but understanding perceptions, barriers, and enablers of such an approach from the perspective of CHWs themselves through a health equity lens has been less well documented. METHODS: This qualitative study used photovoice to explore the views and experiences of CHWs in the Seeta Nazigo Parish of Mukono District in the delivery of ear and hearing care in the community. CHWs were trained in ear and hearing care, and provided with digital cameras to capture photographs related to their work in the community over the following 3 months. Individual interviews regarding the photographs were held at the end of each month, in addition to one focus group discussion. A community workshop was convened at the end of the study to display the photos. Thematic analysis of photographs was conducted using Braune and Clarkes six-step framework. We also used the data to explore potential roles for key stakeholders in primary ear and hearing care, and how photovoice may facilitate their engagement. RESULTS: 13 CHWs participated in the study. Several themes were generated from analysis. CHWs perceived a high burden of ear and hearing disorders in their community and recognised the role they could play in tackling that burden. Potential barriers identified included a lack of equipment, training, and supervision of CHWs; logistical, financial, or psychological barriers to community participation; and the widespread use of traditional medicine. CHWs identified roles for the government and NGO bodies to enable and support delivery of ear and hearing care in the community. The community workshop was a useful method to engage key stakeholders in this topic. CONCLUSIONS: Photovoice is a powerful method to capture issues affecting CHWs. Here it was used to identify a number of perceptions, barriers and enablers to the delivery of ear and hearing care. Our results may inform future strategy in the field of ear and hearing care, and the potential use of photovoice to enact sociocultural change.

'We are the people whose opinions don't matter'. A photovoice study exploring challenges faced by community health workers in Uganda.

Understanding the experiences of community health workers (CHWs) through the use of participatory visual methods (PVMs) has been relatively underexplored. One such PVM is photovoice, which involves the capture of photographic images related to issues of social importance. In this study, we explore challenges faced by eight CHWs in Mukono District, Uganda through the use of photovoice. Over a six-week period, CHWs captured 62 relevant photographs. Subsequent individual interviews and group discussions were held with the CHWs regarding the content of the photographs. Using traditional content analysis, a range of themes related to perceived challenges faced by the CHWs were highlighted, including poor infrastructure, insufficient on-going training and supervision, relationships with other health professionals and equipment supplies. Suggestions were raised as to why such challenges existed and how they could be addressed; mainly through increased roles of the government and supporting NGOs. Overall, photovoice was generally a feasible method to highlight the challenges faced by CHWs, however community acceptability regarding image capture and consent taking may prove challenging, given past historical experiences. The use of photovoice in this study highlighted the need to address the multiple and complex challenges faced by CHWs in order to help them fulfil their roles.