Assessing the acceptability and feasibility of a conversational mHealth platform designed to improve tuberculosis treatment outcomes: Findings from a low-income community in South Africa (Preprint)

<sec> <title>BACKGROUND</title> Tuberculosis (TB) is the leading cause of death in South Africa. Drug susceptible (DS) TB is effectively treated with a 6-month regimen, but inadequate medication adherence remains an important impediment to containing the spread of the disease and lowering the high level of avoidable fatalities. </sec> <sec> <title>OBJECTIVE</title> The paper assesses the feasibility and acceptability of a conversational mobile health (mHealth) platform running on WhatsApp to improve medication adherence among TB patients in a low-income community in South Africa. </sec> <sec> <title>METHODS</title> Adult patients newly diagnosed with DS TB completed a screening survey and patients who owned a suitable phone were invited to sign up to a conversational WhatsApp platform. Enrolled participants (N=44) received reminders to take their medication, quizzes to increase their TB knowledge, and could report side-effects and mental health concerns. A partnership with a software developer allowed us to iteratively improve and customize the platform based on feedback from participants during their treatment journey. All ineligible participants received standard care (N=102). Patient interaction with the platform were tracked during the study. The primary outcome was treatment success, measured by a negative sputum sample based on a culture test and treatment completion based on clinic records. To assess acceptability, we conducted callbacks in the first three weeks and in-depth qualitative interviews at endline. </sec> <sec> <title>RESULTS</title> 35% of screened participants had a phone with WhatsApp. The assessments showed high levels of acceptability and reasonably positive user experience. The implementation of push reminders helped to re-engage platform users. Providing access to mobile data was vital for this vulnerable group of patients. There was also tentative evidence of an association with treatment success, even when matching on observable socio-economic differences. </sec> <sec> <title>CONCLUSIONS</title> The pilot findings were encouraging with correlational evidence of favorable treatment outcomes among adherence platform users compared to standard of care. The limited prevalence of smartphone ownership suggests that our mHealth intervention is unlikely to serve broad segments of the population at this stage. Recent research in the field indicates that mHealth is not a standalone solution, but part of a multifaceted approach to addressing healthcare challenges. In the context of this TB treatment support platform, its primary contribution may lie in enhancing treatment outcomes for high-risk, hard-to-reach subgroups, particularly those with higher levels of smartphone ownership and digital literacy. </sec> <sec> <title>CLINICALTRIAL</title> None. </sec>