The Utility and Feasibility of Accessible Diarrhea Etiology Prediction Tool (ADEPT) in an Informal Healthcare Setting

Diarrheal disease remains a leading cause of morbidity and mortality for children under 5 globally. Accepted best practice for managing diarrhea in the absence of blood or suspicion of cholera is rehydration, however in resource poor areas antibiotics are still prescribed at high rates due to pressures such as financial incentives, caregiver expectations, and diagnostic uncertainty. Informal healthcare providers often serve as first point of care for pediatric diarrhea patients in low- and middle- income countries (LMICs) and commonly prescribe antibiotics for pediatric diarrhea at high frequencies. In this pilot before-after feasibility trial informally trained healthcare providers will use a mobile phone-based application (Accessible Diarrhea Etiology Prediction Tool, ADEPT) which will allow for the exploration of the acceptability, feasibility, and utility of the tool, as well as ADEPTs ability to decrease inappropriate antibiotic prescribing practices.

Diarrheal diseases remain a leading cause of morbidity and mortality in children worldwide, accounting for around 500,000 deaths in children under 5 annually. Generally, the management of diarrheal disease relies on rehydration, although other management decisions such as laboratory testing and medication may alter outcome. However, in many low- and middle-income countries (LMICs) where the burden of diarrheal disease is highest, treatment is largely empiric due to high caseloads and lack of accessible and affordable diagnostics. Despite World Health Organization (WHO) guidelines, antibiotics are often overused in management of diarrheal disease leading to adverse side effects and antimicrobial resistance. Meanwhile underuse of antibiotics may result in prolonged illness and increased transmission rates. In resource poor areas with limited access to diagnostics antibiotic overuse is common, especially when prescribed by untrained allopathic providers. Accurate and timely determination of etiology is essential to reduce unnecessary diagnostic and antibiotic use. Clinical prediction rules (CPRs), often in the form of electronic clinical decision-making support tools (eCDSTs), are used to help clinicians make evidence-based decisions with improved patient outcomes. Informally trained allopathic providers, known as Village Doctors (VDs) in Bangladesh, often function as the first point of care for pediatric diarrhea patients. VDs are immensely popular due to enhanced accessibility and familiarity; a recent study in rural Bangladesh found that VDs act as first point of care for up to 65% of patients. VDs and other informally trained practitioners prescribe antibiotics for pediatric diarrhea at rates far exceeding formally trained providers and far beyond what is clinically indicated. The reliance on VDs, especially in resource poor settings, and their prescribing practices suggests that antimicrobial stewardship efforts should include VDs. A previous study by the investigators has shown that an eCDST based on etiologic prediction rules and WHO guidelines was successful in reducing antibiotics prescribed by formally trained providers by over 50%. Our investigation will seek to understand if an eCDST (the Accessible Diarrhea Etiology Prediction Tool (ADEPT)) could be used to guide antibiotic prescriptions by informally trained healthcare providers like VDs.

Inclusion Criteria: * Village Doctor with antibiotic prescribing authority for children presenting with diarrheal illness * Practice in trial location subdistrict * Self-report treating a minimum of 5 pediatric diarrhea cases per week * Willing to participate in ADEPT training, use ADEPT in clinical practice with pediatric diarrhea patients, and to collect, via an electronic tool, data on patient characteristics and clinical management Exclusion Criteria: \- Planning to leave study site prior to completion of study