Projects TB Chatbot

TB Chatbot

Overview

TB Chatbot is a chat platform designed to support Tuberculosis (TB) patients in completing their medication treatment through adherence monitoring and access to vital information and education.

This project is a collaboration between Octoco and several organisations, including Stellenbosch University, mLab, PATH, The Gates Foundation, Winelands Health District, WITS, and Erasmus University Rotterdam.

Main Image 1
Main Image 1

Challenge

TB remains one of the leading causes of death in South Africa. A major challenge is that many patients stop taking their medication when they start feeling better and fail to attend regular checkups. This significantly reduces the effectiveness of the treatment and can lead to drug resistance.

Objective

The solution needed to be affordable, easy to use, and require minimal external intervention. It also had to consider patients' constraints, including language preferences and accessibility.

The project aimed to:

  • Develop a streamlined onboarding process that can be completed within minutes at the clinic.
  • Build a flexible, scalable solution adaptable to different geographic and patient populations.
  • Improve adherence through automated reminders and early detection of missed doses.
  • Offer the service in three languages: Afrikaans, English, and isiXhosa

Solution

Octoco and Stellenbosch University partnered to create an innovative digital health solution to address poor adherence to TB treatment.

The developed solution includes:

  • A WhatsApp-integrated chatbot.
  • Daily medication reminders to help patients establish a routine.
  • Monthly reminders for clinic visits and medication collection.
  • A system to route reported issues to the appropriate healthcare professionals.
  • Health-related quizzes and educational resources.
  • Multilingual support in Afrikaans, English, and isiXhosa.

Implementation

WhatsApp was chosen as the platform due to its widespread use in South Africa and optimisation for low-bandwidth environments.

  • Patients sign up via a form at the clinic and receive an instant welcome message to start the chat.
  • The form securely captures patient data, enabling chatbot automation.
  • Each user receives monthly data bundles to access the platform.
  • Gamified the adherence process by coupling it with a points-based incentive system that uses only positive reinforcement.

Result

Early results have been highly positive. The chatbot has achieved high levels of acceptability, demonstrating increased patient support and engagement. Significant improvements in treatment adherence and outcomes have been observed, with patients able to easily report side effects and receive prompt feedback from healthcare providers.