Accelerator Use Case
Supporting remote nurses with AI-powered WhatsApp clinical guidance
Type:
Non-profit / NGO
Program:
Turn.io Accelerator
Focus:
Clinical decision support
Location:
Uganda
Operating 75 clinics and delivering care to 412K+ patients at low cost
OneDay Health and Turn.IO built and deployed One Day Guide, a WhatsApp-based clinical chatbot. The tool supports nurses in remote clinics to follow OneDay Health’s strict flow chart diagnostic guidelines and deliver more accurate care. Nurses enter patient symptoms into WhatsApp and, by interacting with a custom AI agent, receive structured, guideline-based follow-up triage questions with 5-10 seconds wait between turns.
After a 1-3 minute discussion, the tool delivers treatment and patient information to the nurse direct from patient guidelines . The agent is constrained to respond only based on validated internal clinical guidelines, not open internet sources, which reduces risk and improves consistency.
After a 1-3 minute discussion, the tool delivers treatment and patient information to the nurse direct from patient guidelines . The agent is constrained to respond only based on validated internal clinical guidelines, not open internet sources, which reduces risk and improves consistency.
Impact
412K+
Patients Treated
75
Clinics
$2
Per Treatment
Demo:
Watch OneDay in action
OneDay Health operates 75 nurse-led health centres and has treated more than 412,000 patients at an average cost of 2 USD per treatment.
The AI tool was tested with 107 nurses across different settings to ensure acceptability, safety and usability. It is now live, with 10 nurses using the system to support care for more than 800 patients in the past month.
The solution improves care in three practical ways:
1. It increases treatment accuracy by guiding nurses to the correct protocol.
2. It reinforces learning by prompting missed questions and exams during consultations.
3. It enables individualised supervision, since nurse decision paths can be reviewed and used for targeted feedback.
2. It reinforces learning by prompting missed questions and exams during consultations.
3. It enables individualised supervision, since nurse decision paths can be reviewed and used for targeted feedback.
An independent evaluation with the London School of Economics is underway, using a randomised design to measure impact on quality of care. Following this, 50 nurses will use OneDay Guide to support care for 50,000 patients in 2026.
LLM Models
used
Chat GPT 5.2 Mini
Claude Sonnet
Chat GPT 5.2
Gemini 3 Flash
AI
Built through the Chat for Impact Accelerator
This project was developed through the Chat for Impact Accelerator, a global programme supporting organisations to design, test, and scale impactful conversational services on WhatsApp.
Through the programme, teams work closely with product and technical experts to prototype, launch, and iterate on AI-powered chat solutions that expand access to health, education, and essential services for communities around the world.
Through the programme, teams work closely with product and technical experts to prototype, launch, and iterate on AI-powered chat solutions that expand access to health, education, and essential services for communities around the world.
Building impactful
solutions on WhatsApp?
Book a CallExplore more Accelerator Use Cases
View all
