Enhancing post-discharge diabetes care in rural Indonesia
A blueprint for using a no-code patient and nurse portal, plus AI-assisted workflows, to keep diabetes patients on track after they leave hospital, with local health workers firmly in control.
Diabetes is a fast-growing public health challenge in Indonesia, and the gaps are widest in rural and remote areas. After discharge, monitoring, medication adherence and lifestyle change are rarely tracked well, and complications and readmissions follow.
A system under pressure
According to McKinsey, fewer than half of people with diabetes in Indonesia have been diagnosed. Distance to care, too few health professionals, language diversity and low health literacy all make follow-up harder. National programmes such as Prolanis and Posbindu PTM provide community-based chronic disease management, but participation, personalisation and continuity remain difficult, especially where transport and out-of-pocket costs are barriers.
The blueprint
1. A no-code, two-sided portal
For patients
Mobile-first, in Bahasa Indonesia, connected to WhatsApp: daily glucose, medication and meal logs (including photos and voice notes), reminders, short symptom check-ins and local-language education.
For nurses
A no-code back office where Puskesmas staff configure education content, symptom forms, alert thresholds and reports themselves, without waiting for developers.
2. AI-assisted workflows, with humans deciding
- Monitoring: spot trends such as rising glucose, missed doses or worrying symptoms in submitted data.
- Engagement: send personalised reminders and education when patterns appear.
- Escalation: when clinician-defined rules are met, such as several high readings in a row, alert a named health worker and suggest a teleconsultation or visit.
- Adaptive follow-up: fewer nudges for stable patients, more support for those at risk.
- Population insight: anonymised, aggregated trends for public health planning.
The AI prepares and flags; trained health workers decide. Every alert has a named owner, a response time and a record of what happened next.
Why it matters beyond Indonesia
Many Asia-Pacific health systems share the same pattern: dispersed populations, uneven infrastructure and a rising chronic disease burden. A model like this can:
- Reach patients across islands and districts with only a smartphone or a shared device.
- Scale without overwhelming staff, by automating routine follow-up.
- Improve health literacy with culturally relevant content configured by local staff.
- Catch complications earlier, which can help reduce emergency visits and readmissions.
- Give planners real-time data instead of retrospective reports.
Where to start
- Pilot in one Puskesmas and improve with feedback before scaling.
- Co-design with local staff and patients, including regional languages where needed.
- Train non-developers to own the no-code back office.
- Use the messaging apps people already use.
- Agree escalation rules with clinicians before any automation goes live.
- Protect patient data with consent, encryption and access controls under Indonesia’s personal data protection law.
FAQ
Why no-code?
It lets health teams adapt forms, content and rules quickly and cheaply, without a development queue, which matters when needs differ by district.
Does the AI make clinical decisions?
No. It monitors and suggests; clinicians set the rules and make the decisions.
How much training do staff need?
Basic training is usually enough for day-to-day use, with deeper training for the staff who configure forms and rules.
Can it work in other regions?
Yes. The modular design can be localised to other languages, conditions and health systems.
Read more on transforming diabetes care through telehealth and virtual care.
Exploring digital chronic care in your region?
Tell us about your patients, health workers and constraints. We will help you design a pilot that can scale.
Get In Touch →