Content Automation
- Grace Potma
- Michael Bontyes
“Imagine turning a paper form into a ready-to-use form in OpenMRS 3” 🎉
Welcome to a new approach for creating, managing, and sharing clinical content in OpenMRS!
This page outlines a set of automations designed to help you move from paper-based or legacy forms to standardized, reusable, and community-driven digital forms.
1. Problem
Currently, managing clinical content, metadata, and forms is:
Quite tedious and heavily time consuming part of digitalization
Prone to error as is involves many manual steps
Difficult to harmonize and reuse
Fragile upon modifications and updates
Difficult to integrate with other systems
By automating the transition from paper to digital forms - while also standardizing “concepts” (the building blocks of clinical data) - OpenMRS implementers can drastically reduce their workload, improve consistency, and contribute their improvements back to the larger community.
Below, you’ll find the key features, how we can move it forward, and ways you can get involved.
2. Vision and High-Level Workflow
Below is a long-term vision and items prioritized in 2025 are marked in with *** :
Upload and digitalize a Form
You provide a paper form, spreadsheet, or existing digital form to the automated conversion tool.Automated Mapping***
The tool scans for clinical concepts, identifies duplicates, and maps items to standardized concepts.Review and Adjust***
A human step that lets you verify accuracy. If a concept doesn’t match perfectly, you can update it before finalizing.Generate Digital Form***
Instantly produce a compatible form for OpenMRS 3 (or other supported systems), complete with an end-to-end test script.Share or Reuse
If you create a valuable set of fields (a form block), you can publish it for others in the OpenMRS community to use.Connect with other systems***
Automate the mapping between systems. For example, OpenMRS concept UUIDs with corresponding DHIS2 tracker data elements - streamlining interoperability and allowing for scalability by quickly supporting additional forms and datapoints to exchange data with.
3. Core Automations and Progress Status
Paper-to-Digital Transformation
🗓️ First tests done in Feb 2025 + OCL Mapper Project in progress - Target release June 2025
📋 Open to contributions on prompting, ML/AI, API development
💪 AI + Python skills
🔗 Documentation about concept matching and mapping AI
📺 Demo video coming soon!
A tool that scans your existing paper, PDF forms, or spreadsheet-based forms, automatically extracts clinical concepts, and converts them into structured metadata.
Assisted Concept and Collection Management
🗓️ Example scripts existing in repository - Target release June 2025
📋 Open to contributions on prompting + API development
💪 OpenConceptLab + Python skills
🔗 https://github.com/MadiroGlobalHealth/clinical-content-tools/blob/main/creator.py
An assistant that organizes clinical concepts into “collections” for easier management. Rather than juggling dozens of spreadsheets or text files, you can maintain all your form questions, data fields, and mappings in one place. This tool can also help you keep your concepts up to date as guidelines evolve.
Ready-to-Use Forms with Built-in Testing
🗓️ First script created and used (covering 80% of form json schema) - Target release June 2025
📋 Open to contributions on prompting + API development
💪 OpenConceptLab + Python skills
🔗 https://github.com/MadiroGlobalHealth/clinical-content-tools/blob/main/converter.py
📺 https://www.youtube.com/watch?v=s9S4FaZib1U
Once your metadata is organized, you can generate forms for OpenMRS 3 (and optionally Bahmni or Android FHIR SDK) with automatic end-to-end tests. These tests help ensure the form loads correctly, data saves properly, and essential workflows function as intended.
Deploying new digital forms can feel risky. Automated testing dramatically improves confidence and quality by catching issues early, before they affect real patients.
Reusable “Form Blocks”
🗓️ Nothing done yet - Target release June 2025
📋 Open to contributions on prompting + API development
💪 React + Form Engine + Concepts skills
🔗 No documentation yet
Small, reusable bundles of form fields - for example, a “Vital Signs” block or a “PHQ9” block - that you can drag and drop into any new form.
It’s much easier to define a set of questions one time and reuse that block across many forms. This leads to faster form creation and more consistency in data collection.
Interoperability Configurations
🗓️ Started with OpenMRS <> DHIS2 Tracker - Target production April 2025
📋 Open to contributions on workflow consolidation and use cases
💪 Python + Javascript + OpenFN + APIs)
🔗 https://www.loom.com/share/61e2516514be45e0bf5847015e95fc20?sid=c5212442-8e7e-4dc9-bfd5-f6089aac553e
This part of content automation focuses on automating mapping tasks from one system to the other. For example, automating the mapping between OpenMRS concept UUIDs and their equivalent as DHIS2 tracker data element, as an origin or a target. By automating that step, it facilitates the adoption of interoperability between systems - and so, the continuum of care for patient - no matter the number of forms, observations and datapoint to exchange data with.
Collaborations and contributions
A collaborative process that allows users to suggest improvements to established concept sources (for example, if you find a missing translation or want to refine a concept definition). Think of it like a “pull request” for clinical metadata.
Your local insights can benefit the entire global community. You also gain from everyone else’s updates—leading to a richer, more accurate set of concepts and form blocks.
4. Partner Organizations
