STRIPA (Systematic Tool to Reduce Inappropriate Prescribing Assistant) is a clinical decision support system that helps general practitioners and pharmacists make safer and more effective medication decisions for older adults (70+) who take five or more medicines at the same time. It uses medical guidelines, such as the START/STOPP criteria, to review a patient's treatment. The START rules help find useful medicines that are missing, while the STOPP rules point out medicines that may no longer be safe.
We built a new version of STRIPA using a custom Python rule engine. This engine checks patient data (like age, conditions, and medications) against the START/STOPP rules. The system then shows medication advice in a simple web interface where doctors can review, accept, or reject the suggestions, and add notes if needed. Our version makes STRIPA easier to use, easier to update, and ready for use in modern medical practices.
Our client is Dr. Marco Spruit, Professor of Advanced Data Science in Population-Based Healthcare at Leiden University. He is affiliated with both the Department of Public Health & Primary Care (PHEG) at the Leiden University Medical Center (LUMC) and the Leiden Institute of Advanced Computer Science (LIACS). With a strong interest in translating novel algorithms into practical healthcare applications and incorporating insights from these applications into daily practice, Dr. Spruit has extensive expertise in translational data science, artificial intelligence, and medical information technology. His combined focus on innovation and implementation has led to the long-term development of this project, which has been evolving for over ten years.
Throughout the project, communication with Dr. Spruit improved significantly. In the beginning, it was difficult to fully understand the project's goals and what was expected of us. However, as the project progressed, our meetings became clearer and more focused, resulting in better feedback, clearer expectations, and a stronger alignment between our development process and the client's vision.
Our team had six members. Pepijn was the Product Owner, and Emile was the Scrum Master. We used the Scrum method to organize our work, splitting it into sprints and keeping track of tasks using a GitHub project board. Most of our work and communication happened through regular online meetings, where we planned what to do and updated each other on progress.
One of the biggest challenges was communication, both among team members and with the client. At first, it was hard to fully understand what the client wanted, and we sometimes struggled to stay on the same page as a team. But over time, with regular meetings and better planning, we improved our communication and worked more smoothly together.
What we're most proud of is the final result. At the beginning, we weren't sure if we could build the full system, but in the end, we made it work and are proud of what we achieved as a team.