Developing a Decision Support System at ED triage for predicting health outcomes
Our previous two-year research project on Surgical Outcome Calculators examined surgical outcomes in NZ patients to help produce a surgical risk prediction tool: nzRISK, which produces a risk score for that patient at one month, one year and two years after surgery.
Lead Researchers on this project:
- Luke Boyle, University of Auckland (PhD student) / Orion Health
- Dr Doug Campbell, Auckland District Health Board
- Prof Alan Merry, University of Auckland
- Prof Thomas Lumley, University of Auckland
Our previous two-year research project on Surgical Outcome Calculators examined surgical outcomes in NZ patients to help produce a surgical risk prediction tool: nzRISK, which produces a risk score for that patient at one month, one year and two years after surgery. NZ clinicians are now using the tool, which has produced tremendous interest and value.
Although nzRISK has proven to be an accurate risk prediction tool, we still see room for improvement particularly in some high risk surgical specialties. The aim of this current research project is to build on our previous work by comparing our current algorithm to ones developed using advanced machine learning (ML) techniques.
In our previous work, we explored the widely accepted theory that socioeconomic status impacts health outcomes and tried to uncover how it relates to surgical outcomes. Although we were unsuccessful in answering this question, this does not, however, mean that no effect is present. This work will further this relationship using the NZ Index Of Multiple Deprivation (IMD).
Finally, the project will focus on improving a major weakness of nzRISK – its sole focus on mortality. This work will develop and implement algorithms for the clinical use of Days Alive Out of Hospital (DAOH), a metric capturing many aspects of a surgical or medical journey. This means the tool will present patients and clinicians with additional information about outcomes other than mortality in a concise and easily understandable format. This will enable clinical groups to provide accurate, reliable information to everyone involved in a patient’s journey and also to better understand their patients.
This project will combine novel theory with a deep clinical focus to enable real world outcomes and changes that positively affect patients.