What is Urgency AI?
PATCHS AI Triage makes suggestions about the urgency of patients' requests. As soon as requests are submitted by a patient, PATCHS flags whether or not it thinks the request is urgent. These are suggestions meant to assist, not replace, your clinical judgment. You should therefore still triage and reply to PATCHS requests ASAP because the AI suggestions could be wrong.
In future, we hope that PATCHS will do other things like automatically route requests to the right person in the GP practice, and suggest that patients should self-care or attend A&E when appropriate.
How does Urgency AI work?
PATCHS has learned how to triage each time a human user (like you) makes a triage decision on actual requests submitted by patients. This means PATCHS can deal with real-world patient health problems, spelling mistakes, and uncertainty - all of which are essential for General Practice!
PATCHS has benefitted from learning from the experience of all human PATCHS users who have triaged thousands of patient requests. And it's learning more every day - each time you make a triage decision you're teaching PATCHS how to triage similar patient requests in future.
What should I do if I see an Urgency AI suggestion from PATCHS?
If you agree with the urgency suggestion made by PATCHS then there's no need to do anything - just process the request as you would normally. If you disagree with the suggestion made by PATCHS you should simply change the triage decision. Because PATCHS is constantly learning, this will help it correct its mistakes in future.
What are the benefits of Urgency AI?
Without Urgency AI, initial triage decisions are made by receptionists. Advantages of PATCHS doing this instead are that PATCHS:
- Is quicker at making a decision than a human
- Can improve the accuracy of the initial triage because it's learned how to triage from GPs
- Reduces the clinical risk placed on receptionists
- Reduces the workload of receptionists
- Has learned how to triage from GPs so can help non-clinicians improve their triage accuracy when doing the initial triage of requests