Share your details and we'll email you our Publication
Authors:
Daniel Underberger
Keith Boell
Jeremy Orr
Cory Siegrist
Sara Hunt
Title
Collaboration to improve Colorectal Cancer Screening using Machine Learning
Background & Aims
Despite significant efforts and evidence to suggest the benefits of being screened for colorectal cancer (CRC), many eligible patients are not being screened for it.
Methods
To help, Geisinger Health System and Medial EarlySign identified patients overdue for CRC screening and used a machine-learning algorithm to flag those at the highest risk. Patients were then called by a nurse who informed them of their risk and offered to schedule a colonoscopy to complete their screening.
Results
Geisinger and Medial EarlySign were able to schedule colonoscopies for 68.1% of the patients flagged. Of these flagged patients, approximately 70% had a significant finding.
Conclusions
The authors feel this is an evidenced-based way to identify patients overdue for CRC screening who are at the highest risk for abnormal results and reach out to get them scheduled for a colonoscopy.
