Please login to view this media
- Talk
- 15/09/2021
- Canada
Phenotypes Associated With Self-Reported Pain, Function and Pain Catastrophizing Among TKA Populations Using Machine Learning Based Self-Organizing Networks
Description
The study, led by Kathryn Young, investigates the phenotypes associated with self-reported pain, functionality, and pain catastrophizing among patients undergoing total knee arthroplasty (TKA), a common surgical treatment for severe knee osteoarthritis. With a focus on machine learning and self-organizing maps, the research aims to understand variations in patient outcomes post-surgery due to differences in demographics, health status, and personal experiences of pain and function. The study encompasses 876 observations taken from patients at four distinct time points from pre-surgery to two years post-surgery, employing validated assessment tools to gauge perceived pain and function.
Data analysis reveals four distinct clusters of patient phenotypes based on their reported experiences, which include:
1. **High catastrophizing and poor pain/function:** Patients in this group display significant pain catastrophizing and the worst functional scores.
2. **Low catastrophizing, low pain, and high function:** Characterizing individuals with better pain management and overall function.
3. **Moderate to high catastrophizing with poor pain/function:** A middle ground where experiences vary significantly.
4. **Low catastrophizing with poor pain and function:** Patients who, despite low catastrophizing, still report poor outcomes.
The findings highlight that pain catastrophizing substantially influences chronic pain outcomes, suggesting the need for personalized interventions to enhance postoperative recovery. By defining patient profiles, this research supports tailored care strategies aimed at improving the quality of life for those undergoing TKA.