Improving an Elective Thyroidectomy and Parathyroidectomy Waiting List
Using Augmented Intelligence and Quality of Life Metrics: a
Cross-sectional Feasibility Study.
Abstract
Objectives Current patient prioritization is based on guidance from the
Federation of Surgical Specialty Associations (FSSA). This approach
fails to identify and prioritize patients with the greatest need and
risk of deterioration. Quality of Life (QOL) metrics and augmented
intelligence (AI) can be used to achieve this. Our feasibility study
aims to assess whether QOL metrics and AI could be used to prioritize an
elective waiting list. Design An augmented intelligence software,
Copeland Clinical Artificial Intelligence (C2-AI, Cambridge England),
was used to analyze physiological data to calculate current and delayed
morbidity and mortality. These patients also completed online QOL
surveys. The combined scores were used to reprioritize the waiting list.
The correlation was assessed between the original FSSA and enhanced
waiting list. Differences in current and delayed mortality and morbidity
were also compared. Results 20 patients were analyzed by C2-AI. An
increase was seen in patient mortality (0.57% to 0.68%,
p<0.01) and morbidity if their procedure was delayed (7.6% to
8.8%, p<0.01). The greatest increases were seen in the risk
of pneumonia, wound infection, and hemorrhage. The enhanced waiting list
failed to correlate with the initial FSSA waiting list prioritization.
Conclusions Our study was able to demonstrate that AI and QOL metrics
can be used to prioritize patients on an elective operative list.
Furthermore, it might be possible to mitigate against complications in
high risk patients due to operative delay. Future studies will
implementation of the enhanced prioritization to asses whether it is
beneficial to patients and the healthcare system.