Integrating Data, Algorithms and Clinical Reasoning for Surgical Risk Assessment

Postoperative Complications Clinical Trial

— IDEALIST  

Official title:

Integrating Data, Algorithms and Clinical Reasoning for Surgical Risk Assessment (IDEALIST)

Clinical Trial Details — Status: Active, not recruiting

Administrative data

• NCT number: NCT02741986
• Other study ID #: IRB201600262-N
• Secondary ID: R01GM110240
• Status: Active, not recruiting
• Start date: October 2016
• Est. completion date: November 2025

Study information

• Verified date: June 2024
• Source: University of Florida
• Contact: n/a
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

Brief Summary: The goal of this study is to implement and test an intelligent perioperative system (IPS) that in real-time predicts risk for postoperative complications using routine clinical data collected in electronic health records. The accuracy of computer-generated risk scores will be compared to physician's risk scores for the same patients. Physicians will be also asked to provide the opinion regarding the computer-generated risk scores using interactive interface with the program. The information regarding the risk scores performance will be collected during the two 6-month periods. The accuracy of IPS and physicians will be compared at the end at those two time periods.

Description:

Postoperative complications significantly increase morbidity, mortality and cost after surgery. In the current clinical practice the prediction of the risk for developing complications after surgery is manly based on physicians' clinical judgment. The predictive accuracy of that judgment is limited and poorly studied. The investigators will design an intelligent perioperative system (IPS) as the set of computer software and algorithms that in real-time predict risk for postoperative complications using routine clinical data in electronic health records. The system is designed as the self-learning system with the ability to interact with physicians and solicit their feedback. This study will compare the clinical judgment of physicians with computer generated risk scores for patients undergoing major surgery. All surgeons and anesthesiologists at large single-center tertiary academic center will be recruited to participate in this study. The IPS system will be implemented in real time and will generate risk scores for postoperative complications for patients planned to undergo surgery performed by the physicians enrolled in the study. Physicians will be asked to provide their risk scores (using visual analog risk scale from 0-100) for the same patients before and after interacting with the IPS. They will also have the opportunity to review computer-generated risk scores and provide their feedback. The information will be collected during two six-month periods. At the end of each 6-months period predicted risk estimates will be compared to the true occurrence of the complications. Predictive performance of physicians' risk scores will be compared to IPS generated risk scores using the comparison between area under the receiver-operating curve (AUC), sensitivity, specificity and positive and negative predicted values.

Recruitment information / eligibility

• Status: Active, not recruiting
• Enrollment: 200
• Est. completion date: November 2025
• Est. primary completion date: November 2025
• Accepts healthy volunteers: No
• Gender: All
• Age group: 18 Years to 100 Years

Eligibility

Inclusion criteria:

Surgeons and anesthesiologists working in adult inpatient operative practices.

Exclusion criteria:

Surgeons and anesthesiologists working in obstetric and pediatric practices.

Related Conditions & MeSH terms

• Postoperative Complications

Intervention

Other:
• Risk estimation
Physicians will be asked to provide their risk scores (ranging from 0 to 100) for postoperative complications prior to surgery and immediately after the surgery for the same patients that IPS is producing the scores. Physicians will provide scores both before and after reviewing the risk scores produced by the IPS. The IPS system will generate risk scores (ranging from 0 to 100) for postoperative complications prior to surgery and immediately after the surgery for patients taken care by the physicians enrolled in the study.

Locations

Country	Name	City	State
United States	UF Health	Gainesville	Florida
United States	UF Health Jacksonville	Jacksonville	Florida

Sponsors (2)

Lead Sponsor	Collaborator
University of Florida	National Institute of General Medical Sciences (NIGMS)

Country where clinical trial is conducted

United States, 

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Area under the receiver operating curve (AUC) of risk estimates	At the end of each 6-months period predicted risk estimates will be compared to the true occurrence of the complications. Predictive performance of physicians' will be compared to IPS by comparing the area under the receiver-operating curves (AUC) for risk estimates.	At the end of six months study period.
Secondary	Sensitivity and specificity of risk estimates	At the end of each 6-months period predicted risk estimates will be compared to the true occurrence of the complications. Predictive performance of physicians' will be compared to IPS by comparing the sensitivity and specificity for risk estimates.	At the end of six months study period.
Secondary	Positive and negative predictive values	At the end of each 6-months period predicted risk estimates will be compared to the true occurrence of the complications. Predictive performance of physicians' will be compared to IPS by comparing the Positive and negative predictive values for risk estimates.	At the end of six months study period.