Machine Learning Models for Predicting Unforeseen Hospital Admissions or Discharges After Anesthesia

Unexpected hospital admissions after ambulatory surgery not only bring discomfort to patients but also causes a decrease in the efficiency of the healthcare system. In addition, unanticipated patient's orientation carry the risk of unsuitable post operative orders. The hypothesis of this project is that artificial intelligence models will outperform traditional models in predicting which patients will require hospital admission after ambulatory surgery or unforeseen hospital discharge after surgery.

Study Overview

• Status: Completed
• Conditions: Pain, Postoperative; Anesthesia Complication; Surgery-Complications
• Intervention / Treatment: Other: Mathematical Prediction of unforseen patient reorientation

• Study Type: Observational
• Enrollment (Actual): 68683

Contacts and Locations

Study Locations

• Belgium
  • Mons, Belgium, 7000
    • Université de Mons

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

All hospitalized or outpatient patients who have undergone anesthesia for a diagnostic or therapeutic procedure, in a scheduled or emergency condition, in the institution's hospitals.

Description

Inclusion Criteria:

• Patient undergoing anesthesia for a therapeutic or diagnostic procedure

Exclusion Criteria:

• Incomplete informatic data
• Error in the encoding system

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Ambulatory Patients; Patient undergoing anesthesia in an ambulatory setting.	Other: Mathematical Prediction of unforseen patient reorientation; The goal of this project is to develop models to predict in the preoperative period which patients will require hospital admission after ambulatory surgery or unforeseen hospital discharge after surgery
Hospitalised Patients; Patient undergoing anesthesia in a hospitalisation setting.	Other: Mathematical Prediction of unforseen patient reorientation; The goal of this project is to develop models to predict in the preoperative period which patients will require hospital admission after ambulatory surgery or unforeseen hospital discharge after surgery

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Rate of patient reorientation	Rate of unforeseen hospital admission after an ambulatory surgery and rate of discharge after an hospitalised surgery	On the day of the operation

Collaborators and Investigators

• Sponsor: HUmani
• Investigators: Principal Investigator: Rémi Florquin, MD, Université de Mons, Belgium

Study record dates

Study Major Dates

• Study Start (Actual): January 1, 2020
• Primary Completion (Actual): June 30, 2024
• Study Completion (Actual): July 30, 2024

Study Registration Dates

• First Submitted: August 30, 2024
• First Submitted That Met QC Criteria: August 30, 2024
• First Posted (Actual): September 3, 2024

Study Record Updates

• Last Update Posted (Actual): October 18, 2024
• Last Update Submitted That Met QC Criteria: October 16, 2024
• Last Verified: October 1, 2024

More Information

Terms related to this study

• Keywords: machine learning; artificial intelligence; anesthesiology; ambulatory surgery; post operative complication
• Additional Relevant MeSH Terms: Pathologic Processes; Postoperative Complications; Pain; Neurologic Manifestations; Pain, Postoperative
• Other Study ID Numbers: HUmani_ODanesth

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: The investigators are not authorized to publish sensitive data by decision of the ethics committee

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No