NCT04877535

Pilot Study for Postoperative Machine Learning

Completed NA Results posted Last updated 4 April 2025

What this trial tests

NA trial testing ML-based report card in Surgery--Complications in 222 participants. Completed in 11 May 2023.

Timeline

3 June 2021

Primary endpoint
11 May 2023

11 May 2023

Quick facts

Lead sponsor	Washington University School of Medicine
Phase	NA
Status	Completed
Study type	INTERVENTIONAL
Allocation	non randomized
Design	parallel
Masking	none
Primary purpose	other
Enrollment	222
Start date	3 June 2021
Primary completion	11 May 2023
Estimated completion	11 May 2023
Sites	1 location across United States

Drugs / interventions tested

• ML-based report card

Conditions studied

• Surgery--Complications — all drugs for Surgery--Complications →

Sponsor

Washington University School of Medicine

Who can join

18 and older, any sex, with Surgery--Complications. Patients with the condition only — healthy volunteers not accepted.

Results — posted to ClinicalTrials.gov

Per-arm endpoint measurements with 95% confidence intervals where reported. Source: trial results section.

Overall Handoff Effectiveness Primary · 8 hours postop

After handoff was completed, receiving nurses were asked: Globally, how effective was the handover 1. Not at all effective 2. Somewhat effective 3. Moderately effective 4. Very effective 5. Extremely effective The item is taken from PMID:25806398 but has no name

Group	Value	95% CI
Stage 1: Usual Care	3.9	± 0.4
Stage 2: Usual Care	3.9	± 0.7
Stage 2: Intervention	4.0	± 0.7

Number of Participants With ML Topics Discussed During Handoff Secondary · 8 hours postop

Binary. A research assistant observed the handoff and recorded if any topics identified by the ML algorithm (in the report card) were discussed included in the handoff

Group	Value	95% CI
Stage 1: Usual Care	0
Stage 2: Usual Care	19
Stage 2: Intervention	47

Number of Participants With Anticipatory Guidance During Handoff Secondary · 8 hours postop

Binary. A research assistant observed the handoff and recorded if expected problems or plans to address expected problems were conveyed, or if no expected problems or plans to address expected problems were conveyed.

Group	Value	95% CI
Stage 1: Usual Care	61
Stage 2: Usual Care	35
Stage 2: Intervention	85

Number of Handoff Receivers Agreeing That They Received All Needed Information Secondary · 8 hours postop

Receiving nurses were asked: Did you receive at handoff all the information you needed to safely take care of this patient? \[Yes, No\]

Group	Value	95% CI
Stage 1: Usual Care	72
Stage 2: Usual Care	37
Stage 2: Intervention	91

Number of Handoff Recipients Self-reporting Referring to Report Card OR Report Card Observed Directly Referred to During Handoff Secondary · 8 hours postop

Handoff was observed by a research assistant. In the intervention group only, each nurse was asked Did you look at or discuss the postoperative report card for this patient? \[Yes, No\] Additionally, the research assistant noted if they observed the report card being referred to be the handoff-giving team. \[Yes, No\] The measure is positive if either the self report or research-assistant recorded a "Yes"

Group	Value	95% CI
Stage 2: Intervention	5
Stage 2: Usual Care	0
Stage 1: Usual Care	0

Sponsor's own description

The objectives of the study are to determine the interpretability, workflow role, and effect on communications of showing report cards containing Machine Learning (ML)-based risk profiles based on pre- and intra-operative data to postoperative providers.

Publications & conference data

No peer-reviewed publications indexed yet for this trial. Completed trials usually publish results within 12-18 months.