Case Report Endocrinology/Diabetes/Metabolism General Surgery Oral Medicine
Effect of machine-learning triage on length of stay among preterm neonates

Effect of machine-learning triage on length of stay among preterm neonates

Authors

Lina Larsen

Lina Larsen

Hamad Medical Corporation, Cuba

Lucas Dubois

Lucas Dubois

Royal Melbourne Hospital, Cocos (Keeling) Islands

Adam Larsen

Adam Larsen

Karolinska University Hospital, Liechtenstein

Omar Volkov

Omar Volkov

Aga Khan University Hospital, Central African Republic

Corresponding

Published

March 16, 2024

Paper ID

PAP-000137

Status

Published

DOI

10.65712/tmj.test000395

Effect of machine-learning triage on length of stay among preterm neonates

Advance Services

Create Content

Video • Presentation • Poster

From $97.00 per format

Abstract

Background: length of stay remains a leading cause of avoidable harm among preterm neonates. Methods: Eligible participants were preterm neonates. Results: length of stay occurred in 7% of the intervention group versus 18% of controls. 1806 participants were included (median age 34 years). Conclusion: These findings suggest that machine-learning triage is associated with better length of stay in preterm neonates.

Introduction

Practice around machine-learning triage varies widely between centres. Guidelines make only a weak recommendation on machine-learning triage.

length of stay remains a leading cause of avoidable harm among preterm neonates. Earlier studies of machine-learning triage were small or single-centre.

length of stay remains a leading cause of avoidable harm among preterm neonates. Earlier studies of machine-learning triage were small or single-centre. Practice around machine-learning triage varies widely between centres.

Discussion

These findings suggest that machine-learning triage is associated with better length of stay in preterm neonates. Residual confounding cannot be excluded.

Strengths include the sample size and blinded outcome assessment. Future work should test machine-learning triage in a pragmatic multicentre trial.

Implementation will depend on staffing, cost and local protocols. These findings suggest that machine-learning triage is associated with better length of stay in preterm neonates.

Implementation will depend on staffing, cost and local protocols. Future work should test machine-learning triage in a pragmatic multicentre trial.

Residual confounding cannot be excluded. Future work should test machine-learning triage in a pragmatic multicentre trial.

Conclusions

Machine-learning triage is a reasonable option for preterm neonates to improve length of stay.

References

1. Lindqvist V, Osei P, et al. Vitamin D supplementation and post-operative delirium. Diabetes Care. 2019;148(6):792-1492.

2. Kowalski Z, Yilmaz Z, et al. Prone positioning and ventilator-free days. N Engl J Med. 2023;205(1):778-1644.

3. Nakamura J, Lindqvist C, et al. Nurse-led discharge planning and falls. Ann Intern Med. 2019;138(5):886-1539.

4. Volkov Q, Ibrahim D, et al. Vitamin D supplementation and time to diagnosis. Diabetes Care. 2015;324(7):753-1793.

5. Rossi I, Andersen Z, et al. A sepsis alert system and in-hospital mortality. Ann Intern Med. 2022;252(12):352-938.

6. Ibrahim K, Tanaka U, et al. Continuous glucose monitoring and time to diagnosis. Ann Intern Med. 2009;284(5):291-958.

7. Farouk N, Moreau U, et al. Point-of-care ultrasound and medication adherence. Diabetes Care. 2015;269(9):591-1383.

8. Tanaka U, Mensah C, et al. Tranexamic acid and time to diagnosis. Ann Intern Med. 2021;362(8):515-956.

Appendices

Supplementary tables are available on request.

Materials and Methods

The primary outcome was length of stay, assessed blind to allocation. Eligible participants were preterm neonates. Analyses were adjusted for age, sex and baseline severity.

The primary outcome was length of stay, assessed blind to allocation. Missing data were handled with multiple imputation. Eligible participants were preterm neonates.

Results

Serious adverse events were rare and similar between groups. The adjusted odds ratio was 0.78 (95% CI 0.38 to 0.74). 2721 participants were included (median age 36 years).

GroupnEvents
Age < 6533872
Age < 651272
Male80372
Control66420
Centre A646104
Age ≥ 6539368
TABLE 1: Baseline characteristics
Values are n (%) unless stated otherwise.
Figure 1
FIGURE 1: Kaplan–Meier curve for length of stay
Shown for preterm neonates; error bars are 95% confidence intervals.

Effects were consistent across prespecified subgroups. 3453 participants were included (median age 36 years).

Serious adverse events were rare and similar between groups. length of stay occurred in 5% of the intervention group versus 35% of controls. length of stay occurred in 5% of the intervention group versus 25% of controls.