VAP risk index: Early prediction and hospital phenotyping of ventilator-associated pneumonia using machine learning

Ali Samadani; Taiyao Wang; Kees van Zon; Leo Anthony Celi

doi:10.1016/j.artmed.2023.102715

VAP risk index: Early prediction and hospital phenotyping of ventilator-associated pneumonia using machine learning

Artif Intell Med. 2023 Dec:146:102715. doi: 10.1016/j.artmed.2023.102715. Epub 2023 Nov 11.

Authors

Ali Samadani¹, Taiyao Wang², Kees van Zon², Leo Anthony Celi³

Affiliations

¹ Philips Research North America, Cambridge, MA, USA. Electronic address: a.a.samadani@gmail.com.
² Philips Research North America, Cambridge, MA, USA.
³ Massachusetts Institute of Technology, Laboratory for Computational Physiology, Cambridge, MA, USA; Beth Israel Deaconess Medical Center, Division of Pulmonary, Critical Care, and Sleep Medicine, Boston, MA, USA.

PMID: 38042602
DOI: 10.1016/j.artmed.2023.102715

Abstract

Background: Ventilator-associated pneumonia (VAP) is a leading cause of morbidity and mortality in intensive care units (ICUs). Early identification of patients at risk of VAP enables early intervention, which in turn improves patient outcomes. We developed a predictive model for individualized risk assessment utilizing machine learning to identify patients at risk of developing VAP.

Methods: The Philips eRI dataset, a multi-institution electronic medical record (EMR), was used for model development. For adult (≥18y) patients, we propose a set of criteria using indications of the start of a new antibiotic treatment temporally contiguous to a microbiological test to mark suspected infection events, of which those with a positive culture are labeled as presumed VAP if 1) the event occurs at least 48 h after intubation, and 2) there are no indications of community-acquired pneumonia (CAP) or other hospital-acquired infections (HAI) in the patient charts. The resulting VAP and no-VAP (control) cases were then used to build an ensemble of decision trees to predict the risk of VAP in the next 24 h using data on patients' demographics, vitals, labs, and ventilator settings.

Results: The resulting model predicts the development of VAP 24 h in advance with an AUC of 76 % and AUPRC of 75 %. Additionally, we group hospitals that are similar in healthcare processes into distinct clusters and characterize VAP prediction for the identified hospital clusters. We show inter-hospital (teaching status and healthcare processes) and cohort-specific (age groups, gender, early vs late VAP, ICU mortality status) differences in VAP prediction and associated symptomologies.

Conclusions: Our proposed VAP criteria use clinical actions to mark incidences of presumed VAP infection, which enables the development of models for early detection of these events. We curated a patient cohort using these criteria and used it to build a model for predicting impending VAP events prior to clinical suspicions. We present a clustering approach for tailoring the VAP prediction model for different hospital types based on their EMR data characteristics. The model provides an instantaneous risk score that allows early interventions and confirmatory diagnostic actions.

Keywords: Clinical suspicion of infection; Early prediction; Inter-hospital comparison; Machine learning; Ventilator-associated pneumonia.

MeSH terms

Adult
Anti-Bacterial Agents / therapeutic use
Cross Infection* / drug therapy
Hospitals
Humans
Intensive Care Units
Machine Learning
Pneumonia, Ventilator-Associated* / diagnosis
Pneumonia, Ventilator-Associated* / drug therapy
Pneumonia, Ventilator-Associated* / epidemiology

Substances

Anti-Bacterial Agents