Detection of Ventilator-Associated Events Among Adults: Comparing a Naïve Bayes Classifier and Logistic Regression
DOI:
https://doi.org/10.46570/utjms-2024-1329Keywords:
ventilator-associated events, machine learning, naïve bayes classifier, logistic regression, predictive modelsAbstract
A Naïve Bayes classifier was used to identify mechanically ventilated patients at risk for ventilator-associated events (VAEs). VAEs, which can lead to ventilator-associated pneumonia (VAP), are considered preventable and have therefore been extensively studied. Data were gathered from May 2013 to August 2016 from hospital surveillance records and matched 1:1 with non-VAE cases using age, sex, unit while intubated, use of vasopressors, and total ventilator days. The dataset was split 7:3 to create samples for training and testing. From the training set, the classifier calculated the conditional probabilities for each factor in the model. After the proposed model was trained, it was validated against the testing data sample. The model demonstrated an overall accuracy of 83.0% (95% CI; 74.8% – 89.5%) and was able to correctly classify 80.4% of VAEs (sensitivity) and 85.7% of non-VAEs (specificity). The PPV and NPV were 81.4% and 84.9%, respectively. The AUC was 0.831 (95% CI; 0.751 – 0.911) and the MCC was 0.662. This study indicates that a Naïve Bayes classifier can be used to classify mechanically ventilated patients at risk for VAEs with moderate accuracy. The model could be used to screen for VAEs at the time of intubation and to strengthen manual surveillance. Further research is needed to validate the model in terms of prognostic and diagnostic use for hospitals and clinical settings with different patient characteristics.
References
Centers for Disease Control and Prevention (CDC). Device-associated Module: VAE. Atlanta (GA): CDC, 2023.
Zhang, R.K., Q. Xiao, S.L. Zhu, H.Y. Lin, and M. Tang, Using different machine learning models to classify patients into mild and severe cases of COVID-19 based on multivariate blood testing. J Med Virol, 2022;94(1): p. 357-365. doi:10.1002/jmv.27352.
Schober, P., and T.R. Vetter, Logistic Regression in Medical Research. Anesth Analg, 2021. 132(2):365-366. doi:10.1213/ANE.0000000000005247.
Chicco, D., N. Tötsch, and G. Jurman, The Matthews correlation coefficient (MCC) is more reliable than balanced accuracy, bookmaker informedness, and markedness in two-class confusion matrix evaluation. BioData Min, 2021. 14(1): p. 13. doi:10.1186/s13040-021-00244-z.
Giang, C., J. Calvert, K. Rahmani, et al., Predicting ventilator-associated pneumonia with machine learning. Medicine (Baltimore), 2021. 100(23): p. 26246. doi:10.1097/MD.0000000000026246.
Wu, Z., Y. Liu, J. Xu, J. Xie, S. Zhang, L. Huang, Y. Huang, Y. Yang, and H. Qiu, A Ventilator-associated Pneumonia Prediction Model in Patients With Acute Respiratory Distress Syndrome. Clin Infect Dis, 2020. 71(Suppl 4): p. 400-408. doi:10.1093/cid/ciaa1518.
Ramirez-Estrada, S., Y. Peña-Lopez, T. Vieceli, and J. Rello, Ventilator-associated events: From surveillance to optimizing management. J Intensive Med, 2022. 3(3): p. 204-211. doi:10.1016/j.jointm.2022.09.004.
Weinberger, J., N. Cocoros, and M. Klompas, Ventilator-associated events. Infectious Disease Clinics of North America, 2021. 35(4): p. 871-899. doi:10.1016/j.idc.2021.07.005.
Klompas, M., K. Kleinman, and M.V. Murphy, Descriptive epidemiology and attributable morbidity of ventilator-associated events. Infect Control Hosp Epidemiol, 2014. 35(5): p. 502-510. doi:10.1086/675834.
He, Q., W. Wang, S. Zhu, et al., The epidemiology and clinical outcomes of ventilator-associated events among 20,769 mechanically ventilated patients at intensive care units: an observational study. Crit Care, 2021. 25(1): p. 44. doi:10.1186/s13054-021-03484-x.
Keneally, R.J., T.J. Peterson, J.R. Benjamin, K. Hawkins, and D. Davison, Making Ventilator Associated Pneumonia Rate a Meaningful Quality Marker. J Intensive Care Med, 2021. 36(11): p. 1354-1360. doi:10.1177/0885066620952763.
Karatas, M., S. Saylan, U. Kostakoglu, and G. Yilmaz. An assessment of ventilator-associated pneumonias and risk factors identified in the Intensive Care Unit. Pak J Med Sci., 2016. 32(4): p. 817-822. doi:10.12669/pjms.324.10381.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 William Barnett, James Bassett, Chad Jaenke, Zachary Holtzapple, Kiran Boyinepally, Warren Back, Aadil Maqsood, Fadi Safi, Ragheb Assaly (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC BY 4.0) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
