Unsupervised machine learning analysis to identify patterns of ICU medication use for fluid overload prediction
Preprint
Abstract
INTRODUCTION Intravenous (IV) medications are a fundamental cause of fluid overload (FO) in the intensive care unit (ICU); however, the association between IV medication use (including volume), administration timing, and FO occurrence remains unclear. METHODS This retrospective cohort study included consecutive adults admitted to an ICU ≥72 hours with available fluid balance data. FO was defined as a positive fluid balance ≥7% of admission body weight within 72 hours of ICU admission. After reviewing medication administration record (MAR) data in three-hour periods, IV medication exposure was categorized into clusters using principal component analysis (PCA) and Restricted Boltzmann Machine (RBM). Medication regimens of patients with and without FO were compared within clusters to assess for temporal clusters associated with FO using the Wilcoxon rank sum test. Exploratory analyses of the medication cluster most associated with FO for medications frequently appearing and used in the first 24 hours was conducted. RESULTS FO occurred in 127/927 (13.7%) of the patients enrolled. Patients received a median (IQR) of 31 (13-65) discrete IV medication administrations over the 72-hour period. Across all 47,803 IV medication administrations, ten unique IV medication clusters were identified with 121-130 medications in each cluster. Among the ten clusters, cluster 7 had the greatest association with FO; the mean number of cluster 7 medications received was significantly greater in patients in the FO cohort compared to patients who did not experience FO (25.6 vs.10.9. p 5 separate 3-hour periods during the 72-hour study window. The most common cluster 7 medications included continuous infusions, antibiotics, and sedatives/analgesics. Addition of cluster 7 medications to a prediction model with APACHE II score and receipt of diuretics improved the ability for the model to predict fluid overload (AUROC 5.65, p =0.0004). CONCLUSIONS Using ML approaches, a unique IV medication cluster was strongly associated with FO. Incorporation of this cluster improved the ability to predict development of fluid overload in ICU patients compared with traditional prediction models. This method may be further developed into real-time clinical applications to improve early detection of adverse outcomes. Questions Can machine learning detect the presence of time-dependent medication administration patterns that are associated with risk of fluid overload in critically ill patients? Findings Using unsupervised machine learning, a unique IV medication cluster was identified that, when combined with the APACHE II score and diuretic use, improved the ability to predict fluid overload in ICU patients. Meaning These findings suggest that machine learning may be an important tool for analyzing IV mediation administration patterns to predict development of fluid overload. Such models may provide insight into areas where medication administration practices could be optimized to mitigate the risk of fluid overload in this patient population.Keywords
This publication has 71 references indexed in Scilit:
- Everything you need to know about deresuscitationIntensive Care Medicine, 2022
- Moving from bytes to bedside: a systematic review on the use of artificial intelligence in the intensive care unitIntensive Care Medicine, 2021
- Association between fluid overload and delirium/coma in mechanically ventilated patientsAcute Medicine & Surgery, 2020
- Liberal versus restrictive fluid therapy in critically ill patientsIntensive Care Medicine, 2019
- Medication regimen complexity is associated with pharmacist interventions and drug‐drug interactions: A use of the novel MRC‐ICU scoring toolJACCP: Journal of the American College of Clinical Pharmacy, 2019
- Fluid OverloadCritical Care Clinics, 2015
- World Medical Association Declaration of HelsinkiJAMA, 2013
- A Practical Guide to Training Restricted Boltzmann Machines2012
- Support-vector networksMachine Learning, 1995
- Principal component analysisChemometrics and Intelligent Laboratory Systems, 1987