Abstract Details

Title Mortality, Discharge Disposition, and Length-of-Stay in Patients with Heart Failure and Acute Ischemic Stroke: Machine Learning Clustering Analysis of 83,000 Patients

Topic Cerebrovascular Disease and Interventional Neurology

Presentation(s) P4 - Poster Session 4 (5:00 PM-6:00 PM)

Poster/Presentation
Number 13-008

Objective We aim to identify patient groups with unique comorbidity profiles in the 2015-2019 National Inpatient Sample (NIS), and assess the contributions of these profiles to mortality, discharge disposition, and length-of-stay in the sample of patients with heart failure and acute ischemic stroke (AIS).

Background Not Applicable

Design/Methods The 2015-2019 NIS was queried using ICD-10 PCS coding to identify patients with both heart failure and AIS. A machine learning clustering analysis evaluated the population based on 50 comorbidities, complications and clinical covariates. Optimal number of clusters was determined using the Davies-Bouldin Index and Calinski-Harabasz Index. Between-cluster multivariate logistic regression analysis was performed to assess risk of mortality and non-routine discharge. Kruskal-Wallis H-Testing was performed to assess variance in length-of-stay between clusters. Statistical analysis was performed using Python.

Results Machine learning analysis identified 83,280 patients and segmented them into four patient clusters, ranging from 3,379 to 56,881 patients. Mortality rate in Cluster 1 was the lowest at 8.14%. Clusters 2-4 demonstrated higher mortality rates of 9.38%, 31.08% and 37.59% respectively [OR Range 1.24-8.11, p<0.001]. Cluster comorbidity profiles are visualized in a heatmap. Cluster 1 displays the highest prevalence of arrhythmia and hyperlipidemia, while Cluster 4 has the highest prevalence of acute kidney failure, myocardial infarction, and sepsis. Compared to Cluster 1, risk of non-routine discharge was highest in Cluster 4 [OR 8.55, p<0.001]. Kruskall-Wallis H-testing and post-hoc pairwise comparison of length-of-stay distributions showed significant (p<0.001) differences between all clusters.

Conclusions This analysis distinguishes four groups with distinct comorbidity profiles. This allows for the assessment of different phenotypic comorbidity presentations of patients with varying mortality, discharge disposition, and length-of-stay, offering a foundation for further analysis to inform future clinical decision-making.

Authors/Disclosures
Ariel Sacknovitz, Medical Student PRESENTER	Mr. Sacknovitz has nothing to disclose.
James M. Beck	Mr. Beck has nothing to disclose.
Adam C. Kiss	Mr. Kiss has nothing to disclose.
Aryan Malhotra	Mr. Malhotra has nothing to disclose.
Mateusz Faltyn, CTO	Mr. Faltyn has nothing to disclose.
Fawaz Al-Mufti, MD (Westchester Medical Center at New York Medical College)	Dr. Al-Mufti has received personal compensation in the range of $0-$499 for serving as a Consultant for Stryker. Dr. Al-Mufti has received personal compensation in the range of $0-$499 for serving as a Consultant for Cerenovus. Dr. Al-Mufti has received personal compensation in the range of $0-$499 for serving on a Scientific Advisory or Data Safety Monitoring board for Revalesio .

��ɫ��

��ɫ��