Abstract
Stroke is a leading cause of mortality and disability worldwide. However, little is known about the predictors of hospital length of stay (LOS) among stroke patients in community hospital settings. This study aimed to identify predictors of hospital LOS among patients admitted for acute ischemic stroke at a community hospital in Southern California. A retrospective chart review was conducted for patients hospitalized with acute ischemic stroke at Arrowhead Regional Medical Center between January 2024 and July 2024. Univariable and multivariable logistic regression analyses were performed to determine predictors of hospital LOS. Receiver operating characteristic (ROC) and area under the curve (AUC) analyses were used to assess the reliability of the logistic regression model. A total of 101 patient records were reviewed, of which 51% had a hospital stay longer than six days. On univariable logistic regression analyses, a total Glasgow Coma Scale score ≤12, National Institutes of Health Stroke Scale (NIHSS) score ≥10, initial glucose level ≥180 mg/dL, hemoglobin level ≤12 g/dL, and admission to the ICU were associated with prolonged hospital stay. On multivariable logistic regression analysis, only NIHSS ≥10, glucose ≥180 mg/dL, and hemoglobin ≤12 g/dL remained significantly associated with prolonged hospital stay. ROC/AUC analysis demonstrated good model reliability, with an AUC of 0.871. Initial clinical evaluation findings of NIHSS ≥10, glucose ≥180 mg/dL, and hemoglobin ≤12 g/dL were significant predictors of prolonged hospital stay among patients with acute ischemic stroke.