Abstract WP366: Prediction of Inpatient Discharge Disposition: Role of the National Institute of Health Stroke Scale

Abstract

Introduction: Severity of stroke is expected to influence length of stay and discharge destination in patients. Ability to predict the likely discharge destination may assist family communication, and efficiently mobilize appropriate hospital resources, and shorten the length of stay. In this study we used NIHSS and ICH scores to predict discharge disposition after inpatient hospitalization. Methods: Data collected prospectively on patients admitted to a comprehensive stroke center was extracted and multiple variables including baseline mRS, daily NIHSS, discharge diagnosis, disposition (home, acute rehab, SNF, Hospice/Death) were analyzed. Results from 402 patient-records over 6 months in this on-going study are presented. Clinical improvement was defined based on change in daily recorded NIHSS. Regression analysis was performed using SPSS 24.0 and hypotheses tested α = 0.05. Results: Ischemic stroke was the most common diagnosis (73%) followed by TIA (15%) and ICH (11%). Average baseline NIHSS was 5.6. Over 3 days of NIHSS recordings, clinical improvement was noted in 174 (48%), no change in 103 (29%), and worsening in 83 (21%). Discharge to home or acute rehab was more frequent with ischemic stroke compared to ICH (70% vs. 42%, p<0.001). Adjusting for age, gender and baseline stroke type and severity, the Day 3 NIHSS score was a significant predictor (p<0.001) of discharge disposition. NIHSS score of > 15 was strongly associated with discharge to SNF (p=0.02), or hospice/death (p=0.005). In this data set, the number of ICH patients was not adequate to draw conclusions re: the ICH score. Conclusion: Commonly available inpatient scores obtained over 3 days can help predict discharge disposition. Use of continuous inpatient data may provide a tool to mobilize and prioritize hospital resources to reduce LOS and achieve predicted discharge disposition.