Abstract 2275: Frailty Independently Predicts Poor Outcomes But Not Quality Of Care: An Analysis Of VA Ischemic Stroke Patients

Abstract

Background: Frailty is a geriatric syndrome characterized by progressive multi-system physiologic decline that is difficult to operationalize clinically. Frailty indices which represent an accumulation of age-related deficits are one method of quantifying frailty, but have not been applied to stroke patients. Our objectives were to create a frailty index (FI) and determine its independent contribution to mortality, discharge disposition and quality of care in ischemic stroke patients. Methods: Chart abstraction data were obtained on a sample of 3965 patients admitted with ischemic stroke to 131 VA hospitals in 2007. Data on 38 deficits were used to create a FI that included: co-morbidities (e.g. previous stroke, MI, diabetes), pre-stroke function (e.g. ambulatory status, functional independence measure), social history (e.g. alcohol, smoking, living alone) and first admission lab values (e.g. creatinine, glucose). The FI (expressed as a %) was calculated for each subject as the number of deficits divided by the maximum number (n=38). Multiple logistic regression models were used to determine the independent effect of FI on in-hospital mortality and discharge home, while adjusting for age, sex, and NIHSS. Quality of care was defined by a composite compliance score expressed as a percent of 14 pre-determined quality indicators; a multivariable linear regression model was used to determine the independent effect of FI on the composite score, after adjusting for age, sex, and NIHSS. Results: The mean FI was 15%. In-hospital mortality was 5% and 70% of patients were discharged home. Multivariable logistic regression results ( Table 1 ) showed that for every 1% increase in the FI, the odds of dying in hospital increased by 4% (p=0.0003) independent of age, sex and NIHSS. Similarly for every 1% increase in the FI, the odds of being discharged home was 3% lower (p <0.0001). Results of the multivariable model of quality of care showed that there was no significant relationship between frailty and the composite compliance score. (p=0.34) Conclusions: Higher FI scores were associated with increased in-hospital mortality and decreased likelihood of going home, independent of age, sex and NIHSS. However quality of care was not affected by frailty. Further work is required to define the relationships between frailty and stroke outcomes and quality of care. A FI based measure of frailty could provide added information to stroke prognostic models and assessments of quality of care processes.