Abstract 30: The Frailty Index: A Novel Predictor Of Stroke Outcomes

Background: There is little research that examines the impact of frailty on stroke outcomes. The objective of our study was to develop a frailty index (FI) and analyze its relationship to stroke outcomes in a nationally representative population-based study of ischemic stroke. Methods: Within a population of 1.3 million, all ischemic strokes among Greater Cincinnati/Northern Kentucky residents in 2005 were ascertained from local hospitals and clinics using ICD-9 discharge codes 430-436 (n=2226). Using previously validated methodology, a FI was created based on 35 age-related deficits including co-morbidities (e.g. prior stroke, dementia), pre-stroke functional status (e.g. as measured by the modified Rankin Scale [mRS], use of walking aids) and admission lab values (e.g. abnormal creatinine, hemoglobin). The FI for each subject (expressed as a %) was calculated as the number of deficits divided by the maximum number (n=35). Multiple logistic regression models were used to examine the association between the FI and in-hospital mortality, poor functional outcome (discharge mRS≥3), and discharge to home, with adjustment for age, sex, race, and NIH Stroke Scale score. Odds ratios were specified for a 3% increase in the FI (which is approximately equivalent to a 1 deficit increase in the FI i.e. 1/35) Results: The median age of the study sample was 73 years and the median FI was 17% (IQR 11%-26%). The in-hospital mortality was 10%. Amongst survivors, 53% were discharged with an mRS≥3 and 68% were discharged home. Logistic regression models demonstrated that after adjustment, for every 3% increase in the FI, the odds of dying in hospital increased by 16% (p<0.01), the odds of being discharged with an mRS≥ 3 increased by 8% (p<0.01) and the odds of being discharged home was decreased by 7% (p<0.01). ( Table ) Conclusions: Frailty was associated with increased in-hospital mortality, poorer outcome at discharge (mRS≥3) and decreased likelihood of being discharged home. Importantly frailty’s effects on stroke outcomes were shown to be independent of age, which suggest that frailty may represent a biological rather than chronological measure of aging. Further work needs to be done to examine the relationship between frailty, its sub-components and stroke outcomes. Frailty indices may represent an important new contribution to prognostic models of stroke.