There are many factors that contribute to the increased length of stay and the high morbidity and mortality rates that are associated with acute ischemic stroke. Get With The Guidelines (GWTG)-Stroke, a stroke registry component of a continuous evidence-driven performance improvement program that includes a web-based patient management tool to provide clinical decision-making support, reporting, and patient education elements, was developed to track, measure, and improve the quality of care and outcomes for patients with acute stroke or transient ischemic attack (TIA) in the United States. Recently, GWTG-Stroke data from the first one million stroke and TIA patients demonstrated that quality of care improved significantly from 2003 to 2009 [Fonarow GC et al. Circulation Cardiovasc Qual Outcomes 2010]. Lee H. Schwamm, MD, Massachusetts General Hospital, Boston, MA, presented a subset of these findings, including only patients with acute ischemic stroke.

A total of 1392 hospitals participated in GWTG-Stroke, with 601,599 patients admitted for ischemic stroke. Of these 1392 hospitals, 301 were considered core hospitals, defined as participating since 2004. There were 287,477 ischemic stroke patients who were admitted to core hospitals. Analysis was performed on the subset of 601,599 patients from all hospitals and on the subset of 287,477 patients who were admitted to participating core hospitals. The median age was 73 years. Fifty-two percent was female, and 73% was Caucasian. Comorbidities included atrial fibrillation (19%), coronary artery disease/prior myocardial infarction (29%), diabetes mellitus (32%), hypertension (79%), and smoking (20%). The primary outcomes were percentage of patients with lengths of stay (LOS) >4 days, which was the median LOS for the entire cohort, percentage that was discharged to the home with or without services, and percentage of in-hospital deaths.

In the large cohort that included all participating hospitals, mortality and LOS decreased significantly between 2003 and 2009 (p<0.001 for both). There was an increasing trend in the rate of patient discharge to the home (p<0.001) as well (Table 1). However, no significant difference in NIH Stroke Scale was observed over time. The subset of core hospitals demonstrated similar results with regard to mortality, LOS, and discharge home (p<0.001 for all). After adjusting for patient characteristics, such as age, gender, comorbidities, hospital arrival mode, and time of presentation (on vs off hours), and hospital characteristics, such as region, number of beds, annual stroke volume, and teaching versus nonteaching hospital, the core hospitals still demonstrated improvement with regard to mortality (p=0.006), LOS (p<0.001), and discharge home (p=0.309) over the 6-year period.

Overall, hospitals that participated in GWTG-Stroke demonstrated substantial reductions in LOS and inhospital mortality in patients with acute ischemic stroke, even after adjusting for patient and hospital characteristics and accounting for hospitals that joined midstudy. Results from GWTG-Stroke are quite promising. However, it is unclear at this point if the improved outcomes that were observed in GWTG-Stroke are owing to improved care, increased adherence to inpatient care process measures, or unmeasured confounders, such as more frequent use of emergency medical services, expanded public education, improvements in in-hospital response times, guideline adherence, and progressive prevention methods, concluded Dr. Schwamm. Further study is needed to establish the origin of these improvements in stroke outcome.