Hospital Variation in Functional Recovery after Stroke

Janet Prvu Bettger, Laine Thomas, Li Liang, Ying Xian, Cheryl D. Bushnell, Jeffrey L. Saver, Gregg C. Fonarow, Eric D. Peterson

Research output: Contribution to journalArticlepeer-review

24 Scopus citations


Background - Functional status is a key patient-centric outcome, but there are little data on whether functional recovery post-stroke varies among hospitals. This study examined the distribution of functional status 3 months after stroke, determined whether these outcomes vary among hospitals, and identified hospital characteristics associated with better (or worse) functional outcomes. Methods and Results - Observational analysis of the AVAIL study (Adherence Evaluation After Ischemic Stroke-Longitudinal) included 2083 ischemic stroke patients enrolled from 82 US hospitals participating in Get With The Guidelines-Stroke and AVAIL. The primary outcome was dependence or death at 3 months (modified Rankin Scale [mRS] score of 3-6). Secondary outcomes included functional dependence (mRS score of 3-5), disabled (mRS score of 2-5), and mRS evaluated as a continuous score. By 3 months post-discharge, 36.5% of patients were functionally dependent or dead. Rates of dependence or death varied widely by discharging hospitals (range: 0%-67%). After risk adjustment, patients had lower rates of 3-month dependence or death when treated at teaching hospitals (odds ratio, 0.72; 95% confidence interval, 0.54-0.96) and certified primary stroke centers (odds ratio, 0.69; 95% confidence interval, 0.53-0.91). In contrast, a composite measure of hospital-level adherence to acute stroke care performance metrics, stroke volume, and bed size was not associated with downstream patient functional status. Findings were robust across mRS end points and sensitivity analyses. Conclusions - One third of acute ischemic stroke patients were functionally dependent or dead 3 months postacute stroke; functional recovery rates varied considerably among hospitals, supporting the need to better determine which care processes can maximize functional outcomes.

Original languageEnglish (US)
Article numbere002391
JournalCirculation: Cardiovascular Quality and Outcomes
Issue number1
StatePublished - Jan 1 2017
Externally publishedYes


  • health services research
  • risk adjustment
  • stroke
  • survivors

ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine


Dive into the research topics of 'Hospital Variation in Functional Recovery after Stroke'. Together they form a unique fingerprint.

Cite this