Using Radiological Data to Estimate Ischemic Stroke Severity

Jason J. Sico, Michael S. Phipps, John Concato, Cynthia Brandt, Carolyn K. Wells, Albert C. Lo, Stephen E. Nadeau, Linda S. Williams, Mark Gorman, John L. Boice, Dawn M. Bravata

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


Background Risk-adjusted poststroke mortality has been proposed for use as a measure of stroke care quality. Although valid measures of stroke severity (e.g., the National Institutes of Health Stroke Scale [NIHSS]) are not typically available in administrative datasets, radiology reports are often available within electronic health records. We sought to examine whether admission head computed tomography data could be used to estimate stroke severity. Materials and Methods Using chart review data from a cohort of acute ischemic stroke patients (1998-2003), we developed a radiographic measure ([BIS]) of stroke severity in a two-third development set and assessed in a one-third validation set. The retrospective NIHSS was dichotomized as mild/moderate (<10) and severe (≥10). We compared the association of this radiographic score with NIHSS and in-hospital mortality at the patient level. Results Among 1348 stroke patients, 86.5% had abnormal findings on initial head computed tomography. The c-statistic for the BIS for modeling severe stroke (development,.581; validation,.579) and in-hospital mortality (development,.623; validation,.678) were generated. Conclusions Although the c-statistics were only moderate, the BIS provided significant risk stratification information with a 2-variable score. Until administrative data routinely includes a valid measure of stroke severity, radiographic data may provide information for use in risk adjustment.

Original languageEnglish (US)
Pages (from-to)792-798
Number of pages7
JournalJournal of Stroke and Cerebrovascular Diseases
Issue number4
StatePublished - Apr 1 2016


  • CT scan
  • Ischemic stroke
  • imaging

ASJC Scopus subject areas

  • Surgery
  • Rehabilitation
  • Clinical Neurology
  • Cardiology and Cardiovascular Medicine

Fingerprint Dive into the research topics of 'Using Radiological Data to Estimate Ischemic Stroke Severity'. Together they form a unique fingerprint.

Cite this