Acute Care/Hospitalization

Regionalizing stroke care to high-volume hospitals may not improve stroke outcomes or lower costs for some regions

Redirecting patients who need complex cardiac and other surgical procedures to hospitals that perform a high volume of such procedures tends to improve patient outcomes due to the expertise of the high-volume hospitals. However, this regionalization of care may not significantly affect the outcome or costs of acute stroke patients in some regions.

Case Western Reserve University investigators, Mark E. Votruba, Ph.D., and Randall D. Cebul, M.D., retrospectively studied 12,150 Medicare patients admitted for acute stroke to 1 of 29 hospitals in Cleveland during a 7-year period. Overall, 15 percent of stroke patients died within 30 days.

For each 100-patient increase in hospital annual volume of stroke patients, risk-adjusted mortality declined 0.9 percentage points, with no significant difference in hospital costs. Yet, for each 1-mile increase in patient distance to the nearest hospital, mortality increased 0.6 percentage points.

This suggests that efforts to direct stroke patients to selected hospitals could have unintended consequences. Only 3 of the 29 hospitals (10 percent) treated more than 250 stroke patients a year. Using this volume threshold would have redirected over 81 percent of patients and increased patients' distance to the nearest stroke-treating hospital an average of 1.44 miles.

This would have resulted in a net reduction in mortality of 0.4 percent. Lower hospital volume thresholds would redirect fewer patients and have negligible effects on mortality. In addition, hospital volume explained only 17 percent of the variation in adjusted mortality rates across hospitals. Indeed, two of the best performing hospitals in the study were those treating the fewest stroke patients.

This analysis makes it difficult to attribute improvements in mortality to a hospital's high volume of stroke patients. However, the authors caution that their findings were based on comparatively old data (1991 to 1997) from one urban area. The study was supported in part by the Agency for Healthcare Research and Quality (HS09969).

See "Redirecting patients to improve stroke outcomes: Implications of a volume-based approach in one urban market," by Drs. Votruba and Cebul, in the December 2006 Medical Care 44(12), pp. 1129-1136.

Return to Contents
Proceed to Next Article