Off-hour admission and in-hospital stroke case fatality in the get with the guidelines-stroke program

Abstract

Background and purpose: Previous reports have shown higher in-hospital mortality for patients with acute stroke who arrived on weekends compared with regular workdays. We analyzed the effect of presenting during off-hours, defined as weekends and weeknights (versus weekdays), on in-hospital mortality and on quality of care in the Get With The Guidelines (GWTG)-Stroke program.

Methods: We analyzed data from 187 669 acute ischemic stroke and 34 845 acute hemorrhagic stroke admissions who presented to the emergency departments of 857 hospitals that participated in the GWTG-Stroke program during the 4-year period 2003 to 2007. Off-hour presentation was defined as presentation anytime outside of 7:00 am to 6:00 pm on weekdays. Quality of care was measured using standard GWTG quality indicators covering acute, subacute, and discharge measures. The relationship between off-hour presentation and in-hospital case fatality was examined using generalized estimating equation logistic regression adjusting for demographics, risk factors, arrival mode, and hospital characteristics.

Results: Half of ischemic stroke admissions and 57% of hemorrhagic stroke admissions presented during off-hours. Among ischemic stroke admissions, the in-hospital case fatality rate was 5.8% for off-hour presentation compared with 5.2% for on-hour presentation (P<0.001). For hemorrhagic stroke admissions, in-hospital case fatality was 27.2% for off-hour presentation compared with 24.1% for on-hour presentation (P<0.001). After adjusting for patient-level and hospital-level factors, presentation during off-hours was significantly associated with higher in-hospital mortality for both ischemic stroke (adjusted OR, 1.09; 95% CI, 1.03 to 1.14) and hemorrhagic stroke admissions (adjusted OR, 1.19; 95% CI, 1.12 to 1.27). No differences were observed between off-hour presentation and any of the quality of care measures.

Conclusions: Off-hour presentation was associated with an increased risk of dying in-hospital, although the absolute effect was small for ischemic stroke admissions (0.6% difference; number needed to harm=166) and moderate for hemorrhagic stroke (3.1% difference; number needed to harm=32). Reducing the disparity in hospital-based outcomes for admissions that present during off-hours represents a potential target for quality improvement efforts, although evidence of differences in the quality of care by time of presentation was lacking.