This prospective cohort investigation collected data from the Cardiac Arrest Registry to Enhance Survival database from 10/1/05 to 12/31/09. Primary outcome was survival to discharge. Hospital characteristics were obtained via 2005 American Hospital Association Survey. A hospital's use of hypothermia was obtained via direct survey. To adjust for hospital- and patient-level variation, multilevel, hierarchical logistic regression was performed. Hospital volume was modeled as a categorical (OHCA/year 鈮?#xA0;10, 11-39, 鈮?0) variable. A stratified analysis evaluating those with ventricular fibrillation or pulseless ventricular tachycardia (VF/VT) was also performed.
The cohort included 4125 patients transported by EMS to 155 hospitals in 16 states. Overall survival to hospital discharge was 35%among those admitted to the hospital. Individual hospital rates of survival varied widely (0-100%). Unadjusted survival did not differ between the 3 hospital groups (36%for 鈮?0 OHCA/year, 35%for 11-39, and 36%for 鈮?0; p = 0.75). After multilevel adjustment, differences in survival across the groups were not statistically significant. Compared to patients at hospitals with 鈮?0 OHCA/year, adjusted OR for survival was 1.04 (CI95 0.83-1.28) among 11-39 annual volume and 0.97 (CI95 0.73-1.30) among the 鈮?0 volume hospitals. Among patients presenting with VF/VT, no difference in survival was identified between the hospital groups.
Survival varied substantially across hospitals. However, hospital OHCA volume was not associated with likelihood of survival. Additional efforts are required to determine what hospital characteristics might account for the variability observed in OHCA hospital outcomes.