Objective:
Wide variation exists in cardiac arrest survival. Historically cardiac arrest research has focused upon clinical pre-arrest and intra-arrest factors to explain this variation in outcomes. In-hospital post-arrest care is increasingly recognized as an important aspect of survival. We sought to identify hospital characteristics associated with improved cardiac arrest survival.
Methods:
We examined all participating hospitals in the University Hospital Consortium (UHC) clinical database with more than 25 adult cardiac arrests in 2012. Cases were identified using International Classification of Diseases, 9th Edition, code 427.5 (cardiac arrest) or 99.60 (CPR), excluding prisoners, pregnant patients, transfers, and hospice patients. We estimated hospital-specific risk-standardized survival rates (RSSRs) using hierarchical logistic regression, adjusting for individual risk of mortality. Institutions in the highest RSSR quartile were compared with those in the lowest three quartiles using Pearson chi-square tests of association.
Results:
UHC institutions admitted 3,686,296 patients in 2012, of which 33,700 patients experienced cardiac arrest. Overall survival was 42.3% (95% CI 41.8-42.9) with median RSSR of 42.7% (IQR 35.5-50.8). Hospitals in the highest quartile of RSSR had higher cardiac arrest volume (median 193 vs. 150, p-value 0.019), higher annual surgical operation volume (21,177 vs. 14,122, 0.007), cared for patients from catchment areas with higher household income ($60,753 vs. $56,424, 0.027), and were more likely to be a trauma (79% vs 59%, 0.024) or cardiac surgery center (91% vs 70%, 0.007). In addition, hospital size (477 vs 415 beds, 0.060) and teaching status (77% vs. 62%, 0.067) demonstrated a trend toward association with higher RSSR.
Conclusion:
Among hospitals in the UHC, those with higher cardiac arrest and surgical case volume, patient household income, and availability of trauma and cardiac surgery were associated with improved RSSR.
857-4 The effect of surgical case volume on outcome following the norwood procedure: An analysis of 29 hospitals
