Impact of Hospital Teaching Status on Out-of-Hospital Cardiac Arrest Outcomes and Resource Utilization in the United States: 2000-2012

Background:  Teaching status has often been studied as a potential systems-level variable associated with patient outcomes; however, limited information regarding the relationship between out-of-hospital cardiac arrest (OHCA) outcomes and teaching status exist.

Methods: Using the National Inpatient Sample database, we selected adults ≥18 years old admitted with a principle diagnosis of non-traumatic OHCA (ICD-9 CM codes 427.5 & 427.41) between 2000 and 2012. Patients were stratified to teaching (THs) or nonteaching hospitals (NTHs). The association of hospital teaching status with length of stay, total charges/case, survival to discharge, and survival with good neurologic outcome was assessed by multivariate logistic regression or generalized linear regression models, adjusting for patient and hospital-level characteristics.

Results: Of 186,483 admitted patients, 86,053 (46.15%) were treated at teaching hospitals during the study period. Overall OHCA incidence in the United States was 3.86 per 10,000 hospital admissions. Patients at teaching hospitals were more often younger, more likely to have VF, and less likely to be Caucasian [all p < 0.001] but with a similar comorbidity burden. Mean length of stay and total charges were higher in teaching hospitals (5 vs. 4 days, p<0.01 & $72,436 vs. $52,917, p<0.01). Risk-adjusted survival and survival with good neurologic outcome were significantly better in THs as compared to NTHs (odds ratio, 1.06; 95% CI 1.02 ̶ 1.14 & 1.19; 95% CI 1.09 ̶ 1.29 respectively). Despite a significant increase in OHCA survival at NTHs between 2000 and 2012 (36.8% to 46.6%, P_trend<0.01), OHCA survival rate at THs remained higher without any significant change in trend (P_trend=0.07) during the study period.

Conclusions:  For the last 12 years in the United States, survival to discharge and survival with good neurologic outcome in patients admitted with out-of-hospital cardiac arrest has been consistently higher in teaching hospitals as compared to non-teaching hospitals. While both have seen increases in the cost of caring for such patients over the last 12 years, this increased expenditure is far less and improvement in survival is far more evident in nonteaching hospitals than in teaching hospitals. Although reassuring and gratifying, these findings merit further explanation.