Background:

In–hospital cardiac arrest (IHCA) outcomes vary widely between hospitals, even after adjusting for patient characteristics, suggesting variations in resuscitation practice as a potential etiology. However, little is known about the actual standard of resuscitation practice across the US.

Methods:

We conducted a nationally representative survey, targeting a stratified random sample of 1,000 hospitals from the American Hospitals Association database, which includes 3,809 US hospitals. Nine strata were chosen to ensure sufficient representation from hospitals of varying volume and teaching status. Internet searches identified each hospital’s Resuscitation Committee Chair or Chief Medical/Quality Officer, to whom the paper–based survey was addressed. Responses were double keyed to ensure accuracy.

Results:

Responses were received from 439 hospitals (Table) with the same distribution of volume and teaching status as the sample population (p = 0.50). Resuscitation committees were more common in both teaching and higher volume hospitals, and when present, were chaired most commonly by pulmonary/critical care (29%) or emergency medicine (26%) clinicians. Hospitalists chaired 9% of the committees. Hospitals were more likely to routinely review cardiac arrest data, if they had a resuscitation committee (78.0% vs 49.3%; P < 0.001) or dedicated staff time for resuscitation (78.8% vs 58.4%, P < 0.001). These results were independent of teaching status and volume, the latter of which (intermediate or high volume) was an independent predictor for both having dedicated staff time and tracking resuscitations. Eighty–nine percent of respondents reported that there was room for improvement in resuscitation practice at their institution and 77% reported at least one barrier to quality, of which lack of a resuscitation champion or inadequate training were the most common. Interestingly, the reported lack of a physician champion was significantly more pronounced in non–teaching hospitals (59% vs 42%, P = 0.02). In non–teaching hospitals, attendings were most likely to lead the resuscitation (83.7%), while physician trainees led the resuscitation in 88.1% of the major teaching hospitals (Figure).

Conclusions:

There is wide variability in resuscitation practice across US hospitals. Some of this variability is associated with hospital teaching status and volume. Further work is required to determine which practices correlate with improved patient outcomes.