LAUSR

Studies of cardiac resuscitation initially alerted us to major disparities in survival between black and white Americans in the out-of-hospital cardiac arrest setting.1,2 The disadvantage of black race was striking, and multiple possible factors were cited, including lower rates of bystander cardiopulmonary resuscitation (CPR),3 the effect of neighborhoods, socioeconomic factors, comorbidities, and delayed arrival of emergency medical services personnel. It was equally disappointing when studies of inhospital arrest by the American Heart Association Get With the Guidelines (GWTG) registry recorded similarly lower survival in black individuals despite the fact that in-hospital identification of cardiac arrest, performance of CPR and defibrillation, and upgrading of in-patient care should not differ by race/ ethnicity. This initial study from the GWTG registry detailed strikingly lower survival to hospital discharge for black patients in general and, more specifically, within hospitals with a higher predominance of black patients.4 In this issue of JAMA Cardiology, Joseph et al5 present longitudinal GWTG registry data from 2000 to 2014 on inhospital cardiac arrest. The promising news is that not only has the risk-adjusted lower survival of black patients disappeared but also that overall survival for both white and black individuals has increased substantially and significantly, including in both predominantly black hospitals and predominantly white hospitals. The authors cite the registry efforts targeted at improving the quality of CPR and resuscitation performance as the likely mechanism for these survival improvements.5 Consistent with this hypothesis is the fact that the largest survival improvements were seen in the frequency of initial return of spontaneous circulation rather than improved survival owing to better treatment later in hospital care. The results are even more impressive for the survival improvement, as the frequency of shockable cardiac arrests deceased significantly during this period, with more asystole and pulseless electrical activity present at the start of resuscitation. The authors make a strong case for the notion that the hospitals that volunteered to participate in GWTG systematically improved their CPR performance, leading to a higher quality of care, frequency of return of spontaneous circulation, and survival to discharge. We would like to believe this is true, yet some caution in the interpretation of these data is warranted. The GWTG program is voluntary, involves some cost, and included only 289 of the nearly 5000 hospitals in the United States and is thus a biased population; GWTG is unlikely to accurately represent typical hospitals of the nation. Whether these hospitals represent the “best practice” in the country or not remains uncertain. Whether the gap in racial disparity in survival has diminished in the rest of the nation is therefore uncertain. A thoughtful reader may ask why we don’t have vital data like these for all hospitals in the country to promote similar improvement in survival across the entire nation. Indeed, a 2015 national report by the Institute of Medicine/National Academy of Medicine6 focused on this challenge, issuing as its number one recommendation the creation of a national cardiac arrest registry specifically to address this very important gap in our ability to protect the health of the public. Additionally, the current study failed to show robust improvement in the timing and quality of resuscitation performance over the same period and failed to show a relationship between the improved survival and the index of defect-free survival. It should be acknowledged that any effort to associate causal factors with outcomes of health care from even an elegant prospective registry is difficult. What other changes in health care were going on between 2000 and 2014 that might provide an alternative explanation for the improving survival rates from in-hospital cardiac arrest? During this 15-year period, there has been a dramatic rise of in-hospital palliative care units and the use of “do not resuscitate” or “do not intubate” advance directives.7 Could this movement have removed unfavorable patients from the denominator of those to be resuscitated and resulted in numerically elevated survival rates? In addition, there has been a growth of rapid response teams in US hospitals over the same period.8 Is a rapid response team more likely to be present when the cardiac arrest occurs and lead to faster treatments? Do rapid response teams prevent cardiac arrests from taking place in the first place or place more patients at risk for cardiac arrest in intensive care units where response to arrest may also be improved? Likewise, the validity of defect-free CPR as a parameter may be questioned. For example, the speed of initiation of CPR was a parameter in the registry, but how truly accurate is that assessment? Moreover, the validity of data on whether defibrillation really occurred within 2 minutes is questionable. These data are typically written down as a best estimate by a health care nursing staff sometime after, not during, a very chaotic cardiac arrest. These medical record elements are subject to profound recall bias, particularly because it is well known that there is a 10% loss of survival for every minute of delay in defibrillation of shockable arrests. Thus, in summary, we are optimistic that this report provides evidence that we have reduced the racial disparity gap within these GWTG hospitals. Despite the well-known limitations in generalizability of the GWTG data, it is good news that rates of survival from in-hospital cardiac arrest have become identical for black and white individuals while overall surAuthor Audio Interview