Rapid Response Team Did Not Reduce Hospital-Wide Codes or Mortality
Abstract & Commentary
By David J. Pierson, MD, Editor, Professor, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, Seattle, is Editor for Critical Care Alert.
Synopsis: With the largest cohort and longest follow-up yet reported, this prospective single-center study found that implementing a rapid response team reduced codes outside the ICU, but had no effect on either hospital-wide code rates or overall patient mortality.
Source: Chan PS, et al. Hospital-wide code rates and mortality before and after implementation of a rapid response team. JAMA 2008;300:2506-2513.
Chan and Aassociates performed a prospective before-and-after cohort study of the effects of implementing a rapid response team (RRT) in a 404-bed tertiary-care academic hospital in Kansas City, MO. They tracked cardiac arrests (codes) both in and out of the ICU as well as overall hospital mortality during two 20-month periods, January 2004 through August 2005 and January 2006 through August 2007. The intervention — implementation of a 3-member ICU-based RRT and an extensive educational effort for ward staff — took place from September through December 2005. Patients were thus enrolled in the study during the same seasonal time periods before and after the intervention. The primary outcome measures were hospital-wide code rates and mortality. The authors undertook extensive measures to adjust for pre-intervention trends and to look for potential confounders that might affect the study variables.
The study included 24,193 patient admissions in the pre-RRT period and 24,978 patient admissions in the post-RRT period. There were clinically small but statistically significant differences in the two populations: Patients admitted during the post-intervention study interval were slightly older, more likely to be male, and more likely to be African-American, and the case-mix estimate was slightly higher. Post-intervention there were 376 RRT activations, for altered mental status (27%), tachycardia (23%), tachypnea (13%), hypotension (12%), and other a priori-determined triggers. Forty-six percent of RRT episodes resulted in transfer to a higher level of care (ICU 41%, telemetry 4%, operating room, or other procedure 1%).
After RRT implementation, non-ICU codes decreased (adjusted odds ratio [AOR], 0.59; 95% confidence interval [CI], 0.40-0.89) relative to ICU codes (AOR 0.95; 95% CI, 0.64-1.43; P = 0.03 for interaction), but there was no change in the rate of hospital-wide codes (AOR 0.76; 95% CI, 0.57-1.01). Hospital-wide mortality did not differ between the pre- and post-intervention periods (3.22 vs 3.09 per 100 admissions; AOR 0.95; 95% CI, 0.81-1.11; P = 0.52). A careful search for response team undertreatment or underuse that might have affected the mortality findings revealed very few instances.
Commentary
This large single-institution study showed that RRT implementation was not associated with reductions in hospital-wide code rates or mortality, although it did document a reduction in codes outside the ICU. One might suggest that, had they included just a few more patients, the small observed differences would have reached statistical significance. However, using post-hoc power calculations the authors determined that, based on the differences observed, a pre-intervention and post-intervention population of 148,000 patients during each period would have been required to have 80% power to detect a 5% mortality reduction at the P = 0.05 level.
The authors discuss several possible reasons for their failure to demonstrate improved survival and reduced hospital-wide code rates after RRT implementation in their hospital. One is that the RRT tended to be called for patients who were not, in fact, about to code, and that those who did go on to cardiopulmonary arrest could not have been helped. Another is that patients triaged by the RRT to the ICU received heightened surveillance by unit staff, such that when they did arrest preparations had been made and a hospital-wide code did not have to be called. A third is that many patients initially treated by the RRT were subsequently made DNR (do-not-resuscitate) as a result of that interaction. As the authors note, it may be that the RRT episode catalyzed end-of-life care discussions in patients that might not otherwise have taken place. However, based on the observed code case-fatality rates prior to implementing the RRT, the authors calculated that as many as 59 more hospital-wide codes would have occurred if the 73 RRT patients who subsequently were made DNR had not been seen by the RRT.
A fourth possible reason for the RRT's failure to improve hospital-wide mortality might be a higher prevalence of DNR status during the second part of the study. Data were not available to the investigators on the numbers of patients designated DNR during the different study periods. Of course, it may also be that no reductions in hospital-wide code rates or mortality occurred after implementing the RRT because such teams do not in and of themselves actually affect these outcomes.
Following studies showing that many codes occur after hours of deterioration that could potentially have been detected and preventive actions taken, the Institute for Healthcare Improvement recommended that hospitals implement RRTs as one of the 6 strategies of the 100,000 Lives Campaign.1 This de facto mandate sent hospitals scrambling nationwide to create, implement, and track the results of RRTs, and a number of before-and-after studies have concluded that these teams both reduce the number of codes and save lives.2 However, while the premise for RRTs appeared sound, the evidence base substantiating their effectiveness has been called into question.3,4 Given the prevalence of medical error, practice variation, and poor adherence to evidence-based practice guidelines, there can be little doubt that better care results in better outcomes. However, this new study will hardly be reassuring to the advocates of RRTs as a means for improving overall quality of health care in hospitals.
References
- Berwick DM, et al. The 100,000 Lives Campaign: Setting a goal and a deadline for improving health care quality. JAMA 2006;295:324-327.
- Pierson DJ. Rapid response systems: Update and critique. Crit Care Alert 2008;15:77-79.
- Wachter RM, Pronovost PJ. The 100,000 Lives Campaign: A scientific and policy review. Jt Comm J Qual Patient Saf 2006;32:621-627.
- Winters BD, et al. Rapid response systems: A systematic review. Crit Care Med 2007;35:1238-1243.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content