NEW YORK NURSE: June 2009
by Christine Malmgreen, RN-BC, MS, MA, CHES, Nursing Research Coordinator, Hudson Valley Hospital Center, Cortland Manor, NY, and NYSNA Research Fellow for the Foundation of New York State Nurses, Central New York Center for Nursing Research Planning Committee.
Do rapid response teams (RRTs) improve patient outcomes? This article presents evidence suggesting the answer to that question needs re-examination.
Rapid response teams (RRT) or systems (RRS) to decrease inpatient deaths have been recommended (Institute for Health Care Improvement (2006); Joint Commission, 2007). Based on that recommendation, acute care facilities have implemented a rapid response protocol. A recent systematic review found only weak evidence to support unquestioned use of RRSs, the term “system” used by these authors to encompass the variety of models.
Systematic reviews are highly disciplined literature reviews. Researchers search for relevant studies on a topic, selecting only those that meet pre-determined inclusion criteria. The process is rigorously defined by protocol. The final report presents a balanced evidence summary, representing “state of the science” (Lo-Biondo-Wood, & Haber, 2006). A meta-analysis may be included that statistically combines data of multiple studies of the same design.
Winters, Cuong, Hunt, Guallar, Berenholtz, and Pronovost (2007) conducted a meta-analysis looking at eight research reports that evaluated the impact of rapid response systems (RRS) on hospital mortality, cardiac arrest and unanticipated admission to the Intensive Care Unit. A total of 10,228 abstracts were examined. It is not uncommon that systematic reviews—with strict exclusion criteria—will result in a very small sample culled from the abstracts found. Weak evidence was found by these researchers that rapid response systems are associated with a reduction in hospital mortality and cardiac arrest rates. Weaknesses in the quality of original studies limited ability to recommend embracing rapid response systems as a standard of care. Most notable was the variety of what were described in the literature as rapid response teams or systems, making comparison difficult. There is a need to standardize what constitutes an RRS. In addition, triggers that herald a need for a “rapid response” and outcomes that can be expected were questioned. Further study is warranted.
As this systematic review demonstrated that only weak evidence exists for a rapid response system, the question becomes, should all acute care settings be required to have one? Let’s take a closer look at the conclusions of these researchers. They did find a reduction in mortality and cardiac arrest. A full explanation of the limitations noted is beyond the scope of this article, however, large randomized controlled trials (RCTs) —the gold standard in biomedical research— were recommended to provide more precise estimates of effects and costs. It is premature, they asserted, to require adopting RRSs as a standard of care without clear evidence from RCTs. In addition, they postulated that cost effectiveness of RRSs in comparison to other interventions that may improve patient outcomes—not necessarily mortality—need to be evaluated. Examples included increased nurse staffing, use of hospitalists or intensivists, and automated monitoring systems. Another issue is the relative importance of RRS triggers. The researcher queried, “Is nursing staff ‘concern’ for a patient’s well-being more important [in detecting a deteriorating medical situation] than specific vital sign abnormalities?” A good question, but where are we left? The burden returns to hospital professionals to evaluate the risks and benefits of RRSs compared with other safety interventions—for example, decreasing patient to nurse ratios, or evaluating the impact of aggressive treatment that may ultimately not add quality to life —that could enable early recognition and appropriate treatment of patients with deteriorating clinical status. Published after this systematic review, a research study at a children’s hospital by Sharek and colleagues (2007, p. 2273) concluded that “implementation of an RRT was associated with a statistically significant reduction in hospital-wide mortality rate” and codes outside the ICU. The monthly code rate decreased by 71.2 %. Clearly, children represent a population for which early detection of deterioration would be highly beneficial. These researchers asserted that saving an estimated 33 children over 19 months of implementation of the RRT without significantly increasing costs was sufficient evidence to continue their RRT.
Take home message—evaluate your own RRS; monitor the criteria (“triggers”) and outcome measures important to your practice in evaluating that RRS. If your hospital is getting the outcomes desired, continue. If not, modify the approach using strategies with the best evidence available—and continue to monitor. On the other hand, you could change what you are monitoring. Winters and colleagues (2007) postulated that a “rapid response” may have “value added” beyond the outcomes currently measured.
Institute for Healthcare Improvement (2006). Rapid Response Teams: Reducing Codes and Raising Morale, Accessed from the Internet, April 14, 2009, http://www.ihi.org/IHI/Topics/CriticalCare/Improvement Stories/FSRapidResponseTeamsReducingCodesandRaisingMorale.htm.
Joint Commission (2007). Good practices for medical emergency teams. Oakbrook, Ill: The Joint Commission International (ISBN: 978-1-59940-000-6).
Lo-Biondo-Wood, G. & Haber, J. (2006). Nursing research: Methods and critical appraisal for evidence-based practice, 6th edition. St Louis: Mosby.
Sharek, P. J., Parast, L. M., Leong, K., Coombs, J. Earnest, K. Sullivan, J. Frankel, L.R. & Roth, S.J. (2007). Effect of a rapid response team on hospital-wide mortality and code rates outside the ICU in a children’s hospital. Journal of the American Medical Association 298 (19): 2267-2274.
Winters, B.D., Cuong, J., Hunt, E.A., Guallar, E., Berenholtz, S. & Pronovost, P.J. (2007). Rapid response systems: A systematic review. Critical Care Medicine 35 (5):1238-1243.
Thanks to the following nursing team at Hudson Valley Hospital Center who read the first draft and provided feedback: Andrew Askew, RN, Melanie Davis, GN, Norma Deraco, GN, Lori Mayo, GN and Barbara Reynolds, RN-CPN, MA.
A special thanks to my Nursing Research Fellowship mentor, Priscilla Sandford Worral, PhD, RN and Patricia Perry, PhD, RN, Research Specialist at the Center for Nursing Research, Veronica M. Driscoll Center for Nursing, for their suggestions and assistance.