When Kendra K. Kay, ACP Member, joined Howard County General Hospital in Columbia, Md., as the medical director of its hospitalist program, she didn't realize that she and the rest of her hospitalist team were just what the staff developing a rapid response system were waiting for.
“They had identified the hospitalist as the logical physician team member,” Dr. Kay said.
Once she had the hospitalist program up and running, the hospital began to roll out the rapid response program, identifying a medical emergency team (MET) staffed by a hospitalist, a critical care nurse and a respiratory therapist. The idea was to detect subtle warning signs and intervene with at-risk patients hours before they would otherwise code, thereby improving outcomes and decreasing transfers to the intensive care unit (ICU).
“Initially I was a little bit concerned about how [the system] would affect hospitalists,” Dr. Kay said. But after a smooth transition, she began seeing payoffs not only in hospitalists' relationships with other doctors and nurses but also in patient outcomes. Rough numbers for the eight-month-old program are starting to show its success: about 10 MET calls made per month and a decrease in codes outside the ICU and codes per thousand discharges.
Howard County and other hospitals are pursuing rapid response systems in part in response to the Institute for Healthcare Improvement's (IHI) 100,000 Lives Campaign, which included a rapid response team initiative as one of six patient safety interventions (). Teams led by nurses are called rapid response teams, while teams led by physicians are called METs, according to terminology developed two years ago at the First International MET/Rapid Response System (RRS) Conference (metconference.com) . The IHI estimates that the more than 3,100 hospitals participating in the campaign, which ran from December 2004 to June 2006, avoided an estimated 122,000 unnecessary deaths through improved care, including some improvement associated with the initiative's best practices. All six of the original safety interventions are now part of the IHI's new 5 Million Lives Campaign, which was announced in December 2006.
Hospitalists and nurses alike are finding that the team approach pays off, leading some to characterize the movement as a step toward a rapid response system in which staff detect deteriorating patient conditions more quickly hospital-wide.
“Many [physicians] were trained to take on the burden of care ourselves. While laudable, I'm not sure it's the safest way of caring for patients,” said Michael A. DeVita, FACP, professor of critical care medicine and internal medicine at the University of Pittsburgh Medical Center Presbyterian Hospital, which formed its rapid response system in 1988.
Instead, multidisciplinary members of a rapid response system learn to rely on each other. Those who have experienced the process say it requires a well-thought-out plan and a strong emphasis on education and communication.
Boosting the trend
Many credit IHI with giving rapid response efforts a boost as well as the practical tools—Web-based materials, national conference calls and a mentoring system—to get a program under way.
Although the IHI has not credited any single intervention or the 100,000 Lives Campaign alone for the national decrease in mortality, “we believe the rapid response team has the potential to be a major source of mortality reduction, larger than the others,” said Joe McCannon, IHI's vice president and campaign manager. He noted that 60% of the 3,100 hospitals that signed up for the campaign participated in the rapid response initiative.
Hospitalists involved in METs may find themselves stretched a bit thinner since most of the rapid response efforts get few additional dollars.
“There are days when it's very busy and [being on the MET] stresses them, but then there's the satisfaction of knowing ... this is a very valuable service,” said Victor A. Morris, ACP Member, assistant chief of staff and director of the hospitalist team at Yale-New Haven Hospital in New Haven, Conn., which began its rapid response program in January 2006.
The early results
To date, much of the data from rapid response programs are very preliminary. According to those who spoke with ACP Hospitalist, team response time is within 5 minutes, often sooner; time spent on a call is between two minutes to an hour, with 20 to 40 minutes being typical; the number of calls per month varies; and cardiac arrest and mortality figures are going down, some say by half.
In addition, the anecdotes are compelling. In one case, a urology nurse watching an 80-year-old diabetic patient's condition worsen repeatedly voiced her concerns to the surgeon with no results. Twenty minutes after she activated the rapid response system, the patient had moved to the ICU; two days later he went home.
“My comment to her was, ‘You saved a life, because his chances of surviving an arrest are something close to zero,’ “ said Michael W. Leonard, MD, physician leader for patient safety at Kaiser Permanente and an IHI faculty member. He estimated that out of 100 calls, a third of the cases go to a higher level of care, a third are corrected and a third offer a chance for a teachable moment.
Still, hard data on the effectiveness of rapid response systems are slim. Because evidence-based medicine is the gold standard, some say there isn't enough proof that rapid response systems are better than any other intervention.
“People need to take a critical view to decide if [rapid response systems] are the best solution in their institution,” said Bradford D. Winters, MD, PhD, assistant professor in the department of anesthesiology and critical care medicine at Johns Hopkins University School of Medicine in Baltimore and coauthor of “Rapid Response Teams: Walk, Don't Run,” which appeared in the Oct. 4, 2006, issue of JAMA.
But systems interventions may never be controlled enough to reach standards used elsewhere in medicine, Dr. DeVita countered. “I believe it is impossible to conduct a randomized clinical trial of a whole hospital system,” he said. Instead, the before-and-after trial may be a better method.
Others criticize the IHI's method of counting lives saved because it doesn't break down the data by intervention or formally account for other factors. However, Andy Hackbarth, senior engineer at the IHI, said that the lives saved numbers were not intended to be considered evidence of the interventions' efficacy and that IHI campaign literature acknowledges the contributions of other initiatives in preventing unnecessary deaths.
Should rapid response systems be required despite the lack of evidence-based data? Dr. Winters said that without evidence, he's concerned that organizations have been considering making mandatory rapid response systems the standard of care. The Joint Commission and the American Association of Medical Colleges, for example, have been looking at rapid response systems.
But Dr. DeVita is comfortable with that. “I think the system should be required,” he said. “Critics should find another mechanism anywhere in medicine that shows similar benefit. To my knowledge there is none.”
Although ACP supported the 100,000 Lives Campaign, it has no official position on rapid response systems and is waiting to see evidence of their effect on outcomes such as death and disability before embracing them as a universal strategy, said Vincenza Snow, FACP, director of clinical programs and quality of care.
The lack of evidence-based data won't stop the rapid response program at Blessing Hospital in Quincy, Ill.
“If you can identify those patients early on and take care of their problems, change their code status ... or move to ICU and aggressively care for them, I can't help but see how outcomes can improve. In our rural hospital, it's worth it,” said Dorothy A. Bybee, RN, BSN, CCRN, director of critical care and the hospital's rapid response facilitator.
Some institutions have already signed up for the IHI's next initiative—those enrolled in the 100,000 Lives Campaign are automatically included in the IHI's 5 Million Lives Campaign—and look ahead to ways to enhance their programs.
As facilities get their rapid response systems into place, they will be able to use them to find system problems, said Kathy Duncan, RN, the IHI's faculty expert for the rapid response intervention. Another goal is to use technology to get one step ahead of the rapid response system. Virginia Mason Medical Center, for example, is beginning to automate its computerized medical record to trigger a call based on vital signs.
Ultimately, the responding team is only one part of a rapid response system that helps identify patients with critical needs, triggers a response team and includes oversight and quality improvement elements, said Dr. DeVita, who served as the course director of the First International MET/RRS Conference.
For now, programs are moving forward. Dr. Winters, the coauthor of the JAMA commentary, heads a program that rolled out at Johns Hopkins last October.
“It's hard to be evidence-based and yet not want to help your patients,” he said. “I'm very excited about our program.”