Rethinking how hospitals react when a patient's health deterioratesSeptember 21st, 2010 in Medicine & Health / Other
The growing use of rapid response teams dispatched by hospitals to evaluate patients whose conditions have suddenly deteriorated may be masking systemic problems in how hospitals care for their sickest patients, says a prominent Johns Hopkins patient safety expert.
In a commentary published in the Sept. 22 issue of the Journal of the American Medical Association, Peter Pronovost, M.D., Ph.D., a professor of anesthesiology and critical care medicine at the Johns Hopkins University School of Medicine, and co-author Eugene Litvak, Ph.D., president of the Institute for Healthcare Optimization, argue that the proliferation of these rapid response teams is just a quick fix to save patients who crash once they are moved out of intensive-care units. Instead, the authors argue, hospitals need to figure out why these patients aren't receiving adequate care and better manage beds to ensure patients are getting what they need whether inside or out of the ICU.
"We need to think about why patients are deteriorating in the first place and do something rather than waiting until they fall off the cliff," says Pronovost, who is also medical director for Johns Hopkins' Center for Innovation in Quality Patient Care. "There's often this decline before patients crash. It's typically not all of a sudden. Why doesn't healthcare prevent the crash or pick it up earlier?"
At the same time, he says, research on whether rapid response teams (RRTs) are effective is equivocal at best.
Patients deteriorate because they acutely get worse despite getting appropriate care or because they are receiving less care than they need, Pronovost says. The latter is thought to be the far larger group as RRTs typically end up bringing ICU-level care to a patient's bedside. The deeper question is why patients are not getting the level of care they need, he says.
Rapid response teams, he says, are really just a "reactionary solution and in many ways a workaround for a broken system and culture."
Rapid response teams are likely to be sent to attend to patients who have been released from the ICU to the floor, where they receive less intensive care. Often a patient is released from the ICU because his bed is needed for a sicker patient. Still, Pronovost says, the answer isn't necessarily adding more beds to the ICU. The ICUs aren't always overcrowded, though there are many times — typically during the middle of the day during the middle of the week — when there are more patients than beds. This forces hospitals to make difficult decisions, and many times that translates into someone being sent to an inadequate unit where they may be at a greater risk of deterioration or even death, he says. Better management of patient flow can go a long way toward improvement, he says, as can having surgeons spread operations out over the course of a week, rather than doing the majority of them early in the week.
The research on the usefulness of rapid response teams is equivocal at best and Pronovost worries that too many hospitals are using them as a crutch instead of addressing deeper problems. "Hope isn't a strategy," he says. "Science has to guide the process."
Meanwhile, the success of a rapid response team is determined through what Pronovost calls "perverse accounting." They are judged, he says, by counting the number of people saved or identified to be sent back to the ICU. "Imagine if we sent everyone from the ICU to the parking lot instead of to the floor," Pronovost says. "The rapid response teams would look like they're doing wonders because they would have to come in and save all of those patients. It's not a rapid response team issue. It's sending them to the proper level of care.
"It's a silly science where you take credit for your own bad decisions."
Provided by Johns Hopkins Medical Institutions
"Rethinking how hospitals react when a patient's health deteriorates." September 21st, 2010. http://phys.org/news/2010-09-rethinking-hospitals-react-patient-health.html