June 21st, 2010
How Teaching Hospitals Could Lead Medicine’s Metamorphosis
By Jonathan Rabinovitz — Stanford Medicine,
An irate doctor wants a blood transfusion for his patient, and he isn’t taking no for an answer.
Fielding the request on a cell phone in one of the intensive care units at Stanford Hospital & Clinics, a resident calmly explains to the more senior physician that the latest research — solid, evidence-based medicine — shows that transfusions in cases with mild anemia, such as this one, lead to increased risk of death.
In fact, there’s a new national protocol against such transfusions that SHC has just adopted. If the doctor tried to order the blood on SHC’s new computerized physician order entry system, it would inform him of the rule to check whether he was sure about the request. But it may take more than computer prompts to change years of habit.
The call takes place as Norman Rizk, MD, leads rounds of the intensive care units, and afterward he commends the resident for standing up for this protocol, even if the physician on the other end of the phone was frustrated. “I’m happy to talk with him about it,” adds Rizk, the medical director of the ICUs. “I would hope that such a practice doesn’t happen on my watch.”
The conversation is another chance for Rizk to drive home a lesson he has been teaching throughout the hospital, and it echoes a point he was talking about earlier in the rounds. “We are using too much blood,” he told the residents, citing a sobering statistic: SHC uses 200 percent more blood than one of its most revered peer institutions, Brigham and Women’s Hospital in Boston.
Limiting transfusions not only benefits patients’ health, Rizk explained, it also is cost-effective: Blood is an expensive and precious resource.
This is a prime example of what Rizk calls the “low-hanging fruit” of clinical effectiveness — practices that can save money while improving care. There are many such opportunities, and SHC is at the forefront of teaching hospitals’ efforts nationwide to capitalize on them. To fully realize these benefits, Stanford and other hospitals are changing the way they provide medical care, emphasizing what’s referred to in the health-care industry, somewhat simplistically, as “quality.” What that means is that hospitals are putting in place a plethora of protocols, based on scientific evidence from definitive studies, and constantly tracking how well these rules are being followed.
Of course, there have always been standards. Still, new information technology and new types of research — coupled with revelations about problems in the way care has been delivered — are transforming the way hospitals practice medicine, making measurement of “quality” central to their missions. For teaching hospitals, that involves a particularly delicate balance between their role as centers for innovative, outside-the-box thinking with this new demand to be more consistent. As a result, they are fostering a different culture — with uniform rules embedded in computerized systems, and management practices that focus attention on efficiency — in place of the ivory tower culture that in another era deferred to whatever the physician-professor deemed best.
Steven Wartman, MD, PhD, president and CEO of the Association of Academic Health Centers, acknowledges the “trend toward more ‘corporate’ management” in a recent editorial, while cautioning that “the new ‘corporate’ paradigm not overshadow the fundamental academic ethos and the creativity, intellectual spirit and unique public standing of these institutions.” This tension is not new, but there is no longer a debate about whether the corporate paradigm is the way to go — the question is how quickly can it be done.
The push for quality can be traced to a 2001 report issued by the Institute of Medicine, Crossing the Quality Chasm: A New Health System for the 21st Century, which drew attention to the gap between what is known to be good, evidence-based health care and the health care that people receive. That gap not only costs lives — two infections that often result from medical missteps, sepsis and pneumonia, kill up to 48,000 Americans per year — but also money for treatment.
Stanford and the other elite teaching hospitals add another wrinkle to the equation. “We can all too easily take for granted when we receive or deliver care that it is of the highest quality, the most cutting-edge, the most sophisticated available, but is this really true?” says Philip Pizzo, MD, dean of the Stanford University School of Medicine. “The reality is that when a bright light is shined on our personal practice, inadequacies and imperfections are likely to be seen.
“This is especially true when we think we are better than we are: After all, if we are members of the Stanford community, we must be on top of the quality pyramid,” he says. “I would argue that once we assume that, we become especially vulnerable to errors and mistakes.”
While safety concerns launched the quality movement, what has propelled it further are reports of doctors driving up costs by overusing expensive technology and costly procedures, particularly in hospitals. Virtually everyone in health policy circles agrees that the increases in health-care spending — it now accounts for 17 percent of our gross domestic product — cannot continue, and that the biggest increases have occurred in connection with complex hospital cases. Arguably the most popular solution is to promote evidence-based standards for care.
The new health-reform law adds to the economic pressure on teaching hospitals. It lowers hospital payments under the Medicare program, and it allocates the existing funds in new ways that reward hospitals for higher-quality treatment and greater efficiencies — and penalize those who perform poorly. To be sure, the law expands the number of people who will have insurance, but there won’t be enough money or doctors to care for all of them if current habits don’t change. There is yet another reason that hospitals are re-engineering how they provide inpatient care: A shift has begun toward reimbursing less for inpatient care and more for outpatient care that keeps people out of hospitals.
So as Rizk, the senior associate dean for clinical affairs at Stanford medical school, leads his trainees on rounds, they provide a glimpse of the future. Unlike a few years ago, they now have three computers on wheels in tow as they stop outside each patient’s room to consult a comprehensive electronic health record; no longer do they need to leaf through a Manhattan-phone-book-sized case file looking — often unsuccessfully — for a test result from a few years ago or an X-ray that was taken a few hours earlier. As appropriate to each case, Rizk gives concise, information-packed lectures on medicine as he has done since joining the Stanford faculty in 1991, but there are some new topics. He talks about the systems for improving efficiency and quality underlying each patient’s care, and he talks about money.
The “big bang” was scheduled to take place at the stroke of midnight on April 25, 2008. At least that’s what some at SHC were calling it, as they waited for the moment that a new electronic health record system would “go live,” culminating two years of intensive planning — rethinking workflows and providing hospital-wide trainings — by more than 300 physicians, nurses and administrative staff members. It was the first stage of an information technology makeover with an estimated cost of some $200 million.
Martha Marsh, the hospital’s president and CEO, says she was as nervous as she had ever been in her entire career as dozens of disparate information technology systems throughout the hospital — many that were unable talk to each other — were replaced with one system.
The new EHR would store patients’ information electronically and uniformly so that all authorized staff could access it; it would also allow them to place orders. The goal was to let physicians, nurses, pharmacists, lab technicians and others call up a patient’s records, view images and test results, and enter instructions, from anywhere at any time. The system would be smart enough to detect possible dosing errors with medication or to call a physician’s attention to a measure they might want to take, given the circumstances of a particular case.
What’s more, the system would eventually make it possible for an automated process of skimming through the records and crunching data from every patient, measuring in real time how the hospital is performing on metrics, then flagging any potential problems. In the future, it could even drill down and determine which units and which clinicians are not following the advised practices. Ultimately, it could be used to compile databases that could provide the fodder for studies of the effectiveness of different treatments, looking not only at patient outcomes but also at costs.
Marsh had good reason to be nervous about the hospital becoming one of the first to adopt such a comprehensive EHR system. Even with all the precautions, if the hospital hit a snag, clinicians might be without vital computer support for precious hours. She had no doubt, though, that it was worth the risk. “Having a great clinical information system is core to the future success of a place like this,” says Marsh. “It won’t make you successful on its own, but you can’t measure the data needed to implement clinical protocols and guidelines without it.
“There are a million quality steps that the system can enable you to do, which you would be hard-pressed to do in its absence,” she adds.
The EHR and other new information technologies are the linchpin of the effort under way at Stanford and other hospitals to — first and foremost — make objective improvements in quality, then capitalize on instances that also reduce costs. The idea is to establish a set of best practices and embed them in the hospital’s information systems, which will encourage their use — a mix that is hoped to be strong tonic for the nations’ hospitals, which have been rife with medical errors and inefficiencies. President Barack Obama and Congress have allocated $20 billion to help health-care providers implement this technology. And the new health-care law adds to the need for such systems by increasing the demands on hospitals and other providers to produce reams of data about their performance.
Marsh understood that the EHR would not succeed unless the hospital organization was streamlined to produce and respond to the new information systems. Under her leadership, the hospital has applied the principles of Six Sigma, (Motorola’s statistics-based strategy to improve its workflows) and lean manufacturing (Toyota’s model for efficiently producing quality cars) to its own operations and services.
Soon after becoming the hospital’s CEO in 2002, before the EHR project was launched, Marsh began to bring in new sets of management skills, people who understood management-science principles and could apply them to a health-care institution. She needed, for instance, to create a position that would oversee quality improvement and medical information, and recruited Kevin Tabb, MD, president of clinical data services for GE Healthcare Information Technologies, for the job. She needed expertise in process redesign, and brought in first Sridhar Seshadri, PhD, also from General Electric, for a new vice president’s post, then added chief operating officer Dan Ginsburg, a senior vice president at Massachusetts General Hospital and a Harvard Business School graduate who had led efforts to redesign care in the network that included MGH. In his previous job, Ginbsurg had, to give one small example, reorganized how patients were seen at a proton beam facility, increasing the number of visits to 37 each day from 29 — without adding staff or hours of operation. “It often sounds totally obvious,” Ginsburg told the Boston Globe, “but until you do it, things won’t change.” Those sorts of redesigns were launched throughout Stanford Hospital before the initial launch of the new clinical information system and continue today.
On that night in April two years ago, the first stage of the EHR transition went relatively smoothly. There were (and continue to be) headaches — doctors and nurses routinely complain, for instance, about trouble finding things within a patient’s EHR — but right out of the gates, some tasks, such as ordering CT scans, became easier. Doctors can also access the system from home to follow up on patients; one even logged in from Florida on that first day. Over the next two years, the EHR was rolled out in the outpatient clinics and expanded to include billing and other business functions. In January, SHC became one of seven health-care institutions to be recognized for having achieved the highest level of clinical information integration by HIMSS Analytics, a wholly owned, nonprofit subsidiary of the Healthcare Information and Management Systems Society.
“Everybody is going to have to do this,” says Tabb, now SHC’s chief medical officer. “We’re very proud because we did it quickly, we did it well and we’re using it well, but it’s still very new.” And he is quick to point out that the hospital is still working on how to weave the technology into everyday practice and how to keep it up to date with the latest changes in medicine. Indeed, the true benefits of SHC’s system have yet to be realized.
Rizk, for instance, wants to be able to see whether the many protocols being adopted in the ICUs are affecting the mortality rate. That should soon be possible, as the system is now running a program that evaluates the expected mortality of every patient admitted to the ICUs, using a national database of ICU patients. It will then report how the results at Stanford compare with those prognoses. Similarly, Rizk wants to track how several of the most expensive and commonly prescribed drugs in the ICUs are being used in the hope of being able to find a less expensive alternative.
“The truth is all that we’ve done is put in place a set of tools,” says Rizk. “We can’t make the changes we need to make without these tools, but we need to learn how to use them.”
For the last half of the 20th century, teaching hospitals flourished as a result of what some term a “virtuous cycle.” The fundamental premise was that they provide the latest, most advanced care not available at their non-academic counterparts, while also carrying out two other missions: They train doctors and conduct research. When things were going smoothly, the academic and the clinical efforts complemented each other: The scientific advances and educational enterprise brought new and greater expertise to the clinical care, and, in turn, the clinical care provided opportunities for study and training.
But the funding that once fueled this juggernaut — government monies for research, education and advanced inpatient care — is less available, and academic health centers are looking to streamline the engine using the tools of EHR, quality improvement and other elements that underlie the new health reform law. Stanford and other teaching hospitals remain committed to giving the most advanced care, but they must find a way to thrive in tighter fiscal times. They need to use their academic strengths to become more efficient.
To understand the situation, take a look at Stanford Hospital, the flagship of one of the nation’s leading academic health centers. It is renowned for taking in the complex cases, known as tertiary and quaternary care, that community hospitals are not equipped to handle: It provides heart, liver, bone marrow and other organ transplants; the latest surgical and drug therapies for stroke; new minimally invasive ways to repair heart damage; and experimental treatments for cancer, among other things. Stanford and other teaching hospitals also disproportionately provide vital emergency services and shoulder much of the care of the neediest. While the major teaching hospitals affiliated with the American Association of Medical Colleges constitute 6 percent of the nation’s hospitals, they account for 41 percent of all charity care, 60 percent of level-1 trauma centers and 50 percent of all transplant services.
As a result of these services and their distinct missions, teaching hospitals tend to be more expensive than their non-teaching brethren. “The cost is by definition higher,” explains Pizzo. “For a variety of reasons, there are more people involved in the provision of care — trainees, students, residents, fellows. All that adds to the cost and, in some ways, decreases the efficiency of the operation.”
Those higher costs weren’t necessarily a problem in the recent past. Revenue from the clinical operation subsidized training and research, but the balance is now in jeopardy. The growing complexity of care has led to growing numbers of trainees — for instance, the Stanford University Medical Center (which includes other units in addition to the hospital) went from 607 residents in 2001 to 900 residents in 2009 — and many of those new positions aren’t covered by federal graduate medical education funds. SHC last year had to devote about $30 million from clinical revenues to cover the cost of training residents. At the same time, the amount spent on medical research by the federal government, when adjusted for inflation, declined during the administration of President George W. Bush, and it’s doubtful, given the poor economy, that the new administration will bring back the boom in funding that characterized previous decades. “This has made the dependency on support from clinical income, generated largely at teaching hospitals, more critical at many academic medical centers,” says Pizzo. “It’s clear that business as usual — and certainly one based simply on growth — is not sustainable.”
One way that teaching hospitals are likely to adapt is by putting more emphasis on research into what’s known as “comparative effectiveness.” It involves doing studies that compare drugs, medical devices, tests, surgeries or ways to deliver health care and then using the evidence from that research to guide care provided by physicians and hospitals. The goal is to provide the basis for new standards and protocols that will not only improve quality, but also cost-effectiveness and efficiency.
Although you might think that the treatment you receive has been carefully evaluated, all too often it has not. “When you shop for a new car, phone or camera, you have lots of information about your choices,” explains the U.S. Agency for Healthcare Research and Quality on its website. “But when it comes to choosing the right medicine or the best health-care treatment, clear and dependable information can be very hard to find.”
There are new funds for this type of research. Congress allocated $1 billion last year for such studies, and the health-care reform law takes it one step further, establishing a permanent trust fund that guarantees a steady stream of support for this type of research. At Stanford, in addition to medical school faculty already doing comparative-effectiveness studies, the medical school in collaboration with Stanford Hospital & Clinics is establishing a new Center for Clinical Excellence Research, under the direction of one of the nation’s leading quality-improvement experts, Arnold Milstein, MD, to boost its efforts. Other academic health centers are ramping up their work in this area as well.
In the meantime, as SHC transforms its model of care, Rizk’s rounds are one of the spots where the rubber meets the road. Like many other Stanford clinical leaders, he has changed his teaching so that he weaves information about cost of care and the need for standardized systems into his discussion of almost every patient on his rounds. And he wants to make his students particularly aware of the guidelines for caring for patients at the end of their lives, which, by one estimate, accounts for $1 of every $3 of spending in the Medicare program.
Indeed, the critical care curriculum for residents lists among its goals and objectives that “residents should be capable of deciding about the utility and appropriateness of this advanced but invasive and expensive care in the overall treatment of patients.” This effort to teach about cost is accompanied by new lessons on end-of-life issues and palliative care, the principles and techniques of managing the ICU, ethical and legal aspects of critical care, and developing and implementing quality improvement through data analysis and careful observation of the ICU’s workings. Such changes are in keeping with guidelines from the Accreditation Council for Graduate Medical Education, which now requires that considerations of cost awareness, risk-benefit analysis and quality improvement be integrated into training programs.
So at each patient’s door, Rizk makes sure that the young doctors follow a set routine and ask the nurse for that room to join the meeting, a system set in place to guarantee that nurses are updated on their cases’ status and that doctors ask nurses about what they’ve observed. (From 2002 through 2006, about 60 percent of “sentinel events” — medical deaths and major injuries of unknown cause — stemmed from communication failures, ranging from doctor-nurse miscues to misinterpreted written instructions, according to a report from the Joint Commission, a nonprofit group that accredits health-care organizations.)
When one of his residents waits until the last moment to put antiseptic gel on her hands before seeing a patient, Rizk gently warns that had she forgotten, he would have handed her a red card, with boldface type saying, “CLEAN HANDS SAVE LIVES,” a practice that is being tracked, measured and reported in many of the biweekly e-mail reports on various quality metrics that are sent to select hospital leaders. (Research from Johns Hopkins University shows that ICUs that do hand-cleaning 100 percent of time bring their infection rate near zero.)
And while Congress dropped the political hot potato of how to handle end-of-life care, Rizk reviews with his residents the steps established under the hospital’s palliative care program to manage the cases in which intensive measures are being taken to stave off death, with little benefit to quality of life. (One of four hospitals now has a palliative care program, and the number has more than doubled since 1999, in response to studies showing that medical care for patients with advanced illness is characterized by inadequately treated physical distress and poor communication between caregivers and patients and their families.)
During rounds, Rizk comes to an elderly patient whose kidneys are failing, whose lungs are falling apart and who is overwhelmed with a host of other problems. Rizk and the residents have tried a panoply of drugs and treatments over the last few days, but her condition has worsened. One resident advocates continuing with such aggressive treatment to prolong the patient’s life.
“Does she know that she is likely to die in two weeks?” Rizk asks quietly. In fact, the patient is now in a coma. To continue dialysis and other highly intrusive and complex measures, Rizk says, is going to add to the patient’s suffering without, as studies show, prolonging her life or improving its quality. It also would be expensive.
There are now systems in place to guide how to proceed, and Rizk refers to them. There will be a meeting of ethicists, caseworkers, nurses, doctors to evaluate the patient’s options, a practice the hospital instituted a few years ago with its launch of a palliative care program.
Rizk explains to the trainees that he wants to be able to work with a family member who understands the choices, and he asks about the relative who was just at the bedside. “Do you know who the surrogate is in this case?” he quizzes them, noting that the answer has very practical consequences. “In the state of California, the doctor gets to pick who the surrogate will be,” he says. “That’s why I’m asking if this relative is going to be able to handle making a decision.” This is not the sort of detail that would’ve been discussed a decade ago.
Rizk requests that one of the residents set up a meeting with the family. It won’t be an easy conversation to have. In many cases, for instance, patients or their families are demanding unnecessary tests and measures, and if Medicare or a physician suggests otherwise — even citing evidence of their ineffectiveness — the initial response could well be an angry one.
But those attitudes can be changed. At least that is what Rizk and other medical practitioners are hoping. While these decisions have a huge effect on cost, the discussions are not about money. “That’s not relevant,” says Rizk. “We’re trying to establish what is best for the patient,” he says, explaining that people often make the mistake of thinking that the best treatment is the most expensive. “If they know you care,” he says, “they’ll listen to what you have to say.
At hospitals these days, “quality” has become a mantra, but, of course, hospitals have always been trying to provide quality. Indeed, when Rizk was appointed head of the ICUs in 1997, then-chief of staff Lawrence Shuer, MD, said Rizk’s mission was to “foster quality care at a time when we have had to be creative in reducing costs.”
Now in health-care circles, “quality” means standardization based on evidence-based protocols, and in the last few years, the number of protocols has multiplied. There are guidelines for running ventilators and modulating blood sugars, and there are control measures for infections, for sedation, for nutrition and for care of patients who suffer cardiac arrest. The hospital board of trustees now has a committee specifically to review how well quality measures are being followed. Rizk chairs another twice-monthly meeting for the ICUs to consider new protocols and to assess how they’re working. In fact, there is now a backlog of protocols waiting to be posted on the hospital’s intranet, he says, noting that while it would be good to have them up more quickly, it is also a sign of the tremendous appetite for enacting evidence-based medicine.
Equally important is the transparency that hospitals are practicing in conjunction with these protocols. Hospitals’ performances on certain practices are being regularly tracked and shared not just within the hospital, but also with the public. There are already a number of websites that post this information and allow a visitor to compare hospitals with each other. And the new health-care law is going to make it possible to compare physicians and look at hospital costs. “Whether we do it or not, someone is going to be measuring our quality,” says Marsh. “That’s the new world we’re in: Public accountability has become really important.”
This transparency appears to be driving hospitals to change.
The Joint Commission hospital accreditation group reported in January that hospital performance nationwide improved continuously between 2002 and 2008 on 12 quality measures, which reflect the best evidence-based treatments for heart attack, heart failure and pneumonia — practices demonstrated by scientific evidence to lead to the best outcomes. The magnitude of national improvement on these measures ranged from 4.9 percent to 58.8 percent.
At Stanford specifically, the hospital last year scored above the norm on most measures. For instance, 100 percent of its patients appropriately received post-operative medication to prevent blood clots as compared with 88 percent for hospitals nationally. Another example is how 98 percent of its patients were properly given the heart medication known as beta blockers in conjunction with surgery as compared with 87 percent for hospitals nationally.
Indeed, from 2006 through the start of 2009, SHC rose from the middle of the pack of 100 hospitals to the top 93rd percentile on compliance in the two dozen or so core quality measures of the care of patients with heart attacks, heart failure, pneumonia and surgical conditions. When these numbers dropped a bit in mid-2009, hospital officials quickly pinpointed where the effort had suffered a slight decline — appropriate use of beta blockers in acute myocardial infarction patients and the consistent ordering of blood cultures for pneumonia patients were two examples — and the number quickly climbed back up.
These changes in quality performance have yet to substantially affect health-care spending, but there’s reason to believe that they will as the health reform provisions are enacted over the next decade. The health-care sector has lagged behind other sectors in adopting the latest information technology — and it takes time for investments in this realm to have an effect. So health-care providers are just beginning to see the increased productivity they can reap from re-engineering their operations. The bottom line: Teaching hospitals will be leveraging technology and management science to get more health out of each dollar spent, and that means a more careful marshalling of resources.
And that is why Rizk is, among other things, so interested in blood.
Earlier this year, Rizk met with hospital leaders to explain how SHC was using twice as much blood products as several of its very best peer institutions, which apparently had adopted the new transfusion protocol more quickly. The group agreed that SHC’s computer system should give a pop-up educational message when someone orders blood for a patient who doesn’t meet the guidelines. Those at the meeting opted for a note that would be more of a gentle nudge than an edict from on high.
Of course, some physicians may be in such a rush that they gloss over the message. Others may have someone else enter the order for them and never see the message at all. Undoubtedly, doctors will need to be engaged on a more direct, personal basis. Even then, some may bridle at the rule regardless of how gingerly it’s raised, seeing it as an example of “cookie cutter” or “cookbook” medicine.
Still, if the numbers don’t improve, Rizk says, more scrutiny could bring about change. The hospital is now keeping track of blood use. The new electronic health record could make it possible to quickly analyze which units are ordering the most and then drill down to determine which physicians are the biggest users. They then could focus the message on those who most need to hear it.
“It’s not the right treatment,” says Rizk, “and it’s a waste of money to be buying all this blood.”