U Magazine
U Magazine
UCLA Health
David Geffen School of Medicine

What Works? What Doesn't?

Through comparative effectiveness, evidence-based medicine can help to shape healthcare of the future.

By Kathy A. Svitil. Illustration by Tina Zemmler

MedMag Fall 10-What WorksFor a consumer, comparison shopping is easy. Need a new refrigerator and want the best model within your budget? Open up Consumer Reports and compare brands. But what if you’re a doctor caring for patients with heart failure? There’s no annual issue of Consumer Reports, or any other publication, summarizing the relative merits of treatment protocols for coronary-artery disease.

But there is “comparative effectiveness” – a hot-button buzzword during the recent legislative debate over healthcare reform. At its most basic, comparative-effectiveness research uses evidence-based studies to directly compare one medical treatment with another to determine which does the most good for what patients and when.

 “Current healthcare reform emphasizes that if you’re responsible for a population of patients, you want to provide and promote the best, most effective care, discourage what is not effective, and when there is a less costly choice with similar or better results, use it,” says Samuel Skootsky, M.D., chief medical officer of the UCLA Faculty Practice Group and UCLA Medical Group. “For example, generic-drug prescriptions are often as effective and cost much less than name brands. Is a generic always the right answer? Of course not. A name brand, at times, might be the right choice. But, there is a huge number of brand-name prescriptions being written when generic versions are equally good.”

To help physicians at UCLA make these choices, Dr. Skootsky has worked with colleagues to develop a series of guidelines for more “rational” care choices. One guideline, for example, summarizes the best approaches – rooted in evidence-based medicine – for ordering diagnostic imaging in a variety of different situations.

“It is sometimes necessary to order an MRI of the back for back pain,” Dr. Skootsky says, “and when you order this kind of imaging, you can order plain MRI or MRI with a contrast agent, or both. But most common back pain issues don’t need the contrast scans.” As a result of the guideline, he says, “we’ve seen a decrease in the number of dual requests for both plain and contrast scans.” To date, Dr. Skootsky and his colleagues have developed guidelines in about 20 areas. “There are maybe 300 we should develop,” he says.

In addition to helping physicians make the best choice from among the various options, “we want to eliminate situations where the right evidencebased treatment is not provided because of an error of omission, where the physician simply didn’t think about a particular therapy,” says cardiologist Gregg Fonarow, M.D. “We also want to guard against overuse of ineffective therapies.”

An innovative program developed more than 16 years ago by Dr. Fonarow, director of the Ahmanson-UCLA Cardiomyopathy Center, associate chief of the Division of Cardiology and co-director of UCLA’s Preventative Cardiology Program, has helped to standardize treatment nationwide of patients who are hospitalized with heart attack, heart failure or stroke. Dr. Fonarow and his colleagues have demonstrated through multiple studies that the evidence-based standardized orders and protocols used in the program – application, for example, of aspirin, beta blockers and angiotensin converting enzyme (ACE) inhibitors coupled with cholesterol-lowering medications, exercise, diet and smoking-cessation programs – have significant positive effects. They have been shown to reduce the danger of recurring cardiovascular events, lower the chance of patients being rehospitalized 30 days and longer after discharge and decrease the risk of death by 50-to-80 percent.

The system, called the Cardiovascular Hospitalization Atherosclerosis Management Program, or CHAMP, has evolved into the American Heart Association’s Get with the Guidelines Program and has been adopted by more than 1,500 hospitals in the U.S., benefiting some 2.5-million patients. CHAMP “was designed to improve how patients are treated and produce highly reliable and safe care for all patients, no matter who the doctor or nurse is,” Dr. Fonarow says. “This is not cookbook medicine. It is a highly reliable system-based approach to provide the most evidence-driven effective care to patients with cardiovascular disease.” In addition, “the system we’ve created is broadly applicable to other disease states as well,” he adds.

Comparative-effectiveness research was among the most controversial elements of this past year’s federal health-reform debate. Critics of the approach worried that it could lead to rationing if researchers recommend low-cost treatments that might be well-suited for most but not all patients. Those concerns were largely addressed by wording within the law that prohibits using the findings from comparative-effectiveness research as mandates, guidelines or recommendations for insurance payment, or to deny coverage.

An independent body, the Patient-Centered Outcomes Research Institute, was established to oversee the research. “This novel undertaking has tremendous promise for measurably improving the quality of healthcare in the United States and the health of all Americans,” says A. Eugene Washington, M.D., M.Sc., vice chancellor for UCLA Health Sciences and dean of the David Geffen School of Medicine at UCLA, who was named chair of the governing board.

MedMag Fall 10-What WorksWhat DoesntBut beyond individual treatment choices, the concept of comparative effectiveness extends to a far broader issue: What is the best way to deliver care for chronic illness? “Right now, we have a certain system for delivering care,” says Tom Rosenthal, M.D., chief medical officer of UCLA Health System. “If you have back pain, for example, you see an orthopaedic surgeon. But this is an expensive way to practice medicine. Maybe it works well for certain illnesses, and maybe it worked when most people died of infectious diseases, but in the 21st century, where you have lots of elderly people with multiple diseases, shuttling them from doctor to doctor is ineffective, inefficient and costly.”

The alternative, Dr. Rosenthal suggests, is a more coordinated system and one in which the services provided by a particular healthcare worker are a better match for his or her skill set. “The idea is to let everyone practice at the highest level his or her education will allow,” Dr. Skootsky concurs. “Doctors do doctor work; nurses, nurse work; medical assistants, medical-assistant work.”

Such efficient utilization of resources will become more necessary as baby boomers age and begin to have more and more chronic health problems. For example, says Dr. Rosenthal, “a person with congestive heart failure could be assigned a nurse who goes to the patient’s house so he or she can avoid 18 doctor visits with different specialists. Cardiologists would still oversee the cases, doctors would still have their roles, but instead of half of it being work that’s below their skill level, the process is more efficient.”

In this vein, Dr. Skootsky and his colleagues have conducted a small pilot study to remotely monitor the weight of patients with congestive heart failure. Even a seemingly mundane gain of five pounds can signal serious problems, as it can mean that the body is holding onto extra fluid because the heart is failing. However, patients might not pay as close attention to their weight as they should, or might not say anything about it to their doctor. And then they land in the emergency room, with real issues, “which can happen very quickly,” Dr. Skootsky says.

Dr. Skootsky’s program is designed to relieve the patient of the responsibility for self-monitoring for such changes. A nurse goes to the patient’s home to introduce the program and set up the telemedicine monitoring equipment. The patient just has to weigh him- or herself daily, and the data are remotely monitored by the telemedicine system. If the patient’s weight is above a predetermined danger threshold, the system sends an alert over the Internet to the nurse. “If the patient is getting into trouble, that’s when the nurse calls. It’s an example of improved patient monitoring by a member of the healthcare team who isn’t the physician. The physician is brought in only when he or she is needed for intervention – a change in treatment, for example.”

Currently, about 30 patients are enrolled in the program, and a few more are added every month. “So far, it is working very well,” Dr. Skootsky says. “The patients like it, the doctors like it, and it appears to be reducing the need for hospitalizations, although we require more patients in the program to be sure.”

This model will be evaluated further in a 1,500-patient comparative-effectiveness study of transition-care programs for elderly congestive-heart-failure patients at the five University of California medical centers and Cedars-Sinai Medical Center. The studies, which are funded by a $9.9-million grant to a team led by UCLA internist Michael Ong, M.D., Ph.D., will compare two different programs that are designed to ease the transition from inpatient to outpatient care. Standard post-discharge care will be provided by each medical center, with structured follow-up telephone calls from a centralized group of nurses, and remote monitoring of patients and as-needed telephone calls similar to Dr. Skootsky’s model.

“Each program is designed to provide a ‘warm’ handoff after the patient is discharged from the hospital but before he or she begins outpatient care, with the goal of reducing hospital readmissions,” Dr. Ong says. “Each also checks in on patients up to six months after discharge,” whether through regular telephone check-ins or remote monitoring. “This is the type of care we should be providing, and trials show that approaches like these are effective at reducing admissions. But they can be very expensive for providers. So the question is, how do we implement them in a way that is just as effective but at lower cost?”

MedMag Fall 10-What Doesn'tSuch evidence-based evaluations are not possible without fairly large patient populations that can provide statistically significant results. “This is a challenge in pediatrics,” says pediatric cardiologist Thomas Klitzner, M.D. “The frequency of chronic diseases in kids is much lower, and the variation in diseases is much greater, so the number of children you can study with homogenous conditions is low.” However, that doesn’t mean it’s impossible to develop more effective and more efficient care for children. “All children with chronic medical conditions will benefit from a less fragmented care-delivery system,” he says. “In pediatrics, we’re working to decrease fragmentation – not for cost reduction as much as for improving quality.”

To that end, Dr. Klitzner, the Jack H. Skirball Professor in the David Geffen School of Medicine at UCLA, and his colleagues developed a novel Pediatric Medical Home program for children with very complex medical conditions that require the coordination of several specialists.

Forty children, each seeing a minimum of two pediatric subspecialists, were initially enrolled in the program, which started in 2003 and follows guidelines established by the American Academy of Pediatrics.

The program involves four basic components: a 60-minute intake appointment, follow-up 40-minute appointments (twice the normal time), access to a bilingual family liaison to help families make their way through the medical system, and a family binder to keep all of the child’s medical information in one location.

“It is a pretty simple system,” Dr. Klitzner says. “If patients don’t have a medical home, their primary medical relationships tend to be with the specialists who take care of their illnesses. So they’ll see a neurologist for a headache and a dermatologist for a skin condition. Each organ system is treated individually, in a vacuum.”

In contrast, he says, the medical home “tries to sit at the center of that medical care, along with the patient. We make sure the neurologist and dermatologist notes are collected in one place, so there is a complete set of notes on the child.” In addition, appointments are coordinated so visits to separate specialists are on the same day, schools are kept abreast of the child’s condition so they will know whether or not the child will be returning to school, and so on. Studies do show that these simple steps have had a big impact, cutting children’s emergency-room visits by 55 percent while also improving patient satisfaction.

“We’re hoping to enroll from 400 to 1,000 patients in the program,” Dr. Klitzner says, adding that he would eventually like to expand the program beyond the UCLA-patient population. “Some 50,000-to-60,000 kids in Los Angeles County have complex cases, so we need to see if this approach will scale,” he says. “I think it is doable, but right now it’s still theoretical.”

It, indeed, is not clear that such tactics will be completely transformational. “There are some in the policy world who hope that comparative-effectiveness research will be a catalyst to make over the American healthcare system. But there’s really no telling how this will play out,” says Dr. Rosenthal. “There won’t be a ‘Big Bang,’ where suddenly everything is different. Change could take 10-plus years, and it might turn out that doing some of these things will not be efficient, or that they can’t be put into practice in the average clinical-care milieu.”

Regardless, identification of the need for this research is a significant step forward and an “important opportunity for academic medical centers like UCLA,” he says. “This direction provides resources for us to study how care is best delivered. And these,” Dr. Rosenthal concludes, “are important questions for us to answer.”

KATHY A. SVITIL is co-director of news at the California Institute of Technology and a former writer and editor for Discover magazine.


Add a comment

Please note that we are unable to respond to medical questions. For information about health care, or if you need help in choosing a UCLA physician, please contact UCLA Physician Referral Service (PRS) at 1-800-UCLA-MD1 (1-800-825-2631) and ask to speak with a referral nurse. Thank you.