Doctors' new focus on evidence
Best-practice tools are improving, so why are some doctors hesitant?
From the April 1997 ACP Observer, copyright © 1997 by the American College of Physicians.
By Deborah Gesensway
The patient wants to know why this year, for the first time, you don't want him to come in for an annual physical exam.
A resident wonders whether she should be adding a low dose of aspirin to the warfarin already prescribed to a patient with a mechanical heart valve who seems to be doing well at the moment.
Patients ask you about vitamin therapies—and whether they should try them.
An increasing number of internists are finding answers to these and other common clinical encounters in the growing discipline known as evidence-based medicine. If health care decision-making can be made to reflect more clearly the results of scientific research, then at least theoretically, physicians can be confident they are doing something that actually improves patient health.
Driving the evidence-based medicine movement is a belief that there must be a better way to keep up on the ever-changing medical literature. There is also the belief that the profession needs to cure its own well-documented problem of doctors being less and less knowledgeable about current best practice the longer they are out of medical school.
Viewed as the bridge linking biomedical research and patient care, evidence-based medicine is defined as "the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients," according to David L. Sackett, MD, a professor of medicine at Oxford University and one of the discipline's founders, writing in the British Medical Journal last year. Behind that simple definition are several philosophical beliefs: that clinical decisions should be based on the best available scientific evidence—which means using ways of thinking that derive from biostatistics and epidemiology—and that not all evidence is equal.
As good as this sounds, however, practitioners are discovering there are limitations on what these new tools can do. The tools are just plain hard to put into practice. And as with all tools, there are fears that evidence-based medicine could be misused.
For instance, evidence-based medicine can offer only limited help in the many gray areas of medicine where there just isn't evidence about a treatment or a therapy. And despite the proliferation of computer technologies like the Internet that make evidence-based medicine possible, evidence-based tools are somewhat coarse in helping doctors translate the results of studies into advice for individuals. And the evidence-based tools must be used in conjunction with techniques that are tricky to master—clinical judgment and patient preferences.
On the plus side, the discipline of evidence-based medicine not only can help doctors keep up to date on the tremendous quantity of new medical research published every month, but it also can help them distinguish among research of differing quality and give them a non-judgmental language for describing to patients, payers and colleagues the reasoning behind their decision-making processes. For some, a single taste of the discipline induces a kind of devotion.
W. Scott Richardson, MD, for instance, an internist and professor at the University of Rochester School of Medicine who has co-authored a new textbook on the topic, sees evidence-based medicine as one way to know that he's not inadvertently hurting patients.
"Everyone has his own story, but the one I think about is how we were taught that all patients after a myocardial infarction who had arrhythmias should get arrhythmia-suppressing drugs," Dr. Richardson said. "But then along came a randomized trial that showed that even though it was our best idea of science at the time and seemed to make sense theoretically, it actually hurt patients," he said. A few examples like this, Dr. Richardson said, and "pretty soon you get to the point where you realize that now that a trial is out, it's very important that we know the results and put them into practice."
According to a series of recent editorials in the ACP Journal Club, the steps to practicing evidence-based medicine include finding the evidence, understanding it, developing evidence-based clinical policies and then applying those policies in the right way for the right patients. Evidence-based methods include computerized literature searching, guidance on how to appraise studies and clinical practice guidelines.
As Dr. Richardson explains, "Evidence-based medicine can be thought of as [a collection of] power tools rather than hand tools" for taking the results of clinical trials and crafting them into clinical decisions about specific patients. "Evidence-based medicine is about individual clinicians with their individual patients trying to use not only the physicians' clinical expertise and the patients' preferences, but also integrating into the decision information that is available and of good quality from the research literature," he said.
But while some practitioners like Dr. Richardson find that evidence-based medicine has transformed the way they practice, other doctors express serious reservations. Evidence-based medicine may be the lingo of the day among clinicians concerned with the quality and cost of health care, but it's far from universally accepted, let alone practiced.
Outside of Canada's McMaster University, where evidence-based medicine was pioneered and infuses training from day one, very few medical schools and training programs teach its methods in any systematic way. Circulation of the year-old medical journal Evidence-Based Medicine, published jointly by ACP and the British Medical Journal, stands at a respectable but small 2,000 in North America. (The concept is more widely accepted in Great Britain, due to the influence of the National Health Service.)
Few would argue that physicians should consider the evidence gleaned from 50 years and billions of dollars of biomedical research when making decisions about patient diagnosis, therapy and treatment, so why does the term "evidence-based medicine" seem to turn off nearly as many doctors as it inspires?
According to Eleanor Z. Wallace, FACP, former ACP Governor for New York and now co-chair of the New York Chapter's evidence-based medicine committee, one problem is that evidence-based medicine is "impossible to do by yourself. The tools aren't ready yet. Most practitioners are just too busy. Many of them are still computer shy. They need training and support."
Resistance to evidence-based medicine is in part fed by computer illiteracy and innumeracy, but experts say that problems also come from the implications for physician autonomy. "Part of the misunderstanding is that people think that the evidence tells them what to do," said R. Brian Haynes, FACP, chief of the health information research unit at McMaster University Medical Center and editor of the evidence-based medicine journals ACP Journal Club and Evidence-Based Medicine. Instead, he said, "the evidence tells them that this is what happens under particular circumstances. You still have to figure out how that pertains to your patient's circumstances, and your patient may differ in ways that are important. That takes a little bit of sorting time.
"On the positive side, the fuss about evidence-based medicine is that there is a lot of new information that we could be harnessing to improve the care of our patients," Dr. Haynes said. "The fuss on the negative side is that it's not all that easy for practitioners to take this new evidence in and incorporate it into practice."
As Dr. Sackett has written, "without clinical expertise, practice risks becoming tyrannized by evidence."
Jonathan M. Ross, FACP, associate professor of medicine at Dartmouth-Hitchcock Medical Center, ACP's Governor for New Hampshire and Chair-elect of ACP's Board of Governors, said he considers the tyranny of evidence a not wholly unexpected side effect, given the fact that the popularization of evidence-based medicine has happened concurrently with the cost-containment and quality-assurance movements in health care.
As an example, he points to HEDIS, the data set the National Committee for Quality Assurance's uses to accredit health plans. While HEDIS measures the mammography rate for a health plan's population, Dr. Ross said that it should ideally measure quality by noting how many patients who understood the risks and benefits and evidence of mammography screening—and decided they wanted to be screened—got mammography.
"We are applying population-based statistics fast and loose in a mindless way because statistically the population will benefit," Dr. Ross said. "But that ignores the fact that women are individual people and should choose to participate or not to participate." Dr. Ross said he can't help but fear sometimes that evidence-based medicine may become a modern, sanitized way of taking individual patient choice out of medical care.
"Evidence-based medicine is definitely a useful tool, but let's not oversell it," Dr. Ross said. "Regardless of what anybody says, medical care is still mainly provided one-on-one. You talk to people and you touch them and you examine them and you counsel them. The tools of clinical epidemiology are not going to change that."
Besides, even after 50 years of intensive investment in biomedical research, there remain vast areas of health care for which there is no evidence from good clinical trials. "The absence of data doesn't mean that something doesn't work," Dr. Ross said. But that's not the way some physicians and payers have interpreted it, as demonstrated by HMOs that justify refusal to pay for certain services by claiming lack of evidence of effectiveness.
In fact, practicing in areas for which there is little evidence is something of a dilemma for evidence-based advocates. "If we don't have the evidence, how do we deal with that?" asked Rosanne M. Leipzig, FACP, who is putting together a center for evidence-based geriatrics at Mt. Sinai Medical Center in New York City. "Do we wait for the studies before we do something for patients? What's our second best choice? I think this is the next stage of evidence-based medicine."
And even where there are good clinical trials and the evidence seems dramatic, appearances can be deceiving. Evidence can be presented in many different statistical formats, and studies have shown that people make different decisions about how they will react to the same evidence depending on the way it is explained.
Robert A. McNutt, FACP, chair of internal medicine at the University of Wisconsin, Milwaukee, cited a study that randomized physicians into two groups to receive the same information about a treatment's effect in two different forms. One received information about relative risk reduction; the other group received information about absolute risk reduction. About 90% of the doctors getting the relative risk reduction information said they would choose the treatment, but only about 20% of the doctors getting the absolute risk reduction figures said they would use it, he said.
To understand the subtleties of the difference between absolute and relative risk reduction, Dr. McNutt said, consider the questions surrounding low-dose aspirin therapy for primary prevention of myocardial infarction. "I may have a 50% reduction in MI deaths (relative risk), but the absolute difference is eight in 1,000 vs. four in 1,000," Dr. McNutt said. "Is it clinically significant? I don't know. But presenting just the single number—the 50% reduction in MI—that's unethical. Looking at that marginal difference around numbers that matter is what evidence-based medicine is about. But it's tricky to do."
In addition to the practical difficulties of putting evidence to work, there are also some psychological reasons why some physicians may be holding evidence-based medicine at bay. "I think it's a control phenomenon," Dr. McNutt said. In theory, he explained, evidence-based medicine gives the youngest resident—and even a patient who understands the biostatistics—the same opportunity to make the best decisions concerning patient care as the most respected, senior clinicians. "Knowledge is power," Dr. McNutt said.
Dr. Wallace said she thinks some of the backlash also is clearly related to the fact that evidence-based medicine is frequently presented as one more way physicians, already bombarded by managed care and financial cutbacks, need to change how they practice.
"Another confusion related to this is that in managed care companies, administrators whose bottom line was cost—not more effective care—have gotten into the fray," Dr. Haynes said. "A lot of administrators have used the term evidence-based medicine to justify their own cost-cutting actions." This helps put physicians on the defensive.
Despite all these drawbacks and limitations, however, evidence-based medicine still appears to offer the best hope for improving the quality of health care today. By pointing out gaps in knowledge and gray areas of evidence, it can help set a research agenda for the next generation. As computers become easier and faster to use and as practitioners take advantage of the growing number of how-to courses on the topic, the barriers to putting it into practice might begin to topple. Plus, proponents of evidence-based medicine are busy creating resources such as ACP's new disk and CD-ROM, "Best Evidence," which provides instant electronic access to research from clinical trials abstracted in ACP Journal Club and Evidence-Based Medicine.
At the giant Minnesota HMO Health Partners, for instance, physicians and administrators explored ways to overcome these barriers during a continuing medical education course last month. They identified everything from accessing the evidence real-time to dealing with colleagues who dispute it and communicating it to patients. "We recognize that it is going to take entities in the system working together in order to advance the cause of evidence-based medicine in the day-to-day practice setting," said Maureen K. Reed, FACP, medical director for contracted care at Health Partners.
"We feel it is our obligation as an organization to improve the health of our members, and if we don't do that based on data and evidence, then what is our basis for it?" she asked. The alternatives, she said, are unacceptable. "Are we just going to keep doing things on a whim or on the basis of what garners the most revenues for the clinic?"
ACP puts evidence-based medicine into its clinical guidelines
Changes in the CEAP process include sppeding up development and working with other organizations
ACP plans to strengthen its 15-year-old Clinical Efficacy Assessment Project (CEAP) by remaining committed to evidence-based medicine, speeding up the development of guidelines and improving their usefulness to busy clinicians.
In a plan approved by College officials last month, CEAP guidelines will continue to be produced by a team that includes ACP staff, committee members and outside experts, but the process will take no longer than 14 months to complete. In the past, many CEAP guidelines have taken more than three years to write and approve, according to Herbert S. Waxman, FACP, the College's Senior Vice President for Education. In addition, the committee will follow a new process for identifying priority topics for guidelines.
The College also will begin to work more closely with other organizations, such as the new research centers being commissioned by the federal Agency for Health Care Policy and Research (AHCPR) and the Cochrane Collaboration. AHCPR recently partnered with two professional medical associations to develop evidence-based reports: the American Heart Association to develop an evidence-based report on valvular heart disease and the American Academy of Pediatrics to develop an evidence-based report on attention deficit hyperactivity disorder. Those reports are expected to be released mid-year. (Information: fax 301-443-7523.)
Despite the broad number of organizations involved in guideline development, a recent ACP task force that reviewed CEAP's mission concluded that the College should remain in the evidence-based guideline-writing business. "The strong consensus was that the College is still doing something that's important and unique enough that we should not abandon the activity," Dr. Waxman said. "But the process has to change dramatically."
In addition to speeding up the time it takes to develop guidelines, the task force recommended the College change the way it disseminates them to "ensure that there's broader knowledge, utilization and ultimately incorporation of the guidelines in patient care with an effect on outcomes," Dr. Waxman said.
The College is now working on new dissemination strategies, including electronic products and perhaps even publication in consumer magazines. "We can empower physicians to do better things, but we can also empower patients who are really our partners in a shared decision-making process," Dr. Waxman said.
Another area of reform of the CEAP process has to do with the "gray areas," those segments of practice where there isn't enough solid evidence to be able to write a useful guideline.
"We have to be responsive to the need of members for guidance, if not guidelines," Dr. Waxman said. "We don't quite know how to do that yet, but maybe it will be by identifying a range of acceptable actions rather than a specific action," Dr. Waxman said. This guidance might take the form of structured reviews that include information on what advice can be generated, he said.
While the program has been overhauled, guideline work has not stopped. The Board of Regents last month approved a new CEAP guideline on screening for thyroid disease. The new guideline expands a 1990 guideline to recommend that physicians screen all women over age 50 using a sensitive thyrotropin test (TSH).
The CEAP guideline places more emphasis on screening than existing guidelines. The rationale is that one in 71 women over age 50 has unsuspected but symptomatic overt hypothyroidism or overt hyperthyroidism that will respond to treatment.
Internist Archives Quick Links
Have questions about the new ABIM MOC Program?
One Click to Confidence - Free to members
ACP Smart Medicine is a new, online clinical decision support tool specifically for internal medicine. Get rapid point-of-care access to evidence-based clinical recommendations and guidelines. Plus, users can easily earn CME credit. Learn more