Putting guidelines to work--lessons from the real world
By Deborah Gesensway
ANNANDALE, Va.--Michael A. Hattwick, FACP, like many internists these days, has an office bookshelf that is bending under the weight of the clinical practice guidelines he has collected over the past few years. There are the rainbow-colored series from the government, ACP's paperbacks on screening and diagnostic tests, a few loose-leaf binders from various managed care companies and utilization review consultants.
But Dr. Hattwick is one physician who is trying to let these guidelines do more than just collect dust.
"I often say to a patient, 'Let me check the guidelines.' No one ever objects to that. They are coming to me for information, and I'm just supplementing my local knowledge with national knowledge," he said. The transplanted Texan has taken it on himself to amass dozens of different guidelines, summarize each one's key recommendations and input the condensations into his notebook computer, which he powers up when the need arises.
"I have a strong feeling that it isn't I, Mike Hattwick, who should decide what the best treatment for back pain is. Yes, I have my opinions and my set of experiences, but so do a lot of other people. That's been the problem. ... Now, however, there are enough guidelines that you can go to those guidelines and not shoot from your own particular experience."
Dr. Hattwick is a perfect example of what some are starting to describe as the second phase of the clinical practice guidelines movement: actually using them.
Conventional wisdom has concluded that the proliferation of guidelines in the last decade has had minimal effect on how medicine is practiced in the United States. Millions of dollars and megawatts of brain power have been spent writing guidelines on everything from breast cancer screening to diabetes treatment, but with a few notable exceptions, medical practice remains marked by the same variability the guidelines were written to eradicate. The gap between what physicians know and what they do for patients appears as wide as ever.
Suddenly, however, a competing school of opinion has emerged. It talks of the Dr. Hattwicks of the profession, with his self-constructed computer database of guidelines relevant to general internal medicine and his automated system for flagging patients whom federal guidelines say need preventive services.
Some guidelines proponents these days are abuzz with numerous success stories, all of which show that guidelines are starting to penetrate medical practice. The stories hail from HMOs, which have boiled down complex guidelines so their practitioners can treat urinary tract infections cheaper and diagnose depression more effectively; from hospitals that have nearly done away with problems with pressure ulcers by instituting guidelines for nurses and doctors; and from peer review organizations, which are beginning to compare the medical charts they are reviewing against clinical practice guidelines instead of against some non-scientifically derived measure of quality.
There even is news of malpractice attorneys, who, it appears, are targeting doctors who fail to follow guidelines and refusing to take cases when they have discovered the physician has followed a guideline. Word comes as well from drug manufacturers that they are changing the way they market drugs so that their products will be in sync with guidelines-influenced formularies.
And evidence abounds of insurance companies and health plans hiring, deselecting and paying physicians according to their success in following particular published guidelines. "If [physicians] don't pre-certify, they don't get paid," said Stanley Stier, MD, associate medical director at Physicians Health Services, an IPA-type HMO in Connecticut and New York. "If they don't comply with our guidelines, we review whether they can continue to participate in our plan."
He ascribes the physician-owned organization's acceptance of guidelines to a basic philosophical shift. "For the most part, we have left judgment up to the physician, but we--and I mean society--have discovered that physicians aren't perfect any more than other people," Dr. Stier said. "The purpose of practice guidelines is to establish basic tenets we all can live with."
According to the Group Health Association of America's 1994 HMO Performance Report, Dr. Stier's organization is not alone anymore. More than 80% of HMOs say they are promoting the use of clinical practice guidelines in their plans. In an effort to counter criticism that their guidelines have made very little difference, the federal Agency for Health Care Policy and Research (AHCPR) has collected a hefty packet of "user stories." And California has passed a law requiring its state medical board to notify all physicians licensed in the state of AHCPR's pain management guidelines.
"You can hear an endless number of stories about this guideline or that guideline being used," said Sean Tunis, ACP Member, health program director at the congressional Office of Technology Assessment (OTA) and author of an article published in Annals of Internal Medicine last summer about internists' attitudes toward clinical practice guidelines. "There is this seeming inconsistency between the notion that guidelines have had no impact--and there are a series of good studies that have come to that conclusion-- and everything we are now seeing in the private sector, where specific guidelines have both been created and implemented [producing] dramatic changes.
"Lots of organizations that are having the greatest success in terms of generating and using guidelines aren't academic," Dr. Tunis said, "and therefore aren't interesting in publishing a record of what they've done. They are just trying to use them."
Whether this fledging use will mature into a medical practice revolution--where "quality" means compliance with guidelines and where notions of physician autonomy are upended--will depend on how effective guidelines prove themselves to be in lifting the quality of health care to new heights. Physicians, in general, have demonstrated they will not use guidelines until they are proved to be useful. But usefulness is exactly what appears to be getting more opportunity to be proved these days. The rise of managed care, with its goal of identifying and promoting value in health care, coupled with the growing sophistication of electronic information technologies, which makes use of guidelines and analysis of their effects an easier task, means experiments with guidelines should only continue and at a faster pace.
"If you have an organization that has real incentive to change practice, and it has the skills and resources to put something together, we are seeing that you can get pretty miraculous things done," Dr. Tunis said. "I think it is becoming increasingly clear that what does not happen is that physicians pick up a guideline from a national organization, whether it be the federal government or their favorite professional society, look at it and go, 'That's what I'm going to do.' "
"Harold C. Sox Jr., FACP, chairman of the department of medicine at the Dartmouth-Hitchcock Medical Center in Lebanon, N.H., is a self-confessed example. As chairman of the U.S. Preventive Services Task Force and a former chairman of ACP's Clinical Efficacy Assessment Project, Dr. Sox has built the bridges designed to connect knowledge with practice, but he has not necessarily crossed them--until now.
For example, he says he still does rectal examinations regularly even though, as he explains, "there is no real evidence that they make a difference and they are not recommended by most organizations as part of a periodic health exam." So why does he do them? Habit, patient expectations and little support from the system to help him and patients change, he answers.
Now, however, the medical staff in his large practice are experimenting with ways to get physicians to comply with colon cancer screening guidelines, focusing particularly on "reverse detailing"-type education. "I think it's a mixture of responding to financial pressure to meet the expectations of the purchasers for quality," Dr. Sox said, "and also a sense that reducing inappropriate variation is part of a new definition of professionalism for our organization."
Dr. Sox's experience is far from unusual. Although guideline originators mostly had expected the consumers of their products to be individual clinicians, there have been few instances of physicians taking the lead on implementing guidelines, according to David Nash, FACP, MBA, director of health policy and clinical outcomes at Thomas Jefferson University in Philadelphia. Instead, those buying in have mostly been those who pay the bills.
The translation dilemma
Guideline writers have also been surprised by how hard it has been to get physicians to use guidelines. In particular, the realization that no guideline is relevant for everyone's practice has been somewhat of a shock.
"It is interesting that many of the people who created the guidelines didn't anticipate this," said William M. Tierney, FACP, professor of medicine at Indiana University School of Medicine and health services researcher at the Regenstrief Institute for Health Care. At Regenstrief, he has been involved in programming five clinical practice guidelines into physicians' online ordering workstations. "[Guidelines developers] anticipated that you would be able to just pick up the booklet and use it. But what we end up doing is ... translating it into something that is operational for our particular needs and our particular institution."
What the organizations and individual practitioners who are motivated to change have learned is that a guideline isn't simply the guideline. To affect clinical practice, the guideline must be "translated" again and again so that it speaks the language of each local practice environment where it will be used.
"We mailed out something like 13 million" copies of the 15 federal clinical practice guidelines, said Douglas B. Kamerow, MD, MPH, director of AHCPR's guideline-writing office. It turns out, he says, "that was only the first step. Then someone has to take the guideline and do something with it. Just holding the book up to your head and saying a mantra is not going to change behavior. That translation step, the implementation, is also what we have to be concerned about."
Says Jonathan B. Brown, PhD, MPP, an investigator at the Center for Health Research in Portland, Ore., "People who draft guidelines worry about translation because they think people will bias their recommendations." Dr. Brown has studied how Kaiser Permanente's Northwest Region implemented the AHCPR depression guidelines, and has concluded that the translation did not bias the guideline. "But on the other hand," he said, "a lot of transformations did occur, which all seemed to be for the good." These included making it shorter, adding practical advice, inserting phone numbers and even, controversially, meshing AHCPR's guideline with local opinion and formularies to spell out exactly which drugs should be prescribed to which types of patients.
As each of these experiments has shown, translations vary widely, depending largely on the goals of the organization implementing them.
Take the goal of reducing costs. Scott R. Weingarten, FACP, MPH, an internist and director of health services research at Cedars-Sinai Medical Center in Los Angeles, has worked with a 20-member staff and a department of case managers to implement guidelines on seven expensive conditions, ranging from pneumonia and chest pain to lumbar laminectomy and knee replacement.
Getting the guidelines used--and care improved and length of stay or other hospital costs lowered as a result--has generally followed this process: The clinical nuances of the guideline are encoded into a computer program that allow case managers to classify individual patients according to risks. When a patient is considered by this program to be a safe subject for the guideline, a case manager contacts that patient's physician via a note attached to the chart. The physician is reminded of the guideline and requested to follow it. The doctor does not have to do so, but the hope is that if he has all the information, he will.
"We've found that no guideline is 100% perfect," Dr. Weingarten explained. For example, Dr. Weingarten found that some guidelines don't save money. That was Cedars-Sinai's experience with a guideline for congestive heart failure patients. Designed to safely reduce health care costs, instead it safely raised costs. That guideline is now in the process of being reassessed, he said. "Our message is that physicians should not assume that all guidelines that have even been written will improve patient care or save money."
As a result of his experience, Dr. Weingarten says he has come to "believe strongly that if a guideline is used to substitute for physicians' careful clinical decision-making, rather than to complement clinical judgment, it could be dangerous."
Across the country, Harvard Community Health Plan (HCHP) has been writing guidelines for 10 years to help its clinicians to lower costs, decrease variation and improve outcomes--or at least not harm them. In general, the driving force behind the guidelines has not primarily been clinicians concerned about variability in practice, but more, HCHP's management focusing on patients' and purchasers' concerns. The goal is to use "quality improvement" to remain successful in a highly competitive market, said Lawrence K. Gottlieb, FACP, director of the Brookline, Mass.-based HMO.
Consider HCHP's recent experience implementing a set of guidelines to improve asthma care. "Asthma is a major contributor to hospitalization, outpatient costs and morbidity, so it was identified as a priority for us," Dr. Gottlieb said.
"It was not up front a priority of clinicians, although soon after we started working on it, they certainly recognized it as a place where they could use some help," he said. "I don't think it's so much that they are unaware that they might be helped. It's just that it's not necessarily what's in their mind on a day-to-day basis when they are trying to take care of patients."
HCHP culled national guidelines on asthma care and then focused its efforts on ways of "getting from a guideline to improved practice," particularly, he said, when "the barriers aren't necessarily knowledge." In this case, what has developed is a system of special case nurses who function for asthma patients as diabetes nurses have for some time--educating and doing follow-up with patients. The guideline implementation also could not have been done without an HMO commitment to distribute peak-flow meters to all asthma patients.
"In this case, it's pretty much of a no-brainer. There aren't physicians who would be resistant to practicing according to these key indicators. So, if there is resistance, it's not out of intellectual resistance. ... The big issue is creating a process of care that doesn't set up barriers to doing what you want people to do."
Guideline developers, from AHCPR to specialty societies including ACP, have now begun listening closely when these pioneer guideline users speak. They are looking for clues about building a better guideline from the beginning. This, they hope, will be the next phase of their movement.
"The bottom line in our activities is improving health status," said AHCPR's Dr. Kamerow. "The question is, 'Is practice changing?' ... But first we have to get feedback about what has been useful and what hasn't been so we can create guidelines that have an easier portability into practice or review criteria or whatever."
Guidelines might stand the tests of the day, let alone the tests of time, and catch on throughout medicine. Or guidelines could slowly shrink in importance, as the process of inducing behavior change in physicians becomes better understood, if guidelines are shown to have a much more limited role in promoting best practices in clinical care.
"If you believe that many inadequacies of care don't necessarily emanate from lack of knowledge, but that there are other barriers, then it is easy to understand why just creating a new knowledge source doesn't do very much. It is just step A," said the OTA's Dr. Tunis. "If somebody is unable to perform a function because they don't have the capability to do it in some way, it doesn't matter how loudly you tell them to do it."
College wants feedback on how its guidelines work--or don't
From the March 1995 ACP Observer, copyright © 1995 by the American College of Physicians.
To make its guidelines more useful to doctors, managed health care plans and others involved in medical quality improvement efforts, ACP is shifting its emphasis from producing more guidelines to evaluating the usefulness of existing and future guidelines. The goal will be to improve physicians' ability to implement guidelines in their practices.
According to Linda White, ACP's Director of Scientific Policy, this new focus will pervade the guidelines development process, from more focused topic selection to publication in easier-to-access electronic formats. The College will also attempt to join with internists in managed care organizations and with those physicians involved in outcomes research to study whether guidelines make a difference for patients.
"We want to get feedback on where they [guidelines] work or don't work to further refine our process," Ms. White said.
Through the Clinical Efficacy Assessment Project (CEAP), ACP has been drafting and publishing clinical practice guidelines since 1981. It has produced about 160 guidelines, about 60 of which remain "active," meaning current, updated and relevant to the practice of medicine. Eleven more, on topics ranging from screening for prostate, breast, colon and ovarian cancers to diagnosing patients with dizziness, are in the pipeline. The CEAP guideline process relies on literature rather than on expert opinion and is generally viewed as helping to raise the scientific standards guidelines nationally are expected to meet.
The CEAP committee recently surveyed a sample of College members to identify about 40 topics of interest to internists. Now, the committee is narrowing those topics into areas in which a guideline could be most useful to practitioners, Ms. White said. For example, instead of taking on the daunting, and probably less than useful, proposition of writing a comprehensive guideline on prevention of osteoporosis--a topic identified by the survey--CEAP will be focusing on the question of how to prevent fractures in someone already diagnosed with osteoporosis, she said.
"Usefulness starts with the selection of the topic," Ms. White said.
In terms of methodology, she said, the College is going to attempt to build on its budding involvement with the Cochrane Collaboration, a multi-institutional, international effort to produce systematic reviews of randomized controlled trials. The idea would be to translate the Cochrane systematic reviews into guidelines of clinically useful information.
In terms of increasing guidelines' accessibility, the new edition of ACP's "Clinical Practice Guidelines" book, scheduled to be released this month, will come with a fully searchable diskette that will include the full text and a structured abstract for each CEAP guideline. Plans are in the works to upload the structured abstracts to ACP Online as well.
In addition, Ms. White said, the College is putting together an online newsletter that will summarize various organizations' guideline efforts and attempt to review guidelines produced by other physician specialty societies, managed care organizations and the federal Agency for Health Care Policy and Research.
Another new ACP effort, called "Consensus to Care," is aimed at bridging the gap between guidelines and practice by producing clinical tools and technologies that educate physicians about guidelines and help them adopt the new knowledge in their practices.
The first study under this project is focusing on ways of educating physicians about Lyme disease. That study, funded by the CDC, will focus on whether a kit of print materials and a videotaped presentation, developed from evidence-based guidelines, help physicians adopt a set of guidelines on better treatment of suspected Lyme disease.
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