Evidence-Based Medicine Examined

April 1, 2009

Evidence-based medicine has been around for 16 years. But even though most physicians agree with the concept, it has not yet become the standard of American practice. Why not? And how can you make evidence-based medicine work in your busy office.

How long does it take for a new treatment for a given disease to make its way into routine patient care - after it’s been proven to be more effective than the previous standard? About 17 years, according to the Institute of Medicine. That huge lag time between the reaching of conclusive evidence for a new protocol and its inclusion into ordinary practice is one big reason that protocols in many practices are obsolete or incomplete, the IOM claims - with only half of today’s patients with many common diseases receiving adequate care.

What’s the problem? Is the research industry failing to get the word out? Nope, just the opposite, in fact. Modern physicians are drowning in a sea of information - and all of this new data is presented as vitally important (though a good bit of it is actually pure dross).

For primary-care physicians alone, that means reading about 90 journals regularly just to keep up. That’s about 15,000 articles per year, or 40 a day - in their spare time.

FIND OUT MOREIntegrating preventive care into your practice
 Maintaining patient confidentiality 
Let your voice be heard

Enter evidence-based medicine. The idea is to provide a framework for treating your patients with the latest proven research backing you up. Most physicians are aware of the evidence-based medicine concept, now 16 years old, and agree with it in principle. What doctor wouldn’t want to apply the latest proven treatments to patient care? And yet, the fact is that evidence-based medicine has not been fully embraced by many physicians - at least, not in practice.

What’s the problem? And how do you bring such a strategy into your practice and still have time for your family and a decent night’s sleep?

A quick primer

The end of World War II marked a turning point for medical research; many countries began investing earnestly, and by the 1970s, there was more information wafting about than any physician could reasonably absorb. “So we took it upon ourselves to try and help with that,” says internist Brian Haynes, a professor at the Michael DeGroote School of Medicine at McMaster University in Hamilton, Ontario. David Sackett, a physician who founded the department of clinical epidemiology at McMaster, pioneered the research effort that led to the development of what was dubbed (in 1993, by fellow researcher Gordon Guyatt) “evidence-based medicine.”

Evidence-based medicine is a systematic, corroborative method of practicing in which your medical decision making is supported by facts. By using evidence-based techniques, you can go beyond using just your current clinical knowledge and experience.

Its proponents argue that evidence-based medicine is not a replacement for the way you practice medicine; it’s an enhancement. You are, in essence, bringing in evidence that gives credence to whatever medical decisions you need to make for your patients.

It is not “cookbook” medicine, as some critics have complained. Evidence-based medicine evangelists note that if you practice with the stance that “that’s the way I’ve always done it,” you’re behaving more robotically than if you incorporate evidence-based protocols, through which you open yourself up to new ideas and scientific exploration. “The evidence is just a tool that can be helpful in some patient situations,” says Haynes. “We’re not inventing a new way of practicing medicine. We’re simply trying to get evidence wedged into the process to support the clinical practice.”

Evidence-based medicine is a logical progression of steps - rules, essentially. Say, for example, you have a 24-year-old male patient with a history of narcotic dependence for back pain. He presents with shortness of breath and increased fatigue five days after his long-acting oxycodone dose was increased. During his workup, which includes an echocardiogram showing cardiomyopathy with ejection fraction of 10 percent, he develops a fever of 39.6 degrees Celsius. Using evidence-based medicine protocols, you would:


  • Formulate a specific clinical question based on a particular patient problem to drive your inquiry. The question can relate to diagnosis, prognosis, treatment, outcome, or any other cogent factor - whatever makes sense for that patient. In this case, you might ask, “Will gentle exercise help, or more meds beyond the beta blockers?”

  • Search current literature for relevant information. Technology is extremely useful here, which we’ll discuss below.

  • Evaluate the evidence. Is it valid, useful, and pertinent to your patient’s situation? After a thorough evaluation of all found studies, you know that yes, there are some additional medications from which he could benefit. Strict compliance is essential. Physical therapy has also shown to be helpful, albeit somewhat less so.

  • Decide how you’ll treat your patient. You know your patient better than any preponderance of evidence. Suppose this patient struggles with both finances and treatment compliance. Knowing that, you decide to try the physical therapy first - it’s cheaper for your patient and more likely something he’ll do.

That’s about it. You stay in complete control of the clinical issue and its resolution. Perhaps you know this particular patient lives too far from the facility he’d need to frequent for the best treatment, or the most- recommended treatment is formidably expensive, and the second-best one is a fifth of the cost. Now you’ve got a set of facts that address risk and feasibility that you can discuss with your patient, and settle on a course of care.

Some realities

So that’s all well and good for one patient, but what if you’ve got 2,500 active charts? You certainly don’t have time to slog through voluminous piles of research every time a clinical situation presents itself to you. “It’s like drinking out of a fire hydrant,” says Chrissie Ott, a family physician from Portland, Ore.

Indeed, about 5,000 new studies come off the press every single day. Some bear important proof of a new treatment for a heinous disease. Others might be useful, but the sample size is too small to be credible. Still others are good for lining your parakeet’s cage. How can you get to what you need expeditiously yet still perform your due diligence?

With a little help from some friends. That means seeking out high-quality resources you can trust to do some legwork for you, so you can quickly check the facts and apply them to your patient’s situations. But where exactly to turn?

Use guidelines. The clinical guidelines proffered by various specialty associations - as opposed to evidence-based medicine, per se - are what some physicians say amount to cookbook medicine. But the guidelines are based on the existing evidence, not the other way around, and they are formed by teams of doctors and researchers who spend their days sifting through all that data. The guidelines save you time.

You’ll find scads of evidence-based guidelines within these organization Web pages, and “all are free and easily accessible,” says Robert Goldszer, chief medical officer at Mt. Sinai Medical Center in Miami Beach. He agrees that evidence-based guidelines can be very helpful. “Even on smaller things, like treatment of upper respiratory infections,” he says. “If people really followed guidelines from evidence-based medicine, we would … be spending a lot less on antibiotics because we wouldn’t have [as many] resistant organisms; we wouldn’t have given so many antibiotics to people.

“That’s a very common issue with evidence-based medicine,” he continues. “Evidence-based medicine for upper respiratory infection is Tylenol, Sudafed, liquids, and stay in bed and it gets better. It’s not the Z-Pak.”

Another issue with guidelines is agreeing on them. Love of autonomy is a prime characteristic of many physicians. Is it possible to agree on standards of care? “It’s a challenge but it’s been done plenty of times by many, many organizations,” says Goldszer, including the CDC, the AMA, and a host of others.

Consider setting up practice-level guidelines to follow based on the guidelines of large, well-established organizations, including:

  • U.S. Department of Health and Human Services - the Centers for Medicare and Medicaid Services and the National Guidelines Clearinghouse;

  • Agency for Healthcare Research and Quality - U.S. Preventive Services Task Force

  • The Joint Commission - Independent, not-for-profit organization that certifies healthcare organizations and programs in the U.S.; and

  • Other large, “blue-chip” medical associations - The American Diabetes Association, the American Heart Association, the AMA, the ADA, and the like.

You can turn to your guidelines whenever you need to add heft to your clinical decisions, especially difficult patients. Show them exactly where you’re getting your facts from. Posting your guidelines - in the office, on your Web site, or both - can be a positive marketing tool, as it shows your practice does everything it can to operate thoughtfully.

Moreover, following proven protocols can protect you from legal threats. “If physicians adhere to guidelines and do what’s recommended by national organizations, then they’re much less likely to have liability,” says Goldszer. “They’re doing the ‘right’ thing. They’re doing the recommended standard of care. If you’re worried about malpractice - which many people are - if you adhere to the guidelines, then you’re practicing standard of care and you’re much safer that way. [Physicians] are worried about standardization, but everything you read about quality improvement is that standardization helps. Variation hurts us. Guidelines and evidence-based medicine are steps towards standardization, and away from variation.”

Use technology. If ever the use of technology made sense for the medical world, organizing and vetting research data might be the most important reason. An EMR, for example, is designed to do so much more than just store electronic records. Many systems include decision-support software architecture to help you treat your patients without wasting time.

Most well-implemented EMRs alert physicians to treatments a patient might need but is not getting, based on the patient’s stored data and regularly updated clinical guidelines.

Reputable EMR developers consult credible guideline sources to set up their alerts, so you can feel good about their basis. You can often customize them, too, to address special needs you may have with your local population or payers.

Still, those evidence-based alerts can be annoyingly intrusive. “Physicians may be getting a whole lot of alerts/reminders they don’t want to see,” admits Sarah Corley, chief medical officer for NextGen, an EMR maker whose alert system is called Healthmonitor. “That’s the difficult part of EMR development: Where’s the best place to present the information, and who should be seeing it? What works for one practice won’t work for another.”

You can put your own EMR data to work, too. “Good EMR software can make it a breeze to maintain a disease registry,” says Cyrus Peikari, a solo internist practicing in Dallas, Texas. “It can allow you to run reports to make sure your diabetics have a recent A1c, lipid screen, etc.” And you’ll know exactly how often and for whom you should be focusing, because you’re evidence-based. Not too much testing, and not too little.

Looking beyond your own office, you’ll find a trove of evidence-based tools and resources through the Internet. Some are more useful than others; the trick is to know where to go so you quickly get good information you can trust. You might use Google, and that can be OK, as long as you know how to critically appraise the research studies you unearth. That takes time, of course. Medically focused Web crawlers, such as PubMed or SearchMedica (a portal that can be accessed through PhysiciansPractice.com, and whose owner, CMPMedica, is Physicians Practice’s corporate parent) will trim out much of the schlock for you.

But you can also find dedicated, Web-based tools to keep you evidence-based. Following are a few biggies often cited by physicians as their go-to sites for evidence:

  • UpToDate - Contains more than 76,000 pages of text, graphics, a drug database, and links to outside study abstracts. About 3,800 physicians from 18 specialties monitor 400 journals, critically appraise the study results contained therein, and incorporate the best information into UpToDate three times a year. This makes it more than simply a “journal watch,” as the actual content of a new study is considered critically. It costs about $500 a year, give or take, with many caveats and subscription types available. Ott has decided that, for her, despite it being “a little pricey,” she considers UpToDate superior to purchasing updated text books.

  • The Cochrane Collaboration - Offering some free access, the Cochrane Collaboration is very clear about avoiding what it calls “conflicted funding,” that is, funding from for-profit industries such as pharmaceutical or medical device companies. Within the collaboration’s library, you can search the “Cochrane Database of Systematic Reviews” for analyses of healthcare interventions. This organization also pulls in studies written in foreign languages.

Full access to the Cochrane Collaboration will cost you about the same as UpToDate. All of its review abstracts, however, are free. One great feature of the reviews: They contain a plain-language “translation” that you can use to help explain the efficacy of a treatment (or lack thereof). Do you have a patient who fervently believes in taking only Vitamin C for a cold? Show her the evidence-based review in the Cochrane Library that concludes, through exhaustive research, an analysis of all available data on that subject: that unless she’s a marathon runner or an alpine skier, Vitamin C will do little to staunch her runny nose.

  • eMedicine- Want to pay zero? eMedicine may work for you, but this WebMD-owned site does accept advertising. It does claim to be peer-reviewed; you’ll have to decide for yourself how to judge clinical studies that have been funded by the for-profit industry, such as pharmaceutical companies.

  • ACP Journal Club - American College of Physicians members have access to this service. You’ll find a decade’s worth of archived articles here, with obsolete material constantly weeded out. Content comes from about 100 clinical journals, with articles going through a stringent criteria acid test.

  • McMaster University - Ground Zero for evidence-based medicine methodology, McMaster’s Health Information Research Unit spends a great deal of time researching and forwarding the subject. Experts there also vet articles in support of the ACP Journal Club. You’ll find a wealth of evidence-based medicine links on the McMaster University Web site, through the Evidence-based Practice Center Web page.

Again, that’s a tiny sample of what’s out there. Note that many products you’ll find on the Web, such as UpToDate and PubMed, are downloadable to your PDA or iPhone, too. For a good list of other evidence-based medicine links and tools, visit

McMaster’s Evidence-based Practice Center

, and then click on “Partners/Links.”

The heart of the matter

By now, the evidence-based medicine ball is rolling downhill, gaining momentum in the medical world. “I’m a big believer in evidence-based, standardized medicine,” says Goldszer. “But it’s variable. Hopefully, some younger physicians coming out are more amenable to using guidelines. I think some of the older are more resistant.”

Is such resistance futile? No, just fairly irrelevant. There’s simply too much important data that physicians need to know, says Haynes. “There’s such wide acceptance now that we have to pay attention to evidence or research. How we do that or when we do that - both are debatable issues.”

Contrarians would be served best by channeling their energies into demanding improvements to the implementation process, rather than mere resistance to evidence-based medicine itself. It’s all part of the time-consuming maturity of a good idea.

And mature it must. “We shouldn’t have to wait for a generation to pass when new practitioners take over before the new evidence is in place,” says Haynes. “It has to be generated at a much faster pace. We don’t want to kill the old practitioners; we want them to catch up.” To be sure, evidence-based medicine offers structure to the complex world of modern healthcare, but it’s far from a cure-all. Pamela Wible, a family physician in Eugene, Ore., notes that it’s “an imperfect tool. It can inform, but it should never dictate. Medicine is an art, when you really get down to it.”

Think of the limitations of evidence-based medicine as a canvas upon which you do your best work. Sometimes, there will be a dearth of evidence to support your decisions.

Granted, change is difficult, and you need time to learn, acknowledges Haynes. “It’s quite difficult to pick up a new way of managing patients - particularly if you’re on your own - or a new drug from a treatment you’re already used to managing. You have to remember how to spell it when prescribing it. You have to remember what to say when the patients ask what the side effects are. You don’t have experience with the drug.”

You’ll also find that people rarely fit into the same mold. Compliance is a particularly unruly wild card. Even if you do determine a certain course of action to be the most effective, says Ott, “if a patient is not organized enough for a certain blood pressure medication, then you make a decision for the ‘best’ treatment. The most important thing is that the blood pressure be treated.”

“You’re not always going to have a study - nor do you need one - for every patient,” Ott adds. “That’s really important to the art of medicine.”

The art, of course, lies within you, when you’re faced with a sick patient and uncertainty swirls about while you decide on a treatment course. You can - and should - turn to techniques and methodologies to assist you in that decision. But in the end, the pressure’s on you to make the right call on behalf of that person.

And that is why evidence-based medicine will never replace actual doctors. We will always need human beings to treat human beings. “I wouldn’t go to a doctor who didn’t have the expertise or experience in the condition I’m trying to get help with. That can’t be replaced; it’s essential to solving health problems,” says Haynes. “But evidence is also very useful in doing that. They’re not against each other, they’re not either/or. The two of them together, expertise and evidence - that’s going to be the best recipe for improving a person’s health.”

Shirley Grace is a former associate editor with Physicians Practice. She can be reached via physicianspractice@cmpmedica.com.

This article originally appeared in the April 2009 issue of Physicians Practice.