Clinical practice has long been art vs. science. But with evidence-based guidelines and electronic decision support, some fear art may be losing the battle, altering a delicate balance that has defined traditional medicine for generations.
Clinical practice has long been part art, part science. But with evidence-based guidelines and electronic health record (EHR) systems with decision support becoming increasingly prevalent, some fear the art of medicine may be redrawn.
And although this trend poses some interesting legal questions long-term, several experts agree that guidelines will never overshadow your most important clinical tool-your intuition, especially when it comes to reaching the right diagnosis.
"Guidelines are just that," says Matthew Finneran, MD, of Wadsworth, Ohio, who practices family medicine and geriatrics. "They don't always account for patient requests-and their fears-which physicians have to address on an individual basis." And the medicolegal aspect of not following guidelines needs to be considered too, he says.
"No guideline fits every patient," even if it is based on a great deal of evidence, adds Carolyn M. Clancy, MD, director of the Agency for Healthcare Research and Quality (AHRQ). "They are just starting points. Clinical judgment will always be important, especially when something just doesn't fit." For example, she says, how a patient looks or the nonverbal cues he or she gives off can be just as essential to achieving an accurate diagnosis as what the patient says. "This is when clinical experience really helps," Clancy says.
Sorting out what a patient says from how he or she says it is the finesse of a good diagnostic physician, agrees American Academy of Family Physicians (AAFP) President Glen Stream, MD, FAAFP, MBI.
"Tests will rarely tell you as much as a patient story will, and a check box can never replace good history-taking," he says.
Amy Compton-Phillips, MD, associate executive director of quality at the Permanente Foundation, compares physician checklists with the checklists pilots go through before takeoff.
"They are not negative. They keep planes from crashing," she says. "Checklists can help us do things better by making is easier to do the right thing."
USING DECISION SUPPORT WELL
Compton-Phillips says Kaiser Permanente encourages practitioners to blend evidence-based care with compassion. Most evidence-based "checklists," she says, are reminders that help doctors meet patient needs, not barriers that prevent them from trying something else. For example, Kaiser Permanente's EHR helps orthopedic surgeons choose which joint replacement is right for each patient.
"We believe in leveraging EHRs to make care better and safer" and improve preventive care and management of chronic problems such as asthma, diabetes, and heart disease, she says. Compton-Phillips says that Kaiser has higher rates of breast cancer survival than some providers, which she attributes to the evidence-based care that its EHRs assist its physicians in providing.
"If we want EHRs to be more than electronic versions of paper charts, we need to use them to help us do things better," such as saving time, money, and resources while improving care, she says.
Evidence-based medicine has a lot of "low-hanging fruit," she says. "There are not many [patients with diabetes] who don't want to know their blood sugar levels or older women who don't want mammograms." Even the best guidelines are not perfect, Compton-Phillips says.
"If one in five patients does not fit the guidelines, then you use your best clinical judgment and experience to treat them," she says. "EHRs and evidence-based medicine help you do the right thing most of the time, but there is no punishment for going past the guidelines."
She adds, however, that Kaiser Permanente encourages learning from such experiences. "Is the guideline wrong? Is the decision support wrong? How can we make it better?"
"In the long run, having information available electronically will transform everything we do," predicts Clancy of the AHRQ.
She believes that EHRs will become tools through which doctors learn as a byproduct of what they do. For example, she says, drugs are used for off-label indications about 20% of the time, but because these experiences rarely are documented, no one else learns from them. Having the information in an electronic record makes it accessible to others.
A well-used EHR can expand the learning that goes on now at continuing medical education meetings and other venues, she says. For example, if a new infectious disease is spreading, an EHR can allow doctors to learn exactly how it is affecting their communities, not just nationally.
Physicians used to have to go a library to research tough cases, she notes. Today, many of them visit PubMed ( http://www.pubmed.com/) in the evenings at home when they need to hunt down an answer. An EHR can allow them to take this action in real time, she says.
Learning new information is a highlight of being a doctor, she adds, and most medical professionals want to keep up with the latest information but find it overwhelming to do so at times. Rapid access to information can make it easier for physicians to make the right decisions regarding care for their patients, not stop them from following their intuition.
ACCEPTING THE INEVITABLE
Moving to more check box-based medicine can be complicated by the simple fact of learning how to navigate an electronic chart instead of a paper one. AAFP President Stream says he often finds himself asking reluctant physicians whether they really think paper charts will still be used in 50 years.
"They say no. I say, 40 years? No. Thirty years? No. Well, I tell them, then let's get started," he says.
Stream believes, however, that this changing work model makes the personal touch and intuition in medicine more important, not less.
"We need to preserve the human aspects of what we do and resist seeing our patients as being punched out of a mold. The art of hunch is more important than ever to help us recognize when the patient in front of us doesn't fit an algorithm," Stream says.
He encourages physicians to view evidence-based guidelines as a way to help them remember specifics, such as how often to order certain tests for a patient with diabetes.
"The power of the computer can help us remember to do the right thing, be our safety net, but blindly following it can lead to mistakes," Stream says.
TREATMENT, NOT DIAGNOSIS
At least one healthcare information expert draws a sharp distinction between using evidence-based guidelines for diagnosis and using them for treatment.
Peter E. Alperin, MD, is vice president of medicine at Archimedes Inc., a startup healthcare technology company focused on mathematics-based simulations of physiology and the healthcare system. Previously, he was a director at Eprocrates, where he was responsible for developing the popular handheld formulary support tool for physicians.
In addition to his work at Archimedes, Alperin maintains a part-time internal medicine practice. He says he believes that data tools cannot replace a physician's skill in gathering histories or performing examinations.
"They do not help you make an accurate diagnosis more quickly, other than having more data such as past computed tomography scans immediately available," Alperin says.
Diagnosis is, in fact, the right time to follow a hunch, he says.
"Hunches are decisions that you cannot fully explain. They can be right, or they can be wrong, but it is more appropriate to follow them on the diagnostic side than the treatment side," he says.
When choosing a treatment, technology and guidelines can suggest best practices, show what has already been tried, and warn about potentially harmful interactions. They also can help apply findings from larger populations to each individual patient's specific situation.
"Don't vary treatments based on hunches. Use data," Alperin advises.
He believes that evidence-based tools can provide better care for patients and will become even more useful than specific physician skill.
"The art of medicine is important, but there is so much variability in healthcare today despite the fact that there may be a proven 'best way.' Embedding those guidelines into treatment decisions can help doctors use them better," Alperin says.
For example, he says that despite the fact that proven best ways exist to prevent blood clots after surgery, surgeons today often follow their own protocols.
"Decisions made off-guidelines are not always better," Alperin says. "Embedding data-based protocols into EHRs may increase the rate of compliance with their recommendations."
For sure, one good side to following the checklists of modern medicine is that doing so can provide some protection for doctors in the event of a malpractice suit.
Checking all the appropriate boxes "shows you followed the standard of care," says Lee J. Johnson, JD, a health law attorney Mount Kisco, New York.
Leaving boxes blank can make it look like you didn't address certain areas, she adds.
"You will have a good defense if you can show you followed the standards that have been spelled out. But don't let this stop you from having new ideas and using your imagination. Just be sure to document your reasoning," says Johnson, an editorial consultant to Medical Economics.
"Look at evidence-based medicine as a baseline to prevent you from missing something obvious," she adds.
Compton-Phillips, however, points out that many evidence-based guidelines are available, and they sometimes disagree, so rigidly following any of them cannot completely protect you from legal action.
"You have to decide what is best for each patient. Doing no harm lets you defend your actions," she says.
Upcoming implementation of new requirements related to the International Statistical Classification of Diseases and Related Health Problems, 10th Revision, or ICD-10, will force doctors to pay more attention to "check boxes" as part of the documentation process.
The new coding system, coming in 2014 if the rule proposed by the U.S. Department of Health and Human Services takes effect, will require much more detailed information and will increase the number of diagnosis codes to about 69,000 from about 13,600.
Alperin believes these changes are a good idea, as they will lead to collection of better data, which can be used to create more individualized care recommendations that consider many more variables, thus yielding better outcomes. This is a goal worth pursing, he says, but he acknowledges it will be time-consuming.
"Everything we do comes at the expense of something else," he says. "This becomes a tax on a doctor's time, giving them fewer minutes to talk to the patient. I can't imagine they will see fewer patients, since they are paid by patient volume."
Having electronic support may be crucial in this new era, Compton-Phillips says.
"Coding is the bane of our existence," she says. "We want to do it well because we know it can lead to better data, but it takes so much time and attention from the practice of medicine. I don't know how physicians who don't have electronic systems will comply."
BE OPEN TO THE FUTURE
Adopting new practice patterns that incorporate evidence-based care well will be a challenge for many years to come, the experts agree.
Above all, Clancy encourages physicians to be patient and remember that the ultimate goal of the changes is to help make it easier for you do the right thing.
"Creating something better never happens overnight. There are always bumps," she says.
Stream encourages physicians to be willing to trust their clinical judgment.
"If the computer is telling you something that you think is wrong, pay attention to that feeling," he says.
"But also consider that the computer may be right," he adds quickly. "Look into it either way."
The bottom line always will be good patient care, which benefits from a strong doctor-patient relationship based on trust.
"We have to tell patients what they need, not always give them what they want," says Conrad Flick, MD, a family physician in Raleigh, North Carolina. He advises, however, that doctors "treat patients like you would your own family. If a patient really wants a prostate-specific antigen test, do it, but note in the chart that you had the conversation about the new recommendations and that you made 'x' decision for 'x' reason."
Learn all you can
Thomas Duffy, MD, a professor of medicine at Yale University and an internist, hematologist, and biomedical ethicist, has seen a lot of changes in medicine in his career. The 75-year-old describes himself as "a product of an older age of medicine" and prides himself on his skills in the art of the diagnosis.
He notes that evidence-based medicine can be valuable, but it can only be applied to an entity that has been correctly identified.
"It doesn't help you recognize the entity," he says.
Nothing can substitute for being well-schooled in the art of taking good histories, performing complete examinations, and having a lengthy clinical experience, he believes.
"It takes ample skills to read a patient, see and hear what others do not, and incorporate that into the repertoire of clinical entities to which we have been exposed," he says. "We don't diagnose entities that we haven't seen before."
Unfortunately, he believes that with physicians now seeing fewer patients during their training, they are encountering fewer unusual entities. This lack of experience can lead to delays in accurate diagnoses.
"We could save so much by creating a population of physicians who are truly well-studied, people who can sit and think without ordering tests," he says.
Duffy's advice to the younger generation? "Be rapacious in milking all that you can from experienced physicians," he says. "There are many things you can't learn in a textbook."