Conventional wisdom about physician expertise generally holds that the longer a physician has been in practice, the better honed his or her clinical skills become. But a new study turns that adage on its head with its conclusion that physicians who have been in practice longer may, in fact, provide lower quality of care.
The problem is not that more experienced physicians are not intelligent, well-trained or competent, and the findings "should not be used to put the blame on older physicians," said study lead author Niteesh K. Choudhry, MD, an instructor of medicine at Harvard Medical School in Boston, Massachusetts. Instead, "the problem is that we've been relying on methods that may not be very effective in requiring that physicians keep up to date."
To address that gap, the traditional methods of keeping physicians' knowledge and clinical practice current need to be reexamined and altered, according to the study. "Over the past 15 or 20 years, there has been a body of literature generated that helps to identify what techniques work and don't work," Dr. Choudhry told Medscape in an interview. "We should refocus our efforts for recertification and relicensing on that." The study linking physician years in practice to quality of care, published in the Feb. 15 issue of the Annals of Internal Medicine, was based on a Medline review of 59 articles from 1996 through 2004 that reported data on 62 groups of outcomes. Studies were included if they were original reports that provided empirical results; measured knowledge, guideline adherence, mortality, or some other type of quality-of-care process; and included years since graduation from medical school, years since certification, or physician age as a potential variable influencing results.
The studies were then categorized into four groups on the basis of whether they evaluated knowledge (such as knowledge of indications for blood transfusions); adherence to standards of care for diagnosis, screening, or prevention (for example, following preventive care guidelines); adherence to standards of care for therapy (such as appropriate prescribing); or health outcomes (mortality, for example).
Results were then classified into six groups on the basis of the association between the length of time a physician had been in practice and his or her performance: consistently negative, partially negative, no effect, mixed effect, partially positive, and consistently positive.
Overall, more than half (52%) of the 62 evaluations captured in the studies showed a negative association between increasing experience and performance for all outcomes assessed, and an additional 21% showed a negative association for some outcomes and none for others.
The study also found that only two evaluations (3%) reported that performance initially increased with longer experience, peaked, then decreased, while 21% found no association. One evaluation (2%) reported increased performance with longer experience in practice for some outcomes, but no association for others, and only one evaluation (2%) found increased performance with more years in practice for all outcomes.
In the area of assessing the knowledge of practicing physicians, for example, all of the 12 studies used reported a negative association between knowledge and increased experience, the study found. For example, after adjusting for specialty and other variables, physicians younger than 40 years were more likely to believe in the value of established therapies that improve survival rates for acute myocardial infarction (AMI), such as thrombolytic agents, aspirin, and beta-blockers, and less likely to believe in the value of therapies that have been disproved, such as prophylactic lidocaine, the study found.
The methodology used in the AMI study was "particularly well done," said Dr. Choudhry, controlling for variables that often can skew a theory about experience and competence. "What this study did was it controlled for a series of patient and physician factors that would have confounded the relationship," he said. "It's hard to dismiss the results out of hand."
The study results call into question the benefit of relying exclusively on continuing medical activities, such as attending lectures or reading journal articles, as a way to remain current in knowledge and practice, Dr. Choudry said. Such methods require only a "passive participation in acquiring knowledge." What is needed instead are activities in which "physicians are interacting with other physicians or using information technology" to measure or build on their clinical skills, he said.
So-called "academic detailing," in which physicians or pharmacists visit physicians' offices to discuss appropriate treatments has proven successful in areas where it has been tested in Canada, Dr. Choudhry said. "One of the big messages of the literature about behavior change is that we need to use more than one thing."
An accompanying editorial cautions, "The profession cannot ignore this striking finding and its implications: Practice does not make perfect, but it must be accompanied by an ongoing effort to maintain competence and quality of care."
To that end, the editorialists representing the American College of Physicians and the American Board of Internal Medicine recommend that professional development apply to a physician's entire career, not just to those in the early professional years.
In addition, the editorial called on physicians to support the concept behind the American Board of Medical Specialties' "maintenance of certification," which replaces one-time certification for life with ongoing performance measurement. "This model of professional development provides a way to identify gaps between current and ideal practice, which is the first step toward acquiring needed new knowledge, skills, and processes of care," the editorialists write.