Published in the May 2015 issue of Today’s Hospitalist
WHY DO PHYSICIANS ORDER TESTS that aren’t supported by evidence? The conventional wisdom holds that in part, the problem stems from the fact that physicians simply can’t keep up with evidence. That has led to a lot of handwringing about how to best educate physicians, primarily through clinical guidelines.
But a recent study of how hospitalists order tests for two common conditions found that even when physicians think they know the evidence behind testing, most will choose to order tests even if the clinical evidence deems them unnecessary. The reasons why have more to do with making patients and family members happy than adhering to the science.
The study, which was published in the Jan. 20 Annals of Internal Medicine, found that a group of about 1,000 hospitalists thought their colleagues would order unnecessary tests at least 52% of the time during routine preoperative evaluations and at least 83% of the time when treating syncope. Tests were defined as unnecessary if clinical guidelines didn’t call for their use.
While those results may have been unexpected, the reasoning behind that overuse came as an even bigger surprise to researchers. The biggest single reason for over-ordering, hospitalists said, was the desire to reassure patients and family members that they are receiving top quality care.
Lead author Allen Kachalia, MD, JD, a hospitalist and chief quality officer at Boston’s Brigham and Women’s Hospital, says that participants were fairly clear about why they thought tests were being overused. “Physicians told us they were often ordering tests to be safe, to reassure patients or family members,” he explains. To a lesser extent, Dr. Kachalia adds, physicians also said that tests were ordered “to make themselves feel comfortable that there was no problem or further evaluation needed.”
Assessing their colleagues
Dr. Kachalia, who is also associate professor of medicine at Harvard Medical School, says his team tackled the issue of overtesting as one way to explore the reasons doctors do not adhere to clinical guidelines. “Physicians often say that they feel pressure to order tests despite the clinical evidence,” he explains. “We wanted to get a sense of whether or not that happens in the hospital setting.”
The research team asked hospitalists to predict how often their colleagues would order tests for syncope and preoperative evaluation. The survey group of hospitalists, who had on average practiced for nine years, reported that overuse of tests was common.
Why didn’t researchers ask hospitalists how they would personally order tests? “We worried that physicians might be less comfortable reporting their own practice patterns,” Dr. Kachalia says.
When researchers asked why hospitalists would overuse testing, 37% of respondents said that scientific evidence would drive their colleagues to overuse. According to Dr. Kachalia, that could mean that physicians were aware of the guidelines and disagreed with the evidence, or they had a misunderstanding of the guidelines.
But an even bigger group “51% “said hospitalists would overuse tests to reassure patients or family members or themselves. For preoperative evaluation, physicians said that 28% of unnecessary tests would be ordered to reassure patients and family members. For syncope, that number came in even higher, at 43%.
Dr. Kachalia hypothesizes that testing overuse may be more common for syncope because the reason for the syncope may be unclear. “When someone passes out,” he explains, “it’s hard for us as physicians to determine the cause. That uncertainty may lead to greater overuse.”
To further explore why unnecessary tests are ordered, researchers created multiple versions of clinical vignettes. In the preoperative evaluation scenarios, for example, hospitalists said that 58% of their colleagues would overuse tests. When the vignette contained no details on the patient’s family members, that number was 52%. But when the vignette was changed to indicate that the patient’s son was a physician who specifically asked for testing, the number jumped to 65%.
While syncope testing overuse was more common, vignette changes produced less variation. When the syncope vignette offered no details about family members’ occupations, overuse of testing came in at 83%. Even when the patient’s wife was identified as an attorney who asked for testing, overuse jumped up only a little, to 85%.
How do liability fears factor in? In the preoperative scenario, researchers noted that overuse increased when legal risk could be perceived as higher because a family member was requesting a test. It’s possible that defensive medicine was on physicians’ minds.
And interestingly, inappropriate testing was used not only to reassure patients and family members. In 23% of preoperative cases and 15% of syncope cases, hospitalists thought their colleagues would overuse testing to reassure themselves that they were making the right decision.
But perhaps the bigger news was that there was no smoking gun in the study linking liability experience and testing overuse. Researchers concluded that the number of years a hospitalist had practiced or a personal history of being sued didn’t have any effect on whether hospitalists would be likely to over-order tests. Researchers had wondered if physicians practicing for many years would be more worried about malpractice risk, but study data didn’t back that up.
There was one area where liability concerns may have affected test use. VA physicians were less likely to overuse tests (39%) than other hospitalists (59%). “Our guess is that these data might have to do with the fact that VA doctors can’t personally be sued,” says Dr. Kachalia. “So the environment in which they practice might be less legally threatening.”
Dr. Kachalia points to another interesting result: The study shows that even in settings in which physicians have little to no financial incentive, they still overuse tests.
Finally, he notes that physicians’ use of testing to reassure patients and family members points to an even bigger issue: how hospitals can balance the growing focus on patient satisfaction scores with the drive for evidence-based medicine.
“We’re going to see what happens as we put more emphasis on patient satisfaction,” he says. “What are we going to do when patients ask for something that is not in line with the evidence? How are we going to make sure we get their experience right as we seek to properly address requests for more care?”
Edward Doyle is Editor of Today’s Hospitalist.