Tweet This
My father is 92 years old, and I am beginning to wonder whether the best thing for his health would be to stay away from doctors. That’s because well intentioned physicians often expose their elderly patients to harmful and unnecessary services out of habit. That’s certainly the message I absorbed after reading a recent issue of JAMA Internal Medicine that published three studies documenting the worrisome frequency with which internists like me over-test and over-treat our patients. I am going to briefly describe these three studies before laying out some ideas about what’s going on here.
One study explored the use of PSA screening among men with limited life expectancy. The PSA blood test is used to screen men for prostate cancer. The test is controversial, with some groups saying there is no evidence it benefits anyone and others saying it is a crucial way to reduce prostate cancer deaths. Despite this controversy, almost everyone agrees that when people have limited life expectancy–when, because of age and other illnesses, they probably have fewer than five years to live–the PSA test does more harm than good. But some physicians nevertheless continue to order PSA tests, even in men close to the end of their lives.
The study, which analyzed data from Veteran’s Affairs medical centers, found out that patients receiving care from “attending physicians”–more senior physicians–were more likely to receive harmful PSA tests than patients receiving care from physicians still in training. Indeed, 40% of patients expected to live five or fewer years received PSA tests from experienced physicians, versus only 25% receiving care from trainees :
The second study looked at carotid artery imaging in people 65 years or older. The carotid arteries are the large vessels on either side of your neck, the ones you can feel your pulse on. They are the main supply of blood to the brain. People who get blockages in their carotid arteries are at risk for strokes.
Carotid imaging with tests like ultrasound can identify narrowing of these important arteries, potentially revealing partial blockages in time to fix them before they fully occlude. In the old days, I’d place my stethoscope on a patient’s neck to listen to the harsh sound of blood squeezing its way through these blockages. Upon hearing a worrisome whoosh, I’d send my patient for imaging and then, if my suspicions were warranted, would refer the patient to a neurovascular surgeon, who would decide whether to perform a procedure to open up the artery.
But now, we physicians are being told to be more cautious. The benefits of all these tests and treatments aren’t so clear in many patients. The risks of the surgery can outweigh the benefits in people with no history of stroke or stroke-like symptoms. Nevertheless, many physicians continue to test and treat aggressively.
In fact the second study, also conducted in the VA setting, showed that the vast majority of carotid imaging studies are conducted in patients where there is no evidence the test will benefit them, and a substantial minority–one in nine–are conducted in patients for whom the test brings more harm than good:
Okay, one more quick study: This one looks at patients who receive aggressive testing or treatment for coronary artery disease, narrowing of the arteries supplying blood to the heart. Specifically, the study looked at the practice of testing for coronary artery disease in patients scheduled to undergo non-heart-related surgeries. When I was trained, I was taught that when patients underwent, say, hip surgeries, the procedure put their bodies under lots of stress. Sometimes, in fact, these patients would experience heart attacks while undergoing the procedures. To avoid these awful events, I was taught to conduct a thorough pre-operative evaluation, including a history and physical exam, an electrocardiogram and, when anything looked suspicious, perhaps even a cardiac stress test (where the patient exercises on a treadmill and we look for signs of heart problems.) Then, when patients revealed evidence of heart disease, we’d fix that problem before the hip surgery, assuming the hip procedure could wait.
Today, the best evidence suggests that such preoperative tests and treatments often do more harm than good. (Are you recognizing a theme here yet?) Some people have false positive stress tests and undergo potentially hazardous tests with no benefit to treatment. Some receive aggressive “revascularization”–surgery or catheter-based treatments to open up their arteries–and these procedures, too, can be risky. The third study showed that aggressive testing and treatment is still common and is most often conducted in people with no symptoms of heart disease, patients for whom there is no evidence that these interventions benefit them.
So what’s going on with all these questionable tests and treatments? It’s not just about the money.
Two of these three studies looked at care in Veterans Affairs medical centers where, if anything, there’s an incentive to provide less aggressive care because of budget constraints.
It’s not about lack of effort to reduce such practices. The VA has been trying to reduce unnecessary tests and treatments for years. Many medical specialties have joined the Choosing Wisely campaign, and have targeted some of the interventions explored in these studies as clinical practices that should be curbed. Admittedly, we could step up all these efforts. But these practices are persisting even in the face of these efforts and in the absence of financial incentive.
So again: What’s happening? It starts with habit, as evidence by the PSA study–where older doctors were more likely to order the test than younger ones. It’s hard to get physicians to stop doing what they’ve gotten used to doing. On top of that, it’s hard for us doctors to shake the idea that more is better–that finding cancer early must be better than finding it late, that opening up clogged arteries must be better than leaving them clogged. Finally, we physicians often don’t receive good information on which patients we should be more or less aggressive in treating. Our electronic medical records don’t alert us that our patient is too old and sick to benefit from the tests we are ordering.
There is no easy fix to all this over-testing and over-treating. Clinicians should work with administrators and EMR vendors to build feedback systems into clinical care processes, ones that don’t untowardly interfere with physician discretion or with the efficiency of seeing patients in the course of busy practices. We need to go beyond identifying which clinical practices we need to avoid and go on to develop systems that help us and our patients avoid them.
