Sam was a healthy man in his sixties who came under the care of a cardiologist colleague. He didn’t smoke, didn’t drink, and wasn’t taking any medications. In fact, he was a world-ranked athlete for his age group in a competitive sport. As part of his compensation package at a powerful firm in downtown Toronto, he went every year for an executive physical. Executive physicals are often offered to business leaders as perks—they visit fancy private clinics with carafes of cucumber water in the waiting room and undergo a whole series of tests that wouldn’t normally be offered in the public system because there is no evidence to support their use. Many of the tests and the doctors’ services are billed to the public system, but employers pay high fees for the components of the service not covered by the public plan.
One year, despite the fact that Sam felt perfectly well, he was subjected to an exercise stress test, “just in case”—it was offered as part of the executive package of services. Some incidental potential abnormalities were identified. He ended up with an angiogram—an invasive test where dye is injected into the patient’s blood vessels to look for blockage of the arteries that supply the heart. Happily, the angiogram confirmed that Sam did not have coronary artery disease. But not before he suffered a stroke on the table, a known complication of the procedure that occurs once in every one to two thousand cases. This healthy athlete will never play his sport again because he’s paralyzed on one side of his body as a direct result of a completely unnecessary test.
You may be thinking “But what if they’d found something?” and wondering whether the risks of screening tests aren’t still worth it to pick up that one cancer or undiagnosed heart condition. Our culture is so shaped by the assumption that all knowledge is helpful that people—both patients and health care providers—find it hard to accept that it can cause more harm than good. We need to stop thinking only about the potential benefits of medical tests and interventions and start talking about their potential harms.
Expensive technology, early diagnosis, and aggressive treatment can save lives—but only when properly applied. When misapplied, they cause suffering and death, not to mention unnecessarily high health care costs. It’s time for us to learn the difference between medicine that helps and medicine that harms. That’s why the third Big Idea is the reduction of unnecessary tests and treatments in Canadian health care.
Most doctors go into medicine because we want to cure illness. That means we like to do stuff that makes people get better, preferably right away. When I train residents and medical students, I sometimes see their eyes glaze over when I start to talk about things like “watchful waiting”—the art of standing back to see what the body will do about a symptom if given a little time. Or I try to describe the “healing interaction,” which is the use of a brief encounter, conversation, and education to help someone through the experience of their ailment rather than ordering tests or prescribing medicines…and I wonder if they think I’m a flake. Medical trainees are high performers, and they’re usually action-oriented. They don’t enjoy uncertainty (who does?), so it’s natural for them to want a clear diagnosis and treatment plan. Low-tech, relationship-based approaches don’t always feel like active options, even though they are. This often means that by the end of an encounter with a patient, many trainees want to order a test, perform a procedure, prescribe a medicine, or do some other action that helps them feel as though they can tie a bow around the interaction and consider the patient’s concern “dealt with.” In fairness, this isn’t a perspective that ends on graduation—though for many doctors it does diminish over time.
Patients, too, often consider tests and interventions to be the mark of a thorough doctor, a doctor who cares and takes their symptom or condition seriously. During my fourth year of medical school I did a month-long rotation in a rural Tanzanian village that had few health care resources. I spent many afternoons in a small, dusty clinic with two rooms and a dirt floor, learning the ropes of tropical medicine from a pair of talented and committed health workers who had little formal education but enormous knowledge of the diseases affecting their community. The first piece of advice they gave me was this: if I wanted patients to feel that I’d helped them, I needed to give them an injection. This was what good doctors did—injections were seen as strong medicine, medicine that would heal.
Lest you find that quaint, let me assure you that in North America our thinking is no different. Good doctors put needles into skin and probes into orifices to ferret out and eliminate the problem. The more high-tech the investigations and the more invasive the solution, the better the medicine is seen to be. Why get an X-ray if an MRI is available? Why advise a patient to lose weight to reduce her knee pain from arthritis when you can prescribe a painkiller instead?
Yet often the very best thing we can do for our patients is to listen—and based on their history and our physical exam, to figure out the most likely diagnosis. We can offer suggestions for improvement that aren’t drug related, and give things time to either heal on their own or declare themselves to be serious. This is sometimes misconstrued by both doctors and patients as “doing nothing.” In fact, physician and patient are doing something—but that something isn’t unnecessary tests and treatments.
I’ve learned this lesson over and over in my career, most poignantly in the birthing unit, where the harm of too much intervention is well documented. For example, we know that when fetal heart rates are continuously monitored while a woman with a low-risk pregnancy is in labour, the likelihood of ending up with a Caesarean section increases. This surgery has risks for both the woman (such as bleeding, infection, and damage to the surrounding organs) and the baby (such as higher rates of admission to the neonatal intensive care unit). What seems like “playing it safe” by closely monitoring the baby’s well-being actually leads to measurable harm.
When I was a new recruit doing obstetrics at Women’s College Hospital, there was a placard on the wall of the nursing station bearing a quote from one of the hospital’s legendary obstetricians: “Don’t just do something, stand there!” It was a wise reminder that when it comes to birth, the hardest and often the best thing is to resist the desire—the compulsion—to intervene.
What does it mean to “stand there”? One of the most important things we do in medicine is to accompany patients through their experience. This is central in primary care, but it’s true in every single part of the health care system. Members of your health care team should educate you about the nature of your disease and what to expect; seek to understand the impact of your symptoms on your ability to function; listen to your concerns about the future; plan the next steps together; and then continually monitor you to see if things change, which might then indicate the need for a test or treatment. We should accompany you watchfully—without exposing you to undue risks. Sometimes that takes longer than ordering a test or prescribing a drug, but it’s the only way to achieve health care’s Triple Aim of improving your health and experience of care while keeping costs from rising.
One of my colleagues who practises palliative care thinks the phrase “doing nothing” is misleading. As he pointed out, many people falsely believe that going to palliative care means doing nothing, yet the nurse-to-patient ratio on a palliative care unit in his hospital is as high as it is on the internal medicine floor, if not higher. “We monitor different things,” he explained. “We don’t monitor blood pressure but we do monitor pain, for example. And we do lots of stuff for people, such as aggressively manage secretions and treat their pain.” He’s found that it’s important for family members to understand this, because it alleviates the concern that loved ones won’t be cared for once they’ve transitioned to palliative care. Like the decision not to test or treat, he told me, that transition “involves a detailed, often time-consuming conversation, and a commitment to follow the patient closely and adjust decision making over time.”
I’m not trying to sell you on the magical ability of the body’s self-healing process. There is ample scientific evidence that overexposure to tests, treatments, and interventions can do more harm than good; sometimes our patients would be better off without our “help.” Consider, for example, the explosive finding in a 2015 study published in The Journal of the American Medical Association: patients admitted to teaching hospitals with high-risk cardiac conditions were less likely to die if they were admitted to hospital when many of the cardiologists were out of town at the annual national cardiology conference. Why? One possible explanation is because when the doctors were away, the patients were less likely to be subjected to high-risk interventions.
How have we ended up in a situation where we’re too frequently exposing patients to harmful interventions? There isn’t a single cause, nor will there be a single solution. Factors that have contributed to the rise of overtesting and overtreatment include the redefinition of disease, the leverage of corporations, the culture of medicine, the fear of litigation, the way we train our doctors, and the influence of patient expectations—to name a few.
When I was in training, in order to be considered diabetic a person’s blood sugar level had to be consistently above 7.0 after fasting for twelve hours. In the early 2000s, a new disease known as “pre-diabetes” emerged, which includes people who have fasting blood sugars above 6.0. Today, the most recent guidelines suggest that even people with sugars below 6.0 are at increased risk, effectively creating a group of pre-pre-diabetics.
Being “pre-sick” doesn’t lead to disability or death; the worst outcome is to eventually become actually sick (and this doesn’t happen to all pre-sick people by any stretch). Nonetheless, in recent years we’ve seen the lowering of diagnostic thresholds for diseases like diabetes and high blood pressure and the emergence of “pre-disease” diagnoses.
It may be good to know that you’re “pre-sick” if it helps you avoid becoming actually sick. But in many cases we don’t reduce that risk—we just give you a label earlier. That label inevitably leads to more tests, and, in some cases, even treatments with side effects and unclear benefits. In my practice, I find it tough to navigate the discussion with patients who are in this “at-risk” category. How can a borderline result help motivate people to make lifestyle changes without frightening them with a label that doesn’t have a clear meaning?
From a population perspective, small changes to testing thresholds for diseases and conditions can have the effect of turning millions of people into patients. When the definition of what constitutes “abnormal” cholesterol levels changed in the United States in 1998, forty-two million new cases of high cholesterol suddenly came into being. That’s more than the entire population of Canada.
Of course, eating well, exercising regularly, getting enough sleep, maintaining a healthy body weight, not smoking, and not drinking would prevent a huge number of diseases and many pre-sick people from getting actually sick. But those lifestyle changes are hard to make, and harder to sustain. So it begins to feel inevitable that the health care industry will push for earlier treatment with drugs instead of focusing on non-drug approaches.
More concerning is that in the case of pre-dementia, pre-diabetes, or pre-hypertension, the current benefits to patients from earlier diagnosis and treatment are often marginal. There are exceptions, of course: Pap smears to identify pre-cancer of the cervix and genetic testing for patients at high risk of developing certain types of breast and ovarian cancer are two examples. These are the exceptions that prove the rule.
Even if it doesn’t improve the health of patients, treating pre-diseases can improve the bottom line of the companies that develop the treatments. A recent study looked at expert panels who were proposing changes to disease definitions that would increase the number of individuals considered to have them. The study found that none of the panels rigorously assessed the potential harms that could result from widening these definitions. And, significantly, most of the panels had a majority of members with financial ties to pharmaceutical companies.
Industry influence extends beyond disease definition. Marketing to both physicians and patients has reached new heights, and is carefully considered and aggressive. Why? Because marketing interactions and relationships exert influence on the prescription pad.
Direct advertising of prescription medicines to consumers is prohibited in Canada. But anyone who watches American television—which is most of us—has seen the ads telling people to “ask your doctor” about x, y, or z “to see if it’s right for you.”
Advertising to patients is effective. Consider an American campaign for migraine medicine that focused on encouraging patients to talk to their doctors about new treatments. When the results of the campaign were reviewed, it was found that if extrapolated across the entire U.S. population, the campaign would have generated around $11.5 million in new prescriptions and almost as much for refills.
It isn’t just the pharmaceutical industry that’s shaping the culture of overtesting and overtreatment. The fear of malpractice suits plays a role as well, with tests and treatments used as a form of self-defence. The term “CYA medicine” isn’t in the medical dictionary, but I probably learned it in my first week of medical school. It refers to “covering your ass”: the act of ordering a test or doing a procedure not because you think the patient needs it, but to avoid being sued if it turns out later that you made the wrong call.
Canadian doctors pride ourselves on practising less CYA medicine than our colleagues in the U.S., where the culture of litigation and the risk of huge medical malpractice suits are extreme. Nevertheless, we can all think of times when we’ve succumbed to the fear of being sued rather than using our best clinical judgment. And if the fear of litigation isn’t in play, there’s always the fear of a complaint or even just disappointing a patient. Colleagues who work in the emergency department often talk about how hard it is to send patients home without ordering such imaging as X-rays, CT scans, or ultrasounds. A person comes in with a twisted ankle. You examine them. You know it isn’t broken. We even have a set of “ankle rules” to help avoid inappropriate imaging in this situation. But that person has been sitting in the waiting room for four hours because they want to be sure their ankle isn’t broken. They start to look pretty irritated when you suggest that an X-ray won’t add much. And if you miss something…So you order the X-ray.
Our approaches to training physicians can reinforce both defensive medicine and a tendency to underestimate (or simply not consider) the potential harms of our actions. Medical trainees are frequently assessed based on their ability to conduct rigorous diagnostic assessments that consider every possible diagnosis—“What are the twenty-two potential causes of this woman’s low platelets?”—but not necessarily on their ability to weigh the costs involved in those assessments. (And by costs, I don’t just mean the financial ones.)
The way we pay doctors can also reinforce a culture of too much medicine. We’ve seen that most Canadian doctors are still paid—at least in part—on a fee-for-service basis, or piecework. In most cases, we don’t get paid for ordering a test or prescribing a medication, but when it’s faster to do so than to engage in thoughtful conversation with a patient, the fee-for-service model has the potential to drive unnecessary tests and prescriptions. Sometimes the link is even clearer: when physicians are paid for performing a test themselves, as is often the case for some heart tests and lung tests, for example, there’s a financial incentive to perform more of those tests. And of course, in procedural specialties like most forms of surgery, the doctor who recommends the treatment will also be paid to perform it.
The risk of “supply-induced demand” in medicine—whereby having a lot of doctors around who need to earn a living generates a rise in intervention rates—is hard to avoid when we consider what leads to overtesting and overtreatment.
An American study that looked at the frequency of stress tests for cardiac patients who were recovering from angioplasty illustrates this point nicely. Current guidelines do not recommend stress tests within two years for patients who’ve had an angioplasty unless they have symptoms. Nonetheless, many patients are subjected to these tests. If the doctors could bill fees for the tests, their patients were significantly more likely to receive them. In other words, people ended up having more tests—some of which involve radiation—when their doctors had a financial incentive to order or perform them. Of course, these are rarely conscious decisions. Physicians are nearly always trying to do what they truly believe is best for their patients. But incentives shape human behaviour just as much in medicine as they do in any other job.
And then, of course, there’s the patient.
In his iconic novel The House of God, read by medical trainees all over the world, Samuel Shem depicts a group of new medical residents who hit the wards in a big hospital and are indoctrinated into the practice of medicine through punitive schedules, sleepless nights, and a cynical view of patient care. The book is a caricature of an extreme universe where the higher calling of medicine is sacrificed in a daily grind of exhaustion, terror, and self-preservation—and it contains just enough truth that trainees often recognize the worst parts of themselves in it. Among the “Laws” passed down from senior to junior trainees is the notion that “the delivery of medical care is to do as much nothing as possible.” As the senior resident explains to them, “It ain’t easy to do nothing, now that society is telling everyone that their body is fundamentally flawed and about to self-destruct. People are afraid they’re on the verge of death all the time.”
Patient demand is a factor in overtesting and overtreatment. Of course, tests and treatments are ordered by health care providers, not patients. When we do a decent job of communicating the evidence about their harm, most reasonable people don’t want unnecessary interventions. But in the era of internet chat rooms and expectations that technology can work miracles, individuals are more likely to ask for specific tests and treatments than ever before. This is especially true if they have a symptom without a clear cause, or a nagging fear in the back of their mind.
Living with uncertainty is hard, but when we try to eliminate it, there are real costs. People like Sam, the healthy executive in my cardiology colleague’s practice, have learned this the hard way. Sam’s unnecessary stress test yielded a false positive result—the test suggested a problem that he didn’t actually have. This led to a cascade of further investigations, culminating in an invasive procedure that yielded disaster.
Of course, not all tests and treatments are unnecessary. But overtesting and overtreatment are significant problems in Canadian health care.
The most fundamental precept of medicine is to avoid harming the patient. But we spend very little time in our training talking about how that principle plays out in the little decisions we make every day.
The Canadian Association of Radiologists has estimated that 30 percent of diagnostic imaging tests ordered may be inappropriate or contribute little to no useful information to the management of the patient’s case. It’s hard to know if this estimate is accurate, but safe to say that a large number of imaging tests done in Canada are unnecessary. The value of testing also needs to be balanced with the risks. While radiation exposure from exams like low-dose CT scans is small, the general consensus is that even low-level radiation exposure carries some risk. That risk is often worth taking when the information gleaned from the scan will make a real difference to the patient’s care. But when the benefit of the test is unclear, the risks loom larger. Furthermore, eliminating inappropriate CT and MRI scans would make a huge dent in our wait times for these important tests for people who actually need them.
Achieving the right balance in testing can be tricky. An example I deal with daily is the Pap test, a simple (if uncomfortable) screening test for cervical cancer. The Pap test is one of the best available screening tests in the world of medicine. It involves taking a tiny sample of cells on the cervix to look for cervical cancer, or more often, pre-cancer cells that might eventually become cancer. When picked up early, these cells are virtually 100 percent treatable, so in essence Pap smears aren’t just a form of cancer detection but also of cancer prevention.
In my office, I often see women who’ve gone to significant lengths to make the time to see me in order to safeguard their health. So I understand why they find it confusing—or even upsetting—when I tell them that screening guidelines have changed, and that for most women, yearly Pap tests are no longer recommended. But it does give us an opportunity to discuss why the recommendation has changed to once every three years, and why that’s an improvement.
Why shouldn’t we screen every year for cervical cancer? Less frequent testing is not a cost-control measure. It’s about getting the maximum benefit from screening while minimizing the risk of harm.
Cervical cancer is a very slow-growing cancer. Frequent screening increases the risk of finding mild abnormalities that the body would have healed by itself over time—often, “abnormal” cells on Pap smears disappear on their own over eighteen to twenty-four months without any intervention. The risk of too-frequent Pap smears is that once these abnormal cells are identified, further testing follows. And invasive, uncomfortable procedures such as cervical biopsies (an experience that no woman would want to repeat unnecessarily) bring on lots of anxiety and other physical risks.
In addition, too-frequent testing increases the chances of a false positive result, suggesting an abnormality that doesn’t actually exist. The chance of a false positive on a single Pap smear lies somewhere between 1 and 10 percent. Of course, the more Pap smears a woman has, the more that risk multiplies. If she has a Pap smear every year between the ages of eighteen and seventy-eight, assuming a 5 percent incidence of false positives, she’d have a 95 percent chance of getting a false positive report during that time.
So it’s better for women to screen often enough to pick up cervical cancer when it’s still 100 percent treatable, but not so often that we pick up fleeting abnormalities that would have healed on their own. It’s also better to decrease the risk of a false positive result. For women at average risk between the ages of twenty-five and seventy, our current understanding of the best practice is therefore to screen not every year, but every three years. That recommendation will almost certainly continue to change, now that more and more girls are being immunized against HPV, the virus that causes cervical cancer. In the future, the Pap smear may become a very infrequent test indeed.
Despite these clear recommendations, some doctors are continuing to perform annual Pap smears on women who are at low risk for cervical cancer. Sometimes this is due to force of habit, but often it’s because the doctor, or the woman, or both overestimate the benefit and underestimate the risk of the test.
The “just in case” philosophy is tough to change for all kinds of testing. Take, for example, a study that looked at Ontario patients undergoing low-risk procedures like cataract removal or screening colonoscopy. The risks of these procedures are so minuscule that the vast majority of people shouldn’t have any heart, lung, or blood tests done before the procedures to ensure that they’re “fit.” Instead, thousands of people are having such unnecessary screening tests, which include electrocardiography (ECG), echocardiography, cardiac stress testing, and chest X-rays. Furthermore, the incidence of these tests varies widely: the researchers found a thirty-fold difference between institutions with the lowest and highest rates of ordering tests.
Overtesting isn’t just limited to screening tests. Even for people who are known to be sick, tests do harm. For example, guidelines from multiple cancer societies recommend against imaging to look for the spread of breast cancer in women with early-stage disease who feel well. This unanimous recommendation is based on the fact that the likelihood of spread is low (less than 1 percent) and the chance of a false positive is high. False positive results can lead to more invasive tests, treatment delays for the cancer they actually do have, and increased anxiety.
Despite the guidelines, a recent Ontario study found that out of twenty-six thousand women with early-stage disease, twenty-two thousand (86 percent) had at least one imaging test to look for spread of their early-stage cancer. The researchers concluded that “despite recommended guidelines, most Ontario women with early-stage breast cancer underwent imaging to detect metastases. Inappropriate imaging in asymptomatic patients with early-stage disease is costly and may lead to harm.”
What do patients think about this? Many of my patients have said that they’d rather not be subjected to inappropriate testing; but for others, it seems hard to imagine saying no to a test. According to a newspaper article about the study I just described, one woman diagnosed with early-stage breast cancer was subjected to a chest X-ray, bone scan, and liver ultrasound even though the guidelines are clear that such tests are inappropriate…and she was glad. “I would hate to be the one percent of the population who would’ve benefited from the test and found out two years later that it was a little bit late to start treatment,” the woman was quoted as saying.
The notion that testing can do more harm than good is difficult to accept. Over the last decade, I’ve had three patients leave my practice because I didn’t want to order particular tests that I felt weren’t warranted. In each of those cases, I thought we’d had a good conversation in which we reached common ground—until they left.
Occasionally I see women in my practice who request mammograms—screening for breast cancer—in their late thirties. (Guidelines suggest that women at average risk should start screening with these tests at age fifty.) What would be the harm if I were to order a mammogram for my healthy thirty-eight-year-old patient “just to get a baseline”?
Let’s imagine that I go ahead and order the test, and the results come back suggesting a shadow that now requires further investigation. There are three possibilities.
The first possibility is that the test is a true positive, and my patient has breast cancer. In a person her age, that’s not common.
The second possibility is that it’s a false positive, and that further testing will reveal that the shadow on the initial test wasn’t cancer. But in the process of sorting that out, she’s at risk of harm. This is exactly what happened to Sam, my colleague’s healthy patient whose stress test showed changes that turned out to be a false alarm, causing him harm in the process.
The harms from a false positive can be physical—as what Sam experienced—or psychological. For some people, when they discover that their result was a false positive, the overwhelming feeling is relief and joy—and they carry on with life. But for others a lingering fear remains, causing more health anxiety about a wider range of issues. One study found that even three years after a false positive mammogram result, women continued to be affected psychologically.
The third possibility is that my patient has been overdiagnosed. This refers to the diagnosis of a “disease” that will never cause symptoms or death during a person’s lifetime. With a false positive, the patient doesn’t actually have the disease. In overdiagnosis, the test result is correct but irrelevant.
Overdiagnosis turns people into patients without making their lives better. Instead, it makes them worse.
One review of all the studies on screening mammograms concluded that “for every 2,000 women invited for screening throughout 10 years, one will avoid dying of breast cancer [true positive and effective intervention] and 10 healthy women, who would not have been diagnosed if there had not been screening, will be treated unnecessarily [overdiagnosis]. Furthermore, more than 200 women will experience important psychological distress including anxiety and uncertainty for years because of false positive findings.” Some of these conclusions are based on studies where mammography used older technology, but the complexity of the issues cannot be resolved only with better technology.
In his book Less Medicine, More Health, Dr. H. Gilbert Welch explains the problem of cancer overdiagnosis like this: in the barnyard pen of cancers, the goal is not to let any of the animals escape and become more deadly. This pen contains turtles, rabbits, and birds. The turtles, or non-lethal cancers, aren’t going anywhere any time soon. The rabbits are the potentially lethal cancers that can jump out at any time and may be stopped with an early intervention. The birds, on the other hand, leave the pen in a blink of an eye. Those birds essentially represent cancer that can’t be successfully treated no matter how early it’s diagnosed. The rabbits can still be helped. The turtles leave patients vulnerable to overdiagnosis.
One such turtle is prostate cancer in men. It’s estimated that nearly 60 percent of men over the age of seventy-nine have cancer cells in their prostates. Yet prostate cancer is rarely the cause of death for these men. In fact, the likelihood that testing will turn up a “problem” that would probably never have harmed the patient is as high as 40–50 percent. This doesn’t mean that some men don’t get aggressive forms of prostate cancer. But a huge number of men carry a “cancer” diagnosis that will never cause them any harm, except of course for the harm caused by investigations, monitoring, and the weight of that dreaded word on them and their families. And for men who go on to have treatment for low-grade prostate cancer, the risks of urinary incontinence and impotence may well outweigh the risks of their cancer ever causing them harm.
As we learn more about overdiagnosis, there’s even discussion in the medical community about how we use the word “cancer.” Our tests have become so precise that abnormalities meeting the definition of cancer can be found under the microscope that would never have bothered the person until we went looking for them. Language is a powerful tool that elicits certain reactions, which in turn activate a social—or in this case medical—script. Hearing the word “cancer” understandably evokes anxiety, fear, and a desire to treat—at all costs—the threat of this disease. Yet sometimes that cancer would never harm the person at all. They may be better off not knowing about their “disease.”
The bottom line on medical testing is this: in some situations and for some people, testing can help us to correctly diagnose disease, guide treatment, and save lives. In other cases, testing won’t help the individual, or worse yet, it will cause harm. The challenge is figuring out which camp a particular test falls into for a particular person. This is a matter for medical judgment, not a mechanical formula. To make that judgment, a health care provider and her patient need to be able to have a mature conversation about both risks and benefits.
Patients know this, and they know they’re subjected to overtesting. One in four Canadians report that their doctor has ordered a test or treatment that they considered to be unnecessary. Nearly a third of those people just ignored their doctor’s advice.
One of my close friends from university had a family doctor for many years whom she loved, but who was very quick on the requisition form and the prescription pad. My friend and her husband joked that Dr. Bresh (not her real name) would send them to a specialist or put them in an MRI machine at the drop of a hat. They referred to the experience of going to see her as “getting Breshed,” meaning receiving a huge workup for any minor complaint. Every time they went to see her, they’d have to make a judgment call about which of her recommendations seemed sound and which ones were overkill.
Something needs to change.