When Doctors Tell Patients What They Don’t Want to Hear

There’s an episode of “The Mindy Project,” Mindy Kaling’s comedy about the life of an obstetrician, that begins in the office of an attractive ob-gyn, Dr. Reed. He sits, beaming, in front of his pregnant patient and her husband. He wears a crown of jewels they have given him, and they exchange pleasantries until the inevitable moment arrives: they need to address the patient’s health. She is obese, and her weight poses many risks to the fetus.

Unwilling to jeopardize the affection of his “favorite patient,” Dr. Reed instead summons the brazen and socially inept Dr. Mindy to do his dirty work. True to form, by the end of the scene, Mindy has offended the patient, which escalates into a shouting match until the patient tells Mindy that she’s the one who needs to lose some weight. Reed emerges, halo intact.

Though the scene marks a bad day for Mindy, I think it also heralds what could turn out to be a bad era for American medicine. Beyond the comedic exaggerations lies an age-old tension: Will our patients still like us if we tell them things they don’t want to hear? The challenge of communicating unpleasant, possibly profoundly upsetting information to patients is timeless. What has changed, however, is that physicians are now being judged, and compensated, based upon their ability to do it.

In October, 2012, Medicare débuted a new hospital-payment system, known as Value-Based Purchasing, which ties a portion of hospital reimbursement to scores on a host of quality measures; thirty per cent of the hospital’s score is based on patient satisfaction. New York City’s public hospitals recently decided to follow suit, taking the incentive scheme one step further: physicians’ salaries will be directly linked to patients’ outcomes, including their satisfaction. Other outpatient practices across the country have also started to base physician pay partly on satisfaction scores, a trend that is expected to grow.

But in a country that spends more per capita on health than any other, with results that remain mediocre in comparison, can we really expect that a nation of more satisfied patients will be a healthier nation over all?

Many insist that we can. One of the leading arguments for pay based on satisfaction, as described in a recent Wall Street Journal article titled “The Talking Cure for Health Care,” is that these incentives will improve patient-doctor communication, which will in turn lead to better health. As the article notes, “Doctors are rude. Doctors don’t listen. Doctors have no time. Doctors don’t explain things in terms patients understand.”

Few object to these generalizations. We’ve all had insensitive doctors who have left us confused and scared. I’m a physician, and I often find myself rushing, interrupting, and overwhelming patients with information. But if the path from good communication to better health is through a better understanding of risk factors and disease, then medicine poses a paradox: how much we understand tends to be inversely related to how well we think physicians have communicated.

Consider, for example, a recent study among patients with chronic kidney disease: the more knowledge patients had about their illness, the less satisfied they were with their doctors’ communication. Another study’s title asks, “How does feeling informed relate to being informed?” The answer: it doesn’t. The investigators surveyed over twenty-five hundred patients about decisions they had made in the previous two years, and found no over-all relationship between how informed patients felt and what they actually knew.

The disconnect between patients’ understanding of disease and their satisfaction with physicians is particularly pronounced for care at the end of life. In a recent study published in the New England Journal of Medicine, oncologists studied patients’ expectations of chemotherapy options. For these patients, with either end-stage colon or lung cancer, chemotherapy may provide some help, but it can also be toxic, and definitely doesn’t provide a cure. Doctors know this, but do patients?

In the study, sixty-nine per cent of patients with lung cancer, and eighty-one per cent of patients with colon cancer, did not understand that chemotherapy was not curative. This finding reminds that we have much to learn about how to communicate medical information to our patients. But it is the second finding that suggests why paying based on patient satisfaction isn’t the way to get us there: the more people understood about the grim nature of their prognosis, the less they liked their physicians.

Understanding that there is no cure for your disease is entirely different from understanding why you need to take a blood-pressure medication. Since I suspect that a bit of denial is precisely what allows the dying to live—see the response of a young, pregnant woman to the news that she has incurable lung cancer in Atul Gawande’s “Letting Go” for a beautiful example—I tend to be more concerned with how to keep people from getting sick in the first place.

And this gets us back to the Mindy problem. Sure, there are nice ways of saying, “You need to lose weight, stop smoking, and take this medication that certainly won’t make you feel better but might very well leave you tired and depressed.” But sometimes there aren’t, and it can be tough to separate how we feel about the message from how we feel about the messenger.

I used to be an avid runner, but have had a slew of running injuries—the most enduring of which is a chronic hamstring problem that has made sitting uncomfortable, and running impossible. But for a long time, my approach to any given injury was simple: run through it.

In my quest for quick fixes, I have seen more orthopedists than I can count. But there was one doctor, Dr. D., who tried to teach me the error of my ways. He told me that the problem was not with my body but with my behavior. He said I didn’t need MRIs or steroid injections but rather to stop running and give myself time to heal. And I, in turn, found much that was wrong with him: he started late, didn’t return phone calls, had bad breath, typed with one finger, and, above all, didn’t seem to listen to me. I decided he was the worst doctor in the world and went searching for a new one.

Many months and doctors later, last year, I found “my person.” Most important, she told me I would run again. That she was so nice, so pretty, and so put together (and she injected my aching gluteal region with steroid every time I asked) only reinforced my sense that I was in the most expert of hands. I loved her as much as I wanted to be her.

If you had mailed me a satisfaction survey, you can imagine which doctor would have gotten a bonus. But in the end, it’s Dr. D who was right. I still can’t run, but had I heeded his advice, I’d likely be back to doing marathons.

The problem with the patient-satisfaction surveys is that they assume we can evaluate specific characteristics of doctors, or hospitals, as distinct from their general likability. But that’s not easy. The halo effect is a well known cognitive bias that describes our tendency to quickly judge people and then assume the person possesses other good or bad qualities consistent with that general impression. The effect is perhaps best described in the many positive attributes we ascribe to someone we find attractive. As the Nobel laureate Daniel Kahneman noted, for example, “If we think a baseball pitcher is handsome and athletic … we are likely to rate him better at throwing the ball, too.”

This tendency has been well demonstrated in our judgments of the competence of political candidates, or our willingness to assume innocence for someone accused of a crime. (See Paul Bloom’s post on the unwarranted empathic response to the attractive face of the Boston Marathon bomber Dzokhar Tzarnaev.) Though there are several factors informing the general likability of physicians beyond how we feel about what they tell us, there is no reason to assume we would be somehow immune to this cognitive bias when it comes time to rate them.

Although we tend to be totally unaware of the effects of these haloes on our own judgments, hospitals and outpatient practices are not. That’s why they are investing millions of dollars in renovated rooms, new foyers, gourmet chefs, and valet parking. These are nice perks, and undoubtedly lead to higher scores across all domains of the satisfaction survey.

But do higher scores on a satisfaction survey translate into better health? So far, the answer seems to be no. A recent study examined patient satisfaction among more than fifty thousand patients over a seven-year period, and two findings were notable. The first was that the most satisfied patients incurred the highest costs. The second was that the most satisfied patients had the highest rates of mortality. While with studies like this one it is always critical to remember that correlation does not equal causation, the data should give us pause. Good medicine, it seems, does not always feel good.

Lisa Rosenbaum is a cardiologist, a Fellow at the Philadelphia V.A. Medical Center, and a Robert Wood Johnson Foundation Clinical Scholar at the University of Pennsylvania.

Photograph: Fox