Uncertainty in Medicine (2016)

16-27 Uncertainty in Medicine

Nancy Benson: When it comes to our health, we don’t allow any room for error. We often want our doctors to provide us with clear, reassuring diagnoses so badly that we can sometimes forget how challenging it is to actually practice medicine.

Steven Hatch: There’s often an expectation that patients have of doctors that they are supposed to radiate this sort of extreme level of confidence and that there is a direct parallel between your confidence in a diagnosis and your competence as a physician.

Benson: That’s Dr. Steven Hatch, assistant professor of medicine at the University of Massachusetts medical school and author of Snowball in a Blizzard: a Physician’s Notes on Uncertainty in Medicine.

Hatch: Physician competence can actually be related to how the level of confidence that you have in a diagnosis and your willingness to explain when you have a very strong sense that something is going on or when you have a weaker sense and that you’re not quite sure what’s actually happening in a patient. We’re trained in medical school to take all these tests where there’s always one right answer. And you get out into the real world and actually what you discover is that a lot of these tests don’t have a nice simple, clean, one right answer. There are certain instances with medical technology where the better than nothing model actually ends up creating situations in which we can end up doing more harm to patients than good because we end up over diagnosing certain diseases because of the uncertainty that’s inherent in certain technologies.

Benson: One example? In 2009, the U.S. preventative services task force recommended against mammograms for women under the age of 50. Hatch says that this advice reflects the need for patients to weigh risks against benefits.

Hatch: What is the personal risk to an individual patient if they get a test and what are the benefits that they can expect to accrue? Mammograms represent a very difficult problem because we know from the studies that screening mammograms have a life saving benefit for women. There’s pretty good evidence for that. We also have pretty good evidence that there are women who are getting over diagnosed. What we’re far less certain about is how many of those women that get over diagnosed have the extreme forms of over-treatment which would be a mastectomy, radiation and the complications that ensue from mastectomy, radiation and chemotherapy.

Benson: There is a similar dilemma for men when it comes to deciding if precautionary screening for prostate cancer is worth the risk of a false diagnosis. Hatch looked at more than 20 years of data to determine the number of men diagnosed with prostate cancer by a PSA test versus how many actually had it.

Hatch: What we found over about 20 years of data that came in is that you could diagnosis many, many men with prostate cancer by having elevated PSA’s and they would go on to receive diagnoses of prostate cancer and treatments which include prostatectomy and chemotherapy and radiation. And at the end of many years of studying, what they found is perhaps one man in a thousand would end up having their life saved while you could have up to 20 or 30 men who could have prostatectomies for cancers that they ended up not having.

Benson: But if there is so much uncertainty involved in medical testing, why do so many people still view these results as unquestionable? Hatch credits this to the media shaping the public’s perception of medicine.

Hatch: The media sometimes is very eager for it’s own reasons to oversell how good medicine is at things and they overgeneralize findings. And then that leaves patients feeling violated when they come to their doctor’s office thinking they’re going to be cured of some disease and the doctor has, y’know, a more sober reality in store for them.

Benson: However, it’s hard for many people to question the unshakable confidence of some doctors in test results, which may make some patients hesitant to discuss their diagnosis.

Hatch: One of the examples that I will give to them is about stomach ulcers which when I was a kid in the 1970’s, the prevailing thought process among mainstream medicine was that ulcers were caused by too much stress. And so you can imagine being a patient who is stressed out and you have had an ulcer, you go to see your doctor and they say ‘You have a stomach ulcer, you have to relax’ and so the patient goes and relaxes and they go back to their doctor on another visit and they’re still feeling pain. And the doctor sort of looks at them and says ‘You’re still too stressed’ and they say it sort of with a, y’know, ‘I don’t believe you’ in your crazy, sort of paternalistic tone that many of us have experienced in medicine. And now, nobody thinks of ulcers as being related to stress. But if you were a patient in 1975 who kept having this stomach pain and your doctor kept telling you to stop being so stressed out and you didn’t think you were stressed, eventually the patient gets a message that the doctor must think I’m crazy.

Benson: As a practicing physician, Hatch makes sure he voices his uncertainties to his patients. He does this in order to minimize the risk of a false diagnosis, even though it occasionally causes conflict with the patient.

Hatch: I’ve had patients who get quite upset with me when they have their own idea about what’s going on and I share with them my skepticism. And I’ve also been on the other side of it as a patient and as a family member of a patient when a very close relative of mine got ill, I had doctors coming up to me saying ‘Y’know, this is what we need to do for your loved one’ and I didn’t actually know the tests at the time. And once I started reading up on the tests, I was really amazed at how confident they were that this was really a major lifesaving technology when in fact, maybe two people in a thousand were going to benefit from it but not 999 people out of 1000.

Benson: In order to decrease the number of false diagnoses, Hatch recommends more transparency with patients. This would allow doctors to uphold their credibility while allowing the patient to become more aware of the details of their diagnosis.

Hatch: If doctors are good at explaining what it is that they know and what it is that they don’t, I don’t think patients will lose confidence in us. I actually think that for the most part, patients are perfectly capable of understanding that tests aren’t always 100 percent positive and 100 percent negative. The issue is in the communication of the nuances, the people to say this test is really, really good and to say this test is so-so. And I think that the more you are able to discuss as a patient with your doctor and from a doctor’s perspective, the more you can discuss with your patient, the more you can arrive at a mutual understanding rather than having it more like a sort of drive-thru fast food situation where you go, you order a bunch of tests, you come to the window, the doctor hands you a diagnosis, y’know, you hand the money, and then you walk out the door happy.

Benson: It’s hard to become comfortable questioning doctors, especially knowing that they are experts in their field. But, Hatch says medical school doesn’t always provide the training necessary to deal with the nuances of each individual case. As difficult as it is for a patient to hear, sometimes the best thing a doctor can say is, ”I don’t know.”

You can find out more about Dr. Steven Hatch and his book Snowball in a Blizzard through links on our webpage, radiohealthjournal.net.

Our writer this week is Michael Wu. Our production director is Sean Waldron. I’m Nancy Benson.