Unmedicine

I saw a local specialist last week who recommended several thousand dollars in tests relating to the obscure, chronic, incurable, but otherwise not particularly awful condition I recently was diagnosed with. Having sprung for a lengthy consultation a couple of months earlier with a Houston doctor who turned out to have been my local guy's medical school professor on this subject, I wasn't convinced the tests were necessary. A phone call to the professor confirmed my view: the tests might or might not produce some interesting results, but they were highly unlikely to uncover any new aspect of my condition for which an immediate treatment was likely to be necessary, or even effective. So that's about $2,000 I don't need to spend (out of pocket, to boot, since I carry only catastrophic coverage and typically pay my own bills in cash).

All this is happening against the backdrop of recent articles questioning the conventional wisdom that screening and early treatment reduce deaths from breast and prostate cancer. I've always assumed I should submit in good grace to frequent mammograms, and that sensible men would get P.S.A. tests. It turns out there's real doubt whether there's any point to the early diagnoses and the treatments they inspire. In the case of breast cancer,
mammography is an inefficient method for detecting breast cancer. It’s much better at finding the indolent cancers that would have never caused harm than it is at finding the nasty, aggressive ones most helped by treatment. Statistics show that for 2,000 women screened by mammography over 10 years, one will be prevented from dying of breast cancer and 10 others will receive treatments for a cancer that would have never become life-threatening. That means that screening causes 10 times as many women to become cancer patients unnecessarily as it prevents from dying from breast cancer.
As noted by H. Gilbert Welch of the Dartmouth Institute for Health Policy and Clinical Practice in Lebanon, N.H., and author of "Should I Be Tested for Cancer?", this is the "overdiagnosis paradox":
“The more overdiagnosis the test causes, the more popular it is because there are more survivors,” he says. “The person who had a breast cancer diagnosed by mammography is tempted to view herself as being helped, but there are two other possibilities that are more likely,” he says. The first is that the person would have fared exactly the same without the mammogram, and the second is that the cancer the mammogram diagnosed was indolent and did not require treatment. “I always hope that the person who found cancer via mammography was helped,” says Welch, but on an individual level it’s impossible to say which category an individual person falls into. Statistically, the vast majority fall into the overdiagnosed category.
In other words, it's entirely possible that the women who survived breast cancer would have been fine even without the treatment, and that those who didn't get well had the sort of cancer that no treatment would have helped anyway. A similar picture emerges for prostate cancer and the P.S.A. test. Undiagnosed prostate cancer is so common that "[a]utopsy studies show that a third of men ages 40 to 60 have prostate cancer, a share that grows to three-fourths after age 85."
“Unfortunately, the evidence now shows that [the P.S.A] test does not save men’s lives,” said Dr. Virginia Moyer, a professor of pediatrics at Baylor College of Medicine and chairwoman of the task force. “This test cannot tell the difference between cancers that will and will not affect a man during his natural lifetime. We need to find one that does.”

. . . [T]here is little doubt that [the P.S.A. test] helps identify the presence of cancerous cells in the prostate. But a vast majority of men with such cells never suffer ill effects because their cancer is usually slow-growing. Even for men who do have fast-growing cancer, the P.S.A. test may not save them since there is no proven benefit to earlier treatment of such invasive disease.
If you'd like to think that modern medicine can delay certain kinds of death, this is bad news. If you're inclined to be fatalistic and to place strict limits on the degree to which you'll let aggressive medicine invade your life, it's actually pretty great. As the author of the article on prostate cancer suggested,
Not knowing what is going on with one’s prostate may be the best course, since few men live happily with the knowledge that one of their organs is cancerous.
Ditto, perhaps, for women and the breast cancer they may be harboring unawares. I've often thought there were medical conditions about my own body I'd just as soon not know about any sooner than I have to, particularly if the only treatment for them would be horrible, expensive, and only ambiguously effective.

7 comments:

MikeD said...

The issue is, that's the kind of decision I want in the patients' hands, not in the governments'. There IS a lot of over-testing done, mostly due to physician concerns over malpractice accusations if they fail to catch something early. This is a side effect of the malpractice industry that I would like to see some reform on.

E Hines said...

What MikeD said. And along the lines of the kind of decision I want in the patients' hands, not in the governments', though, I offer a couple of anecdotes.

In the weeks leading up to my wife's bilateral mastectomy, I had to browbeat her to go see a doctor about her then chronic fatigue. Paranoid that I am, I wanted her to get help, knowing that chronic fatigue is associated with cancer, although it's not at all diagnostic. As part of her checkup, she had a mammography, with came back with each breast loaded with tiny little pin-pricks of dots showing up. The specialist with whom she consulted (and who ultimately did the surgery) didn't think there was anything to the matter, but with the mammography technician's thought that they were suspicious, we insisted on a biopsy. Those results came back iffy in the eyes of the specialist, but he couldn't rule out breast cancer. His recommended course of action was to do a biopsy annually, and when (his word) it came back cancerous (maybe next time, maybe in 20 years), then do the mastectomy. We decided that since the unrolling would lead only to doing the mastectomy, we'd (my wife, actually; at that point, I'd only be standing around with a clipboard looking official) skip that middle part, with its expense and discomfort, and go right to the end game and do the mastectomy. The tissue of each removed breast came back from analysis as malignantly pre-cancerous, and without too many more months before becoming cancerous--waiting for the next year's biopsy might have let the cancer reach its full flower before we did anything.

On the matter of the PSA and prostate cancer, I talked to my doctor about this after the first study came out a year, or so, ago, and asked him why I cared about the test: most prostate cancers are slow-growing, and having reached my geezerdom, I was likely to die of old age before prostate cancer killed me. He agreed that, if the cancer didn't morph into a fast-growing variety, that was true--but a significant number of such cancers (we didn't get into what "significant" meant) morph into the fast-growing variety. I've decided not to run that risk.

And just to lay to rest another potential question, both of these doctors were--and are--very reluctant to cut.

Eric Hines

Texan99 said...

I think everyone here probably will guess correctly that I'm talking solely about information that might help individuals to make decisions for themselves.

I do expect my doctor to be a little affronted, or worried, that I'm not going to have these tests. I'll have to find a way to reassure him that I'm not ever going to sue him for not having done every conceivable test on me just in case.

To make really good decisions about an early cancer test, we need information not only about the survival rates of people who are treated, but about the comparative survival rates of those who are treated and those who are not, including in the latter category those who are never diagnosed. That can be tricky information to come by. In the case of prostate cancer, it's handy that someone decided to do large-scale autopsy studies, because that added information in the "never diagnosed" category. I don't know of any similar autopsy studies re breast cancer.

Every time someone comes out with a new idea for a test that might distinguish aggressive cancers from indolent ones, that helps, too.

Some cancers, like colon cancer and melanoma, respond beautifully to treatment early on but not at all well later on. For them, early screening tests are obviously a good idea, especially since the early treatments are neither nightmarish nor exorbitantly expensive.

Anonymous said...

There's also the consideration of the larger familial picture. For example, if someone's mother, aunt, great-grandmother and sister all developed breast cancer, then a more aggressive monitoring process is certainly called for. If no one has those, and someone is asymptomatic (ie. lumpless and otherwise fine), then I agree with everyone else that it should be a personal call and if the doc balks then see about putting something in writing if necessary.

I wonder how much of the over-testing stems from two things. One is fear of litigation. The second is an unwillingness to accept that death is still not optional. But that's a discussion for another post.

LittleRed1

rcl said...

When I heard about these studies last week my first thought was it was a setup to save money on administering Obamacare. Checking it out I'm impressed with the honesty of the assessments on early testing by the task force. Like any other businessman I expect doctors to pitch their product.

As Little Red notes family history is a strong influence on cancers. I think this strengthens the point that universal testing is overkill. My college girl friend lost her battle with breast cancer in her 40s. Her father died from it as well. She knew she was at risk before ever having a test. She caught it early and prolonged her life for several years.

From your description of the diagnosis T99 I assume the family history is free of your rare malady. Your strategy seems sound to me.

Our extended family has had several "rare" diagnosis over the years including 2011. Different specialists will come up with different diagnosis. It can take a lot of effort to come to a reasonable decision. A second opinion is the bare minimum. Challenging the information is sometimes helpful.

Texan99 said...

Oh, no, my family history is far from free of this malady. My mother died at the age of 35 from breast cancer, in 1959, so long ago that chemo was practically experimental.

But what I have to consider is not only whether I'm at high risk but whether the tests will alter my risk. If tests are alerting us to cancers early on but not changing the outcome very much, what's the point? We know that some people are treated and survive a long time while others are treated and do not. What we know less about is how much longer they survive than they would have without treatment, and how sensitive that differential is to an early diagnosis.

What may be true is that some breast cancers respond much better to early treatment than to late treatment. Other kinds aren't going to respond well no matter how early we intervene, and still others aren't going to be dangerous no matter how late we intervene. It's going to continue to present us with very hard choices (and ambiguous benefits from testing) until we get better at differentiating among these categories.

E Hines said...

But what I have to consider is not only whether I'm at high risk but whether the tests will alter my risk. If tests are alerting us to cancers early on but not changing the outcome very much, what's the point?

You're limiting your consideration to purely medical matters, but there are other risks that accurately predictive tests can expose for mitigation.

If the medical test can accurately tell you what you're likely to die of, and even coarsely when (vis., significantly sooner than your normal life span of 87.1 years), you can, for instance, use that information to alter your financial planning so as to maximize your leavings to your heirs in your known-shorter life span. Or you can lead a more efficiently hedonistic life, so as to make the most of your remaining time.

Just because a medical test gives you no information for a medical course, that test is not pointless.

Eric Hines