DrRich’s Last Word On Breast Cancer Screening

Posted on November 23, 2009
Filed Under General Rationing Issues, Healthcare Reform |

DrRich has taken significant heat for his last two postings on breast cancer screening (the first of which questioned the methods and motives of the USPSTF, and the second of which addressed the subsequent dissembling by Secretary Sebelius regarding same). Among the unpleasantness was a detailed vivisection performed on DrRich by none other than the great Orac. (For regular readers, who are aware that DrRich does not always state in a direct manner what he actually means, DrRich is not being ironic here. He has long been an admirer of Orac - who provides a valuable service to humanity with his extremely entertaining deconstructions of various forms of medical woo - and he highly recommends Orac’s blog.)

DrRich will not attempt to answer the various critiques that have been leveled at him, here and there, regarding his intellect, his presumed political persuasion, his taste, or his parentage. DrRich can tolerate such personal attacks without taking umbrage. For it is certain that if any of these correspondents knew DrRich personally, they would be utterly enchanted by him.

DrRich will, however, address in a general way the most substantive criticisms (including some of those made by Orac) that have been made regarding his interpretation of the USPSTF’s recent recommendations.

Those substantive criticisms, in general, can be summarized with two statements. First, some critics allege, the recommendations made by the USPSTF are actually quite reasonable, and DrRich should immerse himself in the scientific evidence before dismissing those recommendations, and further, by so dismissing them, DrRich has revealed himself to possess (to quote one critic) “monumental ignorance of the history of mammographic screening…”  And second, after whining for years about covert rationing, DrRich should embrace the USPSTF’s new recommendations, since they are at least a step in the right direction toward open rationing.

DrRich cheerfully concedes that he, being a simple country cardiac electrophysiologist, is not an expert in breast cancer screening.  Further, DrRich will admit that while he loves history, and greatly appreciates breasts, he has no desire to immerse himself in the history of mammography. He really doesn’t know what the “best” recommendations on breast cancer screening would look like. Indeed, without any reservation he embraces the possibility that the “best” recommendations, based on the limitations of what we now know, might look a lot like the ones made last week by the USPSTF.

He does not deny the harms of overscreening. DrRich believes that for many decades a lot of overtesting has been done by doctors. Annual chest x-rays to screen for lung cancer (now, thankfully, out of favor), and routine cardiac stress tests and echocardiograms (still being done by too many cardiologists), are good examples of screening overkill. DrRich is also well aware of the harms that can be done by overdiagnosis, and himself declines to have the screening test for prostate cancer.

DrRich is in favor of formulating guidelines derived from looking objectively at the available evidence, and making the best recommendations given the data at hand. He agrees that panels of experts are a reasonable way to approach the construction of such guidelines. He even agrees that panels, like the USPSTF, that consist mainly of unbiased generalists (and not the specialists who may have a vested interest in finding lots of disease, or in performing a lot of well-paid screening tests) are probably the most appropriate group to do this.

Nothing in DrRich’s previous posts, he thinks, indicates otherwise.

DrRich’s problem with the USTSPF and the recommendations it published come down to: a) the methods it used to reach those recommendations, and b) the disgraceful lack of transparency - if not by the USPSTF itself, then by the handlers of this panel - regarding the true import of those recommendations.

Addressing the second point first: Nobody in authority or in the media is telling us that the “recommendations” of the USPSTF, far from being mere recommendations, or even mere strongly-suggested guidelines, are going to be used to directly determine what is covered and not covered by “qualified” insurance plans. Both the House healthcare reform bill and the Senate bill contain language to the effect that in order for preventive services to be covered, those services will have to attain a USPSTF “grade” of A or B. So despite Ms. Sebelius’ soothing blandishments to the contrary, once healthcare reform becomes law, breast cancer screening for women aged 40 - 49, and above age 75, will no longer be offered by approved insurance plans.

Perhaps, given what is known about the effectiveness of breast cancer screening, that is appropriate. But it is not appropriate to deceive women about the import of these new recommendations.

Given that the recommendations made by the USPSTF are meant to be used directly to determine coverage and non-coverage of preventive health services, it seems to DrRich that the methods it has used to make its recommendations on breast cancer screening are, in the long term, much more important than the content of those recommendations themselves. For, those methods may very likely set a precedent for future coverage decisions. We should examine them closely. We should be extremely critical as we do so. Those methods, as applied by this panel, will (forgive DrRich) indeed determine life and death.

Not being an expert in breast cancer screening, DrRich relied on the explanatory document produced by the USPSTF itself to criticize its methods. After all, if understanding those decisions really requires one to become an expert in the science (immersing oneself, say, in the history of mammography), then the recommendations themselves become, for practical purposes, “received knowledge,” that is, wisdom handed down from on high, which the public must simply accept. DrRich thinks we can all agree that if coverage decisions are going to be promulgated with little transparency, it won’t be long before the process will be grossly abused in favor of this interest or that.

And so it is USPSTF’s own words that DrRich used to base his criticism of the panel’s methods, and thus for judging  the precedents it has thereby set for making future recommendations.

To attempt to place a more mathematical framework around his argument, DrRich is postulating that the USPSTF is creating here something like an algebraic formula. It is creating a variable-containing equation for making coverage decisions, and that, in this instance, “breast cancer screening” is merely the variable that is being plugged into the equation. What DrRich is trying to do is to get a first glimpse of what the equation itself may look like. What “functions” (or, what “rules”) may be applied to a medical service to determine whether it is worthwhile or not? We can begin to recognize these rules, DrRich is saying, by examining the logic USPSTF has used in this most recent instance.

In this light, DrRich criticized three specific recommendations of the panel. First, he criticized the recommendation that breast cancer screening begin at age 50 instead of age 40.

DrRich is, of course, well aware that the “old” recommendation to begin at age 40 was controversial and largely arbitrary. But still, in order to change a formal recommendation, however it was reached, there ought to be some objective reason for doing so. So in this spirit, what “rules” did the panel invoke for changing its recommendation from 40 to 50?

By the panel’s own words, screening mammography in women in the 40 - 49 age group appears as effective at reducing mortality as it is in the older age group, and indicates this finding several times within its document.   And as s nearly as DrRich - who is not an expert in the field, but is an electrophysiologist* - can tell, the panel gives only one concrete rationale for changing its recommendations, which is as follows: The panel invokes “a new systematic review, which incorporates a new randomized, controlled trial that estimates the ‘number needed to invite for screening to extend one woman’s life’ as 1904 for women aged 40 to 49 years and 1339 for women aged 50 to 59 years.”

Is this really the whole story? Is it wrong altogether? DrRich does not know. But it is the rationale the panel itself gives us.

What “rule” does this rationale imply? One possibility is: Even for a screening test that is equally effective in reducing mortality, if you must do 40% more tests to save a life in group B as in group A, do not offer the screening test to group B. Other, similarly unhelpful rules can be imagined from this, but most readers will get the drift.

DrRich’s critics can argue that this is not what the panel really meant to say, and that if one really understands the massive, convoluted volumes of clinical trial data, there is actually a compelling reason to offer screening to 50 year-old-women but not 40-year-old women. Again, DrRich cheerfully concedes that this might be true. For that matter (since the panel seems to have trouble articulating why a 50-year-old woman needs to be screened but a 40-year-old woman does not), perhaps the panel really ought to argue that no women should be screened for breast cancer, given the significant downside.

But if that’s the case, DrRich argues that the panel should say so, and not publish a document that tacitly establishes a potentially harmful new rule that can easily be used for arbitrary rationing when “recommendations” are being considered, in the near future, for other preventive medical services.

Second, DrRich criticized the recommendation that women not be taught breast self examination (BSE). In truth, DrRich finds it unlikely that BSE will measurably improve overall survival in any large group of women, and is not really surprised that two large randomized clinical trials failed to show a benefit. He does not actually mind that the panel is recommending that doctors not waste their time teaching these techniques. Doctors haven’t really done that in 20 years anyway.

What DrRich objects to is the method the panel used for making this recommendation, which establishes the following new rule: It is legitimate to take the results of clinical outcomes trials conducted in backward, 3rd world healthcare systems, and directly apply those results to coverage decisions affecting American patients in the American healthcare system.

Does it not strike anyone but DrRich that this seems like a bad rule? It would be like performing a careful statistical analysis of outcomes from a Pee Wee football league, then telling the New England Patriots to abandon the forward pass, because the percentages just aren’t there. If what our Democrat friends really want (as some in the right wing have accused them of wanting) is the Cuban healthcare system, the routine application of this rule would be one sure way of getting there.

Third, DrRich objects to the recommendation that women 75 and older not get breast cancer screening, despite the fact that (from the panel’s own words) breast cancer is the leading cause of death in this age group. The panel justifies this recommendation by noting that there are insufficient data from randomized trials in these patients, and further, that “women of this age are at much greater risk for dying of other conditions that would not be affected by breast cancer screening.” The new rule: Unless there is statistically compelling data to prove otherwise, preventive screening in old people should not be offered, because they’re old.

Some, DrRich understands, will be fine with this rule. He’s not. (Disclosure: DrRich is rapidly becoming old.)

DrRich does not expect that this clarification of his objections to the USPSTF recommendations will change the hearts or minds of those who have been ripping him a new one. He does hope that his critics will at least begin criticizing him for the right thing.

It’s not the panel’s specific recommendations regarding breast cancer screening he objects to, but rather, the methods they have used in reaching those recommendations, and the new “rules” those methods imply.

Given the critical importance of this particular panel over issues of life and death (a level of importance which our political leaders would demonstrably like us not to notice), DrRich believes that the rules by which it operates will become very, very important to many American patients, and that we should not take them lightly.

*DrRich has neurosurgeon friends who, when they wish to express the idea that something they’re doing is not quite as difficult as one might think, but find that what they’re doing is, in fact, neurosurgery, will say, “It’s not exactly electrophysiology!” So being an electrophysiologist bodes at least as well for one’s intellect as, say, staying at a Holiday Inn Express.

Comments

14 Responses to “DrRich’s Last Word On Breast Cancer Screening”

  1. Marilyn Mann on November 23rd, 2009 9:05 am

    Hi Dr Rich

    I don’t know if you saw my comment over at Orac’s place, but my statement that I stopped reading your blog when I saw a post that seemed to promote global warming denialism was not intended to be a personal attack, but rather a statement of fact. That is, in fact, what happened. Of course, it is always possible that I misunderstood your intent in that post. However, the bottom line is that I can only read a certain number of blogs, so I have to pick and choose.

    Also understand that in the past I have had pretty extensive (and unpleasant) interactions with another species of denialists, namely members of The International Network of Cholesterol Skeptics, or people who sympathize with them, such as Eddie Vos, Uffe Ravnskov, Malcolm Kendrick, and so forth. Because some of my family members, including my teenage daughter, have heterozygous familial hypercholesterolemia, and I have seen the death and suffering caused by this disease, I have a particular horror of denialism, whether it is HIV/AIDS denialism, so-called cholesterol skepticism, global warming denialism, or what have you.

  2. DrRich on November 23rd, 2009 9:18 am

    Marilyn,

    I do not take it personally.

    Actually, I do not deny global warming. I believe, in fact, that it’s probably occurring. But I have an open mind, and view with interest some of the evidence that we may be starting to cool.

    What I object to is the notion that the science of global warming is settled, and that no evidence to the contrary is to be admitted. While political questions may be settled (by a consensus, or by force), science can never be truly “settled.”

    Rich

  3. Red Baron on November 23rd, 2009 9:28 am

    Rich, how is this for a paradox: the physicists have already settled this notion any issue can ever be settled. Bottom line: the idea that anything can be settled is epistemologically wrong, especially where you are dealing with chaotic system (remember my obsession with fractals?)

    Just an FYI

    Be well

  4. Red Baron on November 23rd, 2009 10:18 am

    One other point concerning the inability to apply linear logic to non-linear system- I know you already know this backwards and forwards. There is a huge difference between between relative and absolute risk/benefit.

    I know nothing about breast cancer screening so please read the following as illustrative only:

    If screening improves cancer detection by 30% for all groups, but the absolute risk of cancer is only 1% in a younger 40-50 group while it is 10% in an older 50-60 group, the absolute benefit of screening is only 0.3% vs. 7%.

    This is a 23 fold improvement in the absolute benefit of screening.

    Remember, it is a non-linear system. Don’t perpetrate the same mistake you hate seeing in others.

    I know this is not your main point, but you do make this error in your post if you read what upsets them.

    I completely agree with you that the covert aspects of this whole debacle are offensive.

    Anyway, be well

  5. Michael Kirsch, M.D. on November 23rd, 2009 4:02 pm

    This may be DrRich’s last word, but it’s not the last word. Sec’y Sebelius showed a cowardly retreat last week from her own panel’s revised guidelines. As a gastroenterologist, I’m envious of all the attention and hype that Mammogate created. When my colonscopy guidelines are revised, nobody cares. See Our government, Sec’y Sebelius in particular, was shameful. They rejected their own task force. The USPSTF, while not perfect, is known for its caution and objectivity. This past week was a huge defeat for comparative effectiveness research, an endeavor that the administration stated was a high priority item. See http://bit.ly/656CwP

  6. Dr Dan on November 23rd, 2009 4:27 pm

    Dr. Rich: If you’re taking flak, you must be over the target. Keep on posting.

  7. A Skeptic on November 23rd, 2009 4:34 pm

    I still do not see how you can still consider this process “covert.” The existence of the USPSTF panel is public knowledge. The names of the panel members are public knowledge. The methodology used to make the recommendations, as much as you may not agree with it and as much as the general public does not understand it, is public knowledge. The only thing actually “covert” about it is how these recommendations may be used in the future as proposed in a bill that is still quite far from being passed by Congress. The USPSTF has been making recommendations long before this bill was even considered. In your book, you proposed a methodology for open rationing. Although it made sense to me, I’m sure that it would be open to just as much criticism as the methodology used by the USPSTF. It was not that long ago that mammography recommendations called for women to have a “baseline” study in their 30’s. These early baseline mammograms were so plagued by false positive and indeterminate results due to cyclic hormonal effects on normal breast tissue that the recommendation for first mammogram changed to age 40. So, I’m not surprised with the latest proposed changes. Statistically and medically they make sense.

  8. Dr Dan on November 23rd, 2009 5:31 pm

    One more point. Orac seemed to flog you with unusual relish. I wonder if he ever attempted to follow gidelines for screening and found himself the object of a lawsuit? That happened to a colleague of mine who was a real advocate of following USPSTF guidelines and was sued, successfully, by a patient who developed prostate cancer. The linch pin of the case against him was that he had not ordered a PSA. The DRE was normal. The jury ,not bothering to consult Orac, found for the plaintiff and awarded him a nice big settlement.

  9. Paul on November 23rd, 2009 9:45 pm

    Perhaps, the USPSTF was recognizing the state of our Medical care when the benefits are fully established in 2014. Since we may very well be living in a 3rd world economy because of government spending and excessive taxation of the producers of wealth, the observations in 3rd world countries may fit our future situation well. Your main point about the application of statistics which apply to a group should not automatically applied to individuals is well taken.

  10. DrRich on November 24th, 2009 6:26 am

    Skeptic,

    It sounds like you and I simply have different conceptions of what “open rationing” really means. The USPSTF’s actions resemble legitimate open healthcare rationing in the same way that zombies represent real humans. They look similar (in some ways) on the surface, but at the core they are fundamentally different.

    Open rationing, as I have described it (since that is your reference) requires, first and foremost, stating a clear and actionable ethical principle by which we will attempt to resolve the fundamental rationing question, which is: How are we to weigh the competing interests of the individual vs. the collective in a rationing system?

    Then, using that principle as a guide, you need to establish a system of well-defined and widely-broadcast rules to deal with specific, predictable, and common dilemmas that will appear as you make the rationing decisions. (Such as: are we to assign intrinsic “worth” to individuals; how are we to treat illnesses that are thought to be self-induced by lifestyle choices?)

    With these pre-determined, fully-vetted, and very public rules in hand, you can then approach the actual rationing decisions themselves.

    These steps are vital. Without them, the actual rationing decisions will necessarily be arbitrary and capricious.

    My point being: the USPSTF has not taken these vital steps, and their decisions are therefore (predictably) arbitrary and capricious. My post, which purports to “reverse-engineer” the “rules” this panel must have been using, is intended to illustrate this point.

    Rich

  11. DrRich on November 24th, 2009 6:33 am

    Dan,

    Orac is OK. Skewering people is what he does, and he’s quite good at it. As a longtime reader of Orac, I actually thought he was relatively gentle with me. He even (nearly) apologized toward the end for being excessive, which is something I’ve never seen before from him. And while he ridiculed some of my ideas (having misinterpreted my main point, to be sure), he avoided ad hominem attacks. So I have no hard feelings.

    I do, however, have a strange new feeling of brotherhood with the purveyors of reiki.

    Rich

  12. DrRich on November 24th, 2009 6:34 am

    Paul,

    Bingo.

    Rich

  13. DrRich on November 24th, 2009 6:49 am

    Red,

    Your point is correct. I tried to stick with the statements of fact as they appeared in the document published by USPSTF (since this defines the “rules” under which their decisions were made). The sense I got from this document was that USPSTF believes the effectiveness of screening is the same for the younger women as for the older women, and that the thing that justifies differential recommendations is something other than the efficacy of screening. I chose to take this at face value, because I believe it would be unreasonable to expect any “customer” of these recommendations to do otherwise. It’s hard enough to try to reverse-engineer the “rules” they used for rationing, without having to delve into the literature, etc., in an attempt to impute their actual reasoning.

    Rich

  14. Michael Kennedy MD on November 24th, 2009 2:07 pm

    I have a couple of objections not yet mentioned here. One is that physician breast exam is also not to be covered. That seems a bit excessive. Doing away with all healthcare would certainly be cost effective, at least by these standards. Having practiced surgery for 40 years with a breast practice about half my time, I think I know a little. I was also offended by all the lying about harmful effects of false positives, which included “deformity” from biopsy. Do these people not know that biopsy of mammogram lesions is done by needle ?

    Secondly, I watched Debbie something-something lie about the USPSTF as another Representative, named Blackburn, read to her on TV from the House bill. If they can make these decisions, the least they can do is not lie about them.

Leave a Reply