Saturday, January 28, 2012

"Abandoning ship": is debunking ineffective screening and therapy removing hope or just removing risk?

Sometimes I think I sound like a real medical nihilist, since many of my blogs have been about purported treatments that are ineffective, dangerous, done too often, and cost a lot of money (Cardiac stents and profit-driven corruption: do anti-fraud rules address the problem?, December 24, 2010; Greed, corruption and medical procedures: ignoring or suppressing the evidence?, August 12, 2011; Men’s Health? Women’s Health? Valid screening opportunities or “Hallmark Holidays”?, March 15, 2011).  It is a somewhat distressing feeling. I am a doctor; I want to help people, to relieve their suffering, to help facilitate the cure of their diseases. More than that, I am a family doctor, and have a strong preference for prevention, for early detection of disease when it is still treatable rather than waiting for it to be too far gone for treatment to be effective. In addition, I have regularly criticized our health non-system for leaving out too many people, creating financial disincentives for them to seek care early. This leads to their waiting until their diseases become so uncontrolled that they present to the emergency room, then require admission and costly care, making it worse for them (most important) and more expensive for everyone. So I think prevention and early intervention is a really good thing, and it would be great to have tests that could identify disease early in its course so that we can change its trajectory.

But because I want these things to be true doesn’t make them true, as I discuss in the recent post Avastin®, Plan B®, and Magical Thinking, December 31, 2011. Because we can screen you for something doesn’t mean that we should; because a test can be done doesn’t mean that it is a good test. "A good test", in my opinion, is one that has sufficient sensitivity (rate of being positive when a condition is present) or specificity (rate of being negative when a condition is absent) to tell us whether a you have a disease, or if it matters. Because it can be done also doesn’t mean that it is cost-effective. Because a treatment exists doesn’t mean it is a good treatment, a safe treatment, an effective treatment. And, as with most things being sold, the greater the publicity and advertising around it, the more it means someone will be making money on it, which does not exclude its being of benefit, but is certainly not the same thing.

Sometimes, when evidence is discovered that a test or treatment is not of benefit, eventually we stop doing it. "Eventually", however, may be a lot longer than you might think. In a recent “Viewpoint” in JAMA, Prasad, Cifu, and Ionannidis address “Reversals of established medical practice: evidence to abandon ship[1]. They note that while “Ideally, good medical practices are replaced by better ones, based on robust comparative trials in which new interventions outperform older ones and establish new standards of care. Often, however, established standards must be abandoned not because a better replacement has been identified but simply because what was thought to be beneficial was not.” They go on to discuss a number of treatments that have been “standard of care” but were shown by good, randomized controlled trials, to be ineffective or even dangerous (not to say expensive). These include stenting of coronary arteries for stable coronary artery disease (CAD), postmenopausal hormone therapy to prevent CAD, vertebroplasty for osteoporotic fractures, bevucizamab for breast cancer.

Scarily, but unsurprisingly, “true believers” continued to defend these interventions even after the evidence was clear (their livelihoods depend upon it), and in many cases these treatments continue to be offered and performed. “There are thousands of clinical trials, but most deal with trivialities or efforts to buttress the sales of specific products,” and it is only “Rarely [that]…some investigators find the courage to test established ‘truths’ with large, rigorous randomized trials”. Prasad and colleagues have done many of these latter trials; indeed John Ioannidis is the “guru” of debunking treatments with poor evidence (see David H. Freedman’s article in the November 2011 AtlanticLies, damned lies, and medical science.”) The authors have some suggestions, including limiting the role of manufacturers (of drugs, devices, tests) from conducting the trials on them, although they should pay for them: “Large trials of new innovations should be designed and conducted by investigators without conflicts of interest, under the auspices of nonconflicted scientific bodies. Instead of designing, controlling, and conducting the trials, manufacturers may offer the respective budget to a centralized public pool of funding, keeping the trial design and conduct independent.”

Prevention and screening are also subject to the lure of magical-thinking-compounded-by-the-greed-of-the sellers. Screening for prostate cancer with the use of prostate-specific antigen (PSA) testing is a good recent example about which I have written (while the lack of effectiveness of treatments for prostate cancer is would be included in the group of ineffective therapies that Prasad and colleagues have written about). Pap smears are pretty good screening tests for cervical cancer, but most other cancer screening tests (even mammograms and colon cancer screening, probably the next best) are not nearly as good. Every time there is a recommendation to decrease the frequency of screening (Pap smears, mammography) or not do them at all (PSAs or pelvic exams) there is an outcry from people who think that something has been taken from them. In a scientific sense, what has been taken is unnecessary testing that doesn’t lower their risk of bad outcomes, costs money, and can have significant morbidity when positive screens need additional, more invasive tests, to follow up. But, in a more metaphysical sense, what has been taken is hope, the idea that there is something that they can do that will prevent something bad from happening to them. Something that, while perhaps a little risky (if you understand or believe it), is relatively easy. And also, frankly, something someone else, rather than you, can do. Not like, say, dieting or giving up smoking or exercising. And this false concept is encouraged by half-truths promulgated by passionate advocates of interventions with limited proven benefit, whether traditional allopathic or “alternative”.
Even public health advocates sometimes get so passionate about the health benefits of what they are advocating that they may miss other issues. For example, consider a recent discussion on a progressive public health listserve about “portion control” (limiting the amount eaten at one sitting). This is an important and effective method of addressing obesity, which is in fact a major health problem in this country, but the discussion raised many other issues. These included concerns about “blaming the victim” -- expecting self-control (difficult enough) from people who had historically not had that control in the face of major initiatives by fast-food purveyors to push large portions. Another was the economics involved in asking poor people to pay “more for less” while corporations continue to make huge profits from them. It can, and often is, as difficult for public health advocates to recognize problems arising from their positions and passions as it is for providers or manufacturers to back off from theirs.

The difference, of course, is that the former are really invested in the health of the public, while the latter are invested mostly in, well, their investments.

[1] Prasad V, Cifu A, Ioannidis JPA, Reversals of established medical practices: evidence to abandon ship”, JAMA 4Jan2012;307(1):37-8.


David4Peace said...

Don't feel conflicted about raising these issues repeatedly. They're hugely important, and far too few voices are being raised. Overtreatment is a curse on American health, probably at least as dangerous as lack of treatment, and they're bankrupting the system as well. I write about this all the time on my blog at diabetesselfmanagement.

tooearly said...

"But, in a more metaphysical sense, what has been taken is hope, the idea that there is something that they can do that will prevent something bad from happening to them."
Perhaps. But contrast to that what has been inadvertently given to them by all of this fascination with screening: PARANOIA and the constant feeling of risk of significant disease and death that seems to lurk very nearby, when indeed for most of us, it is distinctly not.

Josh Freeman said...

Posted for Sherry Spence:

Very interesting blog; thanks. Everything you wrote struck me as spot on. So it got me thinking, of course. What about the way we develop health-related recommendations and the way that "knowledge" becomes health-care protocol? Given our health-care "non-system" as you call it, have we become far too much of a one-size-must-fit-all nation in that respect? As you pointed out toward the end of the post, we in public health tend to get so passionate that we reject workable strategies because they are not comprehensive enough or do not adequately incorporate the need for social justice in public health interventions. I agree. And sometimes we get so focused on our point that we fail to notice - as I did recently - how easily it can be misunderstood and, therefore, hurtful or harmful.
So, while you make a good point that addressing personal action and responsibility with respect to personal health is a valid public health intervention, I maintain that it risks causing harm if it becomes an unqualified policy and then an unqualified protocol. As someone who has experienced unremitting pain for more than ten years, I must say that when my PCP suggests I exercise portion control, I tell him - trying to suppress my anger - there are times when I simply can't. It seems to me that if the pain were psychic/social/emotional/psychological, that inability would be even more true and the anger at hints of blame would result in my engaging even more in my risky behavior. It seems to me there's a larger issue here than the one of portion control or individual health. When we develop public health interventions that can also be health care protocols, they are often applied to 100% of persons while our data may show they apply to 67% (or 95%, or...) of cases. Because of the way we develop our understandings, I wonder whether we should ever suggest they be applied generally? Do we realize that insurance companies, if not doctors, often take our recommendations to mean the protocol applies in every case? Is anything really always beneficial? And if not, aren't the persons most likely to be harmed those who have the least equity in health-care? Is one underlying problem with achieving social justice in health-care the way our recommendations are stated and how they may be implemented?
Thanks for your always thought-provoking posts,

Total Pageviews