Respectful Insolence

Grant crunch time again yesterday. That means it’s the perfect time once again to dig up something from the archives of old blog and repost it here. This particular piece originally appeared on January 12, 2005, just shy of one month after I started blogging. I’m guessing once again that, because not many people were reading back then, most of you probably haven’t seen this before, and that those of you who have probably don’t remember it. Once again, I’d be interested in feedback from those who haven’t seen this before now that my readership around 10-20 times what it was back then. The only changes I’ve made is to update some of the links and change the formatting.

Just one quick note needs to be made. Since I originally posted this, the use of MRI in the detection of breast cancer has increased, and more and more indications for it are becoming accepted as the standard of care. However, now as 16 months ago, there is still no evidence supporting its use as a screening test for the general population of women, and the only populations for which it is accepted as a screening test for breast cancer tend to be in young women at high risk for breast cancer, either because of a BRCA mutation or because of a very strong family history. Finally, one of these days I’ll explain exactly what I meant by the last sentence in this article.

BATTLING QUACKERY IN “CONVENTIONAL” MEDICINE

On Usenet, in misc.health.alternative, I’ve been criticized for applying the same standards of scientific evidence that I expect in conventional medicine to alternative medicine claims. Sometimes, I’m accused of attacking quackery that I see in the alt-med world, but not speaking ill of conventional medicine. This is a frequent tactic alties use. The problem is, it’s just not true. Here’s an example to help demonstrate:

In April 2003, I got into a brief letter-writing war with an official of one of these companies touting “screening” scans. In this case, a company called AmeriScan was advertising its services on a large local AM radio station. Besides the usual “total body scans,” one of AmeriScan’s services was “screening MRI” for breast cancer. In the ad, the announcer made the astounding claim that mammography misses 2/3 of breast cancers, which is just plain wrong. (In fact, the sensitivity of mammography ranges between 75-90% in most studies in which it is used for screening the general population, which is quite good for a test that is so inexpensive and relatively simple.) That false claim got my attention, and I listened to the rest of the ad.

Other false or dubious claims in the ad included:

  • Mammography misses more breast cancer than it finds. (Patently false.)
  • Mammography will mistake benign tissue as possible cancer in four out of five cases, leading to over 500,000 unnecessary biopsies every year. (A distortion of the real situation, which greatly depends upon the age range of patients and how mammography is being used. It is true that mammography results in many biopsies that are in retrospect “unnecessary,” but that just means that any test that would replace it must be more specific, not just more sensitive.)

I was utterly flabbergasted that a company could use such false and/or misleading statements this on a large AM station in order to sell a test whose validity had not yet been established for that purpose and that insurance companies didn’t pay for. Concerned about the false information being promulgated several times an hour on his station, I politely wrote the program director and expressed to him my concerns that this ad was deceptive in exaggerating the shortcomings of mammography and the usefulness of breast MRI as known at the time. I did not expect that the program director would immediately forward my e-mail to AmeriScan’s Director of Sales and Marketing, Stephanie Darcy, but that’s what he did. Ms. Darcy immediately wrote back to me and, in a rather snarky and self-righteous letter (the complete text of which I’d be happy to forward to almost anyone who asks) presented articles that supposedly backed up AmeriScan’s contention.

I looked up the articles. They were based on two articles (PubMed references here and here) looking at small cohorts of young women with either strong family histories or known BRCA mutations (which predispose women who have them to develop breast and ovarian cancer at a high rate). Thus, these are young women at much higher risk for developing breast cancer than the general population and the very women in whom mammography has long been known to be lacking as a screening tool, with a sensitivity of probably less than 50%. This is because young women have much denser breast tissue. Because mammography relies on seeing densities or calcium deposits in a less dense background, it is not as useful for young women as it is for older women. (Bad news: Breast glandular tissue gradually gets replaced by fat as a woman ages.) However, there is a HUGE difference between screening such a small, defined population in which the incidence of the breast cancer is expected to be very high (and where the sensitivity of the existing screening tool is known to be relatively low) and screening the general population, in which the incidence of the disease is much lower and for whom the existing screening tool is reasonably sensitive. That mammography missed 2/3 of breast cancers in these small groups of young women at very high risk for cancer and MRI caught virtually 100% of them is utterly irrelevant to the question of how useful mammography is as a screening tool for the general population or whether MRI would be better. Certainly, neither article concluded that mammography missed 2/3 breast cancers when used to screen the general population, which is in reality the claim AmeriScan was making. (In fact, talking about why this is so would make a very good topic for another post.) Thus, AmeriScan’s claim was a nonsequitur as far as the evidence Ms. Darcy presented went.

Not content with simply presenting her company’s point of view, Ms. Darcy had to throw around some ad hominems and sarcasm. Here’s an example from her e-mail:

If the medical profession is going to start calling technology and the spread of information ‘irresponsible’ they should point the finger first at mammography. Mammography is an 85 year old technology that lets thousands of women die needlessly every year. It appears, however, from your email that you would rather that we continue to perpetrate the farce of mammography being adequate protection from breast cancer for women. I hope your ‘opinion’ is of comfort to the tens of thousands of women whom mammography fails every year.

[Comment: It is irrelevant how old the technology is, and Ms. Darcy had presented no evidence that her technology could do better at presenting the "needless deaths" she is so concerned about. Further, my opinion was based on then-current evidence, and the efficacy of mammography is supported by decades worth of studies. Hers was based on misinterpreting, possibly intentionally, two pilot studies.]

AmeriScan’s physicians have taken the Hippocratic Oath to do no harm and we will continue to practice what is scientifically proven, and not bow to the undereducated party line. I encourage you to check out the well publicized and continuing scientific debate of whether there is actually any value in mammographic screening whatsoever. In the mean time, we will continue to save women’s lives from this terrible disease. Women absolutely deserve better protection than a mammogram.

[Note also the utter self-righteousness based on very scant data, the dismissing of anyone with the temerity to question AmeriScan as the "undereducated party line" and the statement as fact that their method saves women's lives. Does this sound familiar? It should. It's exactly the sort of stuff you'll find on the websites of many alties. Dr. Day, a seller of alternative "cures" for breast cancer, says similar things on her website about her critics, adding religious threats against her detractors to the mix. You might fall victim to it yourself if you challenge alties too strongly.]

I let Ms. Darcy have it with both barrels in my response. I first cited the flaws in her misuse of the existing scientific data and her faulty reasoning in great detail (as summarized above), and CC’d a copy to the program director, concluding:

Finally, the sarcastic, self-righteous, and condescending comments at the end of your e-mail about my wanting “to continue to perpetrate the ‘farce’ of mammography,” your “hoping my ‘opinion’ is of comfort to the tens of thousands of women whom mammography fails every year,” and your refusal to “bow to the undereducated party line” appear to have been intended to anger me. If they were, then congratulations. You accomplished your mission. I am now royally ticked off. Had you been less condescending and sarcastic, we might have had a reasonable discussion of our disagreement. Does your CEO know that this is how his Director of Sales and Marketing deals with a skeptical surgeon who e-mailed a radio station to express genuine concerns about your advertising campaign? As a cancer surgeon who takes care of breast cancer patients, especially galling to me was your claim that your company will “continue to save women’s lives from this terrible disease” (as if I do not). Unless you can show me a well-designed clinical trial that shows that the use of your MRI protocol for screening results in a measurable survival benefit for the population screened, you have no basis to make that claim. I’m quite certain that any responsible physician on your Medical Advisory Board would be forced to agree with me on this point if pushed.

Only because you made such a big deal of mentioning it in your e-mail in reference to the doctors at your company, I will point out that I took the Hippocratic Oath too. You have not. I take care of breast cancer patients and do translational research into breast cancer (and other cancers) nearly every day. You do not. As is true for all doctors, questioning my devotion to patients or my desire to see that they get the best care available (as you appeared to do) is the best way to make what started as a disagreement turn into outright hostility. Is that what you really wanted to accomplish with your e-mail?

I admit I was in rare form that day. (Damn, sometimes I’m good.)

I also informed Ms. Darcy that I would be happy to speak to her medical director about my concerns any time. Oddly enough, she never took me up on that offer. I contemplated writing or calling AmeriScan to see if I could manage to get a hold of Dr. Craig Bittner, the Founder and CEO, to have a little chat. However, the program director wrote back to me telling me that he was going to require Ameriscan to make changes in their advertisements to take into account my concerns. Within a few weeks, I noticed that AmeriScan’s advertisements were off the radio station. I found out a while later that the State of California had sued AmeriScan for false claims. I found a few months later that AmeriScan went out of business.

I like to think that I had some tiny part in helping put an end to this boondoggle and false advertising designed to take advantage of women’s fears of breast cancer. Certainly, the State of California, at least, agreed with me. I probably had very little to do with AmeriScan’s being shut down, but it’s nice to think that I might have played a small part. At the very least, I gave them a hard time at one large radio station. I only wish more doctors would do the same. What surprised me is that, as far as I was able to learn, no other doctor had complained. I find it hard to believe that, in such a large city, with so many prestigious medical schools, hospitals, and cancer centers, that no other breast surgeon or oncologist had heard these ads. More likely, they did, and just shrugged their shoulders. Certainly, for areas that I’m not knowledgeable enough to argue with confidence, I usually do the same.

I started this post with the provocative word “quackery” as applied to conventional medicine. Did AmeriScan’s service qualify? I’d say it was definitely borderline, if not outright, quackery. (There’s a good reason insurance companies don’t pay for screening MRI yet.) Certainly breast MRI has its uses in young women with high risk of breast cancer (strong family history or known cancer-predisposing mutations), as AmeriScan’s cited papers demonstrated, and more has been published since then supporting a role for MRI in these cases (see here for an example). However, in the case of the general population with an average risk of breast cancer, even now, nearly two years later, there is still no good evidence that I am aware of that MRI is better than mammography, for the simple reason that it is too sensitive. One recent study does show that targeted use of MRI, in which MRI is only done when mammography is abnormal, increases the positive predictive value of studies from 52% to 72% but does not obviate the need for biopsy, but this is not the same thing as showing that MRI is superior as a screening technique for the general population.

Once again, recall what I said regarding evaluating therapies, whether conventional or alternative: If there is one principle I hope to impart here, it is that the claims of conventional medicine and alternative medicine should be treated the same and that they should be held to the same standard of scientific and clinical evidence. I do not differentiate between the two when considering evidence, nor should you. Today’s correllary is that you should treat the claims of any company that is advertising a high-tech test (as Dr. Bittner was unfortunately doing for his screening MRI for the breast and his whole body scans) with just as much skepticism as any altie claiming to cure cancer with diet and prayer or barley and prayer. Look for randomized clinical studies, not testimonials. Make sure that the studies cited actually address the claim that the company is making, as they did not for AmeriScan, which used two small studies with unrepresentative patients at high risk for cancer to back up a deceptive criticism of mammography. Just because it’s high-tech doesn’t mean it’s necessarily better. One day, breast MRI for screening may become the standard of care, but that day is not today.

In addition, ironically the high sensitivity and spacial resolution of MRI may even have unintended consequences for women who do have breast cancer, but, given the extreme length of this post, that is definitely a topic for another day…

Comments

  1. #1 pigilito
    May 23, 2006

    Orac,

    Sort of on your topic: an article in the Times about a letter to the NHS from several leading British physicians asking the NHS to stop providing alternative medicine. They correctly call for the money to be used to pay for proven therapies and drugs.

    http://www.timesonline.co.uk/article/0,,2-2192360,00.html

    Good luck with your grants.

  2. #2 bbclink
    May 23, 2006
  3. #3 Julie Stahlhut
    May 23, 2006

    I think that the relationships among specificity, sensitivity, and prevalence in a screening test should be well-covered in high-school math. (Among other things, it would point out the scientific as well as the civil-liberties-related weaknesses of large-scale screening for drug use in the general population.)

  4. #4 DJ
    May 24, 2006

    What’s Stephanie Darcy doing now, anybody know??

  5. #5 moioci
    May 25, 2006

    Sorry, I gotta ask. You said,
    “the sensitivity of mammography ranges between 75-90% in most studies in which it is used for screening the general population”

    Compared to what standard? It seems that to determine what tumors are missed, you’d have to go around doing mastectomies on pts with negative mammograms and no palpable masses. Tough to get that through IRB, I hope.

  6. #6 not telling
    May 29, 2006

    I have a question regarding mammograms that I am horrified to ask anyone I know. How do they work on someone who is tiny ….. not quite invisible but still pretty little. What is there to read and how flat do breasts get squished? I have a few years to go and I would like to mentally prepare myself.

    And for those with fake boobs, how accurate are they and is there a chance the implant could rupture from a mamogram?

    Enquiring minds want to know if these variables affect the accuracy of mamograms. A medically trained family member already told me that small breasts are a higher risk for death because there is less tissue to remove and the cancer takes up more tissue ratio wise so I am not liking that very much….