EDITOR NOTE: THERE IS AN ADDENDUM, ADDED SEPTEMBER 10.
Besides being a researcher and prolific blogger, I still maintain a practice in breast cancer surgery. It’s one of the more satisfying specialties in oncology because, in the vast majority of cases I treat, I can actually remove the cancer and “cure” the patient. (I use the quotes because we generally don’t like to use that term, given that some forms of breast cancer can recur ten or more years later, but in many cases the term still fits, albeit not as well as we would like.) Granted, I get a little (actually a lot of) help from my friends, so to speak, the multimodality treatment of breast cancer involving surgical oncology, radiation oncology, and medical oncology, but breast cancer that can be cured will be primarily cured with surgery, with chemotherapy, hormonal therapy, and radiation therapy working mostly to decrease the risk of recurrence, either local in the breast or distant elsewhere in the body. Through this multimodality approach, breast cancer mortality has actually been decreasing over the last couple of decades.
However, as a breast cancer surgeon, I not infrequently have to deal with many of the common myths that have sprung up around breast cancer. Some are promoted by quacks; others are just myths that sound plausible but aren’t true. (That’s why they persist as myths.) One such myth has been in the news lately, in particular last week; so I thought now was a good time to take a look as any. Besides, I spent most of the weekend out of town visiting my wife’s family, and I didn’t have a lot of time for this post. So this week sticking to something I know well makes sense and inspired me to make like Harriet Hall and Steve Novella and keep my post to a reasonable length for a change. There’s also so much less mucking about on PubMed and Google that way to make sure I’m not missing something, too.
Those who pay attention to breast cancer news (and, of course, I do have a Google Alert set up for just such news) can’t help but notice that late last week there was a flurry of stories (no, not that kind of flurry) about a study looking at the risk of breast cancer due wearing a bra. I’ll cut to the chase. According to this study, there was no increased risk of breast cancer due to wearing a bra, a result that, to breast cancer specialists, was about as surprising as the observation that the sun rises in the east and sets in the west, water is wet, and gasoline flammable. However, the myth that wearing bras increases the risk of breast cancer is one of those unsinkable rubber ducks (as James Randi would put it) of a myth that just won’t die. (Most recently it showed up in a Facebook- and Twitter-spread article entitled “Your Bra May Be Killing You – Scientists Call For Boycott Of Komen“, as well as on—of course!—NaturalNews.com.) I keep hoping that a study will do to this myth what a wooden stake in the heart does to the vampires in True Blood, but unfortunately this is one of those unsinkable rubber ducks of a myth (as James Randi would put it) that refuses to die. So I consider it worthwhile to briefly (for me) discuss how it came about.
It’s unclear where and how long ago this myth first appeared, but there’s little doubt on when it was first popularized: 1995. That was the year that a book by Sydney Ross Singer and Soma Grismaijer entitled Dressed to Kill: The Link Between Breast Cancer and Bras was published. The central thesis of the book was that bra-wearing is a major cause of breast cancer because of its claimed effect on lymphatic circulation. Basically, the idea was (and still is) that bras interfered with lymphatic drainage and thereby, though unclear mechanisms, caused cancer. The claim was that there are all sorts of “toxins” (of course) that cause cancer and that the lymph vessels drain those “toxins” away from the breast. Thus, if you believe Singer and Grismaijer, these “toxins” are concentrated in the breast by the constriction that bras produce and result in breast cancer.
This concept, that “toxins” somehow accumulate if lymphatic drainage is somehow blocked is one of the key concepts behind the quackery that is “detoxification,” used by quacks of many stripes. Such “detoxification” often consists of diets that result in purging, coffee enemas, chelation therapy to “remove heavy metals,” and many other treatments without a solid basis in science. However, “manual lymphatic drainage” is yet another form of quackery, at least when it’s claimed to “detoxify.” Sometimes, the quackery becomes rather amusing, such as when people come up with ideas like the “Brassage.”
Perhaps the best example of these beliefs can be found in an article entitled “Can a Bra Cause Breast Cancer? The Answer May Surprise You“, in which we find claims like:
Now, if enough pressure is put on your lymphatic system to produce pain and swelling, as in the example of your feet, you’d notice and fix it. Meaning, you’d take off your shoes, put your feet up, and bring the swelling down.
But, what if the pressure were too slight for you to notice? Or, what if it had gone on for so long, you no longer realized your discomfort? This is the case when it comes to bras. Many women simply “get used to” the discomfort of a bra and pay no attention to the indentations bras leave on their skin. Some researchers believe it’s precisely this indentation that, over a period of time, causes long-term mild impairment of the lymphatic function.
One of the largest clusters of lymph nodes is located in the armpit and upper chest area. That lymph cluster filters drainage from the breast, arm, and upper chest. But, if your bra is too tight, it constricts drainage from the lymph nodes in your armpit area.
Drainage to the breastbone lymph nodes also gets constricted, especially by bras with underwire. On top of that, shoulder straps from bras potentially cut off drainage of the lymphatics along the top of the breast to the armpit. Also, pressure from the cups of the bra may hinder drainage down to the liver.
If the researchers are right and this is what’s happening, what does it mean? Simply this: Toxins are not being drained from the breast area, so breast tissues are more susceptible to degeneration and eventual cancer.
Another variant of this concept includes:
Women evolved under conditions where there was BREAST MOVEMENT with every step that they took when they walked or ran. My reading of the scientific literature about lymphatic flow shows me that this may be as important as the constriction factor. Every subtle bounce of the breast while moving, walking, running, etc. gently massages the breast and increases lymphatic flow and thus cleans the breast of toxins and wastes that arise from cellular metabolism.
There are a lot of jokes I could make here, but, through enormous self-restraint, I will refrain, other than to point out that Ralph Reed, PhD has a rather intense interest in the bouncing of breasts.
In their book, Singer and Grismaijer claimed to have found the following (and you will see this list with minor variations in phrasing scattered throughout alternative medicine sites):
- Women who wore their bras 24 hours per day had a 3 out of 4 chance of developing breast cancer (in their study, n=2056 for the cancer group and n=2674 for the standard group).
- Women who wore bras more than 12 hour per day but not to bed had a 1 out of 7 risk.
- Women who wore their bras less than 12 hours per day had a 1 out of 152 risk.
- Women who wore bras rarely or never had a 1 out of 168 chance of getting breast cancer. The overall difference between 24 hour wearing and not at all was a 125-fold difference.
Other claims included that breast cancer is only a problem in cultures where women wear bras. In contrast, in cultures where women don’t wear bras, breast cancer is a rare event. I’m guessing you can tell the major flaw in reasoning there, but let’s go on. Specifically, Singer and Grismaijer claimed to have observed that the Māori of New Zealand, who are integrated into white culture and therefore wear bras, have the same rate of breast cancer, while the aboriginals of Australia, who are bra-free, have practically no breast cancer. They made the same claim for “Westernized” Japanese, Fijians and other bra-converted cultures. They even claimed that women who don’t wear bras have the same risk of breast cancer as males. (Males have an incidence of breast cancer about 100-fold less than that of women.)
Of course, this “study” was not published in a scientific journal. It was published for the first time in Singer and Grismaijer’s book, which was updated in 2005. Naturally, cancer scientists immediately pointed out the massive flaws in the “study,” such as:
- Lack of controlled epidemiological data correlating bra-wearing with the risk for breast cancer
- Lack of proof that the pressure exerted by a bra reduces the flow of lymph
- Lack of proof that lymph contains carcinogens
- Lack of proof that there are carcinogens in the human body that can induce breast cancer
- Existence of published data correlating obesity with post-menopausal breast cancer
This doesn’t even consider the utter lack of even a perfunctory attempt to control for well-established epidemiological risk factors for breast cancer, such as age at menarche, age at menopause, age of first live birth, obesity, Western pattern diet, and others. There was no proper statistical treatment of the data, and the study participants knew the study hypothesis before they took the survey. By any reasonable scientific standard, Singer and Grismaijer’s “study” was a complete mess, but it had a huge impact on popular views of breast cancer and popularized a myth that’s persisted for nearly two decades. Naturally, Singer and Grismaijer reacted to criticism of their study and book in predictable ways, pulling the “garment industry shill gambit,” saying, “The bra industry is a multi-billion dollar enterprise. And billions of dollars are spent each year researching and treating this disease. Ironically, ending breast cancer can cause financial hardship for many people.”
And, most ridiculously:
We have become a breast cancer culture. As a result, our information was seen as a threat to that culture. Sometimes a disease is less threatening than change.
Unfortunately, in that time, very little actual research existed to refute or confirm this myth. A search of PubMed for “bras” and “breast cancer” results in mainly articles on breast edema after breast cancer treatment, designing bras after breast cancer surgery, and cosmetic implications of breast cancer surgery. There is one study from 1991 cited by Singer and Grismaijer as evidence in support of their hypothesis. There’s just one problem: The article said nothing of the sort. Rather, it concluded that “bra cup size (and conceivably mammary gland size) may be a risk factor for breast cancer.” The reason for this association is almost certainly not because women with larger breasts are more likely to wear a bra regularly and for longer periods of time but rather because they have more breast tissue and are more likely to have larger breasts because they are obese. (Obesity is a known risk factor for breast cancer.)
The utility of this latest study, by Chen et al. at the Fred Hutchinson Cancer Center, is that it’s about as rigorous a study as there’s ever been about this issue. Basically, it’s a case-control study that examined women with invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC), comparing them to women without breast cancer matched for relevant risk factors as follows:
Briefly, cases were women between 55 and 74 years of age first diagnosed with invasive breast cancer between January 1, 2000, and March 31, 2004, while residing in the Seattle-Puget Sound area. The Cancer Surveillance System, the region’s population-based cancer registry also participating in the Surveillance, Epidemiology, and End Results program of the National Cancer Institute, was used to identify cases. All ILC cases (identified using ICD-O histology codes of 8520, 8522, and 8524) and a random sample of 25% of the IDC cases (identified using ICD-O histology code of 8500) were targeted for recruitment to enroll equal numbers of ILC and IDC cases. A total of 1,044 out of 1,251 eligible cases were interviewed (83%), consisting of 454 IDC and 590 ILC cases. A common control group, frequency matched 1:1 to the ILC cases within 5-year age groups, was selected from the general population of women living in the three-county area by random-digit dialing.
There were a total of 454 cases with IDC, 590 with ILC, and 469 controls. The authors then looked for differences between cases and controls in various aspects of bra wearing, including bra cup size, recency, average number of hours/day worn, wearing a bra with an underwire, or age first began regularly wearing a bra. They found no correlations. As far as epidemiological studies go, this one is quite good. The only significant weakness, acknowledged by the authors, is that bra wearing was so ubiquitous that it was not possible to examine any differences between women who never wore bras and those who did. So instead the authors concentrated on, in essence, the “dose” of bra wearing; i.e., the number of hours a day on average, cases and controls wore a bra. In this, it is not unlike a previous study of vaccines, in which there was not a true unvaccinated group but there was no correlation between thimerosal dose in vaccines and autism. In epidemiology, it’s not always possible to have a true unexposed group.
In any case, I’m happy to see this study. Before, when confronted with a patient who had heard this myth, I would have to argue on lack of evidence and the extreme implausibility of it all, given that there was no evidence that axillary lymphadenectomy (a.k.a. lymph node dissection) increases the risk of breast cancer. Quite the contrary, in fact. A notable study that looked at breast cancer risk after axillary lymphadenectomy for melanoma found no increased risk of breast cancer.
Sadly, as busted as this myth has been and is now, I fully expect that even this study will not bury it for good.
It occurs to me that I should do more posts about these breast cancer myths. My only concern is that there are so many. I could easily spend the rest of the year, doing a post on a different breast cancer myth each week, and still have more left over. Maybe I’ll do it during Breast Cancer Awareness Month. It’s only three weeks away, after all.
Sadly, as busted as this myth has been and is now, when I first read this study, I fully expected it would not bury it for good. In fact, it hasn’t, as the rotting hand of this myth has pushed its way through the dirt to try to grab the rational and frighten women. Of course. Just the other day, one of the authors of Dressed to Kill, Sydney Ross Singer, published an article on one of the many wretched hives of scum and quackery that I monitor regularly, The Cover-Up Continues: New Study Claims Bra-Cancer Link a “Myth” This response is particularly lame, even for Singer. Get a load of this utter piffle:
A new study was just announced in mainstream media declaring the bra-cancer link does not exist. The study looked into the bra wearing habits of women ages 55 and older who had all worn bras since puberty. They concluded that women should be “reassured” that bras are not causing breast cancer.
Actually, this study supports the bra-cancer link, since all the women in the cancer group were bra wearers.
Notice that I just addressed this specious criticism above. Clearly, Singer does not understand the concept of dose-response. The only way this is a legitimate criticism, at least in the way Singer meant it, would be if the cancer-causing power of bras was so incredible that even a brief exposure to bra wearing causes cancer. He then goes on to cite the very same 1991 paper that I described above as “evidence” that bras cause breast cancer. As I described, that paper shows nothing of the sort. Then there’s this:
Why this study focused only on POST-menopausal women, instead of researching the positive association of pre-menopausal women, suggests a bias and agenda.
Again, the reason for this is simple. Breast cancer is primarily a disease of aging, with the average age of diagnosis being 62. The bra-breast cancer hypothesis is an environmental exposure; so presumably the number of years of “exposure” matters. So it makes perfect sense to look at postmenopausal women, for whom the prevalence of breast cancer is much higher than in pre-menopausal women and for whom the “exposure” to bra wearing would be the longest. There’s no bias or agenda; this was clearly a scientific decision that makes perfect sense. Studying premenopausal women in an exposure/risk factor study like this would have resulted in much reduced sensitivity to detect a true difference in breast cancer prevalence, given how much lower breast cancer prevalence is in them, and would have meant smaller differences in cumulative bra wearing. Clearly, Singer doesn’t understand the basics of case control studies, where one wants to maximize the chances of finding significant differences in the exposure of interest between cases and controls.
None of this stops Singer from ascribing dire motives. Instead of the “pharma shill gambit,” think the clothing industry shill gambit:
It seems there is, indeed, a hidden agenda. The main author, who is a PhD student, did not mention that The Fred Hutchinson Cancer Research Center, where this study was done, receives money annually from a “Bra Dash” fundraiser, where bras are worn on the outside of clothing during a 5K race to raise funds.
It would be very ironic to criticize bras for causing cancer when they are used to raise funds for cancer research.
Uh, no. Seriously. No.
Singer doesn’t stop there, though:
The fact that this poorly designed study without proper controls is used to conclude that there is no bra-cancer link also reveals a bias. The authors should have called for more research.
All this study really shows is that some women who have worn bras for 40 years or longer will get breast cancer and some will not. We already knew that. You can say the same thing about smokers and lung cancer.
Of course, as the bra-cancer link gains increasing public awareness, there are bound to be studies trying to refute it. Billions of dollars annually are at stake. There are also studies that try to refute the cigarette-cancer link. So I take all this as a good sign of progress, since it brings the topic into mainstream consciousness, and these attempts at denial are expected.
Ugh. The “tobacco industry” gambit, the very same gambit beloved of antivaccine advocates. It’s hard to take the criticisms of someone who’s already demonstrated himself to be utterly incompetent at epidemiological research seriously. His smoking analogy is also silly. If there’s one thing research on smoking shows, it’s that dose matters; if you compare a group that has less than a 10 pack-year smoking history, for instance, with a group with more than a 30 pack-year history, for example, you will find a big difference in lung cancer prevalence.
Again, is this study on bra wearing ideal? No. However, it does look at whether there were any differences in the “dose” of bra wearing or the characteristics of bra wearing between cases (women who had had a diagnosis of breast cancer) and controls (those who had not). Ideally, it would be nice to find a group of women who never wore bras who don’t have so many other confounding factors (such as being aboriginals) that comparisons between them and bra-wearing women are highly problematic, but that’s impractical. In the US and other industrialized countries, the vast majority of women wear bras from puberty on for at least their waking hours, and very few women never wear bras. Given that fact, looking for dose-related effects is the best epidemiologists can do. Given the utter implausibility of the hypothesis, it’s even arguable whether the money and resources should have been spent to do this study, but it was done and it was negative—exactly as one would expect.
No, as much as Singer thinks that someday he’ll “show them all,” it’s nothing more than the fallacy of future vindication.