Talk:Risk factors for breast cancer

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Medicine / Hematology-oncology (Rated B-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Medicine, which recommends that medicine-related articles follow the Manual of Style for medicine-related articles and that biomedical information in any article use high-quality medical sources. Please visit the project page for details or ask questions at Wikipedia talk:WikiProject Medicine.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
Taskforce icon
This article is supported by the Hematology-oncology task force.

Bras and breast cancer[edit]

See Talk:Brassiere#Bras and Cancer Risk for previous discussion

One very plausible theory that explains the world-wide epidemiology of breast cancer has to do with the use of bras. Please see this link as well as the long discussion on the brassiere talk page. I know this is controversial but I think it deserves to be mentioned in the article. --Jonathan108 (talk) 17:17, 15 March 2008 (UTC)

I'd suggest a line similar to the one shown by the American Cancer Society under factors with minimal or no risk, since they address the book directly. Most of the other sources do put a line or two to address the book/idea. The National Cancer Institute,, American Cancer Society, and National Institutes of Health -MedLine all make mentons of bra wearing as having no impact on risk. -Optigan13 (talk) 00:41, 21 March 2008 (UTC)

The Cancer Society says: "There are no scientifically valid studies that show a correlation between wearing bras of any type and the occurrence of breast cancer. Two anthropologists made this association in a book called Dressed to Kill. Their study was not conducted according to standard principles of epidemiological research and did not take into consideration other variables, including known risk factors for breast cancer. There is no other, credible research to validate this claim in any way".

I would like to point out that this statement does not cite any evidence refuting the claim. It simply questions the study's methodology. So, there seems no basis for saying that bras pose "little or no risk." --Jonathan108 (talk) 01:17, 21 March 2008 (UTC)

One more thing. The "known risk factors" mentioned in the quote include breast size and obesity, both of which are obviously correlated with bra use. If the bra theory is true, these "known" factors would suddenly become suspect. This would be great news, since bra use is avoidable, while breast size usually is not.

It would be easy to design a study that distinguishes the factors by comparing women with similar breast sizes who wear their bras 24 hrs. per day with those who wear their bras 12 hrs. or less per day. It will just take some researchers willing to think outside the box. --Jonathan108 (talk) 19:50, 21 March 2008 (UTC)

Are you for real? Do you really propose that taxpayers' money be spent on this study? Emmanuelm (talk) 20:06, 21 March 2008 (UTC)

"All truth goes through three stages. First it is ridiculed. Then it is violently opposed. Finally, it is accepted as self-evident." - Arthur Schoepenhauer --Jonathan108 (talk) 20:19, 21 March 2008 (UTC)

"All hoaxes go through three stages. First they are invented, second they are ridiculed. Finally, they are forgotten". Myself, today. Emmanuelm (talk) 14:32, 24 March 2008 (UTC)
You forgot stage four: Finally, they are promoted on Wikipedia. MastCell Talk 16:46, 24 March 2008 (UTC)

[Comment deleted by author, because Mastcell's above comment was added out of sequence]

Research is generally conducted by scientists in an academic setting. This is Wikipedia. When there is research, we will report on it. As there is no research at present, we do not report on it. MastCell Talk 16:48, 24 March 2008 (UTC)


MasteCell, please read [1]
OBJECTIVE: To screen high-risk population of breast cancer by analyzing the risk factors of breast cancer in Guangdong Province...sleeping without bra were the protective factors
(Incidentally and as a personal comment: Are out there women who sleep with a bra?. !!!???. Im am stunned. What a strange world. End of personal comment)
I am talking now as a Wikipedia editor: I would really like to see here statistics about how many women sleep with a bra, and why they do such a thing. Never in my life I have stumbled with such a woman.
Premenopausal women who do not wear bras had half the risk of breast cancer compared with bra users (P about 0.09), possibly because they are thinner and likely to have smaller breasts. [2]
Finally: [3]. Unfortunately I have no access from home to that article.
To sum up: The issue of bras and breast cancer is mentioned in the literature, and should -IMO- be cited in the main article, possibly listed as "inconclusive", with a fair presentation of all the contradictory sources. Randroide (talk) 21:11, 23 October 2009 (UTC)
Just because there's a source for a point of view, does not mean it gets given undue weight per wp:fringe. The one scientific survey did not connect bra use with cancer, but weight (a known risk factor). The article for Dressed to Kill does more than enough to cover this topic including the points and counterpoints. We are not here to Teach the controversy, so to speak.Mattnad (talk) 01:17, 24 October 2009 (UTC)~
Mattnad wrote:The one scientific survey did not connect bra use with cancer. Could you please indicate me where can I read that survey?.
Is out there really just one survey stating that bras are not linked with breast cancer. I just presented here TWO studies connecting (inconclusively, I must say) bras and breast cancer. Therefore, strictly speaking, we should conclude that the "fringe theory" would be that bras are not linked with breast cancer.
At WP:fringe I read:
We use the term fringe theory in a very broad sense to describe ideas that depart significantly from the prevailing or mainstream view in its particular field of study.
Could you please source (with scientific studies, please, not with this or that institution or individual say-so) what is the prevailing or mainstream view regarding bras and breast cancer?. I spent two hours at Pubmed taking a look at what is know about this subject and the only studies I found citing (just citing) if there is (or not) a link between bras and BC were the ones I linked above. Maybe I am not very good looking for info at Pubmed, but frankly I am ignorant about where are the other studies supporting the 'prevailing or mainstream view on this issue. Thank you. Randroide (talk) 12:51, 24 October 2009 (UTC)
There reason you found little to nothing in Pubmed is that it's a nonsense hypothesis. You'll also find few studies about the risks of tight pants causing genital cancer etc. etc.. The study I referred to is by Hsieh CC, Trichopoulos D (1991) Breast size, handedness and breast cancer risk, European Journal of Cancer and Clinical Oncology, vol. 27, Issue 2, pages 131–134.
As for mainstream thinking (with links), here you go:
And please don't come back and say something along the lines of "but there's no study disproving a link" -- that's an old red herring tactic and not consistent with scientific method.Mattnad (talk) 14:56, 24 October 2009 (UTC)
Thank you, but you provided no link to the scientific studies that allow those institutions to claim that there is no bras-BC link. Second time I ask: Where are those studies?. If you have no studies you have no scientific evidence. It is as simple as that. The say-so of three institutions is not enough.
Moreover: The (so far) unsupported opinions of three institutions from the USA are not the "mainstream thinking" about nothing (but about USA-related issues). Please read Wikipedia:WikiProject Countering systemic bias. This is not the Encyclopaedia Americana. Randroide (talk) 15:47, 24 October 2009 (UTC)
What you called "nonsense hypothesis" is supported (with some reservations) by two scientific studies. I see no point in calling "nonsense" to hypothesis sourced at pubmed.
This source you linked [4] is patently incorrect/outdated (I do not know what), as I showed here [5] [6].
I think that it is very clear that Wikipedia policies compell us to mention in the article the debate there seems to exist about this issue Randroide (talk) 15:47, 24 October 2009 (UTC)
I'm not sure which policy you're referring to that require a fringe theory to get significant mention in Wikipedia. The scientific studies you have presented did not draw the definitive conclusion you have. As for your concerns that the institutions I provided are American, national origin of an expert source is not an issue for reliable sources. See WP:RS and while you're at it, see WP:OR. You seem to be taking a tack not supported by sources I'd trust more to understand the big picture than you.Mattnad (talk) 07:21, 25 October 2009 (UTC)

1. I am referring to Wikipedia:Neutral point of view. Please read: All Wikipedia articles and other encyclopedic content must be written from a neutral point of view, representing fairly, and as far as possible without bias, all significant views that have been published by reliable sources. This is non-negotiable and expected of all articles and all editors.

2. That there is a correlation (causative or not, that´s another issue) between bras and Breast cancer is NOT a fringe theory. It is a scientific fact published in reliable sources.

3. I have no "definitive conclusion" about this issue. Moreover: What editors think about this or that is totally irrelevant here. I beg you to stop mentioning what you think I think. It is a waste of time to argue about this.

4. Three institutions (two of them just private institutions) from the USA (a Canadian or Chilean institution is also "American") make not a "mainstream view", much less make two scientific studies published by WP:RS a "fringe theory".

5. and American Cancer Society are just two private institutions. I frankly fail to see how the opinion of two private institutions can make "fringe" the results of two scientific studies publised in reliable sources. Please explain how is that possible. Moreover: The American Cancer Society posts patently incorrect information, as I showed above.

6. Seems that "your" sources "know" that there is no link (causative or otherwise) between bras and breast cancer. Could you please explain us HOW do they "know"?. In Science knowledge comes from experiments/studies. Where are those studies, please?. Or their "knowledge" is just their say-so?. If affirmative: Could you please explain us how a say-so can make "fringe" two scientific studies?.

7. WP:OR is irrelevant in this discussion. "My" sources are good according to WP:RS.

8. I do not want (of course) to write in the article that "bras cause cancer". I want to add something like this:

Reliable Sources A and B state that there is a (qualified) correlation between bras and cancer. Sources X, Y and Z state that there is not such correlation. Randroide (talk) 12:05, 25 October 2009 (UTC)
  • For starters, the Hsieh study suggests bra cup size is a proxy for greater weight and obesity which are known risk factors for breast cancer, and that the apparent risk associated with bra use is more likely a reflection of known risk factors. This is the only known study. The other link you provided seems to be an unanswered question and given you haven't read it (as you state above), how can you be so sure it supports your contentions here?
  • Given the paucity of evidence in the literature for a connection between bra use and cancer, we should not give undue weight to weak correlation that has other more obvious explanations (like breast size and obesity), especially since the Hseih study takes pains to mention cup/breast size and obesity as the explanation rather than bra use itself.
  • WP:NPOV does not mean that we included every and any opinion. That's why there's WP:FRINGE.
  • Your comments about US-centrism are off base: Martha Molete of the South African Cancer society also debunked the hypothesis that bras cause cancer. While here article is no longer posted on their site, they too identified this as myth rather than having a scientific basis. And do you have counter example from other non-US health organizations who claim bras cause cancer? I will add that your interpreation of the word "American" is non-conventional. I'm Canadian and I assure you that we never consider ourselves "American" despite your claims.
  • I don't know what version of WP:RS you're reading, but I doubt you'll find editors reading this page who agree with your contention that the National Institutes of Health and the American Cancer Association are not mainstream. Whether or not they are private does not lessen their authority in this matter.
  • Finally, you should read up on Scientific Method. A hypothesis must be validated with evidence to gain consideration and acceptance. What we don't require is for people to disprove an unsupported hypothesis. Regards, Mattnad (talk) 15:01, 26 October 2009 (UTC)
Raindroide, the position statements of major organizations are a highly reliable and very practical source for determining what mainstream opinion is on a point. That's why we use them in this context: we assume that the major institutions have already considered all published evidence for us, and consequently it's not necessary for us to individually analyze the studies. They've done it for us, and told us their answer. WhatamIdoing (talk) 17:24, 26 October 2009 (UTC)
This is pretty much why WP:WEIGHT exists. PMID 19620080 is not useable here. First of all, it cites the following additional "risk factors" for breast cancer: "prophylactic [sic], family history of breast cancer, bad mood, bad life incidence [sic] and work load." Needless to say, we are not about to add "bad life incidence" or "prophylactic" to the list of breast-cancer risk factors on the basis of this article... right? MastCell Talk 18:09, 26 October 2009 (UTC)
I assume that "bad mood" is a lousy translation of major depression, but I can't account for "bad life incidence" without resorting to rather disreputable claims of past life regression. The incidence of lives (good, bad or otherwise) per individual human is generally accepted as being exactly one. WhatamIdoing (talk) 20:40, 26 October 2009 (UTC)
Or is that the prevalence? And what does "prophylactic" mean, as a risk factor, so I can tell people to stop doing/using/being it? How was "work load" quantified? So many questions... MastCell Talk 04:35, 27 October 2009 (UTC)

Thank you very much for expressing your thoughts.

My comments:

For starters, the Hsieh study suggests bra cup size is a proxy for greater weight and obesity which are known risk factors for breast cancer, and that the apparent risk associated with bra use is more likely a reflection of known risk factors

True. Could you please tell us where do you think that piece of information belongs in Wikipedia.

The other link you provided seems to be an unanswered question and given you haven't read it (as you state above), how can you be so sure it supports your contentions here?

I can not be "sure" about the content of a paper I have not read (How could I be?). That´s obvious. In fact I have never suggested that study hinted at a link between breast cancer and bras. I brougth the link to that study here just because -whatever its content- it is relevant to this issue.

Given the paucity of evidence in the literature for a connection between bra use and cancer, we should not give undue weight to weak correlation that has other more obvious explanations (like breast size and obesity)

Excuse me, I quote from [7]:

Premenopausal women who do not wear bras had half the risk of breast cancer compared with bra users

This is NOT a "weak correlation". This is a very strongBold text one. Of course that correlation is not a proof of causation.

I ask you the same question again: Where is the place at Wikipedia for this STRONG correlation?.

Your comments about US-centrism are off base: Martha Molete of the South African Cancer society also debunked the hypothesis that bras cause cancer. While here article is no longer posted on their site, they too identified this as myth rather than having a scientific basis.

AFAIAC that article does not exist, unless someone tells me how I can gain access to it. How can I comment an article I can not read?.

And do you have counter example from other non-US health organizations who claim bras cause cancer?

I have never typed in my life the sentence "bras cause cancer" (now I just had). Please read Straw man.

The Chinese scientists mentioned do not wearing a bra at night as a protective factor against cancer. The guys at Harvard (Hsieh et alii) wrote "Premenopausal women who do not wear bras had half the risk of breast cancer compared with bra users". I want to know where is the place at Wikipedia to add these facts. That´s all.

I will add that your interpreation of the word "American" is non-conventional. I'm Canadian and I assure you that we never consider ourselves "American" despite your claims.

Uh. You are completely right: American (word). Thank you for improving my knowledge of the English language.

I don't know what version of WP:RS you're reading, but I doubt you'll find editors reading this page who agree with your contention that the National Institutes of Health and the American Cancer Association are not mainstream

Two private organizations do not create a mainstream. They much less cancel two scientific studies.

Plase read:

The statement that all or most scientists or scholars hold a certain view requires reliable sourcing. Without a reliable source that claims a consensus exists, individual opinions should be identified as those of particular, named sources. Editors should avoid original research especially with regard to making blanket statements based on novel syntheses of disparate material. The reliable source needs to claim there is a consensus, rather than the Wikipedia editor [8]

I do not need to resort to editor´s opinions. There is a very clear Wikipedia policy that settles the issue: Unless someone can produce a Reliable Source stating that there is a consensus about this matter, no consensus exists for us here.

We assume that the major institutions have already considered all published evidence for us, and consequently it's not necessary for us to individually analyze the studies. They've done it for us, and told us their answer.

I try to assume nothing. My English teacher told us a tongue-in-cheek "ethymology" of the word assume you probably know about.

Anyway, according to Wikipedia policies "my" sources are higher quality than yours. Please read:

In general, the most reliable sources are peer-reviewed journals and books published in university presses; university-level textbooks; magazines, journals, and books published by respected publishing houses; and mainstream newspapers. Electronic media 'ma also be used.

Finally, MastCell wrote:

PMID 19620080 is not useable here. First of all, it cites the following additional "risk factors" for breast cancer: "prophylactic [sic], family history of breast cancer, bad mood, bad life incidence [sic] and work load." Needless to say, we are not about to add "bad life incidence" or "prophylactic" to the list of breast-cancer risk factors on the basis of this article... right?

As a Wikipedia editor, I am not considering those points ("bad life incidence" or "prophylactic") now. Now I am only considering the issue of bras.

If the wording of the chinese study abstract sounds strange to you, that´s fine. I am not going to discuss that wording. The guys who indexed that study are a valid source according to WP:RS, and that´s my point.

Seems we can not reach a consensus here as editors. Sorry for the inconvenience. MastCell, you have a lot of experience. What do you suggest? Randroide (talk) 17:31, 28 October 2009 (UTC)

Per WhatamIdoing, we prefer the findings of expert organizations who look at all of the relevant data rather than an editor's inexpert interpretation of the abstracts of two studies (one of suspect translation quality) which make only peripheral mention of bra-use. I also noticed you are selective about what you quote. The abstract from Hseih specifically says that weight and breast size are the probable explanation for the increased risk. Exactly opposite to your conclusion, this study is cited as evidence that bras don't cause cancer by Molete and the American Cancer Association. Per the American Cancer Association (who actually read the entire study and not just the abstract like you did):
We know of only one scientifically-conducted epidemiologic study that investigated a possible link between bra use and breast cancer. This study suggested that breast cancer might be less common among women who do not wear bras than among bra wearers. However, breast cancer risk in the two groups was not significantly different, according to standard statistical criteria, and the researchers themselves expressed uncertainty regarding this correlation. They also noted that if there is any connection, it would most likely have occurred indirectly, because women who are obese or have larger breasts are less likely to go braless. Obesity has been identified by numerous studies as a breast cancer risk factor, and having large breasts has been suggested in some studies as increasing breast cancer risk for young women who are not overweight.
Given the thousand of other breast cancer studies out there that DO NOT cite bra use as a risk factor, you might understand why the professional groups (and other medical experts) have not made an issue of this. Finally, If you'd like to see a summary of the Molete article, you can read it in the Dressed to Kill (book) article where you'll also find other external links to coverage of this topic. Per your question of "where" this should be in Wikipedia, the book article is a probably the best place.Mattnad (talk) 19:43, 28 October 2009 (UTC)
Raindroide, you might like to read National Institutes of Health. They are not "private organizations". Collectively, they are the largest public scientific organization in the history of the world.
I've read the Molete article in the past, and can confirm that the Dressed to Kill (book) statements supported by it are as true and accurate a summary as you could make without WP:COPYVIO problems. WhatamIdoing (talk) 20:21, 28 October 2009 (UTC)
The molete article is still available cached here if that helps Raindroide. Mattnad (talk) 20:37, 28 October 2009 (UTC)
Yes. It does help. Thank you very much for the effort you made chasing down that elusive article :-)
I depleted my Wikipedia time for today. I share some IMHO very interesting sources I found about bras and health: [9] [10][11]. IMO these sources should be cited at [[12]]
Given the thousand of other breast cancer studies out there that DO NOT cite bra use as a risk factor, you might understand why the professional groups (and other medical experts) have not made an issue of this
Absence of proof is not proof of absence. Please read Argument from ignorance. Personally, I would like to tell you that currently it is impossible to say with certainty that bras cause or do not cause cancer. Simply there is not enough evidence to say one thing or the other. The only thing we can and should make here as Wikipedia editors is to present what said about this fascinating issue this or that valid source. CU. Randroide (talk) 17:46, 29 October 2009 (UTC)
Hi Raindroid, I appreciate your enthusiasm for this point, but this is not a formal logic debate. It's about whether we give undue weight to a hypothesis that very weakly supported by one, and perhaps two studies, but dismissed by mainstream health organizations. So if you don't accept WP:Weight in this circumstance, or the views of three editors who have taken the time to discuss with you, I suggest you consider alternative channels. Mattnad (talk) 21:29, 29 October 2009 (UTC)

Error in wikicode[edit]

The article shows the section "Age" to be followed immediately by the section "Heredity", but when I proceeded to edit the section "Age" I found it to be followed immediately by another section, "Gender". I studied the text in the article and the code in the edit box, and I discovered an apparent error located between "functional impairment." and "Men with gynaecomastia". The code "<ref name="Perkins 2007">" is not followed by "</ref>" before the next occurrence of "<ref>". I do not know enough about editing references or about the data for this reference to correct this error, so I am placing this information here to alert someone who can make the correction.

-- Wavelength (talk) 17:28, 18 July 2008 (UTC)
Thanks, I corrected it, on the re-used ref tags you have to add a closing /, so "<ref name="Perkins 2007">" -> "<ref name="Perkins 2007" /" It's barely noticable even in the diff[13] and tough to spot. -Optigan13 (talk) 05:02, 19 July 2008 (UTC)

Dubiuous#Environmental causes[edit]

[dubious ] Several citations to unreliable source, paragraphs are copies of each other.Nutriveg (talk) 17:44, 14 December 2008 (UTC)

Renaming to Risk factors of breast cancer[edit]

The current name of this article (Epidemiology and etiology of breast cancer) is misleading. Epidemiology in this case rather would be the study field of the risk factors, and its use here is probably confusing for most readers. Etiology is also confusing here, since only ~5% of breast cancers have a specific genetic cause. The rest is a product of risk factors. Indeed, the whole article currently almost exclusively about risk factors. Therefore, I now rename it. Mikael Häggström (talk) 14:18, 18 September 2009 (UTC)

Bad name change, etiology is not just about genetics, but all causes. Renaming is not so trivial to undo, so you should have discussed it first.--Nutriveg (talk) 17:08, 18 September 2009 (UTC)
I apologize if I was too bold in changing the name. It may be discussed. Still, my point is given; The "cause" of breast cancer is rather general carcinogenesis, and can be learned from e.g. the Carcinogenesis article. It can be said that there actually are no causes of breast cancer specifically, but only risk factors - even regarding genetic predisposition. Therefore, all entries in this article may well be classified as risk factors. Mikael Häggström (talk) 21:17, 18 September 2009 (UTC)
I understand your POV, but as long as there are people with different POV than you (like this one) an article should accommodate their views, and from new research. You can spin off this article, but need to copy somewhere else the discussion of etiology you removed from the intro.--Nutriveg (talk) 22:01, 18 September 2009 (UTC)
I think most readers of this article are less familiar with the term etiology than with risk factors. Still, if if the article should be named more like such articles, then Etiology of breast cancer would be an acceptable alternative. I think Epidemiology was the most important term to leave out here, since incidence and death rates are described elsewhere. Mikael Häggström (talk) 05:45, 19 September 2009 (UTC)
The list that was in etiology is found at pathophysiology in breast cancer. Sorry for late response.Mikael Häggström (talk) 05:36, 2 October 2009 (UTC)

Recent edits[edit]

Why did you remove "Ovary removal"/"Ovary oblation" and Insulin? This is a medical article, please follow WP:MEDRS and avoid using websites.--Nutriveg (talk) 17:16, 18 September 2009 (UTC)

They are not removed - you still find them sorted to where I thought they better belong. And, if you mean the organochlorine reference, I admit I simply copied it here when merging List of breast carcinogenic substances with this article, without paying much attention to the reliability of the references. Mikael Häggström (talk) 21:21, 18 September 2009 (UTC)

Iodine deficiency[edit]

A claim was recently made on this page for iodine deficiency as a risk factor for breast cancer. There's very little work in humans to support this, and relevant scholarly articles make much smaller claims, like 'women with breast cancer have more autoimmune thyroid disease than women without breast cancer' -- but without making any claims about causation. (Also, please notice that "autoimmune thyroid disease" is not exactly iodine deficiency.) Most of these articles are also old by our standards. I don't think that iodine deficiency should be listed unless we can get at least one really solid, mainstream, scholarly source that directly says that iodine deficiency is a risk factor. WhatamIdoing (talk) 20:44, 16 May 2010 (UTC)

Replaced the source.--Nutriveg (talk) 20:26, 17 May 2010 (UTC)
Although these sources are newer, you have still failed to address my objections.
Neither PMID 16025225 nor PMID 18645607 claims that iodine deficiency is a risk factor for developing breast cancer; rather, they say that (high levels of supplementation with) iodine might be a useful treatment for existing estrogen-sensitive breast cancer. "Might help kill existing cancer" is not the same thing as "Might prevent cancer from ever occurring." WhatamIdoing (talk) 20:50, 18 May 2010 (UTC)

Why is "Iodine deficiency" in the section "Environmental factors" rather than "Diet"? Should it go next to "Iodolipids"? Norman21 (talk) 20:54, 18 May 2010 (UTC)

Given the lack of acceptable sources (sources, e.g., that directly say something like, "Iodine deficiency is a risk factor for breast cancer", rather than "Iodine might be a useful treatment for breast cancer"), it should simply be removed. WhatamIdoing (talk) 21:04, 18 May 2010 (UTC)
The sources are enough for me and they are also about prevention: "iodine as an antioxidant and antiproliferative agent contributing to the integrity of normal mammary gland", "protective effects of iodine on breast cancer have been postulated".
You can make improvements over that but not simply delete it.--Nutriveg (talk) 12:11, 19 May 2010 (UTC)
"Protective effects have been postulated" is not sufficiently strong information to include it per WP:DUE. Postulation is just another word for speculation. A comprehensive list of all published speculation isn't appropriate for this article. (Otherwise, we'll have to include things like "Cancer as a divine retribution for sins".) WhatamIdoing (talk) 22:42, 22 June 2010 (UTC)

Progress in nutrition[edit]

I suggest that editors read this very accessible review of that dubious "journal". WhatamIdoing (talk) 20:56, 18 May 2010 (UTC)

Please keep discussion on topic, there is nothing about breast cancer in it, and that website is a unreliable source.--Nutriveg (talk) 12:22, 19 May 2010 (UTC)
My point is that this so-called "journal" is an unreliable source with a lousy reputation. WhatamIdoing (talk) 22:36, 22 June 2010 (UTC)

Epidemiology - article name again[edit]

Sorry to jump in so late, but it should really have been called epidemiology rather than "risk factors". Factors like mammographic density, age, gender fit much better into epidemiology than pass as risk factors. Richiez (talk) 16:13, 15 September 2010 (UTC)

I don't agree. A risk factor is anything that correlates with the likelihood of developing the condition. If women with breast cancer were noticeably more likely to drive red cars, then "driving a red car" would be a legitimate risk factor. WhatamIdoing (talk) 19:25, 15 February 2011 (UTC)


The tobacco stuff is out of date. It turns out that the "suggestive" link to tobacco disappears once you control for alcohol consumption. I've mislaid the source for the moment, but this comment is an aide mémoire to make sure we get that fixed. WhatamIdoing (talk) 17:47, 18 April 2011 (UTC)

French E3N Study[edit]

Someone smarter than I should add a synopsis of the findings in this important study to the discussion: (talk) 17:36, 13 July 2011 (UTC)


This is a significant issue that needs to be more closely examined with a NPOV. Given the increadibly large number of women taking hormonal contraceptives and the as yet undetermined reason for continued increase in breast cancer, as a scientific community, we need to keep on open mind toward all potential causes. We need to be careful that we are not unduly swayed by the position statements of various groups who tout themselves as experts who tend to downplay an association of hormonal contraception and breast cancer. The 1996 study has flaws that Khalenborn's study addresses quite well as was included my edit with 2 additional references (in addition to Kahlenborn's Mayo Clinic study).

Mastcell stated "It seems to me to ignore a large portion of available evidence and expert opinion in the field." However, I didn't delete the prior commentary on the 1996 British study so as to allow it and its commentary to stand on its own merit. I think an objection(NPO) approach would be to leave it as my last edit did which presents the key finding of each study and supporting/referrenced commentary. If additional studies specifically point out methodological flaws or problems with another study (Kahlenborn's for example), then we can add it too and let it stand on its own.

Finally, we need to be careful of ad hominum attacks against investigators who have published material challenging what had previously been considered conventional wisdom. Describing someone as an "anti-abortion activist" is no more relevant to the data than whether or not he or she voted for Obama in the last election. Would you tolerate an editor discrediting references because of accusations that the primary author is a "pro-abortion activist,"? In many areas wikipedia has been able to rise above political, financial, and ideologic special interests. Whether it's Kahlenborn's research on contraception of Brind's on abortion, we should assume good faith in their work and consider it as objectively and openly as any other. No person or group is going to be completely free of bias, in fact the larger and more powerful the organizations are that promote one angle and interpretation, the more likely we should be willing to consider the counterargument and give voice to their valid findings. Nothing less than the physical health and well-being of millions of women is at stake hear. Evidence of a link between hormonal contraception and breast cancer is a mightily inconvenient truth if one is a deeply committed "pro-pill activist." However, in the final analysis we must ask ourselselves: what is more important? The good name of hormonal contraception or the health of women? Frankgyn 03:51, 22 March 2012 (UTC) — Preceding unsigned comment added by Frankgyn (talkcontribs)

In your first paragraph, you ask us to disregard the opinions of expert groups, apparently solely because their findings disagree with your personal view on the topic. I don't think we should minimize the views of expert groups in favor of a single 6-year-old paper which you happen to favor. We need to summarize the current state of knowledge; there are some guidelines on how to do so here. I think we're all aware of the prominence of this topic and have personal (and in some cases professional) interests in preventing cancer. It's not necessary to resort to borderline-manipulative rhetoric about women's health, and that rhetoric sounds particularly odd when combined with your insistence that we disregard all expert opinion except for the one paper you favor.

Let's talk about ad hominems. There is a difference between an ad hominem and a legitimate concern over conflict of interest. Again, you try to have it both ways: you point to drug-company funding to discredit findings you disagree with, but you take umbrage at similar concerns when they're directed at your one favored source (the 2006 Kahlenborn paper, PMID 17036554). When drug companies fund research, that funding is disclosed for a reason - because it's vital to help the reader weigh the evidence, and because considering conflicts of interest is not an "ad hominem" fallacy, but an important part of critical reading.

Let's talk sources. Why, again, should we feature Kahlenborn's 2006 paper to the exclusion of virtually all other sources? What about material from major expert bodies, like the National Cancer Institute? What about more recent literature? There have been many review articles on the topic since 2006, including:

  • Lech et al. 2006, PMID 17056445
  • Cibula et al 2010, PMID 20543200
  • Gaffield et al. 2009, PMID 19751860 (deals specifically with women with a family history of breast cancer, who are presumably at higher-than-average baseline risk)
There are also a number of major studies published since 2006, some including more than 1 million patient-years, virtually all of which show no association between oral contraceptives and breast cancer. Arguably, an up-to-date meta-analysis, including these newer massive studies, would reach different conclusions that did the 2006 Kahlenborn paper, emphasizing the need to use up-to-date evidence. See:
My point is that there is both a wealth of new data since 2006, and a number of more recent review articles... not to mention material from expert bodies like the NCI. The focus of these recent edits has seemed to be to highlight the 2006 Kahlenborn paper and to simultaneously criticize and downplay all other cited works. I don't think we're actually doing a good job informing the reader by quoting that one somewhat dated paper (which happens to reach different conclusions from the majority of other literature). MastCell Talk 05:09, 22 March 2012 (UTC)

MastCell, thank you for the thoughtful feedback. To clarify, I never intended my statement to "keep an open mind toward all potential causes," and "to be careful that we are not unduly swayed by the position statements of various groups" to mean that we should "disregard the opinions of expert groups" as you suggest. It's not about one paper that I personally favor, it's about a willingness to consider compelling data on its own merit. I will review the more recent studies you have graciously listed. Do any of them address the methodology or findings of the Kahlenborn paper?

As for "ad hominem" (thanks for the spelling correction, by the way), I agree with you that there is an important difference between personal attacks (i.e. he's a anti-abortion activist implying that therefore we can't trust him) and legitimate conflicts of interest (as in the case of large corporate bodies with substantial funding from the pharmaceutical industry who have a clear vested interest in publishing data in support of their products). An ad hominem (Latin "to the person") fallacy involves the attack of an individual person (again "anti-abortion"), not so much applicable to an institution/organization , whereas "conflict of interest" occurs when an individual or organization is involved in multiple interests, one of which could possibly corrupt the motivation for an act in the other." (wiki definition). I don't doubt the sincerity of your interest in objective truth (and I trust you don't doubt mine either). However, I think in this venue, we need be especially careful about ad hominem attacks as well as conflicts of interest (COI) problems. For example, if you can make a case that Kahlenborn has a COI, in his research, by all means do so and we can engage in a healthy debate on that point.

I'll play devil's advocate: Do you think his personal opposition to the current legal status of abortion motivates him to show a connection to contraception and breast cancer? If so, should we be be investigating every one else's personal beliefs about abortion as a sort of litmus test before citing their work?--Frankgyn 16:16, 22 March 2012 (UTC) — Preceding unsigned comment added by Frankgyn (talkcontribs)

If someone is prominent enough as a political activist for one side that it is not necessary to "investigate" their "personal beliefs" in order to know what they are (and Kahlenborn is much, much more prominent as an activist than a researcher), their objectivity as a researcher is necessarily called into question. –Roscelese (talkcontribs) 20:39, 22 March 2012 (UTC)
(edit conflict) Response to Frankgyn: Both financial and ideological conflicts of interest can be problematic. My personal viewpoint is that the latter are potentially more problematic, because (in contrast to financial conflicts) ideological conflicts are typically not disclosed to the reader.

Secondly, I think there is a distinction between the roles of researcher and advocate. In a field with political ramifications, one can still often distinguish between people who are researchers first and advocates second, vs. those who are primarily advocates and whose research efforts are limited to the service of their advocacy. Of course, this distinction is somewhat subjective, and largely for that reason I think you're right that this isn't a particularly productive road to go down in this setting.

Moving on to more concrete territory... to address your question about whether later meta-analyses cite and/or address Kahlenborn's findings: the 2010 review (PMID 20543200, full text) commented that the Kahlenborn study "used only case–control studies and crude odd ratio (not adjusted), which could have increased the RR [relative risk] values."

I'd be interested to know what you think: given that a number of large studies have been published subsequent to the 2006 Kahlenborn paper (generally showing no association between oral contraceptives and breast cancer risk), and given that more recent review articles conclude that "in a majority of studies there is no increase in the risk of breast cancer reported in OC users", how should we accurately convey up-to-date medical opinion to the reader? MastCell Talk 21:00, 22 March 2012 (UTC)

Richiez and I did some work on BRCA mutation a while ago, and it appears to be generally accepted that hormonal contraception somewhat increases the risk of breast cancer in that small population, especially when taken as a teenager or as a middle-aged woman. (It is also generally accepted to decrease the risk of ovarian cancer, so the recommendation is that all BRCA carriers take five years of hormones between the ages of 25 and 35.)
The reason this sticks in my mind is because of a nasty catch-22: teenage girls with BRCA mutations are at risk for developing breast cancer at an unusually early age if they take birth control pills (or other hormone treatments), but the medical ethicists forbid these same teenagers from finding out whether or not they've inherited the mutation until they're legally considered to be adults. The ethicists have, in effect, demanded that informed consent not be permitted in this population. WhatamIdoing (talk) 22:57, 22 March 2012 (UTC)
I'm not sure it's correct to say that oral contraceptives increase breast cancer risk in BRCA mutation carriers. The most recent meta-analysis showed no risk increase with modern contraceptives, and any risk increase seemed to be limited to older (pre-1975) formulations (PMID 20537530). But I think it's fair to say that it's a controversial issue. MastCell Talk 04:59, 23 March 2012 (UTC)
That doesn't seem to address the question of age at all. Getting breast cancer when you're 20 is different from getting breast cancer when you're 60. WhatamIdoing (talk) 05:15, 23 March 2012 (UTC)
I might be misunderstanding your point, but the meta-analysis did look at the effect of age at which BRCA mutation carriers began taking oral contraceptives (Table 3). They didn't find a significant difference in risk between women who started OCPs at age <20 and those who started an age >20. But I'm not sure that's what you're getting at? MastCell Talk 19:06, 23 March 2012 (UTC)
I'm more interested here in the difference in the age at which breast cancer is diagnosed, not the age at which the OCPs are taken. The lifetime risk for BRCA1 carriers is already so sky-high that a tiny uptick would be lost in the noise (although at least the older sources have reportedly found such an uptick), but the difference between being diagnosed with breast cancer when you're 20 years old and being diagnosed with breast cancer when you're 60 years old is not something that will be lost in the noise. WhatamIdoing (talk) 00:02, 24 March 2012 (UTC)
Ah, got it. I'm not sure how much evidence there is on the effect of OCPs on age at breast-cancer diagnosis in BRCA mutation carriers. There's Milne et al. 2006 (PMID 15734957), which found no increase in "early-onset" breast cancer, but they didn't really dig into age at diagnosis. Haile et al., 2006 (PMID 17021353) found no increased risk of breast cancer before age 50, which gets at the question at least indirectly. MastCell Talk 06:05, 24 March 2012 (UTC)

I hope that we can at least agree to some mutual acknowledgement that hormonal contraceptives do increase the risk of breast cancer, while also simultaneously agreeing that the exact degree of increased riks is indeed debatable with different studies showing varying results. In this vain, I also believe we should replace the WHO statement removed by Roscelese. The readers (women especially) are entitled to a fair representation of range of data available to date. Perhaps a statement on the WHO positions could be clarified that the initial "Group 1" desisgnation to OC's was first assigned in 1999 (when I believed HRT was also evaluated), and then re-affirmed in 2005. Frankgyn (talk) 22:14, 23 March 2012 (UTC)

My point is that there doesn't seem to be consensus that oral contraceptives increase the risk of breast cancer, although the subject is clearly controversial.

I'm not sure that emphasizing the WHO classification is the best way to inform the reader. The WHO classifications don't address the risk-benefit balance of specific medications. After all, oral contraceptives are clearly powerful anti-carcinogens when it comes to, say, ovarian cancer. And the effect of oral contraception on overall cancer risk is neutral to possibly slightly beneficial. At least some groups have expressed the view that WHO carcinogen ratings have no clinical relevance whatsoever (e.g. PMID 18335326). From a real-world perspective, there is no evidence that oral contraceptives increase the overall cancer risk, so highlighting them as "carcinogens" runs the risk of misinforming by omission. MastCell Talk 22:47, 23 March 2012 (UTC)

Well... I find it a little difficult to believe that OCPs, when taken during perimenopause and labeled "birth control pills" have a materially lower cancer risk than the same hormones taken at the same time and labeled "hormone replacement therapy", especially since the OCPs are a higher dose. How about you?
I think it's generally accepted that, for the typical woman, taking OCPs in her 20s or 30s does not increase her risk of post-menopausal breast cancer. But I'm not sure that the same can be said about taking OCPs in her 40s and early 50s. WhatamIdoing (talk) 00:11, 24 March 2012 (UTC)
There's certainly less evidence about women who take OCPs in their 40s and 50s, particularly since for a long time the FDA set the upper age limit for OCPs at 35 for smokers and 40 for healthy non-smokers. It's actually not hard (at least for me) to believe that OCPs have a different risk profile than post-menopausal hormone replacement. OCPs are given pre-menopausally, when women are producing endogenous estrogen. Hormone replacement therapy is given post-menopausally, at a point when women have stopped producing endogenous estrogen, so it artificially extends a woman's exposure to estrogen. It's entirely plausible for that to result in different risk profiles for pre- vs. post-menopausal estrogen administration. MastCell Talk 05:44, 24 March 2012 (UTC)

I have to agree with WhatamIdoing on her point. If the evidence that HRT (whose estrogen and progesterone levels are much lower than even the lowest dose OCP's) was found to be so harmful that the WHI had to be ended early, it would certainly seem biologically plausible that the much higher doses of estrogen and progesterone of more synthetic varieties would carry substantial risk for increased breast cancer. Mastcell, the article you stated that dismisses the validity or applicability of the WHO carcinogen lists loses all credibility in the following quote from their abstract: "Equating natural compounds like estradiol with defined carcinogens like asbestos..." The inexcusable misrepresentation here is that the hormonal components of OCP's are human-identical molecules. If we are making appeals to expert groups, I think we should be more willing to give ear and print reporting of the findings of this body whose work in this area is to advise the international public of well-established carcinogenic compounds. If saw dust is on that list, I'll be more careful to where a mask to prevent inhaling such. If salty fish in on the list, I'd do well to minimize or avoid consumption of excess salt in preserved fish. As for the ovarian cancer vs breast cancer issue. We need to seriously look at the whole picture and take into consideration how much higher the incidence of breast cancer is than ovarian cancer. Furthermore, we need to consider the increased risk of cervical cancer and liver cancer associated with OCP's while factoring in the decreased risk of uterine cancer. The last time I reviewed such data, the net result leaning heavily in the net increase in cancer--owing in large part to the already high baseline incidence of breast cancer. Frankgyn (talk) 02:09, 24 March 2012 (UTC)

It's entirely plausible for OCPs (given pre-menopausally) to have a different risk profile than post-menopausal hormone replacement. OCPs are given at a time when women are producing endogenous estrogen. Post-menopausal hormone replacement, on the other hand, artificially extends a woman's exposure to estrogen. But that's beside the point.

As for your last sentence: when was the last time you reviewed the data? There is absolutely no evidence that modern oral contraceptives increase overall cancer risk in Western populations. As far as I'm aware, no reputable expert body holds such a position. The largest and most recent study to address the question (Hannaford et al., BMJ 2007, PMID 17855280) found the opposite: an absolute reduction in overall cancer risk in women who used OCPs. Their data suggest that for every 10,000 women who use OCPs, between 10 and 45 cases of cancer will be prevented.

Nor is there any evidence - at all - that modern oral contraceptives increase a woman's risk of dying. Again, if anything, the opposite appears to be true: mortality is generally lower among OCP users. See Hannaford et al., BMJ 2010 (PMID 20223876); Vessey et al., Lancet 2003 (PMID 12885478); Colditz, Annals of Internal Medicine 1994 (PMID 8154642); and so forth. There is a well-known risk to women over 35 who smoke, and these women aren't typically prescribed OCPs, but in the general population I'll reiterate that data do not show an increased risk of cancer overall, nor of death, in OCP users, and may show a benefit for both outcomes. I'm a bit concerned that we're cherry-picking superficially alarming factoids at the expense of providing a thorough and accurate overview of the topic. MastCell Talk 05:44, 24 March 2012 (UTC)

Women do not stop producing estrogen at menopause. The amount of estrogen declines around menopause, but they never stop entirely, and the decline takes several years (after menopause) to reach its new baseline.
My point is this: the very same pills (which the manufacturer intends to be used as OCPs) are given both labels when they're prescribed to perimenopausal women (who are first pre-menopausal and later post-menopausal, and it takes up to a year to recognize the transition clinically). A perimenopausal woman who is using OCPs to control symptoms is not only getting higher levels of these hormones than she would if she took proper HRT pills to control her symptoms, but she's also getting these much higher levels of hormones than she would on HRT when she is post-menopausal.
So despite the official recommendations about what physicians "should" prescribe, she's actually post-menopausal, and she's actually taking OCPs. It's illogical to assume that her risk of breast cancer is not affected by these OCP-delivered hormones.
I think that as a matter of what is physically possible, OCPs are capable of producing breast cancer in certain populations (e.g., if a post-menopausal woman is taking OCPs because she doesn't realize that she's actually post-menopausal, or perhaps if she heard that "HRT" is bad, but "OCPs" are perfectly safe [or cheaper, or easier to get]). It's just that for the most common uses of OCPs (i.e., in younger women), this doesn't appear to be a significant risk. WhatamIdoing (talk) 14:03, 24 March 2012 (UTC)
PMID 20714815 gives two circumstances in which OCPs are a risk factor for breast cancer:
  • current use, and
  • use before the age of 20.
In both cases, these are primarily pre-menopausal breast cancers, i.e., the kind of breast cancer that most patients don't get, and the increases aren't dramatic.
Of course it may well be that it's a case of trading breast cancer for some other kind of cancer (see PMID 20543200), but the chemoprotective value is outside the scope of this particular article. But I think it would be reasonable for us to say more or less what we're saying now: OCP use seems to cause a small proportion of breast cancers. WhatamIdoing (talk) 04:55, 2 April 2012 (UTC)
There seems to be another paper out that addresses this point; OCPs are reported in the popular press as giving women in their 40s a 30% increase in relative risk (over a not very high baseline) for breast cancer. WhatamIdoing (talk) 16:42, 5 May 2012 (UTC)


The most obvious risk factor for breast cancer (besides female sex and having breasts, and right up there with family history) is country of residence. In north and west Europe, age-standardized incidence is in the order of ten times that of south and east Asia. There is even a ASI gradient across Europe, from West to East. The importance of environmental factors is shown in the migration effect, in that a woman migrating at a young age from low- to high-incidence countries acquires the local risk over a period of decades. Geographical and temporal clustering or changes in risk, with migration effects are telling us that there is an important environmental factor. — Preceding unsigned comment added by (talk) 15:53, 26 May 2013 (UTC)

I wonder how women's alcohol consumption compares in NW Europe to SE Asia. WhatamIdoing (talk) 22:58, 26 May 2013 (UTC)