Talk:Abortion–breast cancer hypothesis/Archive 5

Archive 1

←

Archive 3

Archive 4

Archive 5

Breast cancer

I changed the link * Breast Cancer Prevention Institute from the section "pro-life" because after reading about them and studying their page I found no link between them and pro-life movement. They have a position about the correlation between abortion and cancer, but they have not a pro-life approach. If they are in the "prolife" section all the others that are not in the "pro-choice" as the WHO should be moved. —Preceding unsigned comment added by 130.233.34.164 (talk) 09:49, 4 January 2011 (UTC)

Dr. Brind is pro-life as is its other key personnel. - RoyBoy 05:12, 17 January 2011 (UTC)

sources for organizations' opinions

Hi,

In the lede, there is a sentence:

This consensus is supported by major medical bodies, including the World Health Organization, the U.S. National Cancer Institute, the American Cancer Society, the American College of Obstetricians and Gynecologists, and the American Medical Association.<ref name="WHO">"WHO – Induced abortion does not increase breast cancer risk". who.int. Archived from the original on December 14, 2007. Retrieved 2007-12-24.</ref><ref name="rcog_2000">"The Care of Women Requesting Induced Abortion" (PDF). Royal College of Obstetricians and Gynaecologists. p. 9. Retrieved 2008-06-29. Induced abortion is not associated with an increase in breast cancer risk.</ref><ref name="oversight">"Politics & Science – Investigating the State of Science Under the Bush Administration". oversight.house.gov. Archived from the original on March 27, 2008. Retrieved 2008-04-14.</ref><ref name="JASEN"/>

I suggest it would be useful to have a direct source for each medical body: like a press release or a fact sheet they have issued.--Kevinkor2 (talk) 14:01, 11 January 2011 (UTC)

By the way, it is also useful to have general links so that readers can turn up more organizations on their own.--Kevinkor2 (talk) 14:30, 11 January 2011 (UTC)

WHO

Current source is <ref name="WHO">"WHO – Induced abortion does not increase breast cancer risk". who.int. Archived from the original on December 14, 2007. Retrieved 2007-12-24.</ref>

Yay! We can simplify this source to <ref name="WHO">"WHO – Induced abortion does not increase breast cancer risk". who.int. Retrieved 2011-01-11.</ref> because it is back on WHO's web site. There is no need for the archive URL.--Kevinkor2 (talk) 14:30, 11 January 2011 (UTC)

National Cancer Institute

I suggest that <ref>"Abortion, Miscarriage, and Breast Cancer Risk". National Cancer Institute. Retrieved 2011-01-11.</ref> would be a good reference for this. It has a statement:

In February 2003, the National Cancer Institute (NCI) convened a workshop of over 100 of the world’s leading experts who study pregnancy and breast cancer risk. ... They concluded that having an abortion or miscarriage does not increase a woman’s subsequent risk of developing breast cancer.

--Kevinkor2 (talk) 14:30, 11 January 2011 (UTC)

American Cancer Society

I suggest <ref>"Is Abortion Linked to Breast Cancer?". American Cancer Society. Retrieved 2011-01-11.</ref>. Conclusion:

At this time, the scientific evidence does not support the notion that abortion of any kind raises the risk of breast cancer.

--Kevinkor2 (talk) 14:30, 11 January 2011 (UTC)

American College of Obstetricians and Gynecologists

I suggest <ref>"ACOG Finds No Link Between Abortion and Breast Cancer Risk". American Congress of Obstetricians and Gynecologists. July 31, 2003. Retrieved 2011-01-11.</ref>. Lede sentence:

There is no evidence supporting a causal link between induced abortion and subsequent development of breast cancer... .

--Kevinkor2 (talk) 14:30, 11 January 2011 (UTC)

American Medical Association

Unlike the other groups, a quick search did not turn up an obvious source.

Here is the search I used: http://search0.ama-assn.org/search/search?query=abortion+%22breast+cancer%22&database=-1

--Kevinkor2 (talk) 14:30, 11 January 2011 (UTC)

I found an article in WorldNetDaily:

Can doctors be sued over abortion?

http://www.wnd.com/news/article.asp?ARTICLE_ID=26970

A spokesman for the American Medical Association told WND that the organization "doesn't have a policy at all" on whether doctors should inform women about the potential link.

He said the group "only proposes policy changes twice a year," but that he couldn't recall "anyone having proposed that we introduce any resolution that addresses this issue."

Now, I know that WorldNetDaily is considered by many Wikipedia editors as an unreliable source, but a surface reading of AMA's policies on abortion seems to confirm the report:

• Query: http://search0.ama-assn.org/search/pfonline/?chkALL=ALL&query=abortion
• policy H-5.990: The issue of support of or opposition to abortion is a matter for members of the AMA to decide individually, based on personal values or beliefs. The AMA will take no action which may be construed as an attempt to alter or influence the personal views of individual physicians regarding abortion procedures.
• policy H-5.995: Our AMA reaffirms that: (1) abortion is a medical procedure and should be performed only by a duly licensed physician and surgeon in conformance with standards of good medical practice and the Medical Practice Act of his state; and (2) no physician or other professional personnel shall be required to perform an act violative of good medical judgment. Neither physician, hospital, nor hospital personnel shall be required to perform any act violative of personally held moral principles. In these circumstances, good medical practice requires only that the physician or other professional withdraw from the case, so long as the withdrawal is consistent with good medical practice.
• policy E-2.01: The Principles of Medical Ethics of the AMA do not prohibit a physician from performing an abortion in accordance with good medical practice and under circumstances that do not violate the law.

I recommend that we remove American Medical Association from the list of medical bodies that address the alleged link between abortion and breast cancer.

--Kevinkor2 (talk) 14:12, 17 January 2011 (UTC)

I removed American Medical Association in this article and also in the main Abortion article. --Kevinkor2 (talk) 06:35, 27 January 2011 (UTC)

Not direct but still relevant, [1], surely the AMA wouldn't support an operation that they thought would cause breast cancer. WikiManOne (talk) 06:35, 27 January 2011 (UTC)

Also, [2] is from the Journal of AMA.

Article title

I'm sorry if this has been discussed before, but searching the archives found nothing. I think the article title should be Abortion – breast cancer hypothesis, i.e. a spaced en dash, per WP:DASH:

• "Spacing: Disjunctive en dashes are unspaced, except when there is a space within either one or both of the items..."

I know there are occasional exceptions for spaced surnames, but I can see no reason for an exception in this case. Thoughts? --RexxS (talk) 16:38, 11 January 2011 (UTC)

Looks right to me -  Done. - 2/0 (cont.) 07:01, 13 January 2011 (UTC)

Moved the archive subpages as well. --Kevinkor2 (talk) 14:38, 17 January 2011 (UTC)

Oops, forgot about that - thank you. - 2/0 (cont.) 23:03, 20 January 2011 (UTC)

Lead line on response bias

Hello, it was a good call to single out "felt" to be changed, however "concluded" infers they have evidence to say response bias is the cause of positive ABC findings. It's a consensus of the NCI workshop yes, but without positive evidence. Moreover, Melbye contains positive findings; so it isn't black and white. - RoyBoy 18:47, 29 May 2011 (UTC)

Changing it to considered is pretty slick KillerChihuahua, had to think on that one. I'm uncertain why compromise is needed or a full record. If you would like to verify what studies/evidence they actually used for "considered" response bias, no magic bullet exists, no special evidence... they made an educated guess based on existing response bias studies who's datasets (populations) are unrelated to the studies they are critical of. They are listed in this article with details. If the workshop was scientific, "considered" wouldn't fit, that's something laymen and politicians do prior to deciding to take an action. - RoyBoy 04:10, 17 August 2011 (UTC)

When I read this definition for considered, "thought about or decided upon with care: a considered opinion" I felt the need to revert. Using a dataset selected to find response bias; doesn't qualify as careful, nor does it demonstrate bias in studies that endeavored to avoid bias(es) with the case-control methodology. - RoyBoy 04:20, 17 August 2011 (UTC)

Separately OM remains incorrect, as there are no "results" published clarifying how much response bias may actually effect those individual interview case-control studies. The overall concept is based on verified poor studies that show response bias exists in the world in various populations and context(s) (ie. government official interview new mothers -or- a conservative population compared to a liberal one -or- an older cohort when abortion wasn't legal). To conclude/consider this can create positive associations in politically mixed populations, who are not new mothers, dataset not weighted with old individuals, has no substantiation. Add to that positive findings in cohort studies presented at the very same workshop, that by design do not have response bias, is possibly dishonest (a full record may clarify that particular peculiarity). - RoyBoy 04:10, 17 August 2011 (UTC)

I don't know what you mean by "slick." Where I am from, that is often an insult. I suggest you avoid the word if you do not intend to insult, or call someone shady, duplicitous, or otherwise imply dishonesty.

I offered what I hoped would be acceptable phrasing to both parties. You don't seem to want "conclusion" and "hypothesis" is, frankly, unsourced and therefore unacceptable, which is probably why OM, and now AM, and I, have removed it. Do you have a source which states a new hypothesis was formed? If so, please provide it. If not, please cease adding this word. KillerChihuahua^?!?Advice 14:49, 17 August 2011 (UTC)

I meant it as a compliment (here "that" is slick is good, a "person" not-so-good), it's a fine attempt at a compromise and I had to pause. I don't understand your requirement for a "hypothesis was formed". They stated their position (re: response bias) ... wait a sec I can't find it. I thought it was here. Hmmmm... interesting, we may need to remove completely until this can actually be sourced. I consider it important though, as it addresses the preponderance of interview studies that were used for Brind's meta-analysis. - RoyBoy 03:25, 18 August 2011 (UTC)

(pending finding the source) ... there is no direct evidence to back it up; so "considered" implies great care and validation was done to black list dozens of interview based studies done by scientists, demonstrate otherwise and I'd take your (and OM's) point. A hypothesis is an educated guess, it can compromise "form" a sentence in a study / article by a scientist / or in this case a workshop summary, I have no idea what you're driving at. - RoyBoy 03:25, 18 August 2011 (UTC)

The requested source...

J Natl Cancer Inst. 1996 Dec 4;88(23):1759-64. link

Induced abortion and risk for breast cancer: reporting (recall) bias in a Dutch case-control study.

Rookus MA, van Leeuwen FE.

Department of Epidemiology, The Netherlands Cancer Institute, Amsterdam, The Netherlands.

In general, no association has been found between spontaneous abortion (naturally occurring termination of a pregnancy) and the risk for breast cancer. With respect to induced abortion (termination of a pregnancy by artificial means), the results have been more inconclusive. A positive association was found in five studies, no association was found in six studies, and a negative association was found in the only cohort study. It is thought that part of the inconsistency of the reported results may be attributable to reporting (recall) bias, since all but two studies on induced abortion used the case-control design and were based only on information obtained from study subjects. In comparison with breast cancer case patients, healthy control subjects may be more reluctant to report on a controversial, emotionally charged subject such as induced abortion. Thus, differential underreporting may be a cause of spurious associations in case-control studies.

...snip...

Lindefors-Harris et al.(28) were the first to show that differential misclassification bias is a potential problem in case-control studies eliciting information on induced abortions. They compared information about induced abortions reported by patients with breast cancer and control subjects with data from an abortion registry that was used as the gold standard. Control subjects underreported their abortion history more frequently than case patients. This result is indicative of reporting bias, but the study is not fully conclusive because the registry was not complete. Daling et al.(2) concluded that reporting bias in their positive study on induced abortion and breast cancer was unlikely, since they found no association between induced abortion and cervical cancer in the same population. Although this finding argues against reporting bias, patients with cervical cancer may differ from patients with breast cancer in reporting induced abortions, and, furthermore, the actual association between induced abortion and cervical cancer is not known. In a recent case control study by Newcomb et al.(17), a stronger association between breast cancer and induced abortion was found for abortions performed in the United States before legalization than for abortions performed after legalization (RR =1.4 versus 1.1). This suggestion of reporting bias is in line with our findings.

In conclusion, a reliable abortion registry seems to be essential when studying a sensitive issue such as induced abortion. Thus far, few case-control studies have properly examined the validity of the reported information concerning induced abortions. Our study shows that reporting bias is a real problem and that it deserves more quantitative assessment in case-control studies that are based on information from study subjects only.

Clarity: A hypothesis is not an educated guess...It is an integral part of the only method humans have for acuratly predicing the future...

1. Define a question
2. Gather information and resources (observe)
3. Form an explanatory hypothesis
4. Perform an experiment and collect data, testing the hypothesis
5. Analyze the data
6. Interpret the data and draw conclusions that serve as a starting point for new hypothesis
7. Publish results
8. Retest (frequently done by other scientists)

We call this science and it makes the future clear: Abortion is not a predictor of breast cancer. So this baby seal walks into a club... - ArtifexMayhem (talk) 09:06, 18 August 2011 (UTC)

ArtifexMayhem, Rookus has its own section in this article, as does the "gold standard" Lindefors-Harris study. That means they have been reviewed (a long time ago) and are hardly golden! Don't try to tell me what a hypothesis / science is after using shit research to justify an untenable position, that response bias is statistically significant in the real world. When it comes to ABC research you are a neophyte compared to me. To be clear, I respect/like you... but less after bringing up something blatantly covered by the article as if this should be news to me or somehow persuasive.

Lindefors-Harris misclassified women who said they had an abortion, but their gold-standard records said they did not. So instead of recognizing their national registry was laughably incomplete, they decided to say these women "over-reported" their abortion history. Effectively calling them liars on something that makes no sense to lie about. Also part of their dataset came from a Swedish fertility registry, meaning women were interviewed by a government official (authority figure) shortly after giving birth. Do you think a new mother would be more hesitant (than average) about accurately reporting the abortion history ArtifexMayhem? Are new mothers equally weighted (represented) in the interview based studies being posited as being overwhelmed with bias?

As to Rookus, find out for yourself, do more thorough research ... understand the science, rather than trying to explain to someone who's justifiably insulted by your preceding attempt. To help you a bit, the NCI workshop stopped at step 3, making their position unverified and unscientific. - RoyBoy 23:58, 18 August 2011 (UTC)

AfM is absolutely correct: hypothesis has a very distinct set of criteria. "Considered" otoh, can simply mean "that's what they thought." But we cannot state they posited a new hypothesis without a source. We can say "concluded" but you objected, so I offered "considered." But hypothesis is as narrowly defined as theory in science, and neither means "educated guess" - although "posited" can sometimes mean "educated guess." Your source uses "finding" and "concluded" for example, but not "hypothesis."

Thanks for clarifying the word useage. Teh Interwebz can be a difficult place to communicate at times. KillerChihuahua^?!?Advice 11:18, 18 August 2011 (UTC)

And upon researching the matter, no wonder you were confused! Our Hypothesis article actually had the words "educated guess" - added in 2007 by some random IP, later sourced to a 1811 quote. Oh yeah, a 200-y/o quote is an excellent source. *eyeroll* I don't blame you for thinking that was what it meant, RoyBoy, given that. I've fixed the hypothesis article, btw. KillerChihuahua^?!?Advice 11:32, 18 August 2011 (UTC)

In general parlance (which is the editorial style we lean towards) hypothesis is an educated guess. But since this article has quite a bit of science, I concur a scientific application of the word is a reasonable expectation. The word "posited" is good, I intend to use it after spanking ArtifexMayhem. - RoyBoy 23:58, 18 August 2011 (UTC)

Then I guess we'll peel this onion again. I always enjoy a good spanking. This morning I shot an elephant in my underpants. How he got in there I'll never know. - ArtifexMayhem (talk) 01:22, 19 August 2011 (UTC)

LOLz... sorry, I got pissed, Rookus isn't horrible. Their summation is awesomely reasonable "bias has not been ruled out convincingly", somehow the 7-8% left in Lindefors-Harris mutates in the workshop to "response bias encompasses all positive findings" because they assert (without Publishing results / vote tally) cohort studies show "no" ABC association. Verifiably not accurate, cohort studies show "no significant" association for "first trimester" abortions. You lay out the cycle of enlightenment plain as day above, but here we presume they considered the evidence and reached a conclusion, then forgot to 7. Publish results (they're analysis / findings), because they ran out of coffee!? :") Meaning there is nothing to retest, nothing to question, nothing to validate. A problem, or not? I don't see an onion in this case ... at least not yet. - RoyBoy 03:54, 19 August 2011 (UTC)

No presumption needed...

Summary Report: Early Reproductive Events and Breast Cancer Workshop

Summary Report: Early Reproductive Events and Breast Cancer Workshop Posted: 03-04-2003 Updated: 01-12-2010 link Introduction The Early Reproductive Events and Breast Cancer Workshop convened February 24-26, 2003, and the outcomes of the meeting were reviewed and discussed at the joint meeting of the NCI Board of Scientific Advisors (BSA) and Board of Scientific Counselors (BSC) held March 3, 2003. The Workshop was established to provide an integrated scientific assessment of the association between reproductive events and the risk of breast cancer. Participants represented a diversity of breast cancer expertise, including epidemiologists, clinicians, basic scientists and breast cancer advocates. The Workshop evaluated the current strength of evidence of the characteristics of pregnancy related to cancer (epidemiologic studies), the biologic changes resulting from pregnancy that may be involved in modifying breast cancer risk (clinical studies), and the biologic mechanisms identified (animal studies). This report summarizes the epidemiologic, clinical and animal studies findings related to early reproductive events and breast cancer risk, and each finding is given a Strength of Evidence Rating1. Gaps in research knowledge for each scientific area are identified, and recommendations for future research directions are provided. Epidemiologic Findings • Early age at first term birth is related to lifetime decrease in breast cancer risk. (1) • Increasing parity is associated with a long-term risk reduction, even when controlling for age at first birth. (1) • The additional long-term protective effect of young age at subsequent term pregnancies is not as strong as for the first term pregnancy. (1) • A nulliparous woman has approximately the same risk as a woman with a first term birth around age 30. (1) • Breast cancer risk is transiently increased after a term pregnancy. (1) • Induced abortion is not associated with an increase in breast cancer risk. (1) • Recognized spontaneous abortion is not associated with an increase in breast cancer risk. (1) • Long duration of lactation provides a small additional reduction in breast cancer risk after consideration of age at and number of term pregnancies. (1) • Pregnancy-induced hypertension is associated with decreased breast cancer risk. (2) • Maternal DES exposure is associated with an increase in breast cancer risk. (3) ...snip... Boards' Response The NCI Board of Scientific Advisors and Board of Scientific Counselors reviewed and discussed the results of the Early Reproductive Events and Breast Cancer Workshop, and unanimously approved the Workshop findings. One additional gap in our clinical understanding of breast cancer was identified: Do breast cancers diagnosed during pregnancy have different morphologic or molecular characteristics than those diagnosed at other times? It is hoped that the outcomes of this Workshop will help guide the Institute's future research agenda and public communication materials. Regular Evidence Review NCI regularly reviews and analyzes the scientific literature on many topics, including various risk factors for breast cancer. Considering the body of literature that has been published since 2003, when NCI held this extensive workshop on early reproductive events and cancer, the evidence overall still does not support early termination of pregnancy as a cause of breast cancer. To view regular updates on this topic, please go to... Factors of Unproven or Disproven Association: Abortion. 1 Strength of Evidence Ratings: Epidemiology 1 = Well established 2 = Weight of evidence favors 3 = Suggested from human population studies, but speculative 4 = Suggested from laboratory or theoretical considerations but essentially unevaluated in human populations

The considered position of the NCI is very simple: Induced abortion is not associated with an increase in breast cancer risk. The article is the onion. - ArtifexMayhem (talk) 10:25, 19 August 2011 (UTC)

I guess I should have mentioned, we aren't discussing the NCI's overall ABC conclusion, but how to accurately describe their position on interview based studies. Hmmmm, maybe that's the best way to say it: "and the NCI's position on positive findings in interview based studies is they are due to response bias." Is that better? - RoyBoy 02:10, 24 August 2011 (UTC)

If this "publish only the conclusion" logic was applied to the Lindefors-Harris study it's findings would still be 50%. The onion is quite small for the response bias rationale explaining away positive findings. Thanks for hatting the above, so I went to unproven/disproven section, been a few years since I've gone there. It didn't have anything new so I glossed over it last time, but it's all we have on this specific topic. I'll add it to the lead as a ref. Let's get dirty!

As of today we have refs 108 to 113 to look at:

• 108: Rookus MA (1996) -- Have you figured out why this isn't applicable to actual interview based studies?
• 109: Melbye (1997) -- 7 weeks gestation 0.81 RR, 12 weeks gestation 1.38 RR; this is an increase of 57% over 5 weeks. This isn't significant to conclude much, but it does indicate quick use of abortion is well advised, like in China.
• 110: Sanderson (2001) -- Abortion is sanctioned, widely available and required by the one-child policy of the Chinese government. This means the vast majority of Chinese women have the abortion quickly, predominantly after having their first child. This precise situation is unique, and isn't applicable to other countries, though abortion after having children is done elsewhere as we will see.
• 111: Ye Z (2002) -- Ditto, found 2nd trimester abortions had an increased risk; like Melbye (1997). They hypothesize (guess if you insist) this was due to recall bias.
• 112: Mahue-Giangreco M (2003) -- I'm not familiar with this study, and I've spent enough time today on this. If anyone with access to the internet would like to research it, that would be cool.
• 113: Erlandsson G (2003) -- Not familiar, but I decided to research it today given 13 is unlucky. Ah, their dataset is incomplete. [3] Sweden has historically used abortion to limit family size (like China). Abortion history was taken in an antenatal interview (Before birth; during or relating to pregnancy), but any abortion after is not counted. This is a massive gap and means women with abortions after the interview would have been misclassified. Brind wrote a letter, Erlandsson replied. Erlandsson concurs there is a gap (as his study noted then shrugged off), but argues that "less [abortion] exposure" should bring the conclusion to null rather than indicate a protective effect. However, the protective effect observed may simply reflect women having a higher abortion history (at the time of the interview) had more child(ren) earlier in life than average (the control group). Of course it is impossible to say as we cannot verify which women did (or did not) have an abortion.

At first glance they would appear to support "concluded", reading the abstracts may support "considered" (Melbye results), understanding how the studies are not applicable to interview studies with positive findings, and are from countries that use abortion differently, seems closer to "posit" to me. On a study by study basis it mostly relies on Brind's criticism, but I'm comfortable with that given Melbye (1997) shows positive findings are not only in interview based studies subject to (unquantified) response bias. Moreover Rookus 1996 makes it clear response bias has not been established (quantified) to be statistically significant within a population. Something the NCI asserts without evidence. I have to admit, I haven't had this much fun (actual thinking) at Wikipedia in a long time! - RoyBoy 19:17, 20 August 2011 (UTC)

Why are most research references so OLD? (5 years or more) Newton vs Einstein...

As a non scientist, yet a researcher...I was amazed to read this article's enclosed declaration "...There is little or no published research on the topic in the past 5 years." Scanning your references quickly...most articles seem to be from 2008--or earlier!

Are we so SURE that this possible Abortion to Increased Breast Cancer Risk is Done. Settled. A Fully Closed Case? Do we have Absolute Certainty here? Or...politics masquerading as scientific "objectivity"?

Millions of women/girls get abortions. Thus, any possible health risks from abortion should be constantly studied...and regularly restudied. Perhaps this article's almost unabated ZEAL for proclaiming "No Link"...should be toned down to reflect ever changing scientific reality. Which is: New findings appear..Scientific conclusions get amended.

What if---in a different branch of science...we had kept on with Newton's view of physics...and shut our ears to the radical voice of Einstein?Lindisfarnelibrary (talk) 23:42, 7 February 2012 (UTC)

Wikipedia is not a forum for general discussion of the topic. Please make a specific suggestion for the improvement of the article. For instance, a reference to some new scientific study. See also: WP:NOTAFORUM--Harizotoh9 (talk) 19:52, 9 February 2012 (UTC)

I have personally added science done after 2008 to this article, but the ABC issue came to prominence in the 90s, so it isn't surprising a flurry of studies were done then or soon after. Further, Wikipedia makes no proclamations, the NCI and others do. While politics has verifiably effected the analysis of the ABC issue (on both sides), we reflect the current consensus, no ABC link. If this changes, Wikipedia will as well, an encyclopedia does not guess what could happen, we summarize what has happened. As to the Newton v Einstein example, the articles impacted should only change once Einstein's paper(s) were published, not before. Finally, epidemiology studies published over 5 years ago are not "old", if they were on nanotechnology / genetics I'd concede the point, however epidemiology is a mature field. - RoyBoy 22:06, 13 February 2012 (UTC)

JP&S and Brind

Good call on Association of American Physicians and Surgeons, I didn't know in 2006 how conservative and paranoid it could be. But as to why we care about Brind, he is an expert in the field. An expert with an agenda yes, but he's not unique in that regard. - RoyBoy 14:36, 7 July 2013 (UTC)

New paper

I have been informed of WP:MEDRS many, many times, so I will be cautious when dealing with these topics in the future. I have found a recent meta-analysis (i.e. a secondary source) which concluded that "IA [induced abortion] is significantly associated with an increased risk of breast cancer..." I want to know whether it should be included, or whether it challenges the article's current repeated statements that abortion doesn't cause breast cancer. The paper can be viewed here. Jinkinson talk to me What did he do now? 15:52, 11 December 2013 (UTC)

This is of great interest, given its size, location and conclusion. Should wait and see reviews and reaction to this study prior to inclusion to provide context and balance; and avoid recentism. - RoyBoy 04:44, 12 December 2013 (UTC)

Well, I agree we should wait to see whether this paper has any sort of measurable scientific impact before citing it, but it's too late. MastCell ^Talk 07:11, 12 December 2013 (UTC)

Naturally, the right-wingers over at the Washington Times love this paper, and Joel Brind loves it too: [4] Also, it seems this meta-analysis may have been flawed: [5] Jinkinson talk to me What did he do now? 17:27, 12 December 2013 (UTC)

Yes, it's methodologically quite flawed and unlikely to convince anyone with a sophisticated grasp of epidemiology and its pitfalls. But I've long since despaired of the utility of discussing the quality of various studies on this talkpage, and just accepted that someone will eventually throw them all into the article regardless. MastCell ^Talk 23:07, 12 December 2013 (UTC)

If their study / results / flaws become notable, they should be included for completion sake alone, and we provide context etc. Hmmm, wondering out loud if we should mention India & Bangladesh studies as well with their wacky results, Brind dredging them up in reference to this study. Though not notable, so hopefully not. India-Ramchandra Kamath study very small dataset, and tiny control group. Bangladesh-Suraiya Jabeen had a larger dataset, with ABC risk of 2062%, LOLz that's gotta be the best case yet for response bias! Yikes... I wonder if that ironically makes it notable on the ABC issue, dig up the bias there, then you can paint all of Asia with the same political brush. It has the potential to make a better response bias example than Lindefors-Harris or Rookus.

As to sophisticated grasp of epidemiology that should read speculated epidemiology. Hypothesizing "stigma" is responsible for all the positive Chinese result(s) remains a opinion and educated guessing, not science. If there is a study that establishes response bias (and its significance) in one of these provinces for the age group involved, that would be great! Repeating speculation (response bias) repeatedly doesn't magically make it statistically significant in the U.S., Europe or Asia... unless you're in Bangladesh. - RoyBoy 03:30, 13 December 2013 (UTC)

Aditional information must be added

Aditional information must be added, because there is more recent scientific statement on issue: "However, as documented by the Breast Cancer Prevention Institute, the vast majority of studies (57 of 73 worldwide) do show a strong association between IA and an increased risk of breast cancer.[2]" - so it isn't settled problem. See: Abortion and the Risk of Breast Cancer: Information for the Adolescent Woman and Her Parents//American College of Pediatricians – December 2013 [6] — Preceding unsigned comment added by 88.119.128.47 (talk) 10:42, 10 December 2013 (UTC)

The BCPI is known to have always have supported the ABC link. While their observation may be factual, they have a conflict of interest and do not necessarily scrutinize studies prior to repeating positive results. More relevant for Wikipedia, the BCPI isn't widely recognized reliable source on the topic; as an encyclopedia we must give appropriate weight to reliable sources. - RoyBoy 04:38, 12 December 2013 (UTC)

Or, to put the same point more bluntly, the BCPI is an anti-abortion propaganda agency, not a medical or scientific source at all. Goblinshark17 (talk) 06:22, 22 August 2014 (UTC)

WP:SOAP or WP:FRINGE?

I thought the above policy stopped articles like this becoming platforms for "activists".

After the rambling intro, the very first section states:

"A number of major medical organizations such as the American Medical Association, American College of Obstetricians and Gynecologists, World Health Organization and the United States National Cancer Institute have analyzed larger studies regarding induced abortion and breast cancer. Their uniform conclusion is that there is no causative link between induced abortion and breast cancer."

So can anyone then explain why there is a need to then have a 90kb article?? But this is why Wikipedia is a joke. It's tantamount to stating in the article about Elvis that he is dead, but then having another 90kb stating why he isn't!

I will tell you why? It is because articles like this one are platforms for an agenda. The guidelines in WP:SOAP or WP:FRINGE make it clear what to do with articles like this because the "theory" has little or no scientific credibility.

All this article is proving, along with all the other Flat Earth style articles, is the First Follower Theory that states that once any idea (however irrelevant, ignorant or stupid) gets followers it ceases to be bat-shit insane.

This article should be shredded down to the facts what the theory is, it's scientific failings and why some people, irrespective of the empirical evidence, want to believe something because it supports their greater agenda.86.180.228.90 (talk) 13:23, 30 April 2014 (UTC)

I tend to agree with you. But when one Wikipedia editor believes that the scientific community is wrong about something, we tend to end up with these sorts of articles in which they try to prove their case. MastCell ^Talk 16:07, 30 April 2014 (UTC)

There is a big difference between "little" and "no" (flat earth) scientific credibility. Let it be abundantly clear to you the empirical evidence is mixed on ABC risk, but conclusive on causation. Admittedly this could simply be a limitation of epidemiology. But the article details this, and if you bother to scrutinize it, you'd find biology is slightly more hazy than local astronomy; and that agendas exist along the political spectrum. Mastcell, this is about the evidence... remember that next time someone compares this to superstition and folklore. - RoyBoy 03:40, 6 September 2014 (UTC)

Abby Ohlheiser

Why are her opinions published in the mass media quoted? She appears to be a journalist with no relevant academic qualifications. Qemist (talk) 07:09, 10 August 2014 (UTC)

Good point, I'll review it and likely substitute Ohlheiser with an expert opinion(s). - RoyBoy 14:38, 6 September 2014 (UTC)

Beral study indicates the opposite effect--that abortion prevents breast cancer

The Beral study described under the "Meta-analyses" section reports a relative risk of 0.93, which is less than 1.00. Even the entire 95% confidence interval is less than 1.00. This indicates that this study says that abortion prevents breast cancer, although the effect is very weak. For those readers who may not be familiar with epidemiological concepts like relative risk, I have noted this interesting fact in parentheses. Goblinshark17 (talk) 06:19, 3 September 2014 (UTC)

Another study cited in the article--a Scottish record linkage case-control study in the Journal of epidemiology and community health--shows a relative risk of 0.8 for induced abortion, which again indicates a protective effect of induced abortion against breast cancer. Therefore, I am adding a section on studies which show that abortion prevents breast cancer. Goblinshark17 (talk) 15:18, 4 September 2014 (UTC)

It is irresponsible to emphasize Beral. Out of the many studies done on the ABC topic, you've found three indicating the opposite. Not only is this at odds with the majority of the evidence, its been detailed how "protective" effects can be poor statistical adjustment (at least for Beral) The protective effect there obviously is from the difference in child bearing (parity) that were not accounted for, and the protection comes from having children, not having abortion(s). Perhaps something similar befell the Scottish study, either way it should be properly researched prior to being held as some kind of notable result.

I'm removing your brackets, emphasizing and obviously incorrect result from Beral; and removing Beral from your new section. I'll rename the section as well. Unsure if it will survive scrutiny, but I admit the results being different (opposite) is notable insomuch as they are rare. - RoyBoy 03:27, 6 September 2014 (UTC)

Sorry, mixed up the studies; irresponsible to emphasize Harris. That's the one I'll remove. - RoyBoy 03:48, 6 September 2014 (UTC)

I'm not so sure they (studies showing that abortion prevents b/c) are so rare. I have only looked for a short time and only at studies which were already cited in the wiki article. There could be many more. Goblinshark17 (talk) 03:58, 6 September 2014 (UTC)

Forgot to welcome you to the article, Welcome! I changed my mind on Harris for tonight and just toned it down; but repeating those studies in a separate section doesn't sit well with me... unsure if its bad style, or emphasizing potentially bad science. Mastcell, what's your thoughts on that?

Goblinshark17, I'm sure we will disagree on many things (like I do with Mastcell), but I hope like Mastcell you'll ultimately help improve the article. Your enthusiasm is a good start. It is possible I know there are more (not many), especially from China (Asia) as they have historically used abortion differently than Western countries. Bare in mind cohorts (large studies) -- in part -- were done on ABC because the preponderance (quantity) of studies showed a slight ABC association. I've researched this topic for a long time, yes there is more out there (ie. Scottish study)... I'm just skeptical they are studies of note. Also, just because its not in the article, doesn't mean it wasn't considered / reviewed. - RoyBoy 04:25, 6 September 2014 (UTC)

RoyBoy, your proposed new title for my new section "Studies with a Protective Effect" could mean almost anything. How about "Studies which show a Protective Effect"? Or maybe "Studies which report a protective effect"? Goblinshark17 (talk) 04:49, 6 September 2014 (UTC)

Looking at the title again, one of the methods used to keep titles short is to build off the section title. Now that its under "Epidemiological studies", I think we should shorten it to "Protective effect". At the very least "studies" is redundant. - RoyBoy 17:13, 7 September 2014 (UTC)

Could be right, I'm not use to seeing "which" in a title; then again I've been on Wiki hiatus for many moons. - RoyBoy 05:01, 6 September 2014 (UTC)

Discredited, continuing

Someone has removed the second sentence, in which I alert the reader that the ABC hypothesis is at odds with mainstream scientific opinion and contradicted by the major medical professional organizations. As I said before, the reader needs to know this and should not have to wait until the second paragraph to learn it. Accordingly, I am replacing the second sentence. If it is removed again, I will consider that someone is edit-warring against me and respond accordingly. Edit warring is prohibited by Wikipedia policy. Goblinshark17 (talk) 23:10, 7 September 2014 (UTC)

It's discredited

The fact that the ABC hypothesis is discredited should be stated in the first sentence. The reader should not have to wait until the second paragraph to learn that the hypothesis is wrong and that abortion does not cause breast cancer. I have corrected the first sentence accordingly. Goblinshark17 (talk) 20:58, 16 August 2014 (UTC)

This is incorrect. A causal link has been discredited; however whether the risk increases or not isn't certain, likewise with the proposed mechanism. If the mechanism can be found to be wrong or disproven then I would concur. Reverting. Feel free to put discredited in the next paragraph, but do not bold items (reverting "not"), it is poor Wikipedia:Style. - RoyBoy 03:08, 6 September 2014 (UTC)

The reader should not have to wait until paragraph 2 to learn that there is no validity to the ABC hypothesis and that she need not worry about breast cancer if she is considering abortion. Accordingly, I am replacing the word "discredited" in the first sentence.

In fact, there is no need to include the proposed mechanism in the first paragraph. It's not one of the essential things a reader needs to know from paragraph 1. The essential thing most readers need to know from paragraph 1 is that the abc hypothesis is discredited and that if one is considering having an abortion one need not worry about breast cancer. I'm transferring the discussion of the proposed mechanism from paragraph 1 to the section called "Proposed mechanism". That's where a discussion of the proposed mechanism belongs. Goblinshark17 (talk) 04:20, 6 September 2014 (UTC)

That's curious, I don't see any mainstream medical body saying that. Can you clarify your specific reference? I believe they explicitly say causation / link, not correlation. If they said there is "no statistical correlation" that would be incorrect. - RoyBoy 04:36, 6 September 2014 (UTC)

I think we're crossing wires, each editing the TALK article at the same time. I removed my claim above about the mainstream medical organizations at least until I research it further. But I am standing by my proposal to put the discussion of the proposed mechanism in the "Proposed mechanism" section, and to use the word "discredited" in paragraph 1. What is discredited is the idea that having an abortion increases your risk of getting breast cancer. Goblinshark17 (talk) 04:41, 6 September 2014 (UTC)

I can almost see where you're coming from regarding mechanism vs most important message; but I believe this goes back to basic narrative. You need to define what the topic is, prior to saying its wrong, unproven, discredited. Let me create a fictitious topic: Jolt manger is a procedure of jolting mangers with a axe. It's disproven.

It's defined, but in the most basic way; then boom its nothing, move on. :) But hey, I'll see how it looks when you're done. I appreciate you'll now look into what the medical orgs actually say, the position that "abortion increases your risk of getting breast cancer" is discredited scientifically (which is what WP:NPOV cares about) is also untenable. I've acknowledged to others that politically & mainstream media its discredited; but that's irrelevant to NPOV and the scientific evidence as it stands.

It may be helpful, if you also tabulate how many studies show an increased ABC risk; not just a protective effect. Or, are you just interested in studies which show your expected result? - RoyBoy 04:56, 6 September 2014 (UTC)

That's a formidable task.;-) No, obviously I'm interested in studies which show either result, but you and previous editors had already done such a good job of describing studies which find an abc connection that improvement does not seem necessary. However, the reader does need to know that there are studies which show the protective effect as well. Goblinshark17 (talk) 05:08, 6 September 2014 (UTC)

Lolz, indeed it is / was. I had to learn new vocabulary just to fully parse the summaries of the studies. I think I'll end up agreeing on a protective sub-section (I'll move it under Epidemiological studies), but only if its lead gives context by saying how many show no / increased risk. A absolute number may very hard to find, but one could say "majority" or we could grab a numbers from a meta-analysis. - RoyBoy 13:49, 6 September 2014 (UTC)

Separately, after a night sleep looking at the new lead. Wow, that's terrible; unsure why I was so agreeable last night... no I remember, your a bold new editor and we want to encourage you. Reverting: The hypothesis is not discredited either, it remains scientifically valid until shown otherwise. Discredited is down where it belongs, with pro-life pushing an ABC link. I do need to give you credit Goblin (can I call you Goblin?), you've (and the anon / Mastcell above) have given me impetus to create a talk FAQ. - RoyBoy 13:49, 6 September 2014 (UTC)

Two points. ONE: Why include a discussion of the proposed mechanism in paragraph 1? Why not put it in the "Proposed Mechanism" section?

TWO: Regarding FAQ #2. You wrote: "News stories and expert opinions do not supersede the scientific evidence / consensus, which is ambiguous on what risk (if any) there is." Some like the American Cancer Society focus on specific studies to conclude "the scientific evidence does not support the notion that abortion of any kind raises the risk of breast cancer", however this does not reflect the status of the scientific evidence." That's your PERSONAL evaluation of the scientific evidence. You are putting your personal opinion above the ACS, and the Royal College of Obs and Gyns, and the WHO, all three of which deny any increase in risk (not just causal link)! That's unprofessional and it constitutes INDIVIDUAL RESEARCH, doesn't it?

I think we need to appeal this question--whether the first paragraph should say that the ABC hypothesis is "debunked" or "wrong", to a higher wiki authority to resolve this question, as I do not want to get into an edit war with you, but I am still relatively new to wiki and do not know how to initiate an inquiry. Can you initiate one please? Goblinshark17 (talk) 18:20, 6 September 2014 (UTC)

Unfortunately, Wikipedia doesn't have a good mechanism to handle problems like this. It's pretty simple: as you note, this hypothesis obviously lacks any meaningful support in the scientific community. However, RoyBoy (an individual, pseudonymous Wikipedia editor) is convinced that the entire scientific community—the NCI, the American Cancer Society, the WHO, ACOG, etc—has gotten this one wrong. And because he's tenacious, he is able to use Wikipedia as a platform for what is, in essence, his personal essay trying to prove the scientific community wrong. That's what this article is—one editor's extended effort to counter scientific consensus on an issue that he feels strongly about. That runs counter to every tenet of Wikipedia, not to mention any common-sense definition of how a reputable reference work should operate, but there is no solution so long as RoyBoy is willing to keep coming back. Lots of people have come here, made the same points that you (GoblinShark) have, and eventually lost interest and moved on. RoyBoy is still here. And the article still pretends there's an active scientific debate on this issue, because RoyBoy believes there is despite voluminous evidence to the contrary. MastCell ^Talk 20:43, 6 September 2014 (UTC)

Admittedly I should have drafted FAQ with both of you prior to adding it to the talk page. Goblin, have you gotten around to reading what the organizations say, or will you just keep winging it? Clarify for me how WHO "deny any increase in risk". Prior to appealing to a Request for Comment I'd ask you do that much. I'll work on FAQ #2.

RoyBoy, the reference to the first mention of WHO's opinion on the subject (currently footnote number 3) reads: ""WHO – Induced abortion does not increase breast cancer risk". who.int. Archived from the original on 13 January 2011. Retrieved 11 January 2011." From this title I infer that WHO denies any increase in risk. The link itself is dead, so I cannot read the actual statement. I'll see if I can bring it up again by googling, but that will take a little time. Goblinshark17 (talk) 23:48, 6 September 2014 (UTC)

UPDATE: I have searched with google and also searched WHO's own web site but I can find nothing about any alleged abc link. Maybe the documents are now too old to remain online. Goblinshark17 (talk) 01:51, 7 September 2014 (UTC)

FURTHER UPDATE: I found that in the year 2000, the WHO issued a paper entitled "Induced abortion does not increase breast cancer risk". The paper itself is no longer available online, but I think the title alone is enough to infer that WHO takes the position that induced abortion does not increase b/c risk. Goblinshark17 (talk) 02:19, 7 September 2014 (UTC)

FURTHER FURTHER UPDATE: OK, I found it archived, in the second link in footnote #3, indicated by the single word "archived". You are right, the final conclusion is restricted to first trimester abortions. I would argue that since these constitute the overwhelming majority of abortions (at least in USA), the conclusion can safely be generalized to abortions generally. Goblinshark17 (talk) 02:31, 7 September 2014 (UTC)

@Goblinshark17: Another source from the WHO...

Sound epidemiological data show no increased risk of breast cancer for women following spontaneous or induced abortion.

— World Health Organization (2012). Safe abortion : technical and policy guidance for health systems (PDF) (2nd ed.). p. 49.

— ArtifexMayhem (talk) 13:24, 7 September 2014 (UTC)

Thanks for the detailed update(s), :) seems you research as fast as I do. Indeed, the first trimester caveat is one of the last nuggets of scientific honesty on the ABC topic, let the rest (WHO material) fall to the wayside as that's their current position. Mastcell used the same logic. That is wrong and you both should know better, now its you editorializing.

If you genuinely believe this, bare in mind Melbye found a statistically significant result in late term abortions. I find it immensely frustrating otherwise intelligent people will point to record based studies like Melbye and say, see look no significant ABC risk, for the positive results they shrug and ignore them (or in WHO-2012 above, presume them unsound). With positive results ignored / criticized insignificant risk morphs into no risk after sprinkling in politics / editorial. Mastcell (and others) and I have had multiple throw downs on WP:Weight; but these results exist and have not been rejected scientifically. The best on offer is Melbye's expert opinion the late abortion cohort is too small to base a epidemiological conclusion on. I agree 100%, but where we diverge is being able to ignore it outright. I believe WHO's current position acknowledges this indirectly, so should we by not inaccurately conveying a blanket statement from medical associations.

Or to put in short, not all abortions are in the first trimester. - RoyBoy 15:33, 7 September 2014 (UTC)

Are you referencing Melbye 1997? — ArtifexMayhem (talk) 04:11, 8 September 2014 (UTC)

Why is the mechanism discussed? Because the article should define in its lead what "it" is about; prior to determinations of "it". If you remove the mechanism, a reader will be even more under the impression this is just about the pro-life "ABC link". - RoyBoy 22:20, 6 September 2014 (UTC)

Goblin, the "at odds" phrase is better, while I don't see the need for it be the 2nd sentence I can't deny its importance. As to deny, WHO does not deny as they specify "first trimester" in their conclusion. Can you remove that part? - RoyBoy 23:23, 6 September 2014 (UTC)

Mastcell, what do you think of the addition of "and that women who are considering having an abortion do not need to worry about breast cancer when making their decision" in the lead. I feel it bloats it slightly and should go in the Views sub-section. - RoyBoy 23:23, 6 September 2014 (UTC)

RoyBoy, the sentence about women who are considering abortion is of primary importance. The wikipedia article may be read by actual pregnant women who are actually considering abortion and need accurate information. In fact I would speculate that such women are among the most likely readers of the article! Goblinshark17 (talk) 23:48, 6 September 2014 (UTC)

I saw the intent, that why I didn't revert it right away; but I hmmm'd and haah'd as its redundant, heavy handed and slightly condescending. I mean it implies to me, "don't worry about the evidence pregnant women, just look at the medical associations and be done with the topic." I by no means want to concern anyone unnecessarily, but saying "nothing to see here" in the lead... feels wrong, regardless of good intentions. The redundancy nags me (hence bloat, a nono in a lead), "concluded that abortion does not cause breast cancer" immediately before is saying exactly that. Can we shorten to: "and is not a factor for women considering having an abortion"? - RoyBoy 15:33, 7 September 2014 (UTC)

How about "...and should not be a factor for women considering having an abortion"? Goblinshark17 (talk) 17:07, 7 September 2014 (UTC)

Mastcell, I am not a "problem". Sometimes I do make bad edits, but Wikipedia evens that out with others such as yourself. If you and others cannot or choose not to recognize that "no ABC link" does not equate to "no ABC risk", there is a good mechanism to clarify that. This talk page. If the ACS choose to regard Beral study as the gold standard at face value, fine -- but everyone need not follow their example. And generalizing all the conclusions into "no risk" is something a journalist with a deadline and editorial mandate does, not Wikipedia. We can be better than that; I also believe in Wikipedia. It can be at its best when there is disagreement and compromise, again this isn't a problem, its a Feature. Yes? - RoyBoy 22:20, 6 September 2014 (UTC)

RoyBoy, I have inserted two replies to you above, after the points to which they respond. I hope this is not bad TALK PAGE protocol. Goblinshark17 (talk) 23:48, 6 September 2014 (UTC)

Its pretty good, almost missed the reply just above because it was inbetween at the same indent, in that case add an extra indent (just did it) as at first glance it looked part of Mastcell's reply. I prefer the inbetween method myself, because if I reply to you its easier to follow. - RoyBoy 15:33, 7 September 2014 (UTC)

FAQ discussion

I've made changes to the /FAQ, can you please let me know further objections / suggestions? If I have to, we can remove a few points for now to get it re-posted. My initial desire was to address anons cruising by this article and thinking they have a clue because they read a few articles about it. - RoyBoy 22:45, 6 September 2014 (UTC)

No way. The FAQ is incredibly inappropriate. This article is a carefully constructed and lovingly curated violation of Wikipedia's fundamental content policies. Seriously, the article you've built and protected over the years here is a giant middle finger in the face of WP:NPOV and WP:NOR. You've gotten away with it because you're tenacious; other editors come, express surprise and consternation about what you've done here, and then eventually lose interest and move on. I've made peace with it, mostly, because I've learned to pick my battles. But you don't get to write a FAQ giving a quasi-official imprimatur to your violations of site policy. And insofar as we have an ethical responsibility to provide accurate medical information to our readers, you have a lot to answer for. MastCell ^Talk 20:34, 7 September 2014 (UTC)

Studies showing a protective effect, continuing

I have added another study showing a protective effect. This is a Serbian study by Ilic et al.. It is very small and restricted to parous women, but it is notable because of the size of the reported effect: odds ratio is 0.47. Yes, you read that right, 0.47. The women in this study who had induced abortions were LESS THAN HALF as likely to get b/c as women who had no abortions! Ironically, I found out about this study while reading an article by Joel Brind about the ABC "cover up".

Brind made the following criticism of the study: the abortion rate among Serbian women is very high, 89%. So women who had not had abortions were more likely to be infertile or heterosexually inactive women who had had no children. Since having children protects from b/c, Brind says, the protective effect measured was actually the protective effect of having kids.

But Brind overlooked one thing: the study was restricted to women with children. Infertile and heterosexually inactive women were excluded from the analysis! Sorry, Joel, better luck next time.

"The NCI is just another corrupt federal agency like the IRS and the NSA."--Joel Brind.

Goblinshark17 (talk) 06:05, 8 September 2014 (UTC)

UPDATE: I have expanded the "Proponents" section, in particular the description of Joel Brind, adding the fact that he is Christian as well as pro-life (with a reference, RoyBoy). Goblinshark17 (talk) 08:21, 8 September 2014 (UTC)

Australian AMA position

I am not sure this source or this similar source on the AMA website are sufficient to state that the organization itself rejects the ABC hypothesis. Certainly the president does and I would be pretty surprised if the members as a whole generally did not, but I am not finding an actual position paper or other source representing the whole AMA. Removing text for discussion. - 2/0 (cont.) 22:14, 9 September 2014 (UTC)

Hmmmmm. You may be right. Goblinshark17 (talk) 23:08, 9 September 2014 (UTC)

UPDATE: The Canadian Cancer Society, however, is very clear. I have added it. Goblinshark17 (talk) 23:27, 9 September 2014 (UTC)

Looks good to me. Good job! - 2/0 (cont.) 23:51, 9 September 2014 (UTC)

The Cancer Council Australia (Victoria, Queensland etc.) seems to have released a media statement about the alleged ABC link. Relevant for the article? --Sonicyouth86 (talk) 21:20, 10 September 2014 (UTC)

Causative link vs statistical correlation (or "risk")

The following question may lead to improvements in the article:

RoyBoy, you keep saying that a causative link hypothesis is discredited but the "statistical correlation" or "increased risk" hypothesis is unsettled. I would argue that within the field of epidemiology, which looks only at existing cases and does not do controlled experiments, the distinction between statistical correlation and causative link is not valid.

Suppose you observe a statistical correlation between two phenomena. What epidemiological measurement could you make which would tell you whether or not there was a causative link behind the statistical correlation? You could challenge the causative link theory by identifying confounding factors--factors which cause both observed phenomena to occur together without any causative link between them--but what measurement could you make which would affirmatively prove a causative link between the two observed phenomena? There is no answer to this question--in order to establish a causative link, you would need a controlled experiment in which you would impose one of the phenomena and look for the other to occur as a result. This would by definition not be an epidemiological measurement. So I would argue that the distinction you make is invalid, at least within the field of epidemiology.

Comments, replies, or counterexamples? Goblinshark17 (talk) 08:35, 7 September 2014 (UTC)

An important question to pose, I really like your logical inquisitiveness Goblin. In epidemiology they can determine (conclude) a causative link numerous ways, the most simplistic I've come across is 50% increased risk with a Confidence Interval beyond 1.0. Depending on the data set and identified confounding factors this can shift usually up to 100% increase risk. To make solid science it should be replicated.

However, the premise of your question is incorrect as experimentation isn't the only method to determine causation. The Bradford Hill criteria is used to make causal determinations in epidemiology, that's what makes epidemiology science. Specifically I call your attention to biological gradient in relation to Melbye's uncontested results. - RoyBoy 16:55, 7 September 2014 (UTC)

ArtifexMayhem, for the record Melbye contradicts Safe abortion, unless they've included a good reason to call Melbye's record based gradient "unsound"? - RoyBoy 16:55, 7 September 2014 (UTC)

RoyBoy, Safe abortion, may be contradicted by Melbye, but it's still a document issued by WHO which explicitly states that induced abortion does not increase the risk of b/c. Not just denies a causative link, but specifically denies any increase in risk. For abortion generally, not just in the first trimester. That's what you were asking me for above. Accordingly, I will reinstate the sentence saying that WHO denies risk increase. Goblinshark17 (talk) 00:47, 8 September 2014 (UTC)

The phrase "contradicted" seems better to me. I would caution you about cherry picking here, meaning trying to find what you need within WHO to provide an over-arching narrative that isn't entirely accurate. - RoyBoy 23:08, 17 September 2014 (UTC)

Weight

I just removed a large amount of material. Per WP:MEDRS we should not rely on primary studies and we should not rely on old studies. An encyclopedia article is not a historical literature review. Additionally, the material was rife with massive WP:NPOV violations. We do not second-guess sources. We do not re-analyze sources according to our own editorial biases. We do not give a preferred source or point of view the last word on everything. We especially do not qualify the conclusion of a major medical organization with "however". - 2/0 (cont.) 13:05, 9 September 2014 (UTC)

Concur, and thank you for the excellent explanation. — ArtifexMayhem (talk) 19:12, 9 September 2014 (UTC)

Me too, I concur and thank you for the excellent explanation. Let's hope RoyBoy doesn't try to revert your changes or engage in any other funny business. Goblinshark17 (talk) 21:44, 9 September 2014 (UTC)

Solid rationale, though a little cobbled and incorrect; but there was a lot of content so I'll grant cobbled. If you prefer, you can nest responses if easier.

Primary sources, should not; but we can if relevant. Especially if they have notable information: "interesting and in line with the hypothesis of Russo and Russo, the small number of cases of cancer in women in this category of gestational age prompted us not to overstate the finding." We should by-pass this because of editorialized (both in and out of Wikipedia) scientific consensus?

Old studies, which old studies? Bare in mind this is epidemiology, not nanotechnology. Studies don't age the same way, though some methods do.

Historical literature review, while I see what you're getting at this was a fraction of the most relevant research which allowed exploration of core memes on the ABC issue. Interview vs Record, correlation vs causation and most importantly response bias.

NPOV violations, I can see that given your likely application of weight; which sidesteps (removes) what's demonstrated and not. Massive... I take issue with that. Personal issue, however, I'll assume good faith.

Second guess & editorial biases. We (I) took care to show frank expert discussions as part of a narrative(s). Study A, criticism (usually Brind), if another expert or the author(s) counter then its here. Whoever (notable), had the last word, gets the last word. No determination who's right / wrong, last word.

Last word on everything, that's wrong and you should know it. I know it, because I tried to address said objection years ago... with some success. Preferred, no... they (Brind) sometimes get the last word, is that a weight problem? Sometimes he's right. You don't need to be rocket surgeon to figure out "over reporting" in a response bias study wasn't just wrong, it was laughably wrong. Brind said so and the author(s) concurred, while at the same time hypothesizing something else. If the scientific consensus has a reply, then please find it and include it.

Medical organization with "however", running out of time; hmmmm... you do put a however if WHO clarifies "first trimester", or better yet don't make a mostly accurate editorial in the first place.

As to bringing things back, most of the cohorts should be reinstated (even "old" Howe, shows what the fuss was about). I agree with user:Second Quantization, Huang shouldn't be brought back (could even be cited in FAQ of what we avoid).

Everything relating to response bias should be reverted, including Daling. To clarify, response bias is where I see you being incorrect. - RoyBoy 00:09, 18 September 2014 (UTC)

Meta-analyzes, personally I could take or leave them. Problem is, they are (were?) used by a number of sources to substantiate their conclusions / consensus. Seems like the article is incomplete if selection bias isn't explained, and how this relates to meta-studies done on the topic. Prime example of how bias cuts both ways.

Interview, just realized this was completed removed. You've got to be kidding, the primary reason why this topic got any traction with anyone was those "old" studies. Taken individually I'll grant weight, but as a whole, seems to be editorializing to remove them outright. Even the consensus opinions makes note of them within a broader context, as should we, but not just re: bias; but of the arc of research as well. - RoyBoy 01:12, 18 September 2014 (UTC)

It is certainly possible that I cut in places with a machete where a scalpel might have sufficed. I would be open to adding some more material about how the early studies interacted with the culture at the time to produce the movement, preferably sourced to comprehensive secondary sources rather than accompanied by our own editorial analysis and interpretation. What would you suggest? - 2/0 (cont.) 18:21, 18 September 2014 (UTC)

Wow, your Wiki-fu is strong. While mine has admittedly deteriorated since ~2010. As such reviewed the policies in question, with a focus on primary sourcing. I was disappointed to read: "Unless restricted by another policy, reliable primary sources may be used in Wikipedia; but only with care, because it is easy to misuse them.[4] Any interpretation of primary source material requires a reliable secondary source for that interpretation."

Disappointed as I presume you see Brind as not a reliable source? While an essay, WP:Enemy does outline my Wiki-rationale for his prominent inclusion. Personal rationale, is that I know journalists have gotten basics about Brind's position(s) wrong, and this led to my interest in the ABC topic. (BTW, decent Brind article work Goblin) A further question, the Q&A and criticism in a medical journal, is that still a primary or secondary source?

I've dumped the latest studies section here. I'll have to refactor the entire shooting match given Primary sourcing guidelines, likely down to 3-4 subsections providing consensus overview of each study type, its effect on the day, and hopefully a weighted Brind position.

Due warning, I'm going to be vigilant on maintaining response bias content (likely 4th section) & criticism with this in mind, “trust, but verify”. Which is something scientists have neglected to do for ABC response bias. - RoyBoy 14:48, 19 September 2014 (UTC)

Separately, the "overlong" Daling quote is back. That wasn't cool, and I'll go to the mat for it, primary researcher with first hand experience with ABC politics (from both sides). Essential to understanding (or at least being aware) biases occur across the political spectrum on this topic. - RoyBoy 14:48, 19 September 2014 (UTC)

If you want to remove "It's not a matter of believing. It's a matter of what is." then okay, but my preference is simply to have the entire quote per the source. - RoyBoy 14:53, 19 September 2014 (UTC)

I am not sure what you mean by "not cool", as the removal was in my edit summary. Opened a new section for this below.

Brind is reliable in some contexts, but we need to be especially careful in presentation. The Writing for the enemy essay is solid, but I do not see it arguing for a prominent place for Brind's material; rather, I see it urging each of us to make sure the other feels like their favored material is getting a fair shake and is adequately explained.

Part of the point to largely re-writing the article is that we should not be engaging in a "shooting match". I look forward to your work there.

That Economist article is a bit notorious in certain circles. I prefer Why Most Published Research Findings Are False by John P. A. Ioannidis. Please also remember that Wikipedia is not a battleground. - 2/0 (cont.) 16:53, 19 September 2014 (UTC)

A good summary on a bad edit is a bad edit, replied below.

Agreed, it'll be tough but it is what it is. Appreciate it.

Shooting match as in, back and forth? Please elaborate and thank you.

LOL, I think was I trying to find that study when I came across the Economist. Much thanks! While I liked the Economist piece, it was the "trust, but verify" (nutshell) that stuck with me, especially for epidemiology topics, as I knew the "Why Most Published Research Findings Are False" was out there.

Battleground, I've resolved my fair share of disputes. I keep it in mind when others do. - RoyBoy 17:49, 20 September 2014 (UTC)

Rewrite questions

Before I get into this seriously, I need to know how best to proceed with Studies section rewrite.

• Write it chronologically (oldest/newest) or consensus reliability (interviews based, record based, meta-analysis) or scientific reliability (interviews, meta-analysis, record)?

• As secondary sources, can the meta-analysis section remain close to as is; or redone because of shooting match stuff?

• I plan on dumping recent research (ie. anything after Beral) as it doesn't seem to have effected the consensus. Any objections or caveats to that?

• Should I bother including a line about studies showing a protective effect (with context criticism)? Their reliability isn't good, and isn't part of the consensus as far as I can tell.

• Is Q&A and criticism content in journals considered a primary or secondary source?

- RoyBoy 19:04, 20 September 2014 (UTC)

We really need to not recapitulate the entire history of the hypothesis using our own analysis of the primary sources. I think I could like a structure of "this study or set of studies was/were published in $_YEAR, having this effect on the political debate", cited to a secondary source showing specifically the broader effects of publication; a convenience link to the primary literature is good when discussing what sources have said about a particular paper, but we need to be especially careful. This can all go in the #History section; the #Politicization section can probably get the same treatment, come to think of it. The scientific debate itself needs to be presented as essentially settled. - 2/0 (cont.) 19:17, 20 September 2014 (UTC)

I hear you on "recapitulate", but I was asking a simpler question. How best to present the narrative? Is there an FA example to emulate? - RoyBoy 16:29, 21 September 2014 (UTC)

No FA comes to mind, but you might try Vaccine controversies and related articles for inspiration, especially Thiomersal controversy and MMR vaccine controversy. The last article involves scientific fraud, so of course the analogy is not exact. All three topics have significant interplay between scientific research and political and cultural outcry, showing how they feed each other and how a scientifically obsolete paper can continue to have political currency. Disclosure: I have edited this topic area a bit, so reading those articles does not provide a fully independent outside opinion. - 2/0 (cont.) 16:15, 22 September 2014 (UTC)

78 word Daling quote

This quote is overlong and adds little to nothing to the article. Any researcher who is not committing scientific fraud is not "messing with the scientific data to further their own agenda". Daling is a good source, but stating "it sure would be nice if this issue were not politicized" does not further the reader's understanding of any particular point. Giving so much space to a single quote also violates the structural neutrality requirement. - 2/0 (cont.) 16:33, 19 September 2014 (UTC)

That's unexpectedly naive. Daling, as a good source, has a far better perspective of what constitutes scientific "messing" than any of us do (unless you have personal knowledge of how review and criticism works on contentious vs. consensus scientific results). I explained what it adds to the article. Moreover, after your "last word on everything" overstep you're going to remove one of the few alternatives because its long; that's part of the uncool. - RoyBoy 17:41, 20 September 2014 (UTC)

Agree with 2over0, and also note that it is undue weight to the views of one researcher. If we want to discuss politicization of the issue, we should be using overview sources that place all views in context, rather than peacock one view of the many people who have done research here. Yobol (talk) 17:44, 20 September 2014 (UTC)

If you find one, let us know. I don't believe it exists, hence the peacock of one notable view. - RoyBoy 17:56, 20 September 2014 (UTC)

We don't get to throw WP:WEIGHT out the window just because one editor really likes a quote and want to use it. Yobol (talk) 01:53, 21 September 2014 (UTC)

I'm sorry Yobol, can you clarify the policy objection you have here. Melbye's quote is in a secondary reliable source which has context a plenty, its notability on ABC politicization doesn't make this difficult to justify. - RoyBoy 18:26, 20 September 2014 (UTC)

I'm sorry, I have no idea what Melbye has to do with this discussion. My issue is you are cherry picking one quote without regard for WP:WEIGHT. That the lengthy quote also doesn't really add to the readers' understanding of the topic of politicization is just the cherry on top of the objection. Yobol (talk) 01:53, 21 September 2014 (UTC)

My bad, meant to say Daling (not Melbye), reviewing the entire studies section has cluttered me brain. While yes a source properly over viewing the topic is preferred, in lieu of that having Daling is far better than nothing speaking to pro-choice political bias; or is that your intent? One or even many editors "not liking a quote" under the guise of weight is far worse than cherry picking (though I'm unsure how one can cherry pick a complete quote) from a good source. And BTW, using one expert isn't automatically cherry picking, its done all the time in other contexts, ie. criticism of studies, which we don't have a problem with. Daling's quote is also instrumental to establishing some level of confirmation bias in ABC study peer review.

To be clear, removing pro-choice bias from the section does limit the readers understanding on the likely extent of ABC politicization. If you don't personally replace it, it will impact your ability to be taken seriously on the abortion topic, as a start. Also you may have simply missed that the ABC article is about a minority view (generally speaking), as such minority perspectives are given more space by Weight policy (and frankly necessity) to make for a good article. Even User:2over0 might have forgotten on ze first crack at it, as such I can be patient... but resolute (others prefer: stubborn). As to structure, it doesn't have to be a block quote, but according to style I thought it was preferred for long quotes; and I believe it deserves prominence (not just inclusion) in the article. - RoyBoy 16:18, 21 September 2014 (UTC)

What you're proposing is clearly undue weight. You've made it clear that you want to feature this quote not because of its prominence in reliable sources or scholarly discourse, but rather because you personally want to give greater prominence to the viewpoint in question. I agree with the others who have commented here—what you're trying to do here flies in the face of fundamental content policy. MastCell ^Talk 04:41, 22 September 2014 (UTC)

I distinctly remember, made us... uncooperative. Made me not trust your interpretation of Wiki-policy, even when you were right and I was/am wrong (on primary sourcing). It's clear you don't like the quote. WP:UNDUE states "articles specifically relating to a minority viewpoint" requires more space for minority views (Daling) on minority viewpoints (ABC). If reliable minority views are allowed to be treated with contempt, then we can see firsthand what pro-lifers call a "conspiracy". Of course there isn't one, its just good people making poor judgements reinforced by officialdom (and like minded agreement).

You can help things, in numerous ways, by putting it back yourself. Nothing is stopping you from finding a source (speaking to the issue) to balance minority Daling with. But the majority of the section and article already details pro-life bias. If you can't see even the potential for pro-choice bias then that's disappointingly stagnant. Your interpretation of policy on this continues to be wrong / narrow. User:Yobol making the same mistake doesn't change that. - RoyBoy 14:55, 22 September 2014 (UTC)

The quote you are trying to use is being sourced to two two op-eds; not news articles, not scholarly journal articles, but from opinion pieces, which are generally considered low quality sources. To establish that this particular quote deserves any weight at all, you will need to find better sources. In this case, you will need to find a high quality source that not only has this quote in it, but explains why this quote is important in context of the politicization of the subject, so that it actually has content value to the reader. Trying to use an 11 and 17 year old op ed is just not going to cut it. If these are the best sources for this quote, it clearly has no role in this article. Yobol (talk) 01:15, 2 October 2014 (UTC)

As its Daling's opinion, an op-ed is where it would be found. It's importance is self evident, it outlines POV pushing by both sides of the political divide on the ABC issue. While this can be inferred by how positive or negative studies are criticized (or not) having a researcher state it would clarify it to the reader. Wouldn't you agree? Your sourcing requirement is arbitrary, counter-intuitive and likely a double standard if I looked at other sources in the article, and your "11 and 17 year old" is irrelevant, unless Daling has changed her position. - RoyBoy 01:25, 3 November 2014 (UTC)

Um, no, it's importance is established by being published in high quality sources. As it is not published in high quality sources, it's importance is not established and therefore has no place here. Yobol (talk) 02:44, 3 November 2014 (UTC)

Not even close. Its importance is dictated by who said it and what was said. The quality is more than sufficient to meet Wikipedia:Identifying_reliable_sources#News_organizations, the higher quality source requirement comes from you, not policy, unless you can be more specific? - RoyBoy 04:08, 3 November 2014 (UTC)

This is a WP:WEIGHT issue, not a RS issue. It's written in poor quality sources (op-eds) and therefore does not belong here. Feel free to pursue the WP:DR process if you disagree. Yobol (talk) 04:17, 3 November 2014 (UTC)

It's hard to dispute when you don't have a position! Don't say this isn't a RS issue, then say "poor quality source". I specified the sub-section why it has weight. You respond by repeating WP:Weight. That isn't an argument, it isn't even a complete thought. Mastcell notes it isn't prominent, valid point, but that's the nature of a minority unpopular position; which Mastcell neglected in zer's "fundamental content policy" analysis. Likewise Daling isn't interested putting herself out there so that a quote can meet an arbitration threshold you conjure up. This doesn't mean it needs the same prominence as before (given my rethink below), but to eliminate it by saying WP:Weight repeatedly for op-ed's is nonsense; especially when policy points to op-eds as valuable sources for positions. Would it be preferred to have a better source, yes, is it necessary? No. Linking to policy doesn't demonstrate otherwise. - RoyBoy 21:26, 8 November 2014 (UTC)

After reviewing Jasen further during the Studies (History) rewrite, being reminded that Daling was present at NCI 2003 workshop (while this does not nail down Daling's exact position), does make me more circumspect on the Daling quote. Having said that, I do feel there is a middle ground that is best for the article. Don't know what that could be. - RoyBoy 19:35, 28 September 2014 (UTC)

ABC studies remake

Hello all, just finished merging the original Recall Bias into a remade ABC history section which is now chronological, should be easier to follow, and relies more on Jasen rather than primary sources. Everything together (including my notes) currently is 3612 words, compared to the original with 5130 words. So making progress, I'm still working on it to improve flow and shorten everything while maintaining / adding key memes, but more closely following / contextualizing with the scientific consensus narrative. Comments welcome, but its still a work in progress (especially response bias sections, which are draft 0). - RoyBoy 22:03, 8 November 2014 (UTC)

FAQ

Given that this is a controversial topic, having a /FAQ file is really not a bad idea. The current draft would need pretty significant work before we could post it, though. What questions might you want to see in such a document?

1. Why does the article state unequivocally that the hypothesis is not supported? Science can never definitively prove a negative. What about the neutrality policy?
   • The policy that articles be presented from a neutral point of view requires that the article text adhere most closely to the most reliable sources. In this case, major medical organizations state that the evidence does not support a connection, so the article must do so as well. It is true that science cannot prove a negative, but the article must follow the sources in portraying the possibility that the current evidence will be overthrown as unlikely.
2. What about paper 'X'? Why was my referenced text deleted?
   • Individual sources are accorded weight according to how they are treated by other reliable sources. In particular, an individual study should not be used to rebut a large review.
3. What about this expert? Why are their views not described in detail or given any weight?
   • Individual experts can have a large impact on the political and cultural controversies, but scientifically we must defer to the major medical organizations that have commented on the hypothesis.
4. The major medical organizations are in the thrall of the abortion industry!
   • Wikipedia relies on independent reliable sources and is not an appropriate venue for promoting various conspiracy theories.
5. Why not just describe all the relevant papers and let the reader decide?
   • Wikipedia is an encyclopedia, not a historical literature review. The doctors and scientists who study such things have already done the work of synthesizing the primary literature. As an encyclopedia, we summarize this analysis.
6. Why does the article use the term "anti-abortion" rather than "pro-life"?
   • The Neutral point of view policy requires that we avoid biased and loaded terms. This stylistic choice follows that of major newspapers.
7. Why are partisan sources being cited?
   • Partisan sources are reliable for their own opinions and may be important for explaining the cultural and political controversy. Use should be minimized or avoided in other contexts.
8. Why is this idea described as a "hypothesis" instead of a "theory" or "myth" or as the "ABC link"?
   • "Theory" or "link" would imply a degree of acceptance by the medical community that is not evident. "Myth" would imply that there was never any reason to pose the question.

Anything else? - 2/0 (cont.) 11:15, 10 September 2014 (UTC)

Regarding proposed FAQ #!, the article doesn't state unequivocally that the hypothesis is not true; only that it is contrary to mainstream scientific opinion and denied by the major medical professional organizations. Goblinshark17 (talk) 19:56, 10 September 2014 (UTC)

Supplement to question 8, why is it not described as "The ABC Myth"? Goblinshark17 (talk) 21:35, 10 September 2014 (UTC)

Good catch on true - changed to supported as still being simple and something that a newcomer to this article might reasonably ask. Myth is the flip side to theory, so a single answer covering both seems feasible. Took a stab at providing some answers. - 2/0 (cont.) 22:09, 12 September 2014 (UTC)

Pretty decent, definitely better than my first attempt! But we'll deal with this after the above. - RoyBoy 23:24, 17 September 2014 (UTC)

Tweaked 5 and 8. - RoyBoy 21:45, 8 November 2014 (UTC)

Agree with the tweak to five; made it even more straightforward and forthright. Eight is fine. Taking this live, unless there are any further matters to hash out? - 2/0 (cont.) 15:47, 11 November 2014 (UTC)

Chinese article

Just removed a sentence describing a new Chinese meta-analysis, which was sourced to a primary source (the meta-analysis itself). See "Weight" section earlier on this TALK page for reasons for removing such a reference, especially the comments by editor "2/0". If we're gonna describe an individual primary paper like that, then for consistency we would also have to describe other primary papers and the article would degenerate into from an encyclopedia article into a scientific literature review. HandsomeMrToad (talk) 10:44, 24 December 2015 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Abortion–breast cancer hypothesis. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Corrected formatting/usage for http://oversight.house.gov/features/politics_and_science/example_breast_cancer.htm

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

Y An editor has reviewed this edit and fixed any errors that were found.

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—^{cyberbot II}_{Talk to my owner:Online} 05:42, 25 April 2016 (UTC)

Breast Cancer Prevention Institute: not a reliable source

I have removed a short paragraph, recently added by an anonymous editor, which incorrectly stated that "the subject remains an issue of controversy in the field of medical research" and which referred to a chart from the Breast Cancer Prevention Institute. The BCPI is not a reliable source; it is an anti-abortion propaganda organization founded by Dr. Joel Brind (who is already described in the article) to give an appearance of respectability and scientific validity to his discredited abortion-breast-cancer hypothesis. Witness the fact that almost all the papers listed on its "publications" page support the hypothesis; and, except for abortion and hormonal contraception, no other risk for factors for breast cancer are mentioned in the titles of any of the listed publications. Also, several of the listed publications are published in the Journal of American Physicians and Surgeons which is a well-known faux-scientific "journal" dedicated to promoting a right-wing social agenda. (Here is a link to the BCPI's publications page: http://www.bcpinstitute.org/publishedpapers.htm And here is a link to Wikipedia's own entry about the Journal of American Physicians and Surgeons: https://en.wikipedia.org/wiki/Association_of_American_Physicians_and_Surgeons#Journal_of_American_Physicians_and_Surgeons ). HandsomeMrToad (talk) 09:16, 1 June 2016 (UTC)

Agree, good catch. Alexbrn (talk) 09:22, 1 June 2016 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified 5 external links on Abortion–breast cancer hypothesis. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20110113001029/http://www.who.int/mediacentre/factsheets/fs240/en/index.html to http://www.who.int/mediacentre/factsheets/fs240/en/index.html
• Added archive https://web.archive.org/web/20130727105037/http://www.rcog.org.uk/files/rcog-corp/uploaded-files/NEBAbortionSummary.pdf to http://www.rcog.org.uk/files/rcog-corp/uploaded-files/NEBAbortionSummary.pdf
• Added archive https://web.archive.org/web/20080404034430/http://www.washingtonmonthly.com/features/2004/0410.mooney.html to http://www.washingtonmonthly.com/features/2004/0410.mooney.html
• Added {{dead link}} tag to http://www.rcog.org.uk/files/rcog-corp/Abortion%20guideline_web_1.pdf
• Added archive https://web.archive.org/web/20071214064635/http://www.who.int/mediacentre/factsheets/fs240/en/index.html to http://www.who.int/mediacentre/factsheets/fs240/en/index.html
• Added archive https://web.archive.org/web/20130727105037/http://www.rcog.org.uk/files/rcog-corp/uploaded-files/NEBAbortionSummary.pdf to http://www.rcog.org.uk/files/rcog-corp/uploaded-files/NEBAbortionSummary.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 21:10, 2 October 2016 (UTC)