22

Conservapedia's sub-article on Abortion and Breast Cancer claims that there is an undeniable link between breast cancer and abortions:

The vast majority of scientific studies have shown that abortion causes an increase in breast cancer, including 16 out of 17 statistically significant studies.

A new study coming from researchers in Sri Lanka finds women who had abortions ... [had a] 3.42 [increased risk] ... compared with those who kept their baby. ... The Sri Lankan study is the fourth epidemiological study in fourteen months to report an abortion-breast cancer link, including studies from the U.S., China and Turkey.

LifeNews also reported in 2010 that a National Cancer Institute researcher admitted there was a link between abortion and breast cancer. Studies showing that abortion increases breast cancer predate the political controversy. A study at the Fred Hutchinson Cancer Research Center published in 2009 confirmed a 40% increase in risk in breast cancer from abortion. Yet the abortion industry conceals this increased risk, just as the tobacco industry concealed its cancer risk for decades.

Yet the National Cancer Institute's article on Abortion, Miscarriage, and Breast Cancer Risk says the opposite:

They concluded that having an abortion or miscarriage does not increase a woman’s subsequent risk of developing breast cancer.

So which one is correct?

Does Abortion lead to an increased risk of Breast Cancer?

Christian
  • 33,271
  • 15
  • 112
  • 266
Mark Rogers
  • 11,324
  • 11
  • 55
  • 71
  • 21
    Asking any question that originates from "Conservipedia" is probably not a good idea. These are the same people that wanted to deny the "Theory of Relativity" because it just sounded too liberal... They are not beyond outright lying and manufacturing info. – Larian LeQuella Apr 10 '11 at 03:53
  • 16
    @larian all the more reason to debunk them. Somewhere someone is googling 'abortion breast cancer' and a good answer to this question might help him/her in a big way. – justin cress Apr 10 '11 at 14:00
  • by the way, "just as the tobacco industry concealed its cancer risk for decades." on any page ending in 'pedia' makes me lol. – justin cress Apr 10 '11 at 14:01
  • 1
    There is an entire, well-written and footnoted Wikipedia article about this question. The article summary (lead section) says flatly that ["The abortion-breast cancer hypothesis has been the subject of extensive scientific inquiry, and the scientific community has concluded that abortion does not cause breast cancer."](http://en.wikipedia.org/wiki/Abortion_%E2%80%93_breast_cancer_hypothesis) – John Broughton Apr 25 '11 at 14:59

3 Answers3

15

There is one study from 1981 that suggests that there is a positive correlation between early abortion and breast cancer in patients younger than 32:

A first-trimester abortion before FFTP, whether spontaneous or induced, was associated with a 2.4-fold increase in breast-cancer risk (P < 0.005).

Two responses can be found to this study. First, Vessey et al fielded a direct response in 1982:

A recent publication from California in this journal has suggested that both prolonged oral contraceptive use and abortion before first term pregnancy increases the risk of breast cancer in young women. Data are presented on 1176 women aged 16-50 years with breast cancer, interviewed in London or in Oxford, together with a like number of matches control subjects. The results are entirely reassuring, being, in fact, more compatible with protective effects than the reverse. Possible reasons for the differences between the 2 sets of data are discussed.

In addition, Melbye et al in 1997 agreed with this second study. Using a much larger pool than the 1981 study, and with much sounder statistical methodology (not to mention a larger N), they found no relative increase of risk as a result of abortion.

So how can the results of the 1981 study be explained? 1981 was prior to any genetic screenings, so their abstract text of "This increased risk could not be explained by other risk factors" might have to be changed, as several more genetic risk factors have been found since then. Notably, BRCA1 was first described in 1991 and cloned in 1994; Ford et al found in 1998 that

The estimated cumulative risk of breast cancer reached 28% (95% CI 9%–44%) by age 50 years and 84% (95% CI 43%–95%) by age 70 years

Or roughly 1 in 4 by age 50. Women in the general population of the United States, according to SEER, have a 1 in 233 chance of being diagnosed with breast cancer from ages 30-39, and 1 in 69 from ages 40-49.

The fact that the 1981 study limited their cohort to patients who were exceptionally young (below 32 years of age) smacks of selection bias to me, meaning that those patients were carrying some genetic marker like BRCA1 in order to even be included in the study in the first place. Beyond that, it's hard to say whether or not the observed increases were due to noise in the data or not, since they didn't perform genetic testing on the patients. However, if those patients were all carrying those risk factors, those factors would certainly explain how they got cancer so young, in light of other studies finding that abortion is not a correlated factor.

mmr
  • 1,946
  • 1
  • 19
  • 18
  • 6
    Even without selection bias, the correlation is probably real. But the conclusion that *"abortions increase the risk of breast cancer"* does not follow from that. `So how can the results of the 1981 study be explained?` Simple: women who have abortions are more likely in the lower class, and the lower class has a higher risk-factor for cancer. I bet you'd find correlations between abortion and smoking/alcoholism/drug use/other bad habits as well. That doesn't mean that one caused the other; just that they're correlated. – BlueRaja - Danny Pflughoeft Oct 16 '11 at 00:36
  • 1
    @BlueRaja-DannyPflughoeft-- so, there are two assertions there, that 1) lower class women get breast cancer earlier and 2) lower class women have more abortions. Both of which are, I think, beyond the scope of this question, as they could be two questions by themselves. I believe that the brca1 & 2 causal link is probably a better explanation (which is why I used it), because having breast cancer before the age of 32 is extremely rare unless those or similar markers are present. – mmr Oct 16 '11 at 18:22
  • 1
    I meant to show that just because there is a correlation does not mean there's causation; there are other perfectly good explanations for the correlation. I did not mean to say that my explanation is definitely the correct one. It's still entirely possible that abortions cause breast cancer. However, **none of these studies support that claim** - they just don't contradict it. – BlueRaja - Danny Pflughoeft Oct 16 '11 at 21:18
  • @BlueRaja-DannyPflughoeft-- absolutely true. – mmr Oct 17 '11 at 00:29
4

Some abortions definitely increase the risk of breast cancer; whether all abortions do so is disputed.

There's no real dispute about the link between the mother's age when she has her first full-term pregnancy (FFTP) and her likelihood of being diagnosed with breast cancer later in life. The older a woman was when she had her FFTP, the greater her risk of breast cancer.

That means that abortions prior to FFTP almost certainly increase a woman's risk of breast cancer, because they increase the age of FFTP.

If you read the ACS statement on abortion and breast cancer, and just skip to the conclusion at the end, you'll find that it says, "at this time" the scientific evidence does not support a link between abortion and breast cancer. Obviously that's more nuanced than the paraphrase of it which abortion proponents use, that flatly denies any link. But, even so, it is a biased claim, which severely overstates their certainty.

But that's not the biggest problem with the ACS statement. Earlier on the same ACS page you'll repeatedly find this phrase: "After adjusting for known breast cancer risk factors..."

What the ACS inexcusably fails to mention is that among those "known risk factors" is the increased chance of breast cancer associated with delayed FFTP -- which is very often caused by one or more abortions!

Widespread acceptance of elective abortion has caused an increase in the average age at which American women begin bearing children, which is undoubtedly responsible for part of the concurrent increase in the incidence of breast cancer among these women.

The only dispute is whether an abortion done after a woman's FFTP also increases the risk of breast cancer. Many studies have found that it does, but those findings are hotly disputed. You can learn much more about that question here: http://www.abortionbreastcancer.com/index/

Dave Burton
  • 271
  • 2
  • 6
  • 1
    This post is insufficiently referenced. We require notable references for **all** claims. This post has two references, of which *one* supports your point, and the other isn’t notable, nor specific to a particular claim. You also misinterpret scientifically precise language. When scientists say “at this time there is no evidence of X” they mean the same as when lay people say “X is false”. You make it seem as though they are being extra careful here not to overstate conflicting evidence but this isn’t the case. – Konrad Rudolph Jul 04 '13 at 14:38
  • Konrad, “at this time there is no evidence of X” does NOT mean the same as “X is false,” regardless of who says it. The first phrase means "X is unknown." The second phrase means "X is known to be false." Those are contradictory statements. What did I write that you think needs additional references? – Dave Burton Sep 13 '19 at 17:50
  • “there is no evidence for X” absolutely does *not* mean the same as “X is unknown”, that is a fundamental misunderstanding. It only means the same in the absence of good attempts to find evidence. Consider this: Doctors want to know if you have leukaemia. They run a highly sensitive test, which comes back negative. In other words, there’s no evidence that you have breast cancer. But this statement is in the context of running a highly sensitive test. It’s therefore not “unknown” whether you have leukaemia. It’s highly unlikely. – Konrad Rudolph Sep 14 '19 at 13:04
  • Konrad, if the doctor was confident that I don't have leukemia, he'd say, "you don't have leukemia." He would NOT say, "at this time there is no evidence that you have leukemia." THAT statement would suggest that he's still worried about leukemia. He's cautiously hopeful, but not confident, and we probably ought to keep checking, periodically, or else run some more tests, or both. The two statements, "you don't have leukemia," and "at this time there is no evidence that you have leukemia," are not synonymous. – Dave Burton Sep 15 '19 at 17:13
  • 1
    He might be saying that to *you* as a lay person, but that’s now how he would write it in a case study for a scientific journal. You simply need to accept that you are misunderstanding, and consequently misinterpreting, scientific language. This simply isn’t debatable. – Konrad Rudolph Sep 16 '19 at 11:11
  • You are confused, Konrad. Scientific writing supplements common English with additional, domain-specific terminology. But it does not normally invert the meaning of plain English prose. (To be fair, I can think of of one unfortunate exception: in archaeology the abbreviation "BP" is sometimes used when specifying dates/ages, and it actually means "before 1950," even though it stands for "before present.") But it is NEVER the case that “at this time there is no evidence of X” is synonymous with “X is false.” The first statement means X is unknown, the seconds means X is known to be false. – Dave Burton Sep 17 '19 at 12:41
  • 1
    What can I say? You are wrong, plain and simple. You are clearly not a scientist or statistician, and that’s fine, but don’t pretend to know about things which you obviously know nothing about. This isn’t controversial, it’s extremely basic statistical terminology. Take it from an actual scientist: you are woefully wrong, and simply googling or reading an introductory statistical testing textbook would show you that. – Konrad Rudolph Sep 17 '19 at 13:00
  • Konrad, it has been said that when an honest man is shown his error he either ceases to be wrong or ceases to be honest. You've been shown your error, so which shall it be, for you? This is English, not statistics. Even in statistics, the words do not change meaning as you imagine. When someone says, "abortion does not cause breast cancer," they are saying an untruth. Abortion before first full term pregnancy is known to increase the risk of breast cancer later in life, and studies are divided about whether abortion after first full term pregnancy also increases the risk of breast cancer. – Dave Burton Sep 18 '19 at 13:31
  • 1
    This is honestly starting to be embarrassing. Please stop. You’re literally arguing with an expert about his field of expertise, which you know nothing about. Can’t you see how delusional that is? – Konrad Rudolph Sep 18 '19 at 13:38
  • Konrad wrote, "You’re literally arguing with an expert about his field of expertise..." Hahaha! Rookie advice: before changing your claimed "field of expertise" you really ought to adjust your SE profile. ‍‍‍‍‍‍ ‍‍ ‍‍‍‍‍‍ ‍‍ Konrad continued, "Can’t you see how delusional that is?" Yes, I sure do! – Dave Burton Sep 19 '19 at 14:35
  • 1
    I have to assume that you’re actually a troll. My profile literally states that I’m a scientist. It even has a link to my PhD thesis, which (indirectly) shows my former affiliation with a statistical genetics research group, if you did some digging. The stuff that we’re arguing about is literally my bread and butter. – Konrad Rudolph Sep 19 '19 at 14:37
  • Konrad, your profile says you are a scientist and software developer. (So am I, BTW.) Our argument is not about statistics, or genetics, or software development. It is about English. You claimed, "When scientists say 'at this time there is no evidence of X' they mean the same as when lay people say 'X is false'." That's nonsense, of course. – Dave Burton Sep 19 '19 at 14:50
  • Repeating a false claim does not make it true, though it *is* infuriating. Our argument is about English language *in the context of statistics, used in science*. You refuse to believe that ambiguous English expressions can have a precise, slightly different and, to a lay person, surprising meaning in statistics. Yet that is precisely the case. Either way, my first comment was an off-hand remark that should be made more precise, since you are so pugnacious. I had hoped that you had understood this. Clearly not. So here goes: – Konrad Rudolph Sep 19 '19 at 15:17
  • The statement “there is no evidence for X” depends on the amount of potentially available evidence. If no experiment was done, then “there is no evidence for X” means “we don’t know”. Yet it is virtually never used in this context. Instead, the context is virtually always in the presence of more or less compelling experiments, and the meaning thus changes because the amount of uncertainty has decreased. Strictly speaking the meaning is the same, it’s just that lay speakers are not aware of the implications due to lack of precision or context. – Konrad Rudolph Sep 19 '19 at 15:22
2

Here is the link to the original paper by Dolle on PubMed, one of the sources from the article: Risk factors for triple-negative breast cancer in women under the age of 45 years.

Controls were age matched and ascertained via random digit dialing. Oral contraceptive use > or =1 year was associated with a 2.5-fold increased risk for triple-negative breast cancer (95% confidence interval, 1.4-4.3) and no significantly increased risk for non-triple-negative breast cancer (P(heterogeneity) = 0.008). Furthermore, the risk among oral contraceptive users conferred by longer oral contraceptive duration and by more recent use was significantly greater for triple-negative breast cancer than non-triple-negative breast cancer (P(heterogeneity) = 0.02 and 0.01, respectively).

The paper doesn't, at least from the abstract, make reference to abortion, but contraceptives.

However, this paper followed by this research paper: Breast cancer among young U.S. women in relation to oral contraceptive use.

RESULTS: There was no increased incidence of breast cancer associated with ever having used oral contraceptives.

CONCLUSIONS: Long-term oral contraceptive use among young women or use beginning near menarche may be associated with a small excess breast cancer risk, possibly due to susceptibility to genetic damage in breast epithelial cells at ages of high breast cell proliferative activity.

Follow the link for more detail.

So, the initial study found some correlation in data, and then follow up studies with a larger scale testing field, disproving this correlation.

Community
  • 1
Mike
  • 1,147
  • 6
  • 15
  • 6
    First, I think you need to go back and reread the question. The subject here is abortion, not oral contraception. Secondly, I'm not sure I'd use the phrase "debunked," unless the second study was accusing the researchers of fraud, which I don't think is the case. – Scott Hamilton Apr 10 '11 at 19:55
  • 1
    I might have to adjust my post. I was looking at the Dolle paper they listed in their sources, which makes no reference to abortions, at least not from the abstract. As for debunked, I simply mean disproved. I wasn't aware it carried that connotation. – Mike Apr 10 '11 at 20:22
  • 1
    This is a difficult question to answer, because different people have different 'views' as to what constitutes an 'abortion'. The ultra-christian-conservative will consider birth control pills 'abortion' due to their religion. So without knowing precisely they're defining 'abortion' as can greatly influence what 'facts' they're using to support their claim. – Darwy Apr 11 '11 at 08:43
  • 5
    @Darwy Really? I've never heard even the most strident anti-abortion foes claim contraception is the exact same thing as abortion. If they did, they'd have to picket every CVS in the country. In any case, the Conservapedia article is clear they're talking about the medical procedure of abortion ("termination of a pregnancy"), not contraception, and any scientific study should be even clearer about how they're defining their terms. – Scott Hamilton Apr 11 '11 at 17:12
  • Like I was saying, I was addressing this part from the Conservvapedia page: "A study at the Fred Hutchinson Cancer Research Center published in 2009 confirmed a 40% increase in risk in breast cancer from abortion." The source for this statement turned out to be a study that found a link between contraception use and cancer. – Mike Apr 11 '11 at 20:14
  • @Scott I'll direct you to a conversation I had here regarding what is (and isn't) an abortion): http://www.theday.com/article/20110318/OP01/303189987 – Darwy Apr 11 '11 at 20:23
  • 5
    @Mike I'm pretty sure that's a simple mistake, as opposed to an actual statement saying oral contraception and abortion are the same thing. You'll find that in any non-reality based community a surfeit of sources is often used to appear authoritative, and a lot of those sources will be off-topic or even contradictory. As a skeptics resource, we need to stay on topic. Pointing out that a particular source doesn't actually apply the argument is fine, but it doesn't answer the question. If you can show every source of the article is off topic, that would be good too... – Scott Hamilton Apr 11 '11 at 20:51
  • 2
    @Darwy You're conflating a political/moral argument with a medical/scientific one. For the purpose of the question being asked we're only concerned with the medical procedure of abortion and its relationship to breast cancer. Even the Conservapedia article is clear that it's talking about abortion, not contraception, even if some of the sources it links are not on that subject. – Scott Hamilton Apr 11 '11 at 21:01
  • 1
    -1 for a good answer to the wrong question. Would be a fine answer to "does oral contraception give women breast cancer?" – matt_black Oct 15 '11 at 23:08
  • @ScottHamilton Contraceptives sometimes work as abortifacients, so Darwy is not completely wrong in his characterization. I have also heard of strong anti-abortionists claim contraception is very close to abortion, if not exactly the same. The position makes sense if life begins at conception. There is not a big difference between stopping sperm and egg right before joining vs. killing a fertilized egg right after the join. – yters Dec 16 '16 at 17:14