Abortion and maternal mortality: correlation is not causation
I recently discovered an awesome website called Spurious Correlations. It uses hilarious graphs to illustrate the long-understood but frequently ignored truth that correlation is not causation.
|“Number of people who died by becoming tangled in their bedsheets”
and “Total revenue generated by skiing facilities (US).”
Clearly the ski resort industry has blood on its hands.
Sometimes these correlations are purely accidental, as in the above case. Other times, there might be a confounding factor at play. For instance, I would imagine that “Total revenue generated by arcades” and “Computer science doctorates awarded” are both tied to the tech boom in general, which improves video game quality and increases the demand for skilled computer scientists. (But I’d need more data to prove it.)
I really hope this site takes off, because the public’s failure to understand that correlation is not causation can have catastrophic effects.
One of the pillars of the abortion movement is the idea that legalized abortion prevents maternal deaths. They can point to improved maternal mortality rates after the legalization of abortion in the United States and elsewhere. Post hoc ergo propter hoc.
But there are two major problems with that theory. The first is that the numbers themselves are often either wrong or manipulated. Dr. Bernard Nathanson, the co-founder of NARAL, admitted to using inflated figures and noted that “repeating the big lie often enough convinces the public,” which is unfortunately true.
The second problem is the spurious correlation problem. The movement for the legalization of abortion happened to coincide with medical breakthroughs that dramatically improved Americans’ health in general, and in particular, allowed doctors to manage infection, hemorrhage, and other issues that may appear as complications of abortion. It also coincided, in the developing world, with independent foreign aid focused on maternal health issues, such as the provision of midwives.
Untangling all these influences requires serious effort, and when the spurious correlation happens to support your current ideology, there is no incentive to undertake that research. But in 2012—sadly, several decades too late to combat the big lie effectively—researchers published the results of an in-depth study of maternal mortality in Chile, which restored the right to life in 1989. It found that the anti-abortion law did not lead to an increase in women’s deaths; in fact, just the opposite happened. The authors concluded that if a country wants to reduce maternal mortality, it should invest in education, not abortion.
The misapprehension of maternal mortality has grave consequences. Not only are unborn lives lost, but the diversion of money to the ideological goals of the abortion industry reduces the funds available for the desperate, real needs of impoverished people.
Thanks for the article link! The response by the Guttmacher Institute and then the response by the authors to them is worth reading too. I do have friends for whom the 'Back Alley Abortion' argument is enough to convince them to keep it legal, even though they believe it's a human person that's being killed.
According to this article, the average mortality rate for abortion in the US from 1988-97 was about 0.7 deaths per 100,000 abortions, and the risk is not uniform: it's even lower for very early abortions (0.1 for 8 or fewer weeks, 0.4 at the end of the first trimester) and much higher for later abortions. According to the CIA's World Fact Book, no country in the world has a maternal mortality rate of less than 2 deaths per 100,000 births. (In the US, it is disgracefully higher–about 20 deaths per 100,000 live births.) If you took all 800,000 women who have abortions each year in the US (of which about 6 die), and forced them all to carry to term, about 160 of them would die. So unless you dispute those recent numbers, an effective abortion ban would kill about 154 people per year.
Pregnancy and childbirth is dangerous. Legal abortion is less dangerous. Forcing all pregnant women to carry to term whether they want to or not subjects them to risks to which they did not consent, and that is wrong.
So your saying 800,006 deaths is better than 160? Maybe I'm being a little unfair. Let's round up to 320 by saying in every case of in maternal mortality the child dies, too. 800,006 vs. 320.
The solution, obviously, is to kill the 800,006 to save the 320, rather than, say, funnel some of that billion dollar abortion industry money toward better maternity care…
You kind of missed the point of the article, which is that, instead of money being put towards things like medical research, drug trials, scientific studies, and doctor training to REDUCE the maternal mortality rate in the long run, it is being put towards abortion, which is a terrible solution towards reducing maternal mortality in the long run. It is like saying "Well, instead of treating your heart/liver/kidney disease, we are just going to give you a new heart/liver/kidney, because it's easier and we don't have to pay to treat you and/or pay for medical research to treat these diseases." It's bad medicine and it's bad science.
You are assuming that there isn't a statistically significant difference between the potential mortality rates of those who abort, and the general mortality rate. 2/100,000 isn't a number set in stone, it's a pattern we observe (or can observe in other countries). Keep in mind, the vast majority of abortions occur on healthy fetuses, with healthy mothers, whereas the vast majority of women who die in childbirth are either unhealthy themselves, or there is a problem with the fetus.
Now what if you really just confined abortions to severe fetal abnormalities, or life (or avoiding severe injury) to the mother? Would the group of people who would have aborted, but no longer could, have a 2/100,000 mortality rate? Probably not.
According to Table 2 in this article, which compares planned C-sections to vaginal deliveries specifically in healthy women, they still came up with a maternal mortality rate of about 2/100,000, and that was in Canada, which has a noticeably lower overall maternal mortality rate than the US. So the best evidence I can find is that, yes, even if a woman has a low-risk pregnancy, an early abortion is still safer for her than carrying to term.
(That's even disregarding women who have, say, gestational diabetes that develops late, and thus find themselves moved into the "high-risk" category of pregnancy long after an opportunity for an early abortion is past.)
The federal government in the United States does not pay for abortion. Private charities sometimes help, but even they usually can't pay the entire cost of an abortion. Mostly, abortions are paid for out-of-pocket by the pregnant person. Are you seriously suggesting that, if we forbade abortion, these people would donate the few hundred dollars they saved to medical research, rather than putting it towards the (much higher) costs of prenatal care and childbirth costs?
IIRC an abortion in the US usually costs less than a thousand dollars. A hospital birth, with insurance, costs more like $3,500–and the $20,000 the insurance company has to spend is also money that can't go towards medical research.
You want to spend more money on medical research? I don't think anyone here can argue with that. But don't pretend that we could finance that research by having fewer abortions. Childbirth is more expensive than abortion, and every person who chooses not to abort a pregnancy she can't afford is a person who will need that much more help with her baby. You might think that it's worth the price–and I'm not sure I can argue–but don't pretend that the economic argument is in favor of birth over abortion. It's not.
154 is less than 800,000.
Trying to find their definition of "healthy" and failing. Looking at some of these causes of death really put that definition into question. "cardiac arrest"? Are you sure the woman didn't have heart problems beforehand, known or unknown? Healthy, as it seems to be in this paper, is simply no prior problems reported. Now imagine if we put our efforts into more diagnostic and preventative care. That is how you lower the mortality rate there, not by first killing another being.
Considering this is a paper on the difference between cesarean and vaginal deliveries, it's not surprising that they aren't nearly as concerned about the details of the catalyst that brought about the cause of death. They are talking about the status quo, and comparing it with two forms of delivery. They are not looking at the cause of death, they are looking at correlations between deaths and method of delivery. Timing of the deaths isn't even uniform, with only 41 of the subjects dying in hospital/. For the others, as far as we can ascertain, cardiac arrest could have come from nearly getting run over by a train (or it could have come from the birth. The point is, the paper is not clear on that) Did you know the death rate for people in their 30s is about 1/1000? Wow, that is much worse odds than giving birth! I guess it's better to be pregnant than 30+ years old.
Remember, correlation does not imply causation. Can you say with any certainty that pregnancy or labor caused these deaths? This paper can't (and that's fine, that is not the purpose of the paper).
So yeah, essentially, like Tom said, I think you are missing the point of the SPL article.
"Not spending money on abortion" will not free up lots of money to spend on improved prenatal care. It will in fact mean that a lot MORE money has to go into maternal and obstetrical care just to get the quality of care we get now. See my reply to Tom below.
Also, in the study I cited above (which was the best study I could find with Google), yes, "Healthy, as it seems to be in this paper, is simply no prior problems reported." I don't see how you can claim that the risk to an apparently-healthy person being denied the abortion she wants is less than the number in that study.
The debate over the risks of back-alley abortions misses the point here, and the point from a politically anti-abortion perspective is that a relatively small amount of maternal deaths due to their own decisions, while atrocious, is less bad than a much larger amount of deaths due to someone else's decisions.
For reference, the back-alley risk argument might work in regards to some other things, though. For instance, I think that there were cases of biological males who attempted to either perform a self-castration or to get a back-alley castration. Either of these things appears to be more dangerous than getting a castration done in a safe, medical setting by a qualified individual. Thus, perhaps it is a good idea to give males who genuinely want to get castrated this option.
"Forcing all pregnant women to carry to term whether they want to or not subjects them to risks to which they did not consent, and that is wrong."
Well, abortion is *much* more dangerous for the prenatal offspring than childbirth is for the woman.
Also, out of curiosity, are you against the draft in *all* cases?
I might agree with the pro-choicers on this one. After all, if a medical procedure is banned, then one might be less likely to get it safely performed than if this medical procedure is legal. After all, I doubt that *every* doctor who performs such a medical procedure when it is legal will also perform it if it is banned.
However, this point here appears to miss the overall point in this debate, and the overall point is this: What exactly is worse–pregnant people dying due to their own decisions or prenatal humans dying due to someone else's decisions?
I have now read this entire article, and it appears that my previously couple of comments here appeared to be a bit off-topic (but still relevant to the overall abortion debate). Sorry about that.
I've got a question–out of curiosity–what exactly is more dangerous for a woman–an *elective* late-term abortion or giving birth?
Do these pro-choice friends of yours also support making it easy, legal, and safe for people to do other things (such as for males to get castrated) so that these people don't have to get these things done unsafely and/or in back-alleys?
Giving birth is safer than any kind of abortion. And a late-term abortion, whether elective or not, is an extremely dangerous procedure.
Thank you writing about this! Maternal mortality has gone UP in the United States. We are one of the few developed nations where this is the case. And we are also one of the few developed nations where abortion is allowed for any reason up until birth…
Can you please give me some sources for this?
Honestly? I don't know that they've seen past the actual and hypothetical distress that they see women with an unplanned pregnancy experiencing (the friend who has been most specific is a social worker). I can see how it's easier to respond to the emotion that you can see right in front of you, rather than the (usually) intellectual exercise of considering the unborn human and what it deserves.
If the pro-life movement were to succeed in forcing pregnant people to carry to term, then yes, I would support something like the draft; for example, I would support making all healthy young people without uteri register as organ donors, to be called upon to give up kidneys, liver lobes, or anything else that can be a safe living donation. (If some sort of exception for people who didn't have sex, ever, could practically be put into place, I would support that.)
If we keep the legal principle that my body is mine and no fetus may use it without my consent, then we do need to keep the legal principle that your body is yours and the government may not use it in warfare.
Well, abortion is *much* more dangerous for the prenatal offspring than childbirth is for the woman.
And that means that this article doesn't really have a lot of point when addressed to pro-lifers. When you say that 154<800,000, as if it's a complete refutation of everything I've said, what you are saying is that you don't care how many women would die as a result of banning abortion as long as it's significantly less than the number of fetuses saved. So whether the original flawed studies are right or wrong, it doesn't matter to you.
The author appears to be saying that it should matter to me, and to other people who don't care about z/e/fs dying. I'm saying that even if these studies had spurious correlation, looking at the correct numbers still leads to the conclusion that forcing pregnant people to give birth leads to more deaths of real people.
(My other point was that "let's spend money on prenatal care rather than abortion", the article's other suggestion, is absurd because every $800 dollars you don't spend on an abortion is $30,000 dollars someone has to spend on prenatal and obstetric care.)
1st it depends on whether you think the prenatal is a entity that has full moral value.
Lets say it has. How do we deal with the underlying legal/moral principles in non pregnancy cases?
Society allows lethal force against innocent offenders in some states. It also allows lethal force to be used to protect bodily integrity. Though I would not some states don't allow this if the victim caused the stat of affairs.
But nor do we allow bodily compensation.
From legal precedent it would depend on what state you are in and whether bodily compensation is becomes a legal principle.
So your flaw with organ donations and that the fetus is using your body without your consent is that the fetus has absolutely no say in whether or not he/she was conceived. However, people who have sex are consenting to an act which can and does make babies, even when not necessarily intended. This is how the overwhelming amount of babies are conceived.
Eh, no need to apologize. I write articles too and when I check the comments, things seem to get off topic. As long as they relate to abortion, or if there is a side issue in the article, I think it's a worthy comment to be made. It is always good to read the whole thing though, so good for you there. 🙂
When I'm able to use my laptop rather than my phone, yes.
Someone who has leukemia, or failing kidneys, or a failing liver, often had absolutely no say in whether or not they needed a transplant.
By rejecting mandatory organ donation, you seem to be implicitly accepting the following ethical principle: Person A is not entitled to use Person B's body, not even to save Person A's life. Thus, a fetus is not allowed to use another person's uterus, not even to save its own life. Thus, abortion should be a right in the case of pregnancy from rape. You seem to be claiming that people with uteri normally have the right not to be pregnant if they don't want to be, but that they should be regarded as having given up this right when they chose to have sex. This is the responsibility argument.
I've spent a lot of time thinking (and writing) about the responsibility argument and why it shouldn't actually abrogate a person's right not to be pregnant; see this comment (and also this comment, not by me, which tackles the responsibility argument from a legal perspective). If you want to argue the responsibility objection with me, fine…but you're going to have to actually argue it, not just mention it and expect me to accept it.
I have sympathy for patients needing transplants certainly (not saying you were claiming I didn't, just saying). I just don't see the equation for mandatory organ donation.
And I think people should be able to prevent becoming pregnant, although I do have moral issues with forms of birth control and really don't like the idea of us treating our fertility/pregnancy as a disease. You should also pay for your own birth control when it's your sex life. That being said, birth control can and does fail and sex is a known activity which creates babies, again whether we intend for that to happen or not. When this is common knowledge by those participating, yes, responsibility should be taken which means accepting the consequences and choosing parenting or adoption, as another, completely innocent life is at stake.
I also don't believe in abortion even in cases of rape, but I have different arguments for that.
I don't see how you can possibly have a consistent ethical system that allows forced pregnancy even in the case of rape but does not allow for forced organ donation.
I have a consistent ethical system when it comes to life, certainly. I am consistently pro-life. Organ donation is a separate issue, completely.
The unborn child is always an innocent, defenseless human being deserving of the right to life, no matter how they were conceived.
Wow you anti-choicers love to tell lies. Abortion is NOT legal up until birth for any reason in most of the US.
Most abortions that occur after viability are for a fetal defect or the health of the woman and I do not see how anyone could be against an abortion to save a woman's life. But when it is all about the fetus to someone I guess it makes sense.
"Giving birth is safer than any kind of abortion."
That is another lie. An early abortion especially one done with RU-486 is MUCH safer then carrying to term and giving birth.
Consent to sex is not consent to pregnancy. A woman still has a right to control her body and make her own choice about what happens to and inside it.
I'm sorry, but this just reflects how ignorant the public is about how far-reaching the decisions of Roe v. Wade and Doe v. Bolton actually was. I am going to choose to explain this politely, even when you make all *sorts* of assumptions about those who disagree with you.
The decisions from the Supreme Court said that states could regulate abortion in later trimesters. However, abortion must be legal to save the health of the mother. This is to be decided by the abortionist and the mother and thus could really mean anything. Even attending a rock concert was decided as enough of a mental health reason.
And regardless as to if this were to occur in one state or all 50, it still sets the U.S. apart from all but 4 nations of 195.
Okay, so you didn't bother to really look into the analyses then I'm guessing. And considering your tone in replying to my other comments, unfortunately that does not surprise me…
Children dying of cancer are also innocent and defenseless. Mandatory bone marrow donations could save their lives.
I have looked at statistics and rate of death of women who have first trimester abortions versus carrying to term and giving birth. First trimester abortions are much safer for the woman.
Abortion should be legal at any point to save the life of the woman. If you let the woman die then both the woman and the fetus will not survive. How would that be better?
Um no except what I cited says otherwise.
Except that abortion is never medically necessary. An abortion is the direct intention of killing a child. Doctors can induce premature labor or do an emergency c-section. They try to treat both the mother and child. Sometimes the child dies, but indirectly and that is never intended, unlike an abortion.
Really? So there is no such thing an an ectopic pregnancy?
When my friend had a wanted pregnancy and they discovered their was no signs of life it wasn't necessary? You are telling me should she have been forced to carry the dead fetus for another 16 weeks?
Right…. I should believe things from websites that are obviously biased to your point of view. I could find just as many if not more sources that say the opposite. It all depends on your view point and what you want to believe and if you value the woman or the embryo more.
Well the difference of pro-lifers is that we care about the woman and the unborn child just as much.
An abortion is NOT "the direct intention of killing a child".
The intent is to remove/expel an uninvited visitor from your body, much like any parasite or tumor is removed. If the invader happens to die in the process of removal, that is a mere side effect, not the major intent.
You cannot care about them the same when you place the value of the potential life over the feelings and wants of the actual life. They can't be equal.