Aborting a Teen Pregnancy Cuts Risk of Premature Death by a Third

March 15, 2024 11:07 AM

The post title is the headline we arguably should be seeing in news outlets reporting on a recent study published in JAMA. The study was based on health outcomes from more than two million Canadians enrolled in universal health care, examining the impact on the likelihood of premature death (defined as death before age 31) of having a pregnancy as a teenager. Using 1.00 as the reference risk of premature death for someone with no history of teen pregnancy, the study found a health risk of 1.41 for people with a history of teen pregnancy resulting in an induced abortion, and a health risk of 2.10 for people with a history of teen pregnancy that resulted in live birth, stillbirth, miscarriage, or ectopic pregnancy.

The health risk numbers make the math pretty easy: 2.1/1.4 = 150%; 1.4/2.1 = 66.6%. In other words, the data show that a teen pregnancy that ends in anything other than abortion is associated with a 50% greater risk of premature death than having an abortion. Or, conversely, that having an abortion cuts the associated risk of premature death associated with teen pregnancy by a third. This conclusion wasn't the focus of the study, but it seems pretty important. The study's results are at least suggestive of the idea that having an abortion is the safer option, in terms of longevity, for a teenager faced with a pregnancy. (This is, of course, in addition to the known health risks, in the shorter timeframe, of carrying a pregnancy to term.)

But you'd be forgiven for missing this take-away from the way it's getting reported. The New York Times article gives the figures 40% and "more than double" in different places in the article, but doesn't spell out the connection. Worse, a commentary article in JAMA gets it backwards, claiming that "those with pregnancies that ended in induced abortion had the highest risk of premature death (followed by pregnancies that ended in live birth or miscarriage)." An article on The Hill described the study as finding that "while teenagers who had an induced abortion were at a high risk of premature death, people who gave birth or had a miscarriage had a higher risk." Which makes it sound like there's little distinction between the two outcomes.

Interested in your thoughts on whether I've gotten the math or the implications wrong (although you can take it as read that I get that this shows correlation, not necessarily causation). Assuming I haven't got it wrong, it seems like this would be good information to get out there as a counter to the (bad faith) argument that abortion restrictions are somehow about protecting women's health.

posted by mabelstreet (11 comments total)

I think we have to be careful about presuming causality in something like this. Is it that having an abortion reduces the risk of death in the years following or that people who are more likely to have an abortion are less likely to die in the years following? It's probably some of both, but I don't think this study gives us any way to separate those two cases.
posted by ssg at 11:18 AM on March 15 [2 favorites]

We don't have to be too careful, tho.

For instance we can say that banning abortion is bad for peoples' health, and that people criminalizing it are monsters.

We can say that everyone has the right to a free abortion on demand without apologies.
posted by constraint at 11:48 AM on March 15 [9 favorites]

Most of the associated deaths are injuries or "accidental" (which seems to include assault, so I assume it's more accurately "unintended by the victim"). In these circumstances, "induced abortion" seems to be a pretty good proxy for "anyone, anywhere, with some responsibility for this child actually giving a shit."
posted by praemunire at 12:09 PM on March 15 [2 favorites]

Note that the paper doesn’t make any causal claims, but the headline does.
posted by MisantropicPainforest at 12:11 PM on March 15 [2 favorites]

Notable that they stop tracking women who get abortions at 25 but only stop tracking women who give birth at 31. I assume the age discrepancy is already adjusted for somehow in the statistics but don't have the background to know for sure.

Regardless, this may be quite useful for some pushback against all those annoying claims that reproductive rights are bad for women's health akshually.
posted by joannemerriam at 12:30 PM on March 15 [1 favorite]

The Results section right at the beginning of the paper is already throwing me a bit. Is there a typo? They write, "Of those with a teen pregnancy, 60 037 (36.8%) ended in a birth (of which 59 485 [99.1%] were live births), and 106 135 (65.1%) ended in induced abortion."

Anyone know what their definition of "induced abortion" is? Because those numbers make it look like, at minimum, 3048 pregnancies ended in both a birth and an induced abortion. That's definitely not impossible, provided we define abortion as the premature termination of a pregnancy -- which might or might not result in a death -- and provided we define birth in such a way that premature termination of a pregnancy can still result in birth. (I actually like this kind of definition for a variety of reasons, and it also has the interesting consequence that my own son, currently age 12, was born from an aborted pregnancy.) But ... it's really unclear to me whether they have a very specific set of definitions or whether there's a typo.
posted by Jonathan Livengood at 12:40 PM on March 15

I think they have messed up the numbers somehow, because in the main text results they say: "Of all female teenagers who had a pregnancy, 60 037 (36.8%) ended in a birth (of which 59 485 [99.1%] were live births), 106 135 (65.1%) ended in induced abortion, and 17 945 (11.0%) ended in miscarriage or ectopic pregnancy."

I'm guessing they are using the teenagers as the denominator and the pregnancies as the numerator, so the result is more than 100% because some teenagers had more than one pregnancy. Pretty bad way of presenting results though.
posted by ssg at 12:49 PM on March 15 [1 favorite]

I doubt Jama is in the business of typos, but yes you can have an abortion and then give birth. It happens all the time , and is probably the most common outcome when a fetus is incompatible with life after (say, 24 weeks). Depending on local laws.
posted by MisantropicPainforest at 12:49 PM on March 15

Answering my own question: They are doing something very confusing with respect to their choice of statistical unit. When they report counts, the number is a count of pregnancies, but when they report percentages, they are reporting relative to people as the statistical unit. So, the oddity is due to some people having more than one teen pregnancy. At least, this is how I understand note f to Table 1.
posted by Jonathan Livengood at 12:52 PM on March 15 [2 favorites]

I was puzzled by that, too. The answer (in a note to one of the charts) is that it adds up to ore than 100% because some people had multiple pregnancies that ended in different ways.
posted by mabelstreet at 12:54 PM on March 15

yes you can have an abortion and then give birth.

My point was just that this depends very much on how the terms are defined. If, for example, we include in the definition of "abortion" that the pregnancy does not give rise to a birth, then there (trivially) won't be any abortions that result in births. That isn't anything to do with any biological or medical facts, it's bookkeeping. But it's potentially very important bookkeeping, since keeping the books one way or the other can make a difference with respect to policy decisions.
posted by Jonathan Livengood at 1:00 PM on March 15