Abortion and Teenage Depression?

The first can cause the second, right?  Not exactly; in fact, that looks wrong:

 A press release for the study, which is published in the journal Perspectives on Sexual and Reproductive Health, explains, “The researchers found that young women in the study who had an abortion were no more likely to become depressed or have low self-esteem within the first year of pregnancy — or five years later — than their peers who were pregnant, but did not have an abortion.” The data was pulled from the National Longitudinal Study of Adolescent Health, which surveyed 289 girls between the ages of 13 and 18.

The possibility of psychological harm has been appealed to in arguments for parental notification and pre-abortion warnings.  But now it looks as though the harm may well not be there.

 The truth, according to science, is that “on average, abortion does not appear to have major psychological consequences — for adult women or for teens,” says [lead author Jocelyn] Warren. Marie Harvey, a public health professor at Oregon State University, which conducted the research with the University of California, San Francisco, said: “We have policies being made that are not evidence-based, and that have adverse consequences for women’s health.”

Gosh!  Not evidence-based?!?  Who would have guessed?  It makes it sound as though maybe there’s another agenda being acted on.

11 thoughts on “Abortion and Teenage Depression?

  1. What, so teenagers dealing with the hormones associated with pregnancy and the circumstances surrounding their pregnancies might bring have depression regardless of the choice they make? HEAVENS NO.

    That’s always been my issue with the argument that the hormones that follow abortion cause more depression. PREGNANCY hormones cause depression.

  2. Of course, there’s also the job of giving birth to an unwanted child. Must interact with the hormones well.

  3. I’m actually surprised by that. I would think there is a level of depression but it comes from “I did something the world says in wrong” rather than the actual procedure.

  4. LaNeshe, I actually agree. The message “abortion is murder” is pretty visible, and surely the sort of thing that could induce depression. So what’s going on? I suppose it could still be more depressing to give birth to an unwanted child. But I’m also wondering if having an abortion might answer to a very deep need for preserving oneself. Maybe one’s biology is saying something like “You did well! We can now move on.”

    Does any of that sound plausible to you?

  5. I wouldn’t get too excited about this study. It is a tiny, retrospective study with minimal controls. A sample size of 289 for a retrospective study, especially on something as causally complex as this, is virtually meaningless. This is a grad student project published in an obscure journal. If we are going to criticize our political opponents for not being evidence-based (as we should), we need to avoid being selectively gullible when it comes to ‘evidence’ that supports out politics.

    In my fantasy alternative life, I am a science reporter – it frustrates me that media attention to research is doled out entirely on the basis of the newsiness of the topic, and not the quality of the evidence.

  6. For clarification: the lead author is a research associate/post doc, not a grad student. She holds a faculty position. The second author is chair of Health and Human Sciences.

    I can’t speak to the quality of the journal; it’s impact factor is 2.22, which would be low in the physical sciences (for which I have some calibration), but does that make it obscure? It’s a Wilely journal.

  7. Thanks, LaNesha.

    Having looked around a bit, I think it’s important to note that the original survey has over 27,000 subjects, so the validity of the results should depend on how the smaller result was pulled, and not on absolute size, I think. The original survey is in fact the only one that size of US adolescent sexuality.

  8. OK well the authorship point was less important.

    JJ – I am not sure I understand your last point. It is sample size that is relevant to the strength of the results, not how that sample was formulated (although of course a distorted method of choosing the sample could make things worse). No matter how they got their 289 (and I believe it was just straightforwardly the 289 who got pregnant during the relevant year – a little more than 1%, which seems plausible) they only had 289 to go on, and only 69 cases of abortion. When you think of the wide variety of selection biases that might skew the results and wouldn’t be tracked in the demographic data – teenagers who got abortions might be more effectively educated, more resourceful, living in communities with better resources and more progressive attitudes, less religious, who knows what else – the significance of the result strikes me as vanishingly small. Retrospective studies are the weakest kind, and typically have sample sizes in the thousands before they are taken seriously, and that’s even when they are measuring something much less causally and operationally murky than this.

  9. Indeed, the actual journal article says: “The lack of association between abortion and our outcomes could reflect other factors, including
    insufficient sample size to detect an effect. … Furthermore, some unobserved common factors … may drive both the decision to abort (or not abort) and adverse changes in mental health. The failure to include
    such factors could have biased the estimates.”

    I have now read the article – They actually did the interviews themselves, which was not the impression I first had – so I suppose it counts as a prospective rather than a retrospective study. But this changes my concerns about study strength very little.

  10. Interested readers might want to check out Edward Hagen’s work on the (alleged) evolutionary functions of depression, including postpartum depression. His “bargaining” model sort of frames depression as a way of managing/negotiating social support, cooperation, and/or benefits of various kinds. In the case of postpartum depression, the model seems appropriately framed as a way of managing or negotiating paternal investment time/resources in childcare, though depending on the family/social situation (during and toward the end of the Pleistocene) the investments might have involved more than biological fathers and so postpartum depression could have affected the mother/ childcare / family /society time/resource investments of other family and society members. However close to or far from the truth, such claims and the broader topics generally are highly relevant to how we understand and respond to (alleged) psychiatric disorders, how we understand and respond to parental investment needs/theories, and how many of our antecedent (and often socially conditioned) beliefs on these matters can influence our current and developing beliefs about them often without appropriate regard for the epistemic status of our past, present, and future beliefs.

    See in particular Hagen’s 2003 book chapter titled, “The Bargaining Model of Depression”, his 2002 article titled, “Depression as Bargaining: The Case Postpartum”, his 1999 article titled, “The Functions of Postpartum Depression”, and his 2007 article co-authored with H. Clark Barrett titled, “Perinatal Sadness among Shuar Women: Support for an Evolutionary Theory of Psychic Pain”.

    You can find PDF files of these writings on Hagen’s website, listed here:

Comments are closed.