Flawed paper on negative impact of abortion used to help over turn Roe vs. Wade not withdrawn so reviewing panel resigns (contrast with paper on gender dysphoria)

[IwantToRetire]

An independent panel resigned in a row over controversial research about the impact of abortion on the mental health of women, BBC News has been told.

The research, which is still being used in US legal cases about limiting access to abortion, was published in the British Journal of Psychiatry, in 2011.

Last year the panel, which was set up to investigate complaints about the paper, recommended it be withdrawn.

But journal-owner, the Royal College of Psychiatrists, overruled it.

The Royal College said the work had already been fully investigated.

However, BBC Newsnight and the BMJ understand all three panel members, and two other members of the journal's editorial board, resigned in protest. They have called into question the journal's editorial independence.

Full article here https://www.bbc.co.uk/news/uk-66249015

And segment on newsnight at about 20 mins in at BBC iPlayer - Newsnight - Triple By-Election Showdown

What’s your opinion on this?

There is a clear political motivation set out in the article for seeking retraction of the paper (concern it is being used to good effect to overturn laws allowing abortion in the US). Other than suggesting one of the criticisms was combining data in ways that meant women were counted multiple times, the article does not tell us what the criticisms were. It is impossible for me to assess on the strength of that article. It is possible that the author of the paper also had a political motivation, but the article does not suggest this.

It is perfectly possible to be pro-choice and accept (if the research does indeed show this) that women who've had an abortion have an 81% increased risk of developing mental health problems.

Your thoughts??

I'm also strongly pro choice, but agree that a paper should not be withdrawn because it says something I don't like.
Like PaleBlueMoonlight says, it's difficult to assess on the basis of that article. This is why more open debate and information is needed.
The answer from us pro-choice people should be to read the paper carefully and look at its data. It might be right. It might have serious flaws.
If it's the former, then it could be used as a basis for more research on how to support women. If its the latter, then those flaws needed to be highlighted and used as a call for better research with robust data.
Simply ordering them to retract it doesn't help.

I think it's quite unusual to see research published in 2011 (and so undertaken some time before) still being used as evidence supporting or opposing a decision, unless it's being cited as a seminal paper.

I am pro choice but it doesn't look good to want to retract a paper published so long ago because it has been misused. The obvious remedy is to get funding for new research into this topic that could give a more definitive answer. In addition, why couldn't the letter's authors submit a response outlining their concerns? Medical journals often publish these.

I've not read the original article but there is no doubt that retracting the paper would be used by pro choice activists as concealing the evidence.

@IwantToRetire
Your silence reaches so many.

@IwantToRetire
Your silence reaches so many.

Have no idea what you are talking about.

Maybe I dont have any thoughts and posted to get other FWR's thoughts.

And maybe I actually have a lot of committments and haven't yet had time to read article in full, do some research and re listen to newsnight segment.

In the mean time, can only find it funny that someone's only ability to comment and contribute to a thread is to ask just one contributor what they think.

Frankly bizarre - and just to be contrary may not add any opinion at all.

For me the benefit of FWR is to get a wide range of opinions.

Even on this thread people are misquoting the results! 81% had mh problems not 81% increased risk. How does that compare to the general population? Must be 30 to 50% anyway if the oft quoted 1 in 3 kids at school have mh issues before any abortions. And people who have issues might be more likely to abort. It should be retracted as the methodology is flawed and the results are false

But agree new research might be helpful. Also 20 years is pretty current for research so no surprise papers are being used.

I copied it direct from the article.

"It concludes that women who've had an abortion have an 81% increased risk of developing mental health problems."

Haven't read the paper though, is the article inaccurate or an I missing something (perfectly possible)?

Surely, 'an 81% increased risk' doesn't make any sense, does it? What exactly does it mean? If the base rate was 1 in 20 (or whatever figure you like), what would the '81%' increase it to? I can't parse it at all.

It makes perfect sense. An 81% increase in risk would increase 1 in 20 to 1.81 in 20, or from 5% to 9% of women.

I suppose it could maybe mean that your risk of mental health problems is 81% greater than.someone who was pregnant and did not have an abortion, or perhaps than the general population, but both beg loads of questions, such as, over what time period, and what does mental health problems mean?

I know the paper apparently was listed as reference and used in a brief that was submitted to the Supreme Court regarding Roe v. Wade, but I don’t think it had any impact on their decision as womens MH is not mentioned as a rationale in their decision brief.

It’s all about how the states should decide for themselves. More than half the US states also wrote letters to the Supreme Court asking to have Roe v. Wade overturned as well. I think that had the most impact.

Yes sorry the other way round obviously 😜 meaning a small difference not big was my intended point

Doesn’t really matter as it was recommended for retraction due to methodological failures meaning it didn’t meet good practice standards for a systematic review and called into question the results and conclusion.

In other words, the study is worthless.

Surely, 'an 81% increased risk' doesn't make any sense, does it?

I can’t see any way it could possibly make sense. Are they comparing women who had abortions to women who were forced to carry an unwanted pregnancy to term? I doubt it somehow.

Here is an article, it still doesn’t make their methodology clear but I don’t think there is any reasonable way to infer causality in research like that.

https://www.guttmacher.org/news-release/2012/study-purporting-show-link-between-abortion-and-mental-health-outcomes-decisively

It was a systematic review so they did not study any women. They collected and reviewed already published studies. The problem was their inclusion criteria for which studies were included and the fact they misrepresented associations as causal links.

It was a systematic review so they did not study any women

the studies they included did! The 81% increase has to mean they compared women who had abortions to some other group. The big problem is that there is no reasonable way to do that and infer that problems result from abortion and not from unwanted pregnancy or from the life circumstances surrounding unwanted pregnancy. Seems like this researcher did a “systematic review” mostly of her own fundamentally flawed research, which compounds the original problem.

I read another article about systematic reviews very recently; it seems like the problem of systematic reviews of worthless papers is currently getting some attention and various bodies are moving to tighten up the inclusion criteria.
I wonder if the push to retract this specific study is related to that.

Duh, obviously the studies they reviewed did.
“The 81% increase has to mean they compared women who had abortions to some other group”

No it doesn’t mean that they did any comparison at all. A systematic review consolidates prior findings. It shouldn’t do any new comparisons. That simply could be a weighted average of women reported in several or all of the studies with increased risk of MH problems.

“Seems like this researcher did a “systematic review” mostly of her own fundamentally flawed research, which compounds the original problem.”

Yes that was one criticism that she waived inclusion criteria for her own studies and excluded ones that met the criteria but didn’t have the results she agreed with. So clear case of cherry picking.

No it doesn’t mean that they did any comparison at all

“81% increase” means that a comparison was made. I don’t know exactly what you don’t like about that almost-tautological statement but I don’t think it’s worth arguing about!

Yes, but any comparison would not have been made by the authors of the systematic review. That’s what I “don’t like” the fact you don’t seem to understand a difference between a study and a systematic review of numerous studies.

% increase has to mean they compared women who had abortions to some other group.

It may not be another group, it may be the same women pre-abortion or at the point an abortion was undertaken.

If the review had flaws them these should be critiqued. But the fact it was used in the case leading to the overturn of Roe vs Wade is not itself enough and it is quite scary to suggest it might be. Though there may be an argument for judiciary to be trained in how to assess the quality of scientific papers.

Comparing individuals pre- and post treatment is included in the concept of “comparison between two groups.”

Any such comparison in the case of abortion will have exactly the same problem - that you can’t disentangle the effects of an abortion from the effects (and causes) of an unwanted pregnancy. Unless you compare women who have abortions to women who were forced to carry their pregnancy to term.

Yes, but any comparison would not have been made by the authors of the systematic review

That’s not the case. Systematic reviews entail re-analysis of the combined data from multiple studies. But I’m glad we got to the root of the misunderstanding!

That is the case.
They’re analysing the data already gathered by the studies under review, that’s not the same as comparing individual subjects which is gathering new data.