Junk science is cited in abortion ban cases. Researchers are fighting the ‘fatally flawed’ work
The retraction of three peer-reviewed articles prominently cited in court cases on the so-called abortion pill – mifepristone – has put a group of papers by anti-abortion researchers in the scientific limelight.
Seventeen sexual and reproductive health researchers are calling for four peer-reviewed studies by anti-abortion researchers to be retracted or amended. The papers, critics contend, are “fatally flawed” and muddy the scientific consensus for courts and lawmakers who lack the scientific training to understand their methodological flaws.
While some papers date back to 2002, the group argues that now – in the post-Roe v Wade era – the stakes have never been higher. State and federal courts now routinely field cases on near-total abortion bans, attacks on in vitro fertilization and attempts to give fetuses the rights of people.
“When we saw the meta-analysis presented again and again and again – in the briefs to the Dobbs case” that overturned Roe v Wade “and state cases” to restrict abortion, “the concerns really rose,” said Julia Littell, a retired Bryn Mawr professor and social researcher with expertise in statistical analysis.
A meta-analysis is a kind of research that uses statistical methods to combine studies on the same topic. Researchers sometimes use these analyses to examine the scientific consensus on a subject.
Littell was “shocked” by a paper that said women experience dramatic increases in mental health problems after an abortion – primarily because of the paper’s research methods.
Of the 22 studies cited by the meta-analysis, 11 were by the lone author of the paper itself. The meta-analysis “failed to meet any published methodological criteria for systematic reviews” and failed to follow recommendations to avoid statistical dependencies, according to a criticism published in the British Medical Journal (BMJ).
Large scientific bodies have found no evidence to suggest abortion causes increases in mental health problems. The best predictor of a woman’s mental health after an abortion is her health before. What’s more, there is substantial evidence that women who are denied a wanted abortion suffer both mental and financial harms.
From the time it was published, this 2011 meta-analysis has drawn consternation. Still, it remains in the scientific record in a dispute that the 17 authors of the BMJ criticism, including Littell, say goes beyond mere scientific disagreement. The paper has been cited in at least 24 federal and state court cases and 14 parliamentary hearings in six countries.
Dr Chelsea Polis, a reproductive health scientist in New York City, who helped gather the group of academics, says her “concerns with the meta-analysis on abortion and mental health published … are based on it being, in my professional opinion, egregiously methodologically flawed”.
The researcher who wrote the article, Priscilla Coleman, a retired professor from Bowling Green State University in Ohio, has responded to calls for retractions with legal threats and descriptions of conspiracy. She said calls for retraction were “an organized effort to cull professional literature and remove studies demonstrating abortion increases risk of mental health problems to impact the legal status of abortion”.
Since the supreme court overturned the constitutional right to abortion and allowed 21 states to severely restrict or ban the procedure, a series of retractions and investigations show how the scientific community is slowly beginning to re-evaluate work cited in these court cases.
“We’re seeing claims made with legal force behind them, and that’s causing people to look at a lot of this research in a different way,” said Mary Ziegler, a professor of law at the University of California Davis, and an expert on the history of reproduction.
A second author whose work is at the center of the BMJ critique is David C Reardon, a longtime abortion opponent. A 2002 study by Reardon, also published in BMJ, is now under investigation.
BMJ said in a statement that the “issue remains under consideration by our research integrity team”, and that their final decision would be made “public once we have completed our internal process”.
Reardon trained as an engineer, but found his calling in research that claimed a connection between abortion and poor mental health. He founded the Elliot Institute in Illinois, an openly anti-abortion non-profit, to pursue that research.
Today, Reardon is affiliated with the Charlotte Lozier Institute, funded by one of the most powerful anti-abortion campaign organizations in the US, Susan B Anthony Pro-Life America. Reardon also co-authored two of the articles that were retracted before supreme court hearings, both by a colleague at the Lozier Institute. Reardon did not respond to multiple requests for comment.
According to analyses of the literature and experts such as Julia Steinberg, an associate professor of family science at the University of Maryland School of Public Health and a co-author of a recent critique of these studies in BMJ, the science is not in dispute. The “rates of mental health problems for women with an unwanted pregnancy were the same whether they had an abortion or gave birth”, an analysis by the UK’s National Collaborating Center for Mental Health found in 2011. That review was cited as one of the best by the US National Academies of Sciences, Engineering and Medicine, in its own 2018 review of the issue.
Other reviews, such as one from 2009 by the American Psychological Association, found evidence “did not support the claim that observed associations between abortion and mental health problems are caused by abortion per se”.
“One can be pro-choice or pro-abortion or anti-abortion, but still understand what the science says with respect to abortion and mental health,” said Steinberg.
Although matters of scientific integrity may seem academic, they can have concrete impacts on policy in the US post-Roe.
One of the few cases of scientific retractions to break through to the wider public was in Texas, where a federal court relied heavily on two studies in a decision to invalidate the approval of mifepristone – better known as the “abortion pill”.
The case was appealed all the way to the supreme court, where it was heard in March in oral arguments in Alliance for Hippocratic Medicine v FDA. Just weeks before the justices were set to hear the case, and as nearly the entire scientific community screamed about the “junk science” at its heart, the heavily cited studies were retracted by Sage Publications. Even so, the article’s claims remained in briefs before the court, and were cited as evidence by one of the most conservative justices, Samuel Alito.
Like Reardon, Coleman also recently had a paper retracted, this one in Frontiers in Psychology in 2022. The journal said publicly that the paper “did not meet the standards for publication”. Notably, one of the paper’s reviewers also worked at the Lozier Institute. Coleman unsuccessfully sued the journal over its decision to retract. The court ruled against Coleman in March 2023, Frontiers told the Guardian.
Coleman’s 2011 meta-analysis, published in the British Journal of Psychiatry, was also involved in a heated retraction fight in the UK. The first calls for retraction of the article came soon after it was published in 2012.
It was again brought to journal editors in 2022 after the BJP established a research integrity group. “Motivated by strong agreement with” the importance of scientific integrity, said Polis, “I led a group of 16 scholars to summarize and submit our concerns, again, about the Coleman meta-analysis to BJP.”
In response to these concerns, the BJP established an independent panel of experts to investigate. The panel recommended Coleman’s article be retracted, but was overruled by the Royal College of Psychiatrists, the professional association that publishes the BJP. The move prompted members of the independent panel and some editorial board members to resign.
Later reporting that appeared in the BMJ included panel members saying they believed the college declined to retract because they may not have had comprehensive legal cover in the United States. Coleman threatened to sue – twice – according to letters obtained by the BBC.
Although Coleman denied that her legal threats contributed to the BJP’s decision not to retract her study, she said help from attorneys had been important to defending her work.
“I have spent the last two years vigorously defending three of my own articles and without the financial means to hire highly competent lawyers and the time and opportunity to write lengthy rebuttals, the impact could have been very damaging,” said Coleman.
The Royal College of Psychiatrists responded to inquiries from the Guardian by sending a 2023 statement on its decision. That statement read, in part: “After careful consideration, given the distance in time since the original article was published, the widely available public debate on the paper, including the letters of complaint already available alongside the article online, and the fact that the article has already been subject to a full investigation, it has been decided to reject the request for the article to be retracted.” The statement added: “We now regard this matter as closed.”
Coleman has also defended her work when she testified in US courts, including in a Michigan hearing in which she said her study was “not retracted”.
Steinberg said: “That’s what’s really infuriating.”
Coleman “hasn’t even had to admit that she made an error”, she added.
Researchers also called for retraction of a 2009 article in the Journal of Psychiatric Research by Coleman and the anti-abortion activists Catherine Coyle and Vincent Rue. This article too has been under fire for years and even publicly debunked.
In spite of apparent flaws, Coleman included this 2009 article in her metaanalysis, which critics say compounds the errors.
Additionally, authors of the BMJ critique called for a 2005 article in the Journal of Anxiety Disorders by Coleman, Reardon and a Florida State University psychology professor, Jesse Cougle, to be accompanied by an expression of concern.
Ivan Oransky, one of the founders of the Retraction Watch blog, said that although retractions had become more common, they were nowhere near common enough to correct the scientific record. About one in 500 papers are retracted today, but perhaps as many as one in 50 ought to be, he said.
“All it does is further throw into question what the heck value these multibillion-dollar publishing companies are adding,” said Oransky. For critics of the scientific publishing industry, like Oransky, the response shows how flawed studies cited by courts are a “symptom” of problems with publishers, rather than a failure of courts.
To Littell, the solution is in plain sight: “We really need to be publishing fewer papers, better work, better science.”