Updated

A study purporting to find a link between abortions and mental illness does not hold up to scrutiny, according to a new report in the Journal of Psychiatric Research.

The original study, conducted by Priscilla Coleman of Bowling Green State University in Kentucky, has been a source of contention since its publication in 2009, when critics pointed out flaws in the statistical analysis. Those errors triggered a correction by Coleman and her colleagues, but outside researchers found other problems with the paper. Most importantly, they report in the February issue of the journal, the original researchers included mental health ailments not only after abortion, but all across the life span, making it impossible to know whether the psychological problems came before or after the procedure.

"This is not a scholarly difference of opinion; their facts were flatly wrong. This was an abuse of the scientific process to reach conclusions that are not supported by the data," study researcher Julia Steinberg, an assistant professor in the University of California, San Francisco's department of psychiatry, said in a statement. "The shifting explanations and misleading statements that they offered over the past two years served to mask their serious methodological errors."

Abortion debate

The mental health effect of abortion is a hot topic, largely because abortion itself is a matter of vociferous political debate.

High-quality studies on the topic, however, suggest that an elective abortion does not increase the risk for mental health problems. In 2008 an American Psychological Association panel surveyed more than 150 studies on abortion and mental illness and confirmed that while some women experience sadness and grief after an abortion, there is no increased risk of mental health problems for these women. The panel warned, however, that more high-quality studies of abortion were needed, as the task force had to toss out many studies that had serious methodological problems.

Coleman's 2009 paper used data from the National Comorbidity Survey (NCS) in the United States to compare the mental health of 399 women who had an abortion with that of 2,650 women who had never had an abortion. She and her colleagues reported that women who had the procedure had higher rates of anxiety, depression and substance abuse disorders compared with women who had not.

But a 2010 analysis by Steinberg and her colleague Lawrence Finer of the Guttmacher Institute failed to replicate those findings. The exchange continued with a statistical correction by Coleman and her colleagues, but Steinberg and Finer say the correction only unearthed a deeper problem in Coleman's research.

The NCS data included whether the women had ever had a mental illness, and whether they had mental illness symptoms in the month and in the year before they were interviewed, with no data on mental health changes specifically after the abortion. After analyzing the data, Steinberg and Finer found that the only way to get the results Coleman and her colleagues came up with was to use the lifetime mental illness data, not the data from the prior month or year.

The means that many of the women interviewed could have had anxiety, depression or other mental illness before their abortions. [5 Myths About Women's Bodies]

"You just have no way of knowing when the mental health outcome occurred relative to the abortion," Steinberg told LiveScience.

Coleman responds

Coleman confirmed in a response published in the journal that she and her colleagues did use lifetime mental health history "hoping to capture as many cases of mental health problems as possible." She also wrote that because 70 percent of the women interviewed had their abortions before the age of 21, it is likely the mental illnesses came later, in the women's 20s and 30s. But Steinberg said the data can't show whether or not that's the case.

In an email to LiveScience, Coleman wrote that she and her colleagues never asserted that abortions caused the mental health problems. Steinberg declined to comment on Coleman's intentions, but pointed to phrases in the original paper such as "the effects of abortion," which seem to insinuate causality.

The Journal of Psychiatric Research is not retracting Coleman's original paper. However, Steinberg and Finer's analysis was accompanied by a commentary by the journal's editor Alan Schatzberg and Ronald Kessler, the principle investigator of the National Comorbidity Survey.

"Based on our joint review and discussion of the debate, we conclude that the Steinberg-Finer critique has considerable merit and that the Coleman et al. (2009) analysis does not support their assertions that abortions led to psychopathology in the NCS data," Schatzberg and Kessler wrote.

Furthermore, the researchers wrote, studies on the effects of abortion should not compare women who have had the procedure with all other women, as did Coleman and her colleagues as well Steinberg and Finer in their replication study; instead, women who chose to have an abortion should be compared with women who have had unwanted pregnancies who did not choose abortion.

"These strategies should be the focus of future research on the extent to which elective abortions lead to mental disorders," Schatzberg and Kessler wrote.