Is There Sufficient Evidence to Support a Link?
By E. Christian Brugger
WASHINGTON, D.C., SEPT. 14, 2011 (Zenit.org).— Does induced abortion increase a woman's risk of mental health problems? The question has been asked continually over the past several decades with dozens of studies indicating a positive correlation , but a few well-publicized studies are arriving at the opposite conclusion.
An example of the latter is a widely quoted report in 2008 by the American Psychological Association Task Force on Mental Health and Abortion. The report confidently concludes that there is "no evidence sufficient to support the claim" of a positive link between a woman's abortion and increased mental distress. Abortion advocacy groups eagerly jumped on the report to announce that abortion posed no threat at all to a woman's mental health.
The APA report comes under harsh methodological scrutiny in the September issue of the British Journal of Psychiatry (BJPsych) in a study by Priscilla K. Coleman, professor of human development and family studies at Bowling Green State University in Ohio. Coleman claims that the 2008 report, which alleges to collect and analyze all relevant empirical research after 1989, employs inadequate criteria to select which studies to incorporate and which to exclude from its analysis: it ends including the results of "numerous weaker studies" and unjustifiably eliminating "dozens of methodologically sophisticated international studies."
Coleman performs her own comprehensive "meta-analysis" of research published between 1995-2009: the "largest quantitative estimate of mental health risks associated with abortion available in the world literature." She carefully analyzes 22 peer-reviewed studies (15 from the U.S. and 7 from other countries) covering an enormous sample size of 877,181 participants, of whom 163,831 had undergone an induced abortion. Coleman selects for five mental health outcome variables: anxiety, depression, alcohol use, marijuana use and suicidal behavior. She concludes that overall the experience of abortion led to a staggering 81% increased risk of mental health problems across all the variables. The observed effects were strongest when women who had an abortion were compared with women who chose to carry to term. She notes that this is consistent with scientific data demonstrating that delivering a baby actually lowers the risk of mental health problems — that it has "protective effects" on women's mental health (e.g., suicide rates are half for women associated with birth than in the general population).
Critics of Coleman's analysis will no doubt ask: "What about the recent Danish study?" referring to an article published by a Danish research team in January 2011 in the New England Journal of Medicine (after Coleman's final revisions had been submitted to the BJPsych). Using data recorded by the Danish government, the study compares the number of first-time psychiatric visits (for any type of mental disorder) of women before and after abortion. It concludes that since there is no statistically significant difference in the before-and-after numbers, abortion does not increase a woman's risk of psychiatric disorder (but, the study claims, carrying a pregnancy to term does!, prompting the Huffington Post to publish the headline, "Abortion Doesn't Increase Mental Health Risk, But Having A Baby Does").
When the Danish study was released, Coleman published a reply, which LifeSiteNews picked up on Jan. 27, 2011. She argues that four major problems compromise the reliability of its conclusion.
First, numerous studies illustrate that women who are pregnant and actively contemplating abortion face very high levels of stress; and many are in abusive relationships. Consequently, they are likely to have first-time psychiatric visits before their abortion, but nevertheless directly related to their abortion decisions. The Danish researchers, however, take the numbers of pre-abortion psychiatric visits to be the result of factors other than abortion (since they occur before the women have their abortions). And they tendentiously conclude that if the women are suffering from distress caused by factors other than abortion before their abortion, then their post-abortion problems must be related to those same factors. In other words, the abortion must not be the cause of their mental distress.
Second, the Danish study does not control for relevant "third variables" such as "pregnancy wantedness, coercion by others to abort, marital status, income, education, exposure to violence and other traumas, etc." But failure to account for just one of these variables has resulted in multiple other studies being rejected for publication. This study accounts for none of them, yet was accepted for publication by the prestigious New England Journal of Medicine.
Third, the study includes no women with psychiatric histories prior to nine months before their abortion — "only the healthiest of women" are included. But women who have longer-standing histories of mental health problems are most likely to be affected by an abortion. Moreover, women who have had repeated abortions are also excluded, another pool who are likely to suffer problems after their procedures.
Finally, the study only follows the women for 12 months after their abortion. But there is considerable evidence showing that mental health issues arising from abortion do not manifest themselves sometimes for several years.
Coleman asserts that if the study had included all women who underwent an abortion, compared with all who bring their pregnancy to term, and then compared all the number of psychiatric visits of both groups before and after abortion, and controlled for the relevant "third variables" noted above, then the data could be trustworthy. As it is, the study's conclusions lack statistical integrity.
At the outset of her BJPsych article Coleman admonishes her fellow researchers in the explosive field of abortion and mental health to keep the bar high in their standards of research. Scholars, she says, working in "this highly politicized area" must be rigorous in applying "scientifically based evaluation standards in a systematic, unbiased manner when synthesizing and critiquing research findings." If they do not, they "open themselves up to accusations of shifting standards based on conclusions aligned with a particular viewpoint." Indeed. She notes, in addition, that biased research may result in hurting women.
 See, for example, J.M. Thorp Jr., et al. "Long-term physical and psychological health consequences of induced abortion: review of the evidence," Obstet Gynecol Surv., vol. 58, no. 1 (Jan. 2003), 67-79.
* * *
E. Christian Brugger is a Senior Fellow and director of the Fellows Program at the Culture of Life Foundation; and the J. Francis Cardinal Stafford Chair of Moral Theology at St. John Vianney Theological Seminary in Denver, Colorado.