Steinberg
Abortion and mental health: findings from the national comorbidity survey-replication. Steinberg JR, McCulloch CE, Adler NE. Obstet Gynecol. 2014 Feb;123(2 Pt 1):263-70.
- After controlling for numerous factors, women with a history of abortion were 2.3 times more likely to experience subsequent substance use disorders (CI 1.35-3.92).
Criticism by Priscilla Coleman, Ph.D.
Published last week in Obstetrics and Gynecology is the latest attempt by Julia Steinberg and colleagues to manipulate our understanding of the psychological correlates of abortion. The authors analyzed data from the National Comorbidity Survey Replication of 936 US women, ages 18 to 42 to compare women who had reproductive histories consisting exclusively of abortion(s) and childbirth(s).
Steinberg and colleagues note that the only women included in the childbirth group were those who had children under the age of 17 at the time of data collection. The authors’ rationale for eliminating women with children over 18 is based on the need to estimate each woman’s age at the time of her first childbirth in order to perform post-pregnancy mental health comparisons of women who aborted and delivered. For this purpose, they used the age at survey completion and age ranges of the individuals’ oldest child (0-4, 5-12, or 13-17). As they note: “This allowed us to compute a possible range for a woman’s age at first childbirth to determine when mental health problems occurred relative to her first childbirth…The analyses presented used the youngest possible age within the range, because this attributes the maximum number of mental health disorders to be in the post-pregnancy period and the mean age at first childbirth using the youngest possible age was the same as the mean age at first abortion.” Here you have the most central flaw of the Steinberg et al. design…they have only a vague idea of the age of women sampled at the time of their first childbirth, meaning that many mental health problems recorded as occurring post-childbirth may have easily happened prior to birth. Selecting the youngest possible age at first childbirth from a possible 4 year time frame effectively ensures that the post-childbirth profile of mental health is more likely to include very young mothers (who are inclined to experience a birth as a significant stressor), and to incorrectly capture many pre-childbirth mental health outcomes as post-childbirth occurrences. In fact, the authors noted “women who were younger at the time of their pregnancy event were more likely to have post-pregnancy mental health problems.”
This is an excellent example of the horrendous methodological contortions needed to support the claim of no association between abortion and mental health and the breeches of science allowed in our top journals to yield politically correct results. In what other areas of scientific inquiry would such inexact measures be permitted in an article appearing in a top-tier peer-reviewed journal?
If this is not enough to convince you that the results are essentially meaningless, please read on as I highlight additional problems with the paper.
1) Only 44% of abortions are reported. Obviously women who conceal an abortion are more likely to be burdened by the procedure. A significant percentage of women were likely incorrectly classified as childbirth-only when they had an abortion history. A concealment rate of 56% is unacceptable.
2) The authors did not include a comparison group of women who had not had an abortion or childbirth, effectively eliminating 484 women. There is no explanation for excluding this group and we know from prior research that this demographic is associated with a lower risk of mental health problems when compared to women who have had an abortion. For example, in the meta-analysis I published in 2012, the combined effect across several studies revealed a 59% increased risk of mental health problems when comparing women who aborted with women who had not aborted.
3) A large literature on abortion and mental health has revealed that women’s education and marital status at the time of the procedure are critically important variables to control when comparing women who abort to those who deliver. However, the authors state that they did not have access to these data.
4) By including women with only abortion(s) or only childbirth(s) in their reproductive histories, we miss the typical profile of woman in the US who have had an abortion; most will also experience childbirth and after an abortion childbirth may be particularly difficult. A subsequent birth is the point when many women will learn more about fetal development, experience maternal-fetal attachment, and may feel regret and/or remorse for a prior decision to abort, particularly if it was misinformed or coerced.
5) There is a significant body of evidence indicating a dose effect wherein more than one abortion is predictive of a greater risk for post-abortion mental health problems. Steinberg and colleagues included women with a single abortion and with two or more abortions. Why didn’t they separate out these groups and compare mental health status?
6) Instead of reporting the average post-pregnancy time for which mental health outcomes were examined, a logical and much needed figure in order to makes sense of the results reported, here is what the authors chose to report: “The total follow-up time for women was 8,095 years for anxiety disorders, 8,437 years for mood disorders, 8,909 years for impulse-control disorders, 8,923 years for substance use disorders, 9,645 years for eating disorders, and 8,920 years for suicidal ideation. In addition, 50% of women contributed 9 or more years for anxiety disorders, 10 or more years for mood disorders, 12 or more years for impulse control disorders, 12 or more years for substance use disorders, 12 or more years for eating disorders, and 13 or more years for suicidal ideation.” What about the other 50%? Did the length of follow up differ significantly between the abortion and birth groups?
Steinberg and colleagues’ rather transparent motives for publishing this report are apparent at the beginning and at the end of their article. In the second sentence, they note “Understanding whether such a common procedure causes mental health problems is important for clinical practice and policy.” The reader is led to believe that this study published in the prestigious “Green” journal will provide some definitive evidence to address the causal question.As Dr. Steinberg surely learned in her undergraduate training, her design does not permit causal attributions by any stretch of the imagination. This is a correlational study with imprecise measurement and inadequate control variables. Finally in the last paragraph, the authors state “The results reported here show that policies that require women be told that abortion increases their risk of anxiety, depression, and suicide lack an evidence base.” They have not re-analyzed the dozens of studies that have provided a foundation for recently enacted laws. Shame on the editors for permitting publication of this bogus, agenda-driven study.
Additional Notes From David C. Reardon, Ph.D.
1. With only 44% of the expected rate for abortions in this sample, the concealment rate is very high and means that many women with a history of abortion are misclassified by the authors into the control group (delivering women with "no history of abortion"). Given the diluting effect this would have on results, this suggests that negative findings (lack of statistically significant differences) really don't tell us much. On the other hand, positive findings (statistically significant results) are most likely truly significant since the differences must be pronounced enough to still show up despite the dilution effect caused by concealment and misclassification.
2. Even in Model 4, with all the controls in place, higher rates of subsequent mental illness (RR>1) are seen for women admitting a history of abortion in every classification examined. Given the relatively small sample size, the lack of statistical significance in these cases can very likely be due to small sample size (and the concealment/misclassification problem) rather than lack of any true association.
3. It is unfortunate, and inappropriate, that the authors used history of miscarriage as a control variable on several counts. At the very least, women responding to surveys of this type frequently will disguise a past abortion by describing it as a miscarriage. In addition, abortion may contribute to a miscarriage. At the very least, the authors should reanalyze the data to show us how the results would differ if women reporting a first miscarriage were removed from the sample, which is the only way to properly eliminate any confounding effects that may be associated with miscarriage.
4. The authors chose to control for the number of pre-abortion psychiatric episodes/diagnoses, but they failed to explore whether abortion is associated with an increase in the number of episodes subsequent. In other words, the authors made the odd decision to consider the frequency of prior psychiatric episodes but then failed to explore the concern that the intensity and frequency of subsequent mental health problems.
Given the weakness of this study, the authors' conclusions that this study suffices to demonstrate that other studies (including large record based studies) showing associations between abortion and mental illness can be ignored is imprudent and very likely engaging in a Type II error.
Moreover, given these weaknesses, it is especially concerning that the authors dismissed the fact that substance use problems remained significantly associated with abortion despite all the control used to reduce significance, arguing that this outcome, too, could probably be reduced to statistical insignificance if only they were able to find yet more controls to apply to the problem.
In short, the authors rush to dismiss a large body of evidence, including the self reports of self-aware women, suggesting that abortion contributes to mental health problems, based on this analysis of a flawed data set and questionable methodological choices, reaches way too far.