Another response to: ‘Does abortion reduce the mental health risks of unwanted or unintended pregnancy?’ Fergusson et al.,2013

To the Editor

Below, I clarify in three points how my position differs from that of Fergusson and colleagues (Fergusson et al., 2013).

First, the state of the abortion and mental health literature is not in place to combine in a quantitative manner findings from different studies, as Fergusson et al. (2013) do, in order to draw conclusions about the causal relationship between abortion and mental health. This is because there are many poorly conducted studies in the literature and they cannot be combined with well-conducted studies (Kendall et al., 2012; National Collaborating Centre for Mental Health, 2011). In addition, those that are not so poorly conducted control for different factors in analyses and so effect sizes cannot be combined in a systematic manner.

Second, even though the literature is not in a state to conduct a meta-analysis, there are some high-quality studies that provide evidence that abortion relative to birth of an unwanted pregnancy is not associated with higher rates of subsequent mental health problems, suggesting that abortion is not a cause of mental health problems (National Collaborating Centre for Mental Health, 2011). For instance, using the Christchurch study Fergusson and colleagues (2013) found that all 95% confidence intervals of the odds ratios included the value 1, indicating that women who had abortions were not at a significantly increased risk of post-pregnancy anxiety disorders, depression, alcohol misuse, illicit drug use or suicidal behaviour relative to women who gave birth to unwanted pregnancies (Fergusson et al., 2013; Table 2).

Third and lastly, I contend published studies in the literature do not inform policies which grant abortions on mental health grounds, as Fergusson et al. (2013) wish to do. There are two reasons for this. First, as mentioned earlier, many of the studies are so poorly conducted that they cannot answer the causal nature of the relationship between abortion and mental health. Second, studies which are of higher quality inform policies which require women be warned of the negative psychological effects of having an abortion and not policies which grant abortions for mental health reasons. Elaborating on this second point, I will define the concept of target population. A target population is the population to which a study generalizes. To answer the question of whether policies granting abortions on mental health grounds are supported by the evidence we would have one target population in mind – women seeking abortions – and to answer the question of whether policies that inform women of the mental health risks of having an abortion are based on the evidence we would have a different target population in mind – women having unintended pregnancies.

Studies conducted to inform policies granting abortions because continuing the pregnancy would harm women’s mental health should have a target population of women seeking abortions. This is because these policies are only relevant for women who are seeking abortions (and strictly speaking they are relevant for women seeking abortions for mental health reasons). If the policy of granting abortions for mental health reasons is thought of as an intervention, we do not care to provide this intervention to women who are not seeking abortions. This is similar to a smoking cessation treatment. We care to know whether a smoking cessation treatment helps those who smoke, not those who do not smoke.

In contrast, to inform US state policies which require women are told of the negative psychological effects of having an abortion, the target population is women having unintended pregnancies who must decide what to do with their pregnancy – have an abortion or carry it to term. The reason a woman is told of the negative psychological risks is to inform her decision on whether or not to have an abortion. The information is provided with the intention of helping women who have an unintended pregnancy decide to have an abortion or carry the pregnancy to term.