I don’t often blog, but I was recently given cause to do so, and thus here we are. While I’m sure most people have, by now, heard of the sad case of the death of Savita Halappanavar, the summary found here should be reasonably clear and current. This case has very much been at the forefront of Irish political discourse for the last few weeks and rightly so, as it seems that due to a long standing lack of clarity on the law regarding abortion in Ireland a woman may have needlessly died. Abortion has long been a fiercely debated topic in Ireland, with most politicians trying desperately to avoid ever having to address the issue. Indeed it is only after a ruling by the EU court of Human Rights, several high profile cases in the Irish courts and the death of a woman in an Irish hospital that legislation seems to be imminent. However, this has all been covered in depth elsewhere, and with greater clarity than I could manage.
That said, there is one specific claim which I would like to address, and that is the oft repeated nugget on social networks and elsewhere, and which has again surfaced as part of the current Irish debate; that having an abortion makes one six times more likely to commit suicide. This claim is supposedly backed up by a pair of Finish papers, the most recent of which covers maternal deaths from 1987-2000. I intend to examine this claim briefly, and provide some commentary on the same.
Firstly, the data itself. The paper indicates the following;
The first thing you should notice is that the mortality due to suicide after induced abortion (31.9 / 100,000 women) is not six times that of the general population of women (age adjusted 11.8 / 100,000 women) as is implied, but rather a little less than three times that figure. It is around six times the suicide rate of women who are pregnant, but being pregnant and giving birth is in fact a protective factor for suicide; that is, becoming pregnant is generally associated with being less likely to die by suicide than in the general population of women. Should you become pregnant and choose an abortion, you may be at a higher risk of suicide, but that risk has not increased six fold compared to your risk before becoming pregnant. It is misleading to compare these two statistics side by side without explanation – after all, given that a pregnant woman is less than half as likely to commit suicide as she normally would be, one could just as easily say that giving birth double’s your risk of suicide as you are no longer benefiting from being pregnant! Both of these statements misrepresent what is shown in the data. There is nevertheless an increase in the rate of suicide among women who have had an abortion when compared to the general population of women. To put this risk in context, it is worth nothing that Finnish women included in this study and who have had an abortion are still less likely to commit suicide than Finnish people who suffer from the debilitating condition of being male. (Suicide rate of Finnish men c.45 / 100,000 at the midpoint of the period when this data was gathered.) I assume that we’re not going to see the people willing to make this rather alarming claim proceed to suggest that we should take action to curtail the number of males in society, so perhaps they might also take a moment to consider if it is reasonable to use this data to suggest that the number of abortions should be similarly curtailed.
That said, I have no intention of ignoring this statistic; lets take a further look at the data provided in this study and perhaps come to some understanding of what it means. Helpfully recorded along with the number of suicides per 100,000 we also find the number of deaths by homicide and unintentional injuries. In both cases, the risk to women who have had an abortion is higher – in the case of injury almost twice as high as the risk to the general population of women (20.4 vs 10.8 / 100,000) and more than five times higher than the population of pregnant women (20.4 vs 3.9 / 100,000). The risk of homicide displays a similar pattern; women who have had an abortion are more than three times as likely to be killed violently than a women in the general population (7.7 vs 2.1 / 100,000) and eleven times more likely to be murdered than a pregnant women (7.7 vs 0.7 / 100,000). Eleven times more likely! I wonder why those making these claims decided to focus on the comparatively paltry six times increase in the number of suicides among women who have had an abortion when there existed a much larger and more shocking increase in the rate of homicide occurring to the same group? Could it perhaps be that it is much harder and more taxing on one’s credulity to draw a link between abortion and homicide rates than it is to draw a link between abortion and suicide rates? The data doesn’t show a causative abortion-suicide link specifically; rather, those making this claim rely on the reader’s presumption that there must be an intrinsic, common sense and causative link between abortion, guilt and suicide. What this data in fact shows is not that there is a cause and effect relationship between abortion and suicide, but rather that women who are more likely to have an abortion are also more likely to be killed by accident or violence, including suicide, than those who do not have one. To take it a step further – women who in their day to day life are more likely to suffer a fatal injury, become suicidal or be murdered are also more likely to have an unwanted pregnancy, and thus choose to have an abortion. Rather than being a causative factor, an abortion is much more likely to be a symptom of whatever underlying cause is also making these women more likely to be murdered or suffer a fatal accident. While this is in some sense shocking, it is not exactly surprising. Such data was sadly not gathered in this study, but I would not be terribly surprised to find that those women who choose to have an abortion may also be more likely to come from abusive homes or be economically disadvantaged, both factors associated with a higher incidence of violence, mental illness and abortion. While this data does not show that there is no causative link between abortion and suicide, neither does it demonstrate that there is; much like a tasty but ultimately irrelevant sponge cake, it informs us very little in this matter. Rather than being an argument against abortion, this data is instead an argument for greater post abortion support services, and perhaps follow up investigations to look at the social and economic situations of women who choose to have abortions and determine if some form of assistance is appropriate. In summary, while I do not doubt that the notion of a causative link between abortion and suicide will continue to be bandied about, if you feel the urge to perpetuate such a hypothesis please note that the above studies support for your claim is dubious at best.
P.S. In this post I have focused mainly on debunking a specific claim, which in fairness to the authors of the paper they did not themselves make without qualification: the conclusion surmized that “The increased risk of suicide after an induced abortion indicates either common risk factors for both or harmful effects of induced abortion on mental health.” As I’m sure is clear, I tend toward the former suggestion, as the evidence does not goes as far as supporting the latter. In general, the literature has shown either no or little difference in mental health outcomes for unwanted pregnancies which end in abortion or are carried to term. Regardless, the practice of shaming such women should be roundly condemned, as studies have shown a causative relationship between social exclusion and suicide risk. For those interested in more recent and broader ranging studies that specifically address the relationship between abortion and mental health, I would direct you towards a 2011 systematic study by the National Collaborating Centre for Mental Health as a good starting point.