Priscilla Coleman’s recent meta-analysis showing a link between abortion and mental health problems not surprisingly has created a storm, coming as it did in the middle of the recent parliamentary debate over independent abortion counseling.
Whilst the left wing press and the BBC chose to ignore it, other academics in the field have attempted to undermine it in a flurry of letters to the British Journal of Psychiatry.
His letter is available on the BJ Psych website but here are some key excerpts. Fergusson writes:
The letter by Louise Howard and colleagues follows a well-trodden strategy which has been used in a number of reviews to dismiss any evidence suggesting that abortion may have adverse effects on mental health. In this strategy:
a) Methodological criteria are constructed to classify studies
b) No numerical analysis combining study findings is provided
c) Strong conclusions are drawn on the basis of a small number of ‘high quality’ studies
The use of this strategy led the APA review to claim that there was no evidence of adverse effects of single first trimester abortion on the basis of a single study, the conclusions drawn by Charles et al were based on four studies. Howard et al use a similar strategy in which they question the methodological quality of studies reviewed in the Coleman analysis and draw conclusions about the absence of association between abortion and mental health on the basis of a small number of studies. If the evidence is indeed as weak and as limited as these reviewers have claimed the appropriate conclusion to be drawn is that no firm conclusions can be made about the mental health consequences of abortion until further and better research is completed.
In the meantime, Coleman has done the field a service by bringing together a quantitative review of a large number of studies that have addressed the issue of abortion and mental health. What will be clear to any alert analyst is that there is a clear statistical footprint suggesting elevated risks of mental health problems amongst women having abortions, irrespective of the comparison group used or the outcome studied. This evidence clearly challenges the conclusions drawn from the reviews cited above.
Following the Coleman meta-analysis we have conducted a more refined analysis which addresses some of the concerns raised by Howard and colleagues. In this we confined the analysis to only those studies using an unwanted/unintended comparison and data was extracted from the original sources by two reviewers. This search yielded a total of eight studies reporting 14 results. These findings were used to conduct a series of meta-analyses for five separate outcomes: anxiety disorder; mood disorder; alcohol misuse; illicit drug use/misuse and suicidal behaviours. The results tell a clear story in which, with the exception of depression, there are significant small to moderate associations between abortion and increased risks of mental health problems.
The implications of this analysis are inescapable: despite the claims made in previous reviews about the absence of association between abortion and mental health, when data are pooled across studies there is consistent evidence suggesting that women having abortions are at modestly increased risks of mental health problems when compared with women coming to term with unplanned/unwanted pregnancies.
Fergusson is modest in his conclusions but his intervention is significant.
He promises a further review to be published soon.