Inaccuracies, lack of transparency and overconfident conclusions in RC Psych review on abortion and mental health

I previously blogged on the Royal College of Psychiatrists systematic review of the relationship between induced abortion and mental health. The consultation into the review closed on 29 June and the revised review document with be published later this year.

The RCOG will not be able to finalise their own guidance on termination of pregnancy until the RC Psych review is finalised.

I have reproduced below the statement from the Christian Medical Fellowship about the review. CMF’s full submission is available on the CMF website along with a summary of its main points .

Overall CMF has welcomed the RC Psych review on abortion and mental health but has raised concerns about inaccuracies, lack of transparency and overconfident conclusions

CMF Statement

The Department of Health and the Royal College of Psychiatrists (RCPsych) are to be congratulated on funding and carrying out a major review of the impact of abortion on women’s mental health. They, like the Christian Medical Fellowship (CMF), clearly regard this topic as one of great importance for the health and wellbeing of the over 200,000 women who face abortion decisions each year.

The opportunity for the public and stakeholders to review and comment on the RCPsych’s draft consultation document is one we have welcomed. It stands in stark contrast to the Royal College of Obstetricians and Gynaecologists attempt to review its guidance on the care of women requesting induced abortion, which was carried out without proper consultation or thorough research in January this year.

Having consulted widely ourselves for comments on the draft review paper, we have already sent a detailed submission to the RCPsych, which is now available on our website, along with a paper summarising our main points.

While we do have a number of significant concerns with the draft paper, our feedback to the RCPsych on the review has by no means all been negative. We have welcomed the consultation and have acknowledged that this review of the issue is more robust than some previous, widely cited, reviews (e.g. Charles and APA).

We have particularly welcomed the working party’s recommendation that support and monitoring should be provided for all women with an unwanted pregnancy who face the risk of later mental health problems, and for NICE guidance to underpin this support.

One of our main concerns with the RCPysch draft paper is that several of the summary evidence statements are stronger and more definitive than the actual research cited in their review supports. Whilst this might provide strong headlines and summary statements it is at the expense of accuracy and transparency.

On the basis of the available data the conclusions should be more nuanced and tentative.

In particular, the review is wrong to claim that there is no evidence of an elevated risk of mental health problems post-abortion compared to post-pregnancy. This definitive statement is simply not justified by all the data presented in their review and reflects reporting bias. Indeed, the four main research studies selected for this section of the review each found some mental health problems were more common following one or two abortions. It would have been more accurate simply to state, at the very least, that there is conflicting evidence about the relationship between abortion and mental health.

We were also concerned about the lack of stated rationale behind the reviewers’ re-analysis of research data in an important paper by Fergusson. This is important because this reanalysis produced a finding directly contradicting the conclusions of his published paper, which found higher mental health problems for those exposed to abortion.

We have also queried the lack of full transparency over decisions involving data selection and grading, which made it impossible for stakeholders objectively to verify the selection decisions made.

We did moreover note that the review had two different wordings for one of its three main questions. At the very least this reflects some sloppiness in writing but of more concern, the two questions posed could well produce different answers. The answer to the question: ‘Are mental health problems more common in women who have an induced abortion, when compared with women who delivered a live birth?’ (p65) would be ‘yes’. The answer to the alternatively worded question: ‘Are mental health problems more common in women who have an induced abortion, when compared with women who deliver an unwanted pregnancy?’ (p18, p87) depends on the concept of ‘wantedness’ of the pregnancy, which we argue is subjective, difficult to measure and changeable and therefore should be used with more caution than the review uses it.

Interestingly, research in the review shows that, even after controlling for confounding variables, there are higher rates of mental health problems post-abortion, for women with no prior history of mental health problems, compared to women in the general population. Our concern is that while this is stated in the main report, it is not included in the summary evidence statements. It is clearly a finding that women need to be made aware of in order to make a fully informed decision.

Undoubtedly this review will strongly influence any future Government policy and advice on the health consequences and care of women requesting abortion. Therefore it is essential that the final version of the review fairly and accurately reflects the research findings and that the process of study selection and data reanalysis is fully transparent.

This draft is a solid foundation to build upon, but it now requires greater clarification in places along with more caveats and information within the summary statements.

Overall however, if this review does indeed lead to more support and monitoring being offered to all women with an unplanned pregnancy, especially if underpinned by official guidance, it will have had some real long-term value. And that we would certainly welcome.

CMF Public Policy Department
July 2011