Doing Dorries a Disservice

As might be expected, Tom Chivers’ post on Nadine Dorries’ ‘Right to Know’ campaign has drawn a response from Dorries, even if she is merely intemperately parroting the comments of Dr. Peter Saunders of the Christian Medical Fellowship:

Tom Chivers has done the Daily Telegraph a disservice which has up until this article, fairly and accurately reported both our aims and objectives. Misleading statements from journalists on an issue as important as this are frankly distasteful as they condone the suffering many women such as Tanya experience.  Dr Saunders states on his blog……

Tom’s error was to mistakenly cite a letter to the BJPsych as a systematic review of current research on abortion and mental health, when he should have linked to a paper by Charles et al (2008) [1] which was published in the journal ‘Constraception’, a point that he graciously conceded in his own response to those of Dorries and Saunders. That error and a few quibbles over his characterisation of the first of David Fergusson’s recent papers on abortion and mental health, his first article more than stands up to scrutiny despite Dorries’ nonsensical attempt to characterise Saunders’ response as a fisk.

It isn’t, as Tom correctly points out, not least because it fails miserably to address any of the substantive points made in his original post:

However, I would point out that a “fisking” is, according to Wikipedia (and they’re always right), a “point-by-point criticism“. I may be being dense, but I can’t quite see where any of my points – other than one admittedly stupid referencing error – is addressed at all.

From such thin gruel, Saunders manufactures the absurd allegation that Chivers has misrepresented the medical evidence in his original post when, in fact, he did nothing of the sort. His assessment of Fergusson’s paper may well have been a tad harsh in places but on the most important point raised in that article, Dorries’ highly selective and misleading use of data from the Fergusson study, Chivers got it absolutely spot-on:

First, as any GCSE Double Award Science pupil should be able to tell you, correlation does not equal causation. The authors of that article say, very clearly, that “Estimates of attributable risk indicated that exposure to abortion accounted for 1.5 per cent to 5.5 per cent of the overall rate of mental disorders”. “Atrributable risk” refers to the difference in rate of a condition between an exposed population (in this case, those who have had an abortion) and an unexposed population (those who have not). The remaining 24.5 to 28.5 per cent increase is attributable to other factors.

This is, very noticeably, the one point in Chivers’ original article that neither Saunders or Dorries even attempt to address. They’ve been caught blatantly cherry-picking the evidence and have responded in typically dishonest fashion by trying to blow smoke:

He correctly identifies the primary source (after help via friends on Twitter) as an article by Fergusson et al originally published in the British Journal of Psychiatry in 2008 but then goes on, ironically, to make misleading statements about the scientific evidence.

Chivers claims that the Fergusson study is ‘now rather out of date’ and that ‘The BJPsych itself has revisited the topic, with a 2009 systematic review’. Both these statements are simply untrue.

If you follow the link in Chivers’ blog you will see that it does not refer to a BJPsych review at all, but rather to a letter written to the journal by two non-psychiatrists with close links to the pro-choice movement, Sam Rowlands and Kate Guthrie.

Chivers has, as I’ve already noted, copped to his citation error, which amounts to a mistake and not an untruth and while it would be incorrect to say that Fergusson’s study is ‘out of date’ there are very particular reasons why it continues to be relevant in the face of seemingly contradictory evidence from other more recent reviews, none of which Saunders sees fit to mention – if he’s actually aware of those reasons at all.

The Royal College of Psychiatrists has itself acknowledged in a 2008 statement that there are in fact studies which show a link between abortion and mental health problems and is currently carrying out a review of the literature that has not yet been published.

The mere existence of studies that purport to show a link between abortion and mental health is, in itself, neither here nor there. What matters is the quality and reliability of the evidence within those studies, hence the need for RCOG to conduct a comprehensive literature review prior to formally revising their position and guidance on abortion.

The review that Rowlands and Guthrie refer to is not from the editors of the BJPsych at all (or the Royal College) but is actually one by the American Psychological Association which has been widely criticised as I have previously outlined on this blog.

Rowlands and Guthrie’s letter includes eight references, including the APA taskforce report that Saunders refers to as having been ‘widely criticised’ and the paper by Charles et al. that Chivers’ had intended to cite in his original article, which Saunders does not address at all.

Looking on the bright side, Saunders does clear up the mystery of exactly who it is that has been ‘widely’ criticising the APA taskforce report as, as predicted, it turns out to be Priscilla Coleman and her buddies who, apparently, sent the APA a petition letter (containing all of seven signatures) criticising the taskforce’s methods and conclusions.

Having blathered around the edges of Chivers’ article, Saunders continues by introducing his own preferred misrepresentations of the facts:

Why is all this so important? For two reasons.

First most abortions in Britain (about 98%) are carried out on mental health grounds when in fact there is no clear evidence in the literature to show that being denied abortion has mental health consequences. The jury is still out on this question as the RCPsych acknowledges. This makes the vast majority of abortions in Britain technically illegal.

If we leave aside the fact that its common knowledge that the provisions in law for abortion on mental health grounds are little more than a legislative fudge, the simple fact is that it is highly unlikely that RSPsych will even attempt to address the question of whether being denied abortion has mental health consequences precisely because there is no significant research literature on this particular issue for very obvious ethical reasons – one could only obtain such data by forcing women to continue their pregnancies to term against their express wishes.

This is the most serious limitation evident in David Fergusson’s studies, in which the baseline comparison group consists of women who ultimate chose to continue with the pregnancy, irrespective of their initial feelings about falling pregnant, and not women who were compelled to carry a baby to term against their express wishes. One cannot, therefore, rule out the possibilty that at least some women would experience a significant level of distress, equivalent to that reported by women who unwillingly went ahead with an abortion, and would consequently find themselves subject to same elevated risk of subsequent mental health problems as the sub-group identified in Fergusson’s second paper.

Of the three primary groups in the study, women who chose to have an abortion, women who chose to carry their pregnancy to term and women who miscarried, the first two are self-selecting groups and, as a result, freedom of choice and the circumstances under which the choice to continue with or terminate a pregnancy is made must be considered to be highly significant counfounding factors, factors that can only be accounted for in a very limited sense by taking into account women’s feelings about their pregnancy before the choice is made.

Although it is reasonable for Fergusson to question the validity of the current legal arrangements in the UK, under which mental health grounds are used as a cover for abortions undertaken for primarily social reasons, he nevertheless rather overstates the strength of his evidence in posing this question without giving full regard to the limitations imposed on his findings by the absence of a fully valid comparision group.

While we cannot, for obvious ethical reasons, make any strong evidence-based assertions as to the long-term mental health implication of forcing women to continue with a pregnancy against their express wishes, the direct consequences of such practices remain all to evident across much of the developing world [2]:

UNSAFE ABORTION

– The World Health Organization defines unsafe abortion as a procedure for terminating an unintended pregnancy carried out either by persons lacking the necessary skills or in an environment that does not conform to minimal medical standards, or both.

– Worldwide, 48% of all induced abortions are unsafe. However, in developed regions, nearly all abortions (92%) are safe, whereas in developing countries, more than half (55%) are unsafe.

– More than 95% of abortions in Africa and Latin America are performed under unsafe circumstances, as are about 60% of abortions in Asia (excluding Eastern Asia).

– The worldwide unsafe abortion rate was essentially unchanged between 1995 and 2003 (15 and 14 abortions per 1,000 women aged 15–44, respectively). Because the overall abortion rate declined during this period, the proportion of all abortions that are unsafe increased from 44% to 47%.

CONSEQUENCES OF UNSAFE ABORTION

– Worldwide, an estimated five million women are hospitalized each year for treatment of abortion-related complications, such as hemorrhage and sepsis.

– Complications due to unsafe abortion procedures account for an estimated 13% of maternal deaths worldwide, or 47,000 per year.

– Almost all abortion-related deaths occur in developing countries. The highest number occur in Africa.

Additional consequences of unsafe abortion include loss of productivity, economic burden on public health systems, stigma and long-term health problems, such as infertility.

Lack of access to adequate healthcare is, of course, an important factor in the global incidence of unsafe abortions, but across much of Africa and Latin America this goes hand-in-hand with heavily proscriptive abortion laws under which abortion is, if legal at all, permitted only in extremely limited circumstances.

The figures, I think, speak for themselves so let’s get back to Peter Saunders’ ‘fisk’:

Second, there is growing evidence that abortion actually poses a risk to mental health. The significance of Fergusson’s research is that it was a particularly robust randomized, prospective study which had carefully corrected for confounding variables. What makes it doubly interesting is that Fergusson himself comes from a pro-choice perspective so arguably had a vested interest in his research not showing any link between abortion and mental health problems.

As pickiness is the order of the day it seems reasonable to point out that Saunders is mistaken in identifying Fergusson’s research as a prospective study. It isn’t, it’s a moderately-sized longitudinal general health and development study of a birth cohort of 1265 children born in the urban Christchurch region of New Zealand, of which 630 were female. Overall, the Christchurch Health and Development Study, to give its full title, can be considered to be a prospective study only to the extent to which research papers generated from its data reflect upon issues addressed in the study’s original design. Spin-off studies, such as Fergusson’s papers on abortion and mental health, cannot be treated as fully prospective studies because they rely, in part, on data captured after fact rather than contemporaneously during the lifetime of the study, as Fergusson freely acknowledges the second of his two papers on abortion and mental health [3]:

The greatest limitation is that the assessment of abortion-related distress was based on retrospective reports obtained at the age of 30. Such reports may be subject to errors of reminiscence and possible recall bias. For this reason it is important that our findings are replicated using prospective data to assess the links between abortion distress and subsequent mental health.

Saunders is a medical doctor and seemingly well aware of Fergusson’s second paper as its referenced in his organisation’s submission to RCOG’s consultation on its draft guideline on abortion, and yet he seems to incapable of accurately reading Fergusson’s very clear and well-written research papers – that, or he is simply ignoring anything in them that doesn’t suit his preferred line of argument on abortion.

Indeed, as I write this, Saunders has responded on his own blog to Chivers’ response and has resorted to time-honoured politicians’ gambit of responding to the points he’d liked to have been put to him rather the points that Chivers actually raised:

May I also correct a continuing misinterpretation in your new post with respect to the 1.5-5.5% figure as I’m not sure you have yet understood this correctly.

It is easiest to understand if you quote the relevant sentences in Fergusson’s abstract in full.

They read as follows:

‘After adjustment for confounding, abortion was associated with a small increase in the risk of mental disorders; women who had had abortions had rates of mental disorder that were about 30% higher. There were no consistent associations between other pregnancy outcomes and mental health. Estimates of attributable risk indicated that exposure to abortion accounted for 1.5% to 5.5% of the overall rate of mental disorders.’

What I understand this to mean is:

1. Women who have abortions had a 30% higher incidence of mental health problems after correcting for all other possible confounding variables. Ie The risk is real

2. Other outcomes of pregnancy (eg miscarriage, stillbirth, live birth etc) don’t affect mental health in the same way

3. Abortion however only accounts for between 1.5% and 5.5% of all mental health problems in the general population.

Saunders wrongly suggests that Fergusson has corrected, in this paper, for ‘all possible confounding variables’ when, in fact, he hasn’t. In assessing the evidence for causality, Fergusson explicitly states that:

Although the weight of the evidence favours the view that abortion has a small causal effect on mental health problems, other explanations remain possible. In particular it could be suggested that the small association between abortion and mental health found in this study could be explained by uncontrolled residual confounding. As in all naturalistic studies, control of nonobserved sources of confounding is difficult but not impossible and there are several ways in which better control of such confounding might be achieved.

Moreover, this particular paper does not take into account ‘wantedness’, i.e. womens’ feeling about the pregnancy at the time they chose to have an abortion, an issue that was raised in academic criticism of this first paper and explicitly addressed in his second paper, nor does Fergusson control for the implications of a forced pregnancy, i.e. the denial of the option to terminate against the express wishes of a pregnant woman, and this remains a significant limitation despite the ethical limitations that make data on this issue almost impossible to come by.

Last, and by no means least, Saunders is focussing exclusively of Fergusson’s first paper to the exclusion of any reference to or consideration of the findings of his second paper[3]:

Combining the findings of our two studies leads to the following generalisations about the links between unwanted pregnancy, abortion and mental health in this birth cohort.

  1. First, unwanted pregnancy terminated by abortion was associated with modestly increased risks of common mental health problems for women who reported significant distress about the abortion (RR = 1.4–1.8).
  2. Second, unwanted pregnancy terminated by abortion was not associated with significantly increased risks of mental health problems for women who did not report significant distress about the abortion (RR = 1.14–1.24).
  3. Third, unwanted pregnancy that came to term was not associated with significant increases in mental health problems (RR = 1.05–1.11).
  4. Finally, any associations between unwanted pregnancy, abortion and mental health problems were small to moderate, with adjusted relative risks in the region of 1.1–1.8. Estimates of the population attributable risk suggested that exposure to unwanted pregnancy terminated by abortion accounted for fewer than 5% of the mental health problems experienced by women in this cohort.

As we have noted previously, these findings are not consistent with strong pro-life positions that depict unwanted pregnancy terminated by abortion as having devastating consequences for women’s mental health. Equally, however, the findings do not support strong pro-choice positions that claim unwanted pregnancy terminated by abortion is without mental health risks. Rather, the accumulated evidence suggests that unwanted pregnancy terminated by abortion is an event that leads to significant distress in some women, with this distress being associated with a modest increase in risk of common mental health problems.

As one cannot reasonably ascribe this omission to ignorance of the existence of the second Fergusson paper, it seems reasonable to think that Saunders is cherry-picking the evidence to support his preconceived, religiously-motivated, views on abortion – his organisation’s submission to the recent RCOG consulation openly criticises the draft guidelines for ‘not touching on the prevention of abortion‘.

Motive, here, seems fairly obvious.

Taken in isolation, the findings of Fergusson’s first study could be contrued as indicating the existence of a general risk to womens’ long-term mental health associated with induced abortion and, as such, the optimum policy response would incorporate a requirement that all requesting an abortion should be notified of the existence of, and evidence for, this particular risk, although not necessarily one specified in legislation as it would clearly be unethical for any doctor to proceed with a termination without first having given their patient this information.

However, as Fergusson’s second paper clearly indicates, we are not dealing here with a general risk, but one that is specific to a small and very particular sub-group of women in very specific circumstances. Consequently, intervention is only mandated if there is evidence, at the per-abortion counselling stage, that a particular women has presented with a relevant risk factor, at which point it become necessary both to inform the woman of the nature of the risk and make provisions for appropriate aftercare and post-abortion counselling. Women who do not exhibit any relevant risk factors can, however, simply be reassured that the majority of women do not experience any long-term risks to their mental health although they can expect to experience feelings similar to those associated with a bereavement, all of which is a perfectly normal part of the recovery process.

This last approach to pre-abortion counselling, which is fully supported by credible scientific evidence (jncluding Fergusson’s work) does not require legislation, least of all legislation of the kind proposed by Nadine Dorries, and can be readily implemented with nothing more than a few minor revisions to current good practice guidelines.

Saunders cannot but be aware of the fact that Fergusson’s second paper, which updates and directly modifies the findings of his first paper, do not support the arguments advanced by Dorries’ ‘Right to Know’ campaign, not that these hold much water anyway, which is why he finds it necessary t0 resort to the same kind of cherry-picking and quote-mining that has, sadly, become standard practice amongst anti-vaccination campaigners and propenents of so-called ‘alternative’ quack medicine (e.g. homeopathy).

It is entirely indicative of Saunders’ intellectually dishonest approach to the evidence that his own organisation’s submission to the recent RCOG’s consultation on care standards in the provision of abortion services includes, amongst its references a number of papers by Priscilla Coleman and/or David Reardon, the latter being the proud owner of a ‘PhD’ in Bioethics from what was, at the time he obtained his diploma, an unaccredited institution  – almost all the references used by the CMF as ‘evidence’ for a link between abortion and mental health problems are deal with is this earlier post of mine, including Reardon’s ‘academic’ credentials and his bizarre and spectacularly offensive musings on abortion (warning: this last link contains a major violation of Godwin’s Law).

This same submission also continues to peddle the discredited claim that abortion may be a risk factor for breast cancer with the claim that the ‘jury is still out’ on this issue – and relying entirely on only two papers, the most recent of which is now ten years old – when such claims have been rejected by all the major Ob-Gyn and Cancer organisations in the UK, US, etc. on the basis of robust scientific evidence from several large scale cohort studies.

A fair assessment of the quality of the evidence that Saunders and the CMF are relying on can be gleaned from this delightful demolition of paper by Patrick S Carroll, author of one of the two papers cited in the submission to RCOG. This is not the paper included in the submission – that paper has not been peer reviewed and was published only on the website of the anti-abortion organisation, ‘LIFE’ – but a more recent paper that was published in a journal (of sorts), the notorious JPANDS, which is so lacking in credibility that it isn’t listed on Medline. For anyone who fancies skipping the maths, Mark Chu Carroll’s provides the following tidy summation of his findings:

The model is wrong. Invalid models to not produce valid results. Stop. Do not pass go. Do not collect $200. Do not get your paper published in a decent journal. Do get laughed at by people who aren’t clueless jackasses.

As a final word on Saunders’ use of evidence, that’ll do for me very nicely…

… oh, and for future reference, that is how you do a fisk.

References

[1] Charles VE., Polis CB., Sridhara SK., & Blum RW., (2008),  Abortion and long-term mental health outcomes: a systematic review of the evidence, Contraception, 78(6):436–450. (abstract only)

[2] Facts on Induced Abortion Worldwide, (2011), World Health Organisation & Alan Guttmacher Institute.

[3] Fergusson, D. M., Horwood, L. J., & Boden, J. M. (2009). Reactions to abortion and subsequent mental health.  British Journal of Psychiatry, 195, 420–426.

[4] Fergusson, D. M., Horwood, L. J., & Boden, J. M. (2008). Abortion and mental health disorders: Evidence from a 30-year longitudinal study. British Journal of Psychiatry, 193, 444–451.