RSS Subscription Logo

The anscombe bio-ethic centre response to the rcog draft guidance on the case of women requesting induced abortion

 David Albert Jones

Main points:


The Anscombe Bioethics Centre  

The Anscombe Bioethics Centre is the oldest national bioethics centre in the United Kingdom, established in 1977 by the Roman Catholic Archbishops of England and Wales. It was originally known as The Linacre Centre for Healthcare Ethics and was situated in London before moving to Oxford. The Centre engages with the moral questions arising in clinical practice and biomedical research. It brings to bear on those questions principles of natural law, virtue ethics, and the teaching of the Catholic Church, and seeks to develop the implications of that teaching for emerging fields of practice. The Centre engages in scholarly dialogue with academics and practitioners of other traditions. It contributes to public policy debates as well as to debates and consultations within the Church.

Due to the brief time that is provided for this consultation the Centre's response will cover just two general points (on process and ethical principles) and then take for consideration just two of the recommendations in the guidance (on recommendations 7 and 64).  


Process of development and consultation  

In July 2008 the Department for Business, Enterprise and Regulatory Reform issued the third version of the Government's Code of Practice on Consultation. This had itself been the result of a consultation and the aim had been to 'improve the transparency, responsiveness and accessibility or consultations, and help in reducing the burden of engaging in Government policy development.' (BERR 2008:3).  

The Code sets out seven criteria for open and effective consultation (BERR 2008:4):  

Criterion 1. When to consult: Formal consultation should take place at a stage when there is scope to influence the policy outcome.  

Criterion 2. Duration of consultation exercises: Consultations should normally last for at least 12 weeks with consideration given to longer timescales where feasible and sensible.  

Criterion 3. Clarity of scope and impact: Consultation documents should be clear about the consultation process, what is being proposed, the scope to influence and the expected costs and benefits of the proposals.

Criterion 4. Accessibility of consultation exercises: Consultation exercises should be designed to be accessible to, and clearly targeted at, those people the exercise is intended to reach.

Criterion 5. The burden of consultation: Keeping the burden of consultation to a minimum is essential if consultations are to be effective and if consultees’ buy-in to the process is to be obtained.

Criterion 6. Responsiveness of consultation exercises: Consultation responses should be analysed carefully and clear feedback should be provided to participants following the consultation.

Criterion 7. Capacity to consult: Officials running consultations should seek guidance in how to run an effective consultation exercise and share what they have learned from the experience.

These criteria should be reproduced in all government consultation documents. This code is only binding on government departments, not arms-length bodies, professional organisations or other independent agencies. Nevertheless, the criteria provide a touchstone for best practice in consultation against which organisations would do well to measure themselves.

Effective consultation is not only a matter of open (and hence more ethical) policy making but is also a means to test policy against criticism. Opaque processes which are difficult for the professional and academic community to engage with, let alone the public, and perfunctory consultations at a stage where there is little scope to affect the outcome, insulate the exercise from critical comment and hence weaken the quality of the final output. If evidence is not subject to rigorous independent critical review then its value remains untested and hence its conclusions are the less reliable.

The virtue of openness in public life, as enshrined in the Nolan Principles, also explains the role of lay participation in governance of public bodies and the regulation of medicine. The best exemplar of this practice is the General Medical Council but it has relevance to all medical bodies especially where their guidelines are likely to influence public policy. Lay participation may bring valuable experience of service users into the discussion. It also provides an important check and balance against the interests, insularity and defensiveness of professional organisations. It is not only government that requires openness.

Francoise Baylis in an illuminating article has demonstrated how this consultation process can be subverted where it is undertaken with a view to obtaining a particular result. Her paper is an indictment of a particular consultation of the Human Fertilisation and Embryology Authority and its conclusion is worth quoting in full.

It follows that although HFEA’s rule-guided and strategic modes of public consultation on the ethical and social implications of creating human/ animal embryos in research may be legitimate in a strict sense, they fall far short of embracing the democratic ideal of input-oriented legitimacy that “depends on the extensiveness and intensiveness of public participation in the making of policy”. (Baylis 2009: 59 quoting Montpetit 2003:97)

A similar criticism could fairly be levelled against the Royal College of Obstetricians and Gynaecologists (RCOG) in the development of its guidance on The Care of Women requesting Induced Abortion. Rather than being open to evidence and criticism from a wide range of participants, the level of consultation has been narrow and perfunctory.

In the first place the committee clearly excluded any representation from those who have fundamental objections to abortion. It seems to have been selected to represent only those with a prior commitment to delivering and indeed extending the provision of abortion. The committee included representatives from Marie Stopes International and the British Pregnancy Advisory Service. It did not include any members of the RCOG who were not involved with abortion provision.

It is the constitution of the working group which is the first step in seeking balance and credibility, but the committee does not reflect the debate that exists within society, and indeed within the medical profession, but only represents one side of the argument. Neither can it be said that the guidance confines itself to practical matters of delivering abortion and so should be confined to abortionists. On the contrary, the guidance seeks to establish an evidence base on consequences of abortion and best practice on aftercare; it also delineates the duties of those who conscientiously object and discusses many other matters which are of interest to professionals, academics, and members of the public who are not themselves abortionists nor committed to abortion provision. As the guidance explicitly considers ‘legal and ethical aspects’ its membership might reasonably include academics from legal, philosophical and theological fields, including critics of the present practice of abortion in England, Wales and Scotland. It would have been possible to constitute a committee with a larger lay contingent and a wider range of views more representative of society as a whole, and less committed to current practice, but the RCOG has not sought to do this.

Furthermore, in the initial phase of development of the document, ‘members of the Group suggested names of individuals or organisations’ to be consulted. This process reinforces the flaws that seem to be implicit in the initial constitution of the committee. Thus an opaque internal process continued until a draft was produced which only at this point was put on the website of the RCOG for members and others to comment on. Even when the draft guidance finally appeared on the website, there seems to have been little attempt to alert members of the College, or other medical professionals and academics to the consultation. Not least among the flaws in consultation was a failure to make the public aware of the possibility of engagement in this process. Having been posted on 21 January 2011 without a press release or any pro-active attempt to advertise its existence, other than its appearance in the relevant page of the website, the consultation end was initially given as 18 February 2011. This would be less than 5 weeks, in comparison with the 12 weeks that the government lays down for its own consultations as the very minimum that can be considered open and effective.

In the circumstances, it is not surprising that the Anscombe Bioethics Centre, like many other interested parties, only came across the consultation when there were but two days remaining:  the Centre was simply unable to respond within this timeframe. It was only following a question in parliament (Hansard 2011) that the consultation was extended, and then only by a period of one week! This clearly limits the quality of criticism that can be brought to bear on a report of 116 pages with 417 references.

The methodology of the RCOG guidance was to begin with a systematic review of MEDLINE for 2003 to 2010. It does not state which search terms were used but, for example ‘induced abortion’ and related terms for a search conducted on 21 January 2011 for that time period gave 3659 results. This is in addition to many other searches conducted including following the reference trail through journals, as well as to the material still relevant from 1996-2003. The guidance admits that the collation of this quantity of evidence requires ‘sifting’, a process that is qualitative rather than quantitative in character. However it is precisely at this point that it is susceptible to conscious or unconscious bias in selection, in relative weighting, and in presentation. For a balanced evaluation of the evidence a broad range of perspective is required early in the process at the stage when evidence is selected and considered. Without a balanced committee, an open consultation process from early on in the exercise, and sufficient time for high quality feedback, it is difficult for those outside the process to have confidence in the evidence presented.

While this RCOG guidance professes to be ‘evidence-based’, the process by which it has been developed and the meagre time given for peer review severely limits the range of criticism to which it has been exposed and thus the reliability of the final report.


The ethics of abortion

Before considering two of the specific recommendations, it is useful to examine the role of ethics and ethical principles in the guidance. Clinical guidance on the care of patients (as this document seeks to be) is an expression of ethical commitment to patients and thus requires some reflection on the ethics of practice. It is thus appropriate that clinical guidance should have a chapter devoted to ‘Legal and ethical aspects’ (chapter 3) and that the guidance should make a number of references to ethics or ethical issues (fourteen in total). This is a further reason why the committee should have included a range of perspectives, as ethics is not a mere matter of technical knowledge or of expertise confined to practitioners.

Despite the title of chapter 3, the content is in fact dominated by discussion of law and of Codes of Practice with relatively little discussion of the ethical issues as such. The conception of ethics here seems to be rule-based rather than grounded in an understanding of the principles, virtues and goods of human life and the requirements of justice. Most extraordinary of all, there is no virtually no discussion of the issue that makes abortion controversial and which alone explains the diverse laws, prohibitions and restrictions on abortion provision: the status of the unborn child.

In general in obstetrics and gynaecology, where a child is wanted, it is accepted that the doctor has a duty of care both for the mother’s health and for the health of her unborn child. In this context the language of ‘child’ or ‘baby’ is the norm and the use of ‘fetus’ or ‘fetal’ is exceptional, at least in speaking with the expectant mother. The clinical use of ultra-sound has also helped women bond with their babies prior to birth and this, together with other imaging technologies, has allowed the unborn infant to be made visible.

Abortion is ethically problematic and is regarded by many (including but not only Catholics and other religious believers) as unjust because it destroys the unborn human infant. That this is the key issue in relation to the ethics of abortion is known by any GCSE level school child. It is not an obscure point. The issue continues to engage the philosophy and theology journals and to be the subject of extended discussion in scholarly monographs (for example Jones (2004), Coope (2006)). It is utterly inexplicable that it should be completely omitted from a serious discussion of the ethics of abortion.

The duty of care or protection due to the unborn infant is not the only ethical issue in relation to abortion. There are also questions of confidentiality, consent, child protection and so forth. Nevertheless, no account of the ‘evidence’ in relation to the ethics of abortion can ignore the evident significance of the fact that abortion ends the life of the unborn child. This in turn affects our understanding of many other issues. The RCOG guidance acknowledges that the prohibition on abortion in Ireland does not breach the human rights of women, ‘owing to the margin of appreciation allowed to the state in balancing the protection of the life of the unborn with the rights of the mother’ (here referring to ABC v Ireland). It is only in the case where the mother’s life is at stake that Ireland was held to be at fault. In relation to England, Wales, and Scotland the draft guidance states that ‘most abortions are undertaken on ground C: that the pregnancy has not exceeded its 24th week and that continuance would involve risk, greater than if the pregnancy were terminated, of injury to the physical or mental health of the woman’. However there is no explanation as to why such interventions should be limited to the ‘24th week’ or discussion of the moral significance which parliamentarians and many of the public attach (whether rightly or wrongly) to the viability of the fetus.

The guidance thus begs the central ethical issue in the practice of abortion by failing even to acknowledge the duties that might arise as a result of the presence of another human life. There are a number of philosophers who have sought to justify abortion, but they have at least put forward explicit arguments which can then be examined and criticised, and such arguments typically do not attempt to justify all abortions but are more limited in scope. The argument over the status of the unborn has implications not only for the practice of abortion per se but also for issues such as conscientious objection or feticide, which are discussed by the guidance. If the guidance does not acknowledge the key issue in public, political, and academic debate over the ethics of abortion then it cannot evaluate the invocation of conscientious objection in this context, nor can it evaluate the ethics of deliberate feticide.

The remainder of this article will consider two of the RCOG recommendations that bear directly on ethics.


Recommendation 7. Professionals who are ethically opposed to abortion have a duty of care to refer onward in a timely manner women requesting abortion.

It should immediately be noticed that this recommendation involves ethical questions which are not settled by empirical observation but which require reflection on the nature of ethical principles, human rights and the common good.

The guidance alleges that its recommendation is ‘grade C’ which describes as one, ‘that cannot easily be answered by experiment but nevertheless represent good practice’. The ‘evidence’ supporting recommendation 7 is a single reference to the General Medical Council guidance according to which physicians ‘must treat [your] patients with respect, whatever their life choices and beliefs’ (GMC 2008:1). However the quoted principle is clearly insufficient to ground a duty to refer patients to another practitioner for the purpose of procuring abortion.

A doctor could respect a patient fully but not think that abortion would be beneficial (and hence medically indicated) in her particular case, or, some may think, in any case. The fact that a practice is legal and is funded by the NHS is not sufficient to require a doctor to refer where he or she does not think this treatment would be beneficial. This is made clear in the GMC guidance on consent:

If the patient asks for a treatment that the doctor considers would not be of overall benefit to them, the doctor should discuss the issues with the patient and explore the reasons for their request. If, after discussion, the doctor still considers that the treatment would not be of overall benefit to the patient, they do not have to provide the treatment. But they should explain their reasons to the patient, and explain any other options that are available, including the option to seek a second opinion. (GMC 1998: 5(d))

This duty does not extend to finding a doctor who is known to take a different view e.g. of homeopathy or male circumcision for non-medical reasons. It is enough that the doctor informs the patient of his or her right to a second opinion. The guidance on Personal Beliefs and medical practice adds to this the further requirement that the physician must ‘ensure that [the patient] has sufficient information to exercise that right.’ (GMC 2008:21) Nevertheless, this further requirement is still at the level of providing information (or information about where to find information) and of ensuring that a doctor does not withdraw from caring for a patient when that patient has ‘nowhere to turn’ (GMC 2008:22). The GMC does not require doctors to perform abortions other than in the case of danger to the mother’s health, nor does the GMC explicitly require that a doctor refer a patient to another doctor for the specific purpose of abortion. This is an over-interpretation.

A robust understanding of human rights needs to be rooted in the natural law tradition and would not support some contemporary conventions or court judgements as to alleged human rights. Nevertheless, it is interesting to note that where the life of the mother is not in jeopardy, the European Court of Human Rights did not find Ireland at fault for banning such abortions. There is no internationally recognised human right to abortion in these circumstances. Therefore, requiring the doctor to act against his or her conscience in circumstances where the patient has no countervailing right to treatment seems, on the face of it, to infringe the doctor’s human rights as recognised by Article 9 of the convention (See Hammer 1999, Foster 2005, Hill 2010). This has yet to be tested in court but where there is a recognised right not to participate in abortion it is arguable that this may reasonably also cover direct referral for abortion.

Even Frank Chervenak, a keen supporter of legal abortion, acknowledges that ‘sometimes, patients request medical intervention that is not beneficence-based but is clinically safe and effective in achieving the patient’s goal. These procedures have only autonomy-based indications’ (Chervenak 2008:232.e2). In such cases the doctor has no duty to refer directly, though, according to Chervenak, he or she does have a duty to provide information about other practitioners (what he calls ‘indirect referral’). Chervenak’s solution may not resolve the tension between the alleged autonomy rights of the patient and those of the doctor, but it is noticeable that in drawing the line between ‘autonomy-based indications’ and danger to the mother’s life or health he is drawing it where the Abortion Act 1967 drew the right of conscience. There are good reasons to regard the right of conscientious objection as applicable to referral at least in contexts where it would be legally applicable to the termination itself (see also Gerrard 2009).  Furthermore, whether or not the Abortion Act defines this particular right of conscience, it is a right that can be derived from the ethics of good practice and from human rights considerations. Employment law on discrimination is also of relevance here.

TThe RCOG draft guidance takes no cognisance of the debate in peer review journals on the ethics and law of conscientious objection. These legal and ethical arguments are not even discussed. The ‘evidence’ adduced is the over-interpretation of a single quotation from one document of the GMC. In fact there is no good evidence that law, much less ethics, requires a doctor to act against his or her conscience and refer a woman to a colleague or an organisation that will provide the abortion in cases where it the abortion has no clinical justification other than the request of the woman.

The same paucity of ethical discussion is found in another of the RCOGs recommendations, in relation to the deliberate killing of the viable unborn infant.


Recommendation 64. Feticide should be performed before medical abortion after 21 weeks and 6 days gestation to ensure that there is no risk of a live birth.

The ‘should’ in the RCOG recommendation is presumably the ‘should’ of moral or ethical obligation. The guidance asserts that ‘inducing fetal death before medical abortion may have beneficial emotional, ethical and legal consequences’ (line 2621 emphasis added). However, as evidence for this the guide cites only one article (Diedrich 2010) which provides a very cursory treatment of ethical and legal issues, referring only to the dilemma of which hospital to go to for an unplanned delivery and the ethical dilemma of administering hopeless resuscitative measures or withholding care in the case of live birth. The bulk of that article addresses the evidence that feticide improves the health outcome of the woman and concludes that ‘the only RCT of their efficacy in improving D&E… did not show improvement in markers of abortion safety’ (Diedrich 2010: 470).  While most women (92%) expressed a preference for the feticide prior to late termination, Dietrich comments that, the question ‘was posed to patients within the context of a clinical trial in which many of them believed the injection might make their abortion safer. Also, the social acceptability of a positive response may have skewed the results’ (Diedrich 2010: 464). 

The evidence presented as supporting recommendation 64 also includes a citation of the RCOG report on fetal pain which concluded that this did not occur in infants younger than 24 weeks gestation. However, while the issue of fetal pain is marshalled on both sides of the abortion dispute it is not clear what relevance it has to the issue of feticide, the aim of which is to kill the fetus prior to delivery in cases where there is a ‘danger’ that it might survive. The question is not whether the fetus is aware of what is happening to it, but what the intention of the procedure is, and how this should be evaluated. /p>

In termination of pregnancy one aim of the procedure is to expel the fetus (or embryo) from the womb. This may generally be done with the intention of getting rid of the infant, but there are at least some circumstances where death might not be the aim (which is not, however, to say the procedure will necessarily be justified). In contrast the whole aim of the (extra) feticide procedure is to kill the infant as ‘failure to perform feticide could result in a live birth and survival which contradicts the intention of the abortion’. The baseline for the ethical analysis of abortion is that abortion is homicide and that any procedure which effectively brings about the death of the unborn child can only be justified in certain cases where this is the indirect effect of an intervention needed to save the mother’s life or to avert a serious threat to her health.  (An example might be a case where a damaged fallopian tube is removed with an unborn child inside.) Whatever we say about these very different cases feticide, where the specific aim is to kill the unborn child, can never be justified.

It is noteworthy that even among those who are broadly in favour of abortion prior to viability there are a number who would regard feticide as unethical:  ‘It follows that it is ethically impermissible to offer feticide for viable foetuses without anomalies or with less-than-severe anomalies’ (Chervenak 2009: 560.e3). Pullman goes further and suggests that whether the fetus is healthy or disabled, once it is viable then it has a prime facie claim to protection: ‘Thus even those who accept a liberal position with regard to therapeutic abortion, should be concerned about these more recent developments’ (Pullman 2010:353;  see also Gross 2002). Indeed, the current gestation time limits on legal abortion in English law are clearly related to viability, as evident from discussion in parliament and from the language of the Infant Life Preservation Act 1929, which remains in force for actions that fall outside the Abortion Act 1967. However, if it is the intention of the law to protect the viable fetus, and if this expresses the duty of care that the state owes towards the viable fetus, then how can it be ethical deliberately to destroy a viable fetus?

The aim of feticide is nothing other than to kill the unborn child where there is a ‘danger’ that it might be born alive. If the fetus is not viable (in the sense that it might be born alive) then feticide is pointless, but if the fetus is viable the action contradicts the meaning of the law and ethical considerations that are widely recognised, even among those who are not consistently ‘pro-life’. Indeed, if it is the will of parliament to prohibit elective abortion when the fetus is both healthy and viable, then rather than stipulate arbitrary time limits (which are subject to heated and partisan debate on the significance of the evidence), it would seem more rational and honest to declare feticide as a procedure unethical, which will inhibit doctors from inducing abortion in circumstances where live birth is actually a possibility. Rather than having ‘beneficial ethical consequences’, feticide involves the unethical intention to end the life of the viable unborn infant.

FFor reasons of space I have confined myself to just two recommendations. However, it is clear from these two that the committee has neither considered the evidence nor benefitted from critical arguments. There is no discussion of evidence that referral for abortion is covered by the legal right to conscientious objection, nor that feticide is regarded as unethical by a very wide spectrum of moral opinion, and seems to contradict the very rationale of the present law. The examination of these two recommendations has shown a failure even to engage with contrary evidence or argument. This confirms the initial suspicion that the flaws in process have led to systematic bias throughout the guidance and to a careful ‘sifting’ of the evidence. The result is a document which cannot provide policy makers and professionals with credible and reliable guidance. While some recommendations may accurately reflect the evidence others reflect prior commitments, and there is no straightforward way for the reader to distinguish the ‘evidence based policy’ from the ‘policy based evidence’.

Dr David Albert Jones is Director of the Anscombe Bioethics Centre and Visiting Professor at St Mary’s University College.