Article Text

Download PDFPDF
Senior doctors' opinions of rational suicide
  1. Stephen Ginn1,
  2. Annabel Price2,
  3. Lauren Rayner2,
  4. Gareth S Owen2,
  5. Richard D Hayes2,
  6. Matthew Hotopf2,
  7. William Lee2
  1. 1BMJ, London, UK
  2. 2King's College London (Institute of Psychiatry), Department of Psychological Medicine, London, UK
  1. Correspondence to Dr Stephen Ginn, Editorial Registrar, BMJ, BMA House, Tavistock Square, WC1H 9JP, UK; mail{at}


Context The attitudes of medical professionals towards physician assisted dying have been widely discussed. Less explored is the level of agreement among physicians on the possibility of ‘rational suicide’—a considered suicide act made by a sound mind and a precondition of assisted dying legislation.

Objective To assess attitudes towards rational suicide in a representative sample of senior doctors in England and Wales.

Methods A postal survey was conducted of 1000 consultants and general practitioners randomly selected from a commercially available database. The main outcome of interest was level of agreement with a statement about rational suicide.

Results The corrected participation rate was 50%; 363 questionnaires were analysed. Overall 72% of doctors agreed with the possibility of rational suicide, 17% disagreed, and 11% were neutral. Doctors who identified themselves as being more religious were more likely to disagree. Some doctors who disagreed with legalisation of physician assisted suicide nevertheless agreed with the concept of rational suicide.

Conclusions Most senior doctors in England and Wales feel that rational suicide is possible. There was no association with specialty. Strong religious belief was associated with disagreement, although levels of agreement were still high in people reporting the strongest religious belief. Most doctors who were opposed to physician assisted suicide believed that rational suicide was possible, suggesting that some medical opposition is best explained by other factors such as concerns of assessment and protection of vulnerable patients.

  • Psychiatry

Statistics from


The WHO has estimated that approximately a million people die by suicide worldwide every year.1 Many studies indicate that most people who commit suicide have a disturbance of mental functioning.2 This is consistent with there being some people who wish for death, but do not have any impairment of mind. For these people, public health approaches intended to prevent suicide, largely concerned with the elimination of the means by which suicide might be carried out, might be considered infringements on individual liberty. Indeed, the current illegality of UK doctors prescribing medications intended to bring about death represents an unacceptable limit on personal choice for some groups.3 4

Societal trends towards secularism have accompanied shifts in the judgement of suicide from being a moral or religious failure in need of prohibition by the state towards a situation where most suicides have come to be seen as the result of disturbance of mind viewed in a morally neutral manner.2 This was exemplified in the UK by the decriminalisation of the act of suicide in 1961.5 Consistent with these wider trends, contemporary developments in palliative care encourage and afford more opportunity for terminally ill people to contemplate the place and manner of their own deaths.6

Philosophical and theological debates about the morality of suicide are probably as old as human history. Among the Western classics, for example, Plato and the Stoics took contrasting positions on the permissibility of suicide, with Plato arguing that it disgraced the human and the Stoics arguing for its permissibility. Among world religions, Christianity, Judaism and Islam have tended to strongly prohibit suicide, whereas Eastern religions such as Buddhism and Jainism have not. Contrasting accounts of suicide may be found in the thoughts of St Thomas Aquinas and David Hume.7 Aquinas argued that suicide was contrary to natural law and to charity. Part of his argument was that, because every person belongs to a community, self-killing injures the community. David Hume doubted Aquinas' concepts of natural law and charity and thought that, in some circumstances, suicide satisfies both the individual's primary needs and the larger public interest.

Werth and Holdwick have suggested circumscribed criteria under which suicides should not be prevented.8 The proposals are that, for a suicide to be considered rational (and therefore worthy of not being prevented), the person in question must have an unremittingly hopeless medical condition, should make their decision as a free choice, have engaged in a sound decision-making process, and there should have been an assessment by a mental health professional. Proposed criteria for the circumstances under which assisted suicide should be allowable are similar and are also of relevance. Beauchamp7 proposes that assisted suicide could be appropriate with capable patients, where there is mutual decision-making by patient and physician. The decision should be made in a reflective and supportive environment and there should be an ongoing patient–physician relationship. A considered rejection of alternatives is necessary. These criteria are similar to those in unsuccessful attempts to change the law in England and Wales to allow physician assisted suicide (PSA).9

Research examining attitudes concerning end of life decisions has focused on those towards assisted dying, and the pragmatic issue of whether it is appropriate for the law to allow physicians to prescribe medication for this purpose. With this in mind, an examination of the possibility of rational suicide is of worth, as rational suicide is a logical prerequisite for assisted dying to be morally defensible. That is, if a doctor finds it ethical to bring about the death of a patient at the explicit request of that patient, he or she is showing their belief in the possibility of rational suicide.

Many international surveys have found general populations to be in broad agreement with the legalisation of assisted dying in one form or another (reviewed by a 2005 parliamentary report on the subject10). All current and proposed legislation for assisted dying around the world involves medical practitioners. However, doctors are divided, with a majority opposing such legalisation.11 12 Campaigners have identified medical opposition as a key obstacle to legalisation of assisted dying.13

There has been little exploration into the views of physicians concerning rational suicide. In a survey of members of the Royal College of Psychiatrists, Shah et al14 found that 86% of respondents agreed that suicide could be rational, while 7% were neutral and 7% disagreed, a finding they found curious when contrasted against the previous finding that about 90% of people who commit suicide have a mental illness.2 Surveys of attitudes to rational suicide have been undertaken across a range of non-medical health professionals. Werth and Holdwick reviewed the literature in 2000 and found 70–80% acceptability for both rational suicide and physician assisted death. They also found that 20–40% of mental health professionals have had experience working with patients whom they considered to hold suicidal plans that they considered rational.8

Given that belief in rational suicide is a prerequisite for supporting assisted dying legislation, and that decisions on the subject of assisted dying, were it to become legal, are likely to involve senior doctors, we aimed to survey senior doctors practising in England and Wales about their attitudes towards rational suicide and to identify any relationships between these attitudes and other factors such as religious belief and medical specialty.


We sent questionnaires to 1000 senior doctors in England and Wales, who were randomly sampled from the Informa Healthcare Medical Directory 2005/2006, a commercially available directory of medical practitioners. For our purposes, a ‘senior doctor’ was defined as someone who is currently practising as a general practitioner (GP) or a doctor on the specialist register (consultant in the UK) working in any specialty. Retired doctors were excluded, as were directory entries with inadequate postal address information. Other results from this study regarding PSA have already been published elsewhere.11

When completing the survey, we asked those receiving the questionnaire to provide details of their specialty, how long they had been a GP or consultant, their gender, and how much their day to day work involved the management of dying people. We also asked them to record how religious they considered themselves to be. Questionnaires were first posted in February 2007, and non-responders received the questionnaire again 12 weeks later. After a further 6 weeks, we telephoned non-responders, resending questionnaires as required. If it became clear that a potential subject was unable to participate in the survey for reasons of moving, retirement or death, the denominator for the participation rate was adjusted to take account of this. Each questionnaire was given a unique identifying number so that those who responded were not sent another questionnaire. We removed all identifying information before the analysis.

The main outcome of interest was level of agreement with the statement: ‘suicidality cannot be rational under any circumstances’. This was ascertained using a five-point Likert-type scale, which was then converted into a three-point scale consisting of ‘agree’, ‘neither agree nor disagree’ and ‘disagree’ with the possibility of rational suicide.

We gained permission for the study from the Joint Maudsley Hospital and Institute of Psychiatry Research Ethics Committee (Ref 06/Q0706/36).


A response rate of 50% (372/735) was achieved once we had accounted for exclusions (265). We found no differences between responders and non-responders for specialty11 (not shown).

Overall, 72% of the sample agreed that suicide can be rational (table 1), 17% disagreed, and 11% neither agreed nor disagreed. There were no statistically significant differences in the view of rational suicide by specialty. Level of agreement with rational suicide was negatively related to the intensity of religious belief. Responders who reported themselves to be minimally religious were the most in agreement with rational suicide (84%). Responders who considered themselves to be most religious were in least agreement (56%). There was no relationship between the frequency of caring for dying patients and view of the possibility of rational suicide.

Table 1

Degree of support for the possibility of rational suicide by various factors

Doctors who approve of legislation to legalise PAS were more in agreement with the possibility of rational suicide (86%) than those who disagreed with PAS (66%).

We carried out multivariable logistic regression to assess independent predictors of agreement with rational suicide. The two other conditions of disagreeing with the possibility of rational suicide and having no view were combined into a single comparison group. The only independent associations were that doctors who were more religious were more opposed to the possibility of rational suicide, as were doctors who disagreed with legalisation of physician assisted dying (table 2). The tests for trend across the groups were Z=−2.64 p=0.008 for religiousness and Z=−3.04 p=0.002 for support for PSA.

Table 2

Multivariable logistic regression of associations with support for the notion of rational suicide


Most (72%) of our sample of senior doctors in England and Wales reported thinking that rational suicide is possible. There was no significant effect of gender or specialty, but the stronger a doctor reported their religious beliefs to be, the more they were predisposed to disagreement with rational suicide, although even in the most religious group there was only 30% disagreement. This is consistent with other research.15 Unsurprisingly, doctors in support of assisted dying were more likely to report thinking that rational suicide is possible.

This and other samples have previously shown that doctors are divided over assisted dying legislation, with a majority opposed to changing the law to permit this practice.11 12 The opposition of doctors to assisted dying cannot now be explained by this group believing that suicide can never be rational. Other explanations may be the view that it would be impossible in practice to select only those people who rationally wish for suicide, while ensuring vulnerable people receive the protection they need. Second, they may oppose assisted dying legislation because the prospect of medical involvement in bringing about deaths is professionally unacceptable. Third, doctors may feel that suicide, rational or not, is morally prohibited for religious or other reasons.

This study does not allow for straightforward differentiation between these potential explanations. The psychiatrists in our sample, presumably the doctors with the most experience and training in managing suicidal patients, were as supportive of the possibility of rational suicide as the rest of the sample. Indeed, this specialty was the most in agreement with the possibility, although the difference was not statistically significant. A study only examining the views of psychiatrists had a similar finding of 86% of respondents agreeing that rational suicide is possible.14 Previously we have found greater experience of caring for dying patients to be strongly associated with increasing opposition to the legalisation of PAS,11 so it is of interest that greater experience of suicidal patients (as indexed by being a psychiatrist) is not strongly associated with views either for or against the possibility of rational suicide.

As well as the many doctors who agree with the possibility of rational suicide but do not support a change in the law to allow PAS, we noted that some (13 (3.6%)) who opposed the possibility of rationality of suicide nevertheless supported the legalisation of PSA, itself a process to facilitate rational suicides. This could have been because of poor understanding of assisted dying legislation as recently proposed, because of poor appreciation of the questions, or possibly the participants were answering on their view of lethal prescribing for incapacitated individuals, which is not the subject of this paper nor a subject of current mainstream debate in any jurisdiction.

The main strength of this survey is that it was undertaken on a large representative sample of senior UK doctors and contained a single question on the morality of suicide which we think has face validity. The responders and non-responders in the sample were similar on the criteria available (gender and specialty), suggesting no serious problem of response bias. Attitudes of doctors to the idea of rational suicide have been little studied, and not at all in the UK during recent debates on the legalisation of PSA.

This study has some weaknesses. The question on the possible ‘rationality’ of suicide is arguably distinct from the morality of suicide. If ‘rationality’ is interpreted purely as a morally neutral mental ability, then it is possible to believe that suicide may be ‘rational’ but morally impermissible in a similar way to how a certain crime may be seen as rational but still morally impermissible. Probing these subtle distinctions in a questionnaire design, however, we think would be challenging. Contemporary ideas emphasising patient autonomy may influence doctors to select more liberal answers than those which would reflect the views they actually hold; however, were this true, then more liberal answers to the questions about PSA would be expected.

The response rate was modest at 50%. A shorter questionnaire may have resulted in a higher response rate, but this would have meant the collection of fewer important covariates. The doctors in the sample may not be representative of doctors as a whole because of registering with a commercial provider being voluntary. The database was also not up to date, containing relatively few newly registered GPs or specialists. However, we have no reason to believe that this database is inferior in this regard to any other source of details of practising UK doctors readily available to researchers.

We have shown that most senior doctors in England and Wales believe that suicide can be rational. More strongly held religious beliefs were associated with opposition to rational suicide, although, even here, levels of opposition were not high. Further research is needed to investigate the reasons for medical opposition to PAS in light of this. The reasons are likely to concern the problems, in practice, that doctors see in selecting only those people who rationally wish for suicide (a group they may see as small), while giving vulnerable people the protection they need (a group they may see as large). There may also be concerns about the social implications of doctors becoming explicitly linked to a phenomenon that has historically been associated with considerable community suffering and stigma. More detailed questionnaire studies may assist such understanding. Qualitative research could assist the development of such questionnaires and could also allow a deeper understanding of doctor's views on the morality of suicide, as its moral permissibility is also presupposed by the laws on assisted dying currently enacted internationally, as well as the recent proposed legal changes in England and Wales. Such interviews should include doctors with religious belief.


Our thanks to Tony David, Professor of Cognitive Neuropsychiatry at the Institute of Psychiatry, for his advice on this paper.



  • Funding SG is a clinician supported by East London NHS foundation Trust. AP is supported by St Christopher's Hospice. LR is supported by the European Commission's Sixth Framework Programme (contract No LSHC-CT-2006-037777). GSO is supported by the Wellcome Trust. RDH is funded by the NIHR Specialist Biomedical London and Maudsley NHS Foundation Trust and Institute of Psychiatry, King's College London. MH is supported by the Biomedical Research Centre for Mental Health at the Institute of Psychiatry, Kings College London and The South London and Maudsley NHS Foundation Trust. WL is supported by the Medical Research Council.

  • Competing interests None.

  • Ethics approval Joint Maudsley Hospital and Institute of Psychiatry Research Ethics Committee (Ref 06/Q0706/36).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • The concise argument
    D J C Wilkinson