Introduction

Whether physicians or other caregivers should participate in ending the lives of terminally-ill patients is highly controversial. Withholding or withdrawing treatment needed to sustain life is widely practised in western countries, even in countries where it is not legally permitted.For example C.L. Sprung et al. (for the Ethicus Study Group), ‘End-of-life Practices in European Intensive Care Units’, Journal of the American Medical Society 290 (2003): 790-797. In contrast, intervening directly in order to end a patient’s life (euthanasia) or providing the patient with the means to do this (physician-assisted suicide or PAS) is, in most countries, against the law. PAS is legal only in The Netherlands, Belgium, Luxembourg, Germany, Switzerland, and the American states of Oregon, Washington, Vermont, and Montana.

Public Opinion in Western Countries

Public opinion in the United States is sharply divided about PASGallup, ‘Doctor-assisted Suicide is Moral Issue Dividing Americans Most’, 31 May 2011, www.gallup.com/poll/147842/doctor-assisted-suicide-moral-issue-dividing-americans.aspx (accessed 21 July 2013). (Gallup, 2011). In 2012, 55% supported PAS for terminally ill patients and 29% for non-terminal patients suffering from severe pain or disability.Truven Health Analytics, ‘Health Poll: Physician-assisted Suicide’, December 2012, http://truvenhealth.com/truven_insights/npr_assets/NPR_reports_PhysicianAssistedSuicide_1212.pdf
(accessed 21 July 2013). Meanwhile, public opinion in most Western European countries has come to favour PAS.For example, J.A. Rietjens et al., ‘A Comparison of Attitudes towards End-of-life Decisions: Survey among the Dutch General Public and Physicians’, Social Science and Medicine 61 (2005): 1723-1732. In a 2008 telephone poll in France, responses to the question, ‘are you personally in favor or against enacting a law that would authorise a doctor to end the life of a person with an incurable disease and causing unbearable suffering, if this person requests it?’ were 51% completely in favour, 40% somewhat in favour, 6% somewhat against, and 3% completely against.Angus Reid Global Monitor, ‘French Support Doctor-assisted Suicide’, 30 March 2008, www.angus-reid.com/polls/31302/french_support_doctor_assisted_suicide/ ~ (accessed 21 July 2013). In a November 2012 online poll conducted by the Swiss opinion research institute in western European countries where PAS is still against the law, over 70% agreed that each person should be able to choose for themself when and how to die.Agjencia Kombetare e Lajmeve, ‘Europe’s Citizens Call for Assisted Suicide – Opinions from 12 European Countries’, http://noa.al/en/artikull.php?id=257486 (accessed 21 July 2013). Those polled also supported the involvement of doctors in assisted suicide.

The Present Study

The subject of this study was not whether lay people or healthcare professionals regarded PAS as acceptable. Instead, it focused on the views of lay people and healthcare professionals regarding the severity of the penalty physicians who broke the law by providing patients with the means to end their lives should receive. The process by which lay people or healthcare professionals decide whether, in actual cases, physicians should be incarcerated or attributed a token sentence has not previously been studied.

Under French law, euthanasia is equivalent to murder. Article 221-1 of the French Penal Code states: ‘Voluntarily causing the death of another person constitutes murder. It is punished by thirty years of criminal confinement.’Article 221-1, Code Penal (1994), www.legifrance.gouv.fr/affichCodeArticle.do?cidTexte=LEGITEXT000006070719&idArticle=LEGIARTI000006417561&dateTexte=20121103, (accessed 12 November 2012). Article 221-5 states: ‘Trying to end the life of another person by using or administering substances likely to bring about death constitutes poisoning. Poisoning is punished by thirty years of criminal confinement.’Article 221-5, Code Penal, www.legifrance.gouv.fr/affichCodeArticle.do?cidTexte=LEGITEXT000006070719&idArticle=LEGIARTI000006417573&dateTexte=20111217 (accessed 12 November 2012). Article 223-6 of the Penal Code equates PAS with failure to help a person in danger.Article 223-6, Code Penal, www.legifrance.gouv.fr/affichCodeArticle.do?cidTexte=LEGITEXT000006070719&idArticle=LEGIARTI000006417580&dateTexte=20091206 (accessed 12 November 2012).

In addition, physicians who perform euthanasia or PAS put themselves at risk of exclusion from the Ordre des Médecins (Order of Physicians); that is, of being prohibited from practising medicine. Article 38 of the Code of Ethics of the Order of Physicians states:

The physician must continue to care for the dying patient until his final moments, assure by appropriate measures the quality of a life that is coming to an end, safeguard the life of the sick person and comfort his entourage [those around him]. He does not have the right deliberately to provoke death.Code de Déontologie des Médecins, art. 38 (2003), www.conseil-national.medecin.fr/article/
article-38-soins-aux-mourants-euthanasie-262 (accessed 12 November 2012).

Laws against a particular behaviour do not, however, automatically translate into punishment. Examination of cases over the last 15 years in France shows that, in cases of compassionate killing of patients by physicians or members of the family, the law is not strictly applied. These cases are known to the public under the names of their main protagonists: Christiane Malèvre,C. Malèvre, Mes aveux [My confession] (Paris: Laffont, 1999). Marie Humbert,V. Humbert, Je vous demande le droit de mourir [I request the right to die] (Paris: Laffont, 2003). and Laurence Tramois.BMJ, ‘In Brief’, British Medical Journal, International Edition, 334 (2007): 604. In these cases, where there was strong evidence that the patient and the patient’s family requested a life-ending procedure in a state of full awareness, the criminal court either suspended the trial or gave a token sentence. Even in the case in which there was no evidence that the patient had asked for death and there was evidence that the family had not, the sentence was not very severe (for example, one-year suspended).

The French courts’ reluctance to impose on health professionals the punishments stipulated in law is very likely, given the studies described above, to be reflected in the French public. The major purpose of the study was thus to test this expectation. More particularly, its main objectives were (a) to delineate the different views people were likely to have; (b) to try to understand their rationale; and (c) to compare lay people’s views with those of health professionals.

Method

As in many previous studies conducted in the field of empirical bioethics,For example, L. Kpanake, A. Patassi, & E. Mullet, ‘Western African People’s Views on the
Acceptability of Criminal Prosecution for Sexual Transmission of Infectious Diseases’, Sexually Transmitted Infections 89 (2013): 290-294; M.T. Muñoz Sastre et al., ‘Acceptability in France of Induced Abortion for Adolescents’, American Journal of Bioethics 7 (2007): 26-32; N. Teisseyre, E. Mullet & P.C. Sorum, ‘Under what Conditions is Euthanasia Acceptable to Lay People and Health Professionals?’, Social Science and Medicine 60 (2005): 357-368. the method was an application of Information Integration Theory.N.H. Anderson, Unified Social Cognition (New York, NY: Psychology Press, 2008); D.G. Froberg & R.L. Kane, ‘Methodology for Measuring Health-state Preferences-IV: Progress and a Research Agenda’, Journal of Clinical Epidemiology 42 (1989): 675-685.

Participants

The participants were unpaid volunteers. The authors contacted people walking along the pavements of a big city in the south of France; explained the study; asked them to participate; and, if they agreed, arranged where and when to administer the experiment. Of the 110 persons who were contacted, 71 participated, including 14 persons who had already confronted an end-of-life problem regarding a close family member. The authors tried to enrol people from both genders (41 women and 30 men) and different ages (M = 32, SD = 12, range = 18-65 years).

The authors also contacted health care providers working in private offices or in one of the main hospitals in Toulouse. Fifty persons were approached and 23 agreed to participate (46%): 10 generalist physicians (3 women and 7 males, M = 43 years of age), and 13 nurses (11 women and 2 men, M = 42 years). Among them, 78% declared that they had already confronted an end-of life problem in their patients.

Material

The material consisted of 36 cards containing a story of a few lines, a question, and a response scale. The stories were composed according to a four within-subject factor design: Patient’s age x Incurability x Type of suffering x Request, 3 x 2 x 2 x 3. These four information items were in the following order: (a) the patient’s age (35, 60, or 85 years); (b) the level of incurability (or curability) of the illness (completely incurable versus extremely difficult to cure); (c) the type of suffering (extreme physical pain or complete dependence); and (d) the extent to which the patient requests a life-ending procedure, euthanasia or PAS (no request, some form of request, or repeated formal requests). All patients were identified as ‘Mrs X’. The only additional information was, ‘She is currently receiving the best possible treatment’.

A concrete example of a story is the following:

Mrs Endelin is 85 years old. She has a serious illness, totally incurable given current knowledge. She is currently receiving the best possible treatment. She is completely dependent; she cannot breathe by herself and she cannot feed herself. She has asked clearly and repeatedly to resort to euthanasia or physician-assisted suicide.

Under each story was a question and a response scale. The question was: ‘In your opinion, what amount of punishment should the court give for providing this particular patient with the means to resort to physician-assisted suicide?’ The response scale was a 15-point scale with a left-hand anchor of ‘No punishment’ and a right-hand anchor of ‘Very severe punishment’. The cards were arranged by chance and in a different order for each participant.

Procedure

The site was a vacant classroom in the university of Toulouse or the office of the health professional. Each person was tested individually. The participants took 30-45 minutes to complete the ratings. They knew in advance how long the experiment would last. None of them complained about the number of vignettes they were required to evaluate or about the credibility of the proposed situations. Informed consent was obtained from all participants in the study.

Results

As expected, strong individual differences in responses were detected during data gathering. As a result, a K-means analysis was performed on the raw data according to the procedure advocated by Hofmans and Mullet.J. Hofmans & E. Mullet, ‘Towards Unveiling Individual Differences in Different Stages of
Information Processing: A Clustering-based Approach’, Quality and Quantity 47 (2013): 455-464. A four-cluster solution was identified. The patterns of data that correspond to each cluster are shown in Figure 1. Separate ANOVAS using a design of Age x Incurability x Suffering x Request, 3 x 2 x 2 x 3 were conducted on the data of each cluster. Owing to the great number of comparisons, the significance threshold was set at .005.

Figure 1.

Judged level of severity of penalty as a function of patient’s request and curability of illness in each of the four clusters

For 11 participants (64% of whom were men), all the ratings were close to the ‘no penalty’ end of the scale. This cluster was composed of two health professionals (one physician and one nurse) and nine lay people. The mean rating observed in this cluster was 1.11 (SD = .20). No effect was significant.

For 36 participants (42% men), two effects were significant: request and age (see second panel of Figure 1). The penalty rating was higher when the patient had not expressed any request than when the patient had repeatedly requested a life-ending procedure, F(2, 70) = 34.43, p < .001, η²_p = .50. Post-hoc analyses showed that the mean judgment associated with no request (M = 5.83) was significantly different from the mean judgments associated with the two other levels: sometimes (M = 3.74) and repeatedly (M = 2.34), which did not differ significantly. The effect of age, although significant, was weak, F(2, 70) = 10.46, p < .001, η²_p = .23. Post hoc analyses showed that the only significant difference was between 85 years (M = 4.23) and 35 years (M = 3.63). The overall mean rating in this cluster was 3.97 (SD = 1.55): It was called ‘Reluctant to punish’, because even when the patient had not expressed any request, the mean rating corresponded to a low severity level. This cluster was composed of three health professionals (three nurses), and 30 lay participants.

For 30 participants (37% men), the pattern was similar to the one shown in the previous cluster but the curves were steeper. Where the patient had not expressed any request, the rating was very high, close to the maximum value, F(2, 58) = 185.40, p < .001, η²_p = .86. Post-hoc analyses showed that the mean judgment associated with no request (M = 12.89) was significantly different from the mean judgments associated with the two other levels: sometimes (M = 7.63) and repeatedly (M = 2.17), which also differed significantly. The effect of age was again weak, F(2, 58) = 8.36, p < .001, η²_p = .22. The overall mean rating was 7.56 (SD = 0.95). This cluster was called ‘Depending on patient’s request’. It was composed of nine health professionals (six nurses and three physicians) and 21 lay people.

Finally, for the remaining 17 participants (35% men), two effects were significant: patient’s request, F(2, 32) = 56.84, p < .001, η²_p = .78, and level of cura-
bility of the illness F(1, 16) = 13.33, η²_p = .45, p < .005. Post-hoc analyses showed that the mean judgment associated with no request (M = 13.25) was only significantly different from the mean judgments associated with repeated requests (M = 7.48). The overall mean rating was 10.77 (SD = 1.35). This cluster was called ‘Severe penalty’ because, in all cases, the penalty ratings were quite high. It was composed of nine health professionals (six physicians and three nurses) and eight lay people.

The health professionals were significantly less present than the lay persons in the clusters ‘No penalty’ and ‘Reluctant to punish’, and more present in the clusters ‘Depending on request’ and ‘Severe penalty’, Chi²(3) = 13.26, p < .005. A complementary ANOVA using a design of Group (health professionals vs. lay persons) x Age x Incurability x Suffering x Request, 2 x 3 x 2 x 2 x 3, also showed that the group factor was significant, F(1, 92) = 11.84, p < .001, η²_p = .11.

Discussion

We examined how much and under what circumstances French lay people and health professionals think that a physician who provides patients with the means to end their lives should be prosecuted and punished. Eighteen per cent of the participants in this study were quite adamant about the need to punish physicians in this case. This ‘Severe penalty’ position was thus consistent with the laws in France, and of most other countries and American states in which PAS is always illegal. These participants, however, gave lower ratings when patients expressed a will to die, especially if the illness was incurable. This view is parallel to the reduced sentences for PAS given by tribunals in France and in other countries like the United Kingdom.R. Ashcroft, ‘Death Policy in the United Kingdom’, in End-of-life Decision Making: A Cross-
cultural Study, ed. R.H. Blank & J. Merrick (Cambridge, MA: MIT Press, 2005) 197-218.

According to 32% of the participants, physicians of adult patients who suffer from unbearable physical pain or who are completely dependent should not be punished as long as the patients have clearly and repeatedly expressed a will to end their lives, irrespective of their age or of the level of incurability of their illness. This ‘Depending on circumstances’ position was consistent with the laws in Belgium, Germany, Luxembourg, The Netherlands, and Switzerland, where putting the means to commit suicide at the disposition of patients is
allowed for those who clearly request it but not for those who do not request it. Putting the means to commit suicide at the disposition of patients who do not request it would be considered an invitation to suicide and would be prosecuted. The position of these participants is also consistent with, although broader than, the laws in the US states of Oregon and Washington, which address physical suffering, but not complete dependence. It is also consistent with judicial practice in the state of Montana and in France.Humbert, Je vous demande (note 12); L. Tramois (2008), www.lexpress.fr/infos/pers/laurence-tramois.html. Retrieved on November 12, 2012.

For 12% of the participants, punishment was opposed in all cases: the ‘No penalty’ position. For another 38%, punishment was not opposed in all cases, but participants were reluctant to punish the physicians severely even when patients had not expressed a desire to end their lives. This ‘Reluctant to punish’ position is similar to the ‘Depending on circumstances’ position but less severe. We can only speculate about the reasoning of these participants. They may think that patients in these desperate situations do not speak up out of fear that their health care providers will blame them, or their families will misunderstand them, or for religious reasons. To provide them, even unasked, with the means to end their lives is, therefore, to reinforce their autonomy in a way similar to, although more direct than, telling the public in Belgium or The Netherlands about the law permitting PAS. They may alternatively have exhibited deference to the physician – thinking that the physician must have had good reasons. Like the previous group, however, these participants were sensitive to the moral value of patient request, so that the rating of the level of punishment was significantly lower when the patients had expressed a will to end their lives.

Overall, 89% of the lay participants but only 61% of the health professionals considered that physicians who, in response to strong patient request, provide their suffering, terminally-patients with the means to end their lives should not receive any sentence. In contrast, 78% of the health professionals but only 41% of the lay participants indicated that physicians who provide these means to patients who do not request them should be severely punished. Physicians and nurses are less supportive of PAS than lay people, probably because (a) they are consistent with their codes of ethics; (b) they are aware of the seriousness of the role they would play in PAS; and (c) they are aware of their vulnerability in judicial proceedings.M. Guedj et al., ‘The Acceptability of Ending a Patient’s Life’, Journal of Medical Ethics 31 (2005): 311-317.

In spite of several limitations (such as a small sample from only one country), our study is useful in suggesting that citizens balance in their judgments a respect for local laws with a respect for principles of ethics. For the majority, the law against PAS appears to be superseded by respect for the patient’s autonomy and/or empathy with his or her suffering or dependence. For a minority of people, however, ending a life through PAS is wrong and must be punished; it is likely that respect for current law is supplemented by religious injunctions against taking life.G. Baeke, J.-P. Wils & B. Broeckaert, ‘“We Are (Not) the Master of our Body”: Elderly Jewish Women’s Attitudes towards Euthanasia and Assisted Suicide’, Ethnicity and Health 16 (2011): 259-278; M. Kemmelmeier et al., ‘Individualism, Authoritarianism and Attitudes toward Assisted Death: Cross-cultural, Cross-regional and Experimental Evidence’, Journal of Applied Social Psychology, 32 (2002); 60-85. Both lay people and health professionals are therefore divided in their views, but, in general, their views are more nuanced than the laws in France and elsewhere. For most, the laws against PAS can be broken in some circumstances, but not in others. Despite the difference between groups, these findings suggest that French lay people – and, to a lesser degree, French health professionals – have views on PAS closer to those of the parliaments of Belgium, The Netherlands, Switzerland, and the American states of Oregon and Washington than to their own parliament.