Evidence for the practical slippery slope in the debate on physician-assisted suicide and euthanasia

Medical Law Review

Volumn 13, Issue 1, 2005, Pages 17-44

  Smith, Stephen W ^a  

^a UNIVERSITY OF BIRMINGHAM   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 27644454383     PISSN: 09670742     EISSN: None     Source Type: Journal    
DOI: 10.1093/medlaw/fwi002     Document Type: Review

References (116)

1. For the purposes of this article, 'physician-assisted suicide' means any conduct where the physician assists a patient to die (such as by prescribing medication) but the patient performs the final act to bring about death. 'Euthanasia' means any conduct where the physician assists a patient to die by performing the final act which brings about death.
2. A slippery slope argument is one where it is argued that the acceptance of one standard (for example, a legal standard), while not perhaps undesirable in itself, leads to the acceptance of another standard of the same type which is undesirable. Therefore, it is argued that the first standard should not be accepted because it leads to the second, even if that is the only thing wrong with the first standard.
3. The two types of slippery slope argument are often referred to as the logical slippery slope and the empirical or practical slippery slope.
4. Even if one accepts that the empirical slippery slope argument fails at present, which is the argument developed in this article, it does not mean one must be committed to the viewpoint that either PAS or euthanasia should be legalised. There are a myriad of other arguments for and against the legalisation of PAS and euthanasia which would have to be considered before one could reach an informed opinion on the subject in general.
5. J. Keown, Euthanasia, Ethics and Public Policy (Cambridge University Press 2002).
6. D. Enoch, 'Once You Start Using Slippery Slope Arguments, You're on a Very Slippery Slope' (2001) 211 O.J.L.S. 629. Slippery slope arguments are sometimes referred to as 'thin edge of the wedge' arguments, 'camel's nose' arguments and various other terms.
7. E. Volokh, 'The Mechanisms of the Slippery Slope' (2003) 116 Harvard Law Review 1026, 1032. For the purposes of this article, they shall simply be referred to as slippery slope arguments.
8. R.G. Frey, 'The Fear of a Slippery Slope' in G. Dworkin, R.G. Frey and S. Bok (eds), Euthanasia and Physician-Assisted Suicide (Cambridge University Press 1998) at 43-63, 44.
9. Volokh, supra, n. 6 at 1028-9. Generally, A is probably something either considered good or uncertain, or at least something that is not considered too bad in itself. Otherwise, one would use different arguments that focused on the bad qualities of A.
10. Frey, supra, n. 7 at 44.
11. Enoch, supra, n. 6 at 631.
12. Enoch indicates this possibility by referring to the possibility of intervening steps and referring to the bad consequence as 'Z' as opposed to B. Enoch, supra, n. 6 at 631.
13. Volokh, supra, n. 6 at 1030.
14. Frey, supra, n. 7 at 46.
15. Ibid.
16. Enoch, supra, n. 6 at 636.
17. Enoch, supra, n. 6 at 632.
18. This may be referred to as a situation of multipeaked preferences. Volokh, supra, n. 6 at 1048-9.
19. Ibid.
20. For a thorough article on the different ways in which slippery slope mechanisms such as the ones listed work, see Volokh, supra, n. 6.
21. Enoch, supra, n. 6 at 633.
22. Frey has argued that this is not necessarily a causal relationship but one based upon the idea that if we implement A it becomes 'empirically very likely' that we will move to B. Frey, supra, n. 7 at 45.
23. Enoch, however, has argued that what he calls the Essential Premise - that the decision-maker will fail to abide by the distinction between A and B - is a causal connection. Enoch, supra, n. 6 at 631, 637.
24. Enoch, supra, n. 6 at 631.
25. To posit one rationale for this, it may be that, in Jurisdiction 1, the members of the group which supported the granting of specific rights (Standard A) found the provision of those rights to be sufficient for a period of time and did not feel the need to pursue the recognition of marriage. The members of a similar group in Jurisdiction 2 may have been compelled to pursue the recognition of marriage because they were thwarted in their earlier efforts to implement Standard A. It may, then, have been the only option available to them, even though Jurisdiction 2 would have preferred to vote only to grant specific rights. In other words, in Jurisdiction 2, the majority favoured A (the granting of certain rights to homosexual partners) to pre-A (no granting of rights), and B (the recognition of homosexual marriage) to pre-A, but A to B. Since they were not permitted to vote for A, they instead had to vote for B.
26. An American doctor, Leo Alexander, argued that acceptance of euthanasia was a significant factor in leading to doctors participating in the terrible experiments and practices used in Nazi concentration camps. However, Alexander fails to recognise that what the Nazis practised was not anything like what is being proposed under most PAS or euthanasia legislation. See J. Rachels, The End of Life (Oxford University Press 1996) at 175-8.
27. Keown, supra, n. 5 at 72-7.
28. Keown actually refers to this as voluntary active euthanasia or VAE. Keown, supra, n. 5 at xi. The difference in the order of the preceding adjectives makes no difference as far as the arguments are concerned.
29. Active euthanasia may also be contrasted with passive euthanasia where the final act performed by the doctor is usually the removal of treatment (such as a ventilator). The difference between active and passive euthanasia is not important for our purposes.
30. Keown's most explicit explanation of this argument is when he is discussing the logical slippery slope. Keown, supra, n. 5 at 76-7. However, when examining the data, he appears to examine this argument from an empirical viewpoint as well as the logical one.
31. Keown, supra, n. 5 at 72.
32. Keown presents a logical slippery slope which does rely on this point in the book. See Keown, supra, n. 5 at 76-9.
33. One might argue that this is not really a slippery slope argument at all. What Keown is arguing is not a move from A to B, but a move from a small quantity of A to a larger quantity of A. As such, it is not properly considered a slippery slope argument but an argument from consistency - treating like cases alike. Enoch, supra, n. 6 at 644-5. Keown turns this into a slippery slope argument by putting a qualification on A, turning PAS and AVE into two separate things - PAS or AVE as a last resort and PAS or AVE not as a last resort. Certainly if PAS or AVE were simply allowed under any regulation, the argument would not be a slippery slope argument. This would only present a slippery slope argument in situations were PAS and AVE were allowed under the strict policy that they would be used only as a last resort. Consequently, this article will focus on the possible slide towards non-voluntary or involuntary euthanasia rather than on the increase in the rate of PAS and AVE.
34. Keown, supra, n. 5 at 277.
35. Keown, supra, n. 5 at 73.
36. Keown, supra, n. 5 at 75,
37. quoting D. Callahan and M. White, 'The Legalization of Physician-Assisted Suicide: Creating a Regulatory Potemkin Village' (1996) 30 University of Richmond Law Review 1.
38. Keown, supra, n. 5 at 73.
39. Willems et al., 'Attitudes and Practices Concerning the End of Life' (2000) 160 Archives of Internal Medicine 63-8. The United States' doctors were from Oregon. At the time of the study, the legislature of the state of Oregon had voted to legalise PAS, but due to legal challenges the law had not come into effect. The researchers suggest this legal climate might make a difference when attempting to extrapolate these results to the rest of the United States, but note there is no empirical evidence to suggest that physicians in Oregon have different views from the rest of the United States.
40. Ibid. at 67.
41. Ibid. at 64.
42. Ibid. at 65. Information was also gathered about the practices of physicians in the two countries. We shall return to this information when dealing with the second possible practical slippery slope.
43. Ibid.
44. In the United States, 20 per cent endorsed increasing morphine, 22 per cent endorsed PAS and seven per cent endorsed euthanasia. In the Netherlands the figures were 15 per cent, 18 per cent and 14 per cent, respectively. Ibid.
45. Ibid. at 63.
46. According to the authors, 'the fact that Dutch physicians reject the feeling of being a burden as a reason for euthanasia suggests that the liberal legal climate has not led them to find any reason for euthanasia or PAS acceptable'. Ibid. at 67.
47. The researchers indicate that '[t]o our knowledge, this study is the first to compare attitudes and practices regarding euthanasia and PAS using the same questions between different countries'. The author has also not been able to find any other studies of this nature in the literature.
48. J. Wise, 'Public Supports Euthanasia for Most Desperate Cases' (1996) 313 British Medical Journal 1423.
49. S.A.M. McLean and A. Britton, Sometimes a Small Victory (Institute of Law and Ethics in Medicine, University of Glasgow 1996) Appendix 3. PAS was defined in the study as the person's action leading to their own death; euthanasia was defined as when the actions of other persons lead to death.
50. J.K. Mason, R.A. McCall Smith and G.T. Laurie, Law and Medical Ethics, 6th edn (Butterworths Lexis Nexis 2002) at 18.23.
51. E.J. Emanuel, 'Euthanasia and Physician-Assisted Suicide: A Review of the Empirical Data from the United States' (2002) 162 Archives of Internal Medicine 142.
52. For example, one study found that 36.2 per cent of the population supported euthanasia or PAS if the patient had no pain but is concerned about the burden that deterioration might place on the family. Another study found that 29.3 per cent of the population would support euthanasia or PAS if the patient viewed life as meaningless but has no pain. E.J. Emanuel, D.L. Fairclough and L.L. Emanuel, 'Attitudes and Desires Related to Euthanasia and Physician-Assisted Suicide among Terminally Ill Patients' (2000) 284 Journal of the American Medical Association 2460-8,
53. reported in Emanuel, supra, n. 48 at 142.
54. Emanuel, supra, n. 48 at 142-3.
55. Ibid. at 143.
56. Ibid. at 144.
57. Ibid.
58. Ibid. at 145.
59. This has emerged from a large range of studies reported in Emanuel's article. The responses varied from 1.2 per cent in support of euthanasia in a 1993 mailed survey of 265 Rhode Island physicians (8.6 per cent supported PAS in the same survey) to 70 per cent who supported euthanasia in a 1989 mailed survey of 676 San Francisco physicians. Most results, however, were under 50 per cent of those surveyed in support of euthanasia or PAS. The support does increase if the survey asks if PAS or euthanasia is justified 'in some circumstances'. Ibid.
60. Ibid. at 149.
61. J. Griffiths, A. Bood and H. Weyers, Euthanasia and Law in the Netherlands (Amsterdam University Press 1998) at 198.
62. Ibid. A study in 1995 carried out by Holsteyn and Trappenburg provided substantially similar evidence. In that study, 64 per cent of the population stated that euthanasia should 'always be allowed' and only ten per cent stated that it should 'always be forbidden'.
63. Ibid. at 199.
64. H.R. 27 November 1984, N.J. 1985, 106.
65. Ibid. at 198.
66. The study looked at treatment for severely defective newborn babies including those who would be severely handicapped or have Down's syndrome, coma patients, psychiatric patients, senile dementia and the elderly who wish to determine the time of their death. Ibid. at 199-201.
67. R. Posner, Old Age and Dying (University of Chicago Press 1995) at 253,
68. using data from C. Pritchard, 'Is there a Link between Suicide in Young Men and Employment? A Comparison of the UK with other European Community Countries' (1993) 160 British Journal of Psychiatry 750, 753.
69. Posner, supra, n. 62 at 241.
70. Willems et al., supra, n. 36 at 66.
71. P.J. van der Maas, J.J.M. van Delden, L. Pijnenborg and C.W.N. Looman, 'Euthanasia and other Medical Decisions Concerning the End of Life' (1991) 338 The Lancet 669-74.
72. P.J. van der Maas, G. van der Wal, I Haverkate et al., 'Euthanasia, Physician-Assisted Suicide and other Medical Practices Involving the End of Life in the Netherlands, 1990-1995' (1996) 335 New England Journal of Medicine 1699-1705.
73. B.D. Onwuteaka-Philipsen, A. van der Heide, D. Koper, I. Keik-Deerenberg et al., 'Euthanasia and other End-of-Life Decisions in the Netherlands in 1990, 1995 and 2001' (2003) 362 The Lancet 395-9.
74. The first study in 1990 also included a prospective survey. This was not reproduced in the 1995 or 2001 studies.
75. These numbers are based upon the death certificate study. In the interview study, the rates were 1.9 per cent in 1990, 2.3 per cent in 1995 and 2.2 per cent in 2001.
76. Rates according to the physician interview study were 0.3 per cent in 1990, 0.4 per cent in 1995 and 0.1 per cent in 2001.
77. Questions about rates of termination of life without explicit request were not part of the physician interviews in the 1990 study. Rates according to the physician interview study were 0.7 per cent in 1995 and 0.6 per cent in 2001.
78. It should be noted, however, that the deaths in this category are not all of the same type. The cases included in this category include: A very small number of severely disabled newborn babies; a very small number of coma patients; a very small number of cases of inhabitants of institutions for the mentally disabled; a small number of cases of dementia; and a relatively large number of very sick or dying (cancer) patients who are no longer able to make their will known and who are clearly suffering.
79. Griffiths, Bood and Weyers, supra, n. 57 at 226-7.
80. Keown, supra, n. 5 at 95.
81. Keown, supra, n. 5 at 95-101.
82. See H. Kuhse et al., infra, n. 80
83. and L. Deliens et al., infra, n. 82.
84. Airedale N.H.S. Trust v. Bland [1993] A.C. 789 at 886. Lord Mustill states: 'As I understand the position [the doctors] have all, with heavy hearts, taken the ethical decision that since their efforts have run their course it is better from every point of view that Anthony Bland's life should be brought to an end'.
85. Cruzan v. Director, Missouri Department of Health [1990] 496 U.S. 261 at 295 per Scalia J. One of the arguments presented by Nancy Cruzan's parents was that 'preventing her from effectuating her presumed wish to die requires violation of her bodily integrity'.
86. L. Gormally, 'Walton, Davies, Boyd and the Legalization of Euthanasia' in J. Keown (ed.), Euthanasia Examined: Ethical, Clinical and Legal Perspectives (Cambridge University Press 1995) at 113-40, 124-5.
87. This might provide some evidence for Keown's argument that PAS and AVE rates will continue to increase because these practices will be seen less and less as a last resort. However, the Dutch rates have reached a plateau, which would belie this claim.
88. H. Kuhse, P. Singer, P. Baume, M. Clark and M. Rickard, 'End-of-life Decisions in Australian Medical Practice' (1997) 166 Medical Journal of Australia 191. The physician interview component of the Dutch study was not replicated because the researchers considered anonymity to be an essential requirement because doctors may be giving information about acts that were criminal.
89. Ibid. at 195. This category is comparable to the Dutch category of 'terminations of life without explicit request' or non-voluntary and involuntary euthanasia.
90. L. Deliens, F. Mortier, J. Bilsen, M. Cosyns, R. vander Stichele, J. Vanoverloop and K. Ingels, 'End-of-life Decisions in Medical Practice in Flanders, Belgium: A Nationwide Survey' (2000) 356 The Lancet 1806.
91. Ibid. at 1808.
92. A. Fisher, J.I. Fleming, A. Krohn, N. Tonti-Filippini, 'Letter to the Editor' (1997) 166 Medical Journal of Australia 506.
93. Ibid. This criticism has also been made about the Belgian study.
94. Keown, supra, n. 4 at 135.
95. One should note that the allegation of mistranslation printed in the Medical Journal of Australia (where the study was published) does not specify which particular question was mistranslated. Fisher et al., supra, n. 84. Therefore, it is especially difficult to assess this claim.
96. H. Kuhse, P. Singer, M. Clark and P Baume, 'In Reply' (1997) 166 Medical Journal of Australia 507.
97. Keown, supra, n. 5 at 135. The response rate from the Belgian study was 52 per cent.
98. Deliens et al., supra, n. 82 at 1806.
99. Although one would then presume that the figures are underestimated. Thus, one might conclude that the figures put forth in the Australian and Belgian studies are actually too low and should be higher.
100. A. van der Heide, L. Deliens, K. Faisst, T. Nilstun, M. Norup, E. Paci, G. van der Wal and P.J. van der Maas, 'End-of-Life Decision-Making in Six European Countries: Descriptive Study' (2003) 361 The Lancet 345.
101. The response rates of the surveys varied from 75 per cent in the Netherlands to 44 per cent in Italy. The questionnaire used was 'virtually identical' to the 1990 and 1995 Dutch surveys. Country-specific versions were used with a common English version used for comparison. Ibid. at 346.
102. Keown, supra, n. 5 at 136.
103. The 1995 survey lists the reporting rate as being 41 per cent and the results of the most recent study published in The Lancet do not give rates of reporting.
104. Keown, supra, n. 5 at 134.
105. The full reports for all five years the Oregon Death with Dignity Act has been in use (collectively ODHS Reports) are located at the website for the Oregon Department of Human Services at: http://www.dhs.state.or.us/publichealth/chs/pas/pas.cfm (accessed 20 November 2003). Summaries of the data from the full reports have also been printed in the New England Journal of Medicine. See A.E. Chin, K. Hedberg, G.K. Higginshon and D.W. Fleming, 'Legalized Physician-Assisted Suicide in Oregon - The First Year's Experience' (1999) 340 New England Journal of Medicine 577-83
106. A.D. Sullivan, K. Hedberg and D.W. Fleming, 'Legalized Physician-Assisted Suicide in Oregon - The Second Year' (2000) 342 New England Journal of Medicine 598-604
107. A.D. Sullivan, K. Hedberg and D. Hopkins 'Legalized Physician-Assisted Suicide in Oregon, 1998-2000' (2001) 344 New England Journal of Medicine 605-7
108. K. Hedberg, D. Hopkins and K. Southwick, 'Legalized Physician-Assisted Suicide in Oregon, 2001' (2002) 346 New England Journal of Medicine 450-2
109. and K. Hedberg, D. Hopkins and M. Kohn, 'Five Years of Legal Physician-Assisted Suicide in Oregon' (2003) 348 New England Journal of Medicine 961-4.
110. Keown makes much of the 'admission' in the official Oregon Health Department reports that, while there is no evidence of abuse, this could all be a 'cock and bull' story. Keown, supra, n. 4 at 180. Reading the reports, the authors are stating a fact based upon the inability of the Health Department to engage in full-scale investigations of procedures. There is no evidence to indicate that the reports filed with the Oregon Health Department are false.
111. See M. Otlowski, Voluntary Euthanasia and the Common Law (Oxford University Press 2000) at 404-9. Dr Chabot was convicted because he did not have a consultation with a doctor who actually saw the patient. This was contrary to the law even though it was not contrary to law for him to have assisted his patient.
112. Keown, supra, n. 5 at 109-10.
113. J. de Haan, 'The New Dutch Law on Euthanasia' (2002) 10 Med. L. Rev. 57.
114. Ibid. at 64.
115. Ibid. at 64-5.
116. ODHS Reports, supra, n. 94.