7 November 2019

Dear Member of Parliament

Re: The impact of assisted dying on non-assisted suicide rates in New Zealand – incomplete and potentially misleading advice provided to the Justice Select Committee by the Ministry of Justice and Ministry of Health Officials in the Departmental Report on the End of Life Choice Bill.

We have read the Departmental Report on the End of Life Choice Billⁱ that was prepared for the Justice Select Committee by the Ministry of Justice and the Ministry of Health. We are concerned about the quality of the advice contained within that Report with respect to its assessment of the potentially negative impact that implementing an assisted dying (euthanasia and assisted suicide) regime could have on New Zealand’s commitment to reduce our rates of (non-assisted) suicide.

Our specific concern relates to the potential misrepresentation of the analysis in an academic paper by Jones and Paton that forms the basis of the Departmental Report’s conclusions about the possible impact of an assisted dying regime on suicide rates – a concern shared by the authors of the academic paper.

 1. The sponsor of the End of Life Choice Bill, MP David Seymour, has, on various occasions, argued that implementing assisted dying will potentially lead to a reduction in suicide rates because it will give people an alternative to committing “violent suicides”.ⁱⁱ Others, including overseas groups, have made a similar claim.ⁱⁱⁱ

2. This specific claim has been rebutted, notably in a comprehensive quantitative study of the US states that have legalised assisted suicide. This study was carried out in 2015 by David Albert Jones and David Paton who report: “Controlling for various socioeconomic factors, for unobservable state and year effects, and for state-specific linear trends, we found that legalizing PAS was associated with a significant increase in total suicides (including PAS) and no reduction in rates of nonassisted suicide”^iv (emphasis ours).

3. The question that, to date, remains unanswered by the Jones and Paton paper and other research is the extent to which introducing euthanasia and/or assisted suicide^v might create a dynamic that, unintentionally, leads to an accompanying increase in non-assisted suicide rates.

4. Statements that assisted dying and non-assisted suicides are different in nature, and should not be conflated,^vi miss the point. The critical point is whether there is a causal relationship between the two types of actions such that allowing assisted dying will directly or indirectly lead to an increase in the rates of non-assisted suicides.

5. Assessing the exact nature of the relationship between assisted dying and non-assisted suicides is a crucial question for any country contemplating the introduction of either euthanasia and/or assisted suicide. It is particularly crucial for New Zealand given our exceptionally high rates of suicide, especially amongst young men, which (i) have increased year upon year since the 2013/2014 year and (ii) affect Māori at twice the rate of the “European and other” grouping.^vii

6. Commenting on the Jones and Paton publication quoted above, the Departmental Report on the End of Life Choice Bill notes (page 18, paragraph 105) that many submitters to the Justice Select Committee draw on this paper as evidence that allowing assisted dying would increase suicide. The Departmental Report then concludes: “There is no evidence that assisted dying impacts suicide rates.” This conclusion is reiterated on page 123 of the same report at paragraphs 7 and 8: “The [Jones and Paton] study found no significant effect on suicide rates. The total number of suicides only rose if assisted deaths were included.”

7. We are concerned that the conclusion drawn here by Ministry of Justice and Ministry of Health officials, based on the Jones and Paton study, may have given a false sense of reassurance to the Justice Select Committee, and consequently to all MPs, regarding the likely impact of assisted dying on suicide rates in New Zealand should the End of Life Choice Bill pass into law. In the discussion of their findings, Jones and Paton themselves caution: “There are several limitations to the analysis in this study that suggest our results should be treated with some caution” (p. 603).

8. In subsequent correspondence with the authors of the study in question, David Jones has responded (by personal email) that the advice presented in the Departmental Report “does not capture [their] analysis. [The Report’s] sentence is simple and unequivocal whereas we described the results as ‘equivocal’ (page 603)”.^viii

In the opinion of the authors, a close and full reading of their paper draws a more nuanced and more tentative conclusion:

The association [between assisted dying and non-assisted suicides] was ‘positive and significantly so when we do not include state-specific trends’ (page 602) but when these were included ‘the estimated association, although positive, was smaller and no longer statistically significant’ (page 603). However, the inclusion of state-specific trends, while it adds to the robustness of the analysis, asks a lot of the data. The fact that the association is smaller and no longer significant by this test does not necessarily mean there is no effect. A better summary of the paper would be ‘There is some evidence that assisted dying impacts suicides increase’.

9. In addition to this concern, it needs to be noted that a significant limitation of the Jones and Paton study, as a point of reference for the New Zealand context, is that it applies only to assisted suicide and does not include any analysis of jurisdictions that allow euthanasia (such as the Netherlands and Belgium). This is significant for at least three reasons: (i) the proposed New Zealand law will allow both euthanasia and assisted suicide; (ii) when euthanasia is available in a jurisdiction, the overall rates of assisted dying are up to 10 times greater (for example, in 2017, Oregon deaths from assisted suicide were approximately 0.4% of all deaths, while in the Netherlands euthanasia deaths for the same year are reported as 4% of all deaths);ix and (iii) when both methods of assisted dying are available, only an extremely small percentage of eligible candidates choose assisted suicide over euthanasia. ^x

10. Meanwhile, as outlined below, the trends associated with the overall rates of non-assisted suicide for the Netherlands and Belgium (where euthanasia is available) suggest that assisted dying may lead to an increase in non-assisted suicides. While not definitive, until more detailed and substantial research is carried out, the suicide statistics from these two countries show that the question of a direct and/or indirect causal relationship between assisted dying and non-assisted suicides, cannot be ruled out.

11. In the Netherlands, non-assisted suicides have continued to rise in the last 10 years, against the trend of surrounding European countries which all show falling rates of suicides over recent times.^xi In 1991, the (crude) suicide rate per 100,000 population for the Netherlands stood at 11.6, before dropping to a low of 8.4 in 2007 and then increasing steadily again to 11.2 in 2017 (a 33% overall increase in 10 years).^xii Dutch Professor and former long-serving member of a Dutch Euthanasia Review Committee, Theo Boer, notes that the “rise was all the more significant since in exactly that same period we made assisted dying possible for the categories of people that do commit suicide.”^xiii 

12. The reasons for the significant rise in the Netherlands’ suicide rate from 2007 to 2017 might well be related to internal socio-economic realities, such as the severe housing crisis that followed the financial crisis of 2008. However, this cannot be stated categorically, and it is notable that the rates have remained high and not fallen since 2013. 

13. One would also expect that, if the rise in non-assisted suicide numbers in the Netherlands was totally unrelated to the availability of assisted dying, there would be comparable increases in suicide rates in at least some of the other countries surrounding the Netherlands who are similar in terms of age, ethnicity, socio-economic status, religion and language and were affected in a similar way by the economic and housing crisis.

14. It is further worth noting that there was, at the same time, a significant upward trend in the euthanasia and assisted suicide rates recorded in the Netherlands from 2007 on (see Appendix 4).

15. It has been pointed out that in Belgium, which has a similar assisted death regime to the Netherlands, the crude suicide rates have fallen from 19.5 per 100,000 in 2002 (the year that the Belgium parliament legalised euthanasia) to 15.9 in 2016.xiv At first glance, this would seem to undermine the argument that making assisted death legally available contributes to an increase in non-assisted suicides. However, it needs to be noted that, as far back as 1994, Belgium belonged to a group of countries where the rates of suicide were exceptionally high – more than twice as high as the Netherlands.

16. In 1994, Austria (21.7 per 100,000 persons), France (21.6) and Belgium (21.3) had almost identical crude rates of suicide. But whereas by 2015, Austria’s rate was 12.9 and France’s rate was 13.1, a drop of 40% for both countries over that period, Belgium’s rate of 15.8 in 2015 represents a drop of only 26%. To the best of our knowledge, these differences remain unexplained. (Here, as for the Netherlands, the number of assisted suicides is not included in the total suicide numbers.) 

17. Clearly, more research is needed to be able to determine the precise nature of any link between assisted dying and non-assisted suicides.

In conclusion:

18. Based on the data available, it cannot be categorically stated that legalising assisted dying in New Zealand will not further exacerbate our rates of (non-assisted) suicide over time – there is no evidence to support such a claim. To the contrary, there is some statistical evidence that suggests the possibility of a causal link.

19. Given the possibility that the introduction of assisted dying will lead to a further increase in New Zealand’s suicide rates, you, as an MP, need to be certain that there is no causal link between assisted dying and increased rates of non-assisted suicides before passing the End of Life Choice Bill.

20. Until it can be shown beyond reasonable doubt, based on robust evidence, that there is no causal link, and while suicides remain at epidemic levels in New Zealand, in particular for Māori, we maintain that it is too risky to legislate for euthanasia and/or assisted suicide in New Zealand.

21. We believe this is a clear case for applying the ‘Precautionary Principle’: “When human activities may lead to morally unacceptable harm that is scientifically plausible but uncertain, actions shall be taken to avoid or diminish that harm. Morally unacceptable harm refers to harm to humans or the environment that is: threatening to human life or health, or serious and effectively irreversible, or inequitable to present or future generations, or imposed without adequate consideration of the human rights of those affected.” ^xv

22. Given that Australia’s current overall rate of suicides is very similar to that of New Zealand (see Appendix 3), and given that assisted dying has recently been introduced into the State of Victoria, it makes sense to monitor the impact of assisted dying on Victoria’s suicide rates over the next 5 years or more. ^xvi This will, however, only provide partial information since Victoria’s assisted dying regime is limited to cases of self-administration/assisted suicide (excepting where a person is unable to self-ingest as a result of a physical disability). Therefore, because New Zealand’s proposed legislation is much more like Canada’s law (introduced in June 2016) insofar as it would offer both assisted suicide and euthanasia, it would also make sense to monitor suicide rates in Canada.

23. In light of the above, we recommend that the Ministry of Health and the Ministry of Justice immediately review and update the advice given to the Justice Select Committee. 

24. In addition, we recommend that the Ministry of Health and the Ministry of Justice be asked to undertake detailed research that investigates the possibility of a causal link between suicide rates and the practice of assisted dying. This should include an examination of the change in suicide rates in Belgium and the Netherlands in relation to the recent marked increase in the proportion of deaths from euthanasia in those countries. It should also include an examination of emerging trends in Canada and the Australian State of Victoria, over an extended period of time, as the legislation beds in.

Yours sincerely

Dion Howard on behalf of

Dion Howard
Registered Nurse (BN)
Specialist Youth Mental Health

Dr Elliot Bell
Senior Lecturer and Clinical Psychologist
Wellington School of Medicine (Otago) 

Associate Professor Yoram Barak
Consultant Psychogeriatrician

Kirsty Louden-Bell (NZCCP)
Clinical Psychologist

Charlotte Paul (MB, ChB, PhD, FNZCPHM)
Emeritus Professor Preventive and Social Medicine
University of Otago

Māmari Stephens
Reader - Te Kura Tātai Ture - Faculty of Law
Victoria University of Wellington

Warren Brookbanks
Professor of Criminal Law & Justice Studies
AUT Law School

Helen Nicholson
Professor of Anatomy
University of Otago

Dr Christopher K Gale (MBChB, MPH, FRANZCP)
Consultant Psychiatrist
Senior Lecturer, Dept of Psychological Medicine

Dr Mary Gillooly (MBBCh, FRNZCGP, FRANZCP
Consultant Psychiatrist Cert. Old Age Psych)
University of Otago

Dr Christopher Ansley (MBChB, FRANZCP)
Consultant Psychiatrist Consultant

Dr Daniel Svoboda (MBChB, FRANZCP, CACert)
Child and Adolescent Psychiatrist

Dr Helen Bichan ONZM (MBCHB, FRANZCP, FNZCPHM, DHA)
Retired Consultant Psychiatrist

Dr Wayne de Beer (MBBCH, FRANZCP, MClinEd)
Consultant Psychiatrist and Clinical Training Director

Juan Garcia Quiroga (MD, FRANZCP, FAChAM)
Consultant Psychiatrist

Dr John Adams (MB ChB, FRANZCP)
Psychiatrist

Dr Ana Gina Sembrano-Yearley (DClinical Psy)
Clinical Psychologist

Dr John Kleinsman
Researcher, The Nathaniel Centre

Philip Bagshaw
Clinical Associate Professor,
University of Otago, Christchurch

Susan Bagshaw
Senior Clinical Lecturer
University of Otago, Christchurch
Chair, Canterbury Charity Hospital Trust Chair Youth Hub Trust

Dr Richard J. Linscott (PhD, PGDipClPs)
Clinical Psychologist
Senior Lecturer, Department of Psychology
University of Otago

Endnotes
i) Available at https://www.parliament.nz/resource/en-NZ/52SCJUADV74307JU65473/0ef8505fb4c7a82902a17f262f5f1d850a1057c2.

ii) See, for example, https://www.newsroom.co.nz/2017/12/13/68408/big-vote-in-favour-of-euthanasia-bill#. This argument was engaged in the Canadian Supreme Court case and ultimately led to the introduction of Medical Assistance in Dying in Canada (MAiD) in 2016.

iii) See for example EXIT (2018). FAQ: What does physician-assisted suicide mean? Retrieved from https://

www.exit.ch/en/en/faq/: “EXIT’s option of physician-assisted suicide is actually an effective form of suicide

prevention. Living in the certain knowledge of a way out has motivated more than half of the people originally intent on dying to keep enduring their painful lot until they passed away the natural way. Also, EXIT’s physician-assisted suicide keeps people from dying violently at their own hand and alone by themselves.”

iv) David Albert Jones, DPhil; David Paton, PhD. How Does Legalization of Physician-Assisted Suicide Affect Rates of Suicide. Southern Medical Journal 2015; 108 (10): 599 - 604, available at https://pdfs.semanticscholar.org/6df3/55333ceecc41b361da6dc996d90a17b96e9c.pdf. Accepting this finding does not mean that there might not be a few individuals for whom assisted dying acts as an alternative to a self-inflicted violent suicide, as noted by Professor Theo Boer in his submission to the Justice Select Committee, available at https://www.parliament.nz/resource/en-NZ/52SCJUEVI7430748838/863d921ba00bbc623090323d82a8da87321e6c8a, paragraph 7.

v) Assisted suicide is understood as referring to the prescribing of life-ending drugs by a health professional for a person to administer themselves. Euthanasia occurs when a third-party, usually a health professional, administers a life-ending drug to a person.

vi) See, for example, Graham Adams (2019), Euthanasia debate: Conflating youth suicide and assisted dying is a bad move, available at https://www.noted.co.nz/currently/currently-social-issues/euthanasia-debate-nz-conflating-youth-suicide-and-assisted-dying-bad-move. “You don’t have to be very thoughtful to see the clear difference between a depressed teenager who is going through a rough patch and a rational person who wants help to die to avoid the last, most distressing phase of a terminal illness.”

vii) Figures obtained from the “Annual provisional suicide statistics for deaths reported to the Coroner between 1 July 2007 and 30 June 2019”. Available at https://coronialservices.justice.govt.nz/assets/Documents/Publications/Provisional-Figures-August-2019.pdf. With respect to the rates of suicide deaths by ethnicity between July 2007 and June 2019, it is of note that in the 2007/2008 year, the rates for Māori per 100,000 stood at 15.39 while the rates for “European and other” were 13.26, meaning Māori rates were 16% higher. In the 2018/2019 year, the rates for Māori are 28.23 while the rates for “European and other” are 13.46, meaning Māori rates are 110% higher having increased overall by 83%, while rates for “European and other” have increased by a mere 1.5%.

viii) Email received 31 July 2019.

ix) See https://www.canada.ca/en/health-canada/services/publications/health-system-services/medical-assistance-dying-interim-report-april-2019.html in “Summary of Findings”.

x) In Canada, for example, only one case of self-administration of Medical Assistance in Dying (MAiD) was reported from a total number of 2,615 deaths in the period January 1 to October 31, 2018. See https://www.canada.ca/en/health-canada/services/publications/health-system-services/medical-assistance-dying-interim-report-april-2019.html, Table 2. Similarly, as Professor Boer noted in his submission to the Justice Select Committee, in the Netherlands, “97% prefer the doctor to ‘do the act’” (para 6): https://www.parliament.nz/resource/en-NZ/52SCJUEVI7430748838/863d921ba00bbc623090323d82a8da87321e6c8a.

xi) OECD (2019), Crude suicide rates (indicator). Available at https://www.oecd-ilibrary.org/social-issues-migration-health/suicide-rates/indicator/englisha82f3459-en#wrapper (Accessed on 02 October 2019). See also https://data.oecd.org/chart/5Hdm, for a summary of the crude suicide rates for Austria, Belgium, France, Germany and the Netherlands between 1990 and 2015, printed as Appendix 1. (See also WHO age-standardised rates printed as Appendix 2. Both the crude rates and the age-standardised suicide rates show a very similar pattern.)

xii) Statistics obtained from the Centraal Bureau voor de Statistiek and available at: https://www.cbs.nl/en-gb/news/2019/26/fewer-suicide-deaths-in-2018. It should be noted that the OECD figures for the Netherlands for the same period differ slightly from the Centraal Bureau statistics.

xiii See https://www.parliament.nz/resource/en-NZ/52SCJUEVI7430748838/863d921ba00bbc623090323d82a8da87321e6c8a.

xiv) See https://data.oecd.org/chart/5Hdm, printed as Appendix 1.

xv) See: http://www.precautionaryprinciple.eu/.

xvi) See https://data.oecd.org/chart/5HhF, printed as Appendix 3. In 2018, the Australian Bureau of Statistics reported that the State of Victoria was one of only three states with decreasing rates of “death by intentional harm” nationwide in 2017 (621), while Australia as a whole recorded one of its highest rates of suicide per 100,000 people that year (see https://www.bendigoadvertiser.com.au/story/5668329/victoria-records-decrease-in-deaths-due-to-suicide/). The downward trend in Victoria has continued in 2018 with 591 deaths by intentional harm recorded (9.1 deaths per 100,000) (see https://s3-ap-southeast-2.amazonaws.com/lifeinmind/assets/src/uploads/Nationalslides2018.pdf).

Appendix 1

OECD Crude suicide rates for Belgium, France, Austria, Netherlands and Germany: 1990 – 2015

See https://data.oecd.org/chart/5Hdm.

Appendix 2

WHO Age-standardised suicide rates for Belgium, France, Austria, Netherlands and Germany: 2000, 2010, 2015 and 2016

See http://apps.who.int/gho/data/node.main.MHSUICIDEASDR?lang=en.

Appendix 3

OECD Crude suicide rates for New Zealand and Australia: 1990 – 2015

See https://data.oecd.org/chart/5HhF.

Appendix 4

Rates of euthanasia and assisted suicide in the Netherlands from 2002 - 2017

See https://www.theguardian.com/news/2019/jul/15/euthanasia-and-assisted-dying-rates-are-soaring-but-where-are-they-legal.