The public’s viewpoint on the right to hastened death in Alberta, Canada:findings from a population survey study
Donna M. Wilson
RN PhD Professor
1
, Stephen Birch
DPhil Professor
2
, Rod MacLeod
PhD FAChPM Professor
3
,Nurin Dhanji
RN BScN Research Assistant
1
, Jane Osei-Waree
MN Student Research Assistant
1
and Joachim Cohen
PhD Professor
4
1
Faculty of Nursing, University of Alberta, Edmonton, AB, Canada,
2
Department of Clinical Epidemiology andBiostatistics, Centre for Health Economics and Policy Analysis, McMaster University, Hamilton, ON, Canada,
3
Department of General Practice and Primary Health Care, School of Population Health, University of Auckland,Auckland, New Zealand and
4
End-of-Life Care Research Group, Ghent University & Vrije Universiteit, Brussels,Belgium
Accepted for publication 22 August 2012
Correspondence
Donna M. WilsonFaculty of Nursing, University ofAlberta, Edmonton, ABCanada T6G 1C9Tel.: (780) 492 5574Fax: (780) 492 2551E-mail: donna.wilson@ualberta.ca
What is known about this topic
•
Few population surveys or opinionpolls have been conducted and pub-lished so as to provide readily avail-able and credible information onpublic views of assisted suicide andeuthanasia.
•
Legislation permitting euthanasiaand
⁄
or assisted suicide has come toexist in a growing number of countries.
•
Assisted suicide and euthanasiaremain highly controversial.
What this paper adds
•
This population-based surveyrevealed majority public support forthe right to hastened death in oneprovince of Canada.
Abstract
A research study was conducted to determine public opinion in Alberta,a Canadian province, on the controversial topic of death hastening.Questions on the right to hastened death, end-of-life plans and end-of-life experiences were included in the Population Research Laboratory’sannual 2010 health-care telephone survey, with 1203 adults providingresults relatively representative of Albertans. Of all 1203, 72.6% said yesto the question: ‘Should dying adults be able to request and get helpfrom others to end their life early, in other words, this is a request forassisted suicide’? Among all who provided an answer, 36.8% indicated‘yes, every competent adult should have this right’ and 40.6% indicated‘yes, but it should be allowed only in certain cases or situations’. Over50% of respondents in all but one socio-demographic population sub-group (Religious-other) were supportive of the right to hastened death.However, multinomial regression analysis revealed that the experiencesof deciding to euthanise a pet
⁄
animal and developing or planning todevelop an advance directive predicted support, while self-reported reli-giosity predicted non-support. Finding majority public support for deathhastening suggests that legalisation could potentially occur in the future; but with this policy first requiring a careful consideration of the modelof assisted suicide or euthanasia that best protects people who are highlyvulnerable to despair and suffering near the end of life.
Keywords:
assisted suicide, euthanasia, hastened death, population survey,public opinion poll
•
However, approximately one half of all those in favour of the right tohastened death indicated that itshould be allowed in only certainsituations.
•
Majority public support suggestslegalisation could occur in thefuture; a public policy first requir-ing a careful consideration of themodel of assisted suicide and
⁄
oreuthanasia that best protects peoplewho are highly vulnerable to des-pair and suffering near the end of life.
Assisted suicide and euthanasia cannot be performed legally in Canada.However, ongoing discussion of decriminalising death hastening hasoccurred since Sue Rodriguez’ 1992–1993 request to the Supreme Court of Canada for help in ending her life with amyotrophic lateral sclerosis(ALS) at a time of her choosing. The narrow (5–4) ruling against her dem-onstrated much legal, if not also societal, support at that time for her posi-tion. It is ironic that although the Criminal Code was not amended then topermit death hastening, Sue Rodriguez was widely reported as having been euthanised and the person or persons who ended her life were neverprosecuted. Other cases of hastened death have since occurred in Canada, but few have been charged with what Mullock (2010) identified as a‘criminally compassionate’ offence. With the exception of Robert Latimerwho was recently released from prison after serving a 10-year sentence for
200
ª
2012 Blackwell Publishing LtdHealth and Social Care in the Community (2013)
21
(2), 200–208 doi: 10.1111/hsc.12007
euthanising his daughter who had a disability, convic-tions for death hastening are almost non-existent in Canada. Currently, a number of Canadiansare petitioning for the right to have death hasteningassistance, with a Supreme Court judge in the prov-ince of British Columbia (BC) having ruled on June15, 2012 that the Criminal Code prohibition on assistedsuicide violates the rights of people who cannot com-mit suicide without assistance (2012 BCSC 886 Carterv. Canada). This ruling was immediately suspendedfor 1 year to permit the Federal Government torespond, although one petitioner received permissionto obtain assisted suicide after certain conditions have been met. In addition, the Select Committee on Dyingwith Dignity, a group initiated by the National Assem- bly of Quebec 2010 (the government for the largelyfrancophone province of Quebec) has recently releasedtheir report on the open inquiry held in 2010–2011 togain public and professional viewpoints on legalisinghastened death in that province. This report contains12 recommendations to improve access to and qualityof palliative care, and 12 recommendations for assistedsuicide and euthanasia to occur without sanction inthe province. Specifically, recommendation 13 states(after translation into English), ‘the Committee recom-mends that relevant legislation be amended to recog-nize medical aid in dying as appropriate end-of-lifecare if the request made by the person meets the fol-lowing criteria...’ (Dying with Dignity 2012, p. 1). Thisreport and the BC Supreme Court ruling have gener-ated much discussion across Canada.Canada is not the only country wrestling with theissue of what to do when people ask for help to commitsuicide (i.e. assisted suicide) or ask for hastened deathwhen they can no longer commit aided or unaided sui-cide (i.e. euthanasia). Clearly, there are people who seekassisted suicide and euthanasia including those who tra-vel to Switzerland as ‘suicide tourists’. There is consider-able complexity in legislating and conducting hasteneddeath,however,asillustratedbythe Australiansituation.In 1995, the Northern Territory government in that coun-try passed the
Rights of the Terminally Ill Act
, which sanc-tioned ‘the right of a terminally ill person to requestassistance from a medically qualified person to voluntar-ily terminate his or her life in a humane manner’. TheAct was quickly overturned by the Federal Government(Law Council of Australia 2008) and no replacement Acthas been passed since. Hastened death is legally prac-tised now in Switzerland, the Netherlands, Belgium,Luxembourg, and three American states (i.e. Oregon,Washington, Montana); although with some differencesin decisional criteria and mandated death-hastening pro-cesses (Burkhardt
et al.
2006, Cohen
et al.
2006). Onewould expect public support to have been a prerequisitein these jurisdictions for hastened death legalisation. Todate, no population-based polling of Canadians on thetopic of hastened death has been done and the resultspublished for open discussion purposes. Evidence isneeded to guide discussion and inform policy. In thisstudy, we assessed public opinion on the right to has-tened death in Alberta, a Canadian province witharound 3.6 million citizens. We also examined whichsocio-demographic factors and end-of-life experienceswere systematically associated with support for oragainst the right to hastened death.
Research methods
Survey design
The University of Alberta’s Population Research Labora-tory was commissioned to include a series of end-of-lifequestions in their 2010 spring
⁄
summer random-digit-dialling telephone survey, with a University of Albertaresearch ethics committee approving this survey inadvance. The end-of-life questions were developed byan internationalresearchteam after areview of the litera-ture, and the questions were piloted locally with 20expertsinsurveydesign.Minorchangeswerethenmadeto enhance public understanding of the questions andtheir potential answer choices. The sampling plan forgaining information representative of all adults living inAlberta included: (i) a minimum of 1200 participants (thePopulationResearchLaboratoryforpopulationrepresen-tation normallysurveys this number of participants, hav-ing determined that it is an appropriate number, hencethis number was set), (ii) proportional geographical rep-resentation; with 1
⁄
3 of the 1200 expected respondents,respectively,tobefromthe Edmontonmetropolitanarea,the Calgary metropolitan area, or all other rural andurban areas combined, (iii) equal numbers of men andwomen in each of the three regions, (iv) one respondentper household, with these numbers obtained through acomputer-generated random digit-dialling system usinglisted non-business telephone numbers for the province,and (v) multiple call-backs or quota substitutions afterparticipation refusals or 14 call-backs that did not findanyone at home. In May through July 2010, a total of 1203 anonymous adult volunteers were identified as eli-gible and took part in the study. Each subject had beenread a script that informed them they were being askedto take part in a research study, but they could decline toparticipate, and they could answer all or some of thequestions without repercussions. Persons who were inel-igible to participate in the study were those whoreported they were younger than 18 years of age or anon-resident of Alberta, as well as those who wereunwilling or unable to respond to the questions. The
Public support
ª
2012 Blackwell Publishing Ltd
201
obtained sample was representative for the total popula-tion of Alberta in terms of area of residence, gender, being legally married or not and number of children inthe household. Albertans under 45 years of age wereunderrepresented (see Table 1).
Measures
The main outcome measure used in this study was thequestion on hastened death, formulated as follows:‘Should dying adults be able to request and get helpfrom others to end their life early, in other words, thisis a request for assisted suicide’? Respondents coulddecline to answer the question or answer yes, everycompetent adult should have this right; yes, but itshould be allowed only in certain cases or situations; orno. Information was also collected on the respondents’gender, age, number of children in the household, edu-cational attainment, marital status, household and indi-vidual income (grouped into four categories),occupation (dichotomised to health professionals or allother occupations), working or non-working
⁄
retired,religious affiliation, political affiliation and preferredplace of death. Religious affiliation was based on twoquestions, one asking about the religious denominationof the respondent (19 possible categories and a category‘other’ with a possibility to specify) and another askingwhether the respondent considers himself or herself to be a religious person (answer possibilities on a scalefrom 1, ‘strongly disagree’ to 7, ‘strongly agree’ to thequestion ‘I consider myself to be a religious person’).The political orientation of the respondent was gained by the question: ‘If an election were held today, howwould you vote federally’? The question about therespondent’s preferred place of death was asked assuch: ‘Where would you choose to spend your last daysof life’? The respondents could answer: hospital, hos-pice, nursing home, own home, home of a relative orfriend and other as options.
Statistical analyses
Univariate statistics were used to describe views on theright to request and receive hastened death assistance.Kruskal–Wallis tests were used to evaluate bivariateassociations between the acceptance of assistance indying and all independent variables. This test is used forcomparing ordinal variables for two or more groups.With the outcome variable (acceptance of hasteningdeath)beingordinal(no,conditionalyes,yes)ratherthannominal, the Kruskal–Wallis test was deemed moreappropriate than the Pearson chi square test. Statisticalsignificance was set at
P
< 0.05. All independent vari-ables were considered candidates for construction of amultinomial logistic regression with the hastening deathvariable as the dependent variable (no as the referencecategory vs. yes and conditional yes). A stepwise proce-dure was used to select the variables, with only the sta-tistically significant (
P
< 0.05) covariates allowed to stayin the final logistic regression models. Nagelkerkepseudo-
R
2
was used to evaluate the variance explained by the model. SPSS version 19.0 was used for all statisti-cal computations.
Findings
A total of 1203 respondents participated in the survey,50.3% of whom were women, 19% 60 years or older,67.2% in a household without children, 71.8% with somepost secondary education and 28.5% not married(Table 1). In response to the primary research questionof interest: Should dying adults be able to request andget help from others to end their life early; 72.6% repliedyes, 21.3% replied no, and 1.6% declined to answer and4.6% answered ‘don’t know’. The data from all peoplewho did not answer this question or answered ‘don’tknow’ were removed from analysis and among the
Table 1
Socio-demographic characteristics of sample, as com-pared with available population statistics for the Province ofAlbertaSample Population
*
P
-value (
v
2
)
‡
AreaMetropolitan Edmonton 33.3 31.8 0.26Metropolitan Calgary 33.3 33.3 >0.99Other Alberta 33.4 34.9 0.28GenderMale 49.7 49.6 0.95Female 50.3 50.4 0.95Age18–24 5.8 13.9
<0.001
25–34 12.7 20.1
<0.001
35–44 17.2 20.2
0.01
45–54 22.9 19.7
0.01
55–64 22.5 12.5
<0.001
65+ 19.0 13.6
<0.001
Legal marital statusNot married 34.9 35.6 0.61Married 65.1 64.4 0.61Children in householdNo children 67.2 69.5
0.08
1 or more 32.8 30.5
0.08
*
Statistics based on census 2011 for area and for gender, age,marital status, and number of children in household on census2006 (as not yet available in census 2011); all percentagescalculated for population aged 18 and over. Statistics consultedat http://www.statcan.gc.ca.
‡
P
-value calculated through chi square test comparing sampleand population distributions for each presented category vs. allothers.D. M. Wilson
et al.
202
ª
2012 Blackwell Publishing Ltd