Study finds federal survey on assisted dying 'ambiguous and biased'

The new Liberal government has a lot of work to do as it decides how to respond to the Supreme Court’s February ruling on physician assisted dying. One of the major tasks ahead is determining what it will do with the Conservative-appointed External Panel on physician assisted dying.

DWD Canada has expressed repeated concerns about the credibility of the panel, which was tasked to study the issue and report back to Parliament on possible options for federal legislation. Two of the committee’s three members, including chair Harvey Chochinov, had testified in court against the right to physician assisted dying prior to the Supreme Court’s decision in Carter v. Canada.

Now, a new independent report suggests the panel’s online survey — currently the only tool in its arsenal to consult ordinary Canadians about their views on end-of-life choice — is “biased and ambiguous” and thus unlikely to produce results that reflect actual public sentiment on physician assisted dying. The findings pose a serious dilemma for the Trudeau government, which has vowed an evidence-based approach to public policy-making. 

Earlier this fall, DWD Canada commissioned researcher Geordie McRuer, a PhD with the Rotman School of Management who has a background in survey methodology, to provide an independent analysis of the External Panel’s “Issues Book” on physician assisted dying. He examined several factors — including the survey’s design and the makeup of the population that’s likely to respond — that could influence the reliability of the results.

Here are some of the key findings he delivered in his final report:

  • Sampling bias. “Based on my understanding of how The Panel has recruited respondents, the sample will almost certainly not be representative of the Canadian population,” McRuer writes. Hamstrung by the 2.5-month-long federal election for most of the period of the consultation, the Panel “was not actively soliciting respondents through a broadly-targeted marketing campaign or research firm.” However, advocacy organizations (DWD Canada included) with strong positions on assisted dying were still able to mobilize their supporters to share their views, turning the survey from “a research tool into a political tool,” he notes.   
  • Ambiguous questions. McRuer spends much of the report detailing examples of ambiguity in the survey questions, and argues that these ambiguities compromise the validity of the consultation's results. He mentions a section in which respondents are asked to rate, on a scale of 1 to 5, “how concerned” they are about a number of perceived risks related to physician assisted dying (for example, that “patients will be more likely to request assistance from a physician to die because… they feel pressured or influenced by a health care provider?”).'


McRuer says many of the questions in the Issue Book are ambiguous and could be interpreted in several different ways.

“The problem with these questions is that they are actually asking two questions at once,” McRuer points out. Respondents could reasonably believe they’re being asked about the severity of the risk (ie. the degree to which they believe the feared outcome is negative or harmful), the likelihood of this risk occurring, or both at the same time. “Therefore,” McRuer writes, "[one’s] level of ‘concern’ is very difficult to judge. I honestly do not know if I would respond to that question with 1, 5, or something in between. It might swing from one extreme to the other depending on the day.”

  • Insufficient context. The Issues Book, McRuer suggests, doesn’t give enough contextual information to allow respondents, especially lay readers, to make informed responses. For example, the survey uses language from the Supreme Court’s decision like “competent adult person” and “clear consent” without clearly defining what those terms mean in a legal or bioethical context. “The problem of insufficient context,” McRuer writes, “is compounded by subsequent parts of the survey, which contain content suggesting that the Carter decision affected portions of the law that actually remain unchanged. For example, two of the questions (p. 9 and 13) ask respondents to consider a situation in which minors (a 16- and a 17-year-old) seek doctor assistance to end their lives. This question has enormous potential to confuse participants about the nature of the Carter decision," which only applies to adults. 

In his concluding summary, McRuer describes the Issues Book as a “missed opportunity.” He calls the desire to consult the Canadian public “laudable” but says he has “serious doubts” about whether the data collected will be useful to lawmakers going forward.

“There are so many problems with the Issues Book that I do not think that the data can be meaningfully interpreted,” McRuer writes. “The sample of people who were recruited as respondents can tell us nothing about the attitudes of the average Canadian, and the content of the survey is ambiguous and biased.

“In conclusion, I recommend that the results from this survey have little or no influence on legislation that pertains to doctor-assisted dying.”

Click here to download McRuer’s report, "A Methodological Analysis of the Issues Book Survey on Doctor-Assisted Dying."

Be the first to comment

Please check your e-mail for a link to activate your account.