Strengths and weaknesses

One of the weaknesses of the study is the low response rate. This means that response bias cannot be ruled out. Low response rates are a problem in many surveys (10), but analyses indicate that the responses may still be representative (11). The sample was drawn from a panel that is representative of the population, and the answers were weighted based on demographic parameters.

Comparisons with other studies

In 2019, two public opinion surveys were conducted on attitudes towards assisted dying. In one survey, 51% were in favour (12), and in an opinion poll, 77% were in favour of legalising ‘active euthanasia’ in Norway (2). The surveys consisted of a single question about legalisation and ‘Yes’, ‘No’, ‘Don’t know’ response alternatives. Our survey allowed respondents to nuance and qualify their views. They could do this in two ways: by nuancing their support for or opposition to legalisation with the response alternatives ‘Agree to some extent’ and ‘Disagree to some extent’, and by making a distinction between the different situations, and qualifying their support for legalisation as applying to some situations but not others.

In this survey, we have closely followed the questions and definitions from a survey conducted in 2015 (1), but use the term ‘assisted dying’ instead of ‘active euthanasia’. The same questions were used in a survey of doctors’ attitudes from 2016 (5). As with the survey from 2015, our survey shows that there is support for assisted dying, particularly in terminally ill patients with a short remaining life expectancy. As before, there is less support for the legalisation of assisted dying in other situations.

The population is much more in favour of the legalisation of assisted dying than doctors (Table 4). Possible explanations could be the medical profession’s traditional opposition to assisted dying, and doctors’ experiences with the end-of-life phase and knowledge that it is often possible to provide good palliative care.

The population’s support for legalisation seems to have decreased somewhat, or at least not increased, since 2015. However, since both surveys had a low response rate, differences must be interpreted with caution.

In international studies, younger, highly educated, male and non-religious respondents tend to be more supportive of the legalisation of assisted dying (13). In our study, those who were non-religious were more often in favour of legalisation, and no gender difference was found (Table 3). The age group 30–44 years was most in favour of legalisation, and more in favour than the youngest group (< 30 years), which was a surprising finding.

Implications of the findings

The nuances in the response alternatives allow for different interpretations of the results. A widely used interpretation would be to combine those who strongly agree and agree to some extent into a single group of ‘Agree’, and to combine those who strongly disagree and disagree to some extent into a single group of ‘Disagree’. This would mean that 68% of respondents are in favour of physician-assisted suicide in terminally ill patients with a short remaining short life expectancy. The corresponding figure for euthanasia is 59%.

However, there is also another way to interpret the answers: only those who ‘Strongly agree’ and ‘Strongly disagree’ can be considered as absolutely in favour of or opposed to legalisation, respectively. The respondents in the three categories in the middle – ‘Disagree to some extent’, ‘Neither agree nor disagree’ and ‘Agree to some extent’ – could be viewed as not being fully decided on the matter. In this interpretation, the respondents are divided into three: those who are strongly in favour of legislation, those who are strongly against legalisation, and the large group in the middle. Our figures show that the size of this group in the middle varies depending on the specifics of the question, according to what type of assisted dying is being asked about, and what patients it would apply to. This undecided group constitutes between 56% and 67% in the first three questions in the survey (Table 2).

Perhaps this three-way split of respondents into strongly in favour of, strongly against, and neither strongly for nor against, gives just as accurate a picture as the traditional dichotomy of for/against. The three-way split certainly provides better insight into what steers the population’s attitudes towards assisted dying, and what conditions a potential law on assisted dying would have to meet in order to gain support from a majority of the population.

A third of the population is strongly in favour of assisted dying, and is unlikely to be influenced by public discourse or by details in a parliamentary bill. It is conceivable that this category includes people who have had intensely negative experiences as the family of someone who had a difficult death, or people who have a principled position deeply rooted in human rights. A small group, 10%, is strongly against assisted dying, and are unlikely to be swayed into changing their minds. This category includes people whose opposition is rooted in a deeply religious conviction (13), as well as many doctors (5). The final position decided by the large group in the middle could have an impact on potential legislation in the area. We assume that this group includes people who see arguments for and arguments against the legalisation of assisted dying. They are likely to have certain requirements for the design of a potential law on assisted dying, whom it should apply to, and how it should be enforced. This is reflected in the responses to the statements on attitudes towards assisted dying in different situations.

The type of assisted dying matters to the respondents. This is most strongly expressed in response to the statements about assisted dying for patients with mental illness alone, or patients with tiredness of life but no serious illness. Here, 45–50% strongly disagree, while only 3–4% strongly agree that assisted dying should be legalised in these situations. It is therefore mainly in cases of terminal illness with a short remaining life expectancy that Norwegians support the possibility of assisted dying, not in other situations. In light of this, it is interesting to note that existing laws on assisted dying in some countries, such as the Netherlands, allow assisted dying in the case of mental illness alone (14). There is also an ongoing debate in the Netherlands about whether assisted dying should be allowed for older people who have lived what is referred to as a ‘completed life’ (15). In Canada, which introduced assisted dying in 2016, there is now a debate on whether mental illness alone should qualify people for assisted dying (16). It cannot be assumed that assisted dying, if it is legalised, will be limited to people with a somatic disorder with a fatal outcome and a short life expectancy.