Article Text

Download PDFPDF

Dementia and advance directives: some empirical and normative concerns
  1. Karin R Jongsma,
  2. Marijke C Kars,
  3. Johannes J M van Delden
  1. Julius Center for Health Sciences and Primary Care Utrecht, University Medical Center Utrecht, Utrecht, The Netherlands
  1. Correspondence to Dr Karin R Jongsma, Julius Center, University Medical Center Utrecht, Utrecht University, Utrecht 3508 GA, The Netherlands; K.R.Jongsma{at}


The authors of the paper ‘Advance euthanasia directives: a controversial case and its ethical implications’ articulate concerns and reasons with regard to the conduct of euthanasia in persons with dementia based on advance directives. While we agree on the conclusion that there needs to be more attention for such directives in the preparation phase, we disagree with the reasons provided by the authors to support their conclusions. We will outline two concerns with their reasoning by drawing on empirical research and by providing reasons that contradict their assumptions about competence of people with dementia and the (un)importance of happiness in reasoning about advance directives of people with dementia. We will draw attention to the important normative questions that have been overstepped in their paper, and we will outline why further research is required.

  • dementia
  • euthanasia
  • living wills/advance directives
  • ethics
  • competence/incompetence

Statistics from


The authors of the JME paper ‘Advance euthanasia directives: a controversial case and its ethical implications’ articulate concerns about the conduct of euthanasia in persons with dementia based on advance directives.1 We agree with the authors that the practice of the two described cases are ethically problematic and the clinical use of advance euthanasia directives (AEDs) may be limited in its current form because there is too little attention for such directives in the preparation phase. We do, however, disagree with some arguments provided by the authors to support their conclusions because the arguments are based on partial empirical data and the authors circumvent complicated normative questions by drawing on assumptions about advance directives and dementia. In the following, we will elaborate on two concerns with their arguments: competence and decision-making and the (un)importance of happiness. We will end with arguments why we believe the conclusion that there should be more attention for the drafting and completion of A(E)D in clinical practice.

Competence and decision-making

Miller et al argue that people with dementia are impaired to make decisions:

Even mild cognitive impairment (a condition that precedes Alzheimer’s disease by 3–4 years) can cause significant impairment in decision making, with 40% of patients lacking capacity for research decisions. Mrs A’s symptoms had begun 9 years prior to her death; she was formally diagnosed with Alzheimer’s disease 5 years later. Thus, even at the time of diagnosis, she would have been at high risk of incapacity.

The question whether one is competent or not is an important one for completing an advance directive because it will determine whether the directive has any legal standing. Whether people with dementia are able to draft such directives remains unclear, as aside from the empirical studies referred to by the authors, several empirical studies have shown that patients with dementia are able to actively participate in qualitative studies and can respond to open questions in a meaningful way,2–8 as well that they are able to complete an advance directive in the early phases of dementia.9 10 One can therefore not simply conclude on the basis of the diagnosis of dementia or mild cognitive impairment that the person is not able to complete an A(E)D. Assessing competence is a difficult empirical and normative question, and physicians typically have to provide compelling reasons in order to consider a person incompetent for a certain decision. For the discussed case in the paper of the authors, this means that because of a lack of evidence that Mrs A was competent, it cannot be simply deduced that Mrs A was probably not competent.

The authors continue to question Mrs A’s competence because she has changed the language of her AED and used ‘confusing’ language, and argue that:

there is little evidence that Mrs A actually weighed the relevant pros and cons of requesting future EAS. Mrs A mentions that she saw her mother decline from dementia and did not want to experience a similar fate. Although this is certainly one important form of knowing about dementia, there is no evidence that she incorporated other relevant information into her request, perhaps because she was too impaired (especially 3 years after her diagnosis) to appreciate other relevant information. It is not easy to predict the experience of mental decline or its future effects on one’s quality of life. Personality changes due to dementia will vary, and a person may experience moments of lucidity, showing her ‘old self', even with severe dementia. Anticipating the burden of future health decline is complicated because people underestimate their ability to adjust to their health states after developing disease or disability. An AED written when one is no longer able to imagine and incorporate how one might cope with a given condition has limited validity.

The authors point here to several tricky normative questions that have to do with the ‘then-self versus now-self’ questions as the authors acknowledge and also with regard to what is relevant information for consenting and for completing an AED, who may decide what relevant and sufficient information is and how decisions based on limited information should be dealt with. It is not evident that the directive of Mrs A is based on insufficient information, nor that she has not incorporated relevant information in her decision as written in the directive. Contrary to the authors, one could argue that the fact that Mrs A specified her directive to the case of dementia and reconsidered her earlier advance directive is a sign of reflectiveness about this topic. Furthermore, it is also not evident that an advance directive that is based on insufficient information should be ignored altogether. Moreover, the authors assume that the changing behaviour of the person with dementia should be understood to be an ‘ability to adjust’. While it is often observed that people with dementia change their behaviour, it remains contested whether this should be regarded as an adaptation to the disease. Even more contested is what value such ‘adapted’ preferences should be given.11 The authors assume that such change emerge due to adaptation to the disease. However, patients with dementia do not choose what they forget. There is a difference between ‘choosing’ and ‘losing’ abilities and preferences. Even though a patient with dementia might seem to behave similar to a person who has chosen to drop his former beliefs, it is highly questionable whether the patient with dementia and the person who has willingly dropped his beliefs should also be treated in the same manner.12 We have no reason to follow a prior preference that was chosen to be dropped, but it is less clear that the same thing can be said of a preference that is lost due to dementia. The changed behaviour is most likely caused by the loss of the ability to hold on to prior preferences, not by a (sub)conscious choice to change or adapt. Changes in behaviour or a change of preference by patients with dementia may be caused by the disease itself, rather than by a conscious adjustment to the disease.

(Un)importance of happiness

Another argument brought forward by the authors is that patients with dementia may have a wrong image of what dementia is and whether they will suffer.

Additionally, this lack of information may be reinforced by prejudicial attitudes towards disease and disability, with images of incontinence, confusion, agitation and loss of mobility, rather than a more balanced picture incorporating the pleasures and enjoyment that many patients experience.

This argument assumes that you cannot know whether you will be unhappy as a patient with dementia or whether you will still enjoy your life and forget about the fears you had about a future of living with dementia. Because we do not know how our life with dementia will be, we should not bind ourselves with advance directives. Also here the empirical data brought forward by the authors are partial; some studies report annoyance, frustration and boredom.13 14 There is, however, only scarce empirical research conducted into happiness of patients with dementia, due to the lack of reliable and valid ways to measure happiness,15 and happiness is sometimes conflated with well-being or quality of life16 Therefore, we do not know whether patients with dementia who seem happy are indeed happy, and whether patients with dementia who seem unhappy are indeed unhappy.

The authors continue to argue about the relevance of well-being and suffering with the following:

Although it is true that Mrs A suffered daily periods of distress, she also had daily periods in which she appeared happy and content. Every human life has its ups and downs. When do the burdens of life with dementia become sufficiently heavy to justify a finding of unbearable and irremediable suffering? At minimum, Mrs A’s case points to a need for a more thorough examination of this question.

Also, this paragraph raises several normative questions with regard to the importance of happiness and well-being for the authority of advance directives. Arguably, a person drawing an advance directive has foreseen that he might not be able to live according to his preferences and writes the advance directive because he wants to make sure he will still be treated according to these preferences and not according to his later wishes, even if he (seems) happy. Importantly, due to dementia, the person cannot care about his prior preferences anymore.17 The point here is not that the patient with dementia does no longer care, but that he is no longer capable of caring about his prior values. Whether that also should imply that his former values do not matter anymore when these are written in an advance directive is not evident. It is therefore not obvious that in all situations and for all people, current well-being should always take precedence over all other values. A patient with dementia in the more advanced stages cannot live according to his prior values anymore, and it is not evident that only well-being should matter to him.

Need for better integration in clinical care and ethical reflection

Many of the open questions as discussed in this response demand more research to better understand the particularities of dementia and to improve the use of advance directives and AEDs. We agree with the authors that there should be more attention for the drafting and completion of advance directives in clinical practice. Advance directives are difficult to interpret.18 There are many open questions remaining to how advance directives can be used in practice, how they can be interpreted well and what the value of such directives should be when the person utters different wishes when he is demented. A better integration of such directives in the clinical practice, where physicians help to draft and understand the values and wishes of the patient. Moreover, there is a need for empirical and normative research into why patients draft the directive in the first place, how patients understand and values dementia, as well as how ‘crucial’ the interests are that are written in the advance directive. This may provide guidance to physicians when dealing with such tricky questions with persons who are no longer able to care about their prior preferences.



  • Contributors All three authors have contributed to qualify for authorship: substantial contributions to the conception of the paper; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles