As Alzheimer’s disease prevention studies are seeking cognitively healthy but biomarker-positive volunteers, researchers must grapple with ethical implications of divulging amyloid status to symptom-free people. What is the psychological impact? How might knowing one’s dementia risk change behavior? A new study investigating those questions suggests that, when it comes to attitudes about physician-assisted suicide, most people facing new knowledge about their heightened risk of dementia don’t develop a sudden urge to plan to end their lives.

  • Some healthy, amyloid-PET positive people mentioned interest in physician-assisted death.
  • A minority considered PAD as a future possibility.
  • Most did not know PAD is illegal for people with AD in U.S.

In the work, Emily Largent, Jason Karlawish, and colleagues at the University of Pennsylvania, Philadelphia, asked amyloid PET-positive and -negative adults for their views on physician-assisted death (PAD). A majority of people reported having personal, philosophical, or religious beliefs that precluded PAD, and learning their amyloid status did not change their minds. One-fifth of amyloid-positive people said that they would like the opportunity to choose physician-assisted death in the future if they became mentally impaired, were suffering, or became a burden to their loved ones. Few seemed aware that, in the U.S., physican-assisted dying is illegal for people with dementia. The study appeared April 29 in JAMA Neurology.

“Our results suggest the revelation of amyloid positivity did not change attitudes toward PAD—if someone hadn’t considered it a possibility before, learning their amyloid status did not make them think it was now. We don’t see a sudden revision in long-held beliefs,” Largent said. Instead, people who were open to PAD saw their amyloid status as a relevant piece of information to consider when thinking about end-of-life wishes, she told Alzforum. “We hope this information will help inform the current discussion on expanding access to PAD,” she said.

The study indicates that people did not see the positive biomarker as an imminent threat, said Rebecca Dresser, Washington University, St. Louis. “The absence of panic is reassuring,” she wrote to Alzforum. “The interview report highlights the need to ensure that individuals with elevated levels understand the uncertain clinical implications of that finding. We don’t want people to leap to conclusions about their prospects. It would also be good to offer such individuals information about advance planning and end-of-life options (including the eligibility requirements for PAD).”

In physician-assisted death, also called aid in dying, a doctor prescribes a lethal dose of barbiturates, which patients self-administer. This differs from euthanasia, where a doctor causes death, for example by lethal injection. PAD is legal in eight U.S. states and the District of Columbia, as well as in Canada, Colombia, Belgium, the Netherlands, Luxembourg, Switzerland, and parts of Australia. In most countries, people with dementia cannot seek physician-assisted death, and those with slowly developing neurodegenerative diseases are also excluded. Oregon is debating expanding access to people with more conditions (Jefferson Public Radio, 2019). 

The current findings are part of the Study of Knowledge and Reactions to Amyloid Testing (SOKRATES), in which Karlawish and colleagues are tracking attitudes and behaviors of people who learn they have a positive amyloid PET scan and thus are at high risk of developing dementia (Nov 2017 news). The investigators recruited 50 amyloid-positive people, half men, half women, from the Asymptomatic Alzheimer’s A4 prevention study and 30 amyloid-negative people from the Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) study, an observational adjunct to A4. They interviewed all participants four to 12 weeks after, and again one year after they received their PET results, probing their symptoms, feelings, and plans for the future by way of structured interviews. Eighty percent of the amyloid-positive participants had a family history of AD.

While the scientists have yet to publish the complete results of SOKRATES, snippets of data presented at meetings suggest that learning their amyloid PET status hits people differently than other medical tests, Karlawish told Alzforum. “For a cognitively unimpaired older adult, an amyloid imaging result is significant news. It has a real impact on how they think about themselves, how they think other people think about them, and the plans they think about for the future,” he said.

What about planning their own death? The initial post-disclosure interview did not ask about PAD, yet some people spontaneously mentioned they were interested in aid in dying. In the second round of interviews, one year later, the team decided to ask participants if they had been thinking about physcian-assisted death or other ways to end their lives.

Most of the interviewees were familiar with physician-assisted death. Two-thirds of amyloid-positive people said they would not consider PAD for themselves, and their stance stayed the same before and after learning their amyloid status. Several respondents were ambivalent about PAD, and some had made efforts to learn more about it. One in five of the PET-positive participants said they would consider it once symptoms set in. These people were likelier than other participants to have already begun other preparations for the future, such as financial or legal planning.

In the amyloid-negative group, about the same proportion reported they would have considered pursuing physician-assisted death or suicide had their scan been positive. This percentage is about the same as seen for other diseases, Largent told Alzforum.

Even those who ruled out aid in dying for themselves believed it was a personal choice, which could be right for others in a similar situation. To Karlawish’s mind, this suggests a growing societal acceptance of aid in dying for people who want it.

The study has caveats: The number of people surveyed was small and not representative of the general population. Joshua Grill, University of California, Irvine, thinks it will be important to study this in a more generalizable population. “Given the wide-scale availability of biomarker and genetic tests, and especially in the absence of effective therapies, a major concern is the impact disclosure might have on people who learn this information without adequate education and counseling,” he wrote to Alzforum. Grill frequently works with the Pennsylvania group, but was not involved in this study.

Brian Draper, University of New South Wales, Australia, is unsurprised that people who seek out dementia assessment already have a clear idea about their end-of-life plans, and that having biomarker results does not change that. He cautions not everyone may take the news as well, writing, “It does suggest that clinicians might need to consider exploring end-of-life plans in the initial assessment for patients who self-refer, in case there is suicide risk pending.”

Many A4 participants were aware of geographical restrictions to PAD in the U.S. Karlawish told Alzforum one participant talked about “going to Oregon,” when the time came, referencing the state that enacted the first right-to-die law back in 1997. At the same time, few participants knew that under current U.S. law, people with AD cannot legally seek physician-assisted death in any state. Requestors must have a terminal illness and be within six months of death, and have the mental capacity to make two successive written requests for the lethal prescription, then fill it and ingest it themselves. People with AD who are mentally competent to seek PAD are years from death, while those in the terminal stage cannot make the decision.

This differs in the Netherlands, where people need not be near death to request physician-assisted suicide or even euthanasia. Philip Scheltens of VU University in Amsterdam faces the issue of aid in dying regularly in the memory clinic. “We use amyloid biomarkers, either CSF or PET, routinely in diagnosis, and we do have patients who are functioning completely normally, have subjective complaints, and have abnormal biomarkers. We see them every six months, and when they begin to deteriorate, they want to pursue PAD, and we always discuss this,” he told Alzforum. He said this is consistent with the SOKRATES findings, where people said they’d seek aid in dying once cognitive difficulties emerged.

The Netherlands have strict legal criteria and a rigorous procedure for physician-assisted death. Patients must have a definitive diagnosis of an untreatable disease that will lead to inevitable suffering, as affirmed by two other consulting physicians. But they must also be of sound mind and adamant about the desire to die. Scheltens himself is not involved; the ultimate authority is the person’s general practitioner.

Most dementia patients who end their lives through euthanasia or PAD in the Netherlands, do it in the early stages of disease (Jan 2019 BBC report). While many write advanced directives specifying when they’d like to die, such as when they have to go into a nursing home, or when they no longer recognize their children, by the time that day arrives the patient has lost the capacity to make the decision. The Dutch talk about “the five to midnight or five past midnight” dilemma—people don’t want to end their lives at 11:55 p.m., while they are still aware, but if they wait until 12:05 a.m., it’s too late for an autonomous decision. “A lot of people end up in that situation. If you want to do it properly, you have to do it at five before 12,” Scheltens said. “I warn patients that the consequence of that is that you have to do it earlier than you might expect.”

For many, the sense of autonomy and control that comes with access to aid in dying appears to be enough. In the U.S., experience has shown that among terminal patients who qualify for a lethal prescription, one-quarter to one-half never use it.—Pat McCaffrey


  1. This short report represents a critically important topic for the field. Previous surveys and hypothetical studies have found that approximately 10 percent of cognitively unimpaired older adults would want to learn AD genetic or biomarker information to instruct planning suicide (Ott et al., 2016Caselli et al., 2014). Here, Dr. Largent and colleagues present real-world data from participants who have learned their biomarker result through participation in a preclinical AD trial. Approximately 20 percent of people who learned they had “elevated” brain amyloid indicated that they would consider physician-assisted suicide, a notable increase from the hypothetical studies, though the sample size is relatively small.

    Importantly, the experiences of these individuals may represent a “best-case scenario,” since they underwent structured education and counseling before and after learning their biomarker result (Harkins et al., 2015). Given the wide-scale availability of biomarker and genetic tests, and especially in the absence of effective therapies, a major concern is the impact disclosure might have on people who learn this information without adequate education and counseling.


    . A Survey of Knowledge and Views Concerning Genetic and Amyloid PET Status Disclosure. Alzheimers Dement (N Y). 2016 Jan 1;2(1):23-29. PubMed.

    . Public perceptions of presymptomatic testing for Alzheimer disease. Mayo Clin Proc. 2014 Oct;89(10):1389-96. Epub 2014 Aug 26 PubMed.

    . Development of a process to disclose amyloid imaging results to cognitively normal older adult research participants. Alzheimers Res Ther. 2015;7(1):26. Epub 2015 May 12 PubMed.

  2. Bearing in mind that the population is people who are enrolled in clinical trials for early diagnosis/treatment of AD, either because they are biologically at high risk of AD or have been interested enough about their cognitive health to participate in such a study, the findings don't surprise me. I think the two most important findings are that people who self-present for dementia assessment have already got a good idea about their end-of-life plans and this doesn't seem to alter with investigation results. Secondly, there is a broad range of end-of-life plans across the whole spectrum.

    While of course the work needs replication in larger, more representative populations, it does suggest that clinicians might need to consider exploring the end-of-life plans of patients who self-refer in the initial assessment in case there is suicide risk pending.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Among the Healthy, a Positive Amyloid Scan Prompts Questions About Risk

External Citations

  1. Jefferson Public Radio, 2019
  2. Jan 2019 BBC report

Further Reading

Primary Papers

  1. . Attitudes Toward Physician-Assisted Death From Individuals Who Learn They Have an Alzheimer Disease Biomarker. JAMA Neurol. 2019 Jul 1;76(7):864-866. PubMed.