Dementia Status Influences End-of-life Care Decisions Despite Advance Directives

TL;DR

For listenersOffer from Amazon

Turn your wind-down time into listening time

• Thousands of audiobooks, podcasts and originals
• Listen on your phone, tablet or Echo — also offline
• Cancel anytime

Try Audible free →Free trial for new members

As an affiliate, we earn on qualifying purchases.Hide

A randomized survey of nearly 6,000 U.S. adults found participants were less likely to recommend life-sustaining treatment for seriously ill patients with dementia, including when an advance directive requested it. Directives influenced recommendations, but dementia status and the decision-maker’s own preferences also mattered.

A survey experiment involving nearly 6,000 U.S. adults found that participants were about 19 percentage points less likely to recommend life-sustaining treatment for a seriously ill patient with dementia than for one without dementia, even when an advance directive requested treatment. The University of Colorado Anschutz study, published in JAMA Network Open, also found that directives and the surrogate decision-maker’s preferences influenced recommendations.

Researchers presented participants with scenarios involving seriously ill, hospitalized older adults and varied whether each patient had dementia, what an advance directive said, and whether a physician recommended treatment. The outcomes measured were participants’ recommendations in the survey; they do not show what care patients actually received in hospitals.

For a patient with dementia and no advance directive, participants recommended life-sustaining treatment in 15.6% of scenarios. That share rose to 41.0% when the directive requested life-sustaining care and fell to 7.6% when it requested comfort-focused care. For patients without dementia, the corresponding shares were 38.9%, 66.3% and 14.4%.

The study’s lead author, Lauren Hersch Nicholas, a professor of medicine in the University of Colorado Anschutz School of Medicine’s Division of Geriatric Medicine, said assumptions about life with dementia may influence how people think about treatment. The findings indicate that a documented preference can shape recommendations without fully determining them.

Why Directives May Not Settle Care

The findings point to a gap that can arise when a patient cannot communicate: a written directive may not fully protect the person’s stated preferences if a surrogate’s views or assumptions about dementia affect recommendations. The study does not establish that such differences lead to different treatment in practice, but it shows that decision-makers in the survey weighed dementia status alongside the documented wishes.

This matters to people planning for future medical care and to those asked to make decisions for them. Nicholas said selecting a surrogate who understands the patient’s values is part of planning. A decision-maker’s own preferences also shaped responses in the experiment, suggesting that naming someone alone may not convey what the patient would want.

The researchers’ results support discussing specific values and care preferences with a chosen decision-maker, rather than treating the existence of a document as the only measure of planning. The study did not test whether conversations change decisions, so that practical effect remains unestablished by this survey.

How the Survey Tested Decisions

An advance directive records a person’s preferences for future medical care if they cannot communicate or make decisions. A surrogate is someone authorized or designated to make health care decisions on a patient’s behalf when the patient cannot do so. The scenarios in this study contrasted directives requesting life-sustaining treatment with those requesting comfort-focused care, as well as scenarios with no directive.

The research, titled Surrogate Decision-Making for Older Adults With and Without Dementia, was published in JAMA Network Open in 2026. Its randomized online design allowed researchers to compare recommendations across scenarios that differed in patient dementia status and other factors. The study reports responses to those hypothetical cases, not a review of actual clinical records or treatment outcomes.

Nicholas said the results suggest advance care planning involves more than completing a form. She described ongoing conversations as a way for a chosen decision-maker to understand the patient’s values and for preferences to be revisited as circumstances change. The source report also cites previous research suggesting that more than two-thirds of older adults may face a situation in which someone else makes end-of-life decisions, while many have not designated a surrogate or documented preferences.

“Having a document that clearly states your wishes mattered. But so did the preferences of the person making the decision.”

— Lauren Hersch Nicholas, study lead author and professor of medicine at the University of Colorado Anschutz School of Medicine

What the Survey Cannot Establish

The results describe recommendations in hypothetical scenarios, not observed treatment decisions, patient outcomes or the frequency of particular choices in hospitals. The study therefore cannot establish how often dementia status changes care in real clinical settings or whether a specific directive would be followed in an individual case.

The source report does not provide enough detail to assess how recommendations varied when a physician advised treatment, or how closely survey participants reflected the full range of people who may serve as surrogates. It also does not establish why participants responded differently to dementia: assumptions about quality of life are offered as a possible influence, not a demonstrated explanation for every decision.

The reported percentages apply to the study’s survey scenarios. They should not be read as predictions of the care any particular patient will receive. The findings identify influences on recommendations, but the relative role of those influences in actual medical decisions remains unclear.

Planning Beyond the Written Directive

The researchers’ stated implication is to strengthen advance care planning through ongoing conversations about a person’s values and treatment preferences, including with the person selected as surrogate. Nicholas said people should revisit those discussions as health and circumstances change, rather than relying only on whether a directive exists.

The study does not announce a new policy or clinical requirement, and the source report gives no timetable for further research. The next practical question is whether clearer conversations or other planning approaches can help surrogates make recommendations that better reflect patients’ wishes, including when dementia is part of the scenario.

Key Questions

What did the study find about dementia and life-sustaining treatment?

In the survey scenarios, participants were about 19 percentage points less likely to recommend life-sustaining treatment for a seriously ill patient with dementia than for one without dementia, even when an advance directive requested treatment.

Did an advance directive affect participants’ recommendations?

Yes. For patients with dementia, recommendations for life-sustaining treatment rose from 15.6% without a directive to 41.0% when a directive requested that treatment. The directive influenced recommendations but did not erase the difference associated with dementia status.

Did the study track actual treatment in hospitals?

No. Researchers examined answers to hypothetical scenarios in an online survey. The study did not measure what care patients received or the outcomes of real clinical decisions.

What did the researchers say people can take from the findings?

Lead author Lauren Hersch Nicholas said advance care planning should include conversations with the chosen surrogate about the patient’s values and wishes, with those discussions revisited as circumstances change. The survey did not test whether those conversations change treatment decisions.

Source: rss