Legal review maps the world's assisted-dying laws: dementia alone almost never qualifies

A new comparative analysis of every jurisdiction where voluntary assisted dying is lawful finds that dementia, on its own, rarely meets the eligibility bar, even where advance-request laws exist on paper.

A peer reviewed comparative legal analysis published 29 September 2026 in the Journal of Law and the Biosciences has mapped voluntary assisted dying (VAD) laws, policies and guidelines across every jurisdiction where VAD is lawful, and concludes that a diagnosis of dementia alone rarely makes a person eligible. The authors, Casey Haining, Lindy Willmott, Katrine Del Villar and Ben White of the Australian Centre for Health Law Research at Queensland University of Technology, sort existing regimes into two categories: concurrent request models, which require a person to hold decision making capacity at the time VAD is requested and administered, and advance request models, which allow a person to specify in advance the conditions under which they would want VAD once capacity is lost. The paper finds that in all United States and Australian jurisdictions where VAD is lawful, as well as New Zealand, concurrent request rules effectively exclude dementia as a qualifying condition on its own, because the disease progressively removes the capacity the law demands at the point of access.

Where advance request mechanisms do exist, such as in the Netherlands, the review describes a narrower and more conditional pathway: a person must set out triggering conditions in advance, and it then falls to family members and treating doctors to judge, later, whether those conditions have been met in someone who can no longer confirm the decision themselves. The analysis also identifies jurisdictions, including Spain and Austria, that allow access on a non terminal, chronic condition basis but impose an added impairment threshold specifically for conditions like dementia, meaning a person must be more advanced in the illness before qualifying than other applicants. The authors note this layering of extra conditions, on top of general eligibility criteria, as a key regulatory feature across the jurisdictions reviewed.

The review stops short of recommending reform; its authors describe their aim as cataloguing how existing frameworks operate and flagging what that means for access and clinical practice, not prescribing which model is right. For families who have already lived with the uncertainty described in previous reporting on people navigating a dementia diagnosis and conversations about end of life choices, this analysis supplies a concrete regulatory picture: the gap between having a law that mentions advance requests and having a law that someone with dementia can actually use is, on this evidence, still wide in nearly every jurisdiction examined.

The authors say the categorisation has implications both for eligibility debates and for how clinicians and families might expect existing advance request systems to work in practice, without taking a position on whether the criteria should change.