One question, yes or no: "that in the future you may be a caregiver for a person with Alzheimer's disease?" It sits in a survey the way it sits in a family, quietly, between the coffee and the car keys. In a study published in The Journals of Gerontology: Series B, Shana Stites and colleagues at the University of Pennsylvania and NYU put that question to 3,527 adults in the United States, and about 31 percent, 1,090 people, said yes. They are not caregivers. They are people who can see the role coming, and the researchers wanted to know what the view from there looks like.
The design is a randomized vignette experiment, which means it measures beliefs, not behaviour. Every participant read a short story about a fictional person seen at a memory center, with one detail flipped at random: a positive or a negative Alzheimer's biomarker test result. Everyone then completed a modified Family Stigma in Alzheimer's Disease Scale, which sorts reactions into seven domains including expectations of structural discrimination, symptom severity, negative aesthetic attribution, antipathy, social distance, pity and support. Participants who failed a comprehension screener on biomarker results, 286 of them, were excluded before analysis. This was a single sitting, not a study that followed anyone over time.
The split in the results is the story. People who had actually cared for someone with Alzheimer's dementia reported higher stigma than people with no caregiving experience at all on five of the seven domains: they expected the person in the vignette to meet structural discrimination (odds ratio 1.66), to have more severe symptoms (2.60), to be seen through a worse aesthetic lens (3.82), to draw antipathy (2.24) and to be kept at a social distance (2.14). On pity and support, the two domains you might expect experience to soften or sharpen most, they did not differ from everyone else. Expectation ran the other way. People who expected to become an Alzheimer's caregiver, compared with those who did not, reported stronger pity (1.65) and stronger support (1.74), and it was those warmer domains, not the harsher ones, that marked them out.
That asymmetry matters because the timing of the whole conversation is shifting. In May 2025 the US Food and Drug Administration cleared the first blood test used as a diagnostic aid for amyloid pathology, the Lumipulse plasma pTau217 to beta-amyloid 1-42 ratio, for symptomatic adults aged 55 and older, as summarised in a peer-reviewed commentary in Drug Discovery and Therapeutics. A test result now lands years before the crisis does, and it lands on a whole family. Related work from the same group, an online vignette study of 3,548 adults published in Alzheimer's and Dementia: Diagnosis, Assessment and Disease Monitoring in October 2025, found that adding a brain scan to a diagnostic evaluation raised confidence in an Alzheimer's diagnosis across race groups, though Black participants showed the smallest increase, 11.86 points. Certainty arrives early. So does everything that travels with it.
For a family reading this at a kitchen table, the practical reading is modest and specific. If you are the relative who assumes the role is coming, your instinct is likely to be protective rather than distancing, and that is worth knowing about yourself. If you have already done the caregiving, this study suggests something harder: that watching how the world treats a person with dementia can leave you expecting the worst of that world, and expecting more decline than the person in front of you may be showing. The authors frame it as clinical, not personal, noting that understanding these beliefs "may help guide clinical discussions with family members of persons with AD". None of it tells any individual reader what their own family will feel, or what to do about it. That conversation belongs in the clinic room.