Home › Research › Treatments
Potential and biases of large language model simulation for public surveys on Alzheimer's disease therapies
Sato K, Niimi Y, Ihara R, Suzuki K, Iwata A, Iwatsubo T.. Journal of Alzheimer's disease : JAD, 2026-06-27
Topic: Amyloid, tau and blood-test diagnosis
BackgroundWhile large language models (LLMs) have a potential to simulate public-opinion, their reliability for sensitive medical topics like novel Alzheimer's disease (AD) treatments remains unclear.ObjectiveThis study compared LLM-generated and human answers on AD-therapy dilemmas; assessed model and prompting parameter influences; and identified demographic bias.MethodsUsing survey data on late 2023 from 1671 Japanese Trial Ready Cohort Webstudy participants who are presumably cognitively unimpaired, LLM persona profiles guided four LLMs (Gemini-1.5-flash, Gemini-2.0-flash, GPT-4.1-mini, GPT-4o-mini). The models answered a binary question about acceptance towards patient-prioritization or a 5-point Likert question on concern about amyloid-related imaging abnormalities (ARIA) under varied prompt settings. Aggregate similarity was measured with Jensen-Shannon Divergence (JSD) for binary and Earth Mover's Distance (EMD) for Likert scale; while individual agreement used Cohen's κ and Spearman's ρ.ResultsWhile some LLM models achieved fair group-level agreement in both tasks (JSD ≤ 0.05, EMD < 1.0), individual agreement was negligible across any LLM settings (κ, ρ ≈ 0). Adding detailed attributes like living condition, clinical status, or related personal opinions offered limited improvement. Performance was largely stable for most demographic levels, but deteriorated for minority subgroups, such as those with low education or requiring long-term care.ConclusionsOur study demonstrates that current LLMs can approximate aggregate attitudes toward novel AD therapies but cannot predict individual opinions. They can amplify biases in some small subgroups. LLMs may be useful for pre-testing public survey in the field of AD/dementia treatment but should not replace authentic human data.
Cite this study APA, Vancouver, BibTeX · download .bib or .ris
APA 7th
Sato, K., Niimi, Y., Ihara, R., Suzuki, K., Iwata, A., & Iwatsubo, T. (2026). Potential and biases of large language model simulation for public surveys on Alzheimer's disease therapies. Journal of Alzheimer's disease : JAD. https://doi.org/10.1177/13872877261464182
Vancouver
Sato K, Niimi Y, Ihara R, Suzuki K, Iwata A, Iwatsubo T. Potential and biases of large language model simulation for public surveys on Alzheimer's disease therapies. Journal of Alzheimer's disease : JAD. 2026. doi:10.1177/13872877261464182. PMID: 42363810.
BibTeX
@article{sato2026-42363810,
author = {Sato, K. and Niimi, Y. and Ihara, R. and Suzuki, K. and Iwata, A. and Iwatsubo, T.},
title = {{Potential and biases of large language model simulation for public surveys on Alzheimer's disease therapies}},
journal = {Journal of Alzheimer's disease : JAD},
year = {2026},
doi = {10.1177/13872877261464182},
pmid = {42363810},
url = {https://doi.org/10.1177/13872877261464182}
}
Built from the record alzhub holds for this paper, not looked up. Volume, issue and page numbers are not held in the record. Check the citation against the original before you use it.
Key findings
- Aggregate similarity was measured with Jensen-Shannon Divergence (JSD) for binary and Earth Mover's Distance (EMD) for Likert scale; while individual agreement used Cohen's κ and Spearman's ρ.ResultsWhile some LLM models achieved fair group-level agreement in both tasks (JSD ≤ 0.05, EMD < 1.0), individual agreement was negligible across any LLM settings (κ, ρ ≈ 0).
- Adding detailed attributes like living condition, clinical status, or related personal opinions offered limited improvement.
- Performance was largely stable for most demographic levels, but deteriorated for minority subgroups, such as those with low education or requiring long-term care.ConclusionsOur study demonstrates that current LLMs can approximate aggregate attitudes toward novel AD therapies but cannot predict individual opinions.
- They can amplify biases in some small subgroups.
- LLMs may be useful for pre-testing public survey in the field of AD/dementia treatment but should not replace authentic human data.
New to this subject? Start with our plain-English guide: Treatments →
Study details
- Evidence strengthModerate
How this study scored 47/100
Published in a peer-reviewed journal. Scored zero for study design, sample size, effect reported, studied in people and randomised / blinded because the record does not state it. We have no data on journal's citation record, times cited, citations per year, impact for its field and year, cited by studies here, still being cited, senior author's track record, institution's track record and multi-centre collaboration, so those were left out: it earned 27 of the 58 points we could judge it on, which is 47 out of 100.
| Component | Points | Why |
|---|---|---|
| Journal tier | 7 / 10 | Tier 3 — peer-reviewed journal (Journal of Alzheimer's disease : JAD) |
| Journal's citation record | — / 4 | No journal record from OpenAlex — left out of this score |
| Study design | 0 / 12 | No study design recorded — scores zero, not guessed (not stated in the record — scored 0) |
| Strength of evidence | 3 / 5 | Rated “moderate” in our summary |
| Peer-review validation | 10 / 10 | Peer-reviewed. Published as an article in a named peer-reviewed journal. |
| How recent | 5 / 5 | 2026 — within 1 year(s) of the newest study here |
| Sample size | 0 / 4 | Participant count not stated — scores zero (not stated in the record — scored 0) |
| Effect reported | 0 / 3 | No effect size reported — scores zero (not stated in the record — scored 0) |
| Studied in people | 0 / 3 | Not stated whether this was studied in people — scores zero (not stated in the record — scored 0) |
| Randomised / blinded | 0 / 2 | No randomisation or blinding stated — scores zero (not stated in the record — scored 0) |
| Times cited | — / 7 | No citation count (not found in OpenAlex) — left out of this score |
| Citations per year | — / 5 | No citation count or no year — left out of this score |
| Impact for its field and year | — / 6 | No field-weighted citation impact (OpenAlex has none yet) — left out of this score |
| Cited by studies here | — / 6 | Not yet checked against this library's citation graph — left out of this score |
| Still being cited | — / 4 | No yearly citation counts — left out of this score |
| Senior author's track record | — / 4 | Senior author's record not available — left out of this score |
| Institution's track record | — / 3 | No institution listed in OpenAlex — left out of this score |
| Multi-centre collaboration | — / 3 | Institutions not listed — left out of this score |
| Indexed in PubMed | 2 / 2 | Indexed in PubMed (PMID 42363810) |
| Free to read | 0 / 2 | Behind a paywall — scores zero |
| Points | 27 / 58 | Scored on 11 of 20 parts. We have no data for the rest, so they are left out: 27 of the 58 points we could judge it on is 47 out of 100. |
| Total | 47 / 100 | — |
What each of these 20 components measures, and what this score is not →
Read the original
Free copy unknown: no free-copy check has an answer for this record yet. That does not mean it is paywalled.
Related
Studies in the library that share something with this one. The line above each title says what.
Related studies in Treatments
Also tagged “therapies”
5 studies in the library carry the topic tag “therapies”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 5 tagged studies.