Skip to content

Clinical practice guidelines registry — Alzheimer's disease

Last curated: 2026-09-03 (evidence-breadth deepening)

Purpose. This registry catalogues the documents that define, diagnose, test for and treat Alzheimer's disease, by lineage. It records documents — scope, provenance, unique contribution and replacement status. The synthesis of what they recommend, and where they disagree, lives in ../../wiki/guidelines.md. current means the newest located document in its lineage as of the curation date, not that every recommendation incorporates evidence published up to that date.

Every PMID in the 2026-09-03 additions was retrieved through live PubMed E-utilities in that session; earlier records retain their documented 2026-08-31 retrieval provenance. Web-only documents were not reinterpreted in this pass.

Status key. current = newest document in that body/scope lineage. current (aging) = newest located, but its evidence base predates a material development (typically plasma biomarkers, tau PET or disease-modifying therapy). superseded by → = formally or effectively replaced by the named document. retired = marked retired by the issuing body. adjacent = focused document that complements rather than replaces a broad guideline. parallel = a competing document from a different body covering the same scope without replacing it.


1. Master table

# Body Year Region Scope Citation + PMID / URL Status
1 NINCDS–ADRDA 1984 USA / international Clinical diagnosis of AD (probable / possible / definite) McKhann G, et al. Neurology. 1984;34:939-44. PMID 6610841 superseded by → #3
2 NIA-AA 2011 USA / international Introduction to the three-workgroup diagnostic recommendations Jack CR, et al. Alzheimers Dement. 2011;7:257-62. PMID 21514247 framework document; superseded in principle by → #6
3 NIA-AA 2011 USA / international AD dementia McKhann GM, et al. Alzheimers Dement. 2011;7:263-9. PMID 21514250 superseded by → #6
4 NIA-AA 2011 USA / international MCI due to AD Albert MS, et al. Alzheimers Dement. 2011;7:270-9. PMID 21514249 superseded by → #6
5 NIA-AA 2011 USA / international Preclinical AD (research only) Sperling RA, et al. Alzheimers Dement. 2011;7:280-92. PMID 21514248 superseded by → #6
6 NIA-AA 2018 USA / international Research framework; AT(N) biological definition Jack CR, et al. Alzheimers Dement. 2018;14:535-562. PMID 29653606 superseded by → #7 for diagnosis; still used as research vocabulary
7 Alzheimer's Association Workgroup 2024 USA / international Revised criteria for diagnosis and staging; Core 1 / Core 2 Jack CR, et al. Alzheimers Dement. 2024;20:5143-5169. PMID 38934362 current
8 International Working Group (IWG) 2014 International IWG-2 research diagnostic criteria Dubois B, et al. Lancet Neurol. 2014;13:614-29. PMID 24849862 superseded by → #9
9 International Working Group (IWG) 2024 International AD as a clinical-biological construct Dubois B, et al. JAMA Neurol. 2024;81:1304-1311. PMID 39483064 current; parallel to #7 and in explicit disagreement with it
10 Alzheimer's Association (DETeCD-ADRD) 2025 USA Diagnostic evaluation, testing, counselling and disclosure — primary care Atri A, et al. Alzheimers Dement. 2025;21:e14333. PMID 39713942 current
11 Alzheimer's Association (DETeCD-ADRD) 2025 USA Same — specialty care Dickerson BC, et al. Alzheimers Dement. 2025;21:e14337. PMID 39713957 current
12 Alzheimer's Association (DETeCD-ADRD) 2025 USA Validated clinical assessment instruments Atri A, et al. Alzheimers Dement. 2025;21:e14335. PMID 39713939 adjacent; current
13 CCCDTD5 2020 Canada Diagnosis and treatment of dementia, 8 topic areas Ismail Z, et al. Alzheimers Dement. 2020;16:1182-1195. PMID 32725777 current (aging)
14 CCCDTD5 2020 Canada Neuroimaging and fluid biomarkers Brisson M, et al. Alzheimers Dement (N Y). 2020;6:e12098. PMID 33532543 adjacent; current (aging)
15 CCCDTD5 2020 Canada Vascular cognitive impairment management Smith EE, et al. Alzheimers Dement (N Y). 2020;6:e12056. PMID 33209971 adjacent; also registered under vascular dementia
16 CCCDTD5 2022 Canada Deprescribing cognitive enhancers Herrmann N, et al. Alzheimers Dement (N Y). 2022;8:e12099. PMID 35128025 adjacent; current
17 American Academy of Neurology 2018 USA Mild cognitive impairment practice guideline update Petersen RC, et al. Neurology. 2018;90:126-135. PMID 29282327 retired (so marked in PubMed); no located replacement
18 EFNS 2010 Europe Diagnosis and management of AD Hort J, et al. Eur J Neurol. 2010;17:1236-48. PMID 20831773 current (aging) — predates plasma biomarkers, tau PET and DMTs
19 EFNS task force 2012 Europe Neuroimaging in the diagnosis of dementia Filippi M, et al. Eur J Neurol. 2012;19:e131-40, 1487-501. PMID 22900895 adjacent; current (aging)
20 Alzheimer's Association / SNMMI 2013 USA / international Appropriate use criteria, amyloid PET (J Nucl Med version) Johnson KA, et al. J Nucl Med. 2013;54:476-90. PMID 23359661 superseded by → #24
21 Alzheimer's Association / SNMMI 2013 USA / international Same document, Alzheimers Dement version Johnson KA, et al. Alzheimers Dement. 2013;9:e-1-16. PMID 23360977 superseded by → #24
22 Amyloid Imaging Task Force 2013 USA / international Update to the 2013 AUC (J Nucl Med version) Johnson KA, et al. J Nucl Med. 2013;54:1011-3. PMID 23753186 superseded by → #24
23 Amyloid Imaging Task Force 2013 USA / international Same update, Alzheimers Dement version Johnson KA, et al. Alzheimers Dement. 2013;9:e106-9. PMID 23809369 superseded by → #24
24 Alzheimer's Association / SNMMI 2025 USA / international Updated AUC for amyloid and tau PET (Alzheimers Dement version) Rabinovici GD, et al. Alzheimers Dement. 2025;21:e14338. PMID 39776249 current
25 Alzheimer's Association / SNMMI 2025 USA / international Same document, J Nucl Med version Rabinovici GD, et al. J Nucl Med. 2025;66:S5-S31. PMID 39778970 current (dual publication)
26 Alzheimer's Association 2022 International Appropriate use recommendations, blood-based biomarkers Hansson O, et al. Alzheimers Dement. 2022;18:2669-2686. PMID 35908251 superseded by → #27 for specialised care
27 Alzheimer's Association 2025 International Clinical practice guideline, blood-based biomarkers in specialised care (GRADE) Palmqvist S, et al. Alzheimers Dement. 2025;21:e70535. PMID 40729527 current
28 AD and Related Disorders Therapeutic Workgroup 2023 USA / international Lecanemab appropriate use recommendations Cummings J, et al. J Prev Alzheimers Dis. 2023;10:362-377. PMID 37357276 current
29 AD and Related Disorders Therapeutic Workgroup 2025 USA / international Donanemab appropriate use recommendations Rabinovici GD, et al. J Prev Alzheimers Dis. 2025;12:100150. PMID 40155270 current
30 NIA-AA 2012 USA / international Neuropathologic assessment of AD (practical approach) — ABC score Montine TJ, et al. Acta Neuropathol. 2012;123:1-11. PMID 22101365 current
31 NIA-AA 2012 USA / international Neuropathologic assessment of AD (companion) Hyman BT, et al. Alzheimers Dement. 2012;8:1-13. PMID 22265587 current (dual publication)
32 WHO 2019 Global Risk reduction of cognitive decline and dementia WHO, https://www.who.int/publications/i/item/9789241550543, accessed 2026-08-31; summarised in Chowdhary N, et al. Front Neurol. 2021;12:765584. PMID 35082745 current
33 NICE (NG97) 2018, reviewed 2025 England Dementia: assessment, management and support https://www.nice.org.uk/guidance/ng97 current web guideline; reviewed 24 Oct 2025 with a decision not to update recommendations; minor amendments Apr and Jun 2025
34 NICE 2024–2026 appraisal cycle England Lecanemab and donanemab for MCI/mild dementia due to AD https://www.nice.org.uk/guidance/indevelopment/gid-ta11220; https://www.nice.org.uk/guidance/indevelopment/gid-ta11221/documents in development; expected publication TBC; latest drafts remain negative after appeal/reconsultation
35 EMA 2024–2025 European Union Leqembi (lecanemab) marketing authorisation https://www.ema.europa.eu/en/medicines/human/EPAR/leqembi authorised 15 April 2025 (opinion adopted 14 Nov 2024), restricted to ApoE ε4 non-carriers and heterozygotes
36 FDA 2023–2026 USA Leqembi (lecanemab) and Kisunla (donanemab) approvals, labelling and delivery routes https://www.accessdata.fda.gov/drugsatfda_docs/label/2023/761269Orig1s001lbl.pdf; https://www.fda.gov/drugs/news-events-human-drugs/fda-approves-first-home-starting-dose-alzheimers-disease-treatment current; at-home subcutaneous lecanemab starting regimen approved 13 Jul 2026
37 EMA 2025 European Union Kisunla (donanemab) marketing authorisation https://www.ema.europa.eu/en/medicines/human/EPAR/kisunla authorised 24 September 2025, restricted to ApoE ε4 non-carriers and heterozygotes
38 Spanish Society of Neurology 2025 Spain Blood-based biomarker positioning and usage recommendations García-Ribas G, et al. Neurologia (Engl Ed). 2025. PMID 40685136 current; recommends against asymptomatic screening and direct-to-consumer testing
39 American Psychiatric Association 2016 USA Antipsychotics for agitation or psychosis in patients with dementia Reus VI, et al. Am J Psychiatry. 2016;173:543-6. PMID 27133416 current; dedicated symptom-specific document; PubMed record has no abstract, so class/level tables are not reproduced here
40 EFNS–ENS/EAN 2015 Europe Cholinesterase inhibitor plus memantine in moderate-to-severe AD Schmidt R, et al. Eur J Neurol. 2015;22:889-98. PMID 25808982 current (aging); weak GRADE recommendation
41 Korean Dementia Association 2025 South Korea Cholinesterase inhibitors and memantine Kim Y, et al. Dement Neurocogn Disord. 2025;24:1-23. PMID 39944527 current; strong recommendations based on moderate evidence for AD
42 Korean Dementia Association 2025 South Korea Pharmacological treatment of behavioural and psychological symptoms Byeon G, et al. Dement Neurocogn Disord. 2025;24:24-43. PMID 39944528 current; conditional recommendation for antipsychotics; citalopram advised for agitation without a stated grade in the record

Coverage count. 42 documents or maintained sets: 36 journal-indexed records (of which 12 are superseded and 1 retired) plus 6 web-based guideline, regulatory or reimbursement records. Rows count documents, not individual recommendations; dual publications of the same document (rows 20/21, 22/23, 24/25, 30/31) are listed separately because both are citable.


2. Guideline-family entries

2.1 The disease-definition lineage: 1984 → 2011 → 2018 → 2024, with a parallel IWG track

  • 1984 NINCDS-ADRDA (PMID 6610841). Clinical criteria only; states explicitly that the diagnosis cannot be determined by laboratory tests, which serve to exclude other causes. Probable / possible / definite, with definite requiring histopathology.
  • 2011 NIA-AA (PMID 21514247; 21514250; 21514249; 21514248). Three clinical stages, each with its own document. Biomarkers raise or lower diagnostic certainty; MCI criteria carry four levels of certainty. The preclinical document states that its recommendations are "solely intended for research purposes and do not have any clinical implications at this time" — the firewall the 2024 revision removes.
  • 2018 NIA-AA research framework (PMID 29653606). AT(N) classification; disease defined by pathologic processes documented by biomarkers or autopsy, not by symptoms. States that using the framework in general medical practice is "premature and inappropriate."
  • 2024 Alzheimer's Association revision (PMID 38934362). Core 1 sufficient for diagnosis; Core 2 prognostic; integrated biological-and-clinical staging that accommodates copathology and reserve. Deliberately not a workflow guideline.
  • IWG track (PMID 24849862; PMID 39483064). IWG-2 required an appropriate clinical phenotype plus a pathophysiological biomarker. The 2024 IWG recommendation restates that requirement against the AA revision, arguing that most biomarker-positive cognitively normal individuals will not become symptomatic on a proximate timeline and should be described as at risk for AD.

Version hazard. Papers published between 2018 and 2024 that say "AD" may mean AT(N)-defined biology, syndromic AD, or IWG clinical-biological AD. Prevalence and progression estimates are not comparable across these definitions: in the same 259 cognitively unimpaired 70-year-olds, "preclinical AD" prevalence ranged from 9.7% to 22.8% depending on the scheme (Kern 2018, PMID 29653987).

2.2 Diagnostic-evaluation guidelines

  • DETeCD-ADRD (2025), three companion papers (PMID 39713942 primary care; PMID 39713957 specialty care; PMID 39713939 instruments). Developed by modified Delphi from 7,374 publications screened to 133 included. First US diagnostic-evaluation guideline in two decades and the first written to be usable in primary care. Conditions its own claim on health systems providing adequate resources.
  • CCCDTD5 (2020–2022) (PMID 32725777 main; PMID 33532543 imaging/fluid biomarkers; PMID 33209971 vascular cognitive impairment; PMID 35128025 deprescribing). Notable for being the only current family with a dedicated deprescribing document.
  • AAN MCI guideline (2018) (PMID 29282327). Retired in PubMed with no located replacement, which leaves the largest specialty neurology body in the US without a current MCI guideline in the era of plasma biomarkers and anti-amyloid therapy. This is recorded here as a gap, not an oversight of this registry.
  • EFNS (2010, 2012) (PMID 20831773; PMID 22900895). The oldest documents in current use; they predate plasma p-tau217, tau PET and disease-modifying therapy entirely.
  • NICE NG97 (web). Published 20 June 2018; reviewed 24 October 2025 with a decision not to update the recommendations; minor amendments in April 2025 (updating links to MHRA safety advice in recommendation 1.7.10) and June 2025 (recommendation 1.8.11 on eye tests). Guidance page directly fetched 2026-08-31.

2.3 Test-specific appropriate-use documents

  • Amyloid PET AUC 2013 → amyloid and tau PET AUC 2025. The 2013 documents (PMID 23359661 / 23360977, plus the update PMID 23753186 / 23809369) restricted appropriate use to objective cognitive impairment with genuine aetiologic uncertainty and noted the absence of evidence on clinical outcomes. The 2025 update (PMID 39776249 / 39778970) covers 17 scenarios: amyloid PET 7 appropriate / 2 uncertain / 8 rarely appropriate; tau PET 5 / 6 / 6.
  • Blood biomarkers 2022 → 2025. The 2022 appropriate use recommendations (PMID 35908251) permitted use only as pre-screeners with PET/CSF confirmation and advised against stand-alone diagnostic or primary-care use. The 2025 clinical practice guideline (PMID 40729527) is GRADE-based and performance-defined: ≥90% sensitivity and ≥75% specificity for triage; ≥90% sensitivity and ≥90% specificity to substitute for amyloid PET or CSF, in specialised care, in people with objective cognitive impairment. It cautions that many commercially available tests do not meet these thresholds.

2.4 Drug-specific appropriate-use recommendations

  • Lecanemab AUR (2023) (PMID 37357276): MCI or mild AD dementia with confirmed amyloid; APOE genotyping recommended; baseline MRI and institutional preparedness mandatory; patients requiring anticoagulation should not receive lecanemab until more data are available.
  • Donanemab AUR (2025) (PMID 40155270): Clinical Stages 3–4, MMSE 20–30, PET or CSF confirmation, tau PET not required; APOE genotyping should be performed; MRI within 12 months, excluding >4 microbleeds, cortical superficial siderosis or major vascular contribution; surveillance MRI before infusions 2, 3, 4 and 7, before dose 12 in higher-risk patients, and whenever ARIA is suspected; discontinuation may be considered once amyloid clearance is demonstrated, typically at 12–18 months.

2.5 Neuropathologic assessment

  • NIA-AA 2012 ABC score (PMID 22101365 practical approach; PMID 22265587 companion). Combines Thal amyloid phase (A), Braak tangle stage (B) and CERAD neuritic plaque score (C); recognises AD neuropathologic change without cognitive impairment; requires structured assessment of Lewy body disease, vascular brain injury, hippocampal sclerosis and TDP-43 inclusions; specifies minimum brain sampling and staining methods. No successor located.

2.6 Prevention

  • WHO 2019 guidelines on risk reduction of cognitive decline and dementia (web; summarised PMID 35082745), organised into lifestyle and behaviour interventions, interventions for physical health conditions, and specific interventions, developed under Action Area 3 of the Global Action Plan on the Public Health Response to Dementia 2017–2025.

2.7 Regulatory and reimbursement decisions

  • FDA: lecanemab accelerated approval January 2023, converted to traditional approval July 2023; donanemab (Kisunla) approved 2 July 2024. A subcutaneous lecanemab maintenance regimen was approved in 2025 and an at-home subcutaneous starting regimen on 13 July 2026. The delivery approvals do not remove diagnostic-confirmation or ARIA-monitoring requirements. Current FDA pages directly fetched 2026-08-31.
  • EMA: Leqembi was authorised 15 April 2025 after re-examination; Kisunla was authorised 24 September 2025. Both are restricted to ApoE ε4 non-carriers and heterozygotes with confirmed amyloid pathology. Both EPAR pages were directly fetched 2026-08-31.
  • NICE: both appraisals remained in development with expected publication TBC on 31 August 2026. Donanemab's appeal was upheld in part and a new draft consultation ran in March 2026; lecanemab underwent a second draft consultation in March 2026 and a third committee meeting on 8 July 2026. Latest drafts remain negative, but they are not final published guidance. Appraisal pages directly fetched 2026-08-31.

2.8 Symptomatic-treatment recommendation grades

  • EFNS–ENS/EAN combination guideline (2015) (PMID 25808982). Four trials, 1,549 participants: combination therapy improved global impression (SMD −0.20, 95% CI −0.31 to −0.09), cognition (−0.27, −0.37 to −0.17) and behaviour (−0.19, −0.31 to −0.07). GRADE quality was high for behaviour, moderate for cognition/global impression and low for activities of daily living; the resulting recommendation was weak.
  • Korean Dementia Association cognitive-enhancer guideline (2025) (PMID 39944527). Cholinesterase inhibitors in AD and memantine in moderate-to-severe AD receive strong recommendations based on moderate evidence. The document explicitly leaves amyloid-targeting therapy to a later update.
  • Korean Dementia Association behavioural-symptom guideline (2025) (PMID 39944528). Antipsychotics such as risperidone for aggression and psychosis carry a conditional recommendation; citalopram is advised for agitation in AD, with no strength grade stated in the retrieved record. The grading preserves the difference between a detectable symptom effect and an acceptable benefit–harm balance.

3. Disagreements and gaps

# Disagreement Position A Position B
1 Does an abnormal Core 1 biomarker diagnose AD in an asymptomatic person? Yes — AA 2024 (PMID 38934362) No, they are at risk — IWG 2024 (PMID 39483064)
2 Can a blood test substitute for CSF or amyloid PET? Not yet, confirm with PET/CSF — AUR 2022 (PMID 35908251) Yes at ≥90%/≥90% in specialised care — CPG 2025 (PMID 40729527)
3 Should anti-amyloid antibodies be used? FDA: yes, traditional approval EMA: only in ApoE ε4 non-carriers and heterozygotes / NICE: draft non-recommendation in appraisals still under development
4 Is APOE genotyping optional or required before treatment? "Recommended to better inform risk discussions" — lecanemab AUR (PMID 37357276) "Should be performed" — donanemab AUR (PMID 40155270)
5 Is anticoagulation a contraindication? Lecanemab AUR: do not treat pending data (PMID 37357276) FDA labelling: caution; donanemab AUR: silent
6 Cholinesterase inhibitors in MCI AAN: may choose not to offer; must disclose lack of evidence (PMID 29282327) CCCDTD5: deprescribe where the indication was MCI (PMID 35128025)
7 Is tau PET clinically useful? 5 of 17 scenarios appropriate 6 of 17 uncertain — the largest uncertainty block in any current document (PMID 39776249)
8 How strongly should cognitive enhancers be recommended? Korean 2025: strong, moderate-certainty recommendation for ChEIs in AD and memantine in moderate-to-severe AD (PMID 39944527) EFNS–ENS/EAN 2015: weak recommendation for ChEI + memantine despite statistically significant effects, because outcome certainty varies (PMID 25808982)

Gaps identified in this build.

  • No current AAN guideline on MCI or dementia (the 2018 document is retired).
  • No European document later than 2010–2012 covering diagnosis and management as a whole; the EFNS guidelines predate every major diagnostic and therapeutic development of the last decade.
  • No located guideline addressing the insurance, employment or legal-capacity consequences of a biological AD diagnosis in an asymptomatic person.
  • Direct-to-consumer AD blood testing has a negative recommendation from the Spanish Society of Neurology (PMID 40685136), but no retrieved outcomes evidence on people using such services.
  • Non-English national guidelines (German S3, Japanese, Chinese, and most Latin American documents) were not located through PubMed searches in this build and are absent from the table; that is a coverage limit of this registry, not evidence that they do not exist.

4. Watch list

Item Why it is on the list
NICE final guidance on lecanemab and donanemab Both appraisals remain in development after appeal/reconsultation; published guidance will determine NHS access
Extension of the blood-biomarker CPG beyond specialised care The 2025 guideline is explicitly scoped to specialised care despite prospective primary-care accuracy data existing (PMID 39068545)
A replacement for the retired AAN MCI guideline Leaves US neurology without current MCI guidance in the plasma-biomarker era
Any AA/IWG reconciliation document The definitional split is the field's largest unresolved governance problem
Prevention trial results (TRAILBLAZER-ALZ 3, AHEAD 3-45) Positive results would force revision of every appropriate-use document toward preclinical treatment
Post-evoke guidance on GLP-1 agonists The two phase 3 semaglutide trials were null (PMID 41865758); guidance should distinguish metabolic indications from unsupported AD disease modification
Updated neuropathologic criteria incorporating LATE-NC staging and tau density Both are established findings the 2012 ABC score does not capture

5. Maintenance rules

  • Re-query PubMed for each lineage before changing any status; supersession is a claim about the newest located document, not about the field.
  • Re-fetch every web-based guideline, regulatory and reimbursement record and record the HTTP outcome and access date; keep failures visible.
  • Never quote a recommendation from a superseded document without labelling it as such.
  • When a dual publication exists, cite the version the reader is most likely to reach and record both.
  • Record disagreements as disagreements; do not resolve them in this registry.