Protocol · graded by use, three uses
APOE genotype testing (learning your APOE result)
Graded on: risk-reducing health behavior (mostly self-reported) and risk factors after learning the APOE result; dementia, cardiovascular events and death have never been measured in a disclosure trial.
Grade D, tested: a human randomized trial looked for a benefit on the endpoint The outcome a trial measures to decide whether the treatment worked, such as heart attacks, blood pressure or a symptom score. graded and did not find one. How to read a grade →
At a glance
The APOE test reads one gene that comes in three types: e2, e3 and e4. Carrying e4 goes with a higher chance of Alzheimer disease, but most people with one e4 copy do not get it by age 85. In trials, people who learned their result were not more anxious or depressed. Carriers worried more about the test, and more changed their long-term care insurance. Learning the result did not help people eat better, move more or quit smoking. No trial has tested whether knowing it prevents dementia, heart disease or death. People with two e4 copies had the most brain swelling on the new Alzheimer antibody drugs.
- What was measured
- Whether people changed smoking, diet or exercise after learning a gene result, and how they felt about it.
- How big
- Across 18 studies, learning a gene-based risk did not change smoking, diet or exercise.Read from the summary of PMID PubMed ID: the permanent number the US National Library of Medicine gives a published paper. Each one links to that paper's record on PubMed. 26979548.
- Who it applies to
- Adults without dementia, mostly in the United States and Europe.
- Safety
- Learning the result did not raise anxiety or depression in the trials, though carriers worried more about the test. People with two e4 copies had more brain swelling on the new Alzheimer antibody drugs.
- Compared with
- People given no gene result, or a risk estimate without gene information.
- Certainty
- Grade D, resting on: Randomized trials of disclosing APOE results and a meta-analysis A study that pools the numbers from several trials into one combined estimate. of trials of DNA-based risk results (found no benefit on behavior or risk factors, D); risk estimates for e4 carriers come from genetic and population studies and are listed as excluded evidence.
- What would change this grade
- Could rise to C with: Randomized controlled trial People are assigned to the treatment or a comparison by chance, so the groups differ only in what they received. It is the design that can show cause and effect. of APOE disclosure with a structured risk-reduction program against the same program without disclosure. Who: Cognitively unimpaired adults aged 50 to 75, unselected for family history, several thousand per arm. Measured: Incident dementia and cardiovascular events, with measured (not self-reported) behavior and harms from disclosure counted. Length: At least 5 years.
- Cost
- No price recorded yet.
- Sources
- 33 studies cited on this record, listed under Sources.
All three uses graded on this record carry a D, each on its own evidence.
This record carries a separate letter for each thing it has been measured on. The headline grade is the best of them, and it does not transfer to the others.
Hollands et al. (2016; PMID 26979548): DNA-based risk estimates did not change smoking, diet, physical activity or motivation in 18 studies. Leskinen et al. (2021; PMID 33188400) and Fallaize et al. (2016; PMID 27510539): telling people they carried e4 did not make them respond to lifestyle advice better than noncarriers; telling people they were noncarriers may have weakened the response. Chao et al. (2008; PMID 18317253) found carriers more likely than noncarriers to report a behavior change, a self-reported comparison inside the disclosure arm; Christensen et al. (2016; PMID 26810768) found more self-reported changes when coronary risk was added to the disclosure. FINGER (Solomon 2018, PMID 29356827) and US POINTER (Baker 2025, PMID 40720610) found lifestyle programs helped carriers and noncarriers alike, so the result did not change who benefits. Tested, no benefit: D, Route 2 Human randomized trials have been run and did not show benefit on the stated outcome. One of the two routes to the lowest letter on EBL's scale..
The disclosure trials followed people for 1 year (Green 2009, PMID 19605829) to 15 months (Ottman 2026, PMID 42712082) and measured distress, not disease. D here means the trial has not been run, not that one failed.
Tariot et al. (2024; PMID 39320017): in 1,556 cognitively unimpaired carriers, umibecestat caused small declines on cognitive batteries that reversed after washout (harm). Abushakra et al. (2025; PMID 41015981): in 325 e4 homozygotes with early Alzheimer disease, valiltramiprosate did not improve the primary cognitive outcome. Forcano et al. (2025; PMID 40664536): green tea extract added to a lifestyle program in 129 carriers did not meet its primary endpoint The one outcome a trial names in advance as its main test. Results on other outcomes carry less weight.. Yassine et al. (2026; PMID 42315445): DHA did not change cognition or brain volume over 24 months in at-risk adults, 47% of them carriers. Vyas et al. (2024; PMID 37952113): e4 did not modify the effect of vitamin D3 or omega-3s on cognition. Tested, no benefit on the main outcome, one harm: D, Route 2.
Health behavior and risk factors after learning the result (D, tested, no benefit). Dementia, heart disease and death after learning the result (D, never tested). Drugs and supplements chosen because a person carries e4 (D, tested, no benefit on the main measure, mild reversible harm with one drug). Safety: disclosure did not raise anxiety or depression in the randomized trials.
Why this letter
Why this gradeGrade D, tested, no benefit. The reasoning, and what the grade does not cover.
Grade D on Route 2: tested, no benefit on the outcomes the trials measured. A meta-analysis of randomized trials of DNA-based risk results found no effect on smoking, diet, physical activity or motivation (Hollands 2016, PMID 26979548), and in APOE-specific trials people told they carried e4 did not respond to lifestyle advice better than others (Leskinen 2021, PMID 33188400; Fallaize 2016, PMID 27510539). Lifestyle programs improved thinking test scores in carriers about as much as in noncarriers (Solomon 2018, PMID 29356827; Baker 2025, PMID 40720610), so the result did not change who benefited. No randomized trial has measured whether learning the result lowers dementia, heart disease or death (Route 1 No human randomized trial has been run, so the record rests on animal or mechanistic evidence only. One of the two routes to the lowest letter on EBL's scale. for those outcomes). Disclosure did not raise anxiety or depression against nondisclosure in two independent trials (Green 2009, PMID 19605829; Ottman 2026, PMID 42712082).
WHY NOT Cno trial found a health benefit from learning the result; the one positive behavior finding (Chao 2008, PMID 18317253) compared carriers with noncarriers inside the disclosure arm and was self-reported.
WHAT THIS GRADE DOES NOT COVERhow well APOE predicts Alzheimer disease, which comes from genetic and population studies and is described in excluded evidence; the use of APOE status before anti-amyloid antibody treatment for early Alzheimer disease, where trial reports show ARIA is most common in e4 homozygotes (Honig 2024, PMID 38730496; Zimmer 2025, PMID 40063015) but no trial has compared genotype-guided with unguided treatment; and the trials enrolled mostly people with a family history of Alzheimer disease who sought testing and passed psychological screening, so the safety findings do not transfer to people with severe anxiety or depression.
A grade describes the human evidence for a named use. It is not advice, and it is not a judgment on any individual prescription. How the letters are decided →
Evidence notes: meta-analysis of randomized trials, randomized trial
Endpoint graded: Risk-reducing health behavior (mostly self-reported) and risk factors after learning the APOE result; dementia, cardiovascular events and death have never been measured in a disclosure trial (self_report)
Last reviewed: October 2026
The trial that would change this grade.
A positive trial of this shape would move this entry to C.
Why the current trials fall short
Disclosure trials have measured distress, self-reported behavior and blood markers over 1 to 2 years. Only a long trial with clinical outcomes could show whether knowing the result changes health.
Dose, form and timing
Measured outcomes in the disclosure trials: anxiety, depression and test-related distress questionnaires (PMID 19605829; PMID 42712082), self-reported health behavior and insurance changes (PMID 18317253; PMID 15757934), and diet and blood lipids and fatty acids (PMID 33188400; PMID 27510539). None followed participants for dementia, heart disease or death.
As studiedDose, form, population and duration for every use.
A grade describes the intervention that was tested. Where the product on the shelf differs from it, this says so.
Safety
Carriers who learned their result reported more test-related distress than noncarriers (Green 2009, PMID 19605829), and in a large screening cohort distress and anxiety rose most in e4 homozygotes, though below levels of clinical concern (Langbaum 2025, PMID 41319675). People told they carried e4 were 5.76 times more likely than the nondisclosure group to change long-term care insurance (Zick 2005, PMID 15757934). In people selected because they carried e4, the drug umibecestat caused mild cognitive worsening that reversed after stopping (Tariot 2024, PMID 39320017). With anti-amyloid antibodies, brain swelling on MRI (ARIA-E) occurred in 34.5% of e4 homozygotes on lecanemab against 16.8% of e4 carriers overall (Honig 2024, PMID 38730496), and ARIA-E risk with donanemab rose with the number of e4 copies (Zimmer 2025, PMID 40063015).
Cautions documented in human studies. Each caution on this record, sorted by what it rests on.
- Distress: carriers reported more test-related distress, and distress and anxiety rose most in e4 homozygotes, below levels of clinical concern (PMID 19605829 PMID 41319675)
- Insurance: people told they carried e4 were 5.76 times more likely to change long-term care insurance (PMID 15757934)
- Umibecestat: in people selected as e4 carriers, the drug caused mild cognitive worsening that reversed after stopping (PMID 39320017)
- Brain swelling (ARIA-E) on anti-amyloid antibodies: 34.5% in e4 homozygotes on lecanemab, rising with e4 copies on donanemab (PMID 38730496 PMID 40063015)
None recorded.
None recorded.
None recorded.
Does not apply: human harm data are listed above.
Grade history
How this letter has changedLatest: No. 391 · October 2026 · First graded.
APOE genotype testing (learning your APOE result) now has its own record, graded use by use after a full PubMed search in October 2026: Health behavior and risk factors after learning the APOE result (D); Dementia, heart disease and death after learning the APOE result (D); Drugs and supplements chosen because a person carries e4 (D). Health behavior and risk factors after learning the result (D, tested, no benefit). Dementia, heart disease and death after learning the result (D, never tested). Drugs and supplements chosen because a person carries e4 (D, tested, no benefit on the main measure, mild reversible harm with one drug). Safety: disclosure did not raise anxiety or depression in the randomized trials. Applied the written grade definitions under the editor's standing instruction of October 2026 (Arnold Sandlow, D.C.).PMID 19605829 PMID 42712082 PMID 15876213 PMID 18317253 PMID 15757934 PMID 26810768 PMID 26979548 PMID 33188400 PMID 27510539 PMID 25735442 PMID 25499536 PMID 28726810
Every change to a letter on this site is dated and published. The full evidence log →
Sources
- Citations
- Every paper cited on this page is matched to its PubMed record on every build.
- Retractions
- None of these papers is retracted or under a journal expression of concern. PubMed is checked again on every build.
- Read against the papers
- October 2026. By a language model (AI-assisted reading), not yet reviewed by a clinician.
- Outside review
- Not yet done. No reviewer outside EBL has checked this record.
What each of these checks can and cannot show: evidence integrity.
The evidence, in full33 papers, each linked to its PubMed record.
Disclosure against no disclosure. Green et al. (New England Journal of Medicine, 2009; PMID 19605829): REVEAL, 162 adults without symptoms who had a parent with Alzheimer disease randomized to learn or not learn their APOE genotype; over a year, no difference in anxiety (4.5 against 4.4), depression (8.8 against 8.7) or test-related distress (6.9 against 7.5); e4-negative participants had lower test-related distress than e4-positive participants. Ottman et al. (Alzheimer's and Dementia, 2026; PMID 42712082): IDEAL, Latino adults in New York City randomized to Alzheimer risk estimates with APOE (194) or without (180); no difference in genetic-testing impact, traumatic stress or secondary outcomes over 15 months, and 2 against 3 disclosure-related adverse events. Marteau et al. (Risk Analysis, 2005; PMID 15876213): 149 adult children of people with Alzheimer disease; an e4-negative result lowered perceived risk against controls, an e4-positive result did not raise it. Chao et al. (Alzheimer Disease and Associated Disorders, 2008; PMID 18317253): in REVEAL, people told they were e4 positive were more likely than those told they were e4 negative to report an Alzheimer-specific health behavior change at 1 year (adjusted OR 2.73, 95% CI 1.14 to 6.54), although all had been told no proven prevention existed; this compares carriers with noncarriers inside the disclosure arm and is self-reported. Zick et al. (Health Affairs, 2005; PMID 15757934): in 148 REVEAL participants, no difference in health, life or disability insurance purchases, but those who tested positive were 5.76 times more likely than the nondisclosure group to change long-term care insurance. Christensen et al. (Annals of Internal Medicine, 2016; PMID 26810768): 257 adults; adding the coronary risk of APOE to the Alzheimer disclosure did not raise anxiety or depression at 12 months and was followed by more self-reported health behavior changes. Behavior and risk factors. Hollands et al. (BMJ, 2016; PMID 26979548): 18 randomized or quasi-randomized studies of DNA-based risk estimates; no effect on smoking cessation (OR 0.92, 95% CI 0.63 to 1.35), diet (SMD 0.12, 95% CI -0.00 to 0.24) or physical activity (SMD -0.03, 95% CI -0.13 to 0.08), no effect on motivation, and no adverse effects on depression or anxiety; evidence typically low quality. Leskinen et al. (Journal of Nutrition, 2021; PMID 33188400): 188 healthy Finnish adults; those told they carried e4 were not markedly more responsive to lifestyle guidance than the other groups. Fallaize et al. (American Journal of Clinical Nutrition, 2016; PMID 27510539): Food4Me, 1,466 adults; gene-based advice did not change diet differently in e4 carriers and noncarriers, and telling people they were noncarriers may have weakened their dietary response. Hietaranta-Luoma et al. (Journal of Genetic Counseling, 2015; PMID 25735442): 107 healthy Finnish adults; anxiety and threat after disclosure were short-term. Disclosure methods: condensed protocols (Green et al., Alzheimer's and Dementia, 2015; PMID 25499536; 343 adults), telephone (Christensen et al., Genetics in Medicine, 2018; PMID 28726810; 257 adults) and videoconference (Wood et al., Alzheimer's and Dementia, 2026; PMID 42509665; 598 adults) were each comparable to in-person or telephone disclosure on psychological outcomes. Lifestyle and supplements by genotype. Solomon et al. (JAMA Neurology, 2018; PMID 29356827): FINGER, 1,109 at-risk adults aged 60 to 77; the multidomain program improved the neuropsychological test battery in e4 carriers (difference 0.037 per year, 95% CI 0.001 to 0.073) and the effect did not differ significantly from noncarriers (0.023, 95% CI -0.021 to 0.067). Baker et al. (JAMA, 2025; PMID 40720610): US POINTER, 2,111 adults; the structured program's benefit was consistent for e4 carriers and noncarriers (P = .95 for interaction). Vyas et al. (Journals of Gerontology Series A, 2024; PMID 37952113): in 743 older adults in a 2 by 2 factorial trial, e4 did not modify the effects of vitamin D3 or omega-3s on cognition. Treatments in carriers selected by genotype. Tariot et al. (Alzheimer's and Dementia, 2024; PMID 39320017): API Generation Studies, 1,556 cognitively unimpaired e4 homozygotes and heterozygotes with elevated amyloid; the BACE inhibitor umibecestat caused small declines on some cognitive batteries that reversed after washout. Abushakra et al. (Drugs, 2025; PMID 41015981): APOLLOE4, 325 e4 homozygotes with early Alzheimer disease; valiltramiprosate had no significant effect on the primary cognitive outcome at 78 weeks (p = 0.607), slowed hippocampal atrophy, and more than doubled nausea, vomiting and decreased appetite. Forcano et al. (Journal of Prevention of Alzheimer's Disease, 2025; PMID 40664536): PENSA, 129 e4 carriers with subjective cognitive decline; adding green tea extract to a lifestyle program did not meet the primary endpoint (p = 0.061). Yassine et al. (EBioMedicine, 2026; PMID 42315445): 365 adults at risk of dementia, 47% e4 carriers; 2 g a day of DHA raised spinal fluid DHA but did not change brain volume or cognition over 24 months. Anti-amyloid antibodies. van Dyck et al. (New England Journal of Medicine, 2023; PMID 36449413): Clarity AD, ARIA with edema or effusions in 12.6% on lecanemab. Honig et al. (Alzheimer's Research and Therapy, 2024; PMID 38730496): ARIA-E with lecanemab was more common in e4 carriers (16.8%) and most common in e4 homozygotes (34.5%). Perry et al. (Journal of Prevention of Alzheimer's Disease, 2026; PMID 41689888): in heterozygotes and noncarriers, ARIA-E 9%. Sims et al. (JAMA, 2023; PMID 37459141): TRAILBLAZER-ALZ 2, ARIA-E 24.0% on donanemab against 2.1% on placebo An inactive substitute made to look like the real treatment, so participants cannot tell which one they received., 3 treatment-related deaths on donanemab. Zimmer et al. (JAMA Neurology, 2025; PMID 40063015): ARIA-E risk with donanemab rose with the number of e4 copies.
How EBL searched for the trials (1 search)
The search that was run, what it returned, and what was left out and why, so the absence of evidence can be checked as well as its presence.
(APOE[tiab] OR apolipoprotein E[tiab]) AND (disclos*[tiab] OR "genetic risk information"[tiab] OR REVEAL[tiab]) AND (randomized controlled trial[pt] OR randomized[tiab])PubMed watch: does a new study change the grade? (nothing new)
No new paper matching the search behind this grade has been added to PubMed since October 2026.
Last run October 2026. A letter on EBL moves only after a logged review. A new paper never moves a letter by arriving: it is read, the answer is written here, and any change to the letter is published in the evidence log.
Disclosure
Educational use only · Not medical advice · EBL grades published human evidence and does not recommend treatment. Stopping a prescription is a decision for your physician.
Evidence Based Longevity, Inc. is owner-operated and edited by Arnold Sandlow, D.C., a California-licensed chiropractor. Evidence Based Longevity publishes in English only · Machine translations are not produced or reviewed by EBL for accuracy
© Evidence Based Longevity, Inc. All rights reserved
These statements have not been evaluated by the Food and Drug Administration. Any grade on this page describes the strength of published human trial evidence in the populations studied. It is not a recommendation for any individual and is not medical advice. Starting, stopping or changing any supplement or medication is a decision for your physician.
Grades move when new trials publish. A free monthly email lists every grade EBL moved, with the reason and the trial behind it: the sign-up is on the Evidence Log →