Alzheimer's disease · animated companion
Alzheimer's Companion
Alzheimer's disease is the most common cause of dementia. Here it is in moving pictures: watch a protein get cut the wrong way, follow tangles spreading through the brain, see the changes that begin many years before the first forgotten name, and look honestly at what the newest drugs do and do not achieve.
Schematic. Amyloid rises first, tau follows, and memory changes come last.
For students
Every module has an Exam hook box and a citation trail. Finish with flashcards, mnemonics and a quiz.
For patients and families
Look for the In plain words boxes. Bring questions to your own clinician: the app reports what guidelines and studies say, not what you should do.
How sure is each claim?
Research claims carry a label: Established Early evidence Hypothesis. Small numbers link to the Sources list.
Part 1 · The disease
What Alzheimer's is
Dementia is not one disease. It is a pattern of decline that several diseases can cause, and Alzheimer's disease is the most common of them. Press play to build the picture, then see how the odds change with age.
Schematic. Circle sizes are not to scale; the bars and dots are.
Age is the strongest risk factor
Each figure is 1 person in 100. US figures, Alzheimer's Association 2025.
The numbers, and where sources disagree
Counts depend on who is counting, which year, and whether they mean all dementia or Alzheimer's alone. Each card names its source so the figures can be compared fairly.
57 million
people were living with dementia in 2021, over 60% of them in low- and middle-income countries, with nearly 10 million new cases each year. 515 A secondary analysis of GBD 2021 counted 9.84 million new cases and 1.95 million deaths in 2021. 528
These are counts for all dementia, not Alzheimer's alone.
152.8 million by 2050
GBD 2019 projected a rise from 57.4 million in 2019. The age-standardised rate stays almost flat, so the growth comes from more people reaching old age. Projected increases range from 53% in high-income Asia Pacific to 367% in north Africa and the Middle East. 192
Against that, age-specific incidence seems to be falling in high-income countries. In Framingham, 5-year incidence fell from 3.6 to 2.0 per 100 people between the late 1970s and early 2010s, but only among people with at least a high-school diploma. 449
WHO's 57 million (2021) and GBD's 57.4 million (2019) are different years and methods, not a contradiction.
7.2 million
Americans aged 65 and older were living with Alzheimer's dementia in 2025, about 1 in 9. About 22% of people over 65 have mild cognitive impairment, and roughly 5 to 7 million may have MCI due to Alzheimer's, an estimate the report says needs refining. 2
Sources differ The 2025 report projects 13.8 million by 2060. An earlier census-based model reached the same 13.8 million by 2050, a decade sooner. 2,229
$384 billion
US payments in 2025 for health care, long-term care and hospice for people with Alzheimer's or other dementias. Medicare and Medicaid pay $246 billion (64%), and out-of-pocket spending is $97 billion (25%). Annual payments are projected to reach just under $1 trillion by 2050. The lifetime cost of care for one person was estimated at $405,262, about 70% of it borne by families. 2
Worldwide, WHO puts the cost at US$1.3 trillion in 2019, about half of it attributable to informal carers. 515
19.2 billion hours
In 2024, 11.9 million US family members and other unpaid caregivers gave an estimated 19.2 billion hours of care, valued at $413.5 billion. That averages nearly 31 hours per caregiver per week. 2
Worldwide, women provide 70% of the care hours for people living with dementia. 515 See Living with dementia and caregiving.
4.4 vs 2.8 million
Almost two-thirds of older Americans with Alzheimer's dementia are women: 12% of women and 10% of men aged 65 and over. At age 45 the lifetime risk was about 1 in 5 for women and 1 in 10 for men in Framingham data. 2 Globally GBD estimated 1.69 women for every man with dementia in 2019. 192
Debated Why?
- Longevity. Women live longer, and age is the strongest risk factor. Most US incidence studies find no meaningful sex difference at any given age.
- Survival bias. Men who survive to 65 tend to have healthier hearts and vessels.
- Biology Hypothesis Hormonal or reproductive factors, or a possibly stronger effect of APOE e4 in women.
- Social and gender factors. The report says more research is needed. 2
A review argues that symptoms, rates of decline and brain atrophy differ by sex, and that sex should be used to stratify research and care. 170
4 to 8 years
on average for people diagnosed at 65 or older, and some live 20 years. 2 Age at diagnosis matters most:
Baltimore Longitudinal Study of Aging, with no significant difference between men and women. 64 A review found median survival of 3.2 to 6.6 years from diagnosis, and 7 to 10 years from onset in most studies. 487
7th
Alzheimer's was the seventh-leading cause of death in the US in 2022, with 120,122 recorded deaths, up more than 142% since 2000. Death certificates often name pneumonia instead, so the true count is higher. 2 Worldwide, WHO ranks dementia seventh too. 515
Pneumonia is the most commonly identified immediate cause, and swallowing problems can lead to aspiration pneumonia. 2 In 524 autopsies of people with dementia, bronchopneumonia caused 38.4% of deaths and ischaemic heart disease 23.1%, against 2.8% bronchopneumonia in the general older population. 65 In nursing-home residents with advanced dementia, 41.1% developed pneumonia over 18 months, and 46.7% died within six months of it. 358
Part 1 · The disease
The brain that forgets
Tau tangles and amyloid plaques both spread through the brain in a fixed order, but not the same order. Step through the tangle stages, then switch to amyloid and compare where each one starts.
Schematic outlines. Deep structures are dashed. Regions outside a scheme are left unshaded; that does not mean they are spared.
Why memory goes first
The first tangles form where the hippocampal memory circuit begins. Follow a signal around the loop, then watch which neurons are lost, in the order the studies report.
Schematic. Dots on the loop are signals; faded dashed cells are lost neurons. The MRI shapes compound the published yearly rates, for illustration.
Scoring a brain at autopsy: the ABC score
The 2012 NIA-AA guidelines rate three things and combine them into one level of Alzheimer's neuropathologic change. Choose a rating for each and read the result. 246,366
Table after Montine 2012, Table 3. B0 and B1 share a column in the original.
Part 1 · The disease
Amyloid, animated
Amyloid beta is a small fragment cut out of a larger protein, APP, that sits in the neuron's membrane. Whether the fragment gets made depends on which enzyme cuts first. Pick a pathway and press play.
Schematic, not to scale.
From one fragment to a plaque
Schematic. Real aggregation takes years and happens in three dimensions.
How strong is the amyloid case?
Part 1 · The disease
Tau and the tangle
Tau is a normal protein that holds the tracks inside a nerve fibre together. In Alzheimer's it collects too much phosphate, lets go, and folds into tangles. Press play to watch it happen, then see how it spreads.
Schematic, not to scale. Real microtubules are hollow tubes; the ladder is a teaching picture.
Spreading from cell to cell
Schematic network. Each node stands for a region, each line for a connection.
Tau tracks memory better than amyloid
"Weak" and "strong" give the direction each source reports, not a measured coefficient. The only coefficient shown, r = 0.96, is Terry 1991.
Part 1 · The disease
Microglia and the lost synapse
Memory depends on synapses, the contact points between neurons. In Alzheimer's they are lost early, and the brain's own immune cells help remove them. Press play to watch a patch of tissue change, then try the two switches.
Schematic, not to scale. The synapse count is illustrative; the two ticks on the bar are measured human values.
Synapses track memory best
In 1991 Robert Terry's group counted synapses, plaques and tangles in the brains of 15 people with Alzheimer's and 9 controls, and compared each with how the person had scored on cognitive testing. Switch the measure on the x-axis and watch the cloud change shape. 484
Points are schematic, drawn to show the pattern, not Terry's data. The r of 0.96 is the published value.
The other cells, and the other diseases
Astrocytes react too
Activated microglia release IL-1α, TNF and C1q, which push astrocytes into a reactive state, named "A1" in 2017, that supports neurons less well. Such astrocytes were abundant in Alzheimer's brains. 312
But "A1" and "A2" are too simple
A 2021 consensus statement criticised binary astrocyte labels (A1 versus A2, good versus bad) as oversimplified, and recommended describing reactive astrocytes with several measures at once. 143
Microglial genes carry risk
TREM2 R47H raised risk about 2.9-fold. Other rare variants in microglial genes point the same way: PLCG2 P522R is protective (odds ratio 0.68), ABI3 S209F (1.43) and TREM2 R62H (1.67) raise risk. 263,464 More in Genes and inheritance.
A leaky barrier, early
Breakdown of the blood–brain barrier in the hippocampus was found in people with early cognitive problems, whatever their amyloid and tau status. The authors propose it as an independent early marker; the study shows association, not cause. 375
Most brains with Alzheimer's have something else too
Alzheimer's rarely travels alone in older brains. Each figure below comes from a different group of people, so the percentages cannot be added or compared directly.
| Finding | How often | In whom | Source |
|---|---|---|---|
| Pure Alzheimer's pathology, nothing else | 30% | The 50 people with dementia among 141 community autopsies; more than half had several pathologies | 453 |
| Alzheimer's plus infarcts (strokes) | 38% | Same community autopsies; more than one diagnosis nearly tripled the odds of dementia (OR 2.8) | 453 |
| Infarcts, moderate-to-severe CAA, atherosclerosis, arteriolosclerosis | each >30% | Religious Orders Study / Memory and Aging Project brains; each explained 20–30% of decline when present | 209 |
| Some cerebral amyloid angiopathy (CAA: amyloid in vessel walls) | 85–95% | People with Alzheimer's | 114,209 |
| Moderate-to-severe CAA | 13% | Mayo brain bank | 209 |
| LATE (TDP-43) enough to be linked to impairment | about 25% | Community autopsy brains | 378 |
| Alpha-synuclein (Lewy) co-pathology | 41–55% | Alzheimer's brains in 766 autopsies | 432 |
| TDP-43 co-pathology | 33–40% | Same 766 autopsies; linked to age and APOE ε4 | 432 |
Sleep and the "glymphatic" system
The mouse findings
In mice, fluid enters the brain along spaces around arteries and leaves around veins. Deleting the astrocyte water channel AQP4 cut solute clearance by about 70%. Sleep widened the space between cells by about 60% and sped amyloid clearance, and sleep restriction increased plaques. 247,269,526
Other labs disagree
An independent group found transport in brain tissue fitted diffusion, not flow, and did not depend on AQP4. A 2024 mouse study found clearance was lower, not higher, during sleep and anaesthesia. 354,467
What people show
One night without sleep (7 people against 17 controls) slowed clearance of a tracer from most brain regions. Sleep deprivation raised spinal-fluid amyloid by 25–30%, but mainly through more production, not less clearance. 129,330
Where this leaves it
That sleep affects amyloid levels is better supported than any specific glymphatic mechanism in human Alzheimer's. That is a synthesis across the studies above, not any single study's conclusion. Meningeal lymphatic vessels, described in 2015, may be another drainage route; their role in humans is unknown. 330,467,475
Part 1 · The disease
Genes and inheritance
Most Alzheimer's is not inherited in a simple way. One common gene, APOE, shifts the odds. A few rare genes all but guarantee it, and an extra chromosome 21 does too. Pick a genotype, then play the family below.
APOE: one gene, three common versions
Everyone carries two copies of APOE, one from each parent. The three common versions, ε2, ε3 and ε4, differ at only two positions in the protein (112 and 158). 531 Choose a pair to see how often it turns up in people with Alzheimer's compared with ε3/ε3.
Odds ratios compare the odds of Alzheimer's with ε3/ε3, on a log scale. They are group averages, not anyone's personal chance.
Odds ratio is not lifetime risk, and the estimates disagree
An odds ratio of 15 does not mean a 15-times or a near-certain chance. Researchers have turned the odds into lifetime risk two ways, and get different answers. Rows for your chosen genotype are highlighted.
Why they disagree. Case-control and memory-clinic samples carry referral bias; the population cohorts measured MCI or dementia rather than Alzheimer's alone; and most cohorts are of European ancestry, where the ε4 effect is larger than in African ancestry. 21,47,420
What both agree on. ε4 raises risk but does not decide it: many ε4/ε4 people never develop dementia, and Alzheimer's often occurs without ε4. Only 38% of APOE alleles in people with Alzheimer's are ε4. 194,420,531
Dominant inheritance: APP, PSEN1 and PSEN2
A mutation in APP, PSEN1 or PSEN2, or an extra copy of APP, is the only known cause that all but guarantees Alzheimer's. 40 Pick the gene, play the family, and press "Another family" to see that 50% is a chance for each child, not a quota.
Schematic family. DIAN years are cross-sectional estimates relative to the parent's age at onset.
Down syndrome: a third copy of APP
APP, the gene for the protein amyloid beta is cut from, sits on chromosome 21. People with Down syndrome usually have three copies of chromosome 21, so three copies of APP. Drag the age, then switch to the rare case with only two APP copies.
Ages are from separate cohorts; positions on the ladder are to scale.
Resilience, risk scores and testing
APOE3 Christchurch
A woman with the Colombian PSEN1 E280A mutation did not develop MCI until her seventies, about three decades late. She had two copies of APOE3 Christchurch, very high amyloid, and little tau. Among 1,077 carriers, the 27 with one Christchurch copy began to decline at a median 52 years against 47, about 5 years later. 26,421
RELN-COLBOS, 2023
A man with the same PSEN1 mutation stayed cognitively intact until 67 without Christchurch. He carried a gain-of-function variant in Reelin, which binds the same receptors as APOE; in mice it reduced tau phosphorylation. Both resilient people had very high amyloid but little entorhinal tau, suggesting protection acted at the tau stage. 26,323
A673T: protection in APP itself
An Icelandic APP variant next to the beta-secretase site cut amyloid production by about 40% in the lab, and protected against Alzheimer's and against decline in older people without it. 262 See Amyloid, animated.
The Colombian kindred
A 1997 survey of five extended families in Antioquia (nearly 3,000 people) found 128 affected, with onset at a mean 46.8 years. PSEN1 E280A traces to one founder about 15 generations ago. In 449 carriers, median ages were 44 for MCI, 49 for dementia and 59 for death. 8,296,322
GWAS: 75 places in the genome
The largest study, Bellenguez 2022 in Nature Genetics (111,326 cases, 677,663 controls), found 75 risk loci, 42 new, pointing to amyloid and tau processing, microglia and TNF-α signalling. Rare TREM2 R47H raised risk about 2.9-fold. 46,263
Polygenic scores
A score built with APOE, age and sex reached an AUC of 78.2%. Beyond age and APOE ε4, the Bellenguez score separated the top and bottom tenths by only 1.6- to 1.9-fold. Scores discriminate poorly in the general population, work less well outside European ancestry, and have no established clinical use. 46,144,309
Twins: heritability 58–79%
In 11,884 Swedish twin pairs, heritability of Alzheimer's was 58% in the full model and 79% in the best-fitting one; the rest came from non-shared environment. Identical twins also differed less in age at onset. The authors stress non-genetic factors still matter. 190
Is ε4/ε4 its own genetic form?
In 2024, Fortea and colleagues reported Alzheimer's biology in almost all ε4/ε4 people, with symptoms at a mean of 65, and proposed it as a distinct genetic form. Critics answer that near-full penetrance of biology is not near-full penetrance of dementia, citing lifetime risks of about 30–55%. 21,184
Testing: what the guidelines say
Predictive APOE testing
A 1995 ACMG/ASHG statement said APOE should not be used for predictive testing. The 2011 ACMG/NSGC guideline calls APOE susceptibility testing "not clinically recommended" and advises against direct-to-consumer testing; if a person still insists after counselling, it says testing should follow the Huntington disease protocol. 7,201,202
Families with a dominant gene
The guideline says testing for APP, PSEN1 and PSEN2 should be offered to people with early-onset Alzheimer's and a family history, to families with a dominant pattern, and to relatives of a known carrier, using a Huntington-style protocol. Children should not be tested. Only 8.4% of at-risk people in one series asked for predictive testing. 40,201
Genotyping before anti-amyloid drugs
US labels for lecanemab and donanemab say ApoE ε4 testing "should be performed prior to initiation of treatment", because ARIA (brain swelling or bleeding on MRI) is more common in ε4/ε4. This is a different purpose from predicting Alzheimer's. 103,154,155,423 See Anti-amyloid antibodies.
What disclosure did: REVEAL
162 adult children of people with Alzheimer's were randomised to learn their APOE result or not. At 1 year anxiety, depression and distress did not differ. ε4-positive people more often changed health behaviours (adjusted OR 2.73) and were 5.76 times more likely to change long-term care insurance. 72,208,539
Part 2 · Course and diagnosis
Twenty years before symptoms
Alzheimer's changes in the brain long before anyone notices. Biomarkers turn abnormal in a broadly predictable order: amyloid, then tau and nerve-cell damage, then thinking. Play the clock, and switch between families with a dominant gene and the common late-onset form.
Where is this person on the 2024 map?
In 2024 an Alzheimer's Association workgroup proposed two scales: a biological stage from biomarkers (A to D, by how far tau has spread on PET) and a clinical stage from symptoms (0 to 6). The International Working Group (IWG) disagreed about what to call people without symptoms. Set the biomarkers and symptoms, or pick an example. 122,255
Shaded cells: the expected diagonal from 1A to 4–6D. Hatched: no Core 1 biomarker abnormal.
How the model and the definitions changed
Jack 2010, revised 2013
The 2010 model put amyloid first and neurodegeneration (CSF tau, MRI atrophy, FDG) after a variable lag, tracking symptoms. The 2013 update indexed people by time, added person-to-person variation, reordered some markers, and proposed that tau may start on its own in sporadic disease and be accelerated by amyloid. 252,253
DIAN, and its caveats
DIAN's numbers come from 128 mutation carriers placed on a clock set by the parent's age at onset. Longitudinal data from 217 people confirmed amyloid changes first, but disagreed with the cross-sectional timing of CSF tau by more than 10 years and showed CSF p-tau181 falling near onset. The authors warn the results may not apply to sporadic Alzheimer's. 41,345
2011, 2018, 2024
In 2011, diagnosis was clinical and biomarkers were not advised for routine use. In 2018, a research framework defined Alzheimer's biologically and required A+ and T+. In 2024, one abnormal Core 1 result, such as amyloid alone, is enough. 254,255,347
How long each phase lasts
Across 6 cohorts (3,268 people), someone with preclinical Alzheimer's at age 70 was estimated to spend about 10 years preclinical, 4 with MCI and 6 with dementia. Total duration ranged from 24 years starting at 60 to 15 years starting at 80. 501
Part 2 · Course and diagnosis
Stages and symptoms
Alzheimer's takes away abilities in a fairly fixed order, from the most demanding tasks to the most basic ones. Press Play, or pick a stage, and watch which everyday abilities are kept and which are lost.
Schematic. Stages are drawn equal width, not to time. The year spans above come from one model (Vermunt 2019, someone with preclinical Alzheimer's at age 70). Reisberg's own average durations for each FAST stage are widely repeated but could not be checked against the 1988 paper (unverified), so they are not shown. CDR and MMSE are aligned to FAST approximately: the scales were built separately.
After a diagnosis of MCI: who progresses?
Mild cognitive impairment (MCI) means thinking has declined more than expected for age while daily life is still independent. It is a risk state, not a sentence. Two groups of 100 people with MCI are followed for five years below.
Illustration. Hollow figure: no dementia yet. Solid figure: progressed to dementia. It applies each published annual rate to those not yet converted, every year, which is a simplification: real risk differs from person to person.
Not always memory first
Typical Alzheimer's starts with memory, because the first tangles form in the medial temporal lobe. Some people start with vision, language or planning instead. These atypical forms are much more common when the illness starts before 65. 206
Each square is one person in 100. "About one third" is drawn as 33.
Beyond memory: mood and behaviour
Pooled prevalence in Alzheimer's, Neuropsychiatric Inventory. Studies varied a lot.
How long, and how the illness ends
Survival is measured in years
People aged 65 and older live on average 4 to 8 years after a diagnosis of Alzheimer's dementia, and some live as long as 20 years. Across studies, median survival from onset was 3.3 to 11.7 years, most often 7 to 10. 2,487
Age at diagnosis matters most
Median survival was 8.3 years when diagnosed at 65 and 3.4 years at 90, with no significant difference between men and women. Across studies, age was a consistent predictor of death; male sex was a less consistent one. 64,487
Pneumonia is the most common final event
Pneumonia is the most commonly identified immediate cause of death, and swallowing problems can lead to aspiration. In one autopsy series, respiratory causes accounted for 55.5% of deaths in Alzheimer's. 2,65
Advanced dementia in nursing homes
In the CASCADE study of 323 residents with advanced dementia, 54.8% died over 18 months. Eating problems affected 85.8% and pneumonia 41.1%. See Living with dementia for planning and end-of-life care. 358
Part 2 · Course and diagnosis
Diagnosis and biomarkers
A diagnosis is built in layers: first whether thinking has declined, then what else could explain it, then whether Alzheimer's pathology is present. Follow one person through the pathway, or click any step.
Schematic. Real pathways vary: not everyone needs every step, and the order can change. Each box can be clicked or reached with the Tab key.
What a positive blood test really means
Teaching tool This calculator shows arithmetic, not the performance of any real test in any real person. Pick how good a test is (sensitivity and specificity) and how likely the pathology is before testing. The grid then sorts 1,000 people by their true status and their result.
Test presets: the 2025 Alzheimer's Association guideline minimums for a blood test used for triage (90% sensitivity, 75% specificity) or as a substitute for PET or CSF (90% and 90%) 399, and amyloid PET visual reads against autopsy 79. Settings: amyloid positivity in cognitively normal people at 50 256; a primary-care example value (no source here gives one measured figure); 50% with Alzheimer's pathology among 1,213 Swedish patients 398; amyloid PET positivity in IDEAS, among Medicare patients with MCI or dementia of uncertain cause 422.
Filled circle: true positive. Ring: false negative. Square: false positive. Small dot: true negative. Teaching tool, not a diagnostic calculator.
The biomarker toolbox
Amyloid PET
Florbetapir visual reads found moderate-to-frequent neuritic plaques with 92% sensitivity and 100% specificity against autopsy within 2 years. The Centiloid scale runs from 0 (young controls) to 100 (typical mild-to-moderate Alzheimer's); about 12 to 24 Centiloids marked plaque thresholds at autopsy. 79,282,294
Spinal fluid (CSF)
In Alzheimer's, CSF total tau is about 2.5 times and p-tau about 1.9 times control levels, and Aβ42 is about half. The Aβ42/40 ratio classifies better than Aβ42 alone. The FDA permitted the Lumipulse CSF ratio test in May 2022. 149,221,391
Blood p-tau217
Plasma p-tau217 separated clinical Alzheimer's from other neurodegenerative diseases with AUC 0.96. In 1,213 Swedish patients, a p-tau217 test was about 90% accurate, against 61% for primary care doctors and 73% for specialists using standard workup. In May 2025 the FDA cleared the first blood test, for specialised care, "not intended as a screening or stand-alone diagnostic test". 153,397,398
A rule-out test for primary care
In October 2025 the FDA cleared Roche's Elecsys pTau181 plasma test for primary care, to help rule out Alzheimer's amyloid pathology. Its reported negative predictive value was 97.9%, a figure that depends on the low prevalence of the people studied. 433
Tau PET and FDG-PET
Flortaucipir tau PET separated Alzheimer's from other neurodegenerative disorders with about 90% sensitivity and specificity, outperforming MRI volumes. FDG-PET patterns correctly classified 95% of Alzheimer's, 92% of Lewy body and 94% of frontotemporal dementia in a 7-centre study. 370,394
NfL and GFAP
Neurofilament light rises with neurodegeneration of any cause, including MS, ALS and brain injury. GFAP, an astrocyte marker, rises with amyloid even before symptoms. The 2024 criteria use neither for diagnosis. 255,337,402
Screening tests: what they can and cannot do
MMSE
- 11 items, 30 points; about 7 minutes. 59,94
- At a cut-point of 24 in community settings: sensitivity 0.85, specificity 0.90 for dementia. 94
- Strongly affected by schooling: median 29 with 9 or more years, 22 with 0 to 4 years. 95
- Highly verbal and insensitive to mild impairment. 488
- Under copyright; its authors began enforcing their rights in 2000, and licensed forms cost about USD 1.23 each in 2011. 380
MoCA
- About 10 minutes, 30 points; adds executive and visuospatial tasks. In the original study a cut-off of 26 caught 90% of MCI, against 18% for the MMSE; specificity was 87%, against 100% for the MMSE. 374
- The cut-off dispute: the owner keeps 26 or above as normal, with 1 point added for 12 or fewer years of education. A meta-analysis found 26 gives too many false positives, especially in older or less educated people, and the best accuracy at 23. 69,360
- 2025 change: since December 2025, certification is optional for total-score screening but still required to interpret sub-scores. 360
Mini-Cog
- Three-word recall (0 to 3) plus a clock drawing (0 or 2), total 0 to 5. A score of 0 to 2 means a higher likelihood of important impairment; a normal score does not rule it out. 355
- About 3 minutes, and its accuracy was not affected by education or language in the original multilingual study. 59
- In a meta-analysis: sensitivity 0.91, specificity 0.86 for dementia. 490
The limit they share
Screening tools grade severity and prompt further assessment. They do not establish the cause. A highly educated person with early disease can score normally; low education, poor hearing or vision, or a language barrier can lower a score without dementia. 94,95,355,488
Four definitions of Alzheimer's
The criteria have moved from a clinical syndrome toward biology, and the field disagrees about how far to go. Pick a person and see how each framework would label them.
Part 2 · Course and diagnosis
The differential
Not every memory problem is Alzheimer's. How fast it started and what came first are among the strongest clues. Watch eight typical courses unfold on the same clock, then click any row for details.
Typical patterns, schematic: higher line means better function. Solid line: thinking. Dashed line: the domain that leads (behaviour, gait or mood). Real people rarely follow a textbook course, and mixed causes are common.
The differential, condition by condition
Click a condition to open what else the sources say about it.
| Condition | What points to it rather than Alzheimer's | Key tests and findings |
|---|
Delirium or dementia?
Delirium is a medical emergency that can look like dementia, and it often happens on top of dementia. The Confusion Assessment Method (CAM) asks four questions. Switch the features on and off to see how the algorithm combines them.
Teaching illustration of the published algorithm, not a bedside tool. 248
| Feature | Delirium | Typical Alzheimer's dementia |
|---|---|---|
| Onset | Acute: hours to days (CAM feature 1). 248 | Insidious, over months to years. 347 |
| Course | Fluctuates, often within a day. 248 | A clear history of gradual worsening. 347 |
| First domain | Inattention, a required CAM feature. 248 | Usually recent memory. 2,206 |
| Arousal | Level of consciousness may be altered (CAM feature 4). 248 | Dementia criteria require that delirium does not explain the symptoms. 347 |
| Tools | CAM: under 5 minutes, 94 to 100% sensitivity and 90 to 95% specificity in validation. In people with dementia, one study found only 77% sensitivity. 248,367 | History from an informant, cognitive testing, then the workup in Diagnosis and biomarkers. 30 |
| Overlap | Delirium on top of dementia was reported in 22 to 89% of older people with dementia and linked to faster decline, institutional care and death. Fluctuating attention and arousal are also core features of Lewy body dementia. 171,346 | |
The reversible-causes workup
Rates from a 2003 meta-analysis of 39 studies (5,620 people), scaled to 1,000 people. Positions are random.
Routine in the AAN parameter
Screening for depression, B12 deficiency and hypothyroidism, and a non-contrast CT or MRI, in the initial evaluation. This is a 2001 guideline. 284
Only when suspected
Syphilis serology when neurosyphilis is suspected. CSF or genetic testing is considered for atypical cases: onset before 65, rapid onset, or non-memory presentations. 30,284
Medicines
A medication review. The heaviest cumulative use of strong anticholinergics, most commonly tricyclic antidepressants, first-generation antihistamines and bladder antimuscarinics, was associated with dementia (hazard ratio 1.54). 207
Sleep
Sleep-disordered breathing was associated with incident MCI or dementia in 298 older women (odds ratio 1.85), following low oxygen rather than broken sleep. 530
Gait first
When walking changes first, imaging looks for the enlarged ventricles and DESH pattern of normal-pressure hydrocephalus; a CSF tap test may follow. 373
A sudden change
A sudden worsening points to delirium and a search for medical and drug causes, even in someone already living with dementia. 171,248
Part 3 · Treatment
Cholinesterase inhibitors and memantine
The older Alzheimer's drugs act on the synapse, the gap where one neuron signals the next. Watch one cholinergic signal from start to finish, then switch to Alzheimer's or to a drug and play it again.
Schematic, not to scale. The chart counts the molecules in the drawing; it is not measured data.
Memantine at the NMDA receptor
Memantine works on a different system: the NMDA receptor, a calcium channel opened by the transmitter glutamate. The idea behind it is to quiet a constant low-level leak through the channel while still letting a real signal through.
Schematic. The channel has four subunits in reality; two are drawn. Traces are illustrative, not recordings.
How big is the benefit?
Average effects from Cochrane reviews and trials, drawn on the full length of each rating scale so you can see how much of the scale they cover. Then the side effects the labels list, and what happened when people stopped.
Average differences from drug versus placebo. Bars start at zero on each scale.
Published label information
What the US prescribing information (FDA labels) says about starting and raising doses. This is reported as published, for study. It is not dosing advice for any person; dosing is decided with a prescriber.
| Medicine | Labelled stage | Label titration (as published) | Label notes |
|---|---|---|---|
| Donepezil | Mild, moderate and severe | 5 mg once daily in the evening; 10 mg only after 4 to 6 weeks on 5 mg. The 23 mg dose (moderate to severe only) only after at least 3 months on 10 mg. 157 | Half-life about 70 hours; steady state within 15 days. Severe-stage indication added 13 Oct 2006. 119,157 |
| Rivastigmine capsules | Mild to moderate | 1.5 mg twice daily with meals, rising through 3, 4.5 and 6 mg twice daily, at least 2 weeks per step; maximum 12 mg/day. 167 | If doses are missed for more than 3 days, the label says to restart at the lowest dose. 160,167 |
| Rivastigmine patch | Mild, moderate and severe | 4.6 mg/24 h, then 9.5 mg/24 h after at least 4 weeks (the minimum effective dose), then up to 13.3 mg/24 h after at least another 4 weeks. 160 | One patch at a time, replaced every 24 hours; do not reuse a site within 14 days. Medication errors, mostly extra patches left on, caused hospitalisation and rarely death. 160 |
| Galantamine ER | Mild to moderate | 8 mg/day with food, 16 mg/day after at least 4 weeks, optionally 24 mg/day after another 4 weeks. 161 | Maximum 16 mg/day in moderate hepatic or renal impairment. Not recommended in MCI: 13 deaths versus 1 in two 2-year MCI trials (the label notes the placebo death rate was unusually low). 161,326 |
| Memantine IR | Moderate to severe | 5 mg once daily, rising in weekly 5 mg steps to 20 mg/day (10 mg twice daily). 163 | Target 5 mg twice daily in severe renal impairment. 163 |
| Memantine ER | Moderate to severe | 7 mg once daily, rising in weekly 7 mg steps to 28 mg once daily. 162 | 14 mg in severe renal impairment; half-life about 60 to 80 hours. 162 |
| Memantine ER + donepezil | Moderate to severe | For people already stable on donepezil 10 mg. 164 | Approved on bioequivalence with the two drugs taken separately. 164 |
Reading the evidence
Part 3 · Treatment
Anti-amyloid antibodies over time
Lecanemab and donanemab are antibodies infused into the blood that help the brain's immune cells clear amyloid plaque. Follow the published trials month by month: plaque falls fast, ARIA (brain swelling or small bleeds on MRI) shows up early, and the clinical benefit appears slowly and stays modest.
Published trial and label information only. Everything here reports what the trials, labels and regulators published. Whether a particular person is eligible, and whether treatment is right for them, are decisions for specialist clinicians together with the person and their family.
Tissue is schematic, not to scale. Filled dots are published trial values; open dots are sponsor figures or values calculated from published numbers; the lines between them are interpolated shapes. In the open-label view, lecanemab dose ticks after month 18 show the current label's every-4-week maintenance option, not the extension's own schedule.
How big is the benefit?
The same trial numbers are read in two very different ways. Pick a scale, then pick any row to see what it is and how each side reads it. This is a live debate among specialists, so both readings are shown side by side.
Bars: point estimate. Whiskers: published 95% confidence interval or range. Vertical lines: proposed thresholds.
What each antibody binds, and what happened
Eligibility, ARIA, dosing and the history
Regulatory timeline
A reminder. Labels, eligibility rules and coverage differ between countries and change over time. The published criteria summarised here are for understanding the evidence, not for deciding about any one person. Those decisions belong with specialist clinicians.
Part 3 · Treatment
Beyond memory
Agitation, low mood, apathy, suspicion and poor sleep are common in dementia. Published guidance starts with looking for the cause. Press play to follow one behaviour through the DICE approach.
Illustration of a published clinical-reasoning approach. No large trial has tested DICE itself.
Antipsychotics: the risk in 1,000 people
Antipsychotic medicines carry a boxed warning for older people with dementia. The figures below come from the class wording in current US labels, a meta-analysis, and a large study of US veterans.
Each dot is one person. Large filled dots are deaths. Rates from label class wording; counts per 1,000 are arithmetic from those rates.
The other published options and their evidence
What trials and regulators found for the other medicines studied for these symptoms. This is a summary of evidence, not a treatment plan. The only dose mentioned is a published caveat about one trial.
Medicines that can make thinking worse
Part 4 · Living, caring, preventing
Lowering the risk
The 2024 Lancet Commission estimated how much dementia worldwide is linked to 14 risk factors that could, in principle, be changed. Play the wheel to build up its estimate, then pick any factor to see the evidence behind it and the caveats.
Population attributable fractions from the Commission's Figure 9, rounded to whole numbers. Click a segment or a row, or use the buttons below.
What the prevention trials found
A risk factor linked to dementia in observational studies is not proof that changing it prevents dementia. Randomised trials test that. Play the plot to bring in each trial's main result; many are null or small.
Dots are estimates; lines are 95% confidence intervals. A line that crosses the dashed no-effect line means the result is compatible with no effect. Units differ between rows, so compare each with its own zero.
Open questions
WHO 2019 and the Lancet Commission 2024
The two bodies weigh evidence differently. WHO grades recommendations on intervention evidence; the Commission models how much dementia each factor may account for.
| Factor | WHO 2019 recommendation (evidence quality) | Lancet Commission 2024 |
|---|---|---|
| Physical activity | Strong for normal cognition (moderate); conditional for MCI (low) 77 | Inactivity: about 2% 321 |
| Smoking | Stop smoking: strong (low) 77 | About 2% 321 |
| Diet | Mediterranean-like diet: conditional (moderate). Vitamin B, E, PUFA and multi-complex supplements: strong recommendation against (moderate) 77 | Not one of the 14 factors 321 |
| Alcohol | Reduce hazardous drinking: conditional (moderate, observational) 77 | Excessive alcohol: about 1% 321 |
| Blood pressure, diabetes | Manage per WHO guidelines: strong. To reduce dementia specifically: conditional (very low) 77 | Hypertension about 2%, diabetes about 2% 321 |
| Cholesterol | Manage dyslipidaemia in midlife: conditional (low) 77 | High LDL: about 7%, new in 2024 321 |
| Hearing loss | No recommendation: intervention evidence insufficient 77 | About 7%, joint largest 321 |
| Depression, social isolation | No recommendation: insufficient evidence 77 | Depression about 3%, isolation about 5% 321 |
| Cognitive training | Conditional (very low to low) 77 | Not one of the 14 factors 321 |
Part 4 · Living, caring, preventing
Living with dementia and caregiving
For the person living with dementia and for the people beside them. Follow the years after a diagnosis, see which plans and safety questions tend to come up at each point, and what the research says helps the family carer.
The caregiving years
Schematic. The order and pace differ from person to person, and many people never meet some of these points.
The family carer
Bars are drawn to scale within each row only. Each row is a different trial with a different outcome.
Words matter
Dementia Australia's language guidelines ask writers to put the person first. The table shows some of the swaps they suggest. 112
Stigma is common. In a global survey of almost 70,000 people, 62% of healthcare practitioners thought dementia is a normal part of ageing, 35% of carers had hidden a relative's diagnosis from at least one person, and over 85% of people living with dementia said their opinion had not been taken seriously. 9
Where sources disagree. The guidelines advise against "carer burden", preferring neutral wording such as "the impact of supporting someone". Research still measures "caregiver burden", for example with the Zarit Burden Interview used in trials. This page uses "impact" in its own words and "burden" only when naming a research measure. 112,189,297
Part 4 · Living, caring, preventing
Nursing care
Four parts of nursing care for a person living with dementia: assessment, giving the medicines, communication and care approaches, and documentation and safety. Each part has an animation or a tool to try, and each claim links to its source.
ScopeThis module describes what guidelines, drug labels and published studies say. It is not a care plan. It does not replace local protocols, your scope of practice, or the prescriber. Where the sources say no dementia-specific evidence exists, the module says so instead of filling the gap.
1 · Assessment
Delirium on top of dementia, hour by hour
Dementia changes over months and years. Delirium changes over hours and comes and goes. When the two overlap, the sudden change can hide inside the slow one. Press play to watch two weeks pass.
Teaching tool: a schematic curve, not data from any patient.
The 4AT, scored
Teaching tool, not a clinical record
Accuracy: a 2021 meta-analysis (17 studies, 3,702 observations) found pooled sensitivity and specificity of 0.88 each. 486 In 785 acute medical patients aged 70 or over, the 4AT had sensitivity 76% and specificity 94%, while the CAM used in routine conditions had 40% and 100%. 459
PAINAD: pain when the person cannot say
Teaching tool, not a clinical record
The rest of the assessment
2 · Giving the medicines
Some checks before a dose are the same every time. Others depend on the situation. Pick a situation to see what the labels and guidance describe, then press play.
Schematic, not to scale.
Antipsychotic reduction programmes
Antipsychotics used for distress in dementia carry an increased risk of death, and programmes in the UK and the US set out to cut their use. The bars show the published figures.
3 · Communication and care approaches
What do the reviews find for the best-known approaches? The animation puts the Cochrane findings side by side, then shows two studies of how staff talk and organise care.
Effect sizes as reported, with the sign turned so that right always means benefit. Different outcomes, so compare with care.
4 · Documentation and safety
Build an ABC chart entry
An ABC chart records the Antecedent (what came just before), the Behaviour (described as seen and heard) and the Consequence (what happened next). Over many entries, a pattern can point to a need. Choose one of each to assemble a worked entry. The example is fictional.
A Cochrane review of 18 trials found that programmes built on functional analysis, of which ABC recording is the basis, reduced how often challenging behaviour happened and improved caregivers' reactions after the intervention, but not at follow-up. The analysis part could not be separated from the rest of each programme. No study was found that validates the ABC chart itself as a nursing documentation tool in dementia. 364
Capacity, best interests and surrogates
Abuse, mouth care, feeding and quality measures
Gaps the sources name
Part 4 · Living, caring, preventing
Toward a cure
Researchers are trying to interrupt the disease at every step, from the first protein cut to the loss of synapses. Pick a route to drop a barrier at its target and see what the best evidence found. Every route carries an evidence label, and failures are shown as plainly as successes.
Where could we intervene?
Schematic, not to scale. The flow beyond the barrier is a picture of the trial result, not a measurement.
| Route | Target | Evidence | Best result so far |
|---|
Prevention trials: treating before symptoms
These trials give a drug to people who have amyloid or a disease-causing gene but no symptoms yet. The timeline runs from the first readout to the last expected one; solid dots are published results, hollow dots are estimated completion dates from the trial registry.
Dates are publication years for results and registry estimates for trials still running.
The 2025 pipeline in numbers
Counts from the annual pipeline reviews of clinicaltrials.gov.
Open questions
What the next trials test
| Question | Trials | Expected |
|---|
Part 5 · Context and review
History and people
From one woman in a Frankfurt asylum in 1901 to the antibodies approved in 2023 and 2024. Scrub the timeline or press play: each era is drawn as a scene, with the milestone and its source in the panel beside it.
Scenes are illustrations, not historical images. Eras are spaced evenly, not to scale in years. A "via a later review" label means the original paper has no abstract and its content comes through a later history review or Alzforum.
What the history leaves open
Why a separate disease?
Historians still debate why Kraepelin made a new disease from so few cases; one explanation is the rivalry between the Munich and Prague schools. The line between "presenile" Alzheimer's and "senile dementia" was dropped in 1978. 49
The amyloid cascade, three decades on
Hardy and Higgins proposed the amyloid cascade hypothesis in 1992. A 2024 history review notes that the dominant amyloid hypothesis has so far yielded "poor therapeutic results". 49,222 See Amyloid, animated for the evidence on both sides.
Mixed causes in the oldest old
The same 2024 review concludes that mixed dementia, with Alzheimer's changes alongside other brain diseases, is now established as the most frequent cause of dementia in the oldest old. 49
How old was Auguste D.?
One account gives her age as 50 when her symptoms began in 1901; another says she was 56 when she died in 1906. The sources disagree slightly. 49,235
Part 5 · Context & review
Flashcards & mnemonics
Tap the card (or press Space) to flip it. Arrow keys move through the deck. Filter by part, or shuffle.
Mnemonics
Part 5 · Context & review
Quiz
Pick an answer to see the explanation. The mixed quiz draws 20 questions at random; or choose a part.
Part 5 · Context & review
Glossary
Every specialist term used in the app. Type to filter; the arrow on each entry jumps to the module where it matters most.
Part 5 · Context & review
Sources
Every reference returned by the nine research passes behind this app: journal articles, guidelines, drug labels, trial registrations and a few database or search records. The small numbers throughout the app point here.
Research for each area was done separately, so a few works appear under a merged entry. Entries marked check could not be fully verified against the original, so treat them with extra care.