Skip to content

Epidemiology and burden

TL;DR — A single global prevalence for vascular dementia is not defensible because criteria, access to MRI, inclusion of mixed disease, and post-stroke timing change case counts (Pendlebury 2009, PMID 19782001; Sachdev 2014, PMID 24632990). All-cause dementia affected an estimated 57.4 million people in 2019 and is forecast to reach 152.8 million in 2050, but those modeled totals should not be relabeled as vascular dementia (GBD 2019 Dementia Forecasting Collaborators 2022, PMID 34998485). The most quantitative vascular estimates come from stroke cohorts: about 10% have dementia before a first stroke, 10% develop it soon afterward, and more than one-third are affected after recurrent stroke (Pendlebury 2009, PMID 19782001). Hospital-based post-stroke studies estimate any neurocognitive disorder at 53.4%, including 16.5% major disorder, with strong threshold and case-mix effects (Barbay 2018, PMID 30504699). Burden includes executive disability, gait impairment, recurrent vascular events, dependence, caregiver work, and mortality, much of which dementia counts incompletely represent.

Why estimates disagree

Source of variation Direction of effect Evidence
Include pre-stroke dementia Raises post-stroke prevalence Pendlebury 2009, PMID 19782001
Include recurrent strokes Raises prevalence substantially Pendlebury 2009, PMID 19782001
Lower cognitive threshold Raises mild-NCD prevalence Barbay 2018, PMID 30504699
Require memory impairment Misses executive-predominant disease Sachdev 2014, PMID 24632990
Require MRI Improves lesion phenotyping but may select higher-resource cohorts Duering 2023, PMID 37236211
Assign one etiologic label Moves mixed disease between Alzheimer and vascular categories Schneider 2007, PMID 17568013
Hospital recruitment Enriches severe/recurrent stroke and comorbidity Pendlebury 2009, PMID 19782001
Survivor-only follow-up Understates competing mortality Filler 2024, PMID 38101426

All-dementia context

Measure Estimate Population/year Method Source
People living with dementia 57.4 million (95% uncertainty interval 50.4–65.1) Global, 2019 GBD modeling PMID 34998485
Forecast prevalence 152.8 million (130.8–175.9) Global, 2050 demographic/risk forecast PMID 34998485
Proportional increase 166% 2019–2050 derived from modeled totals PMID 34998485

These are contextual denominators, not evidence that a fixed proportion is vascular. Diagnostic substitution and mixed pathology make etiologic fractions especially unstable (O'Brien 2015, PMID 26595643; Schneider 2007, PMID 17568013). O'Brien and Thomas estimated vascular dementia as around 15% of dementia cases after Alzheimer disease, while noting that classification uncertainty is itself a source of that fraction (O'Brien 2015, PMID 26595643). Wolters and Ikram argue that stroke approximately doubles dementia risk and that preventing recurrent stroke remains the practical cornerstone, while urging etiologic studies to include patients regardless of a presumed single cause (Wolters 2019, PMID 31294622).

Jellinger's autopsy syntheses illustrate the same instability: “pure” IVD prevalence in autopsy series ranges from 0.03% to 58% with Western means around 8–15% and Japanese series 22–35%; in a large Viennese series, pure IVD was 9.4% of demented elderly (Jellinger 2002, PMID 12417375; Jellinger 2008, PMID 18525132).

Community incidence rates

The classical community cohorts agree closely on the absolute rate of incident vascular dementia and disagree on nothing except how much of the total it represents.

Cohort Design All-dementia incidence Vascular-dementia incidence Notes
Rotterdam Study (Netherlands) 7,046 dementia-free adults ≥55, 15,135 person-years 10.7 / 1,000 py; rising from 0.6 to 97.2 / 1,000 py across 5-year age bands 1.5 / 1,000 py (AD 7.7) lifetime risk from age 55 was 0.33 in women, 0.16 in men — driven by longevity and very-old-age risk, not by a sex difference before 85 (Ott 1998, PMID 9521184)
NEDICES (central Spain) 3,891 followed a median 3.2 y; 161 incident cases 10.6 / 1,000 py (95% CI 8.9–12.3), European-standardized 1.4 / 1,000 py (95% CI 0.6–2.3); VaD 11.2% of incident cases vs AD 71.4% risk of both AD and VaD rose with the number of vascular risk factors (Bermejo-Pareja 2008, PMID 17727890)
EURODEM pooled (4 European cohorts) 528 incident cases, 28,768 person-years, age ≥65 — cumulative risk to age 95 = 0.04 in both sexes AD incidence at age 90 was 81.7 / 1,000 py in women vs 24.0 in men, but there were no sex differences in vascular-dementia rate or risk (Andersen 1999, PMID 10599770)
Honolulu-Asia Aging Study 3,734 Japanese-American men aged 71–93 age-standardized prevalence 7.6%; cohort prevalence 9.3% 4.2% (primary or contributing) vs AD 5.4%; AD/VaD ratio 1.5 >1 possible cause in 26% of cases; VaD prevalence lower than in Japan but higher than in European-ancestry populations (White 1996, PMID 8805729)

| Hisayama Study (Japan) | 828 dementia-free adults ≥65 followed 17 y; 275 incident cases, 91.2% morphologically evaluated (164 autopsied) | 32.3 / 1,000 py | 9.5 / 1,000 py (AD 14.6; combined dementia 3.8; DLB 1.4) | VaD incidence did not rise steeply after 85, unlike AD and combined dementia; 10-year survival 13.6% vs 29.3% in matched controls (HR 1.67, 95% CI 1.31–2.13), and survival did not differ significantly between subtypes (Matsui 2009, PMID 18977814) |

A sixfold geographic gap in incident vascular dementia is the largest unexplained number on this page. Hisayama's 9.5 per 1,000 person-years against Rotterdam's 1.5 and NEDICES's 1.4 is not a criteria artifact of the usual kind: Hisayama verified 91.2% of its cases morphologically, 164 by autopsy, which is a stronger ascertainment standard than either European cohort applied, and it should if anything have reduced vascular over-labelling. The candidate explanations — higher stroke incidence and hypertension prevalence in mid-century Japanese cohorts, lower AD detection displacing cases into the vascular category, and the environmental gradient the Honolulu-Asia data imply — have never been formally decomposed. Note also that Hisayama's vascular incidence flattened after age 85 while AD's accelerated, which means the vascular fraction of dementia falls with age even where its absolute rate is high.

Two structural facts follow. First, the female excess in dementia is an Alzheimer phenomenon, not a vascular one — EURODEM found identical cumulative vascular-dementia risk in men and women (0.04) while women's cumulative AD risk was 0.22 against men's 0.09 (Andersen 1999, PMID 10599770). Any claim that vascular dementia is sex-patterned needs to specify whether it means incidence (no difference in these data) or absolute case numbers (more women, because more women survive to the ages at which dementia occurs). That said, EURODEM's null is contested. A meta-analysis of 47 studies of community residents aged ≥50 found pooled prevalence of 697 per 10,000 for all-cause dementia, 324 for Alzheimer disease and 116 per 10,000 for vascular dementia, and reported a sex reversal by subtype: at ages 60–69, AD prevalence was 1.9× higher in women (108 vs 56 per 10,000) while VaD prevalence was 1.8× higher in men (56 vs 32 per 10,000) (Cao 2020, PMID 31884487). Prevalence and incidence can diverge if male vascular dementia carries higher case fatality, so the two findings are not strictly contradictory — but no study has reconciled them, and the reconciliation would require sex-stratified survival data that the incidence cohorts did not report. Second, the Honolulu-Asia finding that VaD prevalence tracks migration and environment rather than ancestry — Japanese-American men in Hawaii sitting between Japanese and European-ancestry rates — is the strongest available argument that international differences in the vascular fraction are substantially environmental (White 1996, PMID 8805729).

A national prevalence estimate with a vascular denominator

Most population estimates report all-cause dementia and leave the vascular fraction to inference. The Chinese national cross-sectional survey does not: 46,011 adults aged ≥60 were sampled by multistage stratified cluster methods across 96 sites in 12 provinces (2015–2018), with neuropsychological testing administered in person.

Diagnosis Age- and sex-adjusted prevalence Estimated national cases (≥60 y)
All dementia 6.0% (95% CI 5.8–6.3) 15.07 million (14.53–15.62)
Alzheimer's disease 3.9% (3.8–4.1) 9.83 million (9.39–10.29)
Vascular dementia 1.6% (1.5–1.7) 3.92 million (3.64–4.22)
Other dementias 0.5% (0.5–0.6) 1.32 million (1.16–1.50)
Mild cognitive impairment 15.5% (15.2–15.9) 38.77 million (37.95–39.62)

The vascular fraction here is 27% of dementia cases, well above the ~15% often quoted from Western series (O'Brien 2015, PMID 26595643) and consistent with the autopsy gradient Jellinger reports for Japanese series. Among the risk factors examined, cerebrovascular disease carried by far the largest odds ratio for dementia (5.44, 95% CI 4.95–5.97) — larger than hypertension (1.86, 1.70–2.03), diabetes (2.14, 1.96–2.34), hyperlipidaemia (1.87, 1.71–2.05) or smoking (1.85, 1.67–2.04) — while its association with MCI was far weaker (1.49, 1.36–1.62), which is what a stepwise rather than gradual mechanism predicts. Nine of the identified risk factors were modifiable (Jia 2020, PMID 33271079).

That national survey is, however, the high end of the Chinese estimates, and the disagreement inside the Chinese literature is larger than the disagreement between China and the West. A meta-analysis of 26 studies (100,923 subjects, 977 cases, 1999–2019) put pooled vascular-dementia prevalence at 0.96% (95% CI 0.63–2.1) in adults ≥18, rising with age and higher in north-eastern China, in urban areas, and in men (Jiao 2021, PMID 34178760). A larger and more recent synthesis of 81 observational studies (784,846 participants for prevalence, 1980–2023) estimated pooled vascular cognitive impairment prevalence at 1.54% (95% CI 1.14–1.93), peaking at 2.91% above age 80, with incidence 0.29 per 100 person-years (0.21–0.41) and I² above 90% throughout; gradient-boosting and random-forest models identified age, sex and survey period as the dominant determinants, matching the meta-regression (Liang 2026, PMID 40889799).

These three numbers cannot all be right in the way they are usually quoted. VCI is by construction a superset of vascular dementia, yet the pooled VCI prevalence (1.54%) is essentially identical to the national survey's vascular dementia prevalence (1.6%) and only marginally above the pooled VaD prevalence from an overlapping literature (0.96%). The reconciling possibilities are that the meta-analyses pool studies using dementia-level thresholds under a VCI label, that the national survey's in-person neuropsychological protocol detected cases that community screening studies miss, or that the age denominators (≥60 vs ≥18) are doing most of the work. Until one of these is settled, "the prevalence of vascular dementia in China" should be quoted with its source, its denominator age, and its criteria attached — never as a single figure.

The upstream burden: stroke

Because vascular dementia is downstream of stroke and small-vessel disease, the stroke burden bounds it. In 2019 there were 12.2 million incident strokes (95% UI 11.0–13.6), 101 million prevalent cases, 143 million DALYs and 6.55 million deaths — the second-leading cause of death (11.6% of deaths) and third-leading cause of death and disability combined. Between 1990 and 2019 absolute incident strokes rose 70.0% and prevalent strokes 85.0% while age-standardised incidence fell 17.0% and mortality 36.0%: the improvement is per-capita, the burden is growing. Two subgroup findings matter for future vascular-dementia burden. First, among people younger than 70, prevalence rose 22.0% and incidence 15.0% — younger survivors accumulate more years at risk of delayed cognitive decline. Second, age-standardised stroke mortality was 3.6 times higher and DALY rate 3.7 times higher in low-income than high-income countries. Ischaemic stroke was 62.4% of incidence, intracerebral haemorrhage 27.9%, subarachnoid haemorrhage 9.7%. The five leading risk factors by attributable stroke DALYs were high systolic blood pressure (55.5% of stroke DALYs), high BMI (24.3%), high fasting plasma glucose (20.2%), ambient particulate-matter pollution (20.1%) and smoking (17.6%) (GBD 2019 Stroke Collaborators 2021, PMID 34487721).

At the level of the whole nervous system, the 37 conditions studied in GBD 2021 were collectively the leading group cause of DALYs (443 million, 95% UI 378–521), affecting 3.40 billion people, with stroke ranking first and Alzheimer disease and other dementias fourth by age-standardised DALYs (GBD 2021 Nervous System Disorders Collaborators 2024, PMID 38493795). Falling age-standardised rates alongside rising absolute counts is the same pattern seen in the stroke data, and it is why a declining incidence trend is compatible with an expanding service burden.

Does air pollution act through the vascular route?

Ambient particulate matter is the fourth-largest contributor to stroke DALYs above, which makes the vascular pathway the obvious candidate mechanism for its association with dementia. A formal test did not support it. Among 27,857 Health and Retirement Study participants over 50 followed a mean 10.2 years (4,105 incident dementia), 10-year PM₂.₅ exposure was not associated with incident dementia in fully adjusted models (HR 1.04 per IQR, 95% CI 0.98–1.11); prevalent stroke (HR 1.67, 1.48–1.88) and hypertension (HR 1.15, 1.08–1.23) predicted dementia as expected, but PM₂.₅ was associated with neither stroke (OR 1.08, 0.91–1.29) nor hypertension (OR 0.99, 0.92–1.07), and four-way decomposition found no evidence that either mediated or modified the pollution–dementia relationship (Zhang 2023, PMID 37728927). The exposure contrast (IQR 10.9–14.9 μg/m³) is narrow by global standards, so this is evidence against a vascular mediation pathway at US exposure levels rather than against the association anywhere.

Post-stroke prevalence

Pendlebury and Rothwell reviewed 30 cohorts/7,511 patients. Study methods and case mix explained 93% of variance in early post-stroke dementia rates (Pendlebury 2009, PMID 19782001).

Population definition Dementia estimate (95% CI) Interpretation
Pre-stroke dementia, hospital cohorts 14.4% (12.0–16.8) Referral/case-mix enriched
Pre-stroke dementia, population cohorts 9.1% (6.9–11.3) Lower than hospital cohorts
First-ever stroke; pre-stroke dementia excluded 7.4% (4.8–10.0) Lowest early post-stroke stratum
Recurrent stroke; pre-stroke dementia included 41.3% (29.6–53.1) Highest stratum
Later cumulative incidence 3.0% (1.3–4.7) per year Hospital studies after year one

The authors summarized the clinical pattern as approximately 10% before first stroke, 10% new dementia soon after first stroke, and more than one-third after recurrent stroke (Pendlebury 2009, PMID 19782001). This memorable rule retains the study-era and criteria limitations of its source.

Mild and major post-stroke cognitive disorder

Barbay et al. reviewed 16 hospital studies/3,087 patients and used explicit mild/major thresholds (Barbay 2018, PMID 30504699).

Outcome Pooled prevalence (95% CI)
Any post-stroke NCD 53.4% (46.9–59.8)
Mild post-stroke NCD 36.4% (29.0–43.8)
Major post-stroke NCD 16.5% (12.1–20.8)

Threshold score was the major determinant of total prevalence; recurrent-stroke rate mattered when conservative thresholds were used, and age influenced major-NCD prevalence (Barbay 2018, PMID 30504699). These estimates should not be averaged with dementia-only studies.

Risk architecture

Risk layer Examples Evidentiary interpretation
Background susceptibility age, education/cognitive reserve, prior decline Modifies clinical expression
Vascular exposure hypertension, diabetes, atrial fibrillation, smoking Targets prevention but does not prove etiology in one patient
Brain vulnerability WMH, atrophy, prior infarcts Strongly affects post-stroke trajectory
Index-stroke features severity, strategic location, complications Links acute injury to decline
Recurrent injury recurrent stroke, silent infarcts Accumulates burden
Copathology Alzheimer, Lewy body, TDP-43 Alters phenotype and rate

Across 89 studies and 160,783 patients, baseline cognitive impairment was the strongest predictor of later post-stroke dementia (RR 3.10, 95% CI 2.77–3.47) and of later post-stroke cognitive impairment (RR 2.00, 95% CI 1.66–2.40); diabetes, atrial fibrillation, and WMH were among treatable or potentially actionable factors (Filler 2024, PMID 38101426). A risk factor for later cognition is not automatically proof that modifying it after stroke will prevent dementia.

Geographic and socioeconomic interpretation

Global dementia growth is driven chiefly by demographic expansion and aging, with changes in smoking, BMI, glucose, and education altering forecasts (GBD 2019 Dementia Forecasting Collaborators 2022, PMID 34998485). Vascular cognitive burden is likely shaped by stroke incidence, survival, hypertension detection/control, air pollution, diabetes, and access to acute/secondary care, but vascular-dementia-specific surveillance is too inconsistent to support a uniform regional ranking (Gorelick 2011, PMID 21778438).

Surveillance feature needed Why it matters
Population-based sampling Avoids hospital referral bias
Standard criteria version Makes prevalence comparable
MRI protocol Separates infarct, SVD, CAA, and atrophy patterns
Alzheimer biomarkers Quantifies mixed disease rather than hiding it
Pre-stroke cognition Separates antecedent from incident decline
Recurrent-event capture Avoids underestimating accumulated injury
Death competing risk Prevents survivor bias
Language/education norms Reduces test misclassification

Mortality and disability

Vascular cognitive impairment occurs in populations with high competing risk of recurrent stroke, myocardial infarction, frailty, and death. Dementia itself increases dependence and complicates adherence and rehabilitation, but studies vary in adjustment for vascular disease severity and copathology (O'Brien 2015, PMID 26595643).

A meta-analysis of 78 studies (63,125 people with dementia, 152,353 controls) quantifies the mortality gradient. Any dementia carried HR 5.90 (95% CI 3.53–9.86) for all-cause mortality versus no dementia; mean survival from Alzheimer diagnosis was 5.8 years (SD 2.0); and non-Alzheimer dementias died earlier than Alzheimer disease — HR 1.33 (95% CI 1.21–1.46) for all-cause mortality, 1.12 years shorter survival from diagnosis (95% CI −1.52 to −0.72), and death 1.76 years younger (95% CI −2.66 to −0.85) (Liang 2021, PMID 36097997). In the subtype analysis, shorter survival from disease onset reached significance only for vascular dementia (−1.27 years, 95% CI −1.90 to −0.65) and dementia with Lewy bodies (−1.06 years, 95% CI −1.68 to −0.44) — Lewy body dementia had the highest mortality hazard of all (17.88, 95% CI 5.87–54.46). Heterogeneity exceeded I²=75% for most outcomes and only half the studies were rated good quality, so the ordering is more reliable than the magnitudes. The practical implication is that vascular dementia's burden is compressed into fewer years than Alzheimer disease's, which changes both trial feasibility (competing mortality erodes long-endpoint power) and the shape of the care need.

Economic burden

Vascular-dementia-specific cost data are scarce. A modelling study across 204 countries estimated 9.4 billion hours (95% UI 7.8–11.3) of informal care for 24 brain health disorders in 2021, valued at US$1.7 trillion (1.5–2.1) in forgone earnings, growing 3.2% (2.9–3.5) per year since 2000 — with stroke contributing the most caregiving hours and dementia the most lost earnings (Lastuka 2026, PMID 41748237). Because vascular cognitive impairment sits at the intersection of those two categories, its informal-care cost is double-counted in some accounting frames and invisible in others. Contemporary US Medicare data do separate direct health-care use: among 727,700 beneficiaries with newly identified dementia in 2017–2019, annual cost was $44,896 for vascular dementia, versus $9,034 for beneficiaries without dementia (Parasrampuria 2023, PMID 36749936). This is a payer-cost estimate, not a societal total.

One small population-based study isolates societal cost and points the opposite way from the usual assumption that Alzheimer disease is the costlier diagnosis. Using the Resource Utilization in Dementia instrument in the rural Nordanstig cohort of the Kungsholmen project, societal cost was 23% higher for vascular dementia than for Alzheimer disease (p = 0.02), with the authors emphasising that cardiovascular comorbidity must be counted rather than adjusted away (Wimo 2003, PMID 16191249). The result is preliminary, from one Swedish rural cohort, and more than two decades old. A targeted 2026-09-02 search found contemporary direct health-care costs but no modern VaD-specific societal total.

Burden measurement should include:

  • survival with and without major cognitive disorder;
  • instrumental and basic activities;
  • gait, falls, continence, and mobility;
  • recurrent stroke and hospitalization;
  • caregiver time, distress, and employment effects;
  • institutional care;
  • participation, communication, and decision-making capacity.

These dimensions are not interchangeable with a cognitive score. The 2024 Canadian guideline accordingly places assessment, prevention, management, and caregiver support in one pathway (Swartz 2025, PMID 39822128).

Declining age-standardized stroke rates in some regions could reduce vascular cognitive injury, while improved stroke survival can increase the number living at risk. Better MRI detects more covert disease; wider Alzheimer biomarker use can move mixed cases between labels. Secular change in criteria can therefore resemble biological change.

Apparent trend Alternative explanation
More VCI broader mild criteria or better survival
Less VaD relabeling mixed disease as Alzheimer disease
More post-stroke NCD systematic screening identifies mild cases
Lower dementia incidence risk reduction, cohort education, diagnostic change, or competing death

A systematic review of 43 studies of secular trends found the direction of change to be genuinely country-dependent rather than uniformly downward: prevalence rose (Japan, Canada, France) or was stable (Sweden, Spain, China) in most series, while declines appeared mainly in post-2010 reports from the UK, Sweden, and the USA; incidence was stable or falling in China, Canada, France, Germany, Denmark, Sweden, the Netherlands, the UK, and the USA, and rising in Italy, Japan, Wales, Germany, and the Netherlands — with Germany and the Netherlands appearing in both lists because different datasets disagreed (Stephan 2018, PMID 30347617). Only one study reported trends from a low- or middle-income country (Nigeria, stable incidence). Any claim that vascular dementia is declining because vascular risk factors are better controlled therefore rests on a literature that cannot yet distinguish real change from measurement change, and that has almost no data from the settings where most future cases will occur.

Against the "declining vascular dementia" narrative sits a large contemporary cohort finding in the opposite direction. Among 5.2 million UK primary-care patients, 36,340 people aged 40–75 with atrial fibrillation, no prior stroke, no conventional stroke risk factors, and no anticoagulant were matched to 117,298 controls; over a median 5 years, AF was associated with all-cause dementia (HR 1.17, 95% CI 1.04–1.32) driven entirely by vascular dementia (HR 1.68, 95% CI 1.33–2.12) rather than Alzheimer disease (HR 0.85, 95% CI 0.70–1.03) (Mobley 2024, PMID 38839900). A vascular-dementia excess in exactly the population current risk scores classify as not needing treatment is a burden that better guideline adherence would not remove.

Evidence gaps

  • Population surveillance using harmonized VasCog/VICCCS phenotypes across income settings.
  • Longitudinal studies with both vascular imaging and Alzheimer biomarkers.
  • Incidence estimates separating first stroke, recurrent stroke, covert SVD, and CAA.
  • Burden estimates that include caregiver time and participation.
  • Modeling that treats death and recurrent stroke as competing/intermediate events.

Open questions

  • What fraction of global dementia disability is attributable to preventable vascular brain injury when mixed cases are counted proportionally? (Schneider 2007, PMID 17568013)
  • Do contemporary stroke-unit and secondary-prevention systems reduce delayed cognitive disorder compared with cohorts underlying older estimates? (Pendlebury 2009, PMID 19782001)
  • Which harmonized threshold provides clinically meaningful cross-cultural mild-VCI surveillance? (Barbay 2018, PMID 30504699)
  • How much of forecast dementia growth is modifiable through vascular prevention? (GBD 2019 Dementia Forecasting Collaborators 2022, PMID 34998485)
  • Why is vascular-dementia incidence sex-neutral (cumulative risk 0.04 in both sexes) when almost every vascular risk factor is sex-patterned? (Andersen 1999, PMID 10599770)
  • Is the shorter survival of vascular dementia from onset (−1.27 years vs Alzheimer disease) driven by the cerebrovascular disease, by comorbidity, or by later diagnosis? (Liang 2021, PMID 36097997)
  • Does the vascular-dementia excess in low-conventional-risk atrial fibrillation (HR 1.68) reflect subclinical embolism, hypoperfusion, or shared risk? (Mobley 2024, PMID 38839900)
  • Why is incident vascular dementia 9.5 per 1,000 person-years in Hisayama against 1.4–1.5 in Rotterdam and NEDICES, when Hisayama had the stronger (91% morphological) ascertainment? (Matsui 2009, PMID 18977814; Ott 1998, PMID 9521184; Bermejo-Pareja 2008, PMID 17727890)
  • Is vascular dementia male-predominant in prevalence (1.8× at 60–69) while sex-neutral in incidence, and if so is the difference case fatality? (Cao 2020, PMID 31884487; Andersen 1999, PMID 10599770)
  • How can pooled vascular cognitive impairment prevalence in China (1.54%) be no higher than pooled vascular dementia prevalence from the same literature base (0.96–1.6%)? (Liang 2026, PMID 40889799; Jiao 2021, PMID 34178760; Jia 2020, PMID 33271079)
  • Is the 23% societal cost excess of vascular over Alzheimer dementia real, and does it persist under contemporary stroke care? (Wimo 2003, PMID 16191249)
  • Why does vascular-dementia incidence flatten after age 85 in Hisayama while Alzheimer and combined dementia accelerate? (Matsui 2009, PMID 18977814)
  • Do secular declines reported from the UK, Sweden, and the USA extend to the vascular subtype, and do they occur anywhere outside high-income settings? (Stephan 2018, PMID 30347617)
  • Is the vascular fraction of dementia genuinely ~27% in China against ~15% in Western series, or is the difference diagnostic practice? (Jia 2020, PMID 33271079; O'Brien 2015, PMID 26595643)
  • Why does cerebrovascular disease carry OR 5.44 for dementia but only 1.49 for MCI — is vascular cognitive decline stepwise rather than continuous? (Jia 2020, PMID 33271079)
  • Will the 15–22% rise in stroke incidence and prevalence under age 70 produce a delayed cohort of younger vascular dementia? (GBD 2019 Stroke Collaborators 2021, PMID 34487721)
  • If particulate pollution causes 20.1% of stroke DALYs but does not act on dementia through stroke or hypertension, what is the pathway? (GBD 2019 Stroke Collaborators 2021, PMID 34487721; Zhang 2023, PMID 37728927)

References

  1. Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia. Lancet Neurol. 2009. PMID 19782001
  2. Barbay M, et al. Prevalence of post-stroke neurocognitive disorders in hospital-based studies. Dement Geriatr Cogn Disord. 2018. PMID 30504699
  3. Filler J, et al. Risk factors for cognitive impairment and dementia after stroke. Lancet Healthy Longev. 2024. PMID 38101426
  4. GBD 2019 Dementia Forecasting Collaborators. Estimation of global prevalence of dementia in 2019 and forecast to 2050. Lancet Public Health. 2022. PMID 34998485
  5. Sachdev PS, et al. Diagnostic criteria for vascular cognitive disorders. Alzheimer Dis Assoc Disord. 2014. PMID 24632990
  6. O'Brien JT, Thomas A. Vascular dementia. Lancet. 2015. PMID 26595643
  7. Schneider JA, et al. Mixed brain pathologies account for most dementia cases. Neurology. 2007. PMID 17568013
  8. Duering M, et al. Neuroimaging standards for research into small vessel disease. Lancet Neurol. 2023. PMID 37236211
  9. Gorelick PB, et al. Vascular contributions to cognitive impairment and dementia. Stroke. 2011. PMID 21778438
  10. Swartz RH, et al. Canadian Stroke Best Practice Recommendations: Vascular cognitive impairment, 7th edition update, 2024. Alzheimers Dement. 2025. PMID 39822128
  11. Wolters FJ, Ikram MA. Epidemiology of vascular dementia. Arterioscler Thromb Vasc Biol. 2019;39:1542-1549. PMID 31294622
  12. Jellinger KA. The pathology of ischemic-vascular dementia: an update. J Neurol Sci. 2002;203-204:153-7. PMID 12417375
  13. Jellinger KA. The pathology of "vascular dementia": a critical update. J Alzheimers Dis. 2008;14:107-23. PMID 18525132
  14. Ott A, et al. Incidence and risk of dementia. The Rotterdam Study. Am J Epidemiol. 1998;147:574-80. PMID 9521184
  15. Bermejo-Pareja F, et al. Incidence and subtypes of dementia in three elderly populations of central Spain. J Neurol Sci. 2008;264:63-72. PMID 17727890
  16. Andersen K, et al. Gender differences in the incidence of AD and vascular dementia: the EURODEM studies. Neurology. 1999;53:1992-7. PMID 10599770
  17. Jia L, et al. Prevalence, risk factors, and management of dementia and mild cognitive impairment in adults aged 60 years or older in China: a cross-sectional study. Lancet Public Health. 2020;5:e661-e671. PMID 33271079
  18. GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990-2019. Lancet Neurol. 2021;20:795-820. PMID 34487721
  19. GBD 2021 Nervous System Disorders Collaborators. Global, regional, and national burden of disorders affecting the nervous system, 1990-2021. Lancet Neurol. 2024;23:344-381. PMID 38493795
  20. Zhang B, et al. Hypertension and stroke as mediators of air pollution exposure and incident dementia. JAMA Netw Open. 2023;6:e2333470. PMID 37728927
  21. White L, et al. Prevalence of dementia in older Japanese-American men in Hawaii: the Honolulu-Asia Aging Study. JAMA. 1996;276:955-60. PMID 8805729
  22. Liang CS, et al. Mortality rates in Alzheimer's disease and non-Alzheimer's dementias: a systematic review and meta-analysis. Lancet Healthy Longev. 2021;2:e479-e488. PMID 36097997
  23. Lastuka A, et al. Estimating the costs of informal care for individuals with brain health disorders from 2000 to 2021: a modelling study. Lancet Public Health. 2026;11:e176-e186. PMID 41748237
  24. Stephan BCM, et al. Secular trends in dementia prevalence and incidence worldwide: a systematic review. J Alzheimers Dis. 2018;66:653-680. PMID 30347617
  25. Mobley AR, et al. Thromboembolic events and vascular dementia in patients with atrial fibrillation and low apparent stroke risk. Nat Med. 2024;30:2288-2294. PMID 38839900
  26. Matsui Y, et al. Incidence and survival of dementia in a general population of Japanese elderly: the Hisayama study. J Neurol Neurosurg Psychiatry. 2009;80:366-70. PMID 18977814
  27. Cao Q, et al. The prevalence of dementia: a systematic review and meta-analysis. J Alzheimers Dis. 2020;73:1157-1166. PMID 31884487
  28. Jiao C, et al. The prevalence of vascular dementia in China: a systematic review and meta-analysis from 2009-2019. Iran J Public Health. 2021;50:11-23. PMID 34178760
  29. Liang SZ, et al. Incidence and prevalence of vascular cognitive impairment in China: a systematic review and meta-analysis. Stroke Vasc Neurol. 2026;11:137-147. PMID 40889799
  30. Wimo A, et al. Societal burden and economics of vascular dementia: preliminary results from a Swedish population-based study. Int Psychogeriatr. 2003;15 Suppl 1:251-6. PMID 16191249
  31. Parasrampuria S, et al. Disaggregating the dementia monolith: an analysis of variation in Medicare costs and use by dementia subtype. Alzheimers Dement. 2023;19:3295-3305. PMID 36749936