Half of People With Dementia in Hospital Have Delirium, Often Missed

For people living with dementia, an emergency admission to hospital carries a risk of its own: delirium, a sudden and usually temporary state of confusion. Pooled across 81 studies, about half (48.9%) of older people with dementia in hospital had delirium on top of their dementia^[1], and in older people without dementia, an episode of delirium is linked to a much higher chance of being diagnosed with dementia later^[2].

Many families know the pattern. A person who was managing at home becomes unwell, an ambulance is called, and hours later, in a busy emergency department, they seem far more confused than usual. Staff who have never met them may take that for their dementia, when it may be something new and treatable. This article sets out what research and official statistics from several countries show about emergency admissions, short stays and delirium for people living with dementia, and what guidelines already ask services to do.

Many emergency admissions are for conditions better care at home might prevent

“Ambulatory care-sensitive conditions” are illnesses where timely care outside hospital can often prevent an admission. In the United States, a study of Medicare fee-for-service beneficiaries aged 65 and over with dementia living in the community covered 2.35 million people in 2019. Among those admitted to hospital, 23.0% had an admission for one of these conditions in 2019, and 19.0% in 2021, when hospital use fell during the pandemic^[4]. Before the pandemic, Black and Hispanic people with dementia had probabilities of such an admission 2.0 and 1.6 percentage points higher than White people, and those eligible for both Medicare and Medicaid, a marker of low income, 2.8 points higher^[4]. Gaps like these point to differences in access to care, not in the people themselves.

In Germany, linked data from 16 emergency departments in 2016 covered 99,858 patients aged 70 and over. Several ambulatory care-sensitive conditions were among the most common emergency diagnoses, and 31.2% of patients were triaged as less urgent, yet 64.2% were admitted. Only 45.7% had had a geriatric assessment in outpatient care in the year before^[5]. The 14,511 patients with dementia had 1.18 times the odds of dying in hospital, after adjustment for age, sex, other illnesses and the emergency department^[5].

In England, an analysis of general practices found that admission rates for people with dementia, for both avoidable and other conditions, varied widely between practices, with deprivation and being admitted from home among the risk factors. The authors concluded that guidance “may be applied inconsistently, or local services may have differences in referral criteria”^[6]. Overall admission rates and lengths of stay are covered in our companion study on dementia in hospital.

Almost a third of emergency admissions in England last one night or less

England publishes national indicators on emergency admissions of people with dementia, so it serves here as a worked case. In 2019/20, the latest year published, there were 374,432 emergency admissions of people aged 65 and over with dementia recorded anywhere in their hospital record, an age-standardised rate of 3,517 per 100,000 people in that age group, up from 3,365 in 2016/17^[7]. Because the rate is per head of the whole population aged 65 and over, part of any rise may reflect more people being diagnosed and recorded, not only more admissions.

A second indicator counts short stays. In 2019/20, 31.4% of these emergency admissions (116,133 of 369,344) lasted one night or less^[3]. The Office for Health Improvement and Disparities explains why it tracks this: a change of surroundings can cause people with dementia extra distress, and “Admissions to hospital, particularly ones of short duration should be avoided if at all possible for this population”^[3]. Across England’s 128 unitary authorities, the share ranged from 14.4% to 47.9%^[3]. Some of that spread reflects how hospitals count and code admissions, which the indicator’s own notes warn about, so a high figure is a reason to ask why rather than proof of poor care.

Canada shows the time people with dementia spend in emergency departments. In CIHI’s analysis of Canadian emergency department data for 2015–16, older people with dementia (“seniors” in CIHI’s term) waited the same 1.1 hours as those without for a first physician assessment, but spent a median 6.2 hours in the department against 3.7 hours; in Ontario and Alberta, 43% were admitted to an acute bed against 21% of seniors without dementia^[8]. Long hours in a busy, unfamiliar department sit uneasily with delirium guidance, which asks that people at risk are not moved “within and between wards or rooms unless absolutely necessary”^[9].

Emergency departments miss most delirium, and dementia is one of its clearest warning signs

Delirium is common among older people arriving at emergency departments. A 2022 meta-analysis of 30 studies covering 19,534 older emergency department patients put its pooled prevalence at 15.2%, with the figure varying by region and assessment method^[10]. Dementia stands out among the risk factors. In a US study at one teaching hospital, dementia was one of three factors, with dependence in daily activities and hearing impairment, independently associated with arriving at the emergency department with delirium^[11].

The same study found that most delirium went unrecognised. Of 25 people aged 65 and over with delirium confirmed by research staff, 19 (76.0%) were not recognised as delirious by the emergency physician, and of 16 who were then admitted without the diagnosis, 15 were still unrecognised by the admitting hospital doctor^[11]. Almost all (92.0%) had the hypoactive form, where a person becomes quiet, slow and withdrawn rather than agitated, which is easy to miss^[11]. That study excluded people with severe dementia, so it may understate the problem.

A later Canadian study of 1,493 emergency patients aged 65 and over found that nurses missed delirium in 55.1% of cases and physicians in 50.0%. Fifteen people whose delirium was not recognised were sent home; 6.7% of them (one person) had died a week later, against none of those whose delirium was recognised and 1.1% of the rest of the group^[12]. The numbers are small; the authors judged recognition “sub-optimal at ~50%”^[12]. For people with dementia, the danger is that new confusion is put down to the dementia itself. Where it is hard to tell the two apart, NICE’s guideline is plain: “manage the delirium first”^[9].

About half of older people with dementia in hospital also have delirium

Delirium “superimposed on dementia” has been studied for decades. A 2002 systematic review of 14 studies found that it affected between 22% and 89% of hospitalised and community populations aged 65 and over with dementia^[13]. The largest synthesis since, a 2022 meta-analysis of 81 studies of hospitalised older adults, put the pooled prevalence at 48.9%, highest in the Americas and on orthopaedic wards^[1]. Its authors call dementia “the leading risk factor for delirium”, and found that people with both had longer hospital stays, worse cognitive and functional outcomes, and a higher risk of moving into long-term care and of death than people with dementia alone^[1].

Delirium can also begin after admission. In a US community hospital, 32% of 139 people over 65 with dementia (44 people) developed new delirium during their stay. Those who did had a 25% short-term death rate, longer stays and poorer function at discharge^[14].

In a geriatric ward in São Paulo, Brazil, researchers followed 1,409 hospitalisations of acutely ill people aged 60 and over between 2009 and 2015. Delirium with dementia was found in 26% and delirium alone in 21%. After adjusting for other illness, function and nutrition, delirium with dementia was linked to 2.14 times the hazard of dying in hospital, while dementia on its own was not significantly linked^[15]. In other words, it was the delirium that marked the danger. Twelve months after discharge, neither condition was linked to mortality^[15].

Delirium is linked to faster decline, moves into care and new dementia

The evidence suggests delirium often leaves a lasting mark. A 2010 meta-analysis of high-quality studies that adjusted for age, sex, illness severity and existing dementia found that older people who had delirium had 1.95 times the hazard of death over an average of 22.7 months, and 2.41 times the odds of moving into institutional care (33.4% against 10.7%)^[16]. A 2020 meta-analysis of 24 studies, including 3,562 people who had delirium and 6,987 who did not, found that in every study the group with delirium had worse cognition at follow-up^[17].

For people already living with dementia, delirium is associated with faster progression. In Finland’s Vantaa 85+ study of 553 people aged 85 and over, delirium was associated with 3.1 times the odds of worsening dementia severity, and with losing 1.0 more point a year on a standard memory and thinking test across the whole group^[18].

For people without dementia, delirium is associated with being diagnosed later. In New South Wales, Australia, a study matched 55,211 hospital patients aged 65 and over who had delirium with similar patients who did not, and followed them for more than five years. Those with delirium had three times the risk of a new dementia diagnosis (subdistribution hazard ratio 3.00) and 39% higher risk of death, and each further episode of delirium was linked to a 20% higher risk of dementia^[2]. In Lothian, Scotland, where every emergency admission of people aged 65 and over was assessed at the bedside with the 4AT screening tool between 2017 and 2020, 17.6% of 23,558 people without dementia had delirium on admission^[19]. Delirium was linked to a higher risk of later dementia, most strongly in people without multiple long-term conditions (adjusted hazard ratio 3.38)^[19].

These are observational studies, and they cannot prove that delirium causes dementia; delirium may also reveal a brain already vulnerable. The Australian authors judged that their data “support a causal interpretation”^[2], but the question is not settled. Either way, delirium marks people whose recovery and follow-up after hospital deserve close attention.

Much delirium can be prevented, and guidelines say how

Delirium is not inevitable. A 2015 meta-analysis of 14 studies of multicomponent, non-drug prevention programmes found that in 11 studies the interventions cut the odds of delirium by about half (odds ratio 0.47); in the four randomised or matched trials, the odds of delirium were 44% lower (odds ratio 0.56)^[20]. Falls fell too, while effects on length of stay were not statistically clear^[20]. The ingredients are everyday care done well. The package in NICE guidance, for example, includes clear signage, a visible clock and calendar, explaining to people where they are, encouraging them to drink, looking for and treating infection, and helping them stay mobile^[9].

NICE’s delirium guideline, CG103 (first issued in 2010, with some recommendations amended in 2023), names dementia, past or present cognitive impairment and age 65 or over among the risk factors to check when anyone first arrives in hospital or long-term care^[9]. It asks hospitals to:

• assess people at risk within 24 hours of admission and give “a tailored multicomponent intervention package” delivered by a trained team^[9];
• keep people with staff familiar to them and avoid unnecessary moves between wards or rooms^[9];
• be “particularly vigilant” for hypoactive delirium, “which are often missed”, and, where there are signs of delirium, have a competent practitioner assess the person with the 4AT tool^[9];
• support regular visits from family and friends, and tell people at risk and their families that delirium “is common and usually temporary”, encouraging them to report sudden changes^[9].

The guideline expects that early changes “may be reported by the person at risk, or a carer or relative”^[9]. Families know how the person usually is, and treating them as partners in spotting delirium is the right instinct, but it only works if hospitals ask and listen. Medicines used to manage distress in dementia are a separate issue, covered in our study on antipsychotic prescribing.

What the evidence asks of us

• Treat avoidable emergency admissions as a measure of community care. Admissions for conditions that good care outside hospital can often prevent, and stays of one night or less, say as much about primary, community and care-home support as about hospitals. They should be published routinely, broken down by area, deprivation and ethnicity, as the US and English data show is possible^[4][3]. England’s own national indicators currently stop at 2019/20^[3].
• Screen every older person for delirium at the front door. Lothian’s routine 4AT assessment of every emergency admission aged 65 and over shows that it can be done at scale, and its authors argue for routine delirium assessment on admission for all older adults^[19]. Recognition in emergency departments is around 50%^[12].
• Fund the prevention guidelines already recommend. Multicomponent prevention roughly halves the odds of delirium^[20], and NICE expects it within 24 hours of admission for everyone at risk^[9]. That needs trained staff and time, not only a policy.
• Record delirium and follow people up. Given its links with later dementia and faster decline^[2][18], a delirium diagnosis should be documented in the person’s record, as NICE requires^[9], and should prompt follow-up after discharge.
• Make families partners. Visiting policies and staff training should reflect carers’ part in spotting sudden change^[9].

For people living in care homes, see our study on dementia in care homes; more studies on hospital care, diagnosis and support are in the Dementia & Health research library.

Language: this piece uses person-first language ("people living with dementia", "people with dementia"), as most dementia organisations and people with lived experience prefer, and uses clinical terms such as "Alzheimer's disease" or "mild cognitive impairment" only where a study, dataset or diagnosis refers to that specific condition.

Why we're publishing this

Delirium is common among people living with dementia in hospital, often preventable, and still missed about as often as it is found. For families, a sudden change in someone they love can be frightening, and it matters whether anyone in the hospital notices it. We have brought the international evidence together so that dementia organisations can make the case, with sources, for care that keeps people out of hospital when it can, and safe when they have to go in. The gaps here are in services and systems, not in the people who use them.

How to read this data

Most figures here come from observational studies and meta-analyses, which show associations, not proof of cause. Odds ratios and hazard ratios are reported as the studies give them and are not interchangeable. Delirium prevalence depends heavily on who is assessed, when, and with which tool, which is why estimates range so widely. Administrative data rely on what is coded. England’s indicators count admissions with dementia mentioned anywhere in the record, so they depend on diagnosis and coding, and the latest published year is 2019/20, before the COVID-19 pandemic changed hospital use. The CIHI emergency department figures are for 2015–16; visit and admission shares are for Ontario and Alberta, and time in the department comes from Canadian ambulatory care data. Figures from different countries use different definitions and are not directly comparable, and evidence from low- and middle-income countries is thin.

Use this data

Free to cite with attribution to Understanding Aging. Charities, carers' groups, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research and statistics for general information. It is not medical, care or financial advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else's memory or health, please speak to a GP or another qualified health professional.

Sources

1. Han QYC, Rodrigues NG, Klainin-Yobas P, Haugan G, Wu XV. Prevalence, risk factors, and impact of delirium on hospitalized older adults with dementia: a systematic review and meta-analysis. Journal of the American Medical Directors Association, 2022. europepmc.org/article/MED/34648761
2. Gordon EH, Ward DD, Xiong H, Berkovsky S, Hubbard RE. Delirium and incident dementia in hospital patients in New South Wales, Australia: retrospective cohort study. BMJ, 2024. europepmc.org/article/MED/38537951
3. Office for Health Improvement and Disparities. Dementia: short stay emergency admissions (aged 65 years and over), indicator 91745. Fingertips, data to 2019/20. fingertips.phe.org.uk/api/…indicator_ids=91745
4. Cai S, Qin Q, Veazie P, Conwell Y, Temkin-Greener H. Ambulatory care sensitive hospitalizations and disparities among older adults with dementia before and during the pandemic. Journal of the American Geriatrics Society, 2025. europepmc.org/article/MED/40481707
5. Hartl K, Slagman A, Möckel M, et al. Emergency department presentations of older patients in Germany: high rates of ambulatory care-sensitive conditions and increased odds of inpatient mortality in patients living with dementia. BMC Emergency Medicine, 2026. europepmc.org/article/MED/41495668
6. Eyles E, Redaniel MT, Purdy S, Tilling K, Ben-Shlomo Y. Associations of GP practice characteristics with the rate of ambulatory care sensitive conditions in people living with dementia in England: an ecological analysis of routine data. BMC Health Services Research, 2021. europepmc.org/article/MED/34182996
7. Office for Health Improvement and Disparities. Dementia: directly standardised rate of emergency admissions (aged 65 years and over), indicator 91281. Fingertips, data to 2019/20. fingertips.phe.org.uk/api/…indicator_ids=91281
8. Canadian Institute for Health Information. Dementia in hospitals. Dementia in Canada, CIHI. cihi.ca/en/dementia-in-canada/…/dementia-in-hospitals
9. National Institute for Health and Care Excellence. Delirium: prevention, diagnosis and management in hospital and long-term care (CG103), recommendations. NICE, 2010, amended 2023. nice.org.uk/guidance/cg103/chapter/Recommendations
10. Chen F, Liu L, Wang Y, Liu Y, Fan L, Chi J. Delirium prevalence in geriatric emergency department patients: a systematic review and meta-analysis. American Journal of Emergency Medicine, 2022. europepmc.org/article/MED/35841845
11. Han JH, Zimmerman EE, Cutler N, et al. Delirium in older emergency department patients: recognition, risk factors, and psychomotor subtypes. Academic Emergency Medicine, 2009. europepmc.org/article/MED/19154565
12. Lee JS, Tong T, Chignell M, et al. Prevalence, management and outcomes of unrecognized delirium in a national sample of 1,493 older emergency department patients. Age and Ageing, 2022. europepmc.org/article/MED/35150585
13. Fick DM, Agostini JV, Inouye SK. Delirium superimposed on dementia: a systematic review. Journal of the American Geriatrics Society, 2002. europepmc.org/article/MED/12366629
14. Fick DM, Steis MR, Waller JL, Inouye SK. Delirium superimposed on dementia is associated with prolonged length of stay and poor outcomes in hospitalized older adults. Journal of Hospital Medicine, 2013. europepmc.org/article/MED/23955965
15. Avelino-Silva TJ, Campora F, Curiati JA, Jacob-Filho W. Association between delirium superimposed on dementia and mortality in hospitalized older adults: a prospective cohort study. PLOS Medicine, 2017. europepmc.org/article/MED/28350792
16. Witlox J, Eurelings LS, de Jonghe JF, et al. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA, 2010. europepmc.org/article/MED/20664045
17. Goldberg TE, Chen C, Wang Y, et al. Association of delirium with long-term cognitive decline: a meta-analysis. JAMA Neurology, 2020. europepmc.org/article/MED/32658246
18. Davis DH, Muniz Terrera G, Keage H, et al. Delirium is a strong risk factor for dementia in the oldest-old: a population-based cohort study. Brain, 2012. europepmc.org/article/MED/22879644
19. Penfold RS, MacRae C, Sampson EL, Davis DH, et al. Delirium, long-term conditions, and incident dementia in older adults admitted to hospital for emergency care in Lothian, Scotland: a population-based cohort study. The Lancet Healthy Longevity, 2026. europepmc.org/article/MED/41911932
20. Hshieh TT, Yue J, Oh E, et al. Effectiveness of multicomponent nonpharmacological delirium interventions: a meta-analysis. JAMA Internal Medicine, 2015. europepmc.org/article/MED/25643002