Three and a Half Years to a Dementia Diagnosis: Where the Time Goes

For people living with dementia, the road to a diagnosis is long. Across studies from Europe, North America and Australia, an average of three and a half years passed between the first symptoms and a diagnosis^[1]. Where waits for specialist assessment are measured, as in England’s memory services, they have been getting longer^[2].

Most families can say when they first noticed something was wrong: a missed appointment, a repeated question, a change in mood or in how someone handled money. What follows is often a long stretch of not knowing. People put off seeing a doctor, are told it is just age, wait for a referral, and then wait again for scans, tests and a follow-up appointment. This article brings together what research and national audits show about how long that journey takes in different countries, where along the way the time goes, who waits longest, and what guidelines expect.

On average, three and a half years pass between first symptoms and a diagnosis

The best global summary of how long a diagnosis takes is a 2025 systematic review and meta-analysis by Kusoro and colleagues. It measured time to diagnosis as the interval from symptom onset to a final diagnosis. It found only 13 studies, covering 30,257 participants. Pooling the 10 that reported comparable data gave an average of 3.5 years (95% confidence interval 2.7 to 4.3 years) across all types of dementia, which the authors rated as moderate-quality evidence^[1].

The individual studies behind that figure come from Italy, France, Norway, Sweden, the Netherlands, the United States and Australia, and they show how the interval stretches in different places^[1]:

• In a 2017–2019 Italian cognitive neurology clinic, the mean time from symptom onset to diagnosis was 36.9 months.
• In US National Alzheimer’s Coordinating Center data for 2005–2015, the median time from the onset of cognitive decline to a diagnosis of probable Alzheimer’s disease was 4.5 years.
• In French memory clinics in 2006–2016, people with Alzheimer’s disease had waited a mean of 40.8 months from first signs to referral alone.

An Australian survey of carers in New South Wales in 2007 found the same shape. Families noticed the first symptoms a mean of 1.9 years before anyone consulted a health professional about them, and 3.1 years before a firm diagnosis^[3]. Evidence from outside high-income countries is thin, but it points the same way: in 415 people with frontotemporal lobar degeneration syndromes at 12 sites in six Latin American countries, the mean time to diagnosis was 3.33 years from reported symptom onset, or 2.97 years when onset was dated by diagnostic criteria^[4].

Much of the delay comes before anyone is referred, and many people are never diagnosed

Studies that split the journey into stages find that the longest stretch often comes before a specialist is involved. In a Norwegian study of 223 people with young-onset Alzheimer’s disease seen at university hospitals in 2014–2018, the mean time from first symptoms to diagnosis was 5.5 years. Of that, a mean of 3.4 years passed before the first contact with health services; then 10.3 months from that contact to a first hospital visit, and 14.8 months more from the first hospital visit to diagnosis^[1]. The review’s authors note that delays continued even once people were in the care of specialist clinics^[1].

Once someone does see a doctor, how fast things move differs between health systems. A physician survey in France, Germany, Japan, the UK and the US, published in 2020 and funded by a pharmaceutical company, collected 3,346 patient record forms for people with Alzheimer’s disease. About half (51%) had been diagnosed within six months of first presenting with symptoms, but the share ranged from 65% in Japan to 35% in France^[5]. In France, 39% had waited more than nine months, against 16% in Japan^[5]. Carers, not doctors, were the main driver toward a diagnosis: physician suspicion of Alzheimer’s disease was the trigger in only 20% of cases^[5].

Behind every average wait are people who never reach a diagnosis at all. A 2017 meta-analysis of 23 studies, from community and care-home settings worldwide, estimated that 61.7% of people with dementia had not been detected (95% confidence interval 55.0% to 68.0%). Under-detection was higher in China and India than in Europe and North America, and higher in the community than in residential or nursing care^[6]. Single studies tell a similar story: in the Canadian Study of Health and Aging, 64% of community-dwelling older people found to have dementia reported never having seen a doctor about memory problems^[7], and in a study in Tokyo, 56.4% of those assessed as having dementia had no previous diagnosis^[8]. How many people with dementia have a recorded diagnosis, and how countries measure it, is covered in our article on dementia diagnosis rates.

In England, waits for memory services have grown, and few people are diagnosed within six weeks

Few countries publish routine national data on how long people wait for a specialist dementia assessment. England is an exception, through a national audit run by the Royal College of Psychiatrists, so it serves here as a worked case rather than a typical one. In its 2023 spotlight audit, 125 memory assessment services across England submitted 6,148 case notes for people first assessed from January 2023; services in Wales had registered but withdrew because of clinical pressures^[2].

The audit measured the wait in two parts. From referral to the first assessment appointment, people waited an average of 88 days, up from 66 days in 2021. From that first assessment to a diagnosis took a further 63 days on average, up from 58. Altogether, the average wait from referral to diagnosis rose from 124 days in 2021 to 151 days in 2023^[2]. For more than half of patients the wait was over 18 weeks^[2].

Where someone lives makes a large difference. Average waits from referral to diagnosis ranged from 44 days in one service to 347 days in another. In 32 services, none of the audited patients was diagnosed within six weeks of referral^[2]. The audit links this variation to how services are set up: it says the variation suggests that services are commissioned and resourced differently, and reports that the number of people admitted to memory services each week has risen by 130% since 2009, while average staffing levels have not risen to match^[2].

England’s clinical guideline, NICE NG97, sets out when a GP should refer someone to a specialist dementia diagnostic service such as a memory clinic, after reversible causes of cognitive decline such as delirium, depression, sight or hearing loss, or medicines have been looked into^[9]. It also says dementia should not be ruled out solely because someone scores normally on a cognitive test^[9]. The guideline itself does not set a waiting time. The six-week benchmark is the national recommended timeframe from referral to diagnosis, which the audit quotes while recognising that it will not be achievable for every patient^[2]. Clinicians writing in 2026, one of them NHS England’s national clinical director for dementia, have proposed aligning memory services with the 18-week standard used for most NHS outpatient care, and observe that England is falling behind other countries where median waits for diagnosis are shorter^[10]. England’s recorded diagnosis rate, by contrast, has been rising: 66.3% of people aged 65 and over estimated to have dementia had a recorded diagnosis in September 2025, against an expected number modelled from the Cognitive Function and Ageing Study II, compared with 65.5% a year earlier^[10]. More people are reaching a diagnosis, but they are waiting longer for it.

Younger people, people with frontotemporal dementia and people in poorer areas wait longest

The clearest pattern in the research is that younger people wait longer. Pooling six studies, the Kusoro review put the average time to diagnosis in young-onset dementia, where symptoms start before 65, at 4.1 years (95% confidence interval 3.4 to 4.9), compared with 3.5 years across all ages. Younger age at onset and having frontotemporal dementia were consistently linked with a longer interval^[1]. In the Italian clinic study, the mean time from onset to diagnosis was 41.8 months for young-onset dementia and 30.6 months for late-onset dementia; in a Norwegian sample, people with frontotemporal dementia had waited a mean of 59.2 months, against 39.1 months for Alzheimer’s disease^[1]. The Latin American study found the same: younger age at onset went with a longer time to diagnosis^[4]. The wider picture for younger people is covered in our article on young-onset dementia.

Specialist services seem to help. In an Australian study of 242 people with young-onset dementia referred to a specialist inpatient diagnostic unit, those who had accessed a specialist service had a shorter time to diagnosis than those who did not^[1].

Inequality shows up too. In US Health and Retirement Study data for 1998–2013, only 30.1% of the sample received a diagnosis within three years of the onset of dementia, and Black respondents and those with a high-school education or less were less likely to be diagnosed within that time (hazard ratios 0.73 and 0.65, meaning a lower rate of diagnosis within that window)^[1]. In England’s memory services, the wait from first assessment to diagnosis was longest in the most deprived areas, by 15 days compared with the least deprived^[2]. Our article on ethnicity and dementia looks at these gaps in more detail.

New treatments could make waits much longer unless capacity grows

Treatments that target amyloid in early Alzheimer’s disease have raised a new question: can health systems confirm a diagnosis fast enough? Modelling studies by Mattke and colleagues followed people through the pathway such treatment would need: an initial check in primary care, cognitive testing by a dementia specialist, and confirmation of Alzheimer’s pathology by PET scan or cerebrospinal fluid test, with specialist visits and scans limited by current capacity. If people were referred on the basis of a brief cognitive test, as now, the projected wait for specialist visits in the United States was around 50 months; adding a blood test as a triage step cut this to around 12 months^[11]. For England, they projected a wait of 56 months in 2023, rising to 129 months in 2029, or 17 to 25 months with blood-test triage^[12]. For Germany, the projected wait was about 50 months, with people covered by the statutory health system projected to wait 1.9 times as long as those with private cover^[13].

These are projections built on assumptions about demand and capacity, not measured waits. They suggest that systems already struggling to diagnose people within months could not absorb large new demand without more specialist staff, imaging and simpler first-line tests.

What the evidence asks of us

The figures point to a handful of clear asks for health services, commissioners and governments.

• Measure the wait. Kusoro and colleagues found only 13 studies worldwide and no standard way of measuring time to diagnosis^[1]. Few countries publish routine waiting data for specialist assessment the way England’s national audit does. A shared definition, with clear start and end points, would let countries compare and improve.
• Set and resource a standard. England’s six-week benchmark was met for 10% of people in 2023^[2]. Whether the benchmark is six weeks or 18^[10], a target means little unless staffing grows with referrals; the audit found that it had not^[2].
• Close the gap between services and areas. A range from 44 to 347 days between English services, and longer waits in the most deprived areas^[2], are a matter of commissioning, not of the people waiting.
• Make the first step easier. Much of the delay comes before referral^[1], and carers rather than doctors usually start the process^[5]. Public awareness and primary care that takes concerns seriously, as NICE NG97 asks when it warns against ruling dementia out on a normal test score alone^[9], can shorten that stretch.
• Build routes for younger people. The review’s authors conclude that specialist services are likely to be key to shortening time to diagnosis in young-onset dementia^[1].
• Plan capacity before new treatments arrive. Modelling suggests waits of years without simpler triage such as blood tests^[11][12][13].

A diagnosis is the start of support, not the end of the journey. What happens after it, and how often the promised support arrives, is the subject of our article on post-diagnostic support. More articles 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

Waiting for a diagnosis is common and rarely measured. Months and years without an answer are months and years without the explanation, treatment, planning and support a diagnosis should bring. Dementia charities, carers’ groups and people living with dementia have long made the case for timely diagnosis. We have brought the best available evidence on waiting times from several countries into one place, with sources, so that case can be made with figures that stand up.

How to read this data

Different clocks. Studies start and stop the clock at different points: first symptoms as recalled by carers, first contact with a doctor, referral, first specialist appointment, or final diagnosis. Time from first symptoms (as in the Kusoro review and the Australian carer survey) is always much longer than time from referral (as in England’s audit). Figures measured with different clocks should not be compared directly.

Who is counted. Most studies include only people who were eventually diagnosed, often at specialist clinics, so they say nothing about people never diagnosed. Symptom onset is recalled after the event and is imprecise. The meta-analysis pooled 10 studies of varying size and design, and its authors note the lack of standard criteria^[1].

Countries and years. The review’s studies are almost all from Europe, North America and Australia, and some are old: the Australian carer survey dates from 2007. Evidence from low- and middle-income countries on waiting times is very limited. The five-country physician survey was industry-funded and relies on physicians’ records of a sample of their own patients^[5]. England’s audit covers memory assessment services only, for people first assessed in 2023, and is the most recent round published. The studies of undetected dementia used different methods and years and are not directly comparable. The waiting-time models are projections under stated assumptions, not observed waits.

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. Kusoro O, Roche M, Del-Pino-Casado R, Leung P, Orgeta V. Time to diagnosis in dementia: a systematic review with meta-analysis. International Journal of Geriatric Psychiatry, 2025. europepmc.org/article/MED/40716451
2. Royal College of Psychiatrists. National Audit of Dementia Spotlight Audit in Memory Assessment Services 2023: national report. Healthcare Quality Improvement Partnership, 2024. www.rcpsych.ac.uk/docs/default-source/im…
3. Speechly CM, Bridges-Webb C, Passmore E. The pathway to dementia diagnosis. Medical Journal of Australia, 2008. europepmc.org/article/MED/18976188
4. Magrath-Guimet N, Morello-Garcia F, De Los Santos L, et al. Time to diagnosis in FTLD-associated syndromes in Latin America. Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, 2026. europepmc.org/article/MED/42746372
5. Podhorna J, Winter N, Zoebelein H, Perkins T, Walda S. Alzheimer’s diagnosis: real-world physician behavior across countries. Advances in Therapy, 2020. europepmc.org/article/MED/31933051
6. Lang L, Clifford A, Wei L, et al. Prevalence and determinants of undetected dementia in the community: a systematic literature review and a meta-analysis. BMJ Open, 2017. europepmc.org/article/MED/28159845
7. Sternberg SA, Wolfson C, Baumgarten M. Undetected dementia in community-dwelling older people: the Canadian Study of Health and Aging. Journal of the American Geriatrics Society, 2000. europepmc.org/article/MED/11083319
8. Ura C, Okamura T, Inagaki H, et al. Characteristics of detected and undetected dementia among community-dwelling older people in Metropolitan Tokyo. Geriatrics & Gerontology International, 2020. europepmc.org/article/MED/32291907
9. National Institute for Health and Care Excellence. Dementia: assessment, management and support for people living with dementia and their carers (NG97), recommendations. NICE, 2018. www.nice.org.uk/guidance/ng97/chapter/Recommendations
10. Hazan J, Howard R, Isaacs JD. England’s dementia diagnosis landscape: strengths, limitations, and future directions. International Journal of Geriatric Psychiatry, 2026. europepmc.org/article/MED/42391498
11. Mattke S, Hanson M. Expected wait times for access to a disease-modifying Alzheimer’s treatment in the United States. Alzheimer’s & Dementia, 2022. europepmc.org/article/MED/34569686
12. Mattke S, Tang Y, Hanson M. Expected wait times for access to a disease-modifying Alzheimer’s treatment in England: a modelling study. Journal of Health Services Research & Policy, 2024. europepmc.org/article/MED/37931615
13. Mattke S, Tang Y, Hanson M, et al. Current capacity for diagnosing Alzheimer’s disease in Germany and implications for wait times. Journal of Alzheimer’s Disease, 2024. europepmc.org/article/MED/39302379