While the focus of dementia has been on older adults, people younger than 65 can also develop it. How common is such young-onset dementia? In the July 19 JAMA Neurology, researchers led by Sebastian Köhler, Maastricht University, Netherlands, reported a meta-analysis of 74 studies encompassing 2.8 million adults ages 30 to 64. Overall, they estimate 119 per 100,000 people develop young-onset dementia, amounting to 3.9 million cases worldwide. Of those, Alzheimer’s disease is most prevalent, followed by vascular and frontotemporal dementias. Prevalence increased with age and was similar for men and women. It was lower in more-developed countries.

  • Largest meta-analysis on prevalence of young-onset dementia to date.
  • 3.9 million people ages 30 to 64 estimated to have dementia worldwide.
  • Alzheimer’s most prevalent, followed by vascular and frontotemporal dementias.

“The authors demonstrate that, although still relatively rare, the prevalence is higher than previously estimated,” Michelle Mielke, Mayo Clinic, Rochester, Minnesota, wrote (full comment below). “This information is important for developing services and care for individuals and families affected, and for highlighting the fact that low- and middle-income countries will bear the biggest burden of young-onset dementia,” she added. These findings may also help increase awareness of young-onset dementia in primary care settings. 

Compared to late-onset dementia, people with younger-onset are diagnosed more slowly because physicians rarely see these cases. “Low prevalence directly translates to a lack of familiarity in primary care and frontline neurology clinicians,” David Knopman, Mayo Clinic, Rochester, wrote in a JAMA Neurology editorial. Stevie Hendriks of Maastricht University, the paper's first author, noted that young-onset dementia (YOD) may be missed because it can manifest as depression or burnout, which are much more common in middle age. Zoe Arvanitakis, Rush University, Chicago, agreed. "If someone in their 40s or 50s has cognitive problems and a clear cause is not identified, a primary care physician should refer the person to a neurologist or psychiatrist to determine etiology, consider reversible causes of dementia, and get proper care," she told Alzforum.

To capture a snapshot of young-onset dementia, Hendriks and colleagues scoured the literature for papers that reported prevalence or incidence of dementia among people ages 30 to 64 over the past 30 years. They excluded studies that were limited to cohorts known to have high risk of YOD, such as people with Down’s syndrome and residents of care homes. They also ignored studies that might have underreported dementia, such as those that relied on mortality data or death certificates for diagnosis. In the end, 74 studies, comprising a whopping 2.76 million people, qualified for meta-analysis. Most studies had been conducted in Asia, Europe, North America, and Oceania (see image below). The majority of participants were Caucasian.

Global Studies. This color scale shows that most studies included in the meta-analysis were done in Asia, Europe, and North America. [Courtesy of Hendriks et al., JAMA Neurology, 2021. © 2021 American Medical Association. All rights reserved.]

Hendriks and colleagues estimated that 119 of every 100,000 people worldwide have all-cause YOD. This is about twice two widely cited estimates of 42 and 54 per 100,000 people in the U.K. and Japan, respectively (Harvey et al., 2003; Ikejima et al., 2009). Prevalence was lower in the U.S. than Europe at 115 versus 159 per 100,000, respectively. The authors chalk this up to an older European population. Based on a global population of 7.7 billion, the scientists calculated that 3.9 million middle-aged adults live with dementia, about 10 times less than the estimated 44 million people with late-onset dementia (Nichols et al., 2019). 

Just as for late-onset dementia, young-onset prevalence increases with age, Hendriks found. She analyzed the data by five-year age groups. Prevalence crept up from 1 per 100,000 in the 30s, to 4 and 6 per 100,000 in the early 40s and late 40s, respectively. The pace hastened slightly with 10 and 19 per 100,000 in the early and late 50s, then shot up to 77 per 100,000 in the early 60s (see image below). This leap could indicate underestimates in people younger than 60 because studies on that group are scarce, the authors wrote. Martin Rossor, University College London, agreed. “If the focus is broader and includes diseases associated with significant cognitive impairment, such as HIV and head trauma, then the burden of disease is even greater,” he wrote to Alzforum (full comment below).

Dementia By Age. Prevalence of Alzheimer’s, vascular, and frontotemporal dementias varied worldwide (WSP), in Europe (ESP), and in the U.S. (USP). Numbers are per 100,000 people. [Courtesy of Hendriks et al., JAMA Neurology, 2021. © 2021 American Medical Association. All rights reserved.]

Did prevalence of dementia subtypes differ? Hendriks and colleagues grouped studies that specified diagnoses: 20 reported on Alzheimer’s, 13 on vascular dementia (VaD), and 12 on frontotemporal dementia (FTD). Per 100,000 people ages 30 to 64 worldwide, the authors estimated that 41 have AD, 15 have VaD, and two have FTD. “This result was surprising because we generally think of the prevalence equal between Alzheimer’s and frontotemporal dementia in individuals younger than age 65,” Andrew Budson, Boston University, wrote to Alzforum (full comment below).

Though all subtypes were rare in the 30s and 40s, VaD was the most prevalent, at 0.2 to 0.8 per 100,000. AD and FTD were at most 0.1 and 0.3 per 100,000, respectively. However, the authors caution against over-interpreting the data on people under 50 because they are based on only a few studies in each age range. By the late 50s, AD and VaD prevalence had grown several-fold; by the 60s, AD rates had ballooned to 25 per 100,000 and VaD prevalence to triple that in the 50s. FTD rates were consistently lower, hovering at or below 1 per 100,000 across all ages (see image above).

The scientists also sorted prevalence rates by sex. While women had slightly higher prevalence rates than men, it was not statistically significant. “Many physicians I talked to were not surprised because they see a similar number of men and women with YOD,” Hendriks said. Likewise, Eric Larson at the Kaiser Permanente Washington Health Research Institute, Seattle, did not balk at the age similarity, noting the small rates of YOD coupled with wide variance between studies.

The lower rates in the younger age groups may be underestimates because methodology and demographics complicated that data. The researchers found that cohort-based studies reported prevalence rates almost four times higher than those using health care registry data, and all studies of people 50 or younger relied on the latter. Dementia rates were also higher in middle-income countries compared to high-income countries. Since most studies from high-income countries were registry-based, the authors believe that might contribute to their lower prevalence rates. No data was available on low-income countries.

Some of these issues might be sorted out in future work. Larson believes longitudinal studies will be crucial in this regard, noting that almost a third of young-onset cases were disproved within 15 years of diagnosis in one study (Ron et al., 1979). “If you don’t have any follow-up or access to good-quality medical records, it is hard to be certain of a YOD diagnosis,” he said.

While Hendriks agrees that longitudinal studies are better for verifying a diagnosis, she notes that much has changed since the 1970s. “Since then, medicine has evolved rapidly, and the diagnostic criteria for dementia, including young-onset dementia, have improved,” she told Alzforum. “A few ongoing longitudinal studies on YOD should provide more accurate prevalence numbers once they have sufficient follow-up,” Hendriks added.—Chelsea Weidman Burke


  1. The vast majority of research in the dementia field has focused on late-onset dementia. Historically, initial studies focused on estimating the prevalence and incidence, followed by identifying risk factors and biomarkers, and assessing best methods for differential diagnosis and prognosis. Much less of a focus was on young-onset dementia, mainly because it was so rare compared to late-onset dementia.

    This paper by Hendriks and colleagues attempts to more accurately assess the prevalence of young-onset dementia across the world by incorporating 95 unique studies published to date. They demonstrate that, although still relatively rare, the prevalence is higher than previously estimated, even after excluding studies of specific ethnic populations and specific patient groups at high risk (e.g., HIV dementia, Down’s syndrome) or residents of care homes. Thus, the results are underestimates of the true prevalence of young-onset dementia.

    This information is important for developing services and care for individuals and families affected, and for highlighting the fact that low- and middle-income countries will bear the biggest burden of young-onset dementia. An inherent next step is to better identify risk factors that contribute to young-onset dementia (genetic, sociocultural, and biological) that may help to explain some of the cross-country differences and could lead to potential interventions.

  2. This paper, based on a comprehensive review of published papers that contain information about prevalence of dementia in persons younger than 65, and the accompanying editorial, provide good information about the relative frequency of dementia in younger persons. Compared with late-in-life dementia the frequency is dramatically less. I would classify it as rare and unusual, but nonetheless tragic. Younger persons will usually have longer duration of illness and perhaps a delay in diagnosis. The real public health burden of dementia is late-onset dementias, but those with earlier onset also benefit from good and ongoing care.

    The prevalence seemed higher than I expected, but that’s because the sources are so different from real population-based and longitudinal data sources. What was most interesting was the different distributions of diagnoses—the editorial describe some variation [in diagnoses] based on demographics of people who attend the Mayo Clinic—and their geographic studies. The real value of this paper is for neurologists. It may not be that useful for planners or even public health researchers because there is so much variation in the [study] settings.

  3. The greatest burden of dementia is found in the elderly, hence for clinicians, dementia, and thereby the diagnostic approach, is focused on diseases of the elderly such as Alzheimer’s disease and cerebrovascular disease.

    Young-onset dementia (YOD), however, is not trivial and presents particular challenges for families involved. This systematic review and meta-analysis of the prevalence of YOD by Hendriks et al. provides a valuable reminder of the importance of this patient group and offers valuable data for health care planning.

    Unsurprisingly, the published studies analyzed focused on the diagnoses we normally associate with late-life dementia, such as AD. Moreover, the authors excluded cohort specific studies such as those of HIV and trisomy 21. If our focus is broader and included the many diseases that can be associated with significant cognitive impairment, e.g., HIV, head trauma, etc. then the burden of young-onset disease is even greater.

  4. This article is an important reminder that dementia does not only affect older adults, but younger adults between the age of 30 and 64 as well.

    Most individuals with young-onset dementia were between the ages 45 and 64; between ages 30 and 44 dementia was exceedingly rare. This suggests that clinicians should rarely diagnose a patient younger than age 44 with dementia, and then only with a positive biomarker or early onset genetic marker.

    The most common young-onset dementia was Alzheimer’s disease (worldwide prevalence 41.1 per 100,000), which was more than twice as prevalent as vascular dementia (worldwide prevalence 14.9), and more than 17 times as prevalent as frontotemporal dementia (worldwide prevalence 2.3). This result was surprising because we generally think of the prevalence being equal between Alzheimer’s and frontotemporal dementia in individuals younger than age 65. However, the authors point out that frontotemporal dementia may have been misdiagnosed as another disorder given that pathology and biomarkers were rarely available.

  5. The upper age limit of YOD remains disputed. Perhaps one should try to ascertain an empirical cut-off rather than using 65 years as an administrative one. A number of factors should go into this:

    1. If one looks at the inflection point for prevalence estimates of dementia, perhaps 55 years is closer to where the inflection occurs.

    2. Using 65 years makes the big three—AD, VaD, FTD—the major causes of both YOD and LOD, making the differentiation less meaningful.

    3. If one considers the prevalence of AD, the inflection point is possibly later, and closer to 70 years. YOD is a tragic disease and has some unique characteristics. The definition should try to best capture this.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . The prevalence and causes of dementia in people under the age of 65 years. J Neurol Neurosurg Psychiatry. 2003 Sep;74(9):1206-9. PubMed.
  2. . Prevalence and causes of early-onset dementia in Japan: a population-based study. Stroke. 2009 Aug;40(8):2709-14. PubMed.
  3. . Global, regional, and national burden of Alzheimer's disease and other dementias, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019 Jan;18(1):88-106. Epub 2018 Nov 26 PubMed.
  4. . Diagnostic accuracy in presenile dementia. Br J Psychiatry. 1979 Feb;134:161-8. PubMed.

Further Reading


  1. . The Longitudinal Early-onset Alzheimer's Disease Study (LEADS): Framework and methodology. Alzheimers Dement. 2021 May 21; PubMed.
  2. . Age, Dementia, and Diagnostic Candidacy: Examining the Diagnosis of Young Onset Dementia Using the Candidacy Framework. Qual Health Res. 2021 Feb;31(3):498-511. Epub 2020 Nov 19 PubMed.
  3. . Understanding the causes, symptoms and effects of young-onset dementia. Nurs Stand. 2021 Jan 13;36(1):43-50. Epub 2020 Dec 14 PubMed.
  4. . Receiving a diagnosis of young onset dementia: Evidence-based statements to inform best practice. Dementia (London). 2020 Oct 30;:1471301220969269. PubMed.
  5. . The Prevalence and Subtypes of Young Onset Dementia in Central Norway: A Population-Based Study. J Alzheimers Dis. 2019;69(2):479-487. PubMed.

Primary Papers

  1. . Global Prevalence of Young-Onset Dementia: A Systematic Review and Meta-analysis. JAMA Neurol. 2021 Sep 1;78(9):1080-1090. PubMed.
  2. . Young-Onset Dementia-New Insights for an Underappreciated Problem. JAMA Neurol. 2021 Sep 1;78(9):1055-1056. PubMed.