Dementia remains one of the most feared and least well quantified conditions of aging, and a new analysis of a nationally representative United States survey now offers one of the clearest snapshots to date of how many middle-aged and older Americans say they have been diagnosed with the disorder. Drawing on more than 130,000 responses from the National Health Interview Survey, researchers report that roughly 1.6 percent of adults aged 40 to 85 — about one in every hundred — told interviewers that a physician or other health professional had ever informed them they had dementia. The figure, published in PLOS Aging and Health, may sound modest, but it represents hundreds of thousands of diagnosed individuals in the civilian non-institutionalized population, and the patterns behind it carry significant implications for public health planning.
The study, led by Charllote Boateng and Isaiah Osei Duah Junior, took advantage of the NHIS, an annual survey conducted by the National Center for Health Statistics that samples the US civilian non-institutionalized population. Because the survey is designed to be nationally representative, researchers can apply survey weights to their analyses so that the results reflect the entire country rather than just the people who happened to answer the questionnaire. This statistical machinery matters: unweighted convenience samples, clinic-based cohorts, and registry data all suffer from selection biases that can distort prevalence estimates. The NHIS approach, by contrast, reaches households across every region and demographic stratum, producing estimates that are far more generalizable.
Dementia itself is not a single disease but an umbrella term covering a heterogeneous group of neurodegenerative and cerebrovascular disorders. Alzheimer’s disease is the best-known subtype, but vascular dementia, Lewy body dementia, frontotemporal dementia, and mixed forms all fall within the category. What unites them is progressive cognitive decline — worsening memory, impaired executive function, and eventual difficulty performing basic activities of daily living. Large national surveys typically do not distinguish among these subtypes, which is why the authors of the new study framed their outcome as self-reported all-cause dementia. That choice trades diagnostic granularity for population scale, and it means the findings describe the overall burden of diagnosed dementia rather than the distribution of specific diseases.
The ascertainment method deserves scrutiny. Respondents were classified as having dementia if they answered yes to a binary question about whether they had ever been told by a doctor or other health professional that they had the condition. Self-reported diagnoses of this kind have well-documented strengths and weaknesses. On the one hand, they capture conditions that people know about and that clinicians considered serious enough to communicate. On the other hand, they depend on access to healthcare, on clinicians’ willingness to disclose a diagnosis, and on patients’ recall and willingness to report it. People with undiagnosed dementia, those living in institutions such as nursing homes, and those whose diagnosis was never communicated are all invisible to this measure. The true population prevalence of dementia pathology is therefore almost certainly higher than the 1.6 percent reported here.
Within the analytical sample of 130,312 adults aged 40 to 85, women made up 52.3 percent of respondents. Prevalence climbed steadily with age and with the accumulation of chronic diseases, a pattern consistent with decades of clinical and epidemiological research showing that dementia risk rises sharply in later life and that comorbidity burden is a powerful correlate of cognitive decline. The survey-weighted descriptive statistics allowed the team to characterize the population accurately before moving to the core of the analysis: a series of logistic regression models that examined which factors were independently associated with reporting a dementia diagnosis.
The modeling strategy was sequential. The researchers began with univariate models, examining each candidate factor in isolation, and then built multivariate models that adjusted for all other variables simultaneously. This progression matters because crude associations can be misleading. Older age, for example, is correlated with both dementia and with many chronic diseases; only by adjusting for age can researchers determine whether multimorbidity carries its own independent signal. The final results were expressed as adjusted odds ratios with 95 percent confidence intervals and visualized in a forest plot, the standard graphical format for displaying effect estimates across many predictors at once. Odds ratios above one indicate higher odds of reported dementia; those below one indicate lower odds.
The fully adjusted results painted a nuanced picture. Female sex was associated with lower odds of self-reported dementia, a finding that may seem counterintuitive given that women bear a disproportionate share of dementia cases in absolute terms, largely because they live longer. But after adjusting for age and other factors, the women in this survey were less likely to report a diagnosis — a pattern that could reflect differences in diagnosis rates, reporting behavior, or survival, and which the authors present descriptively rather than as evidence of biological protection. Similarly, respondents who were overweight or had obesity showed lower odds of reported dementia in the adjusted models, an association that echoes a puzzling and much-debated literature on midlife versus late-life body mass index and cognitive outcomes, and one that may also reflect reverse causation, since weight loss often precedes a dementia diagnosis.
On the other side of the ledger, several factors were clearly associated with higher odds. Older age and lower educational attainment topped the list, in line with the well-established cognitive reserve hypothesis, which holds that more years of education build neural and cognitive resources that delay the clinical manifestation of brain pathology. Residence in the western United States was independently associated with higher odds, a regional signal that may reflect differences in healthcare access, diagnostic practice, or population composition. Multimorbidity — the co-occurrence of multiple chronic conditions — emerged as a strong correlate, as did depression and anxiety, both of which have intricate, bidirectional relationships with cognitive decline. Finally, enrollment in Medicaid or other public insurance programs was associated with higher odds of a reported diagnosis, a marker that likely captures socioeconomic disadvantage, poorer baseline health, and patterns of healthcare utilization rather than any effect of the insurance itself.
Taken together, the findings suggest that dementia in middle-aged and older Americans is not distributed randomly across the population but clusters along sociodemographic, educational, physical health, and mental health lines. The authors emphasize that these cross-sectional associations cannot establish causation — the survey captures a single moment in time, and it is impossible to know whether depression precedes cognitive decline, follows it, or shares common causes with it. Still, the consistency of the signals with prior longitudinal research lends weight to the interpretation that education, chronic disease management, and mental health care are meaningful levers for population-level dementia risk.
The public health implications are concrete. The authors argue that strategies promoting educational attainment, improving the management of chronic diseases, and strengthening mental health services may help reduce the burden of dementia in the United States. Each of these targets maps directly onto the independent associations identified in the analysis. With the American population aging rapidly and no disease-modifying therapy yet capable of halting most dementias, modifiable social and medical factors represent some of the most tractable opportunities for prevention. A nationally representative estimate of one diagnosed case per hundred adults over 40 provides both a baseline for tracking progress and a reminder that the dementia burden, while concentrated in older ages, begins to register well before retirement age — and that its social gradient runs deep through education, income, insurance status, and mental health.
Subject of Research: Prevalence and determinants of self-reported all-cause dementia among middle-aged and older adults in the United States
Article Title: Prevalence and determinants of self-reported all-cause dementia among middle-aged and older adults in the United States: Results from a nationally representative survey
Article References: Boateng, C., & Osei Duah Junior, I. (2026). Prevalence and determinants of self-reported all-cause dementia among middle-aged and older adults in the United States: Results from a nationally representative survey. PLOS Aging and Health, 1(3), e0000025. https://doi.org/10.1371/journal.page.0000025
Image Credits: AI Generated
DOI: 10.1371/journal.page.0000025
Keywords: dementia, National Health Interview Survey, prevalence, aging, cognitive decline, multimorbidity, depression, education, health disparities, public health, epidemiology, logistic regression
News Source: Ophelia Keating. (October 8, 2026). One in 100 US Adults Over 40 Reports a Dementia Diagnosis, National Survey Finds. Scienmag.
