News · Brain & Mental Health
42% of older adults with a dementia diagnosis did not report having one, in US survey data
A diagnosis only helps if the person knows about it. Linking survey answers to Medicare records, researchers found dementia went unreported far more often than arthritis, blood pressure or diabetes.
- An observational study of 3,278 US adults aged 65 or older with probable dementia and a diagnosis in Medicare records.
- Across 1998 to 2020 surveys, 42% did not report their dementia diagnosis when asked.
- Among people answering for themselves, 67% did not report it, against 17% for arthritis and 23% for high blood pressure.
- People who did not report it were less likely to have a follow-up visit, a flu shot or a will within a year.
- The study cannot say whether people were never told, forgot, or chose not to say.
Early detection of dementia is a public health goal, backed by memory clinics, screening tools and newer blood tests. But detection only helps if the person, and often their family, know about the diagnosis and can act on it.
A study in JAMA Network Open, a journal of the American Medical Association (JAMA), compared what older Americans told a national survey with what their Medicare records showed. Among people with a recorded dementia diagnosis, 42% did not report having one, a far higher share than for arthritis, high blood pressure or diabetes.
What dementia is and why the diagnosis matters
MedlinePlus describes dementia as a loss of mental functions, such as thinking, memory, and reasoning, that is severe enough to interfere with a person’s daily functioning. It usually worsens over time.
A diagnosis opens doors: to treatment, support services, and planning for money and care while the person can still take part in decisions. The authors note that inadequate disclosure and stigma often leave patients with a documented dementia diagnosis unaware of their condition, which blocks those benefits.
How the dementia reporting study worked
The researchers used the Health and Retirement Study, a long-running national survey of older Americans, linked to Medicare billing records from 1998 to 2020. They focused on people who met the survey’s definition of probable dementia and also had a dementia diagnosis in their claims.
Every two years, participants were asked which conditions a doctor had told them they had. The study compared how often people failed to report dementia with how often they failed to report other diagnosed conditions. It covered 6,158 survey rounds from 3,278 people.
How often dementia went unreported
The gap was large. The adjusted proportion not reporting their diagnosis was 42% among all respondents, and 67% among people answering for themselves rather than through a relative or proxy.
Other conditions were reported far more reliably: 17% of people did not report arthritis, 23% high blood pressure and 39% diabetes. Dementia underreporting was highest before HRS-defined onset (82%), meaning before the survey’s own tests picked up dementia, and stayed high afterwards. Living alone went with higher odds of not reporting the diagnosis.
Why unreported dementia matters for care
The consequences showed up quickly. Patients who underreported were less likely to have a postdiagnosis problem-based visit, a flu vaccination or a witnessed will or trust within a year of diagnosis. Those are exactly the steps a diagnosis is meant to prompt.
The authors conclude that underreporting of dementia diagnosis far exceeded that of other conditions and was associated with reduced postdiagnosis care engagement. The finding points at how diagnoses are communicated, not only at how many are made.
What a survey and records comparison cannot show
The study cannot tell apart the reasons someone did not report a diagnosis: never being told, being told in vague terms, forgetting because of the condition itself, or choosing not to say. Each would call for a different fix.
Billing codes are imperfect, and some diagnoses in claims may be tentative or recorded for testing. The survey’s definition of probable dementia is an estimate, and the data stretch back to 1998, before newer approaches to diagnosis and disclosure.
What this changes for dementia diagnosis and families
For families, the practical step is to ask directly what a memory assessment found, to request a clear explanation, and to ask what support and planning should come next. A diagnosis that is written in a chart but not understood by the patient does little good.
For clinicians and health systems, the study argues for clearer disclosure, follow-up visits after a diagnosis, and involving family members where the patient agrees, especially for people who live alone.
People also ask
What did the study find?
Among 6,158 person-waves from 3,278 people with probable dementia and a claims diagnosis, the adjusted proportion not reporting their diagnosis was 42% overall and 67% among self-respondents. For other conditions it was 17% for arthritis, 23% for hypertension, 39% for diabetes and 46% for depression. Underreporting was 82% before survey-defined onset.
What does underreporting mean here?
When asked in the survey whether a doctor had ever said they had dementia or a memory-related disease, the person said no, even though Medicare billing records showed a dementia diagnosis.
Why would someone not know their diagnosis?
The authors point to inadequate disclosure by clinicians and stigma. Memory problems themselves can also make it hard to remember being told.
What factors were linked to underreporting?
Living alone was linked to higher odds of not reporting the diagnosis, and Medicare Advantage enrollment to lower odds.
Why does it matter if people know?
Knowing allows people to plan finances and care while they can still make decisions, get support, and take part in treatment choices. In this study, people who did not report their diagnosis were less likely to have follow-up care or a will.
What can families do?
Ask the doctor directly what the assessment showed, request that the diagnosis be explained clearly, and ask about local support services and planning. This is general information rather than medical advice.