Yale researchers found a gap between recorded dementia diagnoses and patient reports. Learn what the findings mean for communication, support and care.
On September 17, 2026, Yale News reported that Yale researchers found a large gap between dementia diagnoses recorded in Medicare claims and older adults’ reports of being told about them. The study appeared in JAMA Network Open and was published on August 26, 2026, according to Connecticut Public.
According to Yale News, 67% of participants who answered the survey for themselves said no doctor had told them they had dementia. These participants had probable dementia based on cognitive testing and a dementia diagnosis recorded in Medicare claims. The survey asked about being told of dementia or a memory-related disease.
The researchers used the Health and Retirement Study, a nationally representative survey linked to Medicare claims. They identified participants aged 65 or older using validated cognitive testing. Yale’s report describes thousands of observations from about 3,300 people.
That design brought together different kinds of information: test results, diagnoses recorded in claims and patients’ own survey responses. The central finding was a mismatch between the recorded diagnosis and what participants reported having been told.
Dementia stood out from the comparison conditions in the same participants. Yale reported average underreporting of about 31% for those conditions. The rates were 23% for high blood pressure and 17% for arthritis.
Yale’s summary also reported that underreporting reached 82% in the earliest years after a recorded diagnosis, before measurable cognitive decline. This timing puts the early period after diagnosis at the center of the communication question.
According to Yale News, underreporting was more common among people living alone and among Black older adults. It was also more common among people with less education or lower incomes. People who saw a dementia specialist or were diagnosed in an outpatient clinic reported the diagnosis more often than those diagnosed in a hospital.
Connecticut Public also reported higher odds of nonreporting among people living alone without a caregiver and people without a high school education. For health services, these patterns raise questions about access and support alongside individual recall.
Yale researchers also found associations with care and planning in the following year. Participants who underreported their dementia diagnosis were:
The findings connect the reporting gap with everyday health care and future planning. They help explain why the researchers treated communication as a practical concern rather than a minor difference between records.
Xi Chen, the study’s corresponding author, told Yale News that a diagnosis “that sits in a chart but never registers with the patient delivers almost none of the benefits that early detection is supposed to bring.” His point shifts attention from recording a diagnosis to making the conversation useful.
A practical response is to ask a qualified clinician to explain any diagnosis and what it means for care. This conversation can start with the words used in the medical record rather than assumptions about what a label means. Personal decisions about evaluation and care belong with qualified health professionals.
Questions that may help frame that discussion include:
A trusted person can join if the older adult wants that support. The purpose is to help with questions and understanding while keeping the person at the center of the discussion. For related context, read about bringing memory concerns into a clinical evaluation.
The editorial takeaway for family and friends is to offer help without turning the conversation into a test. Invite questions about what the clinician said. Ask how the person wants support organized instead of deciding that for them.
Discussion can also make room for the person’s priorities: valued relationships, meaningful activities and choices about daily life. A support plan should reflect the person’s priorities rather than treating the diagnosis as their whole identity. FitBrainLab’s guide to protecting identity and everyday choice offers a related perspective.
Chen called for clinicians to have the time, training and support needed to communicate clearly and compassionately, according to Yale News. Connecticut Public reported his call for reimbursement to support disclosure. He also said patients and families should feel able to ask questions.
The practical focus is the discussion after a diagnosis enters the record. Asking what the patient understood and what questions remain gives that discussion a patient-centered starting point. The study also invites researchers to examine how recorded diagnoses become usable information for patients and their chosen supporters.
Nonreporting is not proof of unawareness. Connecticut Public noted that participants may not have been aware of the diagnosis or may have chosen not to report it. The claims-and-survey comparison did not measure the content of a clinical conversation or how a diagnosis was explained.
The result should not become a claim about the awareness of every person with dementia. It concerns self-responding participants selected through cognitive testing and linked claims. Those measures are not the same as confirming what each person heard and understood during a clinical visit.
Several explanations remain possible. Yale’s report discussed stigma, difficulty accepting a diagnosis and inadequate health-system communication. In Connecticut Public’s coverage, Yale neuropsychiatrist Arman Fesharaki-Zadeh described shame and worry about losing independence as possible factors. He also discussed anosognosia, a reduced awareness of one’s medical condition.
The study cannot identify which explanation fits an individual. Neither a survey answer nor membership in a group with higher nonreporting should become a judgment about someone’s honesty or ability to make choices.
The care findings are associations. Yale’s analysis linked underreporting with later differences in doctor visits, flu vaccination and having a will or trust. It does not show that nonreporting caused those differences. Clearer communication is a reasonable goal, but the study does not prove it would change those outcomes.
The practical suggestions were not tested interventions. Asking questions and inviting a trusted supporter follow the authors’ emphasis on communication. The study did not evaluate a family discussion method or a routine of meaningful activities. Those suggestions should guide a respectful conversation rather than carry a promise of improved memory or slower decline.
Yale placed the work in the context of growing attention to early detection and new disease-modifying treatments. The analysis did not measure treatment uptake or the effect of those developments. Its contribution is narrower: it points to a reporting gap that deserves attention alongside efforts to identify dementia earlier.
Future research could look directly at the conversations behind the records. Patient accounts could help identify whether clearer explanations or more opportunities for questions are useful goals. The priority should be to learn from older adults’ experiences rather than infer those experiences from a code alone.
The useful response is neither to assume ignorance nor to dismiss the reporting gap. It is to ask what the person understands, invite their questions and involve a trusted supporter when they want one. That keeps the finding connected to respectful care.
The long-term value of this study lies in a shift of focus: detecting dementia is one task, communicating about it is another. For cognitive health care, a recorded diagnosis should be the beginning of a useful discussion about care and support. It should not replace that discussion.
For older adults and their chosen supporters, the most practical message is to make room for questions while respecting the person’s role in decisions. For health services, the findings give communication a place alongside detection in the research agenda. Progress should be judged not only by what appears in a chart but also by whether people have usable information about their care.
Dementia research headlines leave older adults and their trusted supporters with the task of sorting findings from assumptions, a task FitBrainLab addresses through research-led explanation. Its clear guidance on brain aging helps readers separate established evidence from early or exaggerated claims while keeping memory, independence and an engaged life in view.
Follow FitBrainLab for research-led insights on memory, focus, brain aging, nutrition and mental fitness after 60. Stay connected for new articles, practical guidance and ideas for a sharper, more engaged life.
Build habits that support memory, focus and a curious, connected life.
Read the Blog
