It May Already Be in the Chart

Getting a diagnosis and being told the diagnosis are two separate events, and one of them happens far more often than the other.

Why didn't the doctor say the word?

Because with this disease, unlike almost any other, a great many of them do not.

The Alzheimer's Association's 2015 Facts and Figures report put the numbers side by side. Asked whether they or their caregiver were aware of the diagnosis, seniors diagnosed with these conditions answered:

• Alzheimer's disease — 45%
• Four most common cancers (breast, lung, prostate, colorectal) — 93%
• Cardiovascular disease — 90%
• High blood pressure — 84%
• Arthritis — 81%

And the 45% is the generous reading. Counting only the diagnosed person's own awareness, the figure was 33%.

A family that leaves an appointment confused about what was just said is not misreading the situation. They are inside a well-documented pattern, and one that has a name in national health policy — Healthy People 2020 carried an objective to increase the share of people with diagnosed dementia, or their caregiver, who are aware of the diagnosis, with a baseline around 35%.

Whether to then tell the person themselves is a separate question, and a harder one, taken up in Do You Tell Them They Have Dementia?. This is about the doctor telling the family anything at all.

Could the diagnosis already be in her records?

Quite possibly. Request the notes before you request another referral.

For most families, this is the most useful thirty minutes available.

Under HIPAA, a patient — or a personal representative acting for them — has a right of access to the medical record. Ask the practice for the office visit notes, not a summary. Ask for the problem list, which is where diagnosis codes live. Ask for any cognitive testing results, including the score and the name of the test.

What families find is often one of three things:

That third possibility is common. Less than half of people with Alzheimer's and other dementias have the diagnosis documented in their medical records. So an absence in the chart is not proof of absence in the person.

Why don't doctors disclose it?

Time, training, uncertainty, and a long-standing belief that saying it does not help.

The research on this is unusually candid, and the reasons are mostly structural rather than personal.

Physicians report low confidence recognizing cognitive impairment, little training in dementia care, discomfort with the conversation itself, and doubt about whether an early diagnosis is useful to the patient. Underneath all of it is time. A ten-minute appointment cannot hold a dementia disclosure, and a clinician who knows that often does not start.

Medicare has tried to buy the time back. Since 2017 there has been a dedicated billing code for cognitive impairment care planning — originally G0505, now CPT 99483 — that pays for a comprehensive visit including cognitive testing, functional assessment, medication reconciliation, caregiver assessment and a written care plan. As of 2021 the payment was about $282 in an office setting, it was added to the Shared Savings Program definition of primary care, and it was permanently covered by telehealth.

The catch, documented from the beginning, is that many providers do not know the code exists. Which makes it something a family can raise directly.

Doesn't Medicare require them to check?

Yes — and nobody checks that they checked.

Detection of possible cognitive impairment is a mandatory element of the Medicare Annual Wellness Visit, written into the Affordable Care Act's implementing regulation. It has been required for over a decade.

There is no mechanism to confirm it happens. The Alzheimer's Impact Movement's own policy brief says so plainly: while the visit requires detection of cognitive impairment, there is currently no way to readily track whether physicians conduct these assessments. Some state health departments have resorted to surveying their own doctors to find out.

The U.S. Preventive Services Task Force — the body whose recommendations shape what preventive care gets covered — reviewed screening for cognitive impairment in older adults in 2014, and again in February 2020 in JAMA. Both times it issued an I statement: the evidence is insufficient to assess the balance of benefits and harms of screening. The Task Force was explicit that this is neither a recommendation for nor against screening, and that clinicians should stay alert to symptoms and evaluate as appropriate.

So a family can be told, accurately, both that a cognitive check is a required part of the annual visit and that the national preventive-services body has twice declined to endorse routine screening. That contradiction is real, and it is part of why the system produces so many people who were never told.

The practical reading: the I statement is about screening people with no symptoms. It says nothing about evaluating someone whose family has noticed a change. If there are symptoms, this is not screening — it is a diagnostic workup, and it is standard care.

What about mild cognitive impairment?

Almost nobody gets diagnosed with it, which is why so many families' first diagnosis arrives late.

An analysis of Medicare data from 2017 to 2019, covering 226,756 primary care clinicians and 54,597 practices, compared how much mild cognitive impairment each practice should have expected to see, based on its own patients' age and demographics, against how much it actually diagnosed.

The average detection rate was 8%. Only 0.1% of clinicians and practices diagnosed MCI at anything close to the expected rate. Against roughly 8 million predicted cases in the study population, about 7.4 million went undiagnosed.

That is a detection failure, not a diagnostic disagreement — and it is why so many families' first real conversation about cognition happens after a crisis rather than years before one.

There is a matching failure on the patient side. In 2015 Behavioral Risk Factor Surveillance System data, 55% of adults 45 and older who reported increased confusion or memory loss in the previous year had not discussed it with any health care provider. Even among those whose memory problems were already interfering with work or social activities, 42% had not raised it.

What do I actually ask for?

Specific words, in a specific order, and a request in writing.

• Request the records first. Office visit notes, the problem list, and any cognitive testing with the score and the test name. You may find the answer already there.
• Ask the direct question and wait. "Are you telling me this is dementia?" and then say nothing. The single most useful thing a family member can do in that room is let the silence sit.
• Ask which test was used and what the score was. MMSE, MoCA, Mini-Cog and GPCOG are the common ones. "Her memory seemed fine" is not a test result.
• Ask for a cognitive assessment and care plan visit by name — CPT 99483. It is a separate, longer, billable appointment. It cannot be billed on the same day as the Annual Wellness Visit, so it has to be scheduled on its own.
• Bring an informant. The care-planning code requires an independent historian — someone who knows the person and can describe the changes. That is you, and your account is part of the medical evidence, not a side comment.
• Say what changed, with dates. "She missed three bill payments in June and got lost driving to her sister's in July" carries more weight than "her memory is getting worse."
• If you are told it is normal aging and you do not believe it, ask for a referral to neurology, geriatrics or a memory clinic, and ask for the refusal to be noted in the chart if it is refused.