Annual wellness visits have quietly become a front-line screening tool for cognitive decline, and there’s a legal reason: since 2011, Medicare has required primary care physicians to assess cognitive function as part of every wellness visit for patients 65 and older. This shift wasn’t a trend or passing guideline—it was written into federal regulations after the Affordable Care Act established the Annual Wellness Visit benefit. When your doctor asks you to remember three words, or draws a clock and asks you to fill it in, that’s not a tangent. That’s a mandated part of your preventive care. The urgency behind these cognitive questions has only grown in recent years. About 7.4 million Americans over 65 are living with Alzheimer’s dementia, and that number is projected to nearly double to 13.8 million by 2060.
Yet roughly 61.7 percent of people with dementia remain undiagnosed, and another significant portion are diagnosed so late that treatment options have already narrowed. For the first time in decades, new FDA-approved drugs—lecanemab and donanemab—can slow cognitive decline if caught early enough. That means the brief cognitive check in your annual wellness visit has shifted from optional curiosity to a practical gateway for timely diagnosis and treatment. The cognitive assessment element of the AWV has also become increasingly systematic. Medicare doesn’t mandate a single screening tool, which means your doctor might use the Mini-Cog (a three-minute test combining word recall and a clock-drawing exercise), the Montreal Cognitive Assessment, or observations gathered from concerns raised by family members or the patient’s own Health Risk Assessment. What matters is that cognitive function is documented and tracked—and if concerns emerge, the pathway to more thorough evaluation and potential treatment has been established.
Table of Contents
- How Medicare Made Cognitive Screening a Legal Requirement
- The Underdiagnosis Crisis That Drives Screening
- Specific Screening Tools Now Embedded in Primary Care
- Recent Federal Investments in Dementia Care Detection and Coordination
- Prevalence, Costs, and the Widening Stakes
- How the New Treatments Made Early Screening Actionable
- Why Underdiagnosis Persists Despite Screening Tools
How Medicare Made Cognitive Screening a Legal Requirement
The CMS Manual System transmittal R2159CP, issued in February 2011, codified cognitive assessment as a required element of the Annual Wellness Visit. The regulation is explicit: providers must conduct an “assessment of an individual’s cognitive function by direct observation, with due consideration of information obtained by way of patient report, concerns raised by family members, friends, caretakers, or others.” This language matters because it means the assessment isn’t strictly a test score—it’s an observation-based evaluation that can factor in reports from people who see the patient regularly. That legal foundation has persisted for fifteen years, but the context around it has shifted dramatically. When cognitive screening became a Medicare requirement in 2011, there was no cure for dementia and no disease-modifying treatment.
Doctors screened mostly to identify advanced cases and help families plan. Today, the situation is different: lecanemab received accelerated FDA approval in January 2023 and full approval in July 2023, and donanemab followed with FDA approval on July 2, 2024. Both drugs are designed specifically for early symptomatic Alzheimer’s disease—mild cognitive impairment or mild dementia stages. In the CLARITY-AD clinical trial, lecanemab slowed cognitive decline by 27 percent over 18 months compared to placebo, a modest but measurable benefit in a disease where every month of cognitive preservation matters. However, this benefit comes with a significant tradeoff: amyloid-related imaging abnormalities (ARIA), including brain swelling or microhemorrhages, occur in roughly 1 in 5 treated patients, which is why early detection and medical oversight are crucial.
The Underdiagnosis Crisis That Drives Screening
The reason physicians have become so focused on cognitive screening is the sheer scale of undiagnosed dementia. A global systematic review found that 61.7 percent of people with dementia are undiagnosed in their communities. In the United States specifically, researchers found that among older adults with probable dementia, 39.5 percent were completely undiagnosed, and another 19.2 percent were unaware they had been diagnosed—a combined gap of nearly 58.7 percent. Of patients who do eventually receive a dementia diagnosis, only 54 percent receive it in a timely manner, meaning 46 percent experience either a missed diagnosis or a significantly delayed one.
This underdiagnosis crisis creates a hard reality: without screening, many people who could benefit from early intervention never get identified until symptoms are severe. The Annual Wellness Visit is the primary care system’s practical response to this problem. Because most older adults see their primary care physician at least annually, the AWV represents a scalable moment to ask the memory and thinking questions that might trigger a more formal evaluation. Studies show that when practices integrate structured cognitive screening tools directly into their electronic health records and into the AWV workflow, diagnosis rates and workup rates increase measurably. The limitation worth noting is that even with mandated screening, a patient’s response depends partly on their willingness to report concerns—and some people with early cognitive decline may not recognize or may underreport their symptoms, especially if no family member is present to corroborate concerns.
Specific Screening Tools Now Embedded in Primary Care
The most common brief cognitive tool in primary care is the Mini-Cog, which takes approximately three minutes to administer. It consists of a three-item word recall task (the examiner speaks three words, the patient repeats them, then after a minute or two of other conversation, is asked to recall them) combined with a Clock Drawing Test (the patient is asked to draw a clock showing a specific time). The Mini-Cog is scored 0 to 5, with scores of 0–2 suggesting possible cognitive impairment and scores of 3–5 indicating no impairment. The clock drawing component alone assesses multiple cognitive domains: it requires memory (remembering the task), language comprehension (understanding the instruction), visual-motor coordination, and executive function (planning and execution).
Because it’s free, validated, quick, and requires no special equipment, the Mini-Cog has become the default screening instrument in many primary care settings. When doctors suspect mild cognitive impairment rather than normal aging, they may turn to the Montreal Cognitive Assessment (MoCA), a 30-point test that covers executive function, attention, language, visuospatial skills, memory, and orientation. The MoCA is more sensitive to subtle cognitive changes than the Mini-Cog and better suited for detecting mild cognitive impairment, the gray zone between normal aging and dementia where new treatments are most likely to be effective. However, the MoCA takes 10 to 15 minutes, making it less practical for the brief Annual Wellness Visit itself—it’s more likely to be ordered as a follow-up test in a specialist’s office or memory clinic. The trade-off between brevity and sensitivity is why physicians often use the Mini-Cog as the screening gate and then refer to the MoCA or a neuropsychological evaluation if initial concerns arise.
Recent Federal Investments in Dementia Care Detection and Coordination
Beyond the core requirement, two recent policy changes have intensified the focus on identifying cognitive decline during the AWV. The first is the permanent increase in Medicare reimbursement for the Cognitive Assessment and Care Plan Service (CPT 99483), which rose to approximately $282 in office-based settings as of January 1, 2021. This code is billable once per 180 days and requires a comprehensive multidimensional assessment of cognition, function, safety, neuropsychiatric symptoms, and medication needs, plus a caregiver assessment. The CPT 99483 was also made permanently telehealth-eligible, meaning patients in rural areas or those with mobility limitations can access this more thorough evaluation. The financial pathway this creates is significant: an AWV cognitive screen that identifies concerns can now flow directly into a billable, covered diagnostic workup—a reimbursement structure that didn’t exist when the AWV cognitive requirement was first written.
The second policy shift is the CMS GUIDE Model (Guiding an Improved Dementia Experience), which launched on July 1, 2024, with 390 participating organizations across the country. This is an eight-year voluntary demonstration project designed to test whether paying providers for comprehensive dementia care coordination, caregiver support, and respite services produces better outcomes and lower total costs. Participating practices receive enhanced Medicare payments for dementia-specific care management and caregiver services, which creates organizational incentive to identify and enroll patients with dementia or mild cognitive impairment early. A second cohort of 96 organizations entered the Established Track on July 1, 2025, expanding the model further. Unlike the USPSTF, which concluded in February 2020 that evidence remains “insufficient” to recommend universal cognitive screening of asymptomatic adults 65 and older, these payment models are betting that early identification does lead to better care and outcomes—a practical policy disagreement about screening philosophy.
Prevalence, Costs, and the Widening Stakes
The epidemiological picture makes clear why cognitive screening has become a priority. According to the 2026 Alzheimer’s Disease Facts and Figures published by the Alzheimer’s Association, 7.4 million Americans age 65 and older are currently living with Alzheimer’s dementia, with 74 percent of them aged 75 or older. Prevalence increases dramatically with age: 5.2 percent of those aged 65–74 have Alzheimer’s, compared to 13.8 percent of those aged 75–84, and 35.8 percent of those aged 85 and older. A 2025 study found that the lifetime risk of dementia after age 55 is 42 percent—more than double what was estimated in previous decades—suggesting that dementia is far more prevalent than earlier models predicted. The economic and human costs are staggering.
Alzheimer’s disease was the seventh leading cause of death in the United States in 2022, with 120,122 recorded deaths attributed to it. The total care cost is projected to reach $384 billion in 2025. The burden on families is equally immense: roughly 13 million unpaid caregivers provided more than 19 billion hours of unpaid care in 2025, care valued at more than $446 billion. This creates a system-wide incentive to catch cognitive decline earlier, when family members and healthcare providers can plan and intervene, rather than later, when crisis events force emergency department visits, hospitalizations, and institution-level care. The limitation, however, is that earlier detection doesn’t automatically prevent progression—it extends the knowledge of the disease but doesn’t reverse it. Early detection is valuable because it allows for treatment and planning, but it’s not a substitute for prevention of Alzheimer’s disease itself.
How the New Treatments Made Early Screening Actionable
For decades, the primary value of cognitive screening was prognostic: identifying someone with dementia allowed families to arrange support, adjust living situations, and plan finances. Screening didn’t change the medical course of the disease because no disease-modifying treatment existed. That changed with the approval of anti-amyloid monoclonal antibodies—lecanemab and donanemab—which target the pathological hallmarks of Alzheimer’s disease (amyloid plaques) and have demonstrated modest but measurable slowing of cognitive decline in early stages.
Lecanemab, approved in July 2023 in full form (after an accelerated approval in January), showed a 27 percent slowing of cognitive decline over 18 months in the CLARITY-AD trial, which enrolled roughly 1,800 participants with mild cognitive impairment or mild dementia due to Alzheimer’s disease. Donanemab, approved in July 2024, showed similar magnitude of effect. These drugs only work in the earliest stages of disease—patients with mild cognitive impairment or mild dementia—meaning that someone who has progressed to moderate or advanced dementia will not benefit. That creates an urgent window: cognitive screening in the AWV is now genuinely time-sensitive because detecting someone in the MCI or mild dementia stage unlocks access to a disease-modifying treatment that a three-year delay might eliminate.
Why Underdiagnosis Persists Despite Screening Tools
Despite the legal mandate, the availability of brief screening tools, and the existence of treatment, dementia remains profoundly underdiagnosed. One reason is that primary care practices often lack the time, training, or electronic health record integration to systematically administer and track cognitive screening at every AWV. Another reason is patient and family reluctance: some patients don’t want to know about cognitive decline, and some family members are reluctant to report concerns during a brief office visit. Additionally, cognitive impairment can be subtle in early stages—a person might have mild memory problems that don’t obviously disrupt daily function, and they may not meet the threshold that prompts physician referral to a neurologist or memory specialist.
A practical example illustrates this gap: a 72-year-old woman might score a 3 or 4 on the Mini-Cog at her AWV, technically in the normal range, yet her daughter might report to the doctor that the mother forgets conversations more frequently than she used to. The physician must decide whether to refer for further evaluation or re-check at the next visit. If the practice doesn’t systematize these decisions—if there’s no protocol for tracking borderline scores or acting on family concerns—the patient may not advance to diagnostic testing, and the window for early treatment may close. This underscores why the CMS GUIDE Model and the enhanced reimbursement for CPT 99483 are designed to create organizational systems, not just individual physician decisions: they aim to turn screening into a coordinated diagnostic pathway rather than a one-time question.
- —
