Usage patterns of cognitive assessments during Medicare's Annual Wellness Visit: A national survey

Abstract

INTRODUCTION

Medicare's Annual Wellness Visit (AWV) is a logical opportunity for early detection of cognitive impairment, but recent data for uptake and cognitive assessments during it are lacking.

METHODS

We surveyed Medicare beneficiaries of a nationally representative panel about use of AWV and cognitive assessments and analyzed associations between uptake and beneficiaries’ characteristics.

RESULTS

Of 1871 participants, 80% had an AWV, among whom 31% underwent formal cognitive testing, 35% were asked about memory problems, including 15% having both. Males, rural residents, non‐Hispanic Black beneficiaries, and those having subjective memory problems or a usual source of care were more likely to be probed, but no characteristics were associated with the probability of undergoing cognitive testing.

DISCUSSION

Use of structured cognitive assessments did not increase with higher AWV uptake. Concerningly, individuals at higher risk of cognitive impairment were not more likely to be assessed, calling for policy interventions to increase assessment rates.

Highlights

• Data are lacking on use of cognitive test during Medicare's Annual Wellness Visit (AWV).

• Use of structured cognitive tests did not increase with the uptake of the AWV.

• Individuals at higher risk of cognitive impairment did not receive more testing.

• More high‐risk patients were asked about memory problems despite lack of testing.

1. BACKGROUND

Despite its high prevalence in an elderly population, early‐stage cognitive decline remains underdiagnosed. ¹ , ² This failure to detect mild cognitive impairment (MCI) can have substantial consequences, as opportunities to intervene are missed. ³ Several causes of MCI can be addressed, such as depression, adverse drug effects, substance use, hypothyroidism, and vitamin B12 deficiency. ⁴ Interventions exist even for non‐reversible etiologies, as summarized in a review by Dubois et al. ⁵ Most importantly, a multidomain intervention consisting of diet, exercise, control of cardiovascular risk factors, and cognitive training has been shown to reduce progression in a large trial. ⁶ More recent meta‐analyses point to effectiveness ⁷ , ⁸ and cost effectiveness of such interventions. ⁹

The recent approvals of two treatments for Alzheimer's disease and related dementias (ADRD), the most common etiology of MCI, ¹⁰ , ¹¹ have lent new urgency to the need for early detection. These monoclonal antibodies remove amyloid beta from the brain and were able to slow down disease progression by ≈ 30%. ¹⁰ , ¹¹ As their labels state that they need to be started in early stages, delays in detection and diagnosis can deprive patients of the opportunity to treat this devastating condition.

Introduced in 2011 under the Affordable Care Act, Medicare's Annual Wellness Visit (AWV) would be a logical opportunity to detect cognitive impairment, because it includes an assessment of cognitive state. To fulfill this requirement, physicians may assess cognition through direct patient observation, caregiver reports, or a brief cognitive test (which, however, is not mandated). ¹² Other required items of the AWV include a general health risk assessment; screenings for depression, substance use, and functional ability; and the development of a personalized prevention plan. A physical exam is not part of the AWV.

Indeed, uptake of AWV is associated with an increase in dementia diagnosis. One study ¹³ found that individuals who used the AWVs were 0.47 percentage points more likely to receive a dementia diagnosis within the subsequent 12 months, a 15% increase. Another study ¹⁴ also found that AWV receipt was associated with higher rates of future MCI and ADRD diagnoses. This association was stronger among racial/ethnic minorities. ¹³ , ¹⁵

AWV use has increased over the years. One study, ¹⁶ based on the Medicare Current Beneficiary Survey (MCBS), found that uptake increased from 8.1% to 23.0% between 2011 and 2016 in Medicare fee‐for‐service beneficiaries. Using Medicare claims, another study ¹⁷ reported an uptake of 31.6% in 2018 for the same population. The same study estimated a cumulative uptake rate (i.e., ever having had AWV) of 55% by 2019 among Medicare fee‐for‐service and advantage plan beneficiaries combined, using self‐reported data from a nationally representative survey. More recent data from MCBS suggested an uptake of 45% in 2020. ¹⁸

Studies also noted disparities in the early years of AWV implementation. From 2011 to 2016, non‐White beneficiaries, rural residents, those with lower income and education, and those lacking a usual place of care had lower rates of AWV use. ¹⁶ Additionally, rural health‐care practices had lower rates of AWV uptake than urban practices in 2015. ¹⁹

RESEARCH IN CONTEXT

1. Systematic review: The authors reviewed the literature using commonly used sources (e.g., PubMed) and identified only one study, which is cited, on the use of cognitive assessment during Medicare's Annual Wellness Visit (AWV). No studies investigated the association of receipt of such assessment with individual characteristics.

2. Interpretation: Our finding suggests that the uptake of AWV is increasing, but the use of structured cognitive assessments is not. Concerningly, individuals at higher risk of cognitive impairment are not more likely to undergo such an assessment, suggesting practice style rather than the patient's risk might determine physicians’ decision to test.

3. Future directions: Our study points to the need to better understand a physician's decision to test, and to develop interventions to increase the use of cognitive assessments, improving early detection of cognitive impairment in primary care and timely access to care.

While use is increasing, it is not clear whether it is associated with increased use of brief cognitive assessments during those visits. In the aforementioned study, ¹⁷ fewer than one third of the 55% who ever had an AWV reported having a structured cognitive test during that visit. Little is known about the use of other, less formal forms of cognitive assessments and the differences between those who did and did not receive assessments.

Against this background, the current study leveraged a nationally representative survey panel to determine the current uptake of the AWV and the rate at which cognitive status was assessed during the visit. We differentiated whether a brief cognitive test was administered from whether only a question about cognitive concerns was asked, both of which would satisfy the formal requirements of the AWV. We also examined the differential rates of cognitive assessments of individuals at higher risk of cognitive impairment, namely older individuals, ²⁰ , ²¹ rural residents, ²² those with subjective memory complaints, ²⁰ , ²³ racial/ethnic minorities, ²⁴ and those with lower income or education. ²¹ , ²⁵

2. METHODS

2.1. Participants and procedures

The study used the Understanding America Study (UAS), ²⁶ a nationally representative, probability‐based Internet panel of ≈ 13,500 US residents of 18+ years old, maintained at the University of Southern California. ²⁷ Panel members receive regular invitations to complete surveys about health, lifestyle, and financial situations, including a core collection administered every 2 years since the recruitment. In contrast to opt‐in convenience samples that tend to lack representativeness, ²⁸ , ²⁹ UAS participants are randomly selected from US Postal Service Delivery Sequence Files, with initial and follow‐up contacts through regular mail. Participants are provided with broadband Internet service and a tablet, if they do not have prior access. This design addresses incomplete access to the Internet in the population ³⁰ and recruits truly representative samples comparable to other well‐established longitudinal studies such as the Health and Retirement Study. ³¹

Between December 19, 2023, and February 28, 2024, UAS participants aged 65+ were invited to a survey asking whether they had a Medicare's AWV and whether a cognitive assessment with a structured test or simple questions occurred during that visit. A total of 2490 individuals (81.5%) participated. We excluded 372 (16.0%) respondents who were not enrolled in Medicare Part B and hence not eligible for the AWV, and another 84 (3.4%) who did not know whether they had even been to an AWV. Further, 163 (8.0%) of the 2034 eligible participants were excluded because of missing data for covariates collected prior to the AWV survey (described below), resulting in an analytic sample of 1871 participants.

The study was approved by the institutional review board of the University of Southern California (UPS 14‐00148). Participants provided electronic informed consent for participation. All procedures were in accordance with the principles expressed in the Declaration of Helsinki.

2.2. Measures

2.2.1. Uptake of AWV and cognitive assessment

Participants aged 65+ were asked whether they “ever had Medicare's AWV.” Those who reported affirmatively were asked whether they had a cognitive assessment at their most recent AWV, either in the form of a cognitive test or a question about memory problems. Additionally, we asked participants whether they had any cognitive assessments in the past 12 months, regardless of the AWV. The survey instrument is included in Appendix A in supporting information.

2.2.2. Beneficiaries’ characteristics

Demographic variables included race/ethnicity (non‐Hispanic [NH] White; NH Black; Hispanic; NH other), education attainment (high school or less; some college; bachelor's degree or more), and yearly household income (≤ $39,999; $40,000–$99,999; ≥ $100,000). Age (in years) was recoded as 65 to 69 or 70+, as prior studies found that the AWV uptake was lower among individuals aged 65 to 69. ³² Urbanicity of the residence was derived from the participant's address using the Census Bureau's urban–rural classification. ³³

We considered individuals who reported usually seeking care at a clinic, health center, doctor's office, Health Maintenance Organization, hospital outpatient department, or the US Department of Veterans Affairs as having a usual source of care, as opposed to those typically visiting an emergency room or urgent care clinic or could not name a usual site (Appendix B in supporting information). Subjective memory concerns (Appendix C in supporting information) were operationalized as reporting fair or poor memory or memory decline compared to 2 years ago, as they are associated with increased risk for and etiologies of cognitive impairment. ³⁴ , ³⁵

2.2.3. Analytic approach

We first examined the characteristics of the analytic sample and then determined the uptake rates of AWV and cognitive assessments. Multivariable logistic regression was used to predict the likelihood of having had an AWV using beneficiaries’ characteristics. For individuals who had an AWV, the same model was used to predict the likelihood of having had a cognitive assessment at the most recent AWV. We treated the two forms of assessment, a structured test and a clinician just asking questions about memory problems, as two separate dependent variables, but also considered having either form as a third outcome. To quantify the magnitude of associations, we computed the marginal predicted distribution as adjusted probability due to a predictor. This quantity implies the proportion of observations having a particular outcome that we would have observed if we were able to force all of the study population to have that predictor value, and it is recommended when making population inferences. ³⁶

3. RESULTS

3.1. Respondent characteristics

Table 1 shows the demographic composition of the analytic sample: two thirds were aged 70+, one fifth belonged to racial/ethnic minority groups, 18% graduated from a high school or had less education, 31% had a yearly household income of < $40,000, and 29% lived in a rural area. Additionally, 18% had no usual source of care, and one fifth had subjective memory problems.

TABLE 1.

Characteristics of the sample (n = 1871).

Variable	Frequency	Percent
Ever had an AWV
No	369	19.7
Yes	1502	80.3
Had a cognitive assessment during last AWV a
No	735	48.9
Had a cognitive test	465	31.0
Clinician asked about memory problems	529	35.2
Age
65–69	631	33.7
70+	1240	66.3
Sex
Male	864	46.2
Female	1007	53.8
Race/ethnicity
NH White	1510	80.7
NH Black	142	7.6
Hispanic	91	4.9
NH Other	128	6.8
Education
High school or less	326	17.4
Some college	694	37.1
Bachelor's or more	851	45.5
Yearly household income
$39,999 or less	588	31.4
$40,000–$99,999	858	45.9
$100,000 or more	425	22.7
Urbanicity of residence
Rural	541	28.9
Urban	1330	71.1
Usual source of care
No	345	18.4
Yes	1526	81.6
Self‐reported memory problem
No	1504	80.4
Yes	367	19.6

Abbreviations: AWV, Annual Wellness Visit; NH, non‐Hispanic.

^a Percent was calculated among those who ever had an Annual Wellness Visit. A total of 227 (15.1%) participants had received both types of cognitive assessments during the last Annual Wellness Visit.

3.2. Uptake of AWV

A total of 1502 (80.3%) participants reported ever having had an AWV (Table 1), including 1302 (69.6%) in the past 12 months. Table 2 (column 1) suggests that individuals aged 70+ (adjusted odds ratio [AOR] = 2.14, 95% confidence interval [CI] = 1.68, 2.72) were more likely to have used the AWV, with a covariate‐adjusted probability of 84% (95% CI = 0.82, 0.86), compared to 72% (95% CI = 0.69, 0.76) among those aged 65 to 69, as were those having a usual source of care (AOR = 2.38, 95% CI = 1.82, 3.12), with an adjusted probability of 83% (95% CI = 0.81, 0.85), compared to 68% (95% CI = 0.63, 0.73) among those not having one. Additionally, females were more likely (AOR = 1.44, 95% CI = 1.13, 1.83) and NH other were less likely (AOR = 0.52, 95% CI = 0.34, 0.78) to have an AWV.

TABLE 2.

AOR predicting whether having an AWV, and if so, whether having a cognitive assessment during the most recent visit.

	Ever had an AWV	Had a cognitive test during last AWV	Clinician asked about memory problems during last AWV	Either a cognitive test or simple probing during last AWV
	(n = 1871)	(n = 1502)	(n = 1502)	(n = 1502)
Had usual source of care a	2.38 *** (1.82,3.12)	1.14 (0.83,1.56)	1.50 * (1.09,2.05)	1.32 (0.99,1.76)
Had self‐reported memory problem b	0.93 (0.69,1.24)	1.22 (0.93,1.61)	1.31 * (1.01,1.72)	1.22 (0.94,1.59)
Urban residence c	1.10 (0.84,1.43)	0.87 (0.68,1.11)	0.78 * (0.61,0.99)	0.77 * (0.61,0.97)
70+ years old d	2.14 *** (1.68,2.72)	1.14 (0.89,1.45)	1.18 (0.93,1.50)	1.10 (0.88,1.38)
Female e	1.44 ** (1.13,1.83)	1.09 (0.87,1.37)	0.71 ** (0.57,0.88)	0.83 (0.67,1.02)
NH Black f	1.20 (0.75,1.92)	1.13 (0.74,1.72)	1.56 * (1.04,2.35)	1.34 (0.90,1.99)
Hispanic f	0.87 (0.52,1.46)	1.19 (0.70,2.02)	1.50 (0.90,2.50)	1.54 (0.93,2.55)
NH other f	0.52 ** (0.34,0.78)	0.67 (0.40,1.13)	0.76 (0.46,1.23)	0.72 (0.46,1.13)
Some college g	1.33 (0.96,1.85)	1.29 (0.92,1.81)	1.20 (0.87,1.67)	1.18 (0.86,1.60)
Bachelor or more g	1.32 (0.94,1.87)	1.10 (0.78,1.56)	1.08 (0.77,1.50)	1.07 (0.78,1.47)
$40,000–$99,999 h	1.18 (0.89,1.56)	1.02 (0.78,1.34)	1.02 (0.78,1.32)	0.99 (0.77,1.27)
$100,000 or more h	1.14 (0.80,1.63)	1.13 (0.81,1.58)	0.85 (0.61,1.18)	0.95 (0.69,1.29)
Constant	0.77 (0.50,1.20)	0.31 *** (0.19,0.51)	0.42 *** (0.26,0.67)	0.89 (0.57,1.40)
Pseudo R 2	0.059	0.006	0.017	0.010

Notes: All models were estimated using the logistic regression.

Abbreviations: AOR, adjusted odd ratio; AWV, Annual Wellness Visit; NH, non‐Hispanic.

^* P < 0.05,

^** P < 0.01

^*** P < 0.001.

^a Compared to no usual source of care.

^b Compared to no self‐reported memory problem.

^c Compared to rural residence.

^d Compared to 65–69 years old.

^e Compared to male.

^f Compared to NH White.

^g Compared to high school or less.

^h Compared to yearly household income of ≤ $39,999.

3.3. Use of cognitive assessments during AWV

Contingent upon having had an AWV, 767 (51.1%) participants reported any form of cognitive assessment during the most recent visit: 465 (31.0%) underwent formal cognitive testing, and 529 (35.2%) were asked about memory problems, including 227 (15.1%) who had both (Table 1). Only 39 (2.6%) of respondents had a recent cognitive test outside of the AWV.

3.4. Predictors of use of cognitive assessments during AWV

The probability of having a structured cognitive test was not associated with any individual characteristics (Table 2, column 2), including known risk factors for cognitive impairment: aged 70+ (AOR = 1.14, 95% CI = 0.89, 1.45), having subjective memory problems (AOR = 1.22, 95% CI = 0.93, 1.61), and urbanicity of residence (AOR = 0.87, 95% CI = 0.68, 1.11), among the other characteristics that we included in the analysis.

In contrast, receiving simple probing of memory problems or not (Table 2, column 3) was associated with several characteristics. Individuals having a usual source of care (AOR = 1.50, 95% CI = 1.09, 2.05) and subjective memory problems (AOR = 1.31, 95% CI = 1.01, 1.72) were more likely to be probed. The adjusted probability was 37% (95% CI = 0.34, 0.40) for those having a usual source of care compared to 28% (95% CI = 0.22, 0.34) for those not having one, and 40% (95% CI = 0.35, 0.46) for those having memory problems compared to 34% (95% CI = 0.31, 0.37) for those not having concerns. Urban residents were less likely to be probed (AOR = 0.78, 95% CI = 0.61, 0.99), with a 34% (95% CI = 0.31, 0.36) adjusted probability in contrast to 39% (95% CI = 0.35, 0.44) for rural residents. Further, females were less likely to be probed (AOR = 0.71, 95% CI = 0.57, 0.88), and NH Black beneficiaries were more likely to be probed than NH White beneficiaries (AOR = 1.56, 95% CI = 1.04, 2.35).

When considering having either form—a structured test or simple questions—as having a cognitive assessment (Table 2, column 4), only urbanicity of residence was a significant and negative predictor (AOR = 0.77, 95% CI = 0.61, 0.97). The adjusted probability was 49% (95% CI = 0.46, 0.52) for urban residents, in contrast to a higher rate of 56% (95% CI = 0.51, 0.60) for rural residents.

4. DISCUSSION

The current study uses a nationally representative panel to examine the uptake of Medicare's AWV and patterns of the use of cognitive assessments during the visit. Several results are noteworthy.

First, the overall AWV use continues to increase: from 8% in 2011, when the AWV was first introduced, to 23% in 2016, ¹⁶ to 32% in 2018, ¹⁷ and then to 45% in 2020. ¹⁸ The study reporting the uptake in 2018 also reported a cumulative uptake (i.e., ever had) of 55% by mid‐2019 using the same panel as the current study, compared to 80% by early 2024 in our study, including 70% reporting having a visit in the past 12 months.

Second, we observe fewer disparities in AWV uptake compared to findings from the early years of AWV implementation. We no longer found lower uptake among NH Black or Hispanic beneficiaries, or those with lower income or education, as shown in a 2016 study, ¹⁶ nor among rural residents observed in 2016. ¹⁶ , ¹⁹ Our observation that individuals who had a usual source of care were more likely to take advantage of the benefit is consistent with the pattern observed in 2016. ¹⁶ Additionally, we see that uptake was higher among beneficiaries aged 70+ (84% compared to 72% for those aged 65–69).

Third, approximately half of the respondents stated that they were either asked about cognitive concerns or underwent a formal test during the AWV. However, only ≈ 30% of respondents underwent formal cognitive testing, similar to the aforementioned 2019 study, ¹⁷ whereas 20% were only asked about cognitive concerns.

Fourth, common risk factors of cognitive impairment (e.g., having a subjective memory complaint, older age, being part of a racial/ethnic minority group) were not significantly associated with the probability of receiving a formal cognitive test; in fact, we did not find any associations for the characteristics that we examined. This result is concerning, particularly in contrast to the observation that some beneficiaries (e.g., those with subjective memory complaints or a usual source of care, NH Black, or those living in a rural area) were more likely to be asked by their clinician about memory concerns. These results imply that neither an elevated risk of cognitive impairment nor an established relationship with a clinician increases the likelihood of being formally tested for cognitive state during the AWV, despite some of them being informally probed by the clinician.

Last, we found that rural residents were more likely to report having been asked about cognitive concerns than their urban counterparts. This finding could be surprising, particularly because rural populations are often perceived to experience more challenges in access to health care. ³⁷ One possible explanation is that rural residents are more likely to have a usual source of care than their urban counterparts, ³⁸ , ³⁹ which could make an informal inquiry easier. Further, one study ⁴⁰ found that rural patients with early onset of ADRD are more likely to be diagnosed and exclusively treated by their primary care providers, whereas the urban patients are more likely to be diagnosed and treated by specialists. Therefore, it is possible that primary care clinicians serving the rural population are more aware of the risk of cognitive impairment in older patients. We did not observe a differential usage of a cognitive test during the AWV, but it is possible that a formal evaluation was scheduled for later, if memory concerns were reported during the probing.

It is possible that clinician or practice level factors play a greater role in use of structured cognitive assessments during the AWV than patient characteristics. Several barriers to early detection have been documented: inadequate time to administer tests, ⁴¹ low confidence in diagnosing dementia and interpreting results from structured cognitive assessments, ⁴² and lack of access to specialists and unfamiliarity with community resources to refer patients to after a diagnosis, ⁴³ , ⁴⁴ just to name a few. Despite these challenges, it is less known why some clinicians conduct the testing anyway. While our patient‐level survey data did not allow analyzing clinicians’ behavior, future research could study the clinician‐ and practice‐level factors that promote or deter the use of formal cognitive assessments in routine care.

At the same time, the AWV is the main avenue to such structured assessments of memory and cognition, as we found that only 2.6% of respondents reported having had such tests outside of the AWV. This finding suggests that the guidance on requirements for cognitive assessment during the AWV needs to be strengthened from the current language ¹² of “Assess cognitive function by direct observation or reported observations from the patient, family, friends, caregivers, and others.” While mandating a formal cognitive test for each beneficiary might not be defensible in light of the current position ⁴⁵ of the US Preventive Services Task Force that the net benefit of population‐level cognitive screening is uncertain, a risk‐based approach to test individuals at higher risk, such as those with cardiometabolic risk factors, positive family history, or subjective memory complaints, could be. Several models for such risk prediction have been published recently. ⁴⁶ , ⁴⁷ , ⁴⁸

Another consideration would be to introduce a billing code for brief cognitive tests in recognition of the fact that commonly used tests, such as the Mini‐Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA), take ≈ 10 minutes or more to administer and score. ⁴⁹ Adding such a code would close an important gap in Medicare coverage for memory care, because the program currently pays for evaluation and care planning in patients with established cognitive impairment (CPT 99483) and for a comprehensive neuropsychological assessment (CPT 96132/96133) but not for a brief cognitive assessment to establish the need for those services.

The development of digital tests, preferably self‐administered ones, that have higher sensitivity and specificity for early‐stage cognitive impairment than the MMSE and MoCA could reduce the burden on clinicians and facilitate uptake. ⁵⁰ , ⁵¹ While multiple digital cognitive tests have been developed in recent years, they are typically designed to meet the Medicare billing requirements for the abovementioned comprehensive neuropsychological assessment rather than for a brief cognitive assessment in primary care. Billing Medicare needs to include “several domains including but not limited to; thinking, reasoning and judgment, for example, acquired knowledge, attention, language, memory, planning and problem solving and visual spatial abilities.” ⁵² Simpler tools that are more suitable for screening in primary care are being developed ⁵⁰ but robust evidence for their accuracy in unselected populations remains lacking, partially because the US Food and Drug Administration currently exerts regulatory discretion for low‐risk technologies that merely provide information aid to clinicians in their diagnosis, that is, no evidence is required for market entry. Similarly, physicians’ limited understanding of signs and symptoms of MCI and the differentiation from dementia warrants educational interventions. ⁵³

4.1. Limitations

Our findings are limited in several ways. First, we largely relied on self‐reported data, which could lead to recall bias, in particular regarding the details of cognitive assessment during and outside of the AWV. Our self‐reported AWV uptake rate (70% in the past 12 months) by early 2024 was considerably higher than the uptake rate of 45% ¹⁸ in 2020 based on MCBS, implying that validation of uptake rates against claims data is warranted. Second, it is possible that patients may not perceive, say, a clock drawing task, as a cognitive test, and thus reported incorrectly. Future studies could, for instance, validate patients’ self‐reports against clinicians’ notes. Last, individuals with advanced dementia are both less likely to participate in an online survey and more likely to undergo a structural cognitive assessment. While this limitation might have caused us to under‐detect assessment rates, the findings are still valid for the population with no or early‐stage cognitive impairment.

5. CONCLUSION

In summary, the current study found that uptake of Medicare's AWV is increasing, with some disparities in uptake disappearing, but the use of a structured cognitive assessment during the visit is not. Concerningly, individuals at higher risk of cognitive impairment are not more likely to undergo such an assessment, suggesting practice style rather than risk determines physicians’ decision to test. These findings imply that more education, better tools, and incentives are needed to increase early detection of cognitive impairment in primary care to avoid undetected disease progression and deprivation of patients’ opportunities to intervene early.

CONFLICT OF INTEREST STATEMENT

Soeren Mattke serves on the board of directors of Senscio Systems and the scientific advisory board of ALZpath and Boston Millennia Partners. He has received consulting and/or speaker fees from Biogen, C2N Diagnostics, Eisai, Eli Lilly, Novartis, Novo Nordisk, and Genentech/Roche. The other authors report no conflicts. Author disclosures are available in the supporting information.

CONSENT STATEMENT

The study was approved by the institutional review board of the University of Southern California (UPS 14‐00148). Participants provided electronic informed consent for participation. All procedures were in accordance with the principles expressed in the Declaration of Helsinki.

Supporting information

Supporting Information

Supporting Information

Liu Y, Ozawa T, Mattke S. Usage patterns of cognitive assessments during Medicare's Annual Wellness Visit: A national survey. Alzheimer's Dement. 2025;21:e14539. 10.1002/alz.14539