Validation of Claims Algorithms to Identify Alzheimer's Disease and Related Dementias

Estimates of intra-cluster correlation coefficients from 2018 USA Medicare data to inform the design of cluster randomized trials in Alzheimer's and related dementias

BACKGROUND

Cluster randomized trials (CRTs) are increasingly important for evaluating interventions embedded in health care systems. An essential parameter in sample size calculation to detect both overall and heterogeneous treatment effects for CRTs is the intra-cluster correlation coefficient (ICC) of both outcome and covariates of interest. However, obtaining advance estimates for the ICC can be challenging. When trial outcomes will be obtained from routinely collected data sources, there is an opportunity to obtain reliable ICC estimates in advance of the trial. Using USA national Medicare data, we estimated ICCs for a range of outcomes to inform the design of CRTs for people living with Alzheimer's and related dementias (ADRD).

METHOD

Data from 2018 Medicare Fee-for-Service beneficiaries, specifically, 1,898,812 individuals (≥ 65 years) with diagnosis of ADRD within 3436 hospital service areas (treated as clusters) and 306 hospital referral regions (treated as fixed strata), were used to calculate unadjusted and adjusted ICC estimates for three outcomes: death, any hospitalizations, and any emergency department (ED) visits and three covariates: age, race and sex. We present both overall and stratum-specific ICC estimates. We illustrate their use in sample size calculations for overall treatment effects as well as detecting treatment effect heterogeneity.

RESULTS

The unadjusted overall ICCs for death, hospitalizations, and ED visits were 0.001, 0.010, and 0.017 respectively. Stratum-specific ICCs varied widely across the 306 HRRs: median 0.001, 0.010 and 0.025 for death, hospitalizations, and ED visits respectively and 0.007, 0.001, and 0.080 for age, sex and race. An interactive R Shiny app is provided that allows users to retrieve estimates overlayed on a map of the USA.

CONCLUSIONS

We presented both adjusted and unadjusted ICCs for outcomes as well as unadjusted ICCs for covariates of potential interest from population-level data in the USA and demonstrated how the estimates may be used in sample size calculations for CRTs in ADRD.

High-Risk Medications in Persons Living With Dementia: A Randomized Clinical Trial

Importance

Individuals with Alzheimer disease (AD) and Alzheimer disease-related dementias (ADRD) may be at increased risk for adverse outcomes relating to inappropriate prescribing of certain high-risk medications, including antipsychotics, sedative-hypnotics, and strong anticholinergic agents.

Objective

To evaluate the effect of a patient/caregiver and prescriber-mailed educational intervention on potentially inappropriate prescribing to patients with AD or ADRD.

Design, Setting, and Participants

This prospective, open-label, pragmatic randomized clinical trial, embedded in 2 large national health plans, was conducted from April 2022 to June 2023. The trial included patients with AD or ADRD and use of any of 3 drug classes targeted for deprescribing (antipsychotics, sedative-hypnotics, or strong anticholinergics).

Interventions

Patients were randomized to 1 of 3 arms: (1) a mailing of educational materials specific to the medication targeted for deprescribing to both the patient and their prescribing clinician; (2) a mailing to the prescribing clinician only; or (3) a usual care arm.

Main Outcomes and Measures

Analysis was performed using a modified intention-to-treat approach. The primary study outcome was the dispensing of the medication targeted for deprescribing during a 6-month study observation period. Secondary outcomes included changes in medication-specific mean daily dose and health service utilization.

Results

Among 12 787 patients included in the modified intention-to-treat analysis, 8742 (68.4%) were female, and the mean (SD) age was 77.3 (9.4) years. The cumulative incidence of being dispensed a medication targeted for deprescribing was 76.7% (95% CI, 75.4-78.0) in the patient and prescriber mailing group, 77.9% (95% CI, 76.5-79.1) in the prescriber mailing only group, and 77.5% (95% CI, 76.2-78.8) in the usual care group. Hazard ratios were 0.99 (95% CI, 0.94-1.04) for the patient and prescriber group and 1.00 (95% CI, 0.96-1.06) for the prescriber only group compared with the usual care group. There were no differences between the groups for secondary outcomes.

Conclusions and Relevance

These findings suggest medication-specific educational mailings targeting patients with AD or ADRD and their clinicians are not effective in reducing the use of high-risk medications.

Trial Registration

ClinicalTrials.gov Identifier: NCT05147428.

Regional variation in diagnostic intensity of dementia among older U.S. adults: An observational study

INTRODUCTION

Geographic variation in diagnosed cases of Alzheimer's disease and related dementias (ADRD) could be due to underlying population risk or differences in intensity of new case identification. Areas with low ADRD diagnostic intensity could be targeted for additional surveillance efforts.

METHODS

Medicare claims were used for a cohort of older adults across hospital referral regions (HRRs). ADRD-specific regional diagnosis intensity was measured as the ratio of expected new ADRD cases (estimated using population demographics, risk factors, and practice intensity) compared to observed ADRD-diagnosed cases.

RESULTS

Crude new ADRD diagnosis rate ranged from 1.7 to 5.4 per 100 across HRRs. ADRD-specific diagnosis intensity ranged from 0.69 to 1.47 and varied most for Black, Hispanic, and the youngest (66-74) subgroups. Across all subgroups, ADRD diagnosis intensity was associated with 2-fold difference in receiving an ADRD diagnosis.

DISCUSSION

Where one resides influences the likelihood of receiving an ADRD diagnosis, particularly among those 66-74 years of age and minoritized groups.

HIGHLIGHTS

Rate of new Alzheimer's disease and related dementias (ADRD) case identification varies geographically across the United States. Variation in case identification is greatest in Black, Hispanic, and young-old groups. Intensity of diagnosis (ie, case identification) unrelated to population risk differs across place. Likelihood of receiving an ADRD diagnosis varies 2-fold based on place of residence.

Validation of Claims-Based Frailty Index for Identifying Moderate-to-Severe Dementia in Medicare Beneficiaries

OBJECTIVE

Previous research using the National Health and Aging Trends Study showed that a claims-based frailty index (CFI) could be useful for identifying moderate-to-severe dementia in Medicare claims data. This study aims to validate the findings in an independent cohort.

DESIGN

Retrospective cohort study.

SETTING AND PARTICIPANTS

The study included 658 fee-for-service beneficiaries with dementia who participated in the 2016-2020 Medicare Current Beneficiary Survey in the community-dwelling.

METHODS

We operationalized the Functional Assessment Staging Test (FAST) scale (range: 1-7, stages 5-7 indicate moderate-to-severe dementia) using survey information. CFI (range: 0-1, higher scores indicate greater frailty) was calculated using Medicare claims 12 months before the participants' interview date. Using the previously proposed cut point of 0.280, we calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for identifying moderate-to-severe dementia. Survey procedures were used to account for survey design and weighted to reflect national estimates.

RESULTS

The population had a mean age (SD) of 80.7 (8.9) years, 58.5% female, and 101 beneficiaries (14.8%) had moderate-to-severe dementia. The CFI cut point of 0.280 demonstrated sensitivity 0.49 (95% CI, 0.38-0.59), specificity 0.80 (0.77-0.84), PPV 0.30 (0.23-0.38), and NPV 0.90 (0.87-0.93). Compared with those with a CFI <0.280, beneficiaries with a CFI ≥0.280 had an elevated risk of mortality (2.9% vs 4.1%) over 1 year.

CONCLUSIONS AND IMPLICATIONS

These results confirm our previous findings that CFI among beneficiaries with a dementia diagnosis is a useful measure of moderate-to-severe dementia for Medicare claims data.

Initiation and Persistence of Antipsychotic Medications at Hospital Discharge Among Community-Dwelling Veterans With Dementia

OBJECTIVES

Adults with dementia are frequently prescribed antipsychotic medications despite concerns that risks outweigh benefits. Understanding conditions where antipsychotics are initially prescribed, such as hospitalization, may offer insights into reducing inappropriate use.

DESIGN, SETTING, PARTICIPANTS

Retrospective cohort study of community-dwelling veterans with dementia aged ≥68 with VA hospitalizations in 2014, using Veterans Health Administration (VA) and Medicare data.

MEASUREMENTS

The primary outcome was new outpatient antipsychotic prescription at hospital discharge. We used generalized estimating equations to study associations between antipsychotic initiation and patient, hospitalization, and facility characteristics. Among veterans with antipsychotic initiation, we used a cumulative incidence function to evaluate discontinuation in the year following hospitalization, accounting for competing risks.

RESULTS

4,719 community-dwelling veterans with dementia had VA hospitalizations in 2014; 264 (5.6%) filled new antipsychotic prescriptions at discharge. Antipsychotic initiation was associated with discharge unit (surgical vs medical, OR 0.41, 95% CI 0.19-0.87; psychiatric vs medical, OR 6.58, 95% CI 4.48-9.67), length of stay (OR 1.03/day, 95% CI 1.02-1.05), and delirium diagnosis (OR 2.61, 95% CI 1.78-3.83), but not demographic or facility characteristics. Among veterans with antipsychotic initiation, the 1-year cumulative incidence of discontinuation was 18.2% (n = 47); 15.9% (n = 42) of those who were alive and not censored remained on antipsychotics at 1 year.

CONCLUSIONS

Antipsychotic initiation at hospital discharge was uncommon among community-dwelling veterans with dementia; however, once initiated, antipsychotic persistence at 1 year was common among those who remained community-dwelling. Hospitalization is a contributor to potentially-inappropriate medications in the community, suggesting an opportunity for medication review after hospitalization.

Association of nursing home antipsychotic reduction policies with antipsychotic use in community dwellers with dementia

BACKGROUND

Antipsychotic and other psychotropic medication use is prevalent among community-dwelling older adults with dementia despite the potential for adverse effects. Two Centers for Medicare & Medicaid Services (CMS) initiatives, the National Partnership to Improve Dementia Care ("the Partnership") and the Five Star Quality Rating System for antipsychotic use reporting, have been successful in reducing antipsychotic use in nursing home residents. We assessed if these initiatives had a spillover effect in antipsychotic and other psychotropic medication use among community dwellers with dementia due to potential overlap in prescribers across settings.

METHODS

Among community-dwelling older adults with dementia, we examined psychotropic medication class use (i.e., antipsychotics, antidepressants, anxiolytics, anticonvulsants/mood stabilizers, antidementia) in 2010-2017 Medicare fee-for-service claims using interrupted time series analyses across three periods ("Pre-Partnership": July 1, 2010 to March 31, 2012; "Post-Partnership": April 1, 2012 to January 31, 2015; "Five Star Quality Rating": February 1, 2015 to December 31, 2017).

RESULTS

We included 1,289,401 community dwellers with dementia contributing 26,609,697 person-months. The mean age was 80 years, most were female (70%), approximately 80% were non-Hispanic Whites, 10% were non-Hispanic Blacks, and 5% were Hispanic ethnicity. Antipsychotic use was declining pre-Partnership (β = -0.06, 95% CI: -0.08, -0.05) and post-Partnership (β = -0.02, 95% CI: -0.02, -0.01). Post-Five Star Quality Rating, antipsychotic use remained stable with a nearly flat slope (β = -0.01, 95% CI: -0.01, 0.00). Anticonvulsant and antidepressant use increased and anxiolytic and antidementia medication use decreased among community-dwelling older adults with dementia.

CONCLUSIONS

These two CMS policies on antipsychotic use for nursing home residents were not associated with a spillover effect to community-dwelling older adults with dementia. Strategies to monitor the appropriateness of psychotropic medication use may be warranted for community-dwellers with dementia.

Risk of Dementia Diagnosis After Injurious Falls in Older Adults

Importance

Emerging evidence suggests that mild cognitive impairment, which is a precursor to Alzheimer disease and related dementias (ADRD), places older adults at increased risk for falls. However, the risk that an older adult develops dementia after experiencing a fall is unknown.

Objective

To determine the risk of new ADRD diagnosis after a fall in older adults.

Design, Setting, and Participants

This retrospective cohort study examined Medicare Fee-for-Service data from 2014 to 2015, with follow-up data available for at least 1 year after the index encounter. Participants included adults aged 66 years and older who experienced a traumatic injury that resulted in an emergency department (ED) or inpatient encounter and did not have a preexisting diagnosis of dementia. Data analysis was performed from August 2023 to July 2024.

Exposures

Experiencing a fall compared with other mechanisms of injury, defined by International Classification of Diseases, Ninth Revision (ICD-9) and ICD-10 external cause of injury codes.

Main Outcomes and Measures

The hazard of new ADRD diagnosis within 1 year of a fall, assessed by performing a Cox multivariable competing risk model that controlled for potential confounders while accounting for the competing risk of death.

Results

The study included 2 453 655 older adult patients who experienced a traumatic injury; 1 522 656 (62.1%) were female; 124 396 (5.1%) were Black and 2 232 102 (91.0%) were White; and the mean (SD) age was 78.1 (8.1) years. The mechanism of injury was a fall in 1 228 847 incidents (50.1%). ADRD was more frequently diagnosed within 1 year of a fall compared with other injury mechanisms (10.6% [129 910 of 1 228 847] vs 6.1% [74 799 of 1 224 808]; P < .001). The unadjusted hazard ratio (HR) of incident dementia diagnosis after a fall was 1.63 (95% CI, 1.61-1.64; P < .001). On multivariable Cox competing risk analysis, falling was independently associated with an increased risk of dementia diagnosis among older adults (HR, 1.21 [95% CI, 1.20-1.21]; P < .001) after controlling for patient demographics, medical comorbidities, and injury characteristics, while accounting for the competing risk of death. Among the subset of older adults without a recent skilled nursing facility admission, the HR was 1.27 (95% CI, 1.26-1.28; P < .001).

Conclusions and Relevance

In this cohort study, new ADRD diagnoses were more common after falls compared with other mechanisms of injury, with 10.6% of older adults being diagnosed with ADRD in the first year after a fall. To improve the early identification of ADRD, this study's findings suggest support for the implementation of cognitive screening in older adults who experience an injurious fall that results in an ED visit or hospital admission.

Hospice Enrollment and Central Nervous System-Active Medication Prescribing to Medicare Decedents with Dementia

This study explores the extent to which hospice enrollment is associated with CNS–active medication exposure among Medicare decedents with Alzheimer disease and related dementias.

Identification of older adults with Alzheimer's and related dementias among patients newly diagnosed with cancer: A comparison of methodological approaches

INTRODUCTION

Research efforts to characterize and evaluate care delivery and outcomes for older adults with cancer and comorbid dementia are limited by varied methods used to classify Alzheimer's disease and related dementias (ADRD). The purpose of this study is to evaluate differences in demographic, clinical, and cancer characteristics of people newly diagnosed with cancer and concomitant dementia comparing two common methods to identify ADRD using administrative claims data.

MATERIALS AND METHODS

We conducted a retrospective cohort study using Surveillance, Epidemiology, and End Results (SEER)-Medicare data. Our sample included adults aged 66 years and older with a first primary diagnosis of lung or colorectal cancer between 2011 and 2017. For each cancer diagnosis, we constructed analytical cohorts using the Center for Medicare and Medicaid Services' Chronic Condition Warehouse (CCW) flag and the Bynum-Standard one- and three-year algorithms to capture diagnosed ADRD. We estimated ADRD prevalence using the algorithms and compared Bynum and CCW cohorts on demographic, clinical, and cancer characteristics at cancer diagnosis and survival for lung and colorectal cancer separately.

RESULTS

Among older adults with lung cancer, ADRD prevalence was 4.7% with the one-year Bynum, 6.5% with the three-year Bynum, and 12.5% using the CCW flag. In the colorectal cohort, ADRD prevalence was 5.6% with the one-year Bynum, 7.6% with the three-year Bynum, and 14.1% with the CCW flag. Demographic characteristics were similar across ADRD cohorts. The Bynum cohorts identified higher proportions of individuals with moderate to severe dementia (13.8% and 11.2% versus 7.1% CCW in lung cancer; 13.1% and 10.6% versus 6.8% CCW in colorectal cancer), higher frailty rates (27.4% and 22.7% versus 15.0% CCW in lung cancer; 26.4% and 22.3% versus 14.8% CCW in colorectal cancer). Median survival was lower for the Bynum cohorts compared to the CCW, regardless of cancer type.

DISCUSSION

Findings demonstrate that ADRD prevalence and certain clinical characteristics vary based on dementia ascertainment method and observation period used to classify individuals with ADRD. Considering differences in the cohorts of registry cases generated by the identification method used is essential when interpreting findings related to treatment, utilization, and outcomes within and across cancer cohorts.

Impact of federal antipsychotic use policy in nursing homes on new diagnoses for approved indications in dementia residents

BACKGROUND

Federal policies targeting antipsychotic use among nursing home (NH) residents may have increased reporting of diagnoses for approved uses, including schizophrenia, Tourette's syndrome, and Huntington's Disease (called "exclusionary diagnoses" because they exclude residents from the antipsychotic quality metric). We assessed changes in new exclusionary diagnoses among long-stay NH admissions specifically with dementia following federal policies.

METHODS

Retrospective, quarterly, interrupted time-series analysis (2009-2018) of new long-stay NH residents with dementia and no exclusionary diagnoses reported before NH admission. The National Partnership and the addition of facility level antipsychotic use to the Five Star Quality Rating system were key time exposures. Outcome was quarterly facility level predicted percentage of exclusionary diagnoses within 2 years of admission stratified by NH characteristics.

RESULTS

For 264,095 long-stay admissions, mean percentage of new exclusionary diagnoses was 2.2% before the Partnership. After the Partnership, there was an unadjusted increase in the percentage over time (slope change, 0.044, p = 0.018), but the percentage never exceeded 2.9%. The Partnership contributed to a one-time decrease in diagnoses in NHs with an intermediate percentage of Black residents (-1.29%, p = 0.004). Before the Partnership, diagnoses were increasing among not-for-profit relative to for-profit NHs (0.044; p = 0.012), but after the Partnership, the pattern reversed. For-profit NHs saw an increase (+0.034, p = 0.002); not-for-profit NHs experienced a decrease (-0.014, p = 0.039). Quality Rating modifications had no significant effect.

CONCLUSIONS

Exclusionary diagnosis reporting among long-stay NH residents with dementia, the group most at risk from antipsychotics, did not increase in response to federal policies. Evaluation of reasons for the observed increase in exclusionary diagnoses among non-dementia NH residents is warranted along with continued attention to how to incentivize the appropriate use of medications in residents with dementia that is crucial for high-quality NH care.

Correlates of missed or late versus timely diagnosis of dementia in healthcare settings

INTRODUCTION

There is limited evidence about factors related to the timeliness of dementia diagnosis in healthcare settings.

METHODS

In five prospective cohorts at Rush Alzheimer's Disease Center, we identified participants with incident dementia based on annual assessments and examined the timing of healthcare diagnoses in Medicare claims. We assessed sociodemographic, health, and psychosocial correlates of timely diagnosis.

RESULTS

Of 710 participants, 385 (or 54%) received a timely claims diagnosis within 3 years prior to or 1 year following dementia onset. In logistic regressions accounting for demographics, we found Black participants (odds ratio [OR] = 2.15, 95% confidence interval [CI]: 1.21 to 3.82) and those with better cognition at dementia onset (OR = 1.48, 95% CI: 1.10 to 1.98) were at higher odds of experiencing a diagnostic delay, whereas participants with higher income (OR = 0.89, 95% CI: 0.81 to 0.97) and more comorbidities (OR = 0.94, 95% CI: 0.89 to 0.98) had lower odds.

DISCUSSION

We identified characteristics of individuals who may miss the optimal window for dementia treatment and support.

HIGHLIGHTS

We compared the timing of healthcare diagnosis relative to the timing of incident dementia based on rigorous annual evaluation. Older Black adults with lower income, higher cognitive function, and fewer comorbidities were less likely to be diagnosed in a timely manner by the healthcare system.

A comparison of dementia diagnoses and cognitive function measures in Medicare claims and the Minimum Data Set

BACKGROUND

Gold standard dementia assessments are rarely available in large real-world datasets, leaving researchers to choose among methods with imperfect but acceptable accuracy to identify nursing home (NH) residents with dementia. In healthcare claims, options include claims-based diagnosis algorithms, diagnosis indicators, and cognitive function measures in the Minimum Data Set (MDS), but few studies have compared these. We evaluated the proportion of NH residents identified with possible dementia and concordance of these three.

METHODS

Using a 20% random sample of 2018-2019 Medicare beneficiaries, we identified MDS admission assessments for non-skilled NH stays among individuals with continuous enrollment in Medicare Parts A, B, and D. Dementia was identified using: (1) Chronic Conditions Warehouse (CCW) claims-based algorithm for Alzheimer's disease and non-Alzheimer's dementia; (2) MDS active diagnosis indicators for Alzheimer's disease and non-Alzheimer's dementias; and (3) the MDS Cognitive Function Scale (CFS) (at least mild cognitive impairment). We compared the proportion of admissions with evidence of possible dementia using each criterion and calculated the sensitivity, specificity, and agreement of the CCW claims definition and MDS indicators for identifying any impairment on the CFS.

RESULTS

Among 346,013 non-SNF NH admissions between 2018 and 2019, 57.2% met criteria for at least one definition (44.7% CFS, 40.7% CCW algorithm, 26.0% MDS indicators). The MDS CFS uniquely identified the greatest proportion with evidence of dementia. The CCW claims algorithm had 63.7% sensitivity and 78.1% specificity for identifying any cognitive impairment on the CFS. Active diagnosis indicators from the MDS had lower sensitivity (47.0%), but higher specificity (91.0%).

CONCLUSIONS

Claims- and MDS-based methods for identifying NH residents with possible dementia have only partial overlap in the cohorts they identify, and neither is an obvious gold standard. Future studies should seek to determine whether additional functional assessments from the MDS or prescriptions can improve identification of possible dementia in this population.

Disparities in glaucoma and macular degeneration healthcare utilization among persons living with dementia in the United States

PURPOSE

Dementia is common among patients with primary open angle glaucoma (POAG) and neovascular age-related macular degeneration (nAMD). This study compares visit frequency, diagnostic test utilization, and treatment patterns for POAG and nAMD among persons with vs. without dementia.

METHODS

Optum's de-identified Clinformatics® Data Mart Database (January 1, 2000-June 30, 2022) was used for this study. Two cohorts were created from newly diagnosed POAG or nAMD patients. Within each cohort, an exposure cohort was created of newly diagnosed dementia patients. The primary outcome was the number of visits to an eye care provider. Secondary analyses for the POAG cohort assessed the number of visual field tests, optical coherence tomography (OCT), and glaucoma medication prescription coverage. The secondary analysis for the nAMD cohort included the number of injections performed. Poisson regression was used to determine the relative rates of outcomes.

RESULTS

POAG patients with dementia had reduced rates of eye care visits (RR 0.76, 95% CI: 0.75-0.77), lower rates of testing utilization for visual fields (RR 0.66, 95% CI: 0.63-0.68) and OCT (RR 0.67, 95% CI: 0.64-0.69), and a lower rate of glaucoma prescription medication coverage (RR 0.83, 95% CI: 0.83-0.83). nAMD patients with dementia had reduced rates of eye care visits (RR 0.74, 95% CI: 0.70-0.79) and received fewer intravitreal injections (RR 0.64, 95% CI: 0.58-0.69) than those without dementia.

CONCLUSIONS

POAG and nAMD patients with dementia obtained less eye care and less monitoring and treatment of their disease. These findings suggest that this population may be vulnerable to gaps in ophthalmic care.

Alzheimer Disease and Related Dementia Following Hormone-Modulating Therapy in Patients With Breast Cancer

Importance

Hormone-modulating therapy (HMT) is a widely accepted treatment for hormone receptor-positive breast cancer, although its cognitive effects, including a potential link to Alzheimer disease and related dementias (ADRD), remain understudied.

Objective

To investigate the association between HMT for breast cancer treatment and risk of developing ADRD in women aged 65 years or older.

Design, Setting, and Participants

This cohort study used a comprehensive dataset from the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database to identify patients who did and did not receive HMT treatment within 3 years after the initial diagnosis of breast cancer and assessed their risk of developing ADRD in later life. Individuals with a preexisting diagnosis of ADRD or receiving HMT before the diagnosis of breast cancer were excluded. This study was performed from June 2022 through January 2024.

Exposure

Receipt of HMT.

Main Outcomes and Measures

Risk of ADRD associated with HMT; associations of risk with age, self-identified race, and HMT type. Risk was measured using hazard ratios (HRs) with 95% CIs and adjusted for potential confounders such as demographic, sociocultural, and clinical variables.

Results

Among 18 808 women aged 65 years and older diagnosed with breast cancer between 2007 and 2009 (1266 Black [6.7%], 16 526 White [87.9%], 1016 other [5.4%]), 12 356 (65.7%) received HMT within 3 years after diagnosis, while 6452 (34.3%) did not. The most common age group in both samples was the 75 to 79 years age group (HMT, 2721 women [22.0%]; no HMT, 1469 women [22.8%]), and the majority of women in both groups self-identified as White (HMT, 10 904 women [88.3%]; no HMT, 5622 women [87.1%]). During an average of 12 years of follow-up, 2926 (23.7%) of HMT users and 1802 (27.9%) of non-HMT users developed ADRD. HMT was associated with a 7% lower relative risk of ADRD overall (HR, 0.93; 95% CI, 0.88-0.98; P = .005). The association decreased with age and varied by race. The reduction in ADRD risk associated with HMT was greatest for women aged 65 to 74 years who self-identified as Black (HR, 0.76; 95% CI, 0.62-0.92). This association decreased among women aged 75 years or older (HR, 0.81; 95% CI, 0.67-0.98). Women aged 65 to 74 years who self-identified as White had an 11% relative risk reduction (HR, 0.89; 95% CI, 0.81-0.97), but the association disappeared for women aged 75 years or older (HR, 0.96; 95% CI, 0.90-1.02). Other races showed no significant association between HMT and ADRD. Age- and race-based associations also varied by HMT type.

Conclusions and Relevance

In this retrospective cohort study, hormone therapy was associated with protection against ADRD in women aged 65 years or older with newly diagnosed breast cancer; the decrease in risk was relatively greater for Black women and women under age 75 years, while the protective effect of HMT diminished with age and varied by race in women. When deciding to use HMT for breast cancer in women aged 65 years or more, clinicians should consider age, self-identified race, and HMT type in treatment decisions.

COVID-19 and Influenza Vaccine Coadministration Among Older U.S. Adults

INTRODUCTION

Coadministering COVID-19 and influenza vaccines is recommended by public health authorities and intended to improve uptake and convenience; however, the extent of vaccine coadministration is largely unknown. Investigations into COVID-19 and influenza vaccine coadministration are needed to describe compliance with newer recommendations and to identify potential gaps in the implementation of coadministration.

METHODS

A descriptive, repeated cross-sectional study between September 1, 2021 to November 30, 2021 (Period 1) and September 1, 2022 to November 30, 2022 (Period 2) was conducted. This study included community-dwelling Medicare beneficiaries ≥ 66 years who received an mRNA COVID-19 booster vaccine in Periods 1 and 2. The outcome was an influenza vaccine administered on the same day as the COVID-19 vaccine. Adjusted ORs and 99% CIs were estimated using logistic regression to describe the association between beneficiaries' characteristics and vaccine coadministration. Statistical analysis was performed in 2023.

RESULTS

Among beneficiaries who received a COVID-19 vaccine, 78.8% in Period 1 (N=6,292,777) and 89.1% in Period 2 (N=4,757,501), received an influenza vaccine at some point during the study period (i.e., before, after, or on the same day as their COVID-19 vaccine), though rates were lower in non-White and rural individuals. Vaccine coadministration increased from 11.1% to 36.5% between periods. Beneficiaries with dementia (aORPeriod 2=1.31; 99%CI=1.29-1.32) and in rural counties (aORPeriod 2=1.19; 99%CI=1.17-1.20) were more likely to receive coadministered vaccines, while those with cancer (aORPeriod 2=0.90; 99%CI=0.89-0.91) were less likely.

CONCLUSIONS

Among Medicare beneficiaries vaccinated against COVID-19, influenza vaccination was high, but coadministration of the 2 vaccines was low. Future work should explore which factors explain variation in the decision to receive coadministered vaccines.

Around the EQUATOR with Clin-STAR: Prediction modeling opportunities and challenges in aging research

The 2015 Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) Statement was published to improve reporting transparency for prediction modeling studies. The objective of this review is to highlight methodologic challenges that aging-focused researchers will encounter when designing and reporting studies involving prediction models for older adults and provide guidance for addressing these challenges. In following the 22-item TRIPOD checklist, researchers must consider the representativeness of cohorts used (e.g., whether older adults with frailty, cognitive impairment, and social isolation were included), strategies for incorporating common geriatric predictors (e.g., age, comorbidities, functional status, and frailty), methods for handling missing data and competing risk of death, and assessment of model performance heterogeneity across important subgroups (e.g., age, sex, race, and ethnicity). We provide guidance to help aging-focused researchers develop, validate, and report models that can inform and improve patient care, which we label "TRIPOD-65."

Antidementia Medication Use in Nursing Home Residents

BACKGROUND

Antidementia medication can provide symptomatic improvements in patients with Alzheimer's disease, but there is a lack of consensus guidance on when to start and stop treatment in the nursing home setting.

METHODS

We describe utilization patterns of cholinesterase inhibitors (ChEI) and memantine for 3,50,197 newly admitted NH residents with dementia between 2011 and 2018.

RESULTS

Overall, pre-admission use of antidementia medications declined from 2011 to 2018 (ChEIs: 44.5% to 36.9%; memantine: 27.4% to 23.2%). Older age, use of a feeding tube, and greater functional dependency were associated with lower odds of ChEI initiation. Coronary artery disease, parenteral nutrition, severe aggressive behaviors, severe cognitive impairment, and high functional dependency were associated with discontinuation of ChEIs. Comparison of clinical factors related to anti-dementia drug treatment changes from pre to post NH admission in 2011 and 2018 revealed a change toward lower likelihood of initiation of treatment among residents with more functional dependency and those with indicators of more complex illness as well as a change toward higher likelihood of discontinuation in residents having 2 or more hospital stays.

CONCLUSIONS

These prescribing trends highlight the need for additional research on the effects of initiating and discontinuing antidementia medications in the NH to provide clear guidance for clinicians when making treatment decisions for individual residents.

Association between surgery and rate of incident dementia in older adults: A population-based retrospective cohort study

BACKGROUND

The risk of incident dementia after surgery in older adults is unclear. The study objective was to examine the rate of incident dementia among older adults after elective surgery compared with a matched nonsurgical control group.

METHODS

We conducted a population-based, propensity-matched retrospective cohort study using data from linked administrative databases in Ontario, Canada. All community-dwelling individuals aged 66 years and older who underwent one of five major elective surgeries between April 1, 2007 and March 31, 2011 were included. Each surgical patient was matched 1:1 on surgical specialty of the surgeon at consultation, age, sex, fiscal year of entry, and propensity score with a patient who attended an outpatient visit with a surgeon of the same surgical specialty but did not undergo surgery. Patients were followed for up to 5 years after cohort entry for the occurrence of a new dementia diagnosis, defined from administrative data. Cause-specific hazard models were used to estimate the hazard ratio (HR) and 95% confidence interval (CI) for the association between surgery and the hazard of incident dementia. Subgroup and sensitivity analyses were performed.

RESULTS

A total of 27,878 individuals (13,939 matched pairs) were included in the analysis. A total of 640 (4.6%) individuals in the surgical group and 965 (6.9%) individuals in the control group developed dementia over the 5-year follow-up period. Individuals who underwent surgery had a reduced rate of incident dementia compared with their matched nonsurgical controls (HR 0.88; 95% CI 0.80-0.97; p = 0.01). This association was persistent in most subgroups and after sensitivity analyses.

CONCLUSIONS

Elective surgery did not increase the rate of incident dementia when compared with matched nonsurgical controls. This could be an important consideration for patients and surgeons when elective surgery is considered.

Racial and ethnic differences in disease course Medicare expenditures for beneficiaries with dementia

BACKGROUND

Research on racial and ethnic disparities in costs of care during the course of dementia is sparse. We analyzed Medicare expenditures for beneficiaries with dementia to identify when during the course of care costs are the highest and whether they differ by race and ethnicity.

METHODS

We analyzed data from the 2000-2016 Health and Retirement Study (HRS) linked with corresponding Medicare claims to estimate total Medicare expenditures for four phases: (1) the year before a dementia diagnosis, (2) the first year following a dementia diagnosis, (3) ongoing care for dementia after the first year, and (4) the last year of life. We estimated each patient's phase-specific and disease course Medicare expenditures by using a race-specific survival model and monthly expenditures adjusted for patient characteristics. We investigated healthcare utilization by service type across races/ethnicities and phases of care.

RESULTS

Adjusted mean total Medicare expenditures for non-Hispanic (NH) Black ($165,730) and Hispanic beneficiaries with dementia ($160,442) exceeded corresponding expenditures for NH Whites ($136,326). In the year preceding and immediately following initial dementia diagnosis, mean Medicare expenditures for NH Blacks ($26,337 and $20,429) exceeded expenditures for Hispanics and NH Whites ($21,399-23,176 and 17,182-18,244). The last year of life was responsible for the greatest cost contribution: $51,294 (NH Blacks), $47,469 (Hispanics), and $39,499 (NH Whites). These differences were driven by greater use of high-cost services (e.g., emergency department, inpatient and intensive care), especially during the last year of life.

CONCLUSIONS

NH Black and Hispanic beneficiaries with dementia had higher disease course Medicare expenditures than NH Whites. Expenditures were highest for NH Black beneficiaries in every phase of care. Further research should address mechanisms of such disparities and identify methods to improve communication, shared decision-making, and access to appropriate services for all populations.

How donanemab data address the coverage with evidence development questions

The Centers for Medicare & Medicaid Services (CMS) established a class-based National Coverage Determination (NCD) for monoclonal antibodies directed against amyloid for Alzheimer's disease (AD) with patient access through Coverage with Evidence Development (CED) based on three questions. This review, focused on donanemab, answers each of these CED questions with quality evidence. TRAILBLAZER-ALZ registration trials are presented with supporting literature and real-world data to answer CED questions for donanemab. TRAILBLAZER-ALZ registration trials demonstrated that donanemab significantly slowed cognitive and functional decline in amyloid-positive early symptomatic AD participants, and lowered their risk of disease progression while key safety risks occurred primarily within the first 6 months and then declined. Donanemab meaningfully improved health outcomes with a manageable safety profile in an early symptomatic AD population, representative of Medicare populations across diverse practice settings. The donanemab data provide the necessary level of evidence for CMS to open a reconsideration of their NCD. HIGHLIGHTS: Donanemab meaningfully improved outcomes in trial participants with early symptomatic Alzheimer's disease. Comorbidities in trial participants were consistent with the Medicare population. Co-medications in trial participants were consistent with the Medicare population. Risks associated with treatment tended to occur in the first 6 months. Risks of amyloid-related imaging abnormalities were managed with careful observation and magnetic resonance imaging monitoring.

Care sequences leading to the diagnosis of Alzheimer's disease and related dementias: An analysis of electronic health records

BACKGROUND

We examined the sequences of clinical care leading to diagnoses of Alzheimer's disease and related dementias (ADRD) using electronic health records from a large academic medical center.

METHODS

We included patients aged 65+ with their first ADRD diagnoses from January 1, 2014 to December 31, 2019. Using state sequence analysis, care sequences were defined by the ordering of healthcare utilizations occurred in the 2 years before ADRD diagnosis.

RESULTS

Of 3621 patients (median age 80), nearly half followed a care sequence of having one primary care visit close to their ADRD diagnosis. Additional care sequences included periodic (n = 322, 8.9%) and multiple (n = 416, 11.5%) outpatient visits to primary care and having one (n = 395, 10.9%), multiple (n = 469, 13.0%), or highly frequent (n = 357, 10.7%) outpatient visits to other specialties. Patients' sociodemographic traits contributed to the variability in care sequences.

CONCLUSIONS

Several distinct patterns of care leading to ADRD diagnoses were identified. Integrated care models are needed to promote early identification of ADRD.

HIGHLIGHTS

Dementia patients followed distinct care pathways prior to their dementia diagnoses. Key sociodemographic traits contributed to the variation in the sequences of care. Racial differences in the sequencing of care were also found, but only in women.

Algorithmic identification of persons with dementia for research recruitment: ethical considerations

Underdiagnosis, misdiagnosis, and patterns of social inequality that translate into unequal access to health systems all pose barriers to identifying and recruiting diverse and representative populations into research on Alzheimer's disease and Alzheimer's disease related dementias. In response, some have turned to algorithms to identify patients living with dementia using information that is associated with this condition but that is not as specific as a diagnosis. This paper explains six ethical issues associated with the use of such algorithms including the generation of new, sensitive, identifiable medical information for research purposes without participant consent, issues of justice and equity, risk, and ethical communication. It concludes with a discussion of strategies for addressing these issues and prompting valuable research.

Accountable Care Organization Attribution and Post-Acute Skilled Nursing Facility Outcomes for People Living With Dementia

OBJECTIVES

Under the Accountable Care Organization (ACO) model, reductions in healthcare spending have been achieved by targeting post-acute care, particularly in skilled nursing facilities (SNFs). People with Alzheimer disease and related dementias (ADRD) are frequently discharged to SNF for post-acute care and may be at particular risk for unintended consequences of SNF cost reduction efforts. We examined SNF length of stay (LOS) and outcomes among ACO-attributed and non-ACO-attributed ADRD patients.

DESIGN

Observational serial cross-sectional study.

SETTING AND PARTICIPANTS

Twenty percent national random sample of fee-for-service Medicare beneficiaries (2013-2017) to identify beneficiaries with a diagnosis of ADRD and with a hospitalization followed by SNF admission (n = 263,676).

METHODS

Our primary covariate of interest was ACO (n = 66,842) and non-ACO (n = 196,834) attribution. Hospital readmission and death were measured for 3 time periods (<30, 31-90, and 91-180 days) following hospital discharge. We used 2-stage least squares regression to predict LOS as a function of ACO attribution, and patient and facility characteristics.

RESULTS

ACO-attributed ADRD patients have shorter SNF LOS than their non-ACO counterparts (31.7 vs 32.8 days; P < .001). Hospital readmission rates for ACO vs non-ACO differed at ≤30 days (13.9% vs 14.6%; P < .001) but were similar at 31-90 days and 91-180 days. No significant difference was observed in mortality post-hospital discharge for ACO vs non-ACO at ≤30 days; however, slightly higher mortality was observed at 31-90 days (8.4% vs 8.8%; P = .002) and 91-180 days (7.6% vs 7.9%; P = .011). No significant association was found between LOS and readmission, with small effects on mortality favoring ACOs in fully adjusted models.

CONCLUSIONS AND IMPLICATIONS

Being an ACO-attributed patient is associated with shorter SNF LOS but is not associated with changes in readmission or mortality after controlling for other factors. Policies that shorten LOS may not have adverse effects on outcomes for people living with dementia.

Geographic Variation of Prevalence of Alzheimer's Disease and Related Dementias in Central Appalachia

Background

Alzheimer's disease and related dementias (ADRD) prevalence varies geographically in the United States.

Objective

To assess whether the geographic variation of ADRD in Central Appalachia is explained by county-level sociodemographics or access to care.

Methods

Centers for Medicare and Medicaid Services Public Use Files from 2015- 2018 were used to estimate county-level ADRD prevalence among all fee-for-service (FFS) beneficiaries with≥1 inpatient, skilled nursing facility, home health agency, hospital outpatient or Carrier claim with a valid ADRD ICD-9/10 code over three-years in Central Appalachia (Kentucky, North Carolina, Ohio, Tennessee, Virginia, and West Virginia). Negative binomial regression was used to estimate prevalence overall, by Appalachian/non-Appalachian designation, and by rural/urban classification. Models were then adjusted for county-level: 1) FFS demographics (age, gender, and Medicaid eligibility), comorbidities; 2) population sociodemographics (race/ethnicity, education, aging population distribution, and renter-occupied housing); and 3) diagnostic access (PCP visits, neurology visits, and imaging scans).

Results

Across the 591 counties in the Central Appalachian region, the average prevalence of ADRD from 2015- 2018 was 11.8%. ADRD prevalence was modestly higher for Appalachian counties both overall (PR: 1.03; 95% CI: 1.02, 1.04) and after adjustment (PR: 1.02; 95% CI: 1.00, 1.03) compared to non-Appalachian counties. This difference was similar among rural and urban counties (p = 0.326) but varied by state (p = 0.004).

Conclusions

The relative variation in ADRD prevalence in the Appalachian region was smaller than hypothesized. The case mixture of the dual eligible population, accuracy of the outcome measurement, and impact of educational attainment in this region may contribute to this observation.

Identification of Dementia in Medicare Claims Compared to Rigorous Clinical Assessments in African Americans

BACKGROUND

Evidence indicates the health care system disproportionately misses dementia in African American compared to White individuals. In preliminary data, we examined factors related to dementia identification by the health care system among African Americans.

METHODS

We leveraged linked Medicare fee-for-service claims and detailed annual cohort evaluations in African Americans from 4 cohorts at Rush Alzheimer's Disease Center.

RESULTS

Among 88 African Americans with cognitive impairment (mean = 10 years follow-up), Medicare claims identified dementia <2 years from cohort diagnosis in 55%; 27% were identified 2-9.9 years after cohort diagnosis, and in 18% there was either no claims diagnosis during the study period, or claims identified dementia 10+ years after cohort diagnosis. Claims identification of dementia was related to older age at cohort diagnosis (eg, <2 years between cohort and claims: mean = 82 years; 10+ years/no diagnosis: mean = 77 years, p = .04), lower Mini-Mental State Examination (MMSE) score (<2 years: mean = 24; 10+ years/no diagnosis: mean = 26, p = .04), more depressive symptoms (<2 years: mean = 2.1 symptoms; 10+ years/no diagnosis: mean = 1.2, p = .04), and more comorbidity (<2 years: mean = 5.6 comorbidities; 10+ years/no diagnosis, mean = 3.0, p = .02).

CONCLUSIONS

Among African Americans, preliminary data indicate the health care system most rapidly identifies dementia in older individuals, with worse cognitive and physical health. The health care system may miss opportunities for early support of African Americans with dementia, and caregivers.

Claims-Based Frailty Index as a Measure of Dementia Severity in Medicare Claims Data

BACKGROUND

Dementia severity is unavailable in administrative claims data. We examined whether a claims-based frailty index (CFI) can measure dementia severity in Medicare claims.

METHODS

This cross-sectional study included the National Health and Aging Trends Study Round 5 participants with possible or probable dementia whose Medicare claims were available. We estimated the Functional Assessment Staging Test (FAST) scale (range: 3 [mild cognitive impairment] to 7 [severe dementia]) using information from the survey. We calculated CFI (range: 0-1, higher scores indicating greater frailty) using Medicare claims 12 months prior to the participants' interview date. We examined C-statistics to evaluate the ability of the CFI in identifying moderate-to-severe dementia (FAST stage 5-7) and determined the optimal CFI cut-point that maximized both sensitivity and specificity.

RESULTS

Of the 814 participants with possible or probable dementia and measurable CFI, 686 (72.2%) patients were ≥75 years old, 448 (50.8%) were female, and 244 (25.9%) had FAST stage 5-7. The C-statistic of CFI to identify FAST stage 5-7 was 0.78 (95% confidence interval: 0.72-0.83), with a CFI cut-point of 0.280, achieving the maximum sensitivity of 76.9% and specificity of 62.8%. Participants with CFI ≥0.280 had a higher prevalence of disability (19.4% vs 58.3%) and dementia medication use (6.0% vs 22.8%) and higher risk of mortality (10.7% vs 26.3%) and nursing home admission (4.5% vs 10.6%) over 2 years than those with CFI <0.280.

CONCLUSIONS

Our study suggests that CFI can be useful in identifying moderate-to-severe dementia from administrative claims among older adults with dementia.

Data resource profile: COVid VAXines effects on the aged (COVVAXAGE)

Background

To improve the assessment of COVID-19 vaccine use, safety, and effectiveness in older adults and persons with complex multimorbidity, the COVid VAXines Effects on the Aged (COVVAXAGE) database was established by linking CVS Health and Walgreens pharmacy customers to Medicare claims.

Methods

We deterministically linked CVS Health and Walgreens customers who had a pharmacy dispensation/encounter paid for by Medicare to Medicare enrollment and claims records. Linked data include U.S. Medicare claims, Medicare enrollment files, and community pharmacy records. The data currently span 01/01/2016 to 08/31/2022. "Research-ready" files were created, with weekly indicators for vaccinations, censoring, death, enrollment, demographics, and comorbidities. Data are updated quarterly.

Results

As of November 2022, records for 27,086,723 CVS Health and 23,510,025 Walgreens unique customer IDs were identified for potential linkage. Approximately 91% of customers were matched to a Medicare beneficiary ID (95% for those aged 65 years or older). In the final linked cohort, there were 38,250,873 unique beneficiaries representing ~60% of the Medicare population. Among those alive and enrolled in Medicare as of January 1, 2020 (n = 33,721,568; average age = 73 years, 74% White, 51% Medicare Fee-for-Service, and 11% dual-eligible for Medicaid), the average follow-up time was 130 weeks. The cohort contains 16,021,055 beneficiaries with evidence a first COVID-19 vaccine dose. Data are stored on the secure Medicare & Medicaid Resource Information Center Health & Aging Data Enclave.

Data access

Investigators with funded or in-progress funding applications to the National Institute on Aging who are interested in learning more about the database should contact Dr Vincent Mor [Vincent_mor@brown.edu] and Dr Kaleen Hayes [kaley_hayes@brown.edu]. A data dictionary can be provided under reasonable request.

Conclusions

The COVVAXAGE cohort is a large and diverse cohort that can be used for the ongoing evaluation of COVID-19 vaccine use and other research questions relevant to the Medicare population.

Identifying Medicare beneficiaries with Alzheimer's disease and related dementia using home health OASIS assessments

BACKGROUND

Home health services are an important site of care following hospitalization among Medicare beneficiaries, providing health assessments that can be leveraged to detect diagnoses that are not available in other data sources. In this work, we aimed to develop a parsimonious and accurate algorithm using home health outcome and assessment information set (OASIS) measures to identify Medicare beneficiaries with a diagnosis of Alzheimer's disease and related dementia (ADRD).

METHODS

We conducted a retrospective cohort study of Medicare beneficiaries with a complete OASIS start of care assessment in 2014, 2016, 2018, or 2019 to determine how well the items from various versions could identify those with an ADRD diagnosis by the assessment date. The prediction model was developed iteratively, comparing the performance of different models in terms of sensitivity, specificity, and accuracy of prediction, from a multivariable logistic regression model using clinically relevant variables, to regression models with all available variables and predictive modeling techniques, to estimate the best performing parsimonious model.

RESULTS

The most important predictors of having a diagnosis of ADRD by the start of care OASIS assessment were a prior discharge diagnosis of ADRD among those admitted from an inpatient setting, and frequently exhibiting symptoms of confusion. Results from the parsimonious model were consistent across the four annual cohorts and OASIS versions with high specificity (above 96%), but poor sensitivity (below 58%). The positive predictive value was high, over 87% across study years.

CONCLUSIONS

The proposed algorithm has high accuracy, requires a single OASIS assessment, is easy to implement without sophisticated statistical models, and can be used across four OASIS versions and in situations where claims are not available to identify individuals with a diagnosis of ADRD, including the growing population of Medicare Advantage beneficiaries.

HomeADScreen: Developing Alzheimer's disease and related dementia risk identification model in home healthcare

BACKGROUND

More than 50 % of patients with Alzheimer's disease and related dementia (ADRD) remain undiagnosed. This is specifically the case for home healthcare (HHC) patients.

OBJECTIVES

This study aimed at developing HomeADScreen, an ADRD risk screening model built on the combination of HHC patients' structured data and information extracted from HHC clinical notes.

METHODS

The study's sample included 15,973 HHC patients with no diagnosis of ADRD and 8,901 patients diagnosed with ADRD across four follow-up time windows. First, we applied two natural language processing methods, Word2Vec and topic modeling methods, to extract ADRD risk factors from clinical notes. Next, we built the risk identification model on the combination of the Outcome and Assessment Information Set (OASIS-structured data collected in the HHC setting) and clinical notes-risk factors across the four-time windows.

RESULTS

The top-performing machine learning algorithm attained an Area under the Curve = 0.76 for a four-year risk prediction time window. After optimizing the cut-off value for screening patients with ADRD (cut-off-value = 0.31), we achieved sensitivity = 0.75 and an F1-score = 0.63. For the first-year time window, adding clinical note-derived risk factors to OASIS data improved the overall performance of the risk identification model by 60 %. We observed a similar trend of increasing the model's overall performance across other time windows. Variables associated with increased risk of ADRD were "hearing impairment" and "impaired patient ability in the use of telephone." On the other hand, being "non-Hispanic White" and the "absence of impairment with prior daily functioning" were associated with a lower risk of ADRD.

CONCLUSION

HomeADScreen has a strong potential to be translated into clinical practice and assist HHC clinicians in assessing patients' cognitive function and referring them for further neurological assessment.

Comparing Dementia Classification by Self-Report and Administrative Records in the National Core Indicators-Aging and Disability Survey: A Predictive Modeling Approach

Policymakers are interested in the long-term services and supports (LTSS) needs of people living with dementia. The National Core Indicators-Aging and Disability (NCI-AD) survey is conducted to evaluate LTSS care needs. However, dementia reporting in NCI-AD varies across states, and is either obtained from state administrative records or self-reported during the survey. We explored the implications of identifying dementia from administrative records versus self-report. We analyzed 24,569 NCI-AD respondents age 65+, of which 22.4% had dementia. To assess dementia accuracy by data source, we fit separate logistic regression models using the administrative and self-reported subsamples. We applied model coefficients to the population whose dementia status came from the opposite source. Using the administrative model to predict self-reported dementia resulted in higher sensitivity than using the self-report model to predict administrative dementia (43.8% vs. 37.9%). The self-report model's diminished sensitivity suggests administrative records may capture cases of dementia missed by self-report.

Potentially Harmful Medication Prescribing by the Degree of Physician Specialization in Nursing Home Practice: An Observational Study

OBJECTIVES

The use of anticholinergics, antipsychotics, benzodiazepines, and other potentially harmful medications (PHMs) is associated with particularly poor outcomes in nursing home (NH) residents with Alzheimer's disease and related dementias (ADRD). Our objective was to compare PHM prescribing by NH physicians and advanced practitioners who focus their practice on NH residents (NH specialists) vs non-NH specialists.

DESIGN

Retrospective cohort study.

SETTING AND PARTICIPANTS

We included a 20% random sample of Medicare beneficiaries with ADRD who resided in 12,278 US NHs in 2017. Long-stay NH residents with ADRD were identified using MDS, Medicare Parts A and B claims. Residents <65 years old or without continuous Part D coverage were excluded.

METHODS

Physicians in generalist specialties and advanced practitioners with ≥90% of Part B claims for NH care were considered NH specialists. Residents were assigned to NH specialists vs non-NH specialists based on plurality of Part D claims submitted for that resident. Any PHM use (defined using the Beers Criteria) and the proportion of NH days on a PHM were modeled using generalized estimating equations. Models included resident demographics, clinical characteristics, cognitive and functional status, behavioral assessments, and facility characteristics.

RESULTS

Of the 54,713 residents in the sample, 27.9% were managed by an NH specialist and 72.1% by a non-NH specialist. There was no statistically significant difference in any PHM use [odds ratio (OR) 0.97, 95% CI 0.93-1.02, P = .23]. There were lower odds of prolonged PHM use (OR 0.87, 95% CI 0.81-0.94, P < .001, for PHM use on >75% vs >0%-<25% of NH days) for NH specialists vs non-NH specialists.

CONCLUSIONS AND IMPLICATIONS

Although the use of PHMs among NH residents with ADRD managed by NH specialists was not lower, they were less likely to receive PHMs over longer periods of time. Future work should evaluate the underlying causes of these differences to inform interventions to improve prescribing for NH residents.

Risk Selection and Care Fragmentation at Medicare Accountable Care Organizations for Patients With Dementia

BACKGROUND

Patients with dementia are a growing and vulnerable population within Medicare. Accountable care organizations (ACOs) are becoming Medicare's dominant care model, but ACO enrollment and care patterns for patients with dementia are unknown.

OBJECTIVE

The aim of this study was to compare differences in ACO enrollment for patients with versus without dementia, and in risk profiles and ambulatory care among patients with dementia by ACO enrollment status.

RESEARCH DESIGN

Cohort study assessing the relationships between patient dementia, following-year ACO enrollment, and ambulatory care patterns.

SUBJECTS

A total of 13,362 (weighted: 45, 499,049) person-years for patients [2761 (weighted: 6,312,304) for dementia patients] ages 65 years and above in the 2015-2019 Medicare Current Beneficiary Survey.

MEASURES

We assessed differences in ACO enrollment rates for patients with versus without dementia, and in dementia-relevant ambulatory care visit rates and validated care fragmentation indices among patients with dementia by ACO enrollment status.

RESULTS

Patients with versus without dementia were less likely to be enrolled in (38.3% vs. 44.6%, P<0.001), and more likely to exit (21.1% vs. 13.7%, P<0.01) ACOs. Among patients with dementia, those enrolled versus not enrolled in ACOs had a more favorable social and health risk profile on 6 of 16 measures (P<0.05). There were no differences in rates of dementia-relevant, primary, or specialty care visits. ACO enrollment was associated with 45.7% higher wellness visit rates (P<0.001), and 13.4% more fragmented primary care (P<0.01) spread across 8.7% more distinct physicians (P<0.05).

CONCLUSION

Medicare ACOs are less likely to enroll and retain patients with dementia than other patients and provide more fragmented primary care without providing additional dementia-relevant ambulatory care visits.

Expected and diagnosed rates of mild cognitive impairment and dementia in the U.S. Medicare population: observational analysis

BACKGROUND

With the emergence of disease-modifying Alzheimer's treatments, timely detection of early-stage disease is more important than ever, as the treatment will not be indicated for later stages. Contemporary population-level data for detection rates of mild cognitive impairment (MCI), the stage at which treatment would ideally start, are lacking, and detection rates for dementia are only available for subsets of the Medicare population. We sought to compare documented diagnosis rates of MCI and dementia in the full Medicare population with expected rates based on a predictive model.

METHODS

We performed an observational analysis of Medicare beneficiaries aged 65 and older with a near-continuous enrollment over a 3-year observation window or until death using 100% of the Medicare fee-for-service or Medicare Advantage Plans beneficiaries from 2015 to 2019. Actual diagnoses for MCI and dementia were derived from ICD-10 codes documented in those data. We used the 2000-2016 data of the Health and Retirement Study to develop a prediction model for expected diagnoses for the included population. The ratios between actually diagnosed cases of MCI and dementia over number of cases expected, the observed over expected ratio, reflects the detection rate.

RESULTS

Although detection rates for MCI cases increased from 2015 to 2019 (0.062 to 0.079), the results mean that 7.4 of 8 million (92%) expected MCI cases remained undiagnosed. The detection rate for MCI was 0.039 and 0.048 in Black and Hispanic beneficiaries, respectively, compared with 0.098 in non-Hispanic White beneficiaries. Individuals dually eligible for Medicare and Medicaid had lower estimated detection rates than their Medicare-only counterparts for MCI (0.056 vs 0.085). Dementia was diagnosed more frequently than expected (1.086 to 1.104) from 2015 to 2019, mostly in non-Hispanic White beneficiaries (1.367) compared with 0.696 in Black beneficiaries and 0.758 in Hispanic beneficiaries.

CONCLUSIONS

These results highlight the need to increase the overall detection rates of MCI and of dementia particularly in socioeconomically disadvantaged groups.

Dementias Among Older Males and Females in the U.S. Medicare System With and Without HIV

BACKGROUND

Despite the growing concern that people with HIV (PWH) will experience a disproportionate burden of dementia as they age, very few studies have examined the sex-specific prevalence of dementia, including Alzheimer disease and related dementias (AD/ADRD) among older PWH versus people without HIV (PWOH) using large national samples.

METHODS

We constructed successive cross-sectional cohorts including all PWH aged 65+ years from U.S. Medicare enrollees and PWOH in a 5% national sample of Medicare data from 2007 to 2019. All AD/ADRD cases were identified by ICD-9-CM/ICD-10-CM diagnosis codes. Prevalence of AD/ADRD was calculated for each calendar year by sex-age strata. Generalized estimating equations were used to assess factors associated with dementia and calculate the adjusted prevalence.

RESULTS

PWH had a higher prevalence of AD/ADRD, which increased over time compared with PWOH, especially among female beneficiaries and with increasing age. For example, among those aged 80+ years, the prevalence increased from 2007 to 2019 (females with HIV: 31.4%-44.1%; females without HIV: 27.4%-29.9%; males with HIV: 26.2%-33.3%; males without HIV: 21.0%-23.5%). After adjustment for demographics and comorbidities, the differences in dementia burden by HIV status remained, especially among older age groups.

CONCLUSIONS

Older Medicare enrollees with HIV had an increased dementia burden over time compared with those without HIV, especially women and older subjects. This underscores the need to develop tailored clinical practice guidelines that facilitate the integration of dementia and comorbidity screening, evaluation, and management into the routine primary care of aging PWH.

State Variation in Antipsychotic Use Among Assisted Living Residents With Dementia

OBJECTIVES

More than two-thirds of assisted living (AL) residents have dementia or cognitive impairment and antipsychotics are commonly prescribed for behavioral disturbances. As AL communities are regulated by state-level policies, which vary significantly regarding the care for people with dementia, we examined how antipsychotic prescribing varied across states among AL residents with dementia.

DESIGN

This was an observational study using 20% sample of national Medicare data in 2017.

SETTING AND PARTICIPANTS

The study cohort included Medicare beneficiaries with dementia aged 65 years or older who resided in larger (≥25-bed) ALs in 2017.

METHODS

The study outcome was the percentage of eligible AL person-months in which antipsychotics were prescribed for each state. We used a random intercept linear regression model to shrink estimates toward the overall mean use of antipsychotics addressing unstable estimates due to small sample sizes in some states.

RESULTS

A total of 20,867 AL residents with dementia were included in the analysis, contributing to 194,718 person-months of observation. On average, AL residents with dementia were prescribed antipsychotics during 12.6% of their person-months. This rate varied significantly by state, with a low of 7.8% (95% CI 5.9%-10.3%) for Hawaii to a high of 20.5% (95% CI 16.4%-25.3%) for Wyoming.

CONCLUSIONS AND IMPLICATIONS

We observed significant state variation in the prescribing of antipsychotics among AL residents with dementia using national data. These variations may reflect differences in state regulations regarding the care for AL residents with dementia and suggest the need for further investigation to ensure high quality of care.

Dementia and COVID-19 among Older African American Adults: A Scoping Review of Healthcare Access and Resources

African American/Black communities comprise 12.2% of the U.S. population, with a COVID-19 infection rate of more than 18% and marginal access to healthcare services. This scoping review synthesizes the emerging evidence on healthcare accessibility among older African American adult communities with dementia and COVID-19, as well as the resource requirements for this population during the pandemic. Searches of different databases for empirical studies and other sources on dementia and COVID-19 among older African American adults yielded 13 studies that met the following inclusion criteria: (a) focus on dementia and COVID-19, (b) sampled older African American adults, (c) investigated healthcare accessibility and resources, and (d) published between 2019 and 2022. Following the initial selection of the studies, eight were selected for relevance based on the Population, Concept, and Context (PCC) inclusion and exclusion criteria. Thematic analysis indicated that older African Americans with dementia and COVID-19 experienced longer delays in accessing timely healthcare, including transportation, intensive care units (ICUs), and mechanical ventilation. They also had reduced healthcare resources associated with a lack of health insurance, low financial resources, and an increased length of hospital stay, which further aggravated the negative effects of comorbid dementia and COVID-19 infections. Evidence showed that racial and age disparities affected older African American adults with dementia and COVID-19, resulting in lower healthcare access and marginal resources. This is consistent with historical and systemic inequities in meeting the healthcare needs of people of color in the United States, which was compounded for older African Americans during the COVID-19 pandemic.

Design Considerations for Embedded Pragmatic Clinical Trials of Advance Care Planning Interventions for Persons Living With Dementia

Advance care planning (ACP) is an important part of comprehensive care for persons living with dementia (PLWD). While many trials have established the efficacy of ACP in improving end-of-life communication and documentation of care preferences, there remains a gap in clinical usage. Embedded pragmatic clinical trials (ePCTs) may facilitate the uptake of evidence-based care into existing healthcare by deploying efficacious ACP interventions into real-world settings. However rigorous conduct of ePCTs of ACP for PLWD presents several unique methodological considerations. Here we describe a framework for the construction of these research studies, with a focus on distinguishing between the target of study: the PLWD, their care partners, or both. We outline specific considerations at each step of the research study process including 1) participant identification/eligibility, 2) participant recruitment/enrollment, 3) intervention implementation, and 4) outcome selection/ascertainment. These considerations are weighed in further detail by describing the approaches from three published trials. Specifically, we consider how potential challenges were overcome by tradeoffs in study design. Finally, we offer directions for future growth to advance ePCTs for ACP among PLWD and catalyze future research.

Time-lagged relationships between a decade of air pollution exposure and first hospitalization with Alzheimer's disease and related dementias

Alzheimer's disease and related dementias (ADRD) poses substantial health challenges among an aging population. One of the primary challenges in studying ADRD is that biological processes underlying these ailments begin decades prior to diagnosis. Previous studies indicate a relationship between ADRD and air pollution exposure to both fine particulate matter (PM2.5) and nitrogen dioxide (NO2) but are limited in their interpretation because they consider exposure measurements at a single time point. Our retrospective cohort study considered 27 + million Medicare enrollees in the United States followed up to 17 years and matched with highly accurate annual air pollution exposure measurements for PM2.5, NO2, and summer ozone. We applied distributed lag models and estimated the lagged associations between air pollution and odds of first hospitalization with ADRD. We found significantly increased odds due to overall PM2.5 and NO2 exposure and time-lagged exposure 10 and 8 years prior to admission, respectively. Furthermore, we found the connection between air pollution exposure and increased odds of first hospitalization with ADRD exists at air pollution levels below current National Ambient Air Quality Standards set by the US Environmental Protection Agency, with the steepest increase in odds occurring at low concentrations of PM2.5. Our findings are the first to show that air pollution exposures from as many as 10 years prior to the admission are related to increased odds of hospitalizations with ADRD. As there are no clear treatments available for ADRD, identifying modifiable risk factors such as air pollution exposure may make significant contributions towards prevention or delayed disease progression.

Code-Based Algorithms for Identifying Dementia in Electronic Health Records: Bridging the Gap Between Theory and Practice

Code-based algorithms are crucial tools in the detection of dementia using electronic health record data, with broad applications in medical research and healthcare. Vassilaki et al.'s study explores the efficacy of code-based algorithms in dementia detection using electronic health record data, achieving approximately 70% sensitivity and positive predictive value. Despite the promising results, the algorithms fail to detect around 30% of dementia cases, highlighting challenges in distinguishing cognitive decline factors. The study emphasizes the need for algorithmic improvements and further exploration across diverse healthcare systems and populations, serving as a critical step toward bridging gaps in dementia care and understanding.

Incidence of New Dementia Diagnosis in Veterans Admitted to Nursing Homes After Heart Failure Hospitalization

BACKGROUND

Hospitalization with heart failure (HF) may signal an increased risk of Alzheimer's disease and related dementias (ADRD). Nursing homes routinely assess cognition but the association of these results with new ADRD diagnosis in a population at high risk of ADRD is not known.

OBJECTIVE

To determine the association between nursing home cognitive assessment results and new diagnosis of dementia after heart failure hospitalization.

METHODS

This retrospective cohort study included Veterans hospitalized for HF and discharged to nursing homes, from 2010 to 2015, without a prior diagnosis of ADRD. We determined mild, moderate, or severe cognitive impairment using multiple items of the nursing home admission assessment. We used Cox regression to determine the association of cognitive impairment with new ADRD diagnosis during 365 days of follow-up.

RESULTS

The cohort included 7,472 residents, new diagnosis of ADRD occurred in 4,182 (56%). The adjusted hazard ratio of ADRD diagnosis was 4.5 (95% CI 4.2, 4.8) for the mild impairment group, 5.4 (95% CI 4.8, 5.9) for moderate impairment, and 4.0 (95% CI 3.2, 5.0) for severe impairment compared to the cognitively intact group.

CONCLUSION

New ADRD diagnoses occurred in more than half of Veterans with HF admitted to nursing homes for post-acute care.

Comparing Lookback Periods to Ascertain Alzheimer's Disease and Related Dementias

Claims data are a valuable resource for studying Alzheimer's disease and related dementias (ADRD). Alzheimer's disease and related dementias is often identified using a list of claims codes and a fixed lookback period of 3 years of data. However, a 1-year lookback or an approach using all-available lookback data could be beneficial based on different research questions. Thus, the purpose of this study was to compare 1-year and all-available lookback approaches to ascertaining ADRD compared to the standard 3-year approach. Using a cohort of Veterans hospitalized for heart failure (N = 373, 897), our results suggested high agreement (93% or greater) between the lookback periods. The 1-year lookback period had lower sensitivity (60%) and underestimated the prevalence of ADRD. These results suggest that 1-year and all-available lookback periods are viable approaches when using claims data.

Psychotropic medication prescribing in assisted living and nursing home residents with dementia after the National Partnership

BACKGROUND

The Centers for Medicare & Medicaid Services implemented the National Partnership to Improve Dementia Care in Nursing Homes (the Partnership) to decrease antipsychotic use and improve care for nursing home (NH) residents with dementia. We determined whether the extent of antipsychotic and other psychotropic medication prescribing in AL residents with dementia mirrored that of long-stay NH (LSNH) residents after the Partnership.

METHODS

Using a 20% sample of fee-for-service Medicare beneficiaries with Part D, we conducted a retrospective cohort study including AL and LSNH residents with dementia. The monthly prevalence of psychotropic medication prescribing (antipsychotics, antidepressants, anxiolytics/sedative-hypnotics, anticonvulsants/mood stabilizers, benzodiazepines, and antidementia medications) was examined. We used an interrupted time-series analysis to compare medication prescribing before (July 1, 2010-March 31, 2012) and after (April 1, 2012-December 31, 2017) the Partnership in both settings.

RESULTS

We identified 107,931 beneficiaries with ≥1 month as an AL resident and 323,766 beneficiaries with ≥1 month as a LSNH resident with dementia, including 1,923,867 person-months and 4,984,405 person-months, respectively. Antipsychotic prescribing declined over the study period in both settings. After the launch of the Partnership, the rate of decline in antipsychotic prescribing slowed in AL residents with dementia (slope change = 0.03 [95% CLs: 0.02, 0.04]) while the rate of decline in antipsychotic prescribing increased in LSNH residents with dementia (slope change = -0.12 [95% CLs: -0.16, -0.08]). Antidepressants were the most prevalent medication prescribed, anticonvulsant/mood stabilizer prescribing increased, and anxiolytic/sedative-hypnotic and antidementia medication prescribing declined.

CONCLUSIONS

The federal Partnership to reduce antipsychotic prescribing in NH residents did not appear to affect antipsychotic prescribing in AL residents with dementia. Given the increase in the prescribing of mood stabilizers/anticonvulsants that occurred after the launch of the Partnership, monitoring may be warranted for all psychotropic medications in AL and NH settings.

Associations between dementia diagnosis and end-of-life care utilization

BACKGROUND

Dementia is a leading cause of death for older adults and is more common among persons from racial/ethnic minoritized groups, who also tend to experience more intensive end-of-life care. This retrospective cohort study compared end-of-life care in persons with and without dementia and identified dementia's moderating effects on the relationship between race/ethnicity and end-of-life care.

METHODS

Administrative claims data for 463,590 Medicare fee-for-service decedents from 2016 to 2018 were analyzed. Multivariable logistic and linear regression analyses examined the association of dementia with 5 intensive and 2 quality of life-focused measures. Intensity measures included hospital admission, ICU admission, receipt of any of 5 intensive procedures (CPR, mechanical ventilation, intubation, dialysis initiation, and feeding tube insertion), hospital death, and Medicare expenditures (last 30 days of life). Quality of life measures included timely hospice care (>3 days before death) and days at home (last 6 months of life). Models were adjusted for demographic and clinical factors.

RESULTS

54% of Medicare decedents were female, 85% non-Hispanic White, 8% non-Hispanic Black, and 4% Hispanic. Overall, 51% had a dementia diagnosis claim. In adjusted models, decedents with dementia had 16%-29% lower odds of receiving intensive services (AOR hospital death: 0.71, 95% CI: 0.70-0.72; AOR hospital admission: 0.84, 95% CI: 0.83-0.86). Patients with dementia had 45% higher odds of receiving timely hospice (AOR: 1.45, 95% CI: 1.42-1.47), but spent 0.74 fewer days at home (adjusted mean: -0.74, 95% CI: (-0.98)-(-0.49)). Compared to non-Hispanic White individuals, persons from racial/ethnic minoritized groups were more likely to receive intensive services. This effect was more pronounced among persons with dementia.

CONCLUSIONS

Although overall dementia was associated with fewer intensive services near death, beneficiaries from racial/ethnic groups minoritized with dementia experienced more intensive service use. Particular attention is needed to ensure care aligns with the needs and preferences of persons with dementia and from racial/ethnic minoritized groups.

Medicare spending associated with a dementia diagnosis among older adults

BACKGROUND

Over 6 million Americans have Alzheimer's Disease or Related Dementia (ADRD) but whether spikes in spending surrounding a new diagnosis reflect pre-diagnosis morbidity, diagnostic testing, or treatments for comorbidities is unknown.

METHODS

We used the 1998-2018 Health and Retirement Study and linked Medicare claims from older (≥65) adults to assess incremental quarterly spending changes just before versus just after a clinical diagnosis (diagnosis cohort, n = 2779) and, for comparative purposes, for a cohort screened as impaired based on the validated Telephone Interview for Cognitive Status (TICS) (impairment cohort, n = 2318). Models were adjusted for sociodemographic and health characteristics. Spending patterns were examined separately by sex, race, education, dual eligibility, and geography.

RESULTS

Among the diagnosis cohort, mean (SD) overall spending was $4773 ($9774) per quarter - 43% of which was spending on hospital care ($2048). In adjusted analyses, spending increased by $8400 (p < 0.001), or 156%, from $5394 in the quarter prior to $13,794 in the quarter including the diagnosis. Among the cohort in which impairment was incidentally detected using the TICS, adjusted spending did not change from just before to after detection of impairment, from $2986 before and $2962 after detection (p = 0.90). Incremental spending changes did not differ by sex, race, education, dual eligibility, or geography.

CONCLUSION

Large, transient spending increases accompany an ADRD diagnosis that may not be attributed to impairment or changes in functional status due to dementia. Further study may help reveal how treatment for comorbidities is associated with the clinical diagnosis of dementia, with potential implications for Medicare spending.

Self-reported measures of limitation in physical function in late midlife are associated with incident Alzheimer's disease and related dementias

BACKGROUND

Even small improvements in modifiable Alzheimer's disease and related dementias (ADRD) risk factors could lead to a substantial reduction of dementia cases.

AIMS

To determine if self-reported functional limitation associates with ADRD symptoms 4-18 years later.

METHODS

We conducted a prospective longitudinal study using the Health and Retirement Study of adults aged 51-59 years in 1998 without symptoms of ADRD by 2002 and followed them to 2016. Main exposure variables were difficulty with activities of daily living, mobility, large muscle strength, gross motor and upper limb activities. The outcome was incident ADRD identified by the Lange-Weir algorithm, death, or alive without ADRD. We fit two GEE multinomial models for each measure: (1) baseline measure of function and (2) change in function over time.

RESULTS

In the model with baseline only and outcome, only difficulty with mobility associated with future ADRD across levels of difficulty with near dose-response effect (risk ratios (RR) difficulty with 1-5 functions respectively, compared with no difficulty: 1.82; 2.70; 1.73 2.81; 4.03). Mobility also significantly associated with ADRD when allowing for change over time among those with 3, 4 or 5 versus no mobility limitations (RR 1.76; 2.36; 2.37).

DISCUSSION

The results infer that an adult in midlife reporting difficulty with mobility as well as those with no mobility limitations in midlife but who later report severe limitations may be at increased risk of incident ADRD.

CONCLUSIONS

Self-reported measures of mobility limitation may be early indicators of ADRD and may be useful for public health planning.

Trends in US Medicare Decedents’ Diagnosis of Dementia From 2004 to 2017

Question

To what degree did the diagnosis of Alzheimer disease and related dementias (ADRD) change at the end of life between 2004 and 2017?

Findings

Among 3 515 329 Medicare fee-for-service decedents, the percentage who received an ADRD diagnosis within 2 years of death increased from 34.7% in 2004 to 47.2% in 2017. The likelihood of receiving an ADRD diagnosis particularly increased in the inpatient, hospice, and home health settings; individual characteristics and service use were stable over time, while the intensity of end-of-life care declined on most measures.

Meaning

Dying with an ADRD diagnosis has become more common among older US decedents, potentially owing to increased awareness and temporal changes in billing.

Importance

Alzheimer disease and related dementias (ADRD) have received considerable attention among clinicians, researchers, and policy makers in recent years. Despite increased awareness, few studies have documented temporal changes in the documentation of ADRD diagnoses despite its new importance for risk adjustment for health plans in Medicare.

Objective

To assess trends in frequency of ADRD diagnosis in the last 2 years of life from 2004 to 2017, as well as any associated changes in billing practices, characteristics of the population with diagnosed ADRD, and intensity of end-of-life care.

Design, Setting, and Participants

This is a serial cross-sectional study of older adult decedents (67 years or older) from 2004 to 2017 using a 20% sample of fee-for-service Medicare decedents. An ADRD diagnosis within the last 2 years of life was identified using diagnosis codes from inpatient, professional service, home health, or hospice claims, requiring the standard claims algorithm that required at least 1 claim and a more stringent algorithm that required at least 2 claims. Trends in ADRD diagnosis among decedents were used to lessen influence of new diagnostic technologies for early stage disease. Demographic characteristics, selected comorbidities, place of death, and health service use at the end-of-life were also examined. Data were analyzed from July 9, 2020, to May 3, 2021.

Exposures

Calendar year 2004 to 2017.

Main Outcome and Measure

An ADRD diagnosis within 2 years of death.

Results

Among the included 3 515 329 Medicare fee-for-service decedents, when adjusted for age and sex, the percentage of older decedents with an ADRD diagnosis increased from 34.7% in 2004 to 47.2% in 2017. The trend was attenuated (25.2% to 39.2%) using a stringent ADRD definition. There was an inflection in the curve from 2011 to 2013, the time at which additional diagnoses were added to Medicare claims and the National Alzheimer Care Act was enacted. The ADRD diagnosis frequency increased considerably in inpatient (49.0% to 67.3%), hospice (12.2% to 42.0%), and home health (10.1% to 28.7%) claims. However, individual characteristics, number of visits, and hospitalizations were similar across the study period, and the intensity of end-of-life care declined on most measures.

Conclusions and Relevance

In this cross-sectional study, nearly half of older Medicare decedents had a diagnosis of ADRD at the time of death. From 2004 to 2017, the percentage of older adult decedents who received an ADRD diagnosis increased substantially prior to announcement of the addition of ADRD to Medicare risk adjustment strategies.