Accountable care organizations and Medicare payments for residents with ADRD in disadvantaged neighborhoods

Alzheimer s & Dementia, Journal Year: 2025, Volume and Issue: 21(3)

Published: March 1, 2025

Abstract INTRODUCTION Accountable care organizations (ACOs) are well positioned to promote coordination. However, robust evidence of ACOs’ impact on Medicare payments for residents with Alzheimer's disease and related dementias (ADRD) in disadvantaged neighborhoods remains limited. METHODS Using a 2016 2020 longitudinal dataset, we examined the effects ACO enrollment people newly diagnosed ADRD, focusing neighborhood Social Vulnerability Index (SVI) its subcategories. Multivariable generalized estimating equation (GEE) models were applied. RESULTS was associated significantly reduced total across all SVI The highest cost savings observed among ADRD patients living high proportions racial ethnic minorities. Results also showed that higher quality ACOs lower payments. DISCUSSION have great potential save health‐care costs beneficiaries socially vulnerable neighborhoods, particularly those residing areas minority populations. Highlights disadvantage levels. reductions varied by specific indicators social vulnerability. Highest found proportion racial/ethnic Cost greatest ACOs.

Language: Английский

Hospital Artificial Intelligence/Machine Learning Adoption by Neighborhood Deprivation

Medical Care, Journal Year: 2025, Volume and Issue: 63(3), P. 227 - 233

Published: Jan. 3, 2025

Objective: To understand the variation in artificial intelligence/machine learning (AI/ML) adoption across different hospital characteristics and explore how AI/ML is utilized, particularly relation to neighborhood deprivation. Background: AI/ML-assisted care coordination has potential reduce health disparities, but there a lack of empirical evidence on AI’s impact equity. Methods: We used linked datasets from 2022 American Hospital Association Annual Survey 2023 Information Technology Supplement. The data were further Area Deprivation Index (ADI) for each hospital’s service area. State fixed-effect regressions employed. A decomposition model was also quantify predictors implementation, comparing hospitals higher versus lower ADI areas. Results: Hospitals serving most vulnerable areas (ADI Q4) significantly less likely apply ML or other predictive models (coef = −0.10, P 0.01) provided fewer AI/ML-related workforce applications -0.40, 0.01), compared with those least Decomposition results showed that our specifications explained 79% between Q4 Q1–Q3. In addition, Accountable Care Organization affiliation accounted 12%–25% differences utilization various measures. Conclusions: underuse economically disadvantaged rural areas, management electronic record suggests these communities may not fully benefit advancements AI-enabled care. Our indicate value-based payment could be strategically support AI integration.

Language: Английский

Higher than expected telemedicine use by racial and ethnic minority and cognitively impaired Medicare beneficiaries

Health Affairs Scholar, Journal Year: 2025, Volume and Issue: 3(1)

Published: Jan. 1, 2025

Although pandemic-era telemedicine flexibilities may have preserved access to care, concerns remain that been inequitably distributed among older adults, especially those with mild cognitive impairment or dementia (MCID). As are set fully expire on December 31, 2024, we aimed examine and future-intended use Americans help inform post-pandemic policy design. We hypothesized would be disproportionately underutilized adults MCID racial ethnic minority status. used nationally representative survey data from the Health Retirement Study analyzed 10 075 Medicare beneficiaries aged >50 years during 2020-2022 by cognition across beneficiaries-level characteristics such as age, gender, insurance status, education, multimorbidity. Results were adjusted weights nonresponse rates for national representativeness. Contrary our hypothesis, compared White beneficiaries, Hispanic Black normal reported 44% 57% greater use, respectively, while use. Our findings suggest utilization was common groups MCID.

Language: Английский