The Impact of Ambulatory Dementia Care Models on Hospitalization of Persons Living With Dementia: A Systematic Review

This systematic review presents an overview of the existing dementia care models in various ambulatory care settings under three categories (i.e., home- and community-based care models, partnership between health systems and community-based resources, and consultation models) and their impact on hospitalization among Persons Living with Dementia (PLWD). PRISMA guidelines were applied, and our search resulted in a total of 13 studies focusing on 11 care models. Seven studies reported that utilization of dementia care models was associated with a modest reduction in hospitalization among community-residing PLWD. Only two studies reported statistically significant results. Dementia care models that were utilized in specialty ambulatory care settings such as memory care showed more promising results than traditional primary care. To develop a better understanding of how dementia care models can be improved, future studies should explore how confounders (e.g., stage of dementia) influence hospitalization.

Virtual Interdisciplinary Collaboration in Statewide Implementation of the MIND at Home Dementia Care Program

MIND at Home, a well-researched holistic, family-centered dementia care coordination program, provides collaborative support to community-dwelling persons living with dementia (PLWD) and their informal care partners (CP). Through comprehensive home-based assessment of 13 memory-care domains covering PLWD and CPs, individualized care plans are created, implemented, monitored, and revised over the course of the illness. Non-clinical Memory Care Coordinators (MCCs) working with an interdisciplinary team provide education and coaching to PLWD and their identified CP, and serve as a critical liaison and resource and between families, medical professional, and formal and informal community resources. This paper will describe a statewide pilot implementation of the program within a health plan across diverse sites in Texas and will present qualitative and quantitative descriptions of a key component of the program's effective translation to practice, the virtual collaborative case-based learning sessions. Health plan teams completed online interactive training modules and an intensive in-person case-based training with the Johns Hopkins team prior to program launch, and then engaged in weekly, hour-long virtual collaborative sessions that included health plan teams (site-based field teams, health plan clinical supervisory and specialty personnel [RNs, pharmacists, a geriatric psychiatrist, behavioral health specialists] and Johns Hopkins MIND program experts and geriatric psychiatrists. To date, the program has enrolled 350 health plan members, conducted 65 virtual collaborative sessions, and provided 423 CME/CEU units to team members. We will provide an overview of virtual collaborative session structure, participant contributions and discussion topics, case complexity, as well as didactic learning topics covered.

Considering the Technological and Sensory Needs of Patients With Cognitive Impairment in the Era of Telehealth

During the COVID-19 pandemic, telehealth has become an important means of delivering memory care. Telehealth that is responsive to the technological abilities and preferences as well as the sensory needs of persons living with dementia is critical to advancing access to care. We conducted a systematic review to investigate the use of telehealth among older adults with cognitive impairment. The search yielded 3,551 titles and abstracts that led to 17 full-text articles. Studies showed that telehealth can be used for routine care, cognitive assessment and telerehabilitation with good efficacy and satisfaction. Three studies investigated telemedicine delivery in the home and 16/17 studies relied on support staff and care partners to navigate technology. No studies reported adaptations to account for sensory impairments and 5/17 studies excluded individuals with sensory impairments. This talk will review barriers and facilitators totelehealth for older adults with cognitive impairment and adaptations to address sensory needs.

Diversity and Variation in Assisted Living Care, Nationally and Over Time

Assisted living is generally understood to offer a greater degree of privacy and independence than a nursing home; most residents pay privately, with some receiving support from state subsidies and Medicaid; regulation and oversight are the purview of state agencies. Within these broad parameters, however, one assisted living community may look quite different from another across the country, or down the street, in its resident population and the regulations that govern its operating license. The purpose of this symposium is to explore that variation. The papers leverage an in-depth review of changes in assisted-living regulation from 2007 to 2019 and a methodology to identify Medicare beneficiaries in assisted living using ZIP codes. To set the stage, the first paper examines variation across assisted living licenses to identify six regulatory types and compare their populations’ characteristics and health-care use. The second paper analyzes trends over time in the clinical acuity of assisted living residents associated with changes in nursing home populations. The third paper investigates racial disparities in assisted living associated with memory-care designations and proportions of Medicaid recipients. The fourth investigates how regulation of hospice providers in assisted living affect end-of-life care and place of death. The final paper describes requirements related to care for the residents with mental illness in seven states. The symposium concludes with an expert in long-term care disparities and quality discussing the implications for policymakers, providers, and the population needing long-term care in assisted living.

Passive Infrared Motion Sensors Improved the Detection Accuracy of Nocturnal Agitation

Actigraphy has been used to detect agitation in persons with dementia, although this technology must be worn by participants. Another promising sensing methodology is passive infrared (PIR) motion, which provides continuous, low-cost, and unobtrusive data, and may also improve the detection of agitated periods. Using data from the MODERATE (Monitoring Dementia-Related Agitation Using Technology Evaluation) study, we compared the predictive value of detecting agitation in a male participant, who was 64 years old with Alzheimer’s disease (AD), living in a memory care unit, and monitored with actigraphy on his wrist and four PIR motion sensors within his living quarters. The participant’s medical record indicated that he experienced agitation during 17 nights over 96 consecutive days. 929,037 data points were captured for analysis. From each night, the features extracted from the actigraphy wearable included total and standard deviation of activity counts, activity counts in the most and the least active hours, and median activity counts in one hour. Features extracted from the PIR motion sensors included dwell durations in the areas around bed, sofa, front door and bathroom, and the number of transitions between these areas. Using logistic regression to predict agitated periods, comparable classification performances were achieved using these two sets of features (AUC = 0.74 for wearable and AUC = 0.71 for PIR motion sensors). When these two sets of features were combined, the classification performance showed notable improvement (AUC = 0.83). This study points to the value of utilizing PIR motion sensors for detecting dementia-related agitation.

Staff Perspectives on Lessons Learned From an Interdisciplinary Memory Care Coordination Program

MIND at Home is a home-based care coordination program for persons living with dementia (PLWD) and their informal care partners (CP). Assessments, care planning and coordination is delivered by trained non-clinical Memory Care Coordinators (MCCs), working together on an interdisciplinary team with nurses and geriatric psychiatrists. We report qualitative results from program staff (two nurses and eight MCCs) who implemented the program in the context of two clinical trials. Care team respondents answered open-ended questions covering 5 domains pertaining to: helpful skillsets; positive and challenging factors aspects of care coordination; barriers to care coordination for clients; and improvements suggestions/resources to strengthen the program. Compassion, finding common ground, listening, organization, and time management were reported as critical skills. Staff enjoyed team collaboration, being in and learning about the community, increasing CP confidence and mastery when caring for a PLWD. Reported challenges included documentation in EHR, accessing/navigating resources, driving long distances, unsafe neighborhoods, ambiguous assessment tools, and working with low engagement clients. Common barriers faced by clients (as reported by staff) were financial struggles/poverty, and lack of insurance coverage for needed services. Staff suggested several improvements: better communication strategies, integration with LTSS services and medical providers, 24-hour program hotline, continuous education for staff, simplified data collection and care delivery tracking process. This presentation on the experience of MIND at Home trained nurses and MCCs provides deep insight on how this and similar care coordination programs might be successfully implemented or strengthened.

Dementia Care Coaching: A Pilot to Evaluate Acceptability and Feasibility in Care Communities

The Alzheimer’s Association (AA) Dementia Care Practice Recommendations (DCPR) outline ten recommendations to achieve quality care with a person-centered focus. The AA has developed tools to assist care communities (CC) to evaluate their status within the recommendations by working with a trained coach to maximize adoption and implementation of these recommendations. The purpose of this pilot was to evaluate the acceptability and feasibility of pairing trained DCPR coaches with CC teams to implement the DCPR tools. Seven CCs were recruited and four received the DCPR overview and self-assessment. Of the four CC, one withdrew and did not receive the intervention. The remaining three were located in a suburban area, nonprofit, and with memory care units. Data was collected from November 2019 through March 2020. Nine CC staff participated with a mean age 35.8 years and had worked for 11.8 years. Baseline mean scores on the Organizational Readiness to Implementing Change (ORIC) scale were 4.6 for the commitment domain and 4.4 for the efficacy domain. Mean scores on the Nursing Home Employee Satisfaction Survey were high. Sixty-nine percent of CC participants were satisfied with their jobs (greater than 4). Findings from mid-project interviews with the coaches revealed difficulty with scheduling appointments, significant efforts needed to get the “right” people at the table and need for the DCPR tools to be more user-friendly. No post-intervention results were collected due to closing of the CCs to visitors due to COVID. The DCPR tools shows promise and are being evaluated in additional CCs.

Racial Disparities in Use of Memory Care and Prevalence of Duals in Assisted Living

Assisted living (AL) communities with memory care licenses are disproportionately located in affluent and predominantly White communities and Black older adults are underrepresented in AL. But little is known about characteristics of AL that care for Black residents. We estimated the association of facility-level characteristics as proxy measures for AL resources, such as memory care designations and percentage of dual-eligible residents, across low (0-5%), medium (5-10%) and high (>10%) percentages of Black residents. We found broad differences among communities in the three levels of Black-resident prevalence. High percentage of Black residents was associated with large differences in the percentage of Medicaid-enrolled residents (high 54% duals [s.d.=34], med 28% [31], low=13% [22], p<0.001). ALs with high Black populations were less likely to have a memory-care designation than ALs with medium and low percentages of Black residents (high 4.7% memory care, med 11%, low 17%).

COVID-19 Mortality In Adult Congregate Living Settings Across Five U.S. States

The magnitude of COVID-19 mortality in adult congregate living settings other than nursing homes (NH) is unknown. To address this, we created an individual property level dataset for five U.S. states (Colorado, Connecticut, Florida, Georgia, and Pennsylvania) using multiple public and private sources. The data included information on each observation’s state and county, level of care (LOC), the estimated number of residents, COVID-19 deaths through December 31, 2020, and county-level cases of COVID-19 per 100,000. We restricted our sample to market grade properties with 25+ residents, for which we able to estimate resident and LOC information. We defined LOC as County, non-congregate (CN), Independent Living (IL), Assisted Living (AL), Memory Care (MC), and Nursing Home (NH). We used multilevel, multivariable logistic regression models to estimate the expected death rate for each LOC controlling for differences in reported COVID-19 infections and county and state reporting differences. We identified 3,059 properties that met our inclusion criteria (69 CN, 477 IL, 1,118 AL, 179 MC, and 1,216 NH). We estimated deaths per 1,000 persons of 4.4 (95% CI: 4.0-4.8) for CN, 4.4 (3.9-4.9) for IL, 16.2 (14.7-17.9) for AL, 50.3 (44.4-56.8) for MC, and 32.0 (29.3-34.9) for NH. The order of death rate severity was the same in each state, except MC in Georgia. Additional research is needed to evaluate whether death rate differences resulted from congregate living risks, from COVID mortality risk factors at each LOC, or a combination of factors.