The number of people living with dementia (PLWD) is growing. PLWD are often cared for at home by an informal caregiver, but this care is often not sufficient, resulting in costly hospitalizations and other unnecessary and avoidable use of health services. In addition, many PLWD are transferred to costly long-term care facilities despite their preference to live at home. One way to improve care for PLWD is to work with their primary care doctors to provide better quality of care at lower costs for their patients. To accomplish this, we propose to collaboratively implement the MIND at Home Dementia Care Coordination Program into primary care clinics. The program expands the skills of existing primary care staff to the level of Memory Care Coordinators (MCCs), who will work with a larger primary care team on combining the benefits of clinic-based services with home-based services that support PLWD, their families, and care partners.

Two health care organizations will enroll 150 people in the MIND at Home program for 3 months at a time. The program includes one home visit per month, a comprehensive needs assessment (which assesses medical, nonmedical, social, and environmental issues), the subsequent development and implementation of an individualized care plan, and unlimited contact with the MCC for the PLWD, their family, and care partner. The primary care team, including the MCC, will also have access to weekly virtual sessions focused on dementia and including short lectures and the discussion of specific case examples.

Rates of monthly hospitalizations among participants and emergency room (ER) visits and number of medications the PLWD takes every month will be collected from the health care organization. The hypothesis is that the rate of hospitalizations and ER visits will decrease, and the number of medications will also decrease. We hope to positively impact the quality and costs of care associated with caring for PLWD.

This pilot seeks to test the feasibility of implementing the MIND at Home program into primary care in a racially, ethnically, and geographically diverse population of PLWD to prepare for a larger study that will determine this program's effectiveness and spread it broadly into primary care clinics across the country.

From the perspectives of health systems and overall society, MIND at Home will reduce costs, improve primary care team satisfaction, and preserve the dignity and independence of PLWD by enabling them to age at home.

Integration of evidence-based collaborative care management models such as the MIND at Home Dementia Care Coordination Program within Primary Care (PC) aligns with the Centers for Medicare and Medicaid Services (CMS) initiatives to provide advanced PC services to an especially complex and costly patient group and aligns with AHRQ's team-based Collaborative Care Model. This project will embed and test the feasibility of a novel best practice-based approach, MIND at Home, within PC to enhance and elevate the role of existing PC staff to Memory Care Coordinators (MCCs), increase PC access to interdisciplinary collaborative care, and systematically combines the benefits of clinic-based services with home-based assessment to support family-centered care planning and implementation for PLWD and their care partners. Weekly virtual collaborative learning sessions that include geriatric psychiatry consultants augment the PC care team's work to support the development and mastery of dementia assessment and care management skills and confidence at the PC sites. Our overarching goal is to test and establish feasibility, acceptability, fidelity, and sample size/referral rate data for MIND at Home in PC to prepare for a future multi-site embedded pragmatic trial.

Aim 1: Evaluate the feasibility and validity of eligible PLWD identification, referral, and enrollment in a best practice-based dementia care coordination program (MIND at Home) at 3 primary care clinics. Algorithms based on CMS's Chronic Care Warehouse definition of Alzheimer's Disease (AD) or AD and Related Dementias (ADRD) will be deployed in each respective clinic's electronic health record (EHR) to identify eligible PLWD and standardized referral and enrollment protocols will be enacted. Referral, recruitment, patient acceptance, and attrition rates will be calculated to assess the feasibility of use of the algorithm and of the recruitment protocol. Health equity in recruitment and enrollment will be assessed by comparing demographics (race/ethnicity) of program enrollees with the background population e.g., clinic ADRD patient panel, and by comparing enrollment of rural vs. suburban PC sites.

Aim 2: Evaluate the feasibility, acceptability, and fidelity of implementing MIND at Home in 3 primary care clinics in 2 geographically and demographically diverse integrated health systems. Using a pragmatic trial design, 150 community-residing PC patients with an ADRD diagnosis will be enrolled, each for a 3-month period. Program enrollees will receive office- plus home-based dementia care assessment, individualized care planning, and implementation by an interdisciplinary PC-based team, supported by regular case-based learning sessions. Implementation will be assessed by collecting a standard set of delivery workflow process, clinical, and health care utilization metrics. The investigators will assess feasibility through referral, enrollment, and retention rates, number of MCC-dyad contacts, and clinic staff participation in weekly tele-collaboratives. Fidelity will be assessed through tracking the completion of intervention components and PC team self-evaluation. The investigators will survey clinic staff on perceived program acceptability, usefulness, and challenges/unintended consequences. Program implementation will be tailored to accommodate diverse dyads (PLWD + care partner), which will be closely monitored and tracked.

Aim 3: Evaluate the feasibility of ascertainment of patient-level outcomes over time using electronic health record (EHR) data. The primary outcome (hospital transfers including admissions, ER visits, and observation stays without admissions), and secondary outcomes (number of medications for enrolled patients) will be gathered using EHR data. Accuracy of ascertainment will be tested through chart review. All clinical outcomes will be collected as time-stamped events for each enrolled patient retrospectively for 3 months prior to enrollment, during enrollment (3 months), and 1-month post-enrollment (total observation=7 months). All outcomes will be assessed by race, ethnicity, and rurality of the PLWD.

Embedding a collaborative, best-practice-based approach such as MIND at Home into PC is a potentially powerful strategy to organize care, improve quality, reduce costs, and maximize the population-level benefit for PLWD. This proposal tests the feasibility of implementing MIND at Home into PC in a racially, ethnically, and geographically diverse PLWD population in order to prepare for a multisite embedded pragmatic clinical trial (ePCT) to evaluate effectiveness, and eventually support broader dissemination and uptake in PC.

Inclusion Criteria:

• Community residing adults, ≥18 years
• People Living with Dementia (PLWD) defined by a diagnosis using algorithms based on CMS' Chronic Care Warehouse definition of Alzheimer's Disease, Related Dementias, Related Disorders, and Senile Dementia.11 This definition includes those who have an eligible diagnosis code on any eligible claim (i.e., inpatient, outpatient, skilled nursing facility (SNF), or home health visit or stay) in the past three years or a diagnosis on the patient problem list.
• Actively receiving primary care services at one of three primary care practices within two participating health care organizations selected as study sites.
• Have a reliable care partner who speaks English (or a language spoken by the Memory Care Coordinator).
• Willing to participate in all study home visits and related activities for the entire length of the study (3 months).

Exclusion Criteria:

• PLWD in crisis, e.g., show signs of abuse, neglect, extreme risk of danger to self or others), will be connected to appropriate services, but will be excluded from the study.

Analysis was conducted on intervention and data validation patients only and per-protocol.

Adverse event data were collected on each intervention patient from date of study enrollment through the conclusion of the 3-month study period.

We did not collect AEs on Care Partners nor were we able to given that they were not necessarily patients of the same health systems with which we partnered.

We did not collect AEs on patients in the data validation arm as we had no contact with those patients.

In both of these groups, All-Cause Mortality, Serious, and Other (Not Including Serious) Adverse Events were not monitored/assessed.

This was designed as a pilot feasibility study. The intent was to determine the feasibility of (1) implementing the program in primary care, and (2) collecting patient-level data from health systems. Therefore it was not powered to find statistically significant results, although we still conducted an analysis and submitted results. Per our protocol, we collected only minimal data on Care Partners. They were not always patients of the same health system as the PLWD.