Abstract
The emergency department evaluates many patients with undiagnosed cognitive impairment and presents an opportune setting to facilitate early detection and referral to memory care specialists. We evaluated a novel care navigation pathway that facilitated referrals of ethnoculturally diverse individuals with suspected cognitive impairment from geriatric professionals embedded in the emergency department to dementia specialist care. We compared rates of referrals and successful appointment attendance for patients in this pathway compared to patients in a traditional primary care provider referral pathway. The care navigation team successfully identified and mitigated multiple barriers to accessing dementia specialist care, thereby increasing access.
Keywords
Introduction
Alzheimer's disease and related dementia's (AD/ADRD) prevalence will double by 2060. 1 Dementia care specialists fulfill a crucial role in diagnosing and caring for individuals with AD/ADRD and serve as the only access point for novel disease modifying therapies. 2 Traditionally, primary care providers (PCPs) function as the referral conduit for subspecialist referrals but multiple barriers to a successful referral exist including PCP office visit time constraints, required assessments and diagnostics (e.g., brain imaging, cognitive screening) 3 and insurance pre-approval. Emergency department (ED) discharge of cognitively impaired individuals often includes recommendations for specialty care follow-up, yet individuals and their caregivers may face barriers to successful referral. Difficulty navigating the healthcare system (e.g., inability to fully comprehend referral and follow-up instructions or forgetting important health information details for referral forms) leads to high risk for adverse outcomes, which can result in a cycle of crisis-oriented care.4–6 Racial and ethnic minorities experience additional barriers to specialist access,7,8 including lack of insurance coverage, scarce care services in local communities, and experiences of discrimination in the healthcare system. 9
EDs triage many undiagnosed patients with mild cognitive impairment (MCI) and dementia. 4 Though not appropriate to diagnose AD/ADRD during an acute health crisis, the ED may present an opportunity to facilitate early detection and referral to dementia specialists, 5 if they are able to overcome the barriers of a traditional PCP referral pathway after ED discharge. In this pilot quality improvement project, we evaluated the impact of a direct referral pathway from the ED to dementia specialist care coupled with a referral navigation team (“care navigation pathway”) compared to the usual care referral pathway where ED patients are instructed to follow-up with their PCP who is tasked with making the referral to dementia specialist care (“traditional referral pathway”). We compared the rate of eligible referrals sent for dementia specialist care and successful appointment attendance. In addition, we summarized the barriers faced by patients and the interventions employed to support successful attendance of a dementia specialist appointment. We hypothesized the care navigation pathway would increase access to dementia specialist care, and the care navigation team would identify and address barriers to specialist access.
Methods
Population and setting
Pilot participants were individuals age ≥65 identified by the Age-Friendly ED (AFED) consult service at the University of California San Francisco (UCSF) to have possible cognitive impairment during ED visits between May 1, 2022 and December 31, 2023. From May 1, 2022 to February 28, 2023, the care navigation team received and managed referrals directly from the ED (care navigation pathway), and from March 1, 2023 to December 31, 2023, the patient's PCP received and managed referrals (traditional referral pathway). The UCSF AFED focuses on age-friendly care for older adults through the services of an interprofessional team including a geriatric trained nurse practitioner, social worker, and pharmacist. Individuals who present to the ED and receive the AFED consult service may present for any medical issue(s), which may or may not be related to cognitive impairment. The AFED consult team completes a modified comprehensive evaluation that includes cognitive, functional, and caregiver assessment. The assessments are the AWOLi (Age, ability to spell WORLD backwards, orientation to place, illness severity for assessment of delirium, independent historian answering about patient's memory changes), Nu-DESC (Nursing Delirium Screening Scale), ISAR (Identification of Seniors at Risk screening for predicting adverse outcomes for elderly in the ED), STEADI (Stop Elderly Accidents, Deaths, and Injuries screening for fall risk assessment), IADLs (Instrumental activities of daily living), ADLs (activities of daily living), Mini-COG, AD8 (eight-item informant interview to differentiate aging and dementia), and bCAM-Delirium (Brief Confusion Assessment Method-Delirium). Patients considered to have a high probability of underlying cognitive impairment and likely to benefit from structured evaluation and care by a dementia specialist are referred by the AFED team.
Care navigation pathway
The care navigation team based at the UCSF Memory and Aging Center consisted of a behavioral neurologist, a nurse practitioner hospitalist, a social worker, and a clinical research coordinator. From May 1, 2022 to February 28, 2023, the care navigation team received 32 referrals for AFED participants shared directly by the UCSF AFED consult service via referral order and inbox direct message sent via the electronic health record (EHR, Epic Systems®). Upon receipt of AFED referral, patients were called by a member of the care navigation team (in most cases the clinical research coordinator or social worker) and consented to participation. The care navigation team worked directly with clinic staff to facilitate referrals that were received, identifying and addressing barriers related to the referral until individuals successfully attended their initial evaluation visit at the Memory and Aging Center. Outcomes for referrals were followed up to 60 days after February 28, 2023 to ensure all referrals were followed to completion. All activities by care navigation team members (e.g., telephone calls, emails, conversations with care providers, etc.) were documented to capture data regarding barriers encountered in seeking care at the center and the interventions used to facilitate appointment attendance. The care navigation team met bimonthly and communicated via email and EHR between meetings to develop and implement solutions to identified barriers.
Traditional referral pathway
In the traditional referral pathway ((March 1, 2023–December 31, 2023); 31 referrals received before December 31, 2023 followed up to 60 days to ensure referrals were followed to completion), patients who received an AFED consultation and were recommended for referral to the Memory and Aging Center were instructed to follow-up with their PCP. The PCP was alerted via the communicated AFED consult note recommendations and tasked with submitting a referral to the center. Direct communication from the AFED team to PCPs, via electronic means or regular mail, avoided complete reliance upon the patient to communicate the need to place a referral to the center for further evaluation of cognitive impairment. The care navigation team was informed when these individuals were evaluated in the AFED and discharged back to the community via EHR messaging to permit date monitoring and allow tracking of when referrals were received by community PCPs. The care navigation team provided no extra support to patients referred by PCPs as they sought to establish care in the Memory and Aging Center. The Memory and Aging Center team processed the referrals as they were received according to standard clinic protocol. We did not call PCPs to capture data regarding barriers they faced in placing a referral due to lack of available resources for this task.
All study procedures were approved by the UCSF Institutional Review Board.
Results
Participant demographics are listed in Table 1. The mean age of the 32 individuals in the care navigation pathway was 78 years (range 60–95), and the mean age of the 31 in the traditional referral pathway was 80 years (range 64–95). About half were female (59.4% care navigation team pathway, 51.6% traditional referral pathway). About 80% of participants had a caregiver (81.3% care navigation pathway, 80.6% traditional referral pathway) and 100% of participants in both pathways had a PCP. Medicare A&B was the most common insurance in both the care navigation pathway (71.9%) and in the traditional referral pathway (54.8%).
Demographics of participants by referral pathway.
An eligible referral was sent by the AFED team for 29 out of 32 patients in the care navigation pathway, and 7 out of 31 patients by PCPs in the traditional referral pathway (Table 2, χ2 = 29.5, p < 0.01). The 3 ineligible referrals sent by the AFED team were patients with a high concern for active delirium. For individuals with eligible referrals sent, the median time between ED visit and referral submission was 1 day for the care navigation pathway and 39 days for the traditional pathway. Once an eligible referral was sent, the time to initial scheduled visit at the Memory and Aging Center did not differ between the two pathways (median = 80 days for traditional referral pathway; median = 85 days for care navigation pathway) (Table 2), and 100% of the eligible referrals in both pathways were scheduled. Fifteen (46.9%) of care navigation pathway participants had a visit scheduled and attended successfully, compared to 7 (22.6%) participants from the traditional referral pathway (Table 2, χ2 = 1.7, p = .19).
Outcomes by referral pathway.
In the care navigation pathway, 22/30 (68.8%) patients used the care navigation team support (Table 3). Most of these individuals needed assistance addressing scheduling and visit logistical barriers (46.9%). Repeated care navigation team efforts were needed to schedule initial visits given patient challenges due to other commitments and medical issues. Rescheduling unexpected cancellations by the patients, addressing lack of transportation to the Memory and Aging Center, and addressing the need for special assistance handicap services for in-person visits due to physical impairment were also identified. The care navigation team interventions addressed insurance barriers for 21.9% of individuals by obtaining prior authorizations to avoid substantial out-of-pocket cost to attend a clinic visit. Some individuals had active social issues impeding care (9.4% of cases) (Table 3). These included lack of stable housing and lack of a working telephone. The care navigation team engaged social workers and case managers for these patients. Although 90.6% of patients in the care navigation pathway had an eligible referral sent and 100% of these were scheduled, only 46.9% attended an appointment. Reasons for nonattendance were patient unresponsiveness or lack of interest pursuing care (35.3%), death (17.6%), active medical issues (11.8%), pursuit of dementia care at another center (11.8%), and other reasons (Table 3).
Barriers identified by care navigation pathway.
Visit scheduling and logistical barriers included limited windows of availability to attend a visit, unexpected cancellations, lack of transportation to visit, and need for special assistance due to limited mobility and physical impairment.
Insurance barriers included requirement of prior authorization.
Social issues included unstable housing and lack of access to a telephone or cell phone to communicate with medical care providers.
Additional materials needed from PCP included additional medical records.
Discussion
The care navigation team successfully increased access to dementia specialist care among this cohort of diverse individuals with suspected cognitive impairment who utilized emergency care services. The pathway, which featured direct referrals from the AFED team and provided navigation support to patients and caregivers, was associated with an eligible referral sent for 91% of patients, 100% of those referrals scheduled, and 47% of patients successfully attending an appointment. In contrast, the traditional referral pathway was associated with an eligible referral sent for 23% of patients, 100% of those scheduled, and 23% of patients successfully attending an appointment.
For patients in the care navigation pathway, the most frequent barriers to successful scheduling of an eligible referral and attendance of an initial visit were scheduling and visit logistics, which may reflect the complexity of coordinating follow-up care for those who utilize emergency care services. Active medical and non-medical issues among this population may interfere with attempts to obtain medical care.5 Dedicated support and resources could help ensure unanticipated challenges do not result in failure to follow-up. The next most encountered barrier was related to insurance coverage. Insurance coverage requiring pre-authorization and potential out of pocket costs create additional barriers.10–13 Expanded coverage with streamlined authorizations for dementia specialty care services could reduce the impact of this barrier. Urgent social issues were the third most encountered barrier, which may reflect the vulnerability of this population. Of note, approximately 30% of participants did not require any additional assistance from the care navigation team, suggesting there is a subset of individuals who can successfully navigate the referral process independently after an eligible referral is sent. These individuals represent the minority of cases.
The traditional referral pathway delayed referrals by a median of 38 days, and 72% of patients in this pathway never had an eligible referral sent by their PCP, in contrast to only 9% when referrals were sent directly. These findings suggest that dementia clinics that accept direct referrals from an AFED, even if they don’t offer navigator support, are likely to significantly improve access opportunities. Previous work has demonstrated PCPs face numerous challenges in the successful coordination of care to specialists.14,15 Considering alternative referral pathways to traditional PCP referral may improve timely access to specialty care. This is particularly important given novel disease-modifying therapies for Alzheimer's disease which provide greatest benefit to individuals with mild impairment. 16
Even with the efforts of the care navigation pathway, 53% of individuals in this pathway did not attend their initial scheduled visit. While some of these patients missed the scheduled visit because of critical illness or death (35%), others were unresponsive or expressed a lack of interest in pursuing care (35%). Future efforts to address this by a care navigation team might include more frequent authorized contact with caregivers and family, mailing information in diverse languages that discusses the benefits of early diagnosis and current therapies, and collecting further data on reasons for lack of interest in pursuing care.
It is important to consider the limitations of this pilot. We examined a patient population in a specific geographic region of the US, which may limit generalizability. We were not adequately powered to evaluate the impact of socioeconomic factors on access. Our pilot data was collected for a specified period and the referral pathways could not be performed simultaneously or randomized. Our care navigation services were primarily carried out by a research coordinator and a social worker; however, a neurologist and nurse practitioner also provided oversight and participated in the navigation, which raises questions about the scalability and costs of the model. While we think that this care model could be standardized and delivered by a patient navigator or social worker with minimal involvement from additional medical professionals, the scalability and costs of this care model remain to be tested. We also were not able to collect data regarding barriers that PCPs faced in placing a referral, which is an important consideration for future work.
Footnotes
Acknowledgments
The authors have no acknowledgments to report.
Author contributions
Charles C Windon (Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Writing – original draft; Writing – review & editing); Stephanie Pun (Conceptualization; Data curation; Investigation; Methodology; Project administration; Resources; Writing – original draft; Writing – review & editing); Mitchel Erickson (Conceptualization; Investigation; Methodology; Project administration; Resources; Validation; Writing – original draft; Writing – review & editing); Ashley Jackson (Data curation; Investigation; Project administration; Resources; Validation; Writing – original draft; Writing – review & editing); Hannah Ruben (Investigation; Methodology; Project administration; Resources; Writing – original draft); Phil Smith (Investigation; Project administration); Heather Tirona-Bito (Investigation; Project administration); Nhat Bui (Conceptualization; Investigation; Methodology; Project administration); Angelo Tumaneng (Investigation; Project administration); Nida Degesys (Conceptualization; Investigation; Methodology; Project administration); James Hardy (Conceptualization; Investigation; Methodology; Project administration); Anna Harris (Conceptualization; Project administration); Katherine L Possin (Conceptualization; Funding acquisition; Investigation; Methodology; Resources; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing)
Funding
The authors would like to report the following funding sources: Charles Windon – R35 AG072362, AACSFD-21-872476, Kate Possin - NINDS U01 NS128913, NIA R01 AG074710-01, the Global Brain Health Institute.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability
Data sharing is not applicable to this article as no datasets were generated or analyzed during this study.