Over six million Americans have Alzheimer's disease or an Alzheimer's disease related dementia (AD/ADRD), a number projected to reach 13 million by 2050 1. Healthcare costs for AD/ADRD exceeded 305 billion in 2020, and dementia is now the sixth leading cause of death in the United States 2. More than 50% of older adults with AD/ADRD visit the emergency department (ED) each year 3, 4 and those who do are seriously ill, have significant healthcare costs (e. g. , hospital admissions, repeat ED visits) and mortality 4-6. In persons living with dementia (PLWD) who have serious illness, an ED visit is a critical event and an opportunity to address the unmet needs that preceded it 7. This is even more true when the PLWD is discharged directly back to the community from the ED, another transition where gaps in care can be exacerbated. Emergency medicine developed as a specialty to treat the acutely ill and injured, most of whom were previously healthy, fully functional and without cognitive impairment. Today, many emergency patients are older, suffer from complex medical conditions such as AD/ADRD, and need both medical and social services 3, 4. While many PLWD live alone and have no one to assist them, care partners play a central role supporting the health of PLWD, even if they do not live with the PLWD or provide hands-on assistance with daily activities. Nonetheless, emergency clinicians rarely engage each PLWD and their care partner as a dyad 8, 9. When a PLWD has a care partner, emergency care must be reimagined with the dyad as the unit of care rather than the individual PLWD alone. However, effective, scalable ED-initiated interventions for PLWD (inclusive of dyadic care) are not yet ready for widespread dissemination. This JAGS Special Collection highlights the largest and most comprehensive pragmatic trial aimed at improving ED-to-community transitions for PLWD. The interventions are embedded in diverse health systems (racially, ethnically, and geographically) and will offer evidence highly relevant to real-world practice. Together, this suite of papers provides a comprehensive understanding of not only these interventions but also the process by which they were optimized. The overall goal of Emergency Departments Leading the Transformation of Alzheimer's and Dementia Care (ED-LEAD) is to optimize and study three interventions, each focused on ED-to-home care transitions, through an embedded, pragmatic, multifactorial trial. The interventions are grounded in the concept of a “triadic encounter, ” 10 during which the care recipient (the PLWD) and the care partner are clearly identified and interact with the clinicial team as active participants and decision-makers. The primary goal is to decrease ED revisits and hospitalizations while reducing suffering for PLWD with serious illness and their care partners who visit the ED. Redesigning emergency care processes, delivering a community-paramedic-led transitions intervention, and providing nurse-led telephonic case management can each reduce ED visits and other healthcare use for seriously ill PLWD 11-15. Primary Palliative Care for Emergency Medicine, an Emergency Care Redesign intervention focused on older adults with serious illness, included new and intentional workflows reinforced by digital alerts and structured collaboration between sites that increased the identification of advance care plans and enlisted multidisciplinary support 11-13. Community Paramedic-led Transitions Interventions, a post-ED visit program for PLWD, which used one home visit and three post-discharge phone-based visits, demonstrated 75% reduced odds of an ED revisit within 30 days 15. Finally, several Telephonic Case Management programs have shown efficacy, the most prominent of which are as follows: (1) Care Ecosystem, a telephonic program focused on advance care planning, behavioral and psychological symptoms of dementia, and caregiver needs 14; and (2) the UCLA Comprehensive Disease Management Program 16, an initial in-person comprehensive assessment with telephonic follow-up by community-based organizations to monitor implementation and changing needs. Classical randomized controlled trials (RCTs) determine whether an intervention as a treatment package performs better than an alternative or standard-of-care. They do not tell us which core functions make positive contributions to the overall effect, and whether the inclusion of one key element affects the impact of another. For this reason, many multi-component interventions deemed effective are not efficient and are therefore never scaled 17. Factorial design, as used in this large pragmatic trial 18, allows us to study the impact of three interventions at once, enabling efficiency and scalability of those that are most effective. These three interventions are specific to different settings across the transitions continuum and use different pragmatic delivery mediums. The complexity of caring for seriously ill PLWD requires studying multiple approaches in “real world” settings. We will study these three interventions in a high-risk, seriously ill population with a high rate of repeat ED revisits and hospitalizations: PLWD who visit the ED. These three interventions embody seven key elements from the National Transitions of Care Continuum (NTOCC), which inform the design of care transitions (Figure 1) to achieve a successful transition from ED to home. Whereas all three interventions incorporate all seven elements, each individual intervention carries particular emphases distinct from the other two. For this reason, when two or more are in conjoint integrated delivery, such combinations may yield the greatest benefit. The ED visit uses Emergency Care Redesign 19 to conduct a comprehensive dementia-informed assessment and benefits from interdisciplinary face-to-face care, trusted and engaged healthcare (ED) in the moment of crisis, and immediate diagnostic services not available elsewhere. Community Paramedic-led Transitions Intervention 20 embodies a trusted healthcare agent informed by a home visit where key information about environment and relationships may be leveraged for better care. Nurse-led Telephonic Care Management 21 benefits from focused interactions with dyads outside of a distracting ED environment and a longer-term relationship, which may work best when informed by details transferred from the Emergency Care Redesign assessment. Employing some combination of these three interventions offers the opportunity for: (1) better engagement and accountability; (2) comprehensive identification of addressable needs; (3) engaged and better-informed dyads; (4) reinforcement for medication and other management; and (5) transition planning with more opportunities for arranging follow-up care and accompanying information transfer. The three interventions are embedded in large health systems and studied in a pragmatic, multifactorial trial with the ED as the unit of randomization. We built upon existing research partnerships among diverse healthcare systems that vary in geographic location, size, payor mix, racial and ethnic diversity, and academic versus community focus. We optimized multi-intervention strategies by implementing all three interventions in two pilots EDs while simultaneously building out resources for each individual ED and developing systems that will be used in the full-scale trial. When programs are randomized in different combinations, slight variations to implementation processes may occur. Beginning in September 2025, we launched the first of five waves in the cluster-randomized factorial trial, with a new wave every 3 months. Overall, there are eight treatment arms across 79 EDs within 14 healthcare systems. Medicare claims will be used to follow 19, 200 PLWD anticipated to visit these EDs during the intervention period to determine outcomes, including ED revisits within 30 days (primary), 14 days and 6 months (secondary) of discharge; hospitalization within 14 days, 30 days and 6 months of discharge (secondary) ; and healthy days at home within 6 months of discharge (secondary). We selected the three potentially complementary interventions based on their readiness for real-world implementation both alone and in differing combinations. We have standardized the core functions of ED-LEAD across all three interventions to include: (1) identification and engagement with a care partner of the PLWD; (2) triadic encounter with assessment of both the PLWD and care partner health and well-being; (3) referral to community services; and (4) use of informatics to support clinical decisions. The actions within each of the four core functions will differ in form by provider type, the timing as it relates to the ED visit, and the mode of communication. For PLWD, and particularly for those with serious illness, an ED visit is a critical event and an opportunity to address the unmet needs that preceded this visit. The goal of ED-LEAD is to turn an ED visit from a crisis into an opportunity to improve the health and well-being of PLWD and their care partners. Evidence from ED-LEAD will help inform workforce needs, dementia training, care redesign and novel care delivery models while advancing the science of pragmatic trials, and helping health systems to consider which interventions in which combination may best support the PLWD they care for. We will use the ED-LEAD learnings and findings to serve as a blueprint for other groups who wish to make real-world, scalable change in health care delivery. Concept and design: C. R. G. , A. A. B. , J. C. , K. G. , M. N. S. , and K. S. G. Preparation of manuscript: All. The authors have nothing to report. Research reported in this publication was supported by the National Institute on Aging of the NIH under award number U19 AG078105-01A1. This research was also funded in part through the National Cancer Institute P30 CA008748. This manuscript is the result of funding in whole or in part by the National Institutes of Health (NIH). It is subject to the NIH Public Access Policy. Through acceptance of this federal funding, NIH has been given a right to make this manuscript publicly available in PubMed Central upon the Official Date of Publication, as defined by NIH. Sponsors of this work had no role in the design, methods, subject recruitment, data collection, analysis, or preparation of this paper. The authors declare no conflicts of interest.
Grudzen et al. (Thu,) studied this question.