This 60-minute symposium will present three leading approaches to dementia palliative care in the home, hospital, and hospice, followed by a synthesis and moderated discussion of common principles and implementation challenges.Laura Hanson, MD (10 minutes) will discuss findings from the ADRD-PC trial of inpatient dementia-specific palliative care consultation across five hospitals. Key takeaways include the value of early identification, structured goals-of-care conversations, and transitional support.Greg Sachs, MD (10 minutes) will describe lessons from the IN-PEACE trial, which tested a nurse- or social worker-led telephone intervention for people with moderate-to-severe dementia living at home. The session will highlight practical strategies for scalable delivery of dementia palliative care in the community.Maribeth Gallagher, DNP (10 minutes) will explore the unique role of hospice in dementia care, including transitions from community based supportive care, eligibility challenges, care at the end of life, and grief support for caregivers.David Lynch, BMBS (10 minutes) will synthesize common themes—structured communication, caregiver engagement, interdisciplinary delivery—and lead a discussion of how these models align with emerging payment structures such as GUIDE and Chronic Care Management.Moderated Discussion and Audience Q&A (20 minutes) will invite attendees to reflect on opportunities for expanding access to equitable, high-quality palliative care for people with dementia and their caregivers.

Meredith Bock, MD; Siqi Gan, MPH; Melissa Aldridge, PhD; Krista L. Harrison, PhD; Kristine Yaffe, MD; Alexander K. Smith, MD; John Boscardin, PhD; Lauren J. Hunt, PhD

Introduction Providing care management, treatment and support to patients with Alzheimer’s Disease and Related Dementias (ADRD) is a difficult task for health systems. Over the past 20 years, interventions designed to improve outcomes for patients living in the community with dementia and their care partners have moved progressively, but separately, from large scale trials and pragmatic models of collaborative care. Given the projected increase in the number of people living with dementia coupled with the realignment of payment for services to be value-based and provided in the community, system-level approaches are needed to address the complex needs of patients with a dementia diagnosis and their care partners. We designed a statewide, pragmatic trial to evaluate virtual delivery of an evidence-based dementia collaborative care program on patient healthcare utilization and medication use.Methods and analysis The Aging Brain Care Virtual (ABCV) program is a 12-month embedded, cluster randomized, usual care controlled trial designed to test the effectiveness of a virtual dementia collaborative care program in 24 Indiana University Health primary care clinics (12 intervention, 12 control) across the state of Indiana, enrolling 860 persons living with dementia (430 intervention, 430 control) and their care partners. ABCV relies on a tailored approach in which dyad needs are identified during virtual visits and addressed with standardized protocols previously tested in a randomized controlled trial delivered in person. The ABCV trial will measure emergency department utilization (primary outcome) and appropriate medication use (secondary outcome) at 12 months using electronic medical record data. Additionally, this study will use semi-structured interviews with care partners and clinicians to explore the implementation context, process and outcomes of the ABCV program.Ethics and dissemination Ethics approval was obtained from the Indiana University Institutional Review Board (20249). Research findings will be published in peer-reviewed journals and presented at scientific conferences.Trial registration number NCT06245499.

Importance Care management benefits community-dwelling patients with dementia, but studies include few patients with moderate to severe dementia or from racial and ethnic minority populations, lack palliative care, and seldom reduce health care utilization. Objective To determine whether integrated dementia palliative care reduces dementia symptoms, caregiver depression and distress, and emergency department (ED) visits and hospitalizations compared with usual care in moderate to severe dementia. Design, Setting, and Participants A randomized clinical trial of community-dwelling patients with moderate to severe dementia and their caregivers enrolled from March 2019 to December 2020 from 2 sites in central Indiana (2-year follow-up completed on January 7, 2023). Electronic health record screening identified patients with dementia; caregivers confirmed eligibility, including dementia stage. Intervention The intervention consisted of monthly calls from a trained nurse or social worker and evidence-based protocols to help caregivers manage patients' neuropsychiatric symptoms, caregiver distress, and palliative care issues (eg, advance care planning, symptoms, and hospice) (n = 99). Usual care caregivers received written dementia resource information and patients received care from usual clinicians (n = 102). Main Outcomes and Measures The primary outcome was Neuropsychiatric Inventory Questionnaire (NPI-Q) severity score (scores range from 0-36, with higher scores indicating worse patient symptoms). Secondary outcomes included Symptom Management in End-of-Life Dementia scores, caregiver depression (Patient Health Questionnaire-8) scores, caregiver distress (NPI-Q distress) scores, and combined ED and hospitalization events. Outcomes were assessed quarterly for 24 months or until patient death. Results A total of 201 dyads were enrolled (patients were 67.7% female; 43.3% African American; mean [SD] age, 83.6 [7.9] years); 3 dyads withdrew and 83 patients died over the course of the study, with at least 90% of eligible dyads in both groups completing each of the quarterly assessments. For the dementia palliative care vs usual care groups, mean NPI-Q severity scores were 9.92 vs 9.41 at baseline and 9.15 vs 9.39 at 24 months, respectively (between-group difference at 24 months, -0.24 [95% CI, -2.33 to 1.84]). There was no significant difference in the rate of change in NPI-Q severity from baseline between groups over time (P = .87 for the group and time interaction). There were no significant differences in the secondary outcomes, except that there were fewer combined ED and hospitalization events in the dementia palliative care group (mean events/patient, 1.06 in dementia palliative care vs 2.37 in usual care; between-group difference, -1.31 [95% CI, -1.93 to -0.69]; relative risk, 0.45 [95% CI, 0.31 to 0.65]). Conclusions and Relevance Among community-dwelling patients with moderate to severe dementia and their caregivers, dementia palliative care, compared with usual care, did not significantly improve patients' neuropsychiatric symptoms through 24 months.

CONTEXT:The symptom burden in persons dying with advanced dementia outside long-term care facilities has not been well studied. OBJECTIVES:To determine the symptom burden in persons dying with advanced dementia in the community setting. METHODS:The sample consisted of patient-caregiver dyads enrolled in the 24-month IN-PEACE trial of home-based care management for advanced dementia. Postmortem interviews were administered to caregivers of persons with advanced dementia who had recently died. The primary outcome was the Comfort Assessment in Dying-End of Life in Dementia (CAD-EOLD) scale. A secondary symptom measure was the Symptom Management in End of Life Dementia (SM-EOLD). Results were analyzed and compared to previous studies using the CAD-EOLD. RESULTS:Of the 83 persons with dementia who died during 24 months of follow-up, postmortem caregiver interviews were completed in 80 cases of which 79 completed the CAD-EOLD. Patients' mean age was 83.1; 68% were women, 57% white, and 37% African American. Most patients (62%) died at home and only 32% died in a hospital or nursing home. Hospice enrollment occurred in 75% of participants. Symptom management as reflected in CAD-EOLD scores was comparable to previous studies of persons with dementia dying in nursing homes. Compared to baseline SM-EOLD scores, end-of-life symptom scores were 4 points better. Intervention and control group outcomes were similar. CONCLUSIONS:Some community-dwelling persons with advanced dementia can die at home with symptom management comparable to that received in nursing homes.

ImportanceCare management benefits community-dwelling patients with dementia, but studies include few patients with moderate to severe dementia or from racial and ethnic minority populations, lack palliative care, and seldom reduce health care utilization.ObjectiveTo determine whether integrated dementia palliative care reduces dementia symptoms, caregiver depression and distress, and emergency department (ED) visits and hospitalizations compared with usual care in moderate to severe dementia.Design, Setting, and ParticipantsA randomized clinical trial of community-dwelling patients with moderate to severe dementia and their caregivers enrolled from March 2019 to December 2020 from 2 sites in central Indiana (2-year follow-up completed on January 7, 2023). Electronic health record screening identified patients with dementia; caregivers confirmed eligibility, including dementia stage.InterventionThe intervention consisted of monthly calls from a trained nurse or social worker and evidence-based protocols to help caregivers manage patients’ neuropsychiatric symptoms, caregiver distress, and palliative care issues (eg, advance care planning, symptoms, and hospice) (n = 99). Usual care caregivers received written dementia resource information and patients received care from usual clinicians (n = 102).Main Outcomes and MeasuresThe primary outcome was Neuropsychiatric Inventory Questionnaire (NPI-Q) severity score (scores range from 0-36, with higher scores indicating worse patient symptoms). Secondary outcomes included Symptom Management in End-of-Life Dementia scores, caregiver depression (Patient Health Questionnaire-8) scores, caregiver distress (NPI-Q distress) scores, and combined ED and hospitalization events. Outcomes were assessed quarterly for 24 months or until patient death.ResultsA total of 201 dyads were enrolled (patients were 67.7% female; 43.3% African American; mean [SD] age, 83.6 [7.9] years); 3 dyads withdrew and 83 patients died over the course of the study, with at least 90% of eligible dyads in both groups completing each of the quarterly assessments. For the dementia palliative care vs usual care groups, mean NPI-Q severity scores were 9.92 vs 9.41 at baseline and 9.15 vs 9.39 at 24 months, respectively (between-group difference at 24 months, −0.24 [95% CI, −2.33 to 1.84]). There was no significant difference in the rate of change in NPI-Q severity from baseline between groups over time (P = .87 for the group and time interaction). There were no significant differences in the secondary outcomes, except that there were fewer combined ED and hospitalization events in the dementia palliative care group (mean events/patient, 1.06 in dementia palliative care vs 2.37 in usual care; between-group difference, −1.31 [95% CI, −1.93 to −0.69]; relative risk, 0.45 [95% CI, 0.31 to 0.65]).Conclusions and RelevanceAmong community-dwelling patients with moderate to severe dementia and their caregivers, dementia palliative care, compared with usual care, did not significantly improve patients’ neuropsychiatric symptoms through 24 months.Trial RegistrationClinicalTrials.gov Identifier: NCT03773757

Abstract Dementia care management programs in the community demonstrate some benefits. Study limitations include limited numbers of people living with dementia (PLWD) with advanced disease or from minoritized populations; lack of palliative care components; and limited success reducing health care utilization. IN-PEACE tested dementia care management integrated with palliative care for PLWD with moderate-severe disease in the community and their caregivers. 201 PLWD-caregiver dyads were randomized to either a dementia care coordinator (99) or usual care (102) and followed for 24 months. Outcomes were neuropsychiatric symptoms (NPI-Q severity) and symptom management (SM-EOLD) in PLWD; distress (NPI-Q distress) and depression symptoms (PHQ-8) in caregivers; and the combined measure of ED visits/hospitalizations. Outcomes were assessed quarterly. Separate mixed effects models were run for each symptom/distress measure and a zero-inflated Poisson model compared the mean number of ED/hospitalization events. Subgroup analyses were conducted based on baseline NPI-Q severity, sex, race, income, and health system. There were no statistically significant differences between groups in any symptoms or distress measures in PLWD or caregivers. PLWD receiving the intervention, however, had substantially fewer ED/hospitalization events (means 1.06 events versus 2.37, p < 0.007). The intervention reduced the proportion of PLWD who had one or more ED/hospitalization events (78.4% of controls vs. 50.5% of intervention, p < 0.001). The relative risk reduction was 35.6% for an event, absolute risk reduction 27.9%, and number needed to treat (NNT) of 3.6. PLWD with higher NPI-Q at baseline and Black PLWD experienced greater reductions in ED visit hospitalization events.

Context Communication quality in the hospital impacts outcomes like satisfaction, depression, and anxiety for families, and assessment tools must be efficient and reliable. Objectives We developed the Quick FICS-5 and -10, shorter versions of the 30-item Family Inpatient Communication Survey (FICS). Methods The Quick FICS were developed from the original FICS study of hospitalized patients 65+ and their surrogates. The development sample came from the original FICS-30 scale. The validation sample came from a randomized controlled trial of surrogates for adult ICU patients. Participants were family members of patients on medical ICU or inpatient medicine services at three hospitals in a Midwest metropolitan area. We evaluated validity and reliability using factor analysis, internal consistency, and associations with surrogate psychological and decision-making outcomes. Results In the development sample of 364 patient/surrogate dyads, most surrogates were adult children (66.8%), female (70.9%), and white (68.9%). We identified 5-item and 10-item surveys. Exploratory factor analysis supported an overall communication score for the FICS-5 and FICS-10, as well as information and emotional support subscales for the FICS-10. There was high internal reliability (Cronbach's alpha was 0.83 for the FICS-5 and 0.90 for the FICS-10; information and emotional support subscales were 0.89 and 0.75 respectively). There was good predictive validity when comparing FICS scores to outcomes six to eight weeks after discharge, including anxiety (P = -0.13; P = 0.0234), and satisfaction with the hospital stay (P = 0.48; P < 0.0001). Similarly, the validation sample (n = 188) revealed Cronbach's alpha ranging from 0.81 to 0.93 and significant correlations (P < 0.05) with concurrent distress, anxiety, depression, and decision regret. Conclusions The Quick FICS offers efficient, valid, and reliable evaluation of communication quality in the hospital that can be useful for research and quality improvement.

Background: Dementia affects 6.5 million persons in the U.S., a number which is expected to double by 2060. More than half of persons with dementia die at home, creating an enormous burden for both patients and caregivers. However, there is a paucity of research on community-based palliative care interventions for advanced dementia.Objectives: The Indiana Palliative Excellence in Alzheimer's Care Efforts (IN-PEACE) study is a randomized trial to test the effectiveness of a collaborative predominantly telehealth home-based intervention for persons with advanced dementia residing in the community and their primary, informal caregivers. The primary aim is to determine if this palliative care focused supportive intervention is superior to usual care in reducing neuro-psychiatric symptoms of dementia. Secondarily, intervention effects on other patient symptoms (e.g., pain), caregiver distress and depression, and emergency department (ED)/hospitalization events are examined. Methods: The study population consists of participant pairs comprising a person with dementia and their primary, informal caregiver. The person with dementia must be >= 65 years old, with a clinical diagnosis of moderate to severe dementia. A total of 201 demographically and socioeconomically diverse participant pairs have been randomized to the IN-PEACE care coordination intervention (n = 99) or usual care (n = 102). Outcome as-sessments are conducted at baseline, and quarterly for up to 2 years (3, 6, 9, 12, 15, 18, 21, and 24 months).Discussion: IN-PEACE results will inform care for the large number of individuals with advanced dementia residing in the community and enable informal caregivers to provide effective home-based care. Trial registration: clinicaltrials.gov Identifier: NCT03773757.

INTRODUCTION hospital transfers and admissions are critical events in the care of nursing home residents. We sought to determine hospital transfer rates at different ages. METHODS a cohort of 1,187 long-stay nursing home residents who had participated in a Centers for Medicare and Medicaid demonstration project. We analysed the number of hospital transfers of the study participants recorded by the Minimum Data Set. Using a modern regression technique, we depicted the annual rate of hospital transfers as a smooth function of age. RESULTS transfer rates declined with age in a nonlinear fashion. Rates were the highest among residents younger than 60 years of age (1.30-2.15 transfers per year), relatively stable between 60 and 80 (1.17-1.30 transfers per year) and lower in those older than 80 (0.77-1.17 transfers per year). Factors associated with increased risk of transfers included prior diagnoses of hip fracture (annual incidence rate ratio or IRR: 2.057, 95% confidence interval (CI): [1.240, 3.412]), dialysis (IRR: 1.717, 95% CI: [1.313, 2.246]), urinary tract infection (IRR: 1.755, 95% CI: [1.361, 2.264]), pneumonia (IRR: 1.501, 95% CI: [1.072, 2.104]), daily pain (IRR: 1.297, 95% CI: [1.055,1.594]), anaemia (IRR: 1.229, 95% CI [1.068, 1.414]) and chronic obstructive pulmonary disease (IRR: 1.168, 95% CI: [1.010,1.352]). Transfer rates were lower in residents who had orders reflecting preferences for comfort care (IRR: 0.79, 95% CI: [0.665, 0.936]). DISCUSSION younger nursing home residents may require specialised interventions to reduce hospital transfers; declining transfer rates with the oldest age groups may reflect preferences for comfort-focused care.

Communication quality between clinical care teams and families impacts important outcomes like satisfaction, depression, and anxiety. Assessment tools must be efficient and reliable to be useful in the clinical environment. We report the development of 5 and 10-item versions of the validated 30-item Family Inpatient Communication Survey (FICS). Data were from 364 surrogate decision makers (SDMs) for incapacitated older adults in the ICU. Most SDMs were adult children (66.8%). SDMs were 70.9% female, 68.9% white, with a mean age of 58.3. Exploratory factor analysis revealed high internal reliability for the single-factor FICS5 (α= .88) and two-factor FICS10 (α= .93). The FICS10 reliably measures two subscales: information (α= .91) and emotional support (α=.81). Good discriminant and predictive validity were demonstrated when comparing total scores to outcomes at 6-8 weeks after hospital discharge, including anxiety (correlation coefficient (ρ)= -.13; p=.0234 ), depression (ρ =-.15; p=.0076), decision regret (ρ =--.15; p=.0066), and satisfaction (ρ =-..48; p<.0001). Repeating analysis with a new sample (n=188) revealed similar results with Cronbach’s alpha ranging from .81 to .93. The FICS5 revealed significant associations (p<.05) at 6-8 weeks after discharge with distress (-.22), while the FICS10 demonstrated significant associations with distress (-.28), anxiety (-.20), depression (-.19) and decision regret (-.27). Confirmatory factor analysis indicated adequate fit (CFI: FICS5= .994, FICS10= .994; RMSEA: FICS5= .093, FICS10= .103). The FICS provides clinicians and interventionists with a reliable, low burden tool to evaluate communication quality and respond quickly, which could impact satisfaction and other important outcomes for patients and families.

Background: Behavioral, psychological, and physical symptoms are prevalent in advanced dementia, as well as major contributors to poor quality of life, health care costs, caregiver burden, and nursing home placement. Objectives: To determine the frequency and severity of symptoms in persons with advanced dementia living in the community, as well as the association between symptoms and satisfaction with care, and the identification of factors associated with symptom burden. Design: Baseline data from a clinical trial testing the effectiveness of collaborative care home-based management for patients with advanced dementia. Setting/Subjects: Two hundred and one patient-caregiver dyads from an urban area in the United States, who were still residing in the community. Measurements: Caregivers completed the Symptom Management in End-of-Life Dementia (SM-EOLD) and Satisfaction with Care in End-of-Life Dementia (SWC-EOLD) scales. Results: Patients' mean age was 83.1; 67.7% were women, and most were either White (50.2%) or African American (43.8%). Most (88.1%) had severe dementia (Functional Assessment Staging Tool [FAST] stage 6 or 7). SM-EOLD mean score was 29.3 (on 0-45 scale) and SWC-EOLD score was 32.6 (on 10-40 scale). Pain, agitation, anxiety, and resistiveness to care were present at least weekly in ≥40% of patients. Multivariable linear regression modeling showed that higher neuropsychiatric symptom severity (assessed by the Neuropsychiatric Inventory), increased caregiver strain, and higher medical comorbidity were all independently associated with increased symptom burden. Satisfaction with care was high and had only a modest correlation (r = 0.20) with symptom burden. Conclusions: Community-dwelling patients with advanced dementia and their caregivers may benefit from home-based palliative care interventions to identify and manage burdensome symptoms.

People with Alzheimer’s disease or related dementias and diabetes mellitus (ADRD-DM) are at high risk for hypoglycemic events. Their cognitive impairment and psychosocial situation often hinder detection of hypoglycemia. Extending use and benefits of continuous glucose monitoring (CGM) to people with ADRD-DM could improve hypoglycemia detection, inform care, and reduce adverse events. However, cognitive impairment associated with ADRD presents unique challenges for CGM use. This commentary proposes applying the human-centered design process to CGM, investigating design solutions or interventions needed to integrate CGM into the health care of patients with ADRD-DM. With this process, we can identify and inform CGM designs for people with ADRD-DM, broadening CGM access, increasing detection and treatment of the silent threat posed by hypoglycemia.

Abstract The elevated risks associated with transferring nursing home residents to the hospital are problematic, but identifying which transfers can be avoided is complex. Using billing claims to determine “avoidability” based on hospital discharge diagnostic codes ignores resource constraints, clinical comorbidities, and asymmetrical information between nursing home staff making the transfer decision at the onset of clinical changes and hospital billing departments following treatment and diagnostic procedures. Conversely, relying on clinical staff assessments at the time of transfer may be an impractical and resource-intensive strategy to drive payment reform and improve quality. Using Medicare claims data representing emergency department and hospitalization transfers from 38 nursing facilities in Indiana from 2016-2020, we compared classification of transfers using a claims-based algorithm and trained nurse assessments of avoidability. Among 960 transfers, nurses judged 48.4% were potentially avoidable while 30.8% were classified as such using claims data. Of concordant assessments, 15.3% were avoidable and 36.0% as not avoidable. Of discordant assessments, 33.1% were judged avoidable by nurses only and 15.5% via the claims-based algorithm (Kappa=0.0153). Discordance was most frequent among transfers with heart failure (64%, n=42), psychosis (74.5%, n=34), acute renal disease (50%, n=28); and lowest among urinary tract infections (31.3%, n=64). No resident demographic or clinical characteristics were associated with discordance (age, race, sex, cognitive function scale, activities of daily living, or CHESS scale). High discordance in determining avoidability may be driven by presentation of symptoms or other condition-specific factors. Policies to reduce avoidable hospitalizations must not rely on overly simplistic approaches for identification.

Studies of patient decision-making use many different measures to evaluate the quality of decisions and the decision-making process, partly to determine whether the ethical goals of informed consent, patient autonomy, and shared decision-making have been achieved. We describe these measures, grouped under three main approaches, and review their limitations, leading to three conclusions. First, no measure or combination of measures can provide a complete assessment of decision quality. Second, the quality of a decision is best characterized vaguely, for instance as "good," "satisfactory," or "poor," and these categorizations depend on qualitative judgments that go beyond quantitative measures. Third, bioethicists should focus on identifying and addressing poor or problematic decisions, rather than trying to incrementally increase decision quality, quantified by a measure. Decision-quality measures can be useful in research and in advancing important goals of bioethics, as long as the challenges of defining and measuring decision quality are recognized.

Introduction Faculty development programs encouraging clinician educators' scholarship have been established at many medical schools. The same is true for programs that address the isolation and loneliness many faculty members feel in their day-to-day clinical work and administration. Few programs have explicitly combined development of scholarship and sense of community. Aim The goals of the Advanced Scholars Program for Internists in Research and Education (ASPIRE) are as follows: (1) provide training in scholarship development including research methods, implementation, and dissemination; (2) provide expert mentoring and support for professional development; and (3) create a greater sense of campus community. Setting ASPIRE scholars are clinician educators in the Department of Medicine at Indiana University School of Medicine. Program Description The program runs 18 months, includes intensive mentoring, covered time for scholars and mentors, resources, and two half-day educational sessions per month focused on scholarship and community development. Program Evaluation Institutional leaders' public statements and actions regarding ASPIRE were documented by program leadership. Data collected from ASPIRE mentors and scholars through interviews and free text survey responses were analyzed using an immersion/crystallization approach. Two central themes were identified for both scholars and mentors: benefits and challenges of the program. Benefits included mentors, program design, community development, increased confidence, skills development, improved patient care, and institutional impact. Challenges included time to accomplish the program, balance of community-building and skills development, and lack of a clear path post-ASPIRE. Discussion Combining skills-based learning with safe psychological space were judged important elements of success for the ASPIRE program. Conversations are ongoing to identify opportunities for scholars who have completed the program to continue to pursue scholarship, expand their skills, and build community. We conclude that the program both is feasible and was well-received. Sustainability and generalizability are important next steps in ensuring the viability of the program.

BACKGROUND:Clinical trials are needed to study topics relevant to older adults with serious illness. Investigators conducting clinical trials with this population are challenged by how to appropriately define, classify, report, and monitor serious and non-serious adverse events (SAEs/AEs), given that some traditionally reported AEs (pressure ulcers, delirium) and SAEs (death, hospitalization) are common in persons with serious illness, and may be consistent with their goals of care. OBJECTIVES:A multi-stakeholder group convened to establish greater clarity on and new approaches to address this critical issue. PARTICIPANTS:Thirty-two study investigators, members of regulatory and sponsor agencies, and patient stakeholders took part. APPROACH:The group met virtually four times and, using a collaborative approach, conducted a survey, select interviews, and reviewed regulatory guidance to collectively define the problem and identify a new approach. RESULTS:SAE/AE challenges fell into two areas: (1) definitions and classifications, including (a) implausible relationships, (b) misalignment with patient-centered care goals, and (c) well-known associations, and (2) reporting and monitoring, including (a) limited guidance, (b) inconsistent standards across regulators, and (c) Data Safety Monitoring Board (DSMB) member knowledge gaps. Problems largely reflected practice norms rather than regulatory requirements that already support context-specific and aggregate reporting. Approaches can be improved by adopting principles that better align strategies for addressing adverse events with the type of intervention being tested, favoring routine and aggregate over expedited reporting, and prioritizing how SAE/AEs relate to patient-centered care goals. Reporting plans and decisions should follow an algorithm underpinned by these principles. CONCLUSIONS:Adoption of the proposed approach-and supporting it with education and better alignment with regulatory guidance and procedures-could improve the quality and efficiency of clinical trials' safety involving older adults with serious illness and other vulnerable populations.

1.Describe the importance and meaning of POLST concordance.2.Describe study findings about factors predictive of POLST concordance. POLST is widely used to document the treatment preferences of nursing facility residents as orders, but it is unknown how well POLST reflects current preferences (concordance) and what factors are associated with concordance. To identify factors associated with concordance between existing POLST orders and current treatment preferences reported by nursing facility residents and/or their surrogate. Chart reviews were conducted at 29 Indiana nursing facilities. A stratified random sample of nursing facility residents (n = 123) and surrogates of residents without decisional capacity (n = 152) were interviewed about current treatment preferences. Concordance was determined by comparing existing POLST orders in the medical record with current treatment preferences for resuscitation, medical interventions, and artificial nutrition reported by residents or their surrogate in the research interview. Concordance was higher for orders about resuscitation (85.8%) than for orders about medical interventions (65.3%) and artificial nutrition (66.9%). Concordance was higher when the decision-maker was the surrogate versus the resident (68.9% vs. 31.1%, p < .001). In univariate analyses, better performance on a measure of cognitive functioning (OR 1.07, 95% CI 1.02 – 1.13, p < .01) and the surrogate as the decision-maker (OR 2.87, OR 1.73 – 4.75, p < .001) were both predictive of POLST concordance. POLST concordance is higher when decisions were made by surrogates of incapacitated nursing facility residents in comparison to residents who made their own decisions, and better cognition was predictive of concordance, reinforcing recommendations to periodically review POLST orders and involve family members in POLST discussions.

Background POLST is widely used to document the treatment preferences of nursing facility residents as orders, but it is unknown how well previously completed POLST orders reflect current preferences (concordance) and what factors are associated with concordance. Objectives To describe POLST preference concordance and identify factors associated with concordance. Design Chart reviews to document existing POLST orders and interviews to elicit current treatment preferences. Setting POLST-using nursing facilities (n = 29) in Indiana. Participants Nursing facility residents (n = 123) and surrogates of residents without decisional capacity (n = 152). Measurements Concordance was determined by comparing existing POLST orders for resuscitation, medical interventions, and artificial nutrition with current treatment preferences. Comfort-focused POLSTs contained orders for do not resuscitate, comfort measures, and no artificial nutrition. Results Overall, 55.7% (123/221) of residents and 44.7% (152/340) of surrogates participated (total n = 275). POLST concordance was 44%, but concordance was higher for comfort-focused POLSTs (68%) than for non-comfort-focused POLSTs (27%) (p < 0.001). In the unadjusted analysis, increasing resident age (OR 1.04, 95% CI 1.01-1.07, p < 0.01), better cognitive functioning (OR 1.07, 95% CI 1.02-1.13, p < 0.01), surrogate as the decision-maker (OR 2.87, OR 1.73-4.75, p < 0.001), and comfort-focused POLSTs (OR 6.01, 95% CI 3.29-11.00, p < 0.01) were associated with concordance. In the adjusted multivariable model, only having an existing comfort-focused POLST was associated with higher odds of POLST concordance (OR 5.28, 95% CI 2.59-10.73, p < 0.01). Conclusions Less than half of all POLST forms were concordant with current preferences, but POLST was over five times as likely to be concordant when orders reflected preferences for comfort-focused care. Findings suggest a clear need to improve the quality of POLST use in nursing facilities and focus its use among residents with stable, comfort-focused preferences.