ABSTRACTBackgroundOlder adults are frequent users of the Emergency Department (ED), with a significant proportion presenting with pre‐existing or acute cognitive impairment. While negative post‐ED outcomes associated with cognitive status are well documented, their direct impact on care processes and resource allocation within the hospital remains poorly understood. This study aims to quantify how different cognitive profiles affect costs and care needs for acutely ill older adults.MethodsWe conducted a secondary analysis of a prospective cohort study at a single, tertiary care hospital. We included patients aged ≥ 65 years admitted to the hospital through the ED. They were stratified into three groups based on the brief Confusion Assessment Method (bCAM) and the 10‐Point Cognitive Screener (10‐CS): normal cognition, cognitive impairment without delirium, and delirium. Primary outcome was cost of care. Resource utilization, characterized by the number of medical specialties involved, geriatric consultation, type of inpatient bed allocated from the ED, time to hospitalization, and patient satisfaction, were explored as secondary outcomes. Multiple regression models were used to assess associations, adjusting for sociodemographic factors, clinical severity, and geriatric vulnerability.ResultsThe sample comprised 824 patients: 429 (52.1%) with normal cognition, 165 (20.0%) with delirium, and 230 (27.9%) with cognitive impairment without delirium. Clinical severity, but not cognitive status, was independently associated with costs (B = 0.18; 95% CI: 0.08, 0.27). Delirium was independently associated with allocation to high‐complexity bed and receiving a geriatric consultation. Cognitive impairment was independently associated with a greater number of specialties involved.ConclusionsClinical severity showed the strongest association with costs. In contrast, cognitive profiles were independently associated with the care pathway and complexity, with delirium linked to higher‐acuity allocation and preexisting cognitive impairment without delirium to broader multidisciplinary involvement. Recognizing these distinct cognitive profiles is fundamental for anticipating care demands and optimizing resource allocation for this vulnerable population.

ABSTRACTBackgroundTransitions of care are high‐risk periods for patient safety in the emergency department (ED), particularly for patients who are still in the diagnostic process and are discharged with diagnostic uncertainty. Care transitions must be improved for these patients, as one third of discharged ED patients have diagnostic uncertainty. Yet there are no validated measures that assess the quality of care transitions from the ED, limiting the ability to assess the impact of interventions. Thus, we developed and validated the ED Transitions (EDT) measure.MethodsThis mixed methods study was conducted across a large healthcare system in three phases: item generation, cognitive interviewing, and large‐scale validation. Scale items were generated by experts and then iteratively refined using feedback from cognitive interviews (n = 11). The measure was then validated on a large sample of patients (n = 301) recently discharged from the ED. Exploratory structural equation modeling (ESEM) was employed to assess factor structure. Bivariate correlations were used to assess discriminant and convergent validity using the Care Transition Measure (CTM‐3) and the Communication Assessment Tool‐Teams (CAT‐T).ResultsThe measure was iteratively refined by way of an expert panel and cognitive interviews which resulted in a 15‐item measure to be used for validation. The validation sample (n = 301) was 62% women, 49% White, and the majority having Medicare and/or Medicaid (68%). Sequential comparisons between confirmatory factor analyses and ESEM resulted in a final 10‐item two‐factor structure. Reliability was excellent (0.93), and bivariate correlations indicated positive correlations between the EDT, CTM‐3, and CAT‐T.ConclusionThe EDT measure demonstrates content validity, structural validity, convergent validity, discriminant validity, and high internal consistency (i.e., reliability). This newly developed patient reported outcome measure can be used in future clinical and research work to better understand the impact of ED interventions on quality‐of‐care transitions for patients with diagnostic uncertainty.

ABSTRACTBackgroundPeople with dementia are often hospitalized from the emergency department (ED) for conditions that could potentially be treated in the outpatient setting. In scenarios of clinical equipoise—where either admission or discharge may be appropriate—little is known about how emergency physicians make disposition decisions.MethodsWe conducted a qualitative study with attending emergency physicians at a single academic medical center. Interviews included open‐ended questions about disposition decision‐making in general as well as structured case scenarios using a think‐aloud protocol. We used an inductive approach to code interview data and elucidate themes about the factors emergency physicians consider in disposition decisions for people with dementia in situations of clinical equipoise.ResultsWe interviewed twenty‐one physicians with an average of 10.5 years of clinical practice (range 5–21 years). Five themes emerged: (1) assessing availability of, reliability of, and burden faced by caregivers; (2) anticipating clinical trajectory and feasibility of outpatient follow‐up; (3) identifying acute cognitive changes that may reflect serious underlying illness; (4) navigating medicolegal concerns; and (5) weighing potential harms of hospitalization. Among these, caregiver assessments and clinical trajectories were discussed most frequently, with physicians describing informal and variable approaches to caregiving assessments. While many physicians recognized the risk of discharging a delirious patient, few acknowledged that hospitalization itself could contribute to delirium or other harms. Overall, emergency physicians’ decisions were shaped not only by patient characteristics, but by the surrounding health system's capacity to support timely and safe outpatient care.ConclusionsED disposition decisions for people with dementia often depend on caregiving context and health system constraints. Improving alignment of admission decisions with patients’ and care partners’ values and preferences may require structured approaches to assessing caregiver capacity and communicating risks of hospitalization. Health system investment in outpatient infrastructure and alternatives to admission may better support emergency physicians in making safe, patient‐centered decisions.

ABSTRACTObjectivesPain is common among patients presenting to the emergency department (ED) but is frequently underdetected and undertreated in older people living with dementia (PLWD). This systematic review examined whether dementia‐specific pain assessment tools improve pain management compared with usual care in the ED.MethodsWe conducted a systematic review and have reported the methods and results following PRISMA (PROSPERO: CRD420251044828). Eligible studies included randomized, quasi‐experimental, and observational designs enrolling ED patients aged ≥ 65 years with dementia or cognitive impairment. Interventions were pain assessment tools developed for PLWD, and comparisons were with standard pain scales. Primary outcomes were patient‐reported outcome measures and analgesia administration; secondary outcomes included repeated pain scores, ED revisits, functional decline, mortality, and adverse events. Five databases (Ovid MEDLINE, Embase, Cochrane Library, CINAHL, PsycInfo) and two clinical trial registries were searched without language or date restrictions on April 22, 2025, and December 16, 2025, respectively. Two reviewers independently screened, extracted data, and assessed risk of bias using Cochrane RoB‐2.ResultsOf 987 records identified, 18 underwent full‐text review, and one study met eligibility criteria. Fry et al. (2017) conducted a multicenter, cluster‐randomized controlled trial of 602 older adults with suspected long bone fractures, comparing the Pain Assessment in Advanced Dementia (PAINAD) tool with standard pain scales. No significant differences were observed in median time to first analgesia (83 vs. 82 min, p = 0.42) or proportion receiving analgesia within 60 min (28% vs. 32%, p = 0.19). Evidence certainty was rated very low.ConclusionsEvidence on dementia‐specific pain assessment tools in the ED is extremely limited. Available data suggest PAINAD does not improve timeliness of analgesia, underscoring the urgent need for rigorous studies to guide pain management for PLWD in the ED.

ABSTRACTObjectiveTo evaluate the diagnostic accuracy of a novel point of care (POC) high‐sensitivity troponin (hs‐cTn) assay, used alone or incorporated within validated decision aids, for acute myocardial infarction (AMI) in the prehospital setting.MethodsA pre‐specified secondary analysis of the Prehospital Evaluation of Sensitive Troponin (PRESTO) prospective diagnostic accuracy study, conducted in four ambulance services and 12 Emergency Departments (EDs; February 2019–March 2020). Paramedics included consenting adults with suspected AMI and no other reason for conveyance. Clinical data and venous blood were collected at the scene, and samples conveyed to hospital with participants. Plasma samples were later analyzed for hs‐cTn using a novel POC hs‐cTn assay (Abbott Point of Care i‐STAT hs‐TnI). The target condition was an adjudicated index diagnosis of type 1 AMI.ResultsOf 817 consenting participants, 704 were eligible for inclusion in this analysis, with type 1 AMI occurring in 89 (12.6%). At the limit of detection (< 2 ng/L), POC hs‐cTn had 100.0% sensitivity (95% CI 95 9%–100.0%) but only 4.6% specificity (95% CI 3.1%–6.5%). A Troponin‐only Manchester Acute Coronary Syndromes (T‐MACS) very‐low risk outcome identified 134 (19.7%) patients for non‐conveyance with 98.9% sensitivity (95% CI 94.9%–100.0%), 99.3% negative predictive value (NPV, 95% CI 95.0%–99.9%), and 22.5% specificity (95% CI 19.2%–26.1%). A low‐risk modified HEART score identified 150 (22.0%) patients with 93.2% sensitivity (95% CI 85.8%–97.5%), 96.0% NPV (91.6%–98.1%), and 24.3% specificity (95% CI 20.9%–27.9%). In an exploratory analysis, hs‐cTn < 5 ng/L identified 295 (41.9%) patients with 98.9% sensitivity (93.9%–100.0%), 99.7% NPV (97.7%–100.0%), and 47.8% specificity (95% CI 43.8%–51.8%).ConclusionsThis novel POC hs‐cTn assay achieves high sensitivity and NPV when used alongside the T‐MACS decision aid, but efficiency may be greater at a 5 ng/L threshold without requiring any decision aid.Trial RegistrationClinicalTrials.gov identifier: NCT03561051