Aren't you forgetting something? Cognitive screening beyond delirium in the emergency department

Abstract

Patients with cognitive impairment, particularly those living with dementia, are more likely to be admitted to the hospital and die after an emergency department (ED) visit.1, 2 Nonetheless, cognitive assessments at the ED are scarce and primarily focused on delirium.3, 4 Although delirium is a central aspect of acute geriatric care, patients with pre-existing cognitive impairment without delirium are at high risk of going undetected.5 Up to 63% of older adults hospitalized due to acute illnesses have some cognitive impairment.6 Thus, identifying cognitive impairment in the ED became a priority, particularly in the absence of delirium or previous dementia diagnosis.5 Given the insufficient research on the assessment and prognostic implications of delirium and cognitive impairment comorbidity,5 we aimed to describe their prevalence and prognostic implications for adverse in-hospital outcomes in acutely ill older patients admitted from the ED. We conducted a prospective cohort study of patients aged ≥65 who were admitted from the ED of a general tertiary hospital in Sao Paulo, Brazil, between November 2021 and April 2022. Trained research assistants assessed eligibility and completed standardized interviews with all patients (and their proxies) admitted from the ED. The baseline assessment at the ED included sociodemographic factors, medical history, comorbidities, and illness acuity.7 Delirium was screened using the brief Confusion Assessment Method (bCAM),4 and, when delirium was absent, pre-existing cognitive impairment was assessed using the 10-point cognitive screener (10-CS).8, 9 At the end of hospitalization, medical investigators blinded to the baseline interview reviewed the electronic medical records to document the outcomes of interest: length of hospital stay (LoS); in-hospital adverse events (c); and in-hospital death. The local Institutional Ethics Review Board approved the study, and patients or their proxy provided informed consent. We compared baseline and outcome variables according to the cognitive status groups in the ED: normal; cognitive impairment no delirium; or delirium. We used multivariable logistic regression models to explore associations between cognitive status and in-hospital outcomes. From 1178 admissions (Supplementary Figure S1), we included 830 patients with a mean age of 80 ± 9 years, 47% females (Table 1). We detected cognitive impairment in 401 (48%) patients – 171 were classified as having delirium and 230 as cognitive impairment no delirium. Cognitive impairment no delirium, and delirium were associated with older age, female sex, less education, higher comorbidity, and illness acuity (Table 1). Of note, 53% of the patients with cognitive impairment no delirium did not have a clinical history of memory complaints or a pre-existing diagnosis of dementia. Compared to older patients with normal cognition, those with cognitive impairment had more in-hospital adverse events, longer LoS, and higher in-hospital mortality, regardless of the presence of delirium (Table 1). Similar to those with delirium, patients with cognitive impairment no delirium presented higher in-hospital adverse outcomes than those with normal cognition, even after adjusting for confounders (Figure 1). We found that almost half of acutely ill older patients in the ED had impaired cognition. One in five patients experienced delirium. Moreover, among those without delirium, we observed that one in three patients had cognitive impairment according to 10-CS scores. Without a standard strategy for cognitive screening in the ED, cognitive impairment in the absence of delirium would remain highly undetected,3, 5 as most patients identified as cognitively impaired in the 10-CS did not have a previous dementia diagnosis or substantial cognitive complaints. Like delirium, cognitive impairment without delirium was associated with increased in-hospital adverse events, prolonged LoS, and in-hospital mortality, capturing risks beyond routine ED measures.1-3 Although barriers exist (e.g., resource-constrained settings and stressors related to acute illnesses),9, 10 our findings encourage broader cognitive screening for older patients in the ED. Our study should be interpreted considering our methodological approach. The freely available 10-CS takes only 2 min to administer, does not include tasks requiring additional supplies (e.g., pen, paper, image sheets), and is not influenced by a patient's ability to read, draw, or write.9 While these characteristics are advantages to implementing cognitive screening in the ED, we cannot reject the possibility that other practical cognitive batteries might have even greater predictive power in acutely ill older patients.10 Additionally, we did not explore proxy-based cognitive tools that would have allowed us to determine baseline cognitive impairment regardless of delirium. Finally, our results should be confirmed in studies involving more diverse populations and examining patient-centered outcomes (e.g., detailed investigation of cognitive status, disability, and healthcare use after discharge).9 In conclusion, combining the bCAM with 10-CS, two quick and easy-to-administer tools for detecting delirium and cognitive impairment, effectively identified acutely ill older patients at risk for in-hospital adverse outcomes. Moraes and Curiati: study concept and design, acquisition of data, data interpretation, and manuscript preparation. Avelino-Silva: study concept and design, data interpretation, and manuscript preparation. Morinaga: acquisition of data, data interpretation, and manuscript preparation. Apolinario and Suemoto: data interpretation and manuscript preparation. Aliberti: study concept and design, study supervision, acquisition of data, data interpretation, and manuscript preparation. The authors thank those who supported this study, particularly the ProAGE team, who cared for our patients and helped implement our Geriatric ED program. The study was funded by the Geriatric Center for Advanced Medicine, Hospital Sírio-Libanês, São Paulo. Gabriel Stanziola de Moraes is supported by an undergraduate research scholarship under the Institutional Program of Scientific Initiation Scholarships (PIBIC) from Brazilian Council for Scientific and Technological Development (Bolsa de iniciação cientifica CNPq 117947/2022-9). Claudia Kimie Suemoto receives a research fellowship from Brazilian Council for Scientific and Technological Development (Bolsa de produtividade em pesquisa CNPq 303883/2021-9). Márlon Juliano Romero Aliberti is supported by a post-doc scholarship from HCFMUSP with funds donated by Nubank under the #HCCOMVIDA scheme. The funders had no role in the design, methods, data acquisition, analysis, preparation, review, or approval of the manuscript and decision on its submission. The authors declare that there is no conflict of interest. The funders had no role in the design, methods, data acquisition, analysis, preparation, review, or approval of the manuscript and decision on its submission. Figure S1. Flowchart of the study participants. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts