Dataset Information

Machine learning-assisted screening for cognitive impairment in the emergency department.

ABSTRACT:

Background/objectives

Despite a high prevalence and association with poor outcomes, screening to identify cognitive impairment (CI) in the emergency department (ED) is uncommon. Identification of high-risk subsets of older adults is a critical challenge to expanding screening programs. We developed and evaluated an automated screening tool to identify a subset of patients at high risk for CI.

Methods

In this secondary analysis of existing data collected for a randomized control trial, we developed machine-learning models to identify patients at higher risk of CI using only variables available in electronic health record (EHR). We used records from 1736 community-dwelling adults age > 59 being discharged from three EDs. Potential CI was determined based on the Blessed Orientation Memory Concentration (BOMC) test, administered in the ED. A nested cross-validation framework was used to evaluate machine-learning algorithms, comparing area under the receiver-operator curve (AUC) as the primary metric of performance.

Results

Based on BOMC scores, 121 of 1736 (7%) participants screened positive for potential CI at the time of their ED visit. The best performing algorithm, an XGBoost model, predicted BOMC positivity with an AUC of 0.72. With a classification threshold of 0.4, this model had a sensitivity of 0.73, a specificity of 0.64, a negative predictive value of 0.97, and a positive predictive value of 0.13. In a hypothetical ED with 200 older adult visits per week, the use of this model would lead to a decrease in the in-person screening burden from 200 to 77 individuals in order to detect 10 of 14 patients who would fail a BOMC.

Conclusion

This study demonstrates that an algorithm based on EHR data can define a subset of patients at higher risk for CI. Incorporating such an algorithm into a screening workflow could allow screening efforts and resources to be focused where they have the most impact.

SUBMITTER: Yadgir SR

PROVIDER: S-EPMC8904269 | biostudies-literature | 2022 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning-assisted screening for cognitive impairment in the emergency department.

Yadgir Simon R SR Engstrom Collin C Jacobsohn Gwen Costa GC Green Rebecca K RK Jones Courtney M C CMC Cushman Jeremy T JT Caprio Thomas V TV Kind Amy J H AJH Lohmeier Michael M Shah Manish N MN Patterson Brian W BW

Journal of the American Geriatrics Society 20211013 3

<h4>Background/objectives</h4>Despite a high prevalence and association with poor outcomes, screening to identify cognitive impairment (CI) in the emergency department (ED) is uncommon. Identification of high-risk subsets of older adults is a critical challenge to expanding screening programs. We developed and evaluated an automated screening tool to identify a subset of patients at high risk for CI.<h4>Methods</h4>In this secondary analysis of existing data collected for a randomized control tr ...[more]

PMID: 34643944

Similar Datasets

Project description:ObjectiveTo predict hospital admission at the time of ED triage using patient history in addition to information collected at triage.MethodsThis retrospective study included all adult ED visits between March 2014 and July 2017 from one academic and two community emergency rooms that resulted in either admission or discharge. A total of 972 variables were extracted per patient visit. Samples were randomly partitioned into training (80%), validation (10%), and test (10%) sets. We trained a series of nine binary classifiers using logistic regression (LR), gradient boosting (XGBoost), and deep neural networks (DNN) on three dataset types: one using only triage information, one using only patient history, and one using the full set of variables. Next, we tested the potential benefit of additional training samples by training models on increasing fractions of our data. Lastly, variables of importance were identified using information gain as a metric to create a low-dimensional model.ResultsA total of 560,486 patient visits were included in the study, with an overall admission risk of 29.7%. Models trained on triage information yielded a test AUC of 0.87 for LR (95% CI 0.86-0.87), 0.87 for XGBoost (95% CI 0.87-0.88) and 0.87 for DNN (95% CI 0.87-0.88). Models trained on patient history yielded an AUC of 0.86 for LR (95% CI 0.86-0.87), 0.87 for XGBoost (95% CI 0.87-0.87) and 0.87 for DNN (95% CI 0.87-0.88). Models trained on the full set of variables yielded an AUC of 0.91 for LR (95% CI 0.91-0.91), 0.92 for XGBoost (95% CI 0.92-0.93) and 0.92 for DNN (95% CI 0.92-0.92). All algorithms reached maximum performance at 50% of the training set or less. A low-dimensional XGBoost model built on ESI level, outpatient medication counts, demographics, and hospital usage statistics yielded an AUC of 0.91 (95% CI 0.91-0.91).ConclusionMachine learning can robustly predict hospital admission using triage information and patient history. The addition of historical information improves predictive performance significantly compared to using triage information alone, highlighting the need to incorporate these variables into prediction models.

Project description:ObjectivesWe aimed to describe emergency department (ED) care transition interventions delivered to older adults with cognitive impairment, identify relevant patient-centered outcomes, and determine priority research areas for future investigation.DesignSystematic scoping review.Setting and participantsED patients with cognitive impairment and/or their care partners.MethodsInformed by the clinical questions, we conducted systematic electronic searches of medical research databases for relevant publications following published guidelines. The results were presented to a stakeholder group representing ED-based and non-ED-based clinicians, individuals living with cognitive impairment, care partners, and advocacy organizations. After discussion, they voted on potential research areas to prioritize for future investigations.ResultsFrom 3848 publications identified, 78 eligible studies underwent full text review, and 10 articles were abstracted. Common ED-to-community care transition interventions for older adults with cognitive impairment included interdisciplinary geriatric assessments, home visits from medical personnel, and telephone follow-ups. Intervention effects were mixed, with improvements observed in 30-day ED revisit rates but most largely ineffective at promoting connections to outpatient care or improving secondary outcomes such as physical function. Outcomes identified as important to adults with cognitive impairment and their care partners included care coordination between providers and inclusion of care partners in care management within the ED setting. The highest priority research area for future investigation identified by stakeholders was identifying strategies to tailor ED-to-community care transitions for adults living with cognitive impairment complicated by other vulnerabilities such as social isolation or economic disadvantage.Conclusions and implicationsThis scoping review identified key gaps in ED-to-community care transition interventions delivered to older adults with cognitive impairment. Combined with a stakeholder assessment and prioritization, it identified relevant patient-centered outcomes and clarifies priority areas for future investigation to improve ED care for individuals with impaired cognition, an area of critical need given the current population trends.

Project description:ObjectivesTo identify research and practice gaps to establish future research priorities to advance the detection of cognitive impairment and dementia in the emergency department (ED).DesignLiterature review and consensus-based rankings by a transdisciplinary, stakeholder task force of experts, persons living with dementia, and care partners.Setting and participantsScoping reviews focused on adult ED patients.MethodsTwo systematic scoping reviews of 7 medical research databases focusing on best tools and approaches for detecting cognitive impairment and dementia in the ED in terms of (1) most accurate and (2) most pragmatic to implement. The results were screened, reviewed, and abstracted for relevant information and presented at the stakeholder consensus conference for discussion and ranked prioritization.ResultsWe identified a total of 1464 publications and included 45 to review for accurate tools and approaches for detecting cognitive impairment and dementia. Twenty-seven different assessments and instruments have been studied in the ED setting to evaluate cognitive impairment and dementia, with many focusing on sensitivity and specificity of instruments to screen for cognitive impairment. For pragmatic tools, we identified a total of 2166 publications and included 66 in the review. Most extensively studied tools included the Ottawa 3DY and Six-Item Screener (SIS). The SIS was the shortest to administer (1 minute). Instruments with the highest negative predictive value were the SIS (vs MMSE) and the 4 A's Test (vs expert diagnosis). The GEAR 2.0 Advancing Dementia Care Consensus conference ranked research priorities that included the need for more approaches to recognize more effectively and efficiently persons who may be at risk for cognitive impairment and dementia, while balancing the importance of equitable screening, purpose, and consequences of differentiating various forms of cognitive impairment.Conclusions and implicationsThe scoping review and consensus process identified gaps in clinical care that should be prioritized for research efforts to detect cognitive impairment and dementia in the ED setting. These gaps will be addressed as future GEAR 2.0 research funding priorities.

Project description:IntroductionAs the population ages, Alzheimer's disease and related dementias (ADRD) are becoming increasingly common in patients presenting to the emergency department (ED). This study compares the frequency of ED use among a cohort of individuals with well-defined cognitive performance (cognitively intact, mild cognitive impairment (MCI), and ADRD).MethodsWe performed a retrospective cohort study of English-speaking, community-dwelling individuals evaluated at four health system-based multidisciplinary memory clinics from 2014-2016. We obtained demographic and clinical data, including neuropsychological testing results, through chart review and linkage to electronic health record data. We characterized the frequency and quantity of ED use within one year (6 months before and after) of cognitive evaluation and compared ED use between the three groups using bivariate and multivariate approaches.ResultsOf the 779 eligible patients, 89 were diagnosed as cognitively intact, 372 as MCI, and 318 as ADRD. The proportion of subjects with any annual ED use did not increase significantly with greater cognitive impairment: cognitively intact (16.9%), MCI (26.1%), and ADRD (28.9%) (p = 0.072). Average number of ED visits increased similarly: cognitively intact (0.27, SD 0.72), MCI (0.41, SD 0.91), and ADRD (0.55, SD 1.25) (p = 0.059). Multivariate logistic regression results showed that patients with MCI (odds ratio (OR) 1.62; CI = 0.87-3.00) and ADRD (OR 1.84; CI = 0.98-3.46) did not significantly differ from cognitively intact adults in any ED use. Multivariate negative binomial regression found patients with MCI (incidence rate ratio (IRR) 1.38; CI = 0.79-2.41) and ADRD (IRR 1.76, CI = 1.00-3.10) had elevated but non-significant risk of an ED visit compared to cognitively intact individuals.ConclusionThough there was no significant difference in ED use in this small sample from one health system, our estimates are comparable to other published work. Results suggested a trend towards higher utilization among adults with MCI or ADRD compared to those who were cognitively intact. We must confirm our findings in other settings to better understand how to optimize systems of acute illness care for individuals with MCI and ADRD.

Dataset Information

Machine learning-assisted screening for cognitive impairment in the emergency department.

Background/objectives

Methods

Results

Conclusion

Publications

Machine learning-assisted screening for cognitive impairment in the emergency department.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets