Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysis
Abstract Objectives Efficient and accurate emergency department (ED) triage is critical to prioritize the sickest patients and manage department flow. We explored the use of electronic health record data and advanced predictive analytics to improve triage performance. Methods Using a data set of ove...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-08-01
|
| Series: | Journal of the American College of Emergency Physicians Open |
| Online Access: | https://doi.org/10.1002/emp2.13003 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849412638760501248 |
|---|---|
| author | Dana R. Sax E. Margaret Warton Oleg Sofrygin Dustin G. Mark Dustin W. Ballard Mamata V. Kene David R. Vinson Mary E. Reed |
| author_facet | Dana R. Sax E. Margaret Warton Oleg Sofrygin Dustin G. Mark Dustin W. Ballard Mamata V. Kene David R. Vinson Mary E. Reed |
| author_sort | Dana R. Sax |
| collection | DOAJ |
| description | Abstract Objectives Efficient and accurate emergency department (ED) triage is critical to prioritize the sickest patients and manage department flow. We explored the use of electronic health record data and advanced predictive analytics to improve triage performance. Methods Using a data set of over 5 million ED encounters of patients 18 years and older across 21 EDs from 2016 to 2020, we derived triage models using deep learning to predict 2 outcomes: hospitalization (primary outcome) and fast‐track eligibility (exploratory outcome), defined as ED discharge with <2 resource types used (eg, laboratory or imaging studies) and no critical events (eg, resuscitative medications use or intensive care unit [ICU] admission). We report area under the receiver operator characteristic curve (AUC) and 95% confidence intervals (CI) for models using (1) triage variables alone (demographics and vital signs), (2) triage nurse clinical assessment alone (unstructured notes), and (3) triage variables plus clinical assessment for each prediction target. Results We found 12.7% of patients were hospitalized (n = 673,659) and 37.0% were fast‐track eligible (n = 1,966,615). The AUC was lowest for models using triage variables alone: AUC 0.77 (95% CI 0.77–0.78) and 0.70 (95% CI 0.70–0.71) for hospitalization and fast‐track eligibility, respectively, and highest for models incorporating clinical assessment with triage variables for both hospitalization and fast‐track eligibility: AUC 0.87 (95% CI 0.87–0.87) for both prediction targets. Conclusion Our findings highlight the potential to use advanced predictive analytics to accurately predict key ED triage outcomes. Predictive accuracy was optimized when clinical assessments were added to models using simple structured variables alone. |
| format | Article |
| id | doaj-art-5f78cac4c4e4452c925ccd6b19045159 |
| institution | Kabale University |
| issn | 2688-1152 |
| language | English |
| publishDate | 2023-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of the American College of Emergency Physicians Open |
| spelling | doaj-art-5f78cac4c4e4452c925ccd6b190451592025-08-20T03:34:22ZengElsevierJournal of the American College of Emergency Physicians Open2688-11522023-08-0144n/an/a10.1002/emp2.13003Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysisDana R. Sax0E. Margaret Warton1Oleg Sofrygin2Dustin G. Mark3Dustin W. Ballard4Mamata V. Kene5David R. Vinson6Mary E. Reed7Department of Emergency Medicine Kaiser East Bay and Kaiser Permanente Northern California Division of Research OaklandCaliforniaUSAKaiser Permanente Northern California Division of Research OaklandCaliforniaUSAUber San FranciscoCaliforniaUSADepartment of Emergency Medicine Kaiser East Bay and Kaiser Permanente Northern California Division of Research OaklandCaliforniaUSADepartment of Emergency Medicine Kaiser San Rafael and Kaiser Permanente Northern California Division of Research OaklandCaliforniaUSADepartment of Emergency Medicine Kaiser San Rafael and Kaiser Permanente Northern California Division of Research OaklandCaliforniaUSADepartment of Emergency Medicine Roseville, and Kaiser Permanente Northern California Division of Research OaklandCaliforniaUSAKaiser Permanente Northern California Division of Research OaklandCaliforniaUSAAbstract Objectives Efficient and accurate emergency department (ED) triage is critical to prioritize the sickest patients and manage department flow. We explored the use of electronic health record data and advanced predictive analytics to improve triage performance. Methods Using a data set of over 5 million ED encounters of patients 18 years and older across 21 EDs from 2016 to 2020, we derived triage models using deep learning to predict 2 outcomes: hospitalization (primary outcome) and fast‐track eligibility (exploratory outcome), defined as ED discharge with <2 resource types used (eg, laboratory or imaging studies) and no critical events (eg, resuscitative medications use or intensive care unit [ICU] admission). We report area under the receiver operator characteristic curve (AUC) and 95% confidence intervals (CI) for models using (1) triage variables alone (demographics and vital signs), (2) triage nurse clinical assessment alone (unstructured notes), and (3) triage variables plus clinical assessment for each prediction target. Results We found 12.7% of patients were hospitalized (n = 673,659) and 37.0% were fast‐track eligible (n = 1,966,615). The AUC was lowest for models using triage variables alone: AUC 0.77 (95% CI 0.77–0.78) and 0.70 (95% CI 0.70–0.71) for hospitalization and fast‐track eligibility, respectively, and highest for models incorporating clinical assessment with triage variables for both hospitalization and fast‐track eligibility: AUC 0.87 (95% CI 0.87–0.87) for both prediction targets. Conclusion Our findings highlight the potential to use advanced predictive analytics to accurately predict key ED triage outcomes. Predictive accuracy was optimized when clinical assessments were added to models using simple structured variables alone.https://doi.org/10.1002/emp2.13003 |
| spellingShingle | Dana R. Sax E. Margaret Warton Oleg Sofrygin Dustin G. Mark Dustin W. Ballard Mamata V. Kene David R. Vinson Mary E. Reed Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysis Journal of the American College of Emergency Physicians Open |
| title | Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysis |
| title_full | Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysis |
| title_fullStr | Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysis |
| title_full_unstemmed | Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysis |
| title_short | Automated analysis of unstructured clinical assessments improves emergency department triage performance: A retrospective deep learning analysis |
| title_sort | automated analysis of unstructured clinical assessments improves emergency department triage performance a retrospective deep learning analysis |
| url | https://doi.org/10.1002/emp2.13003 |
| work_keys_str_mv | AT danarsax automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis AT emargaretwarton automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis AT olegsofrygin automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis AT dustingmark automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis AT dustinwballard automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis AT mamatavkene automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis AT davidrvinson automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis AT maryereed automatedanalysisofunstructuredclinicalassessmentsimprovesemergencydepartmenttriageperformancearetrospectivedeeplearninganalysis |