Using machine-learning risk prediction models to triage the acuity of undifferentiated patients entering the emergency care system: a systematic review

Jamie Miles1, Janette Turner2, Richard Jacques2, Julia Williams3, Suzanne Mason2
1Yorkshire Ambulance Service, Brindley Way, Wakefield, WF2 0XQ, UK
2School of Health and Related Research, 3rd Floor, Regent Court (ScHARR), 30 Regent Street, Sheffield, S1 4DA, UK
3University of Herfordshire, Hatfield, Herfordshire, UK

Tóm tắt

Abstract Background The primary objective of this review is to assess the accuracy of machine learning methods in their application of triaging the acuity of patients presenting in the Emergency Care System (ECS). The population are patients that have contacted the ambulance service or turned up at the Emergency Department. The index test is a machine-learning algorithm that aims to stratify the acuity of incoming patients at initial triage. This is in comparison to either an existing decision support tool, clinical opinion or in the absence of these, no comparator. The outcome of this review is the calibration, discrimination and classification statistics. Methods Only derivation studies (with or without internal validation) were included. MEDLINE, CINAHL, PubMed and the grey literature were searched on the 14th December 2019. Risk of bias was assessed using the PROBAST tool and data was extracted using the CHARMS checklist. Discrimination (C-statistic) was a commonly reported model performance measure and therefore these statistics were represented as a range within each machine learning method. The majority of studies had poorly reported outcomes and thus a narrative synthesis of results was performed. Results There was a total of 92 models (from 25 studies) included in the review. There were two main triage outcomes: hospitalisation (56 models), and critical care need (25 models). For hospitalisation, neural networks and tree-based methods both had a median C-statistic of 0.81 (IQR 0.80-0.84, 0.79-0.82). Logistic regression had a median C-statistic of 0.80 (0.74-0.83). For critical care need, neural networks had a median C-statistic of 0.89 (0.86-0.91), tree based 0.85 (0.84-0.88), and logistic regression 0.83 (0.79-0.84). Conclusions Machine-learning methods appear accurate in triaging undifferentiated patients entering the Emergency Care System. There was no clear benefit of using one technique over another; however, models derived by logistic regression were more transparent in reporting model performance. Future studies should adhere to reporting guidelines and use these at the protocol design stage. Registration and funding This systematic review is registered on the International prospective register of systematic reviews (PROSPERO) and can be accessed online at the following URL: https://www.crd.york.ac.uk/PROSPERO/display_record.php?ID=CRD42020168696 This study was funded by the NIHR as part of a Clinical Doctoral Research Fellowship.

Từ khóa


Tài liệu tham khảo

Murphy KP. Machine learning: a probabilistic perspective. London: the MIT press; 2012.

Bektas F, Eken C, Soyuncu S, Kilicaslan I, Cete Y. Artificial neural network in predicting craniocervical junction injury: an alternative approach to trauma patients. Eur J Emerg Med. 2008 Dec;15(6):318–23.

Walsh P, Cunningham P, Rothenberg SJ, O’Doherty S, Hoey H, Healy R. An artificial neural network ensemble to predict disposition and length of stay in children presenting with bronchiolitis. Eur J Emerg Med. 2004;11(5):259–64.

Molaei S, Korley FK, Soroushmehr SMR, Falk H, Sair H, Ward K, et al. A machine learning based approach for identifying traumatic brain injury patients for whom a head CT scan can be avoided. In: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. Institute of Electrical and Electronics Engineers Inc.; 2016. p. 2258–61.

Vanhouten JP, Starmer JM, Lorenzi NM, Maron DJ, Lasko TA. Machine learning for risk prediction of acute coronary syndrome.

Harrison RF, Kennedy RL. Artificial neural network models for prediction of acute coronary syndromes using clinical data from the time of presentation. Ann Emerg Med. 2005 Nov 1;46(5):431–9.

Weber EJ. Triage: Making the simple complex? Emerg Med J. 2018;36(2):64–5.

O’Keeffe C, Mason S, Jacques R, Nicholl J. Characterising non-urgent users of the emergency department (ED): a retrospective analysis of routine ED data. PLoS One. 2018;13(2):1–14.

Stewart J, Sprivulis P, Dwivedi G. Artificial intelligence and machine learning in emergency medicine. Emerg Med Australas [Internet]. 2018 Dec;30(6):870–4. Available from: http://doi.wiley.com/10.1111/1742-6723.13145.

Berlyand Y, Raja AS, Dorner SC, Prabhakar AM, Sonis JD, Gottumukkala R V., et al. How artificial intelligence could transform emergency department operations. Am J Emerg Med [Internet]. 2018 Aug;36(8):1515–7. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0735675718300184.

Grant K, McParland A. Applications of artificial intelligence in emergency medicine. Univ Toronto Med J. 2019;96(1):37–9.

Liu N, Zhang Z, Wah Ho AF, Ong MEH. Artificial intelligence in emergency medicine. J Emerg Crit Care Med. 2018;2(4):82–82.

Aacharya RP, Gastmans C, Denier Y. Emergency department triage: an ethical analysis. BMC Emerg Med. 2011 Oct 7;11:16.

Brousseau DC, Hoffmann RG, Nattinger AB, Flores G, Zhang Y, Gorelick M. Quality of primary care and subsequent pediatric emergency department utilization. Pediatrics. 2007 Jun 1;119(6):1131–8.

Simpson R, Croft S, O’Keeffe C, Jacques R, Stone T, Ahmed N, et al. Exploring the characteristics, acuity and management of adult ED patients at night-time. Emerg Med J. 2019 Sep 1;36(9):554–7.

McCusker J, Karp I, Cardin S, Durand P, Morin J. Determinants of emergency department visits by older adults: a systematic review. Acad Emerg Med [Internet]. 2003 Dec 1 [cited 2020 Mar 6];10(12):1362–70. Available from: http://doi.wiley.com/10.1197/S1069-6563(03)00539-6.

Latham LP, Ackroyd-Stolarz S. Emergency department utilization by older adults: a descriptive study. Can Geriatr J. 2014 Dec 1;17(4):118–25.

Lehmann CU, Barr J, Kelly PJ. Emergency department utilization by adolescents. J Adolesc Heal. 1994;15(6):485–90.

Ziv A, Boulet JR, Slap GB. Emergency department utilization by adolescents in the United States. Pediatrics. 1998 Jun 1;101(6):987–94.

Steyerberg EW. Clinical prediction models: a practical approach to development, validation, and updating [Internet]. New York, NY: Springer New York; 2009. (Statistics for Biology and Health). Available from: http://link.springer.com/10.1007/978-0-387-77244-8.

Moons KGM, de Groot JAH, Bouwmeester W, Vergouwe Y, Mallett S, Altman DG, et al. Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist. PLoS Med [Internet]. 2014 Oct 14 [cited 2020 Mar 1];11(10):e1001744. Available from: https://dx.plos.org/10.1371/journal.pmed.1001744.

Wolff RF, Moons KGM, Riley RD, Whiting PF, Westwood M, Collins GS, et al. PROBAST: a tool to assess the risk of bias and applicability of prediction model studies. Ann Intern Med. 2019 Jan 1;170(1):51–8.

Macaskill P, Gatsonis C, Deeks J, Harbord R, Takwoingi Y. Analysing and presenting results. In: The Cochrane collaboration, editor. Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy [Internet]. Version 1. 2010. Available from: http://srdta.cochrane.org/.

Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med [Internet]. 2009 Jul 21 [cited 2020 Mar 6];6(7):e1000097. Available from: https://dx.plos.org/10.1371/journal.pmed.1000097.

Azeez D, Ali MAM, Gan KB, Saiboon I. Comparison of adaptive neuro-fuzzy inference system and artificial neutral networks model to categorize patients in the emergency department. Springerplus. 2013;2(1):1–10.

Caicedo-Torres W, Hernando Pinzon G. A machine learning model for triage in lean paediatric emergency departments. Montes y Gómez M, Escalante HJ, Segura A, Murillo J de D, editors. 2016;10022(November 2016):259–70. Available from: http://link.springer.com/10.1007/978-3-319-47955-2.

Cameron A, Rodgers K, Ireland A, Jamdar R, McKay GA. A simple tool to predict admission at the time of triage. Emerg Med J [Internet]. 2015;32(3):174–9. Available from: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=med11&NEWS=N&AN=24421344.

Dinh MM, Russell SB, Bein KJ, Rogers K, Muscatello D, Paoloni R, et al. The Sydney Triage to Admission Risk Tool (START) to predict emergency department disposition: a derivation and internal validation study using retrospective state-wide data from New South Wales, Australia. BMC Emerg Med [Internet]. 2016;16(1):1–7 Available from: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=med12&NEWS=N&AN=27912757.

Dugas AF, Kirsch TD, Toerper M, Korley F, Yenokyan G, France D, et al. An electronic emergency triage system to improve patient distribution by critical outcomes. J Emerg Med. 2016;50(6):910–8.

Golmohammadi D. Predicting hospital admissions to reduce emergency department boarding. Int J Prod Econ. 2016;182(September):535–44.

Goto T, Camargo CAJ, Faridi MK, Freishtat RJ, Hasegawa K. Machine learning-based prediction of clinical outcomes for children during emergency department triage. JAMA Netw open [Internet]. 2019;2(1):e186937. Available from: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=prem&NEWS=N&AN=30646206.

Hong WS, Haimovich AD, Taylor RA. Predicting hospital admission at emergency department triage using machine learning. PLoS One [Internet]. 2018;13(7):1–13 Available from: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=medl&NEWS=N&AN=30028888.

Kim D, You S, So S, Lee J, Yook S, Jang DP, et al. A data-driven artificial intelligence model for remote triage in the prehospital environment. PLoS One. 2018;13(10):1–14.

Kim SW, Li JY, Hakendorf P, Teubner DJJO, Ben-Tovim DI, Thompson CH. Predicting admission of patients by their presentation to the emergency department. EMA - Emerg Med Australas. 2014 Aug;26(4):361–7.

Kwon Jmyoung, Lee YY, Lee YY, Lee S, Park H, Park J. Validation of deep-learning-based triage and acuity score using a large national dataset. PLoS One [Internet]. 2018;13(10):1–10. Available from: https://doi.org/10.1371/journal.pone.0205836.

Kwon J, Jeon K-H, Lee M, Kim K-H, Park J, Oh B-H. Deep learning algorithm to predict need for critical care in pediatric emergency departments. Pediatr Emerg Care [Internet]. 2019 Jul;1. Available from: http://insights.ovid.com/crossref?an=00006565-900000000-98117.

Levin S, Toerper M, Hamrock E, Hinson JS, Barnes S, Gardner H, et al. Machine-learning-based electronic triage more accurately differentiates patients with respect to clinical outcomes compared with the emergency severity index. Ann Emerg Med. 2018;71(5):565-574.e2.

Li J, Guo L, Handly N. Hospital admission prediction using pre-hospital variables. 2009 IEEE Int Conf Bioinforma Biomed BIBM 2009. 2009;283–286.

Meisel ZF, Pollack CV, Mechem CC, Pines JM. Derivation and internal validation of a rule to predict hospital admission in prehospital patients. Prehospital Emerg Care. 2008;12(3):314–9.

Newgard CD, Hsia RY, Mann NC, Schmidt T, Sahni R, Bulger EM, et al. The trade-offs in field trauma triage. J Trauma Acute Care Surg [Internet]. 2013;74(5):1298–306 Available from: http://insights.ovid.com/crossref?an=01586154-201305000-00017.

Olivia D, Nayak A, Balachandra M. Machine learning based electronic triage for emergency department. In 2018. p. 215–21. Available from: http://link.springer.com/10.1007/978-981-13-2907-4_19.

Raita Y, Goto T, Faridi MK, Brown DFM, Camargo CA, Hasegawa K. Emergency department triage prediction of clinical outcomes using machine learning models. Crit Care [Internet]. 2019 22;23(1):64. Available from: https://ccforum.biomedcentral.com/articles/10.1186/s13054-019-2351-7.

Rendell K, Koprinska I, Kyme A, Ebker-White AA, Dinh MM. The Sydney Triage to Admission Risk Tool (START2) using machine learning techniques to support disposition decision-making. EMA - Emerg Med Australas. 2019;31(3):429–35.

Seymour CW, Kahn JM, Cooke CR, Watkins TR, Rea TD. During out-of-hospital emergency care. 2010;304(7):747–54.

van Rein EAJ, van der Sluijs R, Voskens FJ, Lansink KWW, Houwert RM, Lichtveld RA, et al. Development and Validation of a Prediction Model for Prehospital Triage of Trauma Patients. JAMA Surg [Internet]. 2019;154(5):421–9 Available from: http://search.ebscohost.com/login.aspx?direct=true&db=cin20&AN=136501962&site=ehost-live.

Wang S-T. Construct an optimal triage prediction model: a case study of the emergency department of a teaching hospital in Taiwan. J Med Syst [Internet]. 2013t 29;37(5):9968. Available from: http://link.springer.com/10.1007/s10916-013-9968-x.

Zhang X, Kim J, Patzer RE, Pitts SR, Patzer A, Schrager JD. Prediction of emergency department hospital admission based on natural language processing and neural networks. Methods Inf Med [Internet]. 2017;56(05):377–89 Available from: http://www.thieme-connect.de/DOI/DOI?10.3414/ME17-01-0024.

Zlotnik A, Alfaro MC, Pérez MCP, Gallardo-Antolín A, Martínez JMM. Building a decision support system for inpatient admission prediction with the Manchester triage system and administrative check-in variables. CIN Comput Informatics, Nurs [Internet]. 2016 May;34(5):224–30. Available from: http://content.wkhealth.com/linkback/openurl?sid=WKPTLP:landingpage&an=00024665-201605000-00006.

Zmiri D, Shahar Y, Taieb-Maimon M. Classification of patients by severity grades during triage in the emergency department using data mining methods. J Eval Clin Pract. 2012;18(2):378–88 Available from: http://doi.wiley.com/10.1111/j.1365-2753.2010.01592.x.

Debray TPA, Damen JAAG, Snell KIE, Ensor J, Hooft L, Reitsma JB, et al. A guide to systematic review and meta-analysis of prediction model performance. BMJ. 2017;356.

Takwoingi Y, Quinn TJ. Review of diagnostic test accuracy (DTA) studies in older people. Age Ageing. 2018;47(3):349–55.

Mirhaghi A, Heydari A, Mazlom R, Hasanzadeh F. Reliability of the emergency severity index: meta-analysis. Sultan Qaboos Univ Med J. 2015;15(1):e71–7.

Liu N, Lin Z, Cao J, Koh Z, Zhang T, Bin HG, et al. An intelligent scoring system and its application to cardiac arrest prediction. IEEE Trans Inf Technol Biomed. 2012;16(6):1324–31.

American Heart Association. Cardiac arrest [Internet]. 2020 [cited 2020 Jun 2]. Available from: https://www.heart.org/en/health-topics/cardiac-arrest#:~:text=About Cardiac Arrest,the wake of other symptoms.

Collins GS, Ogundimu EO, Cook JA, Le Manach Y, Altman DG. Quantifying the impact of different approaches for handling continuous predictors on the performance of a prognostic model. 2016;.

Riley RD, van der Windt D, Croft P, Moons KGM, editors. Prognosis research in health care [Internet]. Oxford University Press; 2019. Available from: http://www.oxfordmedicine.com/view/10.1093/med/9780198796619.001.0001/med-9780198796619.

Sun G-W, Shook TL, Kay GL. Inappropriate use of bivariable analysis to screen risk factors for use in multivariable analysis. Vol. 49, J Clin Epidemiol. 1996.

Caetano SJ, Sonpavde G, Pond GR. C-statistic: a brief explanation of its construction, interpretation and limitations. Eur J Cancer. 2018;90:130–2.

Moons KGM, Altman DG, Reitsma JB, Ioannidis JPA, Macaskill P, Steyerberg EW, et al. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): explanation and elaboration. Ann Intern Med. 2015 Jan 6;162(1):W1–73.

Moons KGM, Wolff RF, Riley RD, Penny ;, Whiting F, Westwood M, et al. PROBAST: a tool to assess risk of bias and applicability of prediction model studies: Explanation and Elaboration Annals of Internal Medicine RESEARCH AND REPORTING METHODS. Ann Intern Med [Internet]. 2019 [cited 2020 Mar 8];170:1–33. Available from: www.probast.org.