Artificial Intelligence (IA) Advanced Triage Tool for G&O Emergencies

Emergencies Clinical Trial

— TIAGO  

Official title:

Evaluation of an Advanced Triage Tool for Gynecology and Obstetrics Emergencies Based on Artificial Intelligence Algorithms

Clinical Trial Details — Status: Active, not recruiting

Administrative data

• NCT number: NCT05382000
• Other study ID #: IIBSP-TIA-2021-81
• Status: Active, not recruiting
• Phase: N/A
• Start date: May 11, 2022
• Est. completion date: December 1, 2024

Study information

• Verified date: May 2024
• Source: Fundació Institut de Recerca de l'Hospital de la Santa Creu i Sant Pau
• Contact: n/a
• Is FDA regulated: No
• Study type: Interventional

Clinical Trial Summary

Triage represents the first opportunity to classify patients who come to an Emergency Department (ED) and to be able to identify, prioritize high-risk patients and efficiently allocate the limited resources that are available. Therefore, the purpose of triage in the ED is to prioritize patients, detecting those that are urgent (that cannot wait to be attended). Urgency is defined as that clinical situation with the capacity to generate deterioration or danger to the health or life of the patient, depending on the time elapsed between its appearance and the establishment of an effective treatment, which determines a healthcare episode with significant intervention needs in a short period of time. There are currently six triage systems or models systematically structured into 5 levels.

Although simple in concept, the practice of triage is challenging due to time pressure, the limitations of available information, the various medical conditions of the patients, and a great reliance on intuition on the part of the professionals who perform it. which conditions a great variability in it. On the other hand, almost half of adult ED visits nationwide are classified as level 3 in a 5-level structured triage system, which makes level 3 a heterogeneous group with patients with diverse pathologies, in which triage is not capable of accurately differentiating them, and this inability poses safety risks for the most severely ill patients ("under-triage") and may influence the accuracy and efficiency in resource allocation when patients with low acuity are overrated. Therefore, it seems necessary to develop new triage procedures that allow us to improve their accuracy and reduce inter-individual variability.

TIAGO is a prospective, single-center, observational, comparative study to determine the validity of the Mediktor ® Triage and its effectiveness with respect to the current triage system and the "gold standard" (physician's diagnosis).

Description:

Prospective interventional, comparative study to determine the validity of the Mediktor ® Triage and its effectiveness with respect to "Model Andorra of Triage" (MAT) system and the "gold standard" (doctor's diagnosis).

Obtaining informed consent. Participation in the study will be offered to all those patients who attend the Emergency Department of Gynecology and Obstetrics of the Hospital de la Santa Creu i Sant Pau during the study period and who meet the inclusion criteria. An information sheet on the study will also be provided to each patient.

Sequential triage Once the patient's consent has been obtained, the patient will be assessed sequentially in the same triage space. Initially, a nurse from the Gynecology and Obstetrics Service will classify the patient in the triage box using the MAT system according to the usual practice. Then, another professional from the center, trained in the use of the Mediktor Hospital ® tool, and who has not been present in the conventional sorting, will perform the advanced triage in the same space, both professionals being blind to the result of each of the tools. In the event that the first triage performed with MAT gave the investigators an emergency level 1, the triage with the Mediktor tool would not be performed, since in this case the immediate care of the patient would be prioritized.

Attention in the Emergency Service Once the sequential triage is completed, the patient will return to the Gynecology and Obstetrics Emergency Room. The patient's care will be performed according to usual clinical practice, following the triage assessment performed with the MAT system.

Data recovery and entry in Data Recovery Form (DRF) All data of the study variables will be retrieved from the emergency report issued in the area of Gynecology and Obstetrics Emergencies and will be entered in an electronic DRF for further analysis and processing.

Evaluation of Effectiveness

Main evaluation variables Level of triage assigned by MAT, Level of triage assigned by Mediktor Hospital

Secondary evaluation variables

• Affiliation variables (administrative): Date of birth, sex, residence, financing, date and time of arrival in the emergency room or administrative record, form of arrival in the emergency room (own foot, ambulance, etc.…), reason for the urgency ( common illness, traffic accident, school…).

• Triage variables: Triage date and time, triage duration time, coded clinical reason for consultation, readmission within 72 hours, readmission reason, triage level, number of reevaluations, triage level of each reevaluation.

• Care variables: date and time of the emergency room visit, request for additional test, type of additional test, analytical parameters requested (if applicable), diagnosis according to International Classification of Diseases (ICD) (primary and secondary), most important procedures performed, patient admission > 24h), date and time of admission, urgent surgery, time of stay in the emergency room (LOS) which is the time that the patient remains in the emergency room from the time of admission to the hospital until the patient is discharged or hospitalized. Number of patients admitted to the hospital and number of patients discharged. Number of emergency consultations / revisits in the first 72 hours after discharge.

• Variables of discharge: circumstance of discharge or reason for emergency discharge (home discharge, hospital admission, transfer to another center, voluntary discharge, escape, exit…), identification of the transfer center, date and time of discharge, date and time of administrative discharge, departure transport, cause of success, time spent in emergencies, registration canceled

Recruitment information / eligibility

• Status: Active, not recruiting
• Enrollment: 450
• Est. completion date: December 1, 2024
• Est. primary completion date: August 1, 2024
• Accepts healthy volunteers: No
• Gender: Female
• Age group: 18 Years and older

Eligibility

Inclusion Criteria:

• Being over 18 years

• Understand and accept the study procedures

• Sign the informed consent.

Exclusion Criteria:

• Not being able to understand the nature of the study and/or the procedures to be followed

• Not signing the informed consent

• Be under 18 years of age

• Emergency level 1 through current triage system

Related Conditions & MeSH terms

• Emergencies

Intervention

Diagnostic Test:
• Advanced triage tool for Gynecology and Obstetrics emergencies based on artificial intelligence algorithms.
After the conventional triage, a second independent doctor will make the suit with the Mediktor Hospital tool.. In less than 3 minutes and with an average of 14 questions, Mediktor performs an interrogation very similar to what an emergency doctor would do. The professional version allows the health professional to modify the course of the questions in the middle of the evaluation, if he considers it necessary to go deeper into some aspect of the anamnesis. The system allows you to see in real time the diseases that Mediktor considers possible during the evaluation. At the end of the triage process, Mediktor offers the level of urgency and a list of possible diagnoses based on the signs and symptoms answered. The professional can change the level of urgency if he considers it beneficial for the patient. Once the two triages (Conventional and Mediktor) have been carried out, the patient will be seen according to the care protocols of the center.

Locations

Country	Name	City	State
Spain	Hospital de la Santa Creu i Sant Pau	Barcelona

Sponsors (1)

Lead Sponsor	Collaborator
Fundació Institut de Recerca de l'Hospital de la Santa Creu i Sant Pau

Country where clinical trial is conducted

Spain, 

References & Publications (10)

Berlyand Y, Raja AS, Dorner SC, Prabhakar AM, Sonis JD, Gottumukkala RV, Succi MD, Yun BJ. How artificial intelligence could transform emergency department operations. Am J Emerg Med. 2018 Aug;36(8):1515-1517. doi: 10.1016/j.ajem.2018.01.017. Epub 2018 Jan 4. No abstract available. — View Citation

Elias P, Damle A, Casale M, Branson K, Churi C, Komatireddy R, Feramisco J. A Web-Based Tool for Patient Triage in Emergency Department Settings: Validation Using the Emergency Severity Index. JMIR Med Inform. 2015 Jun 10;3(2):e23. doi: 10.2196/medinform.3508. Erratum In: JMIR Med Inform. 2015 Jun 15;3(3):e24. — View Citation

Julian-Jimenez A, Palomo de los Reyes MJ, Lain Teres N. [Coment on the original article: modelo predictor de ingreso hospitalario a la llegada al servicio de Urgencias]. An Sist Sanit Navar. 2012 Sep-Dec;35(3):493-6; author reply 497-9. doi: 10.23938/ASSN.0113. No abstract available. Spanish. — View Citation

Kuriyama A, Urushidani S, Nakayama T. Five-level emergency triage systems: variation in assessment of validity. Emerg Med J. 2017 Nov;34(11):703-710. doi: 10.1136/emermed-2016-206295. Epub 2017 Jul 27. — View Citation

Levin S, Toerper M, Hamrock E, Hinson JS, Barnes S, Gardner H, Dugas A, Linton B, Kirsch T, Kelen G. Machine-Learning-Based Electronic Triage More Accurately Differentiates Patients With Respect to Clinical Outcomes Compared With the Emergency Severity Index. Ann Emerg Med. 2018 May;71(5):565-574.e2. doi: 10.1016/j.annemergmed.2017.08.005. Epub 2017 Sep 6. — View Citation

Moll HA. Challenges in the validation of triage systems at emergency departments. J Clin Epidemiol. 2010 Apr;63(4):384-8. doi: 10.1016/j.jclinepi.2009.07.009. Epub 2009 Oct 28. — View Citation

Moreno Barriga E, Pueyo Ferrer I, Sanchez Sanchez M, Martin Baranera M, Masip Utset J. [A new artificial intelligence tool for assessing symptoms in patients seeking emergency department care: the Mediktor application]. Emergencias. 2017 Dic;29(6):391-396. Spanish. — View Citation

Mueller B, Kinoshita T, Peebles A, Graber MA, Lee S. Artificial intelligence and machine learning in emergency medicine: a narrative review. Acute Med Surg. 2022 Mar 1;9(1):e740. doi: 10.1002/ams2.740. eCollection 2022 Jan-Dec. — View Citation

Raita Y, Goto T, Faridi MK, Brown DFM, Camargo CA Jr, Hasegawa K. Emergency department triage prediction of clinical outcomes using machine learning models. Crit Care. 2019 Feb 22;23(1):64. doi: 10.1186/s13054-019-2351-7. — View Citation

Storm-Versloot MN, Ubbink DT, Kappelhof J, Luitse JS. Comparison of an informally structured triage system, the emergency severity index, and the manchester triage system to distinguish patient priority in the emergency department. Acad Emerg Med. 2011 Aug;18(8):822-9. doi: 10.1111/j.1553-2712.2011.01122.x. — View Citation

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Number of patients with equivalence between emergency triage classifications	Correspondence of emergency grading between Advanced IA Triage Tool (Mediktor Hospital) and the current triage system.	3 days
Secondary	Number of patients with the same diagnosis on advanced triage tool and emergency discharge report (gold-standard)	Assess the correlation between the pre-diagnosis provided by the advanced triage tool and the diagnosis offered by the physician in the emergency discharge report.	3 days
Secondary	Number of patients with good correlation between complimentary tests requested by the advanced triage tool with gold-standard	Assess the correlation between the complementary tests proposed by the advanced triage tool and those requested by the doctors during the emergency room visit, following the care protocols of the center.	3 days