Triage: an investigation of the process and potential vulnerabilities

Comparative analysis of ChatGPT, Gemini and emergency medicine specialist in ESI triage assessment

INTRODUCTION

The term Artificial Intelligence (AI) was first coined in the 1960s and has made significant progress up to the present day. During this period, numerous AI applications have been developed. GPT-4 and Gemini are two of the best-known of these AI models. As a triage system The Emergency Severity Index (ESI) is currently one of the most commonly used for effective patient triage in the emergency department. The aim of this study is to evaluate the performance of GPT-4, Gemini, and emergency medicine specialists in ESI triage against each other; furthermore, it aims to contribute to the literature on the usability of these AI programs in emergency department triage.

METHODS

Our study was conducted between February 1, 2024, and February 29, 2024, among emergency medicine specialists in Turkey, as well as with GPT-4 and Gemini. Ten emergency medicine specialists were included in our study but as a limitation the emergency medicine specialists participating in the study do not frequently use the ESI triage model in daily practice. In the first phase of our study, 100 case examples related to adult or trauma patients were extracted from the sample and training cases found in the ESI Implementation Handbook. In the second phase of our study, the provided responses were categorized into three groups: correct triage, over-triage, and under-triage. In the third phase of our study, the questions were categorized according to the correct triage responses.

RESULTS

In the results of our study, a statistically significant difference was found between the three groups in terms of correct triage, over-triage, and under-triage (p < 0.001). GPT-4 was found to have the highest correct triage rate with an average of 70.60 (±3.74), while Gemini had the highest over-triage rate with an average of 35.2 (±2.93) (p < 0.001). The highest under-triage rate was observed in emergency medicine specialists (32.90 (±11.83)). In the ESI 1-2 class, Gemini had a correct triage rate of 87.77%, GPT-4 had 85.11%, and emergency medicine specialists had 49.33%.

CONCLUSION

In conclusion, our study shows that both GPT-4 and Gemini can accurately triage critical and urgent patients in ESI 1&2 groups at a high rate. Furthermore, GPT-4 has been more successful in ESI triage for all patients. These results suggest that GPT-4 and Gemini could assist in accurate ESI triage of patients in emergency departments.

Emergency department triage decision-making by registered nurses: An instrument development study

AIM

To develop and psychometrically test the triage decision-making instrument, a tool to measure Emergency Department Registered Nurses decision-making.

DESIGN

Five phases: (1) defining the concept, (2) item generation, (3) face validity, (4) content validity and (5) pilot testing.

METHODS

Concept definition informed by a grounded theory study from which four domains emerged. Items relevant to the four domains were generated and revised. Face validity was established using three focus groups. The target population upon which the reliability and validity of the triage decision-making instrument was explored were triage registered nurses in emergency departments. Three expert judges assessed 89 items for content and domain designation using a 4-point scale. Psychometric properties were assessed by exploratory factor analysis, following which the names of the four domains were modified.

RESULTS

The triage decision-making instrument is a 22-item tool with four factors: clinical judgement, managing acuity, professional collaboration and creating space. Focus group data indicated support for the domains. Expert review resulted in 46 items with 100% agreement and 13 with 66% agreement. Fifty-nine items were distributed to a convenience sample of 204 triage nurses from six hospitals in 2019. The Kaiser-Meyer-Olkin measures indicated that the data were sufficient for exploratory factor analysis. Bartlett's test indicated patterned relationships among the items (X2 (231) = 1156.69). An eigenvalue of >1.0 was used and four factors explained 48.64% of the variance. All factor loadings were ≥0.40. Internal consistency was demonstrated by Cronbach's alphas of .596 factor 1, .690 factor 2, .749 factor 3 and .822 for factor 4.

CONCLUSION

The triage decision-making instrument meets the criteria for face validity, content validity and internal consistency. It is suitable for further testing and refinement.

IMPACT

The instrument is a first step in quantifying triage decision-making in real-world clinical environments. The triage decision-making instrument can be used for targeted triage interventions aimed at improving throughput and staff education.

STATISTICAL SUPPORT

Dr. Tak Fung who is a member of the research team is a statistician.

STATISTICAL METHODS

Development, validation and assessment of instruments/scales. Descriptive statistics.

REPORTING METHOD

STROBE cross-sectional checklist.

IMPLICATIONS FOR THE PROFESSION AND/OR PATIENT CARE

The TDI makes the complexity of triage decision-making visible. Identifying the influence of decision-making factors in addition to acuity that affect triage decisions will enable nurse managers and educators to develop targeted interventions and staff development initiatives. By extension, this will enhance patient care and safety.

Triage-clinical reasoning on emergency nursing competency: a multiple linear mediation effect

BACKGROUND

Triage is the first step in providing prompt and appropriate emergency nursing and addressing diagnostic issues. Rapid clinical reasoning skills of emergency nurses are essential for prompt decision-making and emergency care. Nurses experience limitations in emergency nursing that begin with triage. This cross-sectional study explored the mediating effect of perceived triage competency and clinical reasoning skills on the association between Korean Triage and Acuity Scale (KTAS) proficiency and emergency nursing competency.

METHODS

A web-based survey was conducted with 157 emergency nurses working in 20 hospitals in South Korea between mid-May and mid-July 2022. Data were collected utilizing self-administered questionnaires to measure KTAS proficiency (48 tasks), perceived triage competency (30 items), clinical reasoning skills (26 items), and emergency nursing competency (78 items). Data were analyzed using the PROCESS macro (Model 6).

RESULTS

Perceived triage competency indirectly mediate the relationship between KTAS proficiency and emergency nursing competency. Perceived triage competency and clinical reasoning skills were significant predictors of emergency nursing competency with a multiple linear mediating effect. The model was found have a good fit (F = 8.990, P <.001) with, a statistical power of 15.0% (R² = 0.150).

CONCLUSIONS

This study indicates that improving emergency nursing competency requires enhancing triage proficiency as well as perceived triage competency, which should be followed by developing clinical reasoning skills, starting with triage of emergency nurses.

Incidence of overtriage and undertriage and associated factors: A cross-sectional study using a secondary data analysis

BACKGROUND

Improving triage accuracy for accurate patient identification and appropriate resource allocation is essential. Little is known about the trend of triage accuracy, and factors associated with mistriage vary from study to study.

AIM

To identify incidence and risk factors of mistriage, such as overtriage and undertriage.

DESIGN

This is a cross-sectional study.

METHODS

The data came from the National Emergency Department Information System database in 2016-2020. All patients 15 years and older visiting emergency departments in Korea were assessed for eligibility, and 20,641,411 emergency patients' data were used. Multivariable logistic regressions were conducted to confirm the associated factors with overtriage and undertriage compared to expected triage. Demographic characteristics, disease-related signs and triage-related factors were independent variables.

RESULTS

Expected triage decreased from 96.8% in 2016 to 95.7% in 2020. Overtriage (0.5%-0.7%) and undertriage (2.4%-3.3%) increased. The occupation that performed triage the most (over 85%) was nurses. Associated factors with overtriage were demographic characteristics (40-64 age group, female), disease-related signs (known disease, direct visit) and triage-related factors (regional emergency medical centre). Risk factors to undertriage were disease-related signs (systolic/diastolic blood pressure and pulse rates within normal range).

CONCLUSIONS

While the acuity degree remained within the recommended range, the accuracy of triage decreased, and there was a gradual increase in mistriaged cases. Nurses have performed most of the triage and played a key role in expected triage. Associated factors with overtriage were demographic characteristics, disease-related signs and triage-related factors and risk factors to undertriage were disease-related signs.

PATIENT OR PUBLIC CONTRIBUTION

No patient or public contribution.

IMPLICATIONS FOR THE PROFESSION

Nurses should be aware of what factors are associated with mistriage and why the factors cause mistriage to improve the triage accuracy because they are responsible for the majority of the triage assessments.

Understanding triage assessment of acuity by emergency nurses at initial adult patient presentation: A qualitative systematic review

BACKGROUND

Nurses make complex triage decisions within emergency departments, which significantly affect patient outcomes. Understanding how nurses make these decisions and why they deviate from triage algorithms facilitates interventions that work with their decision-making processes, increasing acceptability and effectiveness.

AIMS

This qualitative systematic review aimed to understand decision-making processes emergency nurses use to make acuity decisions during triage assessment at initial patient presentation.

METHODOLOGY

Medline, CINAHL and Academic Search Complete were systematically searched to 15th December 2022. Data were analysed using thematic synthesis. Established themes were reviewed with GRADE-CERQual to evaluate certainty of evidence.

RESULTS

28 studies were included in the review. Data analysis uncovered three superordinate themes of holistic reasoning, situational awareness, and informed decision-making. The findings show nurses value holistic assessments over algorithms and rely on knowledge and experience. They also assess the wider situation in the emergency department.

CONCLUSIONS

This review presents new perspectives on nurses' decision-making processes about patient's acuity. Nurses holistically gather information about patients before translating that information into acuity scores. These actions are informed by their knowledge and experience; however, the wider situation also impacts their decisions. In turn, the nurses use interpretations of patients' acuity to control the wider situation.

Factors contributing to patient safety during triage process in the emergency department: A systematic review

BACKGROUND

Triage is a dynamic environment in which large numbers of people can present. It presents a vulnerable assessment point, as a triage nurse must assess a patient's urgency level and analyse their health status and expected resource needs. Given the critical nature of triage, it is necessary to understand the factors contributing to patient safety.

OBJECTIVES

To identify and examine the factors contributing to patient safety during the triage process.

METHODS

A systematic review of the literature was undertaken, and a thematic analysis of the factors contributing to patient safety during the triage process. PubMed, CINAHL, Web of Sciences, Science Direct, SAGE, EMBASE and reference lists of relevant studies published in English until March 2022 were searched for relevant studies. The search protocol has been registered at the PROSPERO (CRD42019146616), and the review was conducted using the PRISMA criteria.

RESULTS

Out of 5366 records, we included 11 papers for thematic synthesis. Identified factors contributing to patient safety in triage are related to the emergency's work environment, such as patient assessment, high workload, frequent interruptions and staffing, and personal factors such as nurse traits, experience, knowledge, triage fatigue and work schedule.

CONCLUSIONS

This review shows that patient safety is influenced by the attitude, capabilities and experiences of triage nurses, the time when nurses can dedicate themselves to the patient and triage the patient without disruption. It is necessary to raise awareness among nursing administrators and healthcare professionals to provide a safe triage environment for patients.

RELEVANCE TO CLINICAL PRACTICE

This review highlights the evidence on the factors contributing to patient safety in the triage process. Further research is needed for this cohort of triage nurses in the emergency department concerning ensuring patient safety.

PATIENT OR PUBLIC CONTRIBUTION

No patient or public contribution was required to design or undertake this review.

Missed nursing care in emergency departments: A scoping review

BACKGROUND

Patient safety is a global health priority. Errors of omission, such as missed nursing care in hospitals, are frequent and may lead to adverse events. Emergency departments (ED) are especially vulnerable to patient safety errors, and the significance missed nursing care has in this context is not as well known as in other contexts.

AIM

The aim of this scoping review was to summarize and disseminate research about missed nursing care in the context of EDs.

METHOD

A scoping review following the framework suggested by Arksey and O'Malley was used to (1) identify the research question; (2) identify relevant studies; (3) select studies; (4) chart the data; (5) collate, summarize, and report the results; and (6) consultation.

RESULTS

In total, 20 themes were derived from the 55 included studies. Missed or delayed assessments or other fundamental care were examples of missed nursing care characteristics. EDs not staffed or dimensioned in relation to the patient load were identified as a cause of missed nursing care in most included studies. Clinical deteriorations and medication errors were described in the included studies in relation to patient safety and quality of care deficiencies. Registered nurses also expressed that missed nursing care was undignified and unsafe.

CONCLUSION

The findings from this scoping review indicate that patients' fundamental needs are not met in the ED, mainly because of the patient load and how the ED is designed. According to registered nurses, missed nursing care is perceived as undignified and unsafe.

The first positive evidence that training improves triage decisions in Greece: evidence from emergency nurses at an Academic Tertiary Care Emergency Department

BACKGROUND

Triage refers to the process of patient prioritisation in the emergency department (ED). This is based on the severity of the patient's illness and is performed by emergency nurses (ENs). This has a pivotal role in ensuring patient safety and in ensuring that the ED operates smoothly - so continuous and accurate training are essential. As Emergency Nursing has been formally established in Greece since 2019, it is of the uppermost importance that all Greek ENs should be trained in the use of a standardised triage system. The present study aimed to evaluate the effect of triage training of ENs in the use of the Swiss Triage System (STS) after an intervention of one week.

METHODS

The effect of triage training was studied experimentally by comparing performance before and one week after training. A sample of thirty-six ENs from the University Department of Emergency Medicine at AHEPA University Hospital took part. The role of training in triage by the STS was assessed by completing the same self-administered questionnaire before and after a 45-minute e-learning program (presentation video of STS but with simulation scenarios) which was available during the period of a week. The post-training test was taken 2 weeks later, after the training process.

RESULTS

The most promising finding was that there was a significant improvement in the number of correct answers after the training in triage (p<0.001). A significant improvement was also detected (p<0.001) in the questions that tested vigilance in providing safe health services by ENs, whereas there was no significant association between the number of correct answers and years of emergency experience or level of education, - either before or after the intervention.

CONCLUSIONS

Triage training seems to successfully improve effective and efficient triage. To the best of our knowledge, this is the first study that has demonstrated that triage training has a significant positive impact on triage performance by ENs in Greece. It is planned to support these findings by real time studies in an ED.

Strategies to improve the quality of nurse triage in emergency departments: A realist review protocol

AIM

The purpose of this realist review was to assess what works, for whom and in what context, regarding strategies that influence nurses' behaviour to improve triage quality in emergency departments (ED).

DESIGN

Realist review protocol.

METHODS

This protocol follows the PRISMA-P statement and will include any type of study on strategies to improve the triage process in the ED (using recognized and validated triage scales). The included studies were examined for scientific quality using the Mixed Methods Appraisal Tool. The framework for this realist review is based on the Behaviour Change Wheel (BCW) and the context-mechanism-outcome (CMO) models.

DISCUSSION

Nurses and ED decision makers will be informed on the evidence regarding strategies to improve the quality of triage and the factors required to maximize their effectiveness. Research gaps may also be identified to guide future research projects on the adoption of best practices in ED nursing triage.

Association between triage nurses' job satisfaction and professional capability: Results of a mixed-method study

AIM

This study aims to examine factors related to the job satisfaction of triaging nurses and their professional capability in the clinical setting.

BACKGROUND

Triage is a complex process that relies on making decisions in favour of the patient and his treatment. The professional capability of a triaging nurse is an important psychological construct of job satisfaction.

METHODS

The study used a mixed-method methodology, with data collection based on an explanatory research design. The research instrument in the quantitative part was a survey questionnaire, and in the qualitative part, a semi-structured interview. The results were integrated using the 'Pillar Integration Process'.

RESULTS

There are significant relationships between professional capability and job satisfaction. Six main topics were exposed: characteristics and traits, work organization, safety is the key, burdening circumstances, capability and self-evaluation.

CONCLUSION

Professional capability is associated with job satisfaction. The necessary managerial changes should be made to achieve job satisfaction and develop professional competence while focusing on already trained and competent triage nurses, as satisfied triage nurses will stay longer in the institution.

IMPLICATIONS FOR NURSING MANAGEMENT

The manager's job is to be aware of the level of job satisfaction, take care to develop their employee's professional capability and take action in case of disrupted balance.

Developing and Validating an Emergency Triage Model Using Machine Learning Algorithms with Medical Big Data

Objective

To establish an emergency triage model through the statistical analysis of big data during a particular time period from a hospital information system to improve the accuracy of triage in emergency department (ED).

Methods

A total of 276,164 patients who visited the Emergency Medicine Department of Beijing Chao-Yang Hospital from 2017 to 2020 were included in this study, including 123,392 men and 152,772 women aged from 14 to 112 years. The baseline characteristics (age and gender) and medical records (patient's condition, body temperature, heart rate, breathing, blood pressure, consciousness, and oxygen saturation) of the patients was collected. The data samples were randomly allocated, with 80% as the training set and 20% as the testing set. The patients were divided into levels I, II, III, and IV in accordance with a four-level triage standard. We selected the effective Extreme Gradient Boosting (XGBoost) algorithm as our emergency classification prediction model. The XGBoost model was applied to simulate the thinking process of triage nurses, and the De Long's test was used to compare the receiver operating characteristic (ROC) curve of different models. The P value was obtained by calculating the variance and covariance of area under the curve (AUC) values of different ROC curves.

Results

Level I had 4960 (1.8%) patients, level II had 25,646 (9.29%), level III had 130,664 (47.31%), and level IV had 114,894 (41.6%). The XGBoost model was built following a logic exercise based on the traditional manual pre-inspection and triage results. After verification, the prediction accuracy was 82.57%. The AUC of each disease severity level (levels I, II, III, and IV) was 0.9629, 0.9554, 0.9120, and 0.9296, respectively.

Conclusion

The emergency triage prediction model, which achieved a relatively strong accuracy rate, can reduce the work intensity of medical workers and improve their working efficiency.

Machine learning methods applied to triage in emergency services: A systematic review

BACKGROUND

In emergency services is important to accurately assess and classify symptoms, which may be improved with the help of technology. One mechanism that could help and improve predictions from health records or patient flow is machine learning (ML).

AIM

To analyse the effectiveness of ML systems in triage for making predictions at the emergency department in comparison with other triage scales/scores.

METHODS

Following the PRISMA recommendations, a systematic review was conducted using CINAHL, Cochrane, Cuiden, Medline and Scopus databases with the search equation "Machine learning AND triage AND emergency".

RESULTS

Eleven studies were identified. The studies show that the use of ML methods consistently predict important outcomes like mortality, critical care outcomes and admission, and the need for hospitalization in comparison with scales like Emergency Severity Index or others. Among the ML models considered, XGBoost and Deep Neural Networks obtained the highest levels of prediction accuracy, while Logistic Regression performed obtained the worst values.

CONCLUSIONS

Machine learning methods can be a good instrument for helping triage process with the prediction of important emergency variables like mortality or the need for critical care or hospitalization.

Facilitators and Barriers of the Triage Process based on Emergency Nurses' Experience with the Korean Triage and Acuity Scale (KTAS): A Qualitative Content Analysis

PURPOSE

Since 2016, the Korean Triage and Acuity Scale (KTAS) algorithm has been applied to the triage process in the emergency departments (EDs) of Korea. This study aimed to investigate the facilitators of and barriers to a well-run triage function based on how Korean emergency nurses perceived the triage process and their experiences with it.

METHODS

Data were collected using focus group interviews from June 2018 to January 2019. Twenty emergency nurses were divided into 2 junior and 4 senior groups based on their level of clinical experience. All interviews were recorded as they were spoken and transcribed. Data were analyzed using qualitative content analysis.

RESULTS

The participants recognized the need for the KTAS algorithm to efficiently classify emergency patients, and were working on it properly. According to the data, we extracted 4 themes and 20 sub-themes. Four themes were as follows; (1) awareness about the necessity of triage, (2) facilitators to triage process, (3) barriers to triage process, and (4) suggestions for the establishment and development of triage.

CONCLUSIONS

From the findings of this study, various vulnerabilities of the triage process were identified, and solutions were suggested from the emergency nurses' perspective. Educational, staffing, financial support, and periodic updates of the KTAS are needed to promote the triage process in the future.

Establishing Research Priorities for the Emergency Severity Index Using a Modified Delphi Approach

INTRODUCTION

The accuracy of an initial ED triage decision has been reported to drive the clinical trajectory for ED patients, and, therefore, this assessment is critical to patient safety. The Emergency Severity Index-a 5-point score assigned by a triage nurse and based on disease acuity, patient potential for decompensation, and anticipated resource use-is used both in the United States and internationally. In the US, the Emergency Severity Index is used by up to 94% of the academic medical center emergency departments. In 2020, the Emergency Nurses Association acquired the intellectual property rights to the Emergency Severity Index and is responsible for its maintenance and improvement.

OBJECTIVE

The purpose of this study was to establish a research agenda for the improvement of individual and institutional understanding and use of the Emergency Severity Index.

METHODS

Modified Delphi process was used with 3 rounds of data collection.

RESULTS

Round 1 yielded 112 issues, which were collapsed into 18 potential research questions in 4 general categories: education and training (6 questions), workplace environment (3 questions), emergency care services (7 questions), and special populations (2 questions). These questions were used in round 2 to establish importance. Round 3 yielded a rank ordering of both categories and research questions.

DISCUSSION

The research priorities as set through the use of this modified Delphi process align well with current gaps in the literature. Research in these areas should be encouraged to improve the understanding of educational, environmental, and process challenges to emergency nurses' triage decisions and accuracy of Emergency Severity Index assignments.

An updated national survey of triage and triage related work in Sweden: a cross-sectional descriptive and comparative study

BACKGROUND

Triage and triage related work has been performed in Swedish Emergency Departments (EDs) since the mid-1990s. According to two national surveys from 2005 to 2011, triage was carried out with different triage scales and without guidelines or formal education. Furthermore, a review from 2010 questioned the scientific evidence for both triage as a method as well as the Swedish five level triage scale Medical Emergency Triage and Treatment System (METTS); nevertheless, METTS was applied in 65% of the EDs in 2011. Subsequently, METTS was renamed to Rapid Emergency Triage and Treatment System (RETTS©). The hypothesis for this study is that the method of triage is still applied nationally and that the use of METTS/RETTS© has increased. Hence, the aim is to describe the occurrence and application of triage and triage related work at Swedish Emergency Departments, in comparison with previous national surveys.

METHODS

In this cross-sectional study with a descriptive and comparative design, an electronic questionnaire was developed, based on questionnaire from previous studies. The survey was distributed to all hospital affiliated EDs from late March to the middle of July in 2019. The data was analysed with descriptive statistics, by IBM SPSS Statistics, version 26.

RESULTS

Of the 51 (75%) EDs partaking in the study, all (100%) applied triage, and 92% used the Swedish triage scale RETTS©. Even so, there was low concordance in how RETTS© was applied regarding time frames i.e., how long a patient in respective triage level could wait for assessment by a physician. Additionally, the results show a major diversion in how the EDs performed education in triage.

CONCLUSION

This study confirms that triage method is nationally implemented across Swedish EDs. RETTS© is the dominating triage scale but cannot be considered as one triage scale due to the variation with regard to time frames per triage level. Further, a diversion in introduction and education in the pivotal role of triage has been shown. This can be counteracted by national guidelines in what triage scale to use and how to perform triage education.

Influence of electronic triage decision-support on hospital admission, left without being seen and time to physician initial assessment in the emergency department

OBJECTIVE

To explore the impact of the implementation of eCTAS, a real-time electronic decision-support tool, on hospital admission, rate of left without being seen, and time from triage to physician initial assessment.

METHODS

We conducted a cohort study using population-based administrative data from all Ontario emergency departments (EDs) that had implemented eCTAS for 9 months. We compared 6 months post-eCTAS data to the same 6 months the previous year (pre-eCTAS). We included triage encounters of adult (≥ 18 years) patients if they had one of 16 pre-specified, high-volume presenting complaints. Multivariable logistic regression and quantile regression models informed the effect of eCTAS on outcomes.

RESULTS

We included data from 354,176 triage encounters from 31 EDs. There was a change in the distribution of triage scores post-eCTAS, with fewer patients classified as CTAS 2 and CTAS 3, and more patients classified as CTAS 1 and CTAS 4. Overall, hospital admission decreased post-eCTAS (adjusted OR: 0.98; 95% CI: 0.97 to 1.00), with fewer CTAS 2 and more CTAS 3 and CTAS 4 patients admitted post-eCTAS. The rate of left without being seen increased (2.8% vs. 3.0%; adjusted OR: 1.07; 95% CI: 1.03 to 1.11) post-eCTAS, while time to physician initial assessment proved similar pre and post-eCTAS.

CONCLUSIONS

eCTAS implementation had little impact on admission, rate of left without being seen and time to physician initial assessment. eCTAS appears to reclassify patients from higher to lower acuity scores, resulting in higher admission rates for CTAS 3 and CTAS 4 patients. It remains unknown if this reclassification is appropriate.

The challenge of interprofessional collaboration in emergency department team triage - An interpretive description

AIMS

To explore emergency nurses' and physicians' experience of collaboration and collective decision-making when triaging older Emergency Department patients within the interprofessional team triage system.

DESIGN

Qualitative.

METHODS

Semi-structured interviews were conducted with seven nurses and five physicians. Transcripts were analysed via Interpretive Description between September 2016-May 2017.

RESULTS

'Negotiating collaboration' was developed as the main theme. Three subthemes influenced the negotiation process: Participants described divergent opinions on how an optimal triage system should work ('preferences for triage systems'); they had conflicting perceptions of each profession's role ('role perceptions'); and they expressed different coping strategies regarding 'perceived time pressure'. The compatibility of participants' views on these sub-themes determined whether the nurse and physician were able to successfully negotiate their collaboration. These themes became more evident when the team triaged older ED patients.

CONCLUSION

Improving interprofessional team triage requires working with the involved nurses' and physicians' values and beliefs. The strengths of both professions need to be considered and a flexible approach to collaboration established according to the patients' situations.

IMPACT

Emergency Department leaders need to consider nurses' and physicians' values and beliefs to promote interprofessional collaboration in team triage.

Vulnerability: A concept synthesis and its application to the Emergency Department

AIM

The aim of this concept synthesis was to add clarity to the concept of vulnerability with application to the Emergency Department (ED) by critiquing, analysing and amalgamating published concept analyses.

BACKGROUND

The concept of vulnerability has been used widely, however it has various meanings. A clearer understanding of vulnerability and application to the ED may help healthcare professionals provide high quality care responsive to the needs of vulnerable individuals.

METHOD

Nine concept analyses of vulnerability were retrieved using Medline, CINAHL, and PsycINFO databases. After extracting data on each analysis, Walker and Avant's concept synthesis method was used to structure this synthesis, with a thematic synthesis approach used in the analysis.

FINDINGS

Four themes associated with vulnerability emerged from the synthesis. The first theme, vulnerability as a journey, reflected elements within an individual's life that perpetuate and exacerbate vulnerability. The second theme, vulnerability as susceptibility and risk, highlighted intrinsic and extrinsic elements that contribute to a state of risk. The third theme, positive and negative repercussions, emphasised lessons that can be learned from experience, with the fourth theme of a shared understanding indicating the importance of understanding the concept of vulnerability for patient care.

CONCLUSION

Findings from this synthesis highlight the multiple elements associated with a vulnerable state, evident in the context of the ED. With multiple ED-specific elements contributing to vulnerability, clarity of the term is important to inform ED-specific interventions designed to meet the needs of vulnerable populations.

Effects of environmental distractors on nurse emergency triage accuracy: a pilot study protocol

Background

The clinical decisions of emergency department triage nurses need to be of the highest accuracy. However, studies have found repeatedly that these nurses over- or underestimate the severity of patient health conditions. This has major consequences for patient safety and patient flow management. Workplace distractors such as noise and task interruptions have been pointed to as factors that might explain this inaccuracy. The use of a serious game reproducing the work environment during triage affords the opportunity to explore the impact of these distractors on nurse emergency triage accuracy, in a safe setting.

Methods/design

A pilot study with a factorial design will be carried out to test the acceptability and feasibility of a serious game developed specifically to simulate the triage process in emergency departments and to explore the primary effects of distractors on nurse emergency triage accuracy. Eighty emergency nurses will be randomized into four groups: three groups exposed to different distractors (A, noise; B, task interruptions; C, noise and task interruptions) and one control group. All nurses will have to complete 20 clinical vignettes within 2 h. For each vignette, a gold standard assessment will be determined by experts. Pre-tests will be conducted with clinicians and certified emergency nurses to evaluate the appeal of the serious game.

Discussion

Study results will inform the design of large-scale investigations and will help identify teaching, training, and research areas that require further development.

Consistency of triage scores by presenting complaint pre- and post-implementation of a real-time electronic triage decision support tool

Objective

eCTAS is a real‐time electronic decision‐support tool designed to standardize the application of the Canadian Triage and Acuity Scale (CTAS). This study addresses the variability of CTAS score distributions across institutions pre‐ and post‐eCTAS implementation.

Methods

We used population‐based administrative data from 2016–2018 from all emergency departments (EDs) that had implemented eCTAS for 9 months. Following a 3‐month stabilization period, we compared 6 months post‐eCTAS data to the same 6 months the previous year (pre‐eCTAS). We included triage encounters of adult (≥17 years) patients who presented with 1 of 16 pre‐specified, high‐volume complaints. For each ED, consistency was calculated as the absolute difference in CTAS distribution compared to the average of all included EDs for each presenting complaint. Pre‐eCTAS and post‐eCTAS change scores were compared using a paired‐samples t‐test. We also assessed if eCTAS modifiers were associated with triage consistency.

Results

There were 363,214 (183,231 pre‐eCTAS, 179,983 post‐eCTAS) triage encounters included from 35 EDs. Triage scores were more consistent (P < 0.05) post‐eCTAS for 6 (37.5%) presenting complaints: chest pain (cardiac features), extremity weakness/symptoms of cerebrovascular accident, fever, shortness of breath, syncope, and hyperglycemia. Triage consistency was similar pre‐ and post‐eCTAS for altered level of consciousness, anxiety/situational crisis, confusion, depression/suicidal/deliberate self‐harm, general weakness, head injury, palpitations, seizure, substance misuse/intoxication, and vertigo. Use of eCTAS modifiers was associated with increased triage consistency.

Conclusions

eCTAS increased triage consistency across many, but not all, high‐volume presenting complaints. Modifier use was associated with increased triage consistency, particularly for non‐specific complaints such as fever and general weakness.

Exploring the complex interactions of baseline patient factors to improve nursing triage of acute coronary syndrome

Emergency department (ED) nurses need to identify patients with potential acute coronary syndrome (ACS) rapidly because treatment delay could impact patient outcomes. Aims of this secondary analysis were to identify key patient factors that could be available at initial ED nurse triage that predict ACS. Consecutive patients with chest pain who called 9-1-1, received a 12-lead electrocardiogram in the prehospital setting, and were transported via emergency medical service were included in the study. A total of 750 patients were recruited. The sample had an average age of 59 years old, was 57% male, and 40% Black. One hundred and fifteen patients were diagnosed with ACS. Older age, non-Caucasian race, and faster respiratory rate were independent predictors of ACS. There was an interaction between heart rate by Type II diabetes receiving insulin in the context of ACS. Type II diabetics requiring insulin for better glycemic control manifested a faster heart rate. By identifying patient factors at ED nurse triage that could be predictive of ACS, accuracy rates of triage may improve, thus impacting patient outcomes.

Risk of mortality and cardiopulmonary arrest in critical patients presenting to the emergency department using machine learning and natural language processing

Emergency department triage is the first point in time when a patient's acuity level is determined. The time to assign a priority at triage is short and it is vital to accurately stratify patients at this stage, since under-triage can lead to increased morbidity, mortality and costs. Our aim was to present a model that can assist healthcare professionals in triage decision making, namely in the stratification of patients through the risk prediction of a composite critical outcome-mortality and cardiopulmonary arrest. Our study cohort consisted of 235826 adult patients triaged at a Portuguese Emergency Department from 2012 to 2016. Patients were assigned to emergent, very urgent or urgent priorities of the Manchester Triage System (MTS). Demographics, clinical variables routinely collected at triage and the patients' chief complaint were used. Logistic regression, random forests and extreme gradient boosting were developed using all available variables. The term frequency-inverse document frequency (TF-IDF) natural language processing weighting factor was applied to vectorize the chief complaint. Stratified random sampling was used to split the data into train (70%) and test (30%) data sets. Ten-fold cross validation was performed in train to optimize model hyper-parameters. The performance obtained with the best model was compared against the reference model-a regularized logistic regression trained using only triage priorities. Extreme gradient boosting exhibited good calibration properties and yielded areas under the receiver operating characteristic and precision-recall curves of 0.96 (95% CI 0.95-0.97) and 0.31 (95% CI 0.26-0.36), respectively. The predictors ranked with higher importance by this model were the Glasgow coma score, the patients' age, pulse oximetry and arrival mode. Compared to the reference, the extreme gradient boosting model using clinical variables and the chief complaint presented higher recall for patients assigned MTS-3 and can identify those who are at risk of the composite outcome.

Experiences of nurses working in a triage area: An integrative review

OBJECTIVES

The aim of the study is to review the literature on the experiences of nurses working in triage within emergency departments in hospitals.

REVIEW METHOD

This is an integrative review based on Cooper's five-stage framework.

DATA SOURCES

Primary research articles published from January 2008 to January 2018 were identified from seven databases: PubMed, Embase, Cumulative Index to Nursing and Allied Health Literature, PsycINFO, Cochrane, ProQuest, and Scopus. A manual search of the end references from the published studies was also conducted to ensure a comprehensive search.

REVIEW METHOD

The included studies were evaluated independently by two authors based on the Joanna Briggs Institute appraisal checklist for Interpretive and Critical Research and Descriptive/Case Series to ensure methodological rigour and validity of the review.

RESULTS

The literature review included 35 articles; of which, 18 were qualitative studies, 13 were quantitative studies, and four were mixed-method studies. The findings from this review were categorised into three themes: (1) making judgement, (2) service delivery, and (3) effective communication.

CONCLUSIONS

The results from the review suggest that triage is a complex process involving decision-making and interprofessional communication. Nurses need to make judgement, consider factors affecting service delivery, and engage in effective communication, which is central to triaging. Decision-making is contextual and requires judicious judgement. Effective communication with fellow healthcare personnel and with patients and their caregivers is paramount to optimise care delivery. Finally, more support is needed to empower the nurses to cope with the work involved in triaging.

Triage emergency nurse decision-making: Incidental findings from a focus group study

INTRODUCTION

Emergency Department (ED) triage decision-making is a complex process. Triage scales are used to determine patient acuity level, however, they do not provide information on how nurses make triage decisions. This focus group study was part of a larger project to develop a psychometrically sound instrument to measure triage decision-making by nurses; the Triage Decision-Making Instrument. We report important incidental findings that emerged during the study that the participants identified as factors impeding triage decision-making.

METHODS

Three focus groups were conducted with 11 triage Registered Nurses. They commented on the instrument items and identified factors that influence triage decision-making. Transcripts were analyzed using thematic analysis.

RESULTS

Three powerful inter-related themes emerged central to encumbering triage decision-making: competing systems (pre-hospital versus the ED), fluctuating patient volume resulting in 'intra-Canadian Triage and Acuity Scale' triaging, and personal capacity including experience and 'triage fatigue'.

DISCUSSION

The findings demonstrate how interrelated system factors impede nurses triage decision-making. Triage nurses require support in their role and initiatives are needed to reduce the pressure they feel in relation to resolving system issues. Larger system issues and the capacity of the individual decision-makers must be accounted for within the context of increasing effectiveness and safety of ED triage.

Trauma Team Activation: Accuracy of Triage When Minutes Count: A Synthesis of Literature and Performance Improvement Process

Accuracy and timeliness of trauma activations are vital to patient safety. The American College of Surgeons mandates the trauma surgeon's presence within 15 min of the patient's arrival to the emergency department (ED) 80% of the time. In 2015, at this Level II Pediatric Trauma Center, average mean activation times were approximately 16 min and activation accuracy (over- and undertriage) affected 27% of the trauma patient activations. This evidence-based quality improvement project set out to determine the most efficient method of Emergency Medical Services (EMS) intake. Communication Center (Com. Center) recordings were carefully reviewed to identify time when EMS notifies the Com. Center and actual time of trauma activation page. A timeline was formulated with assessment of time to activation and patient triage accuracy. An educational curriculum was developed as an intervention for the Com. Center staff. Education included a decision tree for trauma activations and the development of templates for our electronic health record and prompts to improve accurate activations. After additional focus groups analyzed present ED performance and the industry standard, a policy requiring only paramedic-trained staff was put in place. After implementation of the aforementioned intervention, the Com. Center performance revealed reduction in incorrect activations from 27.3% to 10.7% from 2015 to 2016. Mean activation time in January 2015 was 48.5 min before the intervention and 4.71 min postintervention in December 2016; this is a staggering reduction in activation times of 90%!

Caring for an older person with dementia in the Emergency Department (ED): An Appreciative Inquiry exploring family member and ED nurse experiences

AIMS AND OBJECTIVES

To generate insights about what matters and is valued by family members of older people with dementia in the emergency department. To explore the experiences of emergency nurses looking after older people with dementia in an episode of care.

BACKGROUND

In the emergency department, older people with dementia are at risk of suboptimal care. Little is known of the experiences of family members of being with an older person with dementia in the emergency department or the experiences of emergency nurses looking after older people with dementia in this environment.

DESIGN AND METHODS

Phase 1 Data Analysis of the Discovery Phase of an Appreciative Inquiry study. Study participants were family members of older people with dementia and emergency nurses. Data collection methods included interviews with family members of older people with dementia and 30 hr of participant observation working alongside emergency nurses. This study was guided by the Standards for Reporting Qualitative Research.

RESULTS

Two themes emerged from the analysis: What matters to family members with four subthemes and challenges for family members and nurses in the emergency department with two subthemes.

CONCLUSION

This study demonstrates that some emergency nurses are connecting with family members even in the briefest of clinical encounters. It is feasible for more emergency nurses to do the same more of the time.

RELEVANCE TO CLINICAL PRACTICE

The older person with dementia must be given a triage category of no less than 3 (to be seen by the doctor within the hour) on arrival in the department. Further education is needed to assist emergency nurses to establish rapport and incorporate family member insights as part of care planning and assessment of the needs of the older person with dementia.

Machine Learning-Based Prediction of Clinical Outcomes for Children During Emergency Department Triage

IMPORTANCE

While machine learning approaches may enhance prediction ability, little is known about their utility in emergency department (ED) triage.

OBJECTIVES

To examine the performance of machine learning approaches to predict clinical outcomes and disposition in children in the ED and to compare their performance with conventional triage approaches.

DESIGN, SETTING, AND PARTICIPANTS

Prognostic study of ED data from the National Hospital Ambulatory Medical Care Survey from January 1, 2007, through December 31, 2015. A nationally representative sample of 52 037 children aged 18 years or younger who presented to the ED were included. Data analysis was performed in August 2018.

MAIN OUTCOMES AND MEASURES

The outcomes were critical care (admission to an intensive care unit and/or in-hospital death) and hospitalization (direct hospital admission or transfer). In the training set (70% random sample), using routinely available triage data as predictors (eg, demographic characteristics and vital signs), we derived 4 machine learning-based models: lasso regression, random forest, gradient-boosted decision tree, and deep neural network. In the test set (the remaining 30% of the sample), we measured the models' prediction performance by computing C statistics, prospective prediction results, and decision curves. These machine learning models were built for each outcome and compared with the reference model using the conventional triage classification information.

RESULTS

Of 52 037 eligible ED visits by children (median [interquartile range] age, 6 [2-14] years; 24 929 [48.0%] female), 163 (0.3%) had the critical care outcome and 2352 (4.5%) had the hospitalization outcome. For the critical care prediction, all machine learning approaches had higher discriminative ability compared with the reference model, although the difference was not statistically significant (eg, C statistics of 0.85 [95% CI, 0.78-0.92] for the deep neural network vs 0.78 [95% CI, 0.71-0.85] for the reference; P = .16), and lower number of undertriaged critically ill children in the conventional triage levels 3 to 5 (urgent to nonurgent). For the hospitalization prediction, all machine learning approaches had significantly higher discrimination ability (eg, C statistic, 0.80 [95% CI, 0.78-0.81] for the deep neural network vs 0.73 [95% CI, 0.71-0.75] for the reference; P < .001) and fewer overtriaged children who did not require inpatient management in the conventional triage levels 1 to 3 (immediate to urgent). The decision curve analysis demonstrated a greater net benefit of machine learning models over ranges of clinical thresholds.

CONCLUSIONS AND RELEVANCE

Machine learning-based triage had better discrimination ability to predict clinical outcomes and disposition, with reduction in undertriaging critically ill children and overtriaging children who are less ill.

Machine learning approaches for predicting disposition of asthma and COPD exacerbations in the ED

OBJECTIVE

The prediction of emergency department (ED) disposition at triage remains challenging. Machine learning approaches may enhance prediction. We compared the performance of several machine learning approaches for predicting two clinical outcomes (critical care and hospitalization) among ED patients with asthma or COPD exacerbation.

METHODS

Using the 2007-2015 National Hospital and Ambulatory Medical Care Survey (NHAMCS) ED data, we identified adults with asthma or COPD exacerbation. In the training set (70% random sample), using routinely-available triage data as predictors (e.g., demographics, arrival mode, vital signs, chief complaint, comorbidities), we derived four machine learning-based models: Lasso regression, random forest, boosting, and deep neural network. In the test set (the remaining 30% of sample), we compared their prediction ability against traditional logistic regression with Emergency Severity Index (ESI, reference model).

RESULTS

Of 3206 eligible ED visits, corresponding to weighted estimates of 13.9 million visits, 4% had critical care outcome and 26% had hospitalization outcome. For the critical care prediction, the best performing approach- boosting - achieved the highest discriminative ability (C-statistics 0.80 vs. 0.68), reclassification improvement (net reclassification improvement [NRI] 53%, P = 0.002), and sensitivity (0.79 vs. 0.53) over the reference model. For the hospitalization prediction, random forest provided the highest discriminative ability (C-statistics 0.83 vs. 0.64) reclassification improvement (NRI 92%, P < 0.001), and sensitivity (0.75 vs. 0.33). Results were generally consistent across the asthma and COPD subgroups.

CONCLUSIONS

Based on nationally-representative ED data, machine learning approaches improved the ability to predict disposition of patients with asthma or COPD exacerbation.

Accuracy of emergency department triage using the Emergency Severity Index and independent predictors of under-triage and over-triage in Brazil: a retrospective cohort analysis

Background

Emergency department (ED) triage is performed to prioritize care for patients with critical and time-sensitive illness. Triage errors create opportunity for increased morbidity and mortality. Here, we sought to measure the frequency of under- and over-triage of patients by nurses using the Emergency Severity Index (ESI) in Brazil and to identify factors independently associated with each.

Methods

This was a single-center retrospective cohort study. The accuracy of initial ESI score assignment was determined by comparison with a score entered at the close of each ED encounter by treating physicians with full knowledge of actual resource utilization, disposition, and acute outcomes. Chi-square analysis was used to validate this surrogate gold standard, via comparison of associations with disposition and clinical outcomes. Independent predictors of under- and over-triage were identified by multivariate logistic regression.

Results

Initial ESI-determined triage score was classified as inaccurate for 16,426 of 96,071 patient encounters. Under-triage was associated with a significantly higher rate of admission and critical outcome, while over-triage was associated with a lower rate of both. A number of factors identifiable at time of presentation including advanced age, bradycardia, tachycardia, hypoxia, hyperthermia, and several specific chief complaints (i.e., neurologic complaints, chest pain, shortness of breath) were identified as independent predictors of under-triage, while other chief complaints (i.e., hypertension and allergic complaints) were independent predictors of over-triage.

Conclusions

Despite rigorous and ongoing training of ESI users, a large number of patients in this cohort were under- or over-triaged. Advanced age, vital sign derangements, and specific chief complaints—all subject to limited guidance by the ESI algorithm—were particularly under-appreciated.

Acolhimento com classificação de risco em unidade de pronto atendimento: estudo avaliativo

RESUMO Objetivo Descrever a avaliação da estrutura, processo e resultado do Acolhimento com Classificação de Risco, na perspectiva dos médicos e enfermeiros de uma Unidade de Pronto Atendimento. Método Estudo avaliativo, descritivo, quantitativo, desenvolvido em Santa Catarina. Dados coletados com instrumento validado e adaptado, constituído por 21 itens distribuídos nas dimensões Estrutura (instalações), Processo (atividades e relações no atendimento) e Resultado (efeitos do atendimento). Na análise, aplicaram-se a estatística descritiva, o cálculo do Ranking Médio e o da Pontuação Média. Resultados A amostra foi de 37 participantes. Dos 21 itens avaliados, 11 (52,4%) tiveram Ranking Médio entre 3 e 4, e nenhum atingiu o máximo (5 pontos). A “Priorização dos casos graves” e o “Atendimento primário por gravidade do caso” obtiveram maior Ranking Médio (4,5), enquanto a “Discussão sobre fluxograma” revelou menor Ranking (2,1). As dimensões Estrutura, Processo e Resultado atingiram, respectivamente, as pontuações médias 23,9, 21,9 e 25,5, indicando avaliação Precária (17,5 a 26,1 pontos). Conclusão Há precarização do Acolhimento com Classificação de Risco, em especial no que se refere ao processo, que obteve menor nível de satisfação dos participantes.

Care delays in patients with signs and symptoms of acute myocardial infarction

AIM

More than six million patients with signs and symptoms suggestive of a heart attack present to emergency departments (EDs) in the US each year. Of those diagnosed with acute myocardial infarction (AMI), one third die in the acute phase. Rapid ED triage can reduce the mortality rate, yet there are still delays in patient care. The aim of this study is to explore the relationship between patient and nurse characteristics, patient presentations, delays in triage, and delays in obtaining electrocardiograms (ECGs) of patients with signs and symptoms of AMI.

METHODS

A retrospective correlational study drawing on data from the records of 286 patients with signs and symptoms of AMI.

RESULTS

Delays in triage are related significantly to patients' gender and race. Most patients were triaged by nurses with associate degrees in nursing, a mean age of 45 and a mean of 18 years' experience. An increase in nurse years of experience predicted greater delay in triage. Delays in obtaining ECGs were also significantly greater with more experienced nurses and when patients reported no chest pain.

CONCLUSION

The study adds to the literature on delays in triage and ECGs in care of patients with possible AMI, and further research of the effects of ED crowding and availability of resources in emergency cardiac care is warranted. Studies should identify the processes that cause delays in the emergency care of patients with signs and symptoms of AMI to ensure timely treatment and care.