Emergency Severity Index versus Manchester Triage System: Utilising Routine Data to Compare the Characteristics, Diagnoses and Admission Status of Patients Triaged in 12 German Emergency Departments

Emergency Severity Index versus Manchester Triage System: Utilising Routine Data to Compare the Characteristics, Diagnoses and Admission Status of Patients Triaged in 12 German Emergency Departments | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Emergency Severity Index versus Manchester Triage System: Utilising Routine Data to Compare the Characteristics, Diagnoses and Admission Status of Patients Triaged in 12 German Emergency Departments David Legg, Yves Noel Wu, Myrto Bolanaki, Antje Fischer-Rosinský, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9171354/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract Background: Although the Manchester Triage System (MTS) and Emergency Severity Index (ESI) have proven valid and reliable methods of assessing the urgency of incoming emergency department (ED) patients, there is a dearth of studies that compare their performance. Thus, the aim of this study was to explore differences in the characteristics, diagnosis and admission status of patients triaged to different categories in German EDs that used either the MTS or ESI. Method: This study is part of the INDEED project, a retrospective multicentre study of statutory health insured adult patients who attended one of 16 participating EDs in 2016. Variables of interest included triage system (ESI and MTS), initial triage category (one to five), patient age and sex, admission status (inpatient/outpatient) and diagnosis information (ICD-10-GM-2017) from hospital information systems. Absolute and relative frequencies were calculated for all variables and separate mixed-effects multivariable regressions were ran to test for the association of ESI and MTS categories and the outcome measure admission status. Results: The final sample included routine data from 354,497 ED visits at 12 participating hospital sites (N = 354,497). Nine hospitals used MTS (cases = 276, 205) and three used ESI (cases = 78,292). Differences were found in the proportion of cases across all triage categories (MTS vs ESI): category one (1.0% vs 1.5%), two (9.8% vs 18.9%), three (38.5% vs 39.5%), four (39.0% vs 24.0%) and five (3.4% vs 4.1%). Mixed-effects multivariable regression analyses indicated that in hospitals using MTS it was 3.0 times more likely to be admitted as an inpatient in the urgent triage categories, one, two and three combined, than in the less-urgent categories four and five combined. In hospitals using ESI, patients were 11.6 times more likely to be admitted as an inpatient in triage categories one, two and three than in categories four and five. Conclusion: Distinct differences in the distribution of triage categories were observed in the analysis of routine data from EDs using either MTS or ESI. Hospitals utilising MTS appeared more likely than those using ESI to triage patients into the less-urgent triage categories and had a higher probability of hospital admissions than ESI hospitals in these less-urgent categories. While the results of this study could have also been affected by differences in the underlying patient populations between hospitals applying MTS or ESI, the findings hint towards systematic differences in the clinical outcomes of patients triaged using either system and raise questions concerning the accuracy of triage categorization. Triage Emergency Department Routine-Data Manchester Triage System Emergency Severity Index Figures Figure 1 Figure 2 Figure 3 Key Findings There are currently two standardized triage system being used in Germany: Emergency Severity Index (ESI) and Manchester Triage System (MTS) When comparing the triage systems MTS and ESI there are substantial differences in triage category distribution in different emergency departments ESI patients are in general categorized more urgent which might pose challenges to treat patients within recommended treatment time in crowded emergency departments With MTS on the other hand also patients who were later admitted to hospital were triaged into lower urgency categories which could be an indicator of undertriage. This was especially the case for patients with older age. A higher proportion of patients with ischemic stroke who were categorized in the least urgent triage category using MTS might reflect a concerning lack of sensitivity for stroke symptoms in MTS. The high proportion of patients in urgent triage categories who were later discharged directly from the ED in both systems might be an indicator of overtriage. Introduction Designed to facilitate critical decision making processes at times when Emergency Departments (EDs) face increased demand ( 1 ), triage systems are fundamental to the provision of modern emergency care ( 2 ). By structuring the allocation of finite resources in a systematic manner, these systems play a pivotal role in ensuring care is provided in order of need ( 3 ), especially at times of ED crowding: the continued negative impacts of which ( 4 , 5 ) have made the ongoing review of triage scales increasingly important ( 6 – 8 ). To date, the majority of the evidence base concerning triage scales has been generated by single centre studies ( 9 ). While these studies have shown individual triage systems to be valid and reliable methods for assessing the urgency of incoming patients’ conditions, few studies have compared their performance against one another and as a result there is increasing demand for comparisons of different triage scales’ performance with regards to clinically important outcomes ( 10 , 11 ). Structured triage systems were first introduced in German EDs in 2004 ( 12 ). Since then the utilisation of these initial assessment tools have become standard practice ( 13 ), with structured and validated triage tools becoming a requirement of emergency care provision in 2018 ( 14 ). That said, there is still considerable variation in the type of triage systems used in German EDs. According to the latest available survey data, collected in 2018, the two most widely used triage systems are the Manchester Triage System (MTS) and Emergency Severity Index (ESI) ( 15 ): the underlying mechanisms of which differ substantially. The German version of the MTS utilises a set of 50 presentation flowcharts and underlying algorithms to prioritise patients based on a rapid assessment of their presenting condition ( 16 ). Patients are allocated to one of the following colour coded triage categories, Red (1, Immediate), Orange (2, Very Urgent), Yellow (3, Urgent), Green (4, Standard) and Blue (5, Non-Urgent). The ESI in contrast utilises a simple three step algorithm to assess the urgency of a patient’s presenting condition and then, for less acute presentations with stable vital signs, expected resource need ( 17 ). Patients in this system are allocated into one of the following five categories, “1, Immediate”, “2, High risk of deterioration”, “3, Stable with multiple resources needed”, “4, Stable with only one type of resource needed” and “5, Stable with no resources anticipated other than medication”. Although both of these systems have proven valid and reliable ( 18 ), there is a dearth of large scale empirical studies investigating if and how these disparate systems affect the provision of care differently. For example, while it is widely accepted that both systems share similar shortcomings when it comes to the accurate triaging of patients over the age of 65 ( 19 ), little is known about how these shortcomings compare. To address this deficit, the present study aimed to examine potential differences in the characteristics and clinical outcomes of patients triaged to different ESI or MTS categories in German EDs. Variables of interest included triage system (ESI or MTS), initial triage category (one to five), patient’s age and sex, diagnosis information (ICD-10-GM-2017) and admission status (inpatient/outpatient). Methods Study Design This study is a retrospective analysis of the routine ED data collected by the INDEED project – Utilization and Cross-Sectoral Patterns of Care for Patients Admitted to Emergency Departments in Germany . The sample population included all statutory health insured adult patients who attended one of 16 participating hospitals in 2016. The study protocol was reviewed by the institutional review board of the Charité – Universitätsmedizin Berlin (application: EA4/086/17). A detailed explanation of the dataset, including the data protection concept, has been published elsewhere ( 20 ). Study Setting Unlike other countries where national triage systems are in place ( 11 ), the decentralized German healthcare system ( 21 ) does not require EDs to utilise a specific triage tool. Although structured and validated triage tools became a requirement of emergency care provision in 2018 ( 14 ) during the study period participating hospitals were not required to follow a particular triage system or mode of operation for such. This meant that in addition to utilising different triage systems, the 16 study sites could potentially use the same system in different ways. Participating hospital were identified for exclusion based on the following systematic criteria: 1) comprehensive availability and quality of the variables of interest and 2) uniformity of triage categories across individual study sites that ostensibly utilised the same system. Study sites that showed noticeably different triage category distributions to a majority of other hospitals that utilised the same system were excluded on the grounds that the data indicated a hospital specific adaptation in either triage category attribution or documentation processes. Variables All variables were documented in the respective centre’s hospital information systems. Variables of interest included triage system (ESI or MTS), initial triage category (one = highest risk to five = lowest risk), patient demographic details (age and sex), diagnosis information (ICD-10-GM-2016) and admission status (inpatient/outpatient). Corresponding triage categories from ESI and MTS were coded from one to five based on their respective urgency, with one being the most urgent and five the least. Statistical Analysis Routine hospital data from individual study sites were analysed in subgroups according to their respective triage system (ESI or MTS). Absolute and relative frequencies were calculated and compared for triage categories. Patient demographics and clinical outcomes are presented as absolute and relative frequencies. After the median (med) age and interquartile range (IQR) were calculated, patient age was recoded into a new ordinal variable (18–34, 35–49, 50–64, > 65 years). In all figures depicting the relative frequency of hospitalisation, age group was used as a stratification variable due to the established propensity of both ESI and MTS to under-triage elderly patients ( 19 ). For all cases of hospital admission observed in the “less-urgent” triage categories (four and five), absolute and relative frequencies of associated ICD codes were analysed to provide further insight into these cases. For data anonymization purposes, any results below 10 cases were reported as “<10” and all percentages below 10% were reported as “<10%”. Separate mixed-effects multivariable regression analyses were run to test for associations between ESI or MTS categories and the outcome measure admission status (admitted vs. non-admitted to hospital). The mixed-effects model studied the relationship between less-urgent triage categories (reference=less urgent=categories four and five) as a fixed effect, adjusting for age (per 10 years; reference = 20) and sex (reference = men). Study site was included as a random effect. The results are expressed as crude and adjusted odds ratios including confidence intervals (CI 95%) and p-values. Inter-class correlation was calculated in order to determine how much variation in the outcome measures could be explained by differences amongst the study sites which shared the same triage system and to justify the application of the mixed model. All analysis were performed using the Statistical Package for Social Sciences (IBM SPSS V.27) and the R Project for Statistical Computing (V.4.3) using the packages tidyverse and lme4 ( 22 – 24 ) . Results Sample Population Of the 16 participating study sites, 12 centres were included in this analysis. Four hospitals were excluded: two were excluded due to the high proportion of missing triage data (≥ 75%), one due to all cases being recorded as inpatients, and lastly, one due to evidence of hospital specific adaptation in the triage process. In specific, when compared with other study sites using the same system, we found that a disproportionate proportion of patients were triaged as category one. After analysis of the complete INDEED data set (21), it was determined that this was due to all patients arriving by ambulance being categorised as triage category one. The final sample included routine data from 354,497 ED visits at the 12 participating hospital sites. Nine hospitals used MTS (n = 276,205) and three utilised ESI (n = 78,292) (see Table 1). Table 1 – Proportion of cases by triage categories. and distribution of gender, age and admission status within the respective categories of the different triage systems. All Patients Triage 1 (highest risk) Triage 2 Triage 3 Triage 4 Triage 5 (lowest risk) Triage not assessed MTS ESI MTS ESI MTS ESI MTS ESI MTS ESI MTS ESI MTS ESI n (%) 276,205 (100.0) 78,292 (100.0) 2,509 (< 1.0) 1,179 (1.5) 27,115 (9.8) 14,819 (18.9) 109,083 (39.5) 30,106 (38.5) 107,659 (39.0) 18,767 (24.0) 9,494 (3.4) 3,215 (4.1) 20,345 (7.4) 10,206 (13.0) Women, n (%) 140,945 (51.0) 37,025 (47.3) 1,003 (40.0) 525 (44.5) 13,298 (49.0) 6,899 (46.6) 56,314 (51.6) 14,758 (49.0) 55,118 (51.2) 8,697 (46.3) 4,780 (50.3) 1,433 (44.6) 10,432 (51.3) 4,713 (46.2) Age, Median (IQR) 53 (34,73) 54 (34,73) 66 (50,78) 66 (50,79) 61 (43,76) 64 (46,78) 54 (35,74) 56 (36,74) 54 (33,70) 40 (28,59) 48 (32,66) 41 (27,53) 57 (36,76) 59 (39,76) Inpatient admissions, n (%) 99,933 (36.2) 38,579 (49.3) 2,102 (83.8) 1,125 (95.4) 17,075 (63.0) 11,194 (75.5) 43,852 (40.2) 16,446 (54.6) 26,042 (24.2) 2,083 (11.1) 1,992 (21.0) 223 (6.9) 8,870 (43.6) 7,508 (73.6) Table 1 – Legend: Shown are absolute and relative proportions n (%). Results below 10 cases are reported “<10” and percentages below 1% are reported as “<1.0” due to data protection rules. Abbreviations: ESI - Emergency Severity Index, MTS - Manchester Triage System The proportion of recorded cases in each triage category differed across both systems (see Table 1). The largest differences were observed in categories two (9.1% difference) and four (15.0% difference). Median age was found to be similar across both populations but differed across individual triage categories: most notably in the categories four and five. Hospitals using MTS were visited by more women than those using ESI, and this was reflected in the proportion of women in each triage category with more pronounced differences in categories four and five. Hospitals using MTS had a lower proportion of inpatient cases than hospitals using ESI in categories one, two and three and a higher proportion in categories four and five. Age Group Analysis Hospitals using MTS were more likely than those using ESI to triage patients into lower triage categories as age increased (see Fig. 1). This pattern was most apparent in the decreasing proportion of patients triaged as category four in hospitals using MTS and the increasing proportion of patients triaged as category two in hospitals using ESI. The proportion of category one cases showed the least difference between the two systems. Analysis of patient admission status by age group indicated that a higher proportion of patients triaged as categories four or five were later admitted as inpatients in hospitals using MTS (see Fig. 2). It should be noted that regardless of the triage system used there was a high degree of variance in the inpatient rates across hospitals (MTS: 22.8–49.9%, ESI: 39.0–82.0%). Regarding non-admitted cases the majority of cases was triaged in urgency category four (green) except for older patients in ESI-hospitals but also a high proportion was triaged in the yellow triage category in all age groups (Fig. 3). Characterization of patients who were admitted to hospital in triage categories four and five In hospitals using either system, patients aged 65 or older were the most likely to be admitted to hospital after being given a triage category of four (MTS: 53.6%, ESI: 36.9%). The top five most common diagnosis in patients given a triage score of four and later admitted to hospitals using MTS were: R10-abdominal and pelvic pain (n = 269, 1.7%), N17-acute renal failure (n = 268, 1.7%), I10 - Essential (primary) hypertension (n = 260, 1.7%), J18 – Pneumonia, organism unspecified (n = 253, 1.6%) and M62 – Other disorders of the muscle (n = 239, 1.5%). The top five most common diagnoses in patients given a triage score of four and later admitted in hospitals using ESI were: M54 – Dorsalgia (n = 83, 3.5%), T14 - Injury of unspecified body region (n = 72, 3.0%), R51-Headache (n = 70, 2.9%), R42-Dizziness and giddiness (n = 69, 2.9%) and R20- Disturbances of skin sensation (n = 56, 2.4%). As for triage category five, again patients aged 65 or older were the most likely to be admitted to hospital after being given a triage category of five (MTS: 54.8%, ESI: 38.5%). The top five most common diagnoses in patients given a triage score of five and later admitted in hospitals using MTS were: F10 - Mental and behavioural disorders due to use of alcohol (n = 197, 13.9%), R10 Abdominal and Pelvic pain (n = 56, 3.9%), I63 – Cerebral infarction (n = 34, 2.4%), F32 – Major depressive disorder, single episode, mild (n = 24, 1.7%) and M54 – Dorsalgia (n = 23, 1.6%). The top five most common diagnoses in patients given a triage score of five and later admitted in hospitals using ESI were: R06 – Abnormalities of breathing (n = 11, 3.9%), R10 – Abdominal and Pelvic pain (n = 11, 3.9%), R07- Pain in throat and chest (n < 10, <1%) and E87- Other disorders of fluid, electrolyte and acid-base balance (< 10, < 1%). Regression Analysis The results of the inter-class correlation (ICC) for the random effect “study site” on admission status was small (0.07; 95%-CI 0.04–0.19). When analysing the centres separately, the ICC for EDs using the MTS was 0.06 (95% CI: 0.03–0.18), while for those using the ESI, the ICC was 0.14 (95% CI: 0.04–0.86). The smaller number of clusters (centres) for ESI, fewer than five, likely contributed to the elevated ICC value. The model results (see Table 2) indicated that in hospitals using ESI, patients were 11.6 times more likely to be admitted as inpatients in triage categories one, two and three than in categories four and five. In hospitals using MTS in contrast, ED patients were 3.0 times more likely to be admitted as an inpatient in categories one, two and three than in categories four and five. Table 2 – Generalized Mixed Regression Model for inpatient stay in hospitals using MTS and ESI. MTS Inpatient stay Crude OR (95%-CI) Adjusted OR* (95%-CI) Adjusted p-value* Women , ref. = men 0.90 (0.89–0.91) 0.86 (0.85–0.88) < 0.001 Age, per 10 years , ref. = 20 1.42 (1.41–1.43) 1.38 (1.37–1.38) < 0.001 Triage Categories 1–3 , ref. = 4–5 2.65 (2.60–2.69) 3.03 (2.97–3.09) < 0.001 ESI Inpatient stay Crude OR (95%-CI) Adjusted OR* (95%-CI) Adjusted p-value* Women, ref. = men 1.02 (0.99–1.05) 0.91 (0.87–0.94) < 0.001 Age, per 10 years , ref. = 20 1.39 (1.38–1.40) 1.30 (1.28–1.31) < 0.001 Triage Categories 1–3 , ref. = 4–5 14.16 (13.51–14.84) 11.63 (11.08–12.21) < 0.001 Table 2 - Legend: Triage was used as a fixed effect with categories 4 and 5 (low risk) as a reference category compared to categories 1-3 (high to moderate risk),. Study site was included as a random effect. Abbreviations: MTS - Manchester Triage System, ESI - Emergency Severity Index, ref – reference. *The multiple logistic regression was adjusted for age (per 10 years; reference = 20) and sex (reference = men). OR = Odds Ratio. Discussion We found that the proportion of patients admitted to hospital was higher in the non-urgent triage categories in hospitals that used MTS than in those that used ESI. This finding was supported by the results of the mixed model which showed that within the sample population it was less likely to be admitted as an inpatient in categories four and five than one, two and three in hospitals using ESI than MTS. The rates of hospitalisation imply that the risk of possible under-triage ( 25 ) was greater in hospitals that used MTS than in those that used ESI across all age categories. On the other hand, a much lower proportion was allocated in the lower triage categories in ESI hospitals in general, which indicates a lack of differentiation between higher and lower urgency. This pattern was not only observed among admitted patients, but also in the non-admitted population, where ESI tended to assign fewer cases to the lowest urgency category. One potential explanation for this is that by incorporating resource need into the triage algorithm, ESI is less likely to allocate patients into lower categories than MTS, which only takes into account patients' presenting condition ( 26 ). Taking elderly patients as an example, as these patients are known to require more resources ( 27 ), it follows that they would be allocated to higher triage categories in ESI than in MTS. Expanding on this logic, it also follows that patients who require resources but do not exhibit the genuine life-threatening conditions for which emergency medical services are designed may be seen to faster in ESI than MTS. While exact triage to treatment time goes beyond the purview of this study, under normal circumstances patients given higher triage scores should be seen to faster than those who receive lower categorisation ( 3 , 6 ). As such, individual patients presenting with conditions that are not urgent but require resources beyond medication may be seen to faster in hospitals using ESI due to higher triage categorisation. With that said, the impact of the resource-based approach on the wider population in terms of patient flow is unclear but would possibly lead to longer waiting times until first physician contact also within the majority of patients with urgent triage categories in a crowded ED. This potential lack of differentiation between higher and lower urgency in ESI might therefore minimise the risk of under-triage but lack efficiency for ED patient flow. Although it has been reported that the correct categorisation of patients with low-acuity conditions can increase efficiency in terms of patient flow and reduces waiting times for high-acuity cases ( 9 ), it may be that attributing more patients to higher triage categories negatively impacts waiting times and length of stay. Potential impact aside, it is important to note that in addition to being a resource-based system, ESI is also characterised by a high dependence on individual triager’s experience and intuition ( 10 ). The extent to which differences between ESI and MTS can be attributed to resources is uncertain: the higher degree of subjectivity could be decisive ( 28 ). What is clear however is that triage category alone is an insufficient gauge of acuity regardless of what triage system is used. The proportion of patients given a triage score of four or five that were later admitted to hospital provides further evidence that the sufficiency of initial triage scores are limited to some extent in order to determine which patients can be redirected to other sources of healthcare ( 29 ). Acuity of patients presenting condition can change rapidly and requires consistent monitoring ( 30 ). These results underline that current triage systems - regardless of their design – are not suitable for the purpose of patient streaming or redirection at ED arrival. Limitations A central limitation of this work comes from the nature of triage process itself. The triage process is a complex task ( 31 ) that can be affected by a range of different individual factors, such as training and experience, and external factors such as staffing and crowding ( 28 ). In practice, this means that the triage process can be subject to a high degree of variation ( 32 ). The aggregated results of this study must therefore be treated with caution. More studies comparing ESI and MTS are required to see if the patterns of hospitalization identified herein are representative. While direct comparisons of triage systems within the same population are desirable in this regard; such designs are not without their own limitations. Considering the resources that would be required, including staff training in both respective triage systems, and the potential impact on patient safety, the use of routine data as described in this study provides a cost-effective and efficient means of estimating the impact of each triage system on clinical outcomes. A secondary limitation stems from the focus on possible under triage. While under triage is known to affect patient safety ( 33 ), previous research has indicated that elderly patients can also be given a higher treatment priority than necessary ( 34 ). To fully understand how both systems, compare in term of triage accuracy, it is necessary to take both under- and over-triage into account. This task is complicated by the lack of a true gold standard for true patient urgency ( 25 ): the measure used herein to identify cases of possible under triage namely, hospitalization rates, is only an approximation. Future studies should look to incorporate a measure of over-triage and in-hospital mortality as another indicator of potential under-triage in this regard. This was not possible in the current retrospective study due to commitments to anonymization which could have been infringed by the reporting of small case numbers ( 20 ). Further variables aside, the current method provides a replicable means of investigating if and how these disparate triage systems affect the provision of care differently and in doing so address a significant gap in the current literature. Conclusions Distinct differences in the distribution of triage categories were observed in the analysis of routine hospital data from EDs in Germany using either MTS or ESI. Hospitals utilising MTS, appeared more likely than those using ESI to triage patients into the less-urgent triage categories and had a higher proportion of hospital admissions than hospitals in these less-urgent categories. While the results of this study could have been affected by differences in the underlying populations between hospitals applying either MTS or ESI, our findings hint towards systematic differences in the clinical outcomes of patients triaged using either system and raise questions concerning the accuracy of triage categorization. Declarations Acknowledgements The authors would like to extend their thanks to the participating Emergency Departments: Michael Bernhard (University Hospital Düsseldorf), Hans-Jörg Busch (University Hospital Freiburg), Christian Wrede (Helios Clinical Centre Berlin Buch), Rajan Somasundaram (Charité - Universitätsmedizin Berlin, Campus Benjamin Franklin), Timo Schöpke (Barnim Hospital), Erik Weidmann (Ruppiner Hospital in Neuruppin), Bernhard Flasch (Frankfurt (Oder) Hospital), Heike Höger-Schmidt (Chemnitz Hospital), André Gries (University Hospital Leipzig), Constanze Schwarz (Sana Hospital, Leipzig County), Wilhelm Behringer (University Hospital Jena), Bernadett Erdmann (Wolfsburg Hospital), Sabine Blaschke (University Hospital Göttingen), Sebastian Wolfrum (University Hospital Lübeck). We would also like to thank Dominik Brammen (Magdeburg) for support during the application phase, and Michael Erhart (Berlin) also for supporting application and data extraction. For their valuable contributions to INDEED we would like to thank Natalie Baier, Reinhard Busse, Dominik Brammen, Johannes Drepper, Patrik Dröge, Felix Greiner, Cornelia Henschke, Stella Kuhlmann, Björn Kreye, Christian Lüpkes, Thomas Reinhold, Burgi Riens, Marie-Luise Rosenbusch, Felix Staeps, Kristin Schmieder, Daniel Schreiber, Dominik von Stillfried, Maike Below, Rainer Röhrig, Stephanie Roll, Thomas Ruhnke, Felix Walcher, and Grit Zimmermann (all Germany), and Ryan King (Australia). Funding INDEED was funded by the Innovation Fund of the Federal Joint Committee (Innovationsfonds des Gemeinsamen Bundesausschusses, grant number 01VSF16044). Ethics Approval This study was approved by the ethics committee of the Charité – Universitätsmedizin Berlin (application: EA4/086/17) and was conducted in accordance with the ethical principles of the Declaration of Helsinki. Consent to Participate Written informed consent for participation was not required for this study. This study is a retrospective analysis of routinely collected data from the INDEED project (application: EA4/086/17). The collection and use of these data were approved by the responsible regulatory authorities and carried out in accordance with applicable national legislation, including the relevant provisions of the German Social Code Books (SGB X, V, and I). Individual declarations of consent were not applicable given the retrospective secondary use of routine health care data. Authors contribution David Legg: data curation, formal analysis, investigation, methodology writing – original draft, writing – review & editing , visualization; Yves-Noel Wu : formal analysis, software, methodology writing – original draft, writing – review & editing, correspondence; Myrto Bolanaki : analysis, writing – review & editing; Antje Fischer-Rosinsky : supervision, writing – review & editing; Thomas Keil : writing – review, supervision & editing, funding acquisition; Martin Möckel : conceptualization, methodology, writing – review & editing, supervision, funding acquisition and ; Anna Slagman : conceptualization, methodology, writing – review & editing, supervision, funding acquisition, validation, project administration, resources. References Robertson-Steel I. Evolution of triage systems. Emerg Med J. 2006;23(2):154–5. Dickinson A, Joos S. Barriers to integration of primary care into emergency care: experiences in Germany. Int J Integr care. 2021;21(2). Yancey CC, O'Rourke MC. Emergency department triage. 2020. Di Somma S, Paladino L, Vaughan L, Lalle I, Magrini L, Magnanti M. Overcrowding in emergency department: an international issue. Intern Emerg Med. 2015;10:171–5. Sartini M, Carbone A, Demartini A, Giribone L, Oliva M, Spagnolo AM, et al. editors. Overcrowding in emergency department: causes, consequences, and solutions—a narrative review. Healthcare: MDPI; 2022. Calder S, Platz E. Triage systems. 2014. FitzGerald G, Jelinek GA, Scott D, Gerdtz MF. Emergency department triage revisited. Emerg Med J. 2010;27(2):86–92. van der Linden MC, Meester BE, van der Linden N. Emergency department crowding affects triage processes. Int Emerg Nurs. 2016;29:27–31. Zachariasse JM, van der Hagen V, Seiger N, Mackway-Jones K, van Veen M, Moll HA. Performance of triage systems in emergency care: a systematic review and meta-analysis. BMJ open. 2019;9(5):e026471. Hinson JS, Martinez DA, Cabral S, George K, Whalen M, Hansoti B, et al. Triage performance in emergency medicine: a systematic review. Ann Emerg Med. 2019;74(1):140–52. Peta D, Day A, Lugari WS, Gorman V, Pajo VMT. Triage: A global perspective. J Emerg Nurs. 2023;49(6):814–25. Schacher S, Kuehl M, Gräff I. Some machine’s doin’that for you*–elektronische Triagesysteme in der Notaufnahme. Notfall+ Rettungsmedizin. 2023;26(5):331–8. Christ M, Bingisser R, Nickel CH. Bedeutung der Triage in der klinischen Notfallmedizin. DMW-Deutsche Medizinische Wochenschrift. 2016;141(05):329–35. Bundesausschuss G. Regulations on a tiered system of emergency structures in hospitals 2020 [Available from: https://www.g-ba.de/richtlinien/103/ Wallstab F, Greiner F, Schirrmeister W, Wehrle M, Walcher F, Wrede C, et al. German emergency department measures in 2018: a status quo based on the Utstein reporting standard. BMC Emerg Med. 2022;22(1):5. Gräff I, Goldschmidt B, Glien P, Bogdanow M, Fimmers R, Hoeft A, et al. The German Version of the Manchester Triage System and its quality criteria–first assessment of validity and reliability. PLoS ONE. 2014;9(2):e88995. Grossmann FF, Nickel CH, Christ M, Schneider K, Spirig R, Bingisser R. Transporting clinical tools to new settings: cultural adaptation and validation of the Emergency Severity Index in German. Ann Emerg Med. 2011;57(3):257–64. Möckel M, Reiter S, Lindner T, Slagman A. Triage—primary assessment of patients in the emergency department: An overview with a systematic review. Medizinische Klinik-Intensivmedizin und Notfallmedizin. 2020;115:668–81. Groening M, Wilke P. Triage, screening, and assessment of geriatric patients in the emergency department. Medizinische Klinik-Intensivmedizin und Notfallmedizin. 2020;115:8–15. Fischer-Rosinský A, Slagman A, King R, Reinhold T, Schenk L, Greiner F, et al. INDEED–Utilization and cross-sectoral patterns of care for patients admitted to emergency departments in Germany: Rationale and study design. Front Public Health. 2021;9:616857. Health FMo. The German healthcare system. Strong. Reliable. Proven. www.bundesgesundheitsministerium.de: Federal Ministry of Health. 2018 [Available from: https://www.bun desgesundheitsministerium.de/fileadmin/Dateien/ 5_Publikationen/Gesundheit/Broschueren/ 200629_BMG_Das_deutsche_ Gesundheitssystem_EN.pdf Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. arXiv preprint arXiv:14065823. 2014. Team RC. RA language and environment for statistical computing. R Foundation for Statistical. Computing; 2020. Wickham H, Averick M, Bryan J, Chang W, McGowan LDA, François R, et al. Welcome to the Tidyverse. J open source Softw. 2019;4(43):1686. Brouns SH, Mignot-Evers L, Derkx F, Lambooij SL, Dieleman JP, Haak HR. Performance of the Manchester triage system in older emergency department patients: a retrospective cohort study. BMC Emerg Med. 2019;19:1–11. Kemp K, Mertanen R, Lääperi M, Niemi-Murola L, Lehtonen L, Castren M. Nonspecific complaints in the emergency department–a systematic review. Scand J Trauma Resusc Emerg Med. 2020;28:1–12. Ginsburg AD, e Silva LOJ, Mullan A, Mhayamaguru KM, Bower S, Jeffery MM, et al. Should age be incorporated into the adult triage algorithm in the emergency department? Am J Emerg Med. 2021;46:508–14. Considine J, Botti M, Thomas S. Do knowledge and experience have specific roles in triage decision-making? Acad Emerg Med. 2007;14(8):722–6. Slagman A, Greiner F, Searle J, Harriss L, Thompson F, Frick J, et al. Suitability of the German version of the Manchester Triage System to redirect emergency department patients to general practitioner care: a prospective cohort study. BMJ open. 2019;9(5):e024896. Burström L, Starrin B, Engström M-L, Thulesius H. Waiting management at the emergency department–a grounded theory study. BMC Health Serv Res. 2013;13:1–10. Butler K, Anderson N, Jull A. Evaluating the effects of triage education on triage accuracy within the emergency department: An integrative review. Int Emerg Nurs. 2023;70:101322. Gorick H, McGee M, Wilson G, Williams E, Patel J, Zonato A, et al. Understanding triage assessment of acuity by emergency nurses at initial adult patient presentation: A qualitative systematic review. Int Emerg Nurs. 2023;71:101334. Grossmann FF, Zumbrunn T, Ciprian S, Stephan F-P, Woy N, Bingisser R, et al. Undertriage in older emergency department patients–tilting against windmills? PLoS ONE. 2014;9(8):e106203. Savioli G, Ceresa IF, Bressan MA, Bavestrello Piccini G, Novelli V, Cutti S, et al. Geriatric Population Triage: The Risk of Real-Life Over-and Under-Triage in an Overcrowded ED: 4-and 5-Level Triage Systems Compared: The CREONTE (Crowding and RE Organization National TriagE) Study. J Personalized Med. 2024;14(2):195. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 29 Apr, 2026 Reviewers agreed at journal 19 Apr, 2026 Reviewers invited by journal 17 Apr, 2026 Editor assigned by journal 20 Mar, 2026 Submission checks completed at journal 20 Mar, 2026 First submitted to journal 19 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9171354","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":627889512,"identity":"fa97239d-d1a1-4a84-9b60-85259acc1696","order_by":0,"name":"David Legg","email":"","orcid":"","institution":"Charité - University Medicine Berlin","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Legg","suffix":""},{"id":627889513,"identity":"8d5d5347-0d5f-48ce-a0f6-138e49c61df1","order_by":1,"name":"Yves Noel Wu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIie3OsYrCMBjA8S98UBe1a0WwPoIiVITqvUpLIZO3Ogl2yi0+QAefQxyVD3QpujrcoAg3B1xcDi/pIC6pd9sN+Q8hJPklAbDZ/nN1PUhYA6CeOb8g+gzLCoJ/IFjVBF6QzjEhyVZhy4Fq0Aynn37/w92cYBKWEM49lvNeQcbbr+6CEDuw52aS5wF+C4oF1JbN95RYhuh4TFAJOVylOjAryCClN0UqNybuZrKbg74zcjRhKcX6FWBibSSNnQgU4V2B7nUw31KiSM+L9omR1Akv6mOh71ZEfLxNaZi5m7OUk5GRtB8fwOflyAgA/LRk02az2WxFPzvhS43uV9LGAAAAAElFTkSuQmCC","orcid":"","institution":"Charité - University Medicine Berlin","correspondingAuthor":true,"prefix":"","firstName":"Yves","middleName":"Noel","lastName":"Wu","suffix":""},{"id":627889518,"identity":"43aa66bc-5d14-4041-bced-e3b4d943f02d","order_by":2,"name":"Myrto Bolanaki","email":"","orcid":"","institution":"Charité - University Medicine Berlin","correspondingAuthor":false,"prefix":"","firstName":"Myrto","middleName":"","lastName":"Bolanaki","suffix":""},{"id":627889520,"identity":"ce37ccb1-7018-433b-86a2-26144b316d36","order_by":3,"name":"Antje Fischer-Rosinský","email":"","orcid":"","institution":"Charité - University Medicine Berlin","correspondingAuthor":false,"prefix":"","firstName":"Antje","middleName":"","lastName":"Fischer-Rosinský","suffix":""},{"id":627889522,"identity":"63176c75-74f1-4cf2-994f-d52d67d96469","order_by":4,"name":"Thomas Keil","email":"","orcid":"","institution":"Charité - University Medicine Berlin","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Keil","suffix":""},{"id":627889524,"identity":"be1b2194-75e9-4823-857a-87fa2a9c34b9","order_by":5,"name":"Martin Möckel","email":"","orcid":"","institution":"Charité - University Medicine Berlin","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Möckel","suffix":""},{"id":627889527,"identity":"92890ed8-5021-4f59-b8d5-5f269b2b364d","order_by":6,"name":"Anna Slagman","email":"","orcid":"","institution":"Charité - University Medicine Berlin","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Slagman","suffix":""}],"badges":[],"createdAt":"2026-03-19 15:53:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9171354/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9171354/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107839268,"identity":"7e419ab4-ac97-4c80-b5c4-3b539a1a6016","added_by":"auto","created_at":"2026-04-26 17:17:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":54217,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eProportion of all cases in each triage category stratified by age group. Shown are relative proportion of cases (%). Abbreviations: MTS - Manchester Triage System, ESI - Emergency Severity Index.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"drawingimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9171354/v1/50bd13491137a0b8cab604eb.png"},{"id":107839269,"identity":"674f6b99-5444-40b2-b802-86a834301eba","added_by":"auto","created_at":"2026-04-26 17:17:08","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":61640,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eProportion of \u003c/em\u003e\u003cu\u003e\u003cem\u003ehospitalised (admitted as inpatient)\u003c/em\u003e\u003c/u\u003e\u003cem\u003e cases in each triage category stratified by age group. Shown are relative proportions of all inpatient cases (%). \u0026nbsp;Abbreviations: MTS - Manchester Triage System, ESI - Emergency Severity Index.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"drawingimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9171354/v1/89fddca7831bb1c2e3f3d58a.png"},{"id":107870771,"identity":"838a73f3-ef22-4b70-a678-db9126e235db","added_by":"auto","created_at":"2026-04-27 07:40:36","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":54110,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eProportion of \u003c/em\u003e\u003cu\u003e\u003cem\u003enon-hospitalised (non-admitted outpatient) \u003c/em\u003e\u003c/u\u003e\u003cem\u003ecases in each triage category stratified by age group. Shown are relative proportions of all inpatient cases (%). Abbreviations: MTS - Manchester Triage System, ESI - Emergency Severity Index.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"drawingimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9171354/v1/467690ef8344b8cf6fdc9d3f.png"},{"id":107873051,"identity":"4dd577c6-09f6-4a2f-ad1b-c6c635e682d8","added_by":"auto","created_at":"2026-04-27 08:01:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":511483,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9171354/v1/73e45e06-06ba-4e76-a750-015099b354e3.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eEmergency Severity Index versus Manchester Triage System: Utilising Routine Data to Compare the Characteristics, Diagnoses and Admission Status of Patients Triaged in 12 German Emergency Departments\u003c/p\u003e","fulltext":[{"header":"Key Findings","content":"\u003cul\u003e\n \u003cli\u003eThere are currently two standardized triage system being used in Germany: Emergency Severity Index (ESI) and Manchester Triage System (MTS)\u003c/li\u003e\n \u003cli\u003eWhen comparing the triage systems MTS and ESI there are substantial differences in triage category distribution in different emergency departments\u003c/li\u003e\n \u003cli\u003eESI patients are in general categorized more urgent which might pose challenges to treat patients within recommended treatment time in crowded emergency departments\u003c/li\u003e\n \u003cli\u003eWith MTS on the other hand also patients who were later admitted to hospital were triaged into lower urgency categories which could be an indicator of undertriage. This was especially the case for patients with older age.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eA higher proportion of patients with ischemic stroke who were categorized in the least urgent triage category using MTS might reflect a concerning lack of sensitivity for stroke symptoms in MTS.\u003c/li\u003e\n \u003cli\u003eThe high proportion of patients in urgent triage categories who were later discharged directly from the ED in both systems might be an indicator of overtriage.\u0026nbsp;\u003cbr clear=\"all\"\u003e\u0026nbsp;\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"Introduction","content":"\u003cp\u003eDesigned to facilitate critical decision making processes at times when Emergency Departments (EDs) face increased demand (\u003cspan class=\"CitationRef\"\u003e1\u003c/span\u003e), triage systems are fundamental to the provision of modern emergency care (\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e). By structuring the allocation of finite resources in a systematic manner, these systems play a pivotal role in ensuring care is provided in order of need (\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e), especially at times of ED crowding: the continued negative impacts of which (\u003cspan class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e5\u003c/span\u003e) have made the ongoing review of triage scales increasingly important (\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e–\u003cspan class=\"CitationRef\"\u003e8\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo date, the majority of the evidence base concerning triage scales has been generated by single centre studies (\u003cspan class=\"CitationRef\"\u003e9\u003c/span\u003e). While these studies have shown individual triage systems to be valid and reliable methods for assessing the urgency of incoming patients’ conditions, few studies have compared their performance against one another and as a result there is increasing demand for comparisons of different triage scales’ performance with regards to clinically important outcomes (\u003cspan class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e11\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eStructured triage systems were first introduced in German EDs in 2004 (\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e). Since then the utilisation of these initial assessment tools have become standard practice (\u003cspan class=\"CitationRef\"\u003e13\u003c/span\u003e), with structured and validated triage tools becoming a requirement of emergency care provision in 2018 (\u003cspan class=\"CitationRef\"\u003e14\u003c/span\u003e). That said, there is still considerable variation in the type of triage systems used in German EDs. According to the latest available survey data, collected in 2018, the two most widely used triage systems are the Manchester Triage System (MTS) and Emergency Severity Index (ESI) (\u003cspan class=\"CitationRef\"\u003e15\u003c/span\u003e): the underlying mechanisms of which differ substantially.\u003c/p\u003e \u003cp\u003eThe German version of the MTS utilises a set of 50 presentation flowcharts and underlying algorithms to prioritise patients based on a rapid assessment of their presenting condition (\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e). Patients are allocated to one of the following colour coded triage categories, Red (1, Immediate), Orange (2, Very Urgent), Yellow (3, Urgent), Green (4, Standard) and Blue (5, Non-Urgent). The ESI in contrast utilises a simple three step algorithm to assess the urgency of a patient’s presenting condition and then, for less acute presentations with stable vital signs, expected resource need (\u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e). Patients in this system are allocated into one of the following five categories, “1, Immediate”, “2, High risk of deterioration”, “3, Stable with multiple resources needed”, “4, Stable with only one type of resource needed” and “5, Stable with no resources anticipated other than medication”.\u003c/p\u003e \u003cp\u003eAlthough both of these systems have proven valid and reliable (\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e), there is a dearth of large scale empirical studies investigating if and how these disparate systems affect the provision of care differently. For example, while it is widely accepted that both systems share similar shortcomings when it comes to the accurate triaging of patients over the age of 65 (\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e), little is known about how these shortcomings compare. To address this deficit, the present study aimed to examine potential differences in the characteristics and clinical outcomes of patients triaged to different ESI or MTS categories in German EDs. Variables of interest included triage system (ESI or MTS), initial triage category (one to five), patient’s age and sex, diagnosis information (ICD-10-GM-2017) and admission status (inpatient/outpatient).\u003c/p\u003e "},{"header":"Methods","content":"\u003cp\u003eStudy Design\u003c/p\u003e\u003cp\u003eThis study is a retrospective analysis of the routine ED data collected by the INDEED project – \u003cem\u003eUtilization and Cross-Sectoral Patterns of Care for Patients Admitted to Emergency Departments in Germany\u003c/em\u003e. The sample population included all statutory health insured adult patients who attended one of 16 participating hospitals in 2016. The study protocol was reviewed by the institutional review board of the Charité – Universitätsmedizin Berlin (application: EA4/086/17). A detailed explanation of the dataset, including the data protection concept, has been published elsewhere (\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eStudy Setting\u003c/p\u003e\u003cp\u003eUnlike other countries where national triage systems are in place (\u003cspan class=\"CitationRef\"\u003e11\u003c/span\u003e), the decentralized German healthcare system (\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e) does not require EDs to utilise a specific triage tool. Although structured and validated triage tools became a requirement of emergency care provision in 2018 (\u003cspan class=\"CitationRef\"\u003e14\u003c/span\u003e) during the study period participating hospitals were not required to follow a particular triage system or mode of operation for such. This meant that in addition to utilising different triage systems, the 16 study sites could potentially use the same system in different ways.\u003c/p\u003e\u003cp\u003eParticipating hospital were identified for exclusion based on the following systematic criteria: 1) comprehensive availability and quality of the variables of interest and 2) uniformity of triage categories across individual study sites that ostensibly utilised the same system. Study sites that showed noticeably different triage category distributions to a majority of other hospitals that utilised the same system were excluded on the grounds that the data indicated a hospital specific adaptation in either triage category attribution or documentation processes.\u003c/p\u003e\u003cp\u003eVariables\u003c/p\u003e\u003cp\u003eAll variables were documented in the respective centre’s hospital information systems. Variables of interest included triage system (ESI or MTS), initial triage category (one = highest risk to five = lowest risk), patient demographic details (age and sex), diagnosis information (ICD-10-GM-2016) and admission status (inpatient/outpatient). Corresponding triage categories from ESI and MTS were coded from one to five based on their respective urgency, with one being the most urgent and five the least.\u003c/p\u003e\u003ch2\u003eStatistical Analysis\u003c/h2\u003e\u003cp\u003eRoutine hospital data from individual study sites were analysed in subgroups according to their respective triage system (ESI or MTS). Absolute and relative frequencies were calculated and compared for triage categories.\u003c/p\u003e\u003cp\u003ePatient demographics and clinical outcomes are presented as absolute and relative frequencies. After the median (med) age and interquartile range (IQR) were calculated, patient age was recoded into a new ordinal variable (18–34, 35–49, 50–64, \u0026gt; 65 years). In all figures depicting the relative frequency of hospitalisation, age group was used as a stratification variable due to the established propensity of both ESI and MTS to under-triage elderly patients (\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e). For all cases of hospital admission observed in the “less-urgent” triage categories (four and five), absolute and relative frequencies of associated ICD codes were analysed to provide further insight into these cases. For data anonymization purposes, any results below 10 cases were reported as “\u0026lt;10” and all percentages below 10% were reported as “\u0026lt;10%”.\u003c/p\u003e\u003cp\u003eSeparate mixed-effects multivariable regression analyses were run to test for associations between ESI or MTS categories and the outcome measure admission status (admitted vs. non-admitted to hospital).\u003c/p\u003e\u003cp\u003eThe mixed-effects model studied the relationship between less-urgent triage categories (reference=less urgent=categories four and five) as a fixed effect, adjusting for age (per 10 years; reference = 20) and sex (reference = men). Study site was included as a random effect. The results are expressed as crude and adjusted odds ratios including confidence intervals (CI 95%) and p-values. Inter-class correlation was calculated in order to determine how much variation in the outcome measures could be explained by differences amongst the study sites which shared the same triage system and to justify the application of the mixed model.\u003c/p\u003e\u003cp\u003eAll analysis were performed using the Statistical Package for Social Sciences (IBM SPSS V.27) and the R Project for Statistical Computing (V.4.3) using the packages tidyverse and lme4 (\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e–\u003cspan class=\"CitationRef\"\u003e24\u003c/span\u003e) .\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eSample Population\u003c/p\u003e\n\u003cp\u003eOf the 16 participating study sites, 12 centres were included in this analysis. Four hospitals were excluded: two were excluded due to the high proportion of missing triage data (≥ 75%), one due to all cases being recorded as inpatients, and lastly, one due to evidence of hospital specific adaptation in the triage process. In specific, when compared with other study sites using the same system, we found that a disproportionate proportion of patients were triaged as category one. After analysis of the complete INDEED data set (21), it was determined that this was due to all patients arriving by ambulance being categorised as triage category one. The final sample included routine data from 354,497 ED visits at the 12 participating hospital sites. Nine hospitals used MTS (n = 276,205) and three utilised ESI (n = 78,292) (see Table 1).\u003c/p\u003e\n\u003cdiv\u003e \u0026nbsp;\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 1\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003e– Proportion of cases by triage categories. and distribution of gender, age and admission status within the respective categories of the different triage systems.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"15\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\n \u003cp\u003e\u003cem\u003eAll Patients\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003eTriage 1\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(highest risk)\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\n \u003cp\u003e\u003cem\u003eTriage 2\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c9\" namest=\"c8\"\u003e\n \u003cp\u003e\u003cem\u003eTriage 3\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e\n \u003cp\u003e\u003cem\u003eTriage 4\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e\n \u003cp\u003e\u003cem\u003eTriage 5\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(lowest risk)\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c15\" namest=\"c14\"\u003e\n \u003cp\u003e\u003cem\u003eTriage not assessed\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003eESI\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003eESI\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c7\"\u003e\n \u003cp\u003e\u003cem\u003eESI\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c8\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c9\"\u003e\n \u003cp\u003e\u003cem\u003eESI\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c10\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c11\"\u003e\n \u003cp\u003e\u003cem\u003eESI\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c12\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c13\"\u003e\n \u003cp\u003e\u003cem\u003eESI\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c14\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c15\"\u003e\n \u003cp\u003e\u003cem\u003eESI\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003en (%)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e\u003cem\u003e276,205\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(100.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e78,292\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(100.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e2,509\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(\u0026lt; 1.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e1,179\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(1.5)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e\u003cem\u003e27,115\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(9.8)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c7\"\u003e\n \u003cp\u003e\u003cem\u003e14,819\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(18.9)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c8\"\u003e\n \u003cp\u003e\u003cem\u003e109,083\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(39.5)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c9\"\u003e\n \u003cp\u003e\u003cem\u003e30,106\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(38.5)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c10\"\u003e\n \u003cp\u003e\u003cem\u003e107,659\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(39.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c11\"\u003e\n \u003cp\u003e\u003cem\u003e18,767\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(24.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c12\"\u003e\n \u003cp\u003e\u003cem\u003e9,494\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(3.4)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c13\"\u003e\n \u003cp\u003e\u003cem\u003e3,215\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(4.1)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c14\"\u003e\n \u003cp\u003e\u003cem\u003e20,345\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(7.4)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c15\"\u003e\n \u003cp\u003e\u003cem\u003e10,206\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(13.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eWomen,\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003en (%)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e\u003cem\u003e140,945\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(51.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e37,025\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(47.3)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e1,003\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(40.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e525\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(44.5)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e\u003cem\u003e13,298\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(49.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c7\"\u003e\n \u003cp\u003e\u003cem\u003e6,899\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(46.6)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c8\"\u003e\n \u003cp\u003e\u003cem\u003e56,314\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(51.6)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c9\"\u003e\n \u003cp\u003e\u003cem\u003e14,758\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(49.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c10\"\u003e\n \u003cp\u003e\u003cem\u003e55,118\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(51.2)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c11\"\u003e\n \u003cp\u003e\u003cem\u003e8,697\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(46.3)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c12\"\u003e\n \u003cp\u003e\u003cem\u003e4,780\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(50.3)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c13\"\u003e\n \u003cp\u003e\u003cem\u003e1,433\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(44.6)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c14\"\u003e\n \u003cp\u003e\u003cem\u003e10,432\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(51.3)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c15\"\u003e\n \u003cp\u003e\u003cem\u003e4,713\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(46.2)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eAge,\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eMedian (IQR)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e\u003cem\u003e53\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(34,73)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e54\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(34,73)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e66\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(50,78)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e66\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(50,79)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e\u003cem\u003e61\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(43,76)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c7\"\u003e\n \u003cp\u003e\u003cem\u003e64\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(46,78)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c8\"\u003e\n \u003cp\u003e\u003cem\u003e54\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(35,74)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c9\"\u003e\n \u003cp\u003e\u003cem\u003e56\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(36,74)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c10\"\u003e\n \u003cp\u003e\u003cem\u003e54\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(33,70)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c11\"\u003e\n \u003cp\u003e\u003cem\u003e40\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(28,59)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c12\"\u003e\n \u003cp\u003e\u003cem\u003e48\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(32,66)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c13\"\u003e\n \u003cp\u003e\u003cem\u003e41\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(27,53)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c14\"\u003e\n \u003cp\u003e\u003cem\u003e57\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(36,76)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c15\"\u003e\n \u003cp\u003e\u003cem\u003e59\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(39,76)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eInpatient admissions,\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003en (%)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e\u003cem\u003e99,933\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(36.2)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e38,579\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(49.3)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e2,102\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(83.8)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e1,125\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(95.4)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e\u003cem\u003e17,075\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(63.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c7\"\u003e\n \u003cp\u003e\u003cem\u003e11,194\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(75.5)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c8\"\u003e\n \u003cp\u003e\u003cem\u003e43,852\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(40.2)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c9\"\u003e\n \u003cp\u003e\u003cem\u003e16,446\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(54.6)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c10\"\u003e\n \u003cp\u003e\u003cem\u003e26,042\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(24.2)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c11\"\u003e\n \u003cp\u003e\u003cem\u003e2,083\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(11.1)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c12\"\u003e\n \u003cp\u003e\u003cem\u003e1,992\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(21.0)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c13\"\u003e\n \u003cp\u003e\u003cem\u003e223\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(6.9)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c14\"\u003e\n \u003cp\u003e\u003cem\u003e8,870\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(43.6)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c15\"\u003e\n \u003cp\u003e\u003cem\u003e7,508\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;(73.6)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cem\u003eTable 1 – Legend: Shown are absolute and relative proportions n (%). Results below 10 cases are reported “\u0026lt;10” and percentages below 1% are reported as\u003c/em\u003e \u003cem\u003e“\u0026lt;1.0” due to data protection rules. Abbreviations: ESI - Emergency Severity Index, MTS - Manchester Triage System\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe proportion of recorded cases in each triage category differed across both systems (see Table 1). The largest differences were observed in categories two (9.1% difference) and four (15.0% difference). Median age was found to be similar across both populations but differed across individual triage categories: most notably in the categories four and five. Hospitals using MTS were visited by more women than those using ESI, and this was reflected in the proportion of women in each triage category with more pronounced differences in categories four and five. Hospitals using MTS had a lower proportion of inpatient cases than hospitals using ESI in categories one, two and three and a higher proportion in categories four and five.\u003c/p\u003e\n\u003cp\u003eAge Group Analysis\u003c/p\u003e\n\u003cp\u003eHospitals using MTS were more likely than those using ESI to triage patients into lower triage categories as age increased (see Fig.\u0026nbsp;1). This pattern was most apparent in the decreasing proportion of patients triaged as category four in hospitals using MTS and the increasing proportion of patients triaged as category two in hospitals using ESI. The proportion of category one cases showed the least difference between the two systems.\u003c/p\u003e\n\u003cp\u003eAnalysis of patient admission status by age group indicated that a higher proportion of patients triaged as categories four or five were later admitted as inpatients in hospitals using MTS (see Fig.\u0026nbsp;2). It should be noted that regardless of the triage system used there was a high degree of variance in the inpatient rates across hospitals (MTS: 22.8–49.9%, ESI: 39.0–82.0%).\u003c/p\u003e\n\u003cp\u003eRegarding non-admitted cases the majority of cases was triaged in urgency category four (green) except for older patients in ESI-hospitals but also a high proportion was triaged in the yellow triage category in all age groups (Fig.\u0026nbsp;3).\u003c/p\u003e\n\u003cp\u003eCharacterization of patients who were admitted to hospital in triage categories four and five\u003c/p\u003e\n\u003cp\u003eIn hospitals using either system, patients aged 65 or older were the most likely to be admitted to hospital after being given a triage category of four (MTS: 53.6%, ESI: 36.9%). The top five most common diagnosis in patients given a triage score of four and later admitted to hospitals using MTS were: R10-abdominal and pelvic pain (n = 269, 1.7%), N17-acute renal failure (n = 268, 1.7%), I10 - Essential (primary) hypertension (n = 260, 1.7%), J18 – Pneumonia, organism unspecified (n = 253, 1.6%) and M62 – Other disorders of the muscle (n = 239, 1.5%). The top five most common diagnoses in patients given a triage score of four and later admitted in hospitals using ESI were: M54 – Dorsalgia (n = 83, 3.5%), T14 - Injury of unspecified body region (n = 72, 3.0%), R51-Headache (n = 70, 2.9%), R42-Dizziness and giddiness (n = 69, 2.9%) and R20- Disturbances of skin sensation (n = 56, 2.4%).\u003c/p\u003e\n\u003cp\u003eAs for triage category five, again patients aged 65 or older were the most likely to be admitted to hospital after being given a triage category of five (MTS: 54.8%, ESI: 38.5%). The top five most common diagnoses in patients given a triage score of five and later admitted in hospitals using MTS were: F10 - Mental and behavioural disorders due to use of alcohol (n = 197, 13.9%), R10 Abdominal and Pelvic pain (n = 56, 3.9%), I63 – Cerebral infarction (n = 34, 2.4%), F32 – Major depressive disorder, single episode, mild (n = 24, 1.7%) and M54 – Dorsalgia (n = 23, 1.6%). The top five most common diagnoses in patients given a triage score of five and later admitted in hospitals using ESI were: R06 – Abnormalities of breathing (n = 11, 3.9%), R10 – Abdominal and Pelvic pain (n = 11, 3.9%), R07- Pain in throat and chest (n \u0026lt; 10, \u0026lt;1%) and E87- Other disorders of fluid, electrolyte and acid-base balance (\u0026lt; 10, \u0026lt; 1%).\u003c/p\u003e\n\u003cp\u003eRegression Analysis\u003c/p\u003e\n\u003cp\u003eThe results of the inter-class correlation (ICC) for the random effect “study site” on admission status was small (0.07; 95%-CI 0.04–0.19). When analysing the centres separately, the ICC for EDs using the MTS was 0.06 (95% CI: 0.03–0.18), while for those using the ESI, the ICC was 0.14 (95% CI: 0.04–0.86). The smaller number of clusters (centres) for ESI, fewer than five, likely contributed to the elevated ICC value. The model results (see Table 2) indicated that in hospitals using ESI, patients were 11.6 times more likely to be admitted as inpatients in triage categories one, two and three than in categories four and five. In hospitals using MTS in contrast, ED patients were 3.0 times more likely to be admitted as an inpatient in categories one, two and three than in categories four and five.\u003c/p\u003e\n\u003cdiv\u003e \u0026nbsp;\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 2\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003e– Generalized Mixed Regression Model for inpatient stay in hospitals using MTS and ESI.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"5\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eMTS\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003eInpatient stay\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003eCrude OR (95%-CI)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003eAdjusted OR* (95%-CI)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003eAdjusted\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep-value*\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eWomen\u003c/em\u003e,\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eref. = men\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e0.90 (0.89–0.91)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e0.86 (0.85–0.88)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026lt; 0.001\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eAge, per 10 years\u003c/em\u003e,\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eref. = 20\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e1.42 (1.41–1.43)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e1.38 (1.37–1.38)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026lt; 0.001\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eTriage Categories 1–3\u003c/em\u003e,\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eref. = 4–5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e2.65 (2.60–2.69)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e3.03 (2.97–3.09)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026lt; 0.001\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cstrong\u003eESI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\n \u003cp\u003e\u003cstrong\u003eInpatient stay\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003eCrude OR (95%-CI)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003eAdjusted OR* (95%-CI)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003eAdjusted\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep-value*\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eWomen,\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;ref. = men\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e1.02 (0.99–1.05)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e0.91 (0.87–0.94)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026lt; 0.001\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eAge, per 10 years\u003c/em\u003e,\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eref. = 20\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e1.39 (1.38–1.40)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e1.30 (1.28–1.31)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026lt; 0.001\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\n \u003cp\u003e\u003cem\u003eTriage Categories 1–3\u003c/em\u003e,\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eref. = 4–5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e\u003cem\u003e14.16 (13.51–14.84)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e\u003cem\u003e11.63 (11.08–12.21)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026lt; 0.001\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cem\u003eTable 2 - Legend: Triage was used as a fixed effect with categories 4 and 5 (low risk) as a reference category compared to categories 1-3 (high to moderate risk),. Study site was included as a random effect. Abbreviations: MTS - Manchester Triage System, ESI - Emergency Severity Index, ref – reference. *The multiple logistic regression was adjusted for age (per 10 years; reference = 20) and sex (reference = men). OR = Odds Ratio.\u003c/em\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe found that the proportion of patients admitted to hospital was higher in the non-urgent triage categories in hospitals that used MTS than in those that used ESI. This finding was supported by the results of the mixed model which showed that within the sample population it was less likely to be admitted as an inpatient in categories four and five than one, two and three in hospitals using ESI than MTS.\u003c/p\u003e \u003cp\u003eThe rates of hospitalisation imply that the risk of possible under-triage (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e) was greater in hospitals that used MTS than in those that used ESI across all age categories. On the other hand, a much lower proportion was allocated in the lower triage categories in ESI hospitals in general, which indicates a lack of differentiation between higher and lower urgency. This pattern was not only observed among admitted patients, but also in the non-admitted population, where ESI tended to assign fewer cases to the lowest urgency category. One potential explanation for this is that by incorporating resource need into the triage algorithm, ESI is less likely to allocate patients into lower categories than MTS, which only takes into account patients' presenting condition (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). Taking elderly patients as an example, as these patients are known to require more resources (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e), it follows that they would be allocated to higher triage categories in ESI than in MTS.\u003c/p\u003e \u003cp\u003eExpanding on this logic, it also follows that patients who require resources but do not exhibit the genuine life-threatening conditions for which emergency medical services are designed may be seen to faster in ESI than MTS. While exact triage to treatment time goes beyond the purview of this study, under normal circumstances patients given higher triage scores should be seen to faster than those who receive lower categorisation (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). As such, individual patients presenting with conditions that are not urgent but require resources beyond medication may be seen to faster in hospitals using ESI due to higher triage categorisation. With that said, the impact of the resource-based approach on the wider population in terms of patient flow is unclear but would possibly lead to longer waiting times until first physician contact also within the majority of patients with urgent triage categories in a crowded ED. This potential lack of differentiation between higher and lower urgency in ESI might therefore minimise the risk of under-triage but lack efficiency for ED patient flow.\u003c/p\u003e \u003cp\u003eAlthough it has been reported that the correct categorisation of patients with low-acuity conditions can increase efficiency in terms of patient flow and reduces waiting times for high-acuity cases (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e), it may be that attributing more patients to higher triage categories negatively impacts waiting times and length of stay. Potential impact aside, it is important to note that in addition to being a resource-based system, ESI is also characterised by a high dependence on individual triager\u0026rsquo;s experience and intuition (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). The extent to which differences between ESI and MTS can be attributed to resources is uncertain: the higher degree of subjectivity could be decisive (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWhat is clear however is that triage category alone is an insufficient gauge of acuity regardless of what triage system is used. The proportion of patients given a triage score of four or five that were later admitted to hospital provides further evidence that the sufficiency of initial triage scores are limited to some extent in order to determine which patients can be redirected to other sources of healthcare (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). Acuity of patients presenting condition can change rapidly and requires consistent monitoring (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). These results underline that current triage systems - regardless of their design \u0026ndash; are not suitable for the purpose of patient streaming or redirection at ED arrival.\u003c/p\u003e \u003cp\u003eLimitations\u003c/p\u003e \u003cp\u003eA central limitation of this work comes from the nature of triage process itself. The triage process is a complex task (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e) that can be affected by a range of different individual factors, such as training and experience, and external factors such as staffing and crowding (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). In practice, this means that the triage process can be subject to a high degree of variation (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). The aggregated results of this study must therefore be treated with caution. More studies comparing ESI and MTS are required to see if the patterns of hospitalization identified herein are representative. While direct comparisons of triage systems within the same population are desirable in this regard; such designs are not without their own limitations. Considering the resources that would be required, including staff training in both respective triage systems, and the potential impact on patient safety, the use of routine data as described in this study provides a cost-effective and efficient means of estimating the impact of each triage system on clinical outcomes.\u003c/p\u003e \u003cp\u003eA secondary limitation stems from the focus on possible under triage. While under triage is known to affect patient safety (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e), previous research has indicated that elderly patients can also be given a higher treatment priority than necessary (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). To fully understand how both systems, compare in term of triage accuracy, it is necessary to take both under- and over-triage into account. This task is complicated by the lack of a true gold standard for true patient urgency (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e): the measure used herein to identify cases of possible under triage namely, hospitalization rates, is only an approximation. Future studies should look to incorporate a measure of over-triage and in-hospital mortality as another indicator of potential under-triage in this regard. This was not possible in the current retrospective study due to commitments to anonymization which could have been infringed by the reporting of small case numbers (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Further variables aside, the current method provides a replicable means of investigating if and how these disparate triage systems affect the provision of care differently and in doing so address a significant gap in the current literature.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eDistinct differences in the distribution of triage categories were observed in the analysis of routine hospital data from EDs in Germany using either MTS or ESI. Hospitals utilising MTS, appeared more likely than those using ESI to triage patients into the less-urgent triage categories and had a higher proportion of hospital admissions than hospitals in these less-urgent categories. While the results of this study could have been affected by differences in the underlying populations between hospitals applying either MTS or ESI, our findings hint towards systematic differences in the clinical outcomes of patients triaged using either system and raise questions concerning the accuracy of triage categorization.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eThe authors would like to extend their thanks to the participating Emergency Departments: Michael Bernhard (University Hospital D\u0026uuml;sseldorf), Hans-J\u0026ouml;rg Busch (University Hospital Freiburg), Christian Wrede (Helios Clinical Centre Berlin Buch), Rajan Somasundaram (Charit\u0026eacute; - Universit\u0026auml;tsmedizin Berlin, Campus Benjamin Franklin), Timo Sch\u0026ouml;pke (Barnim Hospital), Erik Weidmann (Ruppiner Hospital in Neuruppin), Bernhard Flasch (Frankfurt (Oder) Hospital), Heike H\u0026ouml;ger-Schmidt (Chemnitz Hospital), Andr\u0026eacute; Gries (University Hospital Leipzig), Constanze Schwarz (Sana Hospital, Leipzig County), Wilhelm Behringer (University Hospital Jena), Bernadett Erdmann (Wolfsburg Hospital), Sabine Blaschke (University Hospital G\u0026ouml;ttingen), Sebastian Wolfrum (University Hospital L\u0026uuml;beck). We would also like to thank Dominik Brammen (Magdeburg) for support during the application phase, and Michael Erhart (Berlin) also for supporting application and data extraction.\u003c/p\u003e\n\u003cp\u003eFor their valuable contributions to INDEED we would like to thank Natalie Baier, Reinhard Busse, Dominik Brammen, Johannes Drepper, Patrik Dr\u0026ouml;ge, Felix Greiner, Cornelia Henschke, Stella Kuhlmann, Bj\u0026ouml;rn Kreye, Christian L\u0026uuml;pkes, Thomas Reinhold, Burgi Riens, Marie-Luise Rosenbusch, Felix Staeps, Kristin Schmieder, Daniel Schreiber, Dominik von Stillfried, Maike Below, Rainer R\u0026ouml;hrig, Stephanie Roll, Thomas Ruhnke, Felix Walcher, and Grit Zimmermann (all Germany), and Ryan King (Australia).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eINDEED was funded by the Innovation Fund of the Federal Joint Committee (Innovationsfonds des Gemeinsamen Bundesausschusses, grant number 01VSF16044).\u003c/p\u003e\n\u003cp\u003eEthics Approval\u003c/p\u003e\n\u003cp\u003eThis study was approved by the ethics committee of the Charit\u0026eacute; \u0026ndash; Universit\u0026auml;tsmedizin Berlin (application: EA4/086/17) and was conducted in accordance with the ethical principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003eConsent to Participate\u003c/p\u003e\n\u003cp\u003eWritten informed consent for participation was not required for this study. This study is a retrospective analysis of routinely collected data from the INDEED project (application: EA4/086/17). The collection and use of these data were approved by the responsible regulatory authorities and carried out in accordance with applicable national legislation, including the relevant provisions of the German Social Code Books (SGB X, V, and I). Individual declarations of consent were not applicable given the retrospective secondary use of routine health care data.\u003c/p\u003e\n\u003cp\u003eAuthors contribution\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDavid Legg:\u0026nbsp;\u003c/strong\u003edata curation, formal\u0026nbsp;analysis, investigation, methodology writing \u0026ndash; original draft, writing \u0026ndash; review \u0026amp; editing\u003cstrong\u003e,\u0026nbsp;\u003c/strong\u003evisualization; \u003cstrong\u003eYves-Noel Wu\u003c/strong\u003e: formal analysis, software, methodology writing \u0026ndash; original draft, writing \u0026ndash; review \u0026amp; editing, correspondence; \u003cstrong\u003eMyrto Bolanaki\u003c/strong\u003e: analysis, writing \u0026ndash; review \u0026amp; editing; \u003cstrong\u003eAntje Fischer-Rosinsky\u003c/strong\u003e: supervision, writing \u0026ndash; review \u0026amp; editing; \u003cstrong\u003eThomas Keil\u003c/strong\u003e: writing \u0026ndash; review, supervision \u0026amp; editing, funding acquisition; \u003cstrong\u003eMartin M\u0026ouml;ckel\u003c/strong\u003e: conceptualization, methodology, writing \u0026ndash; review \u0026amp; editing, supervision, funding acquisition and ; \u003cstrong\u003eAnna Slagman\u003c/strong\u003e:\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003econceptualization, methodology, writing \u0026ndash; review \u0026amp; editing, supervision, funding acquisition, validation, project administration, resources.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eRobertson-Steel I. Evolution of triage systems. Emerg Med J. 2006;23(2):154\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDickinson A, Joos S. Barriers to integration of primary care into emergency care: experiences in Germany. Int J Integr care. 2021;21(2).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYancey CC, O'Rourke MC. Emergency department triage. 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Somma S, Paladino L, Vaughan L, Lalle I, Magrini L, Magnanti M. Overcrowding in emergency department: an international issue. Intern Emerg Med. 2015;10:171\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSartini M, Carbone A, Demartini A, Giribone L, Oliva M, Spagnolo AM, et al. editors. Overcrowding in emergency department: causes, consequences, and solutions\u0026mdash;a narrative review. Healthcare: MDPI; 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCalder S, Platz E. Triage systems. 2014.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFitzGerald G, Jelinek GA, Scott D, Gerdtz MF. Emergency department triage revisited. Emerg Med J. 2010;27(2):86\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Linden MC, Meester BE, van der Linden N. Emergency department crowding affects triage processes. Int Emerg Nurs. 2016;29:27\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZachariasse JM, van der Hagen V, Seiger N, Mackway-Jones K, van Veen M, Moll HA. Performance of triage systems in emergency care: a systematic review and meta-analysis. BMJ open. 2019;9(5):e026471.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHinson JS, Martinez DA, Cabral S, George K, Whalen M, Hansoti B, et al. Triage performance in emergency medicine: a systematic review. Ann Emerg Med. 2019;74(1):140\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeta D, Day A, Lugari WS, Gorman V, Pajo VMT. Triage: A global perspective. J Emerg Nurs. 2023;49(6):814\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchacher S, Kuehl M, Gr\u0026auml;ff I. Some machine\u0026rsquo;s doin\u0026rsquo;that for you*\u0026ndash;elektronische Triagesysteme in der Notaufnahme. Notfall+ Rettungsmedizin. 2023;26(5):331\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChrist M, Bingisser R, Nickel CH. Bedeutung der Triage in der klinischen Notfallmedizin. DMW-Deutsche Medizinische Wochenschrift. 2016;141(05):329\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBundesausschuss G. Regulations on a tiered system of emergency structures in hospitals 2020 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.g-ba.de/richtlinien/103/\u003c/span\u003e\u003cspan address=\"https://www.g-ba.de/richtlinien/103/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWallstab F, Greiner F, Schirrmeister W, Wehrle M, Walcher F, Wrede C, et al. German emergency department measures in 2018: a status quo based on the Utstein reporting standard. BMC Emerg Med. 2022;22(1):5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGr\u0026auml;ff I, Goldschmidt B, Glien P, Bogdanow M, Fimmers R, Hoeft A, et al. The German Version of the Manchester Triage System and its quality criteria\u0026ndash;first assessment of validity and reliability. PLoS ONE. 2014;9(2):e88995.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrossmann FF, Nickel CH, Christ M, Schneider K, Spirig R, Bingisser R. Transporting clinical tools to new settings: cultural adaptation and validation of the Emergency Severity Index in German. Ann Emerg Med. 2011;57(3):257\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026ouml;ckel M, Reiter S, Lindner T, Slagman A. Triage\u0026mdash;primary assessment of patients in the emergency department: An overview with a systematic review. Medizinische Klinik-Intensivmedizin und Notfallmedizin. 2020;115:668\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGroening M, Wilke P. Triage, screening, and assessment of geriatric patients in the emergency department. Medizinische Klinik-Intensivmedizin und Notfallmedizin. 2020;115:8\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFischer-Rosinsk\u0026yacute; A, Slagman A, King R, Reinhold T, Schenk L, Greiner F, et al. INDEED\u0026ndash;Utilization and cross-sectoral patterns of care for patients admitted to emergency departments in Germany: Rationale and study design. Front Public Health. 2021;9:616857.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHealth FMo. The German healthcare system. Strong. Reliable. Proven. www.bundesgesundheitsministerium.de: Federal Ministry of Health. 2018 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.bun\ndesgesundheitsministerium.de/fileadmin/Dateien/\n5_Publikationen/Gesundheit/Broschueren/\n200629_BMG_Das_deutsche_\nGesundheitssystem_EN.pdf\u003c/span\u003e\u003cspan address=\"https://www.bundesgesu\nndheitsministerium.de/fileadmin/D\nateien/5_Publikationen/Gesundheit\n/Broschueren/\n200629_BMG_Das_deutsche_Gesu\nndheitssystem_EN.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBates D, M\u0026auml;chler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. arXiv preprint arXiv:14065823. 2014.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeam RC. RA language and environment for statistical computing. R Foundation for Statistical. Computing; 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H, Averick M, Bryan J, Chang W, McGowan LDA, Fran\u0026ccedil;ois R, et al. Welcome to the Tidyverse. J open source Softw. 2019;4(43):1686.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrouns SH, Mignot-Evers L, Derkx F, Lambooij SL, Dieleman JP, Haak HR. Performance of the Manchester triage system in older emergency department patients: a retrospective cohort study. BMC Emerg Med. 2019;19:1\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKemp K, Mertanen R, L\u0026auml;\u0026auml;peri M, Niemi-Murola L, Lehtonen L, Castren M. Nonspecific complaints in the emergency department\u0026ndash;a systematic review. Scand J Trauma Resusc Emerg Med. 2020;28:1\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGinsburg AD, e Silva LOJ, Mullan A, Mhayamaguru KM, Bower S, Jeffery MM, et al. Should age be incorporated into the adult triage algorithm in the emergency department? Am J Emerg Med. 2021;46:508\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConsidine J, Botti M, Thomas S. Do knowledge and experience have specific roles in triage decision-making? Acad Emerg Med. 2007;14(8):722\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSlagman A, Greiner F, Searle J, Harriss L, Thompson F, Frick J, et al. Suitability of the German version of the Manchester Triage System to redirect emergency department patients to general practitioner care: a prospective cohort study. BMJ open. 2019;9(5):e024896.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurstr\u0026ouml;m L, Starrin B, Engstr\u0026ouml;m M-L, Thulesius H. Waiting management at the emergency department\u0026ndash;a grounded theory study. BMC Health Serv Res. 2013;13:1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eButler K, Anderson N, Jull A. Evaluating the effects of triage education on triage accuracy within the emergency department: An integrative review. Int Emerg Nurs. 2023;70:101322.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGorick H, McGee M, Wilson G, Williams E, Patel J, Zonato A, et al. Understanding triage assessment of acuity by emergency nurses at initial adult patient presentation: A qualitative systematic review. Int Emerg Nurs. 2023;71:101334.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrossmann FF, Zumbrunn T, Ciprian S, Stephan F-P, Woy N, Bingisser R, et al. Undertriage in older emergency department patients\u0026ndash;tilting against windmills? PLoS ONE. 2014;9(8):e106203.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSavioli G, Ceresa IF, Bressan MA, Bavestrello Piccini G, Novelli V, Cutti S, et al. Geriatric Population Triage: The Risk of Real-Life Over-and Under-Triage in an Overcrowded ED: 4-and 5-Level Triage Systems Compared: The CREONTE (Crowding and RE Organization National TriagE) Study. J Personalized Med. 2024;14(2):195.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-emergency-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijem","sideBox":"Learn more about [International Journal of Emergency Medicine](https://intjem.biomedcentral.com/)","snPcode":"12245","submissionUrl":"https://submission.nature.com/new-submission/12245/3","title":"International Journal of Emergency Medicine","twitterHandle":"@IntJEmergMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Triage, Emergency Department, Routine-Data, Manchester Triage System, Emergency Severity Index","lastPublishedDoi":"10.21203/rs.3.rs-9171354/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9171354/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground:\u003c/h2\u003e \u003cp\u003eAlthough the Manchester Triage System (MTS) and Emergency Severity Index (ESI) have proven valid and reliable methods of assessing the urgency of incoming emergency department (ED) patients, there is a dearth of studies that compare their performance. Thus, the aim of this study was to explore differences in the characteristics, diagnosis and admission status of patients triaged to different categories in German EDs that used either the MTS or ESI.\u003c/p\u003e\u003ch2\u003eMethod:\u003c/h2\u003e \u003cp\u003e This study is part of the INDEED project, a retrospective multicentre study of statutory health insured adult patients who attended one of 16 participating EDs in 2016. Variables of interest included triage system (ESI and MTS), initial triage category (one to five), patient age and sex, admission status (inpatient/outpatient) and diagnosis information (ICD-10-GM-2017) from hospital information systems. Absolute and relative frequencies were calculated for all variables and separate mixed-effects multivariable regressions were ran to test for the association of ESI and MTS categories and the outcome measure admission status.\u003c/p\u003e\u003ch2\u003eResults:\u003c/h2\u003e \u003cp\u003eThe final sample included routine data from 354,497 ED visits at 12 participating hospital sites (N\u0026thinsp;=\u0026thinsp;354,497). Nine hospitals used MTS (cases\u0026thinsp;=\u0026thinsp;276, 205) and three used ESI (cases\u0026thinsp;=\u0026thinsp;78,292). Differences were found in the proportion of cases across all triage categories (MTS vs ESI): category one (1.0% vs 1.5%), two (9.8% vs 18.9%), three (38.5% vs 39.5%), four (39.0% vs 24.0%) and five (3.4% vs 4.1%). Mixed-effects multivariable regression analyses indicated that in hospitals using MTS it was 3.0 times more likely to be admitted as an inpatient in the urgent triage categories, one, two and three combined, than in the less-urgent categories four and five combined. In hospitals using ESI, patients were 11.6 times more likely to be admitted as an inpatient in triage categories one, two and three than in categories four and five.\u003c/p\u003e\u003ch2\u003eConclusion:\u003c/h2\u003e \u003cp\u003eDistinct differences in the distribution of triage categories were observed in the analysis of routine data from EDs using either MTS or ESI. Hospitals utilising MTS appeared more likely than those using ESI to triage patients into the less-urgent triage categories and had a higher probability of hospital admissions than ESI hospitals in these less-urgent categories. While the results of this study could have also been affected by differences in the underlying patient populations between hospitals applying MTS or ESI, the findings hint towards systematic differences in the clinical outcomes of patients triaged using either system and raise questions concerning the accuracy of triage categorization.\u003c/p\u003e","manuscriptTitle":"Emergency Severity Index versus Manchester Triage System: Utilising Routine Data to Compare the Characteristics, Diagnoses and Admission Status of Patients Triaged in 12 German Emergency Departments","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-26 17:17:04","doi":"10.21203/rs.3.rs-9171354/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-29T20:14:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"28374885257722935265736725183961318629","date":"2026-04-19T15:03:32+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-17T14:35:28+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-20T08:52:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-20T08:52:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal of Emergency Medicine","date":"2026-03-19T15:36:29+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-emergency-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijem","sideBox":"Learn more about [International Journal of Emergency Medicine](https://intjem.biomedcentral.com/)","snPcode":"12245","submissionUrl":"https://submission.nature.com/new-submission/12245/3","title":"International Journal of Emergency Medicine","twitterHandle":"@IntJEmergMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"01a3c1a1-0379-4ab1-bc62-b67feeabc022","owner":[],"postedDate":"April 26th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-04-29T20:14:39+00:00","index":30,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-26T17:17:04+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-26 17:17:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9171354","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9171354","identity":"rs-9171354","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}