Emergency call triage priority predicts transport decisions and patient disposition in the London Ambulance Service

Emergency call triage priority predicts transport decisions and patient disposition in the London Ambulance Service | 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 call triage priority predicts transport decisions and patient disposition in the London Ambulance Service Konstantinos Papakonstantinou This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8314214/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background This study examines the link between emergency call triage priority and patient outcomes within the London Ambulance Service (LAS), with a focus on how triage decisions affect paramedic treatment and transport choices. As the world’s busiest ambulance service, LAS uses the Advanced Medical Priority Dispatch System under the Ambulance Response Programme to categorise incidents by urgency and allocate resources accordingly. Despite these structured frameworks, concerns remain about the accuracy of triage classifications and their impact on patient care, particularly regarding the risks of over- and under-triage. Methods Using 2023 data, this retrospective cohort study addresses three key questions: whether higher call priorities are linked to increased emergency transport rates; whether lower priorities correspond to more frequent community discharges; and whether paramedic medication administration alters these relationships. The analysis considers transport decisions, destination types (such as emergency departments, specialist units, or community care), and paramedic interventions as proxies for patient acuity. Results Results show that higher triage priorities are generally associated with greater likelihood of emergency transport and need for interventions, while lower priorities more often result in community discharge or referral to alternative care pathways. However, notable mismatches persist; some low-priority cases require urgent intervention, and some high-priority dispatches do not result in transport. Medication administration by paramedics emerged as a significant modifier, often indicating greater clinical severity and influencing transport decisions regardless of initial triage category. Conclusions The findings suggest that while current triage systems broadly align resources with patient needs, mismatches remain that can affect both patient safety and resource efficiency. These insights underscore the need for ongoing refinement of triage protocols and decision-support tools to optimise emergency medical service response and improve patient outcomes in complex urban settings like London. ambulance triage EMS call priority call triage resource allocation Figures Figure 1 Introduction Emergency medical services (EMS) play a pivotal role in delivering acute care to patients in prehospital settings, where rapid decision-making is essential and often guided by limited information. Central to this process is the triage system—a structured method for determining the urgency of each case and allocating resources accordingly. In the United Kingdom, ambulance services rely on established protocols such as the Advanced Medical Priority Dispatch System (AMPDS) and NHS Pathways, which function under the broader framework of the Ambulance Response Programme (ARP) ( 1 ). These systems aim to match clinical need with appropriate responses, yet growing evidence highlights persistent discrepancies between triage classification and patient outcomes. Recent years have seen escalating demand on ambulance services, coupled with increasing scrutiny of resource utilisation and patient flow efficiency. The London Ambulance Service (LAS), the busiest emergency medical service in the world, embodies these challenges at scale. Serving over 8 million residents and responding to nearly two million calls annually, LAS employs ARP categories ranging from immediately life-threatening (Category 1) to non-urgent (Category 5) ( 2 , 3 ). However, concerns persist regarding both over-triage—where patients receive higher priority than clinically warranted—and under-triage, which risks delaying critical interventions. Previous studies have examined the accuracy of dispatch systems in identifying serious conditions, but few have investigated how triage priority correlates with downstream clinical decisions, such as transport modality, destination type, or paramedic intervention ( 4 – 6 ). Furthermore, little is known about how pharmacological treatment administered in the field may influence these relationships. This study aims to address these knowledge gaps by analysing over 850,000 LAS emergency calls from 2023. Specifically, it investigates three primary questions: ( 1 ) whether higher triage categories are associated with increased use of emergency transport; ( 2 ) whether lower triage categories correspond to higher rates of community discharge; and ( 3 ) whether medication administration by paramedics modifies these associations. By examining these interactions, the study seeks to evaluate the real-world effectiveness of current triage protocols and inform future refinements in EMS decision-making and resource allocation. In doing so, it contributes novel insights to the ongoing discourse on triage precision, patient safety, and the optimisation of prehospital care systems in urban environments. Background Ambulance services are under growing pressure to provide timely and effective emergency care in increasingly complex environments. At the core of their function lies the triage process—a mechanism designed to prioritise patient care based on perceived urgency. To manage high call volumes and diverse patient presentations, dispatch centres around the world employ structured triage protocols to categorise emergency calls. The two dominant paradigms used in prehospital settings are structured algorithmic models like the AMPDS and NHS Pathways, and more flexible criteria-based systems that allow greater discretion from experienced call handlers ( 4 ). AMPDS is globally standardised and widely used in countries including the United States, Canada, the UK, and Australia. It employs a scripted series of questions based on the caller's chief complaint to generate a prioritisation code. These codes (Alpha to Echo) are mapped to levels of urgency and associated response protocols. The rigid structure of AMPDS promotes consistency and is particularly useful in environments with high call-taker turnover or limited clinical training. However, this rigidity may sacrifice sensitivity to nuance and clinical judgment. In contrast, NHS Pathways incorporates symptom-driven algorithms and real-time clinical decision support to guide non-clinical call-handlers through the triage process ( 7 ). It is used by several NHS ambulance services and NHS 111, making it integral to emergency and urgent care coordination in England ( 8 ). Despite being robust and widely adopted, both systems are far from perfect. Evidence suggests that NHS Pathways identifies approximately 76% of cardiac arrests from call data, a sensitivity rate comparable to AMPDS ( 8 ). Yet this also means that nearly one in four cases are under-triaged—missed or misclassified as less urgent than they are. This issue is not unique to cardiac arrests; similar performance gaps exist in identifying strokes, sepsis, and other time-critical conditions ( 6 , 9 ). The challenge is balancing over-triage, which burdens resources, with under-triage, which jeopardises patient safety. In practice, dispatch systems tend to err on the side of caution, accepting high rates of over-triage to ensure critical patients are not missed. However, the operational consequences are severe: increased ambulance workload, delayed responses for genuinely urgent cases, and higher stress on emergency departments. To improve triage accuracy, the UK introduced the ARP in 2017. This initiative restructured triage categories and response time targets, aiming to align ambulance dispatch with clinical need rather than speed alone. Crucially, ARP allowed dispatchers up to 240 seconds to triage lower-priority cases, promoting better differentiation between urgent and non-urgent calls ( 1 ). While this change improved flexibility, its real-world effect on patient outcomes remains debated. Studies report persistent over-triage, particularly within Category 2, the largest and most heterogeneous group ( 5 , 10 , 11 ). This mismatch reflects limitations in the dispatch algorithms and the difficulty of interpreting urgency through phone-based symptom descriptions alone. One of the most persistent themes in the literature is the misalignment between the triage priority assigned at the call-taking stage and the actual patient condition as assessed by ambulance clinicians. For instance, a Dutch trauma study reported that 17.7% of high-priority dispatches were downgraded on scene, while 2.3% were upgraded due to underestimated severity ( 12 ). Similar findings from Finland show that only 29% of dispatches labelled “urgent” were confirmed as such by EMS providers, underscoring the system-wide challenge of predictive accuracy ( 13 ). These findings raise critical questions about the downstream effects of misclassification. High rates of over-triage not only strain ambulance availability but also contribute to unnecessary hospital admissions. Conversely, under-triage risks leaving patients with time-sensitive conditions untreated during crucial early windows. A 2022 UK study of COVID-19-related calls found that 3.5% of patients deemed not to require an ambulance experienced serious adverse outcomes (death or organ failure) within 30 days ( 5 ). Once on scene, paramedics make transport decisions based on a combination of clinical, logistical, and systemic factors. Medication administration is often used as a surrogate marker for acuity. Patients receiving interventions such as oxygen, analgesics, or intravenous drugs are more likely to be transported to hospital or specialist centres ( 14 ). However, the presence of pharmacological treatment is not always indicative of conveyance. An increasing number of patients receive care on scene and are referred to community services or advised on self-care. Several other factors influence transport decisions: patient preference, time of day, ED crowding, distance to hospital, paramedic confidence, and fear of litigation. In some cases, the default inclination is to transport “just to be safe,” especially if clinical uncertainty is high or there is a lack of clear guidance on alternatives ( 15 ). Moreover, regional differences in the availability of alternative pathways—such as urgent care centres, mental health teams, or specialist referral hubs—can lead to wide variation in non-conveyance rates. In England, these rates range from 30% to over 50% depending on the service, with no consistent explanation across trusts ( 16 ). Methods Study Design This study employed a retrospective observational design to investigate the relationship between emergency call triage priority and patient outcomes in the LAS. Specifically, it assessed how initial triage classifications influenced subsequent transport decisions and examined the modifying role of medication administration by ambulance clinicians. This design was chosen for its capacity to analyse real-world operational data without introducing experimental interference or behavioural bias, allowing for the evaluation of established EMS processes across a large dataset ( 17 , 18 ). Data Source and Collection Data were obtained from LAS through a Freedom of Information (FOI) request and encompassed all 999 emergency calls received between January 1 and December 31, 2023. The dataset included triage categories, transport modes, medication administration records, and conveyance destinations. In accordance with UK data protection legislation, all data were anonymised prior to release, with no patient-identifiable information included ( 19 , 20 ). The dataset was provided in a fully de-identified, aggregate format to ensure compliance with the General Data Protection Regulation (GDPR) and the Data Protection Act 2018. Due to the dataset’s size (over 850,000 records), additional measures were required to resolve technical challenges during transfer and processing. After initial release issues, the LAS approved a secure external file-sharing solution, and the final dataset was obtained in January 2025. Ethical Considerations This evaluation was classified as a service evaluation under the UK Health Research Authority guidelines, as it involved the analysis of routinely collected, anonymised operational data and did not require changes to patient care or direct patient contact. Accordingly, formal NHS Research Ethics Committee (REC) approval was not required. Data access was granted under LAS’s internal governance framework, and all principles of data minimisation, confidentiality, and responsible use were strictly observed ( 20 ). Variables and Operational Definitions The primary independent variable was the triage priority assigned at the time of the emergency call, classified using ARP categories ranging from Category 1 (life-threatening emergencies) to Category 5 (non-urgent issues). The main outcome variable was the mode of transport, grouped into three categories: ( 1 ) emergency transport (using lights and sirens), ( 2 ) routine or non-emergency transport, and ( 3 ) non-conveyance (including on-scene discharge or referral to alternative services). Secondary variables included (a) the type of receiving destination—differentiated between high-acuity centres (e.g., major trauma units, hyperacute stroke units) and admission-avoidance facilities (e.g., urgent care centres, day clinics)—and (b) whether paramedics administered any medication during the encounter. Medication use was analysed as a binary variable (administered vs. not administered). Data Analysis Data were cleaned and prepared for statistical analysis using IBM SPSS Statistics (Version 29.0.2.0). Descriptive statistics were used to summarise categorical variables, including frequency distributions for triage priority, transport mode, hospital type, and medication administration. Chi-square tests of independence were employed to examine associations between triage priority and transport mode, as well as between triage priority and destination type. Significance was set at p < 0.05. Where appropriate, listwise deletion was used to handle missing data, which represented less than 2% of the dataset ( 21 ). This analytical approach enabled an evaluation of how closely triage classifications aligned with observed clinical outcomes and transport decisions. Additionally, interaction effects involving medication administration were explored to assess whether this factor modified the relationship between triage priority and transport or destination type. Results Triage Priority and Emergency Transport Higher triage priority at the point of call dispatch was strongly associated with increased rates of emergency transport to hospital. Calls classified as Category 1 or 2, denoting the highest levels of urgency, resulted in hospital transport in most cases. Specifically, transport rates exceeded 70% for Category 1 and Category 2 (C1: 74.08%, n = 100230 and C2: 72.72%, n = 383442). In contrast, lower-priority calls, in Categories 3 and 4, were associated with significantly lower transport rates, with fewer than 6% resulting in emergency conveyance (C3: 5.9%, n = 7634 and C4: 1.2%, n = 55). Despite this overarching trend, notable exceptions were observed. A small but clinically significant proportion of Category 3 calls required urgent transport following paramedic assessment on scene, suggesting potential under-triage at the dispatch stage. Conversely, some Category 1 and 2 calls did not result in transport. In such cases, patient refusal, resolution of symptoms on scene, or the selection of alternative care pathways were frequently cited as reasons for non-conveyance. Triage Priority and Community Discharge Lower triage priority was also associated with higher rates of community-based management. Approximately 40% of Category 3 calls were resolved on scene, either through discharge or referral to community or primary care services (see Table). These findings align with the expectation that lower priority calls often involve conditions of lesser acuity. In contrast, patients classified under higher triage categories were far less likely to be managed in the community. Community discharge rates for Category 1 and 2 calls were lower, reflecting the anticipated clinical severity and the corresponding need for further in-hospital evaluation. However, some higher-priority dispatches—particularly those in Category 2—did not result in hospital transport. These cases often involved patient preference for remaining at home or clinical reassessment by attending paramedics that led to a downgrading of urgency. Impact of Medication Administration The administration of medication by paramedics emerged as a significant modifier in the relationship between triage priority and patient disposition. Across all categories, patients who received medication on scene were substantially more likely to be transported to hospital. In lower-priority cases, the use of medication frequently coincided with a decision to escalate care, suggesting that clinical findings during face-to-face assessment occasionally revealed more severe pathology than initially assumed during the call triage process. Among higher-priority calls, medication administration was highly associated with emergency transport to hospital, reinforcing its role as a marker of clinical severity. These findings highlight that while triage categorisation offers an initial framework for resource allocation and response, paramedic-led clinical interventions—particularly pharmacological treatment—can significantly influence the ultimate patient disposition. Discrepancies and Mismatches Despite the overall alignment between triage priority and patient outcome, the data revealed important mismatches. Namely, some low-priority calls required urgent intervention and hospital transport, highlighting the risk of under-triage. Conversely, a proportion of high-priority dispatches resulted in non-transport, often due to over-triage or resolution of symptoms prior to ambulance arrival. These discrepancies underscore the limitations of telephone-based triage and the critical role of paramedic assessment in determining the final patient disposition. Table 1 Key Findings by Triage Category Triage Category Emergency Transport Rate Community Discharge Rate Medication as Modifier in Emergency Transports Medication as Modifier in Routine Transports Category 1 22.7% 25.8% 78.7% of emergency transport patients received medications 36% of patients conveyed received medications Category 2 15.2% 27.1% 65.3% of emergency transport patients received medications 29.9% of patients conveyed received medications Category 3 5.3% 43.1% 62.8% of emergency transport patients received medications 16.6% of patients conveyed received medications Category 4 1.1% 21.1% 67.3% of emergency transport patients received medications 0.98% of patients conveyed received medications The results demonstrate that LAS triage system generally succeeds in aligning resources with patient need, but notable mismatches persist. Medication administration by paramedics demonstrates notable alignment with the triage priority, although its utilisation requires further research. These findings support the need for ongoing refinement of triage protocols and enhanced decision-support tools to further optimise emergency medical service delivery in complex urban environments. Discussion The present study analysed over 850,000 emergency call records from the LAS to investigate the alignment between initial triage category, transport mode, medication administration, and patient disposition. The results highlight a significant degree of inconsistency between triage priority and downstream clinical decisions, offering important insights into the performance and limitations of current emergency medical dispatch systems. This section critically evaluates those findings, situates them within existing literature, and explores implications for EMS policy, clinical safety, and resource optimisation. One of the study’s central findings is the high prevalence of emergency transport among lower-priority triage categories, particularly Category 3 and 4 calls. This phenomenon reflects a broader issue identified in the literature: the systemic tendency towards over-triage in prehospital care. Over-triage—dispatching high-priority responses to patients with low actual acuity—has been observed in multiple jurisdictions. A Swiss study reported an over-triage rate of 78% for dispatches with lights and sirens ( 22 ), while similar patterns were confirmed in Swedish ( 23 ) and Dutch systems ( 24 ). This high rate of over-triage may be driven by several factors. Firstly, existing triage algorithms, such as AMPDS and NHS Pathways, prioritise sensitivity over specificity, seeking to avoid under-triage of critical cases. While this precautionary approach may be ethically justified, it results in unnecessary resource deployment, prolonged ED wait times, and reduced ambulance availability for genuine emergencies ( 6 ). In the LAS data, the tendency for Category 3 and 4 calls to be transported under emergency conditions suggests that either initial triage underestimated urgency, or that paramedics reassessed the patient on scene and overrode dispatch decisions; both of which point to structural inefficiencies. The relationship between medication administration and transport mode was also notable. Patients who received prehospital medications were significantly more likely to be conveyed via emergency transport and taken to high-acuity destinations. This finding aligns with literature suggesting that pharmacological intervention correlates with higher clinical severity ( 14 ). It further supports the use of medication administration as a proxy for patient acuity in retrospective studies where direct clinical data may be unavailable. However, medication data must be interpreted with caution. For instance, analgesics such as paracetamol or low-dose opioids may be administered for patient comfort rather than clinical necessity. Conversely, some high-risk patients, such as those in sepsis, may not receive medication until arrival at hospital due to prehospital time constraints. Thus, while useful as a heuristic, medication use cannot replace comprehensive physiological or diagnostic data in risk stratification. The study also identified substantial rates of non-conveyance in higher triage categories, especially Category 2. Approximately one in five patients initially triaged as requiring urgent response were ultimately discharged on scene. This raises concerns about “category drift”—the reclassification of urgency during the care pathway—which may reflect inaccuracies in initial triage. These findings are consistent with other UK-based studies. Marincowitz et al. found that 16% of patients advised to self-care by NHS Pathways experienced serious adverse outcomes within 30 days, highlighting the potential risks of mis-triage ( 5 ). Similarly, Farhat et al. observed that non-conveyance decisions were often shaped by patient refusal rather than clinical stability, emphasising the need for shared decision-making frameworks ( 15 ). Moreover, the literature suggests that triage errors are not random but systematically associated with certain patient groups. Elderly patients, those with vague symptoms, or with comorbidities are particularly prone to under-triage ( 25 , 26 ). Over-triage, conversely, is more common in young, otherwise healthy individuals reporting chest pain or trauma symptoms without physiological instability. The performance of NHS Pathways and AMPDS, as reflected in this study, echoes broader concerns about the accuracy of algorithm-driven triage. The NHS Pathways system has demonstrated moderate sensitivity for high-acuity conditions like cardiac arrest (76%) but limited specificity, leading to a significant proportion of low-risk patients receiving high-priority responses ( 7 ). In international comparisons, other triage systems show similar trade-offs. In the Netherlands, the Field Triage Decision Scheme achieved only 64.5% sensitivity for detecting serious trauma in children, with a 16.3% under-triage rate ( 24 ). Meanwhile, Criteria-Based Dispatch systems have shown over-triage rates exceeding 70%, calling into question their operational efficiency ( 22 ). Importantly, studies suggest that the lack of a universal gold standard for defining “correct triage” complicates efforts to benchmark system performance. Outcomes are influenced by context, comorbidities, and available alternatives to hospital transport, limiting the utility of crude metrics such as conveyance or admission alone ( 27 ). One promising avenue for improving triage accuracy is the integration of clinical decision support systems (CDSS) and machine learning (ML) tools. These systems can analyse call data, patient history, and environmental context to generate probabilistic risk scores in real time. Studies demonstrate that ML-enhanced triage models outperform conventional algorithms in both sensitivity and specificity, particularly for time-critical conditions such as sepsis and myocardial infarction ( 28 , 29 ). The use of such technologies could mitigate the current binary nature of dispatch decisions and allow more nuanced risk stratification. However, this approach also requires caution. Without transparent design and continuous validation, CDSS may encode existing biases or fail in edge cases, such as rare presentations or atypical demographics ( 26 ). Another strategy gaining attention is secondary triage; reassessing low-acuity cases after initial classification to divert suitable patients to non-ambulance services. Evidence from Eastwood et al. suggests that this approach can safely reduce ambulance dispatches by up to 50% without increasing adverse outcomes ( 30 ). However, the success of such systems depends on the availability of community-based alternatives and public willingness to engage with them. In the LAS context, the use of urgent care centres, mental health crisis teams, and virtual wards remains underutilised. As the study findings indicate, even patients triaged as non-urgent are often transported to hospital, reflecting limited access to or trust in alternative care routes. The study’s results have several operational and policy implications. Firstly, the observed misalignment between triage and outcome highlights the need for feedback mechanisms linking prehospital data with hospital outcomes. Real-time outcome tracking would enable refinement of triage protocols and improve accountability. Secondly, the evidence supports expanded training for call handlers and paramedics in advanced decision-making, especially regarding discharge and non-conveyance protocols. Moreover, the findings advocate for the inclusion of structured risk scores—such as NEWS2 or PRIEST—into dispatch protocols. These scores have shown superior predictive value compared to symptom-based triage in studies of COVID-19 and other acute presentations ( 5 ). Finally, reforming performance metrics is essential. Current targets, such as response time adherence, incentivise high-priority dispatch regardless of clinical appropriateness. A shift towards outcome-based indicators—such as safe non-conveyance, reduced ED crowding, and patient satisfaction—would better reflect system quality and safety. While the present study benefits from a large sample size and operational relevance, it has limitations. Most notably, the absence of linked hospital data restricts outcome assessment to proxy indicators such as transport and medication. Future studies should aim to integrate prehospital and hospital datasets, enabling more definitive conclusions on triage validity and patient outcomes. In addition, qualitative studies exploring the decision-making processes of paramedics and dispatchers could offer valuable insights into behavioural factors influencing triage divergence. Similarly, randomised controlled trials testing revised triage algorithms, or AI-supported dispatch, would provide higher-level evidence to support system redesign. Conclusion This study provides a comprehensive analysis of the relationship between emergency call triage priority and patient outcomes within the LAS, the world’s busiest ambulance service. By retrospectively examining data, the research assessed whether triage priority accurately predicts the need for emergency transport, impacts community discharge rates, and is influenced by medication administration by paramedics. The findings demonstrate that higher triage priorities are generally associated with increased rates of emergency transport and advanced interventions, reflecting greater clinical acuity. Conversely, lower priority calls more frequently result in community discharge or referral to alternative care pathways, aligning with the intended function of structured triage systems. However, the study also identifies significant discrepancies; some low-priority cases required urgent intervention, while certain high-priority dispatches did not result in transport. Medication administration emerged as a modifier, often indicating higher clinical severity and influencing transport decisions independently of initial triage categorisation. These mismatches suggest that while current triage frameworks broadly succeed in aligning resources with patient need, they are not infallible and can impact both patient safety and system efficiency. The results underscore the complexity of prehospital decision making, shaped by clinical protocols, patient preferences, and operational pressures. They highlight the ongoing need for refinement of triage algorithms and decision-support tools to improve the accuracy of prioritisation and resource allocation. Ultimately, this research advocates for continuous evaluation and adaptation of triage systems to better match the dynamic realities of urban EMS, thereby optimising patient outcomes and enhancing the resilience of emergency medical services. Declarations Ethics, Consent to Participate, and Consent to Publish declarations: Not applicable – NHS Health Research Authority deem this study as service evaluation thus have waived the consent to participate and ethical approval requirements. Clinical Trial number: Not applicable. Funding declaration: No funding was received for the study. Competing interests: The author declares that they have no competing interests. Authors' contributions: Single author paper with KP as the sole contributor. The datasets generated and analysed during the current study are available from the corresponding author on reasonable request. The original dataset is publicly accessible online: https://www.whatdotheyknow.com/request/ambulance_call_triage_and_outcom_2/response/2848177/attach/2/FOI%206328.xlsx Acknowledgements: The author wishes to recognise and thank Prof. Jo Roislien for his expert input to the methodology selection and statistical analysis of this project. References Turner J, Jacques R. Ambulance Response Programme Review [Internet]. Sheffield; 2018 May [cited 2025 Mar 10]. Available from: London Ambulance Service. About us- London Ambulance Service [Internet]. 2025 [cited 2025 Mar 10]. Available from: https://www.londonambulance.nhs.uk/about-us/#:~:text=We%20are%20the%20world’s%20busiest,most%20dynamic%20and%20diverse%20cities London Ambulance Service. London Ambulance Service – Integrated Performance Report [Internet]. 2023 [cited 2024 Mar 19]. Available from: https://www.londonambulance.nhs.uk/wp-content/uploads/2022/03/IPR-1.pdf Bohm K, Kurland L. The accuracy of medical dispatch - a systematic review. Scand J Trauma Resusc Emerg Med. 2018;26(1):94. Marincowitz C, Stone T, Hasan M, Campbell R, Bath PA, Turner J, et al. Accuracy of emergency medical service telephone triage of need for an ambulance response in suspected COVID-19: an observational cohort study. BMJ Open. 2022;12(5):e058628. 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. Deakin CD, England S, Diffey D. Ambulance telephone triage using ‘NHS Pathways’ to identify adult cardiac arrest. Heart. 2017;103(10):738. Deakin CD, England S, Diffey D, Maconochie I. Can ambulance telephone triage using NHS Pathways accurately identify paediatric cardiac arrest? Resuscitation. 2017;116:109–12. Deakin CD, Alasaad M, King P, Thompson F. Is ambulance telephone triage using advanced medical priority dispatch protocols able to identify patients with acute stroke correctly? Emerg Med J. 2009;26(6):442–5. Green J, Ewings S, Wortham R, Walsh B. PP13 The NHS pathways (NHSP) medical call triage system and new ‘nature of call’ telephone screening tool, employed prior to NHSP, can accurately identify high acuity patients: analysis of sensitivity and specificity using routine ambulance service data. Emergency Medicine Journal. 2019;36(1):e5.3-e6. Belcher J, Finn J, Whiteside A, Ball S. Association between initial presenting level of consciousness and patient acuity – A potential application for secondary triage in emergency ambulance calls. Australas Emerg Care. 2023;26(3):199–204. Waalwijk JF, Lokerman RD, van der Sluijs R, Fiddelers AAA, Leenen LPH, van Heijl M, et al. Priority accuracy by dispatch centers and Emergency Medical Services professionals in trauma patients: a cohort study. Eur J Trauma Emerg Surg. 2022;48(2):1111–20. Salminen T, Kaartinen K, Roos M, Vaajanen V, Ekstrand A, Setälä P et al. Differences between the dispatch priority assessments of emergency medical dispatchers and emergency medical services: a prospective register-based study in Finland. Scand J Trauma Resusc Emerg Med [Internet]. 2023 Dec 1 [cited 2025 Apr 15];31(1):1–10. Available from: https://sjtrem.biomedcentral.com/articles/ 10.1186/s13049-023-01072-2 Villani M, Nehme E, Cox S, Anderson D, Reinders N, Nehme Z. Outcomes of adult patients discharged at scene by emergency medical services. Emergency Medicine Journal [Internet]. 2024 Aug 1 [cited 2025 May 9];41(8):459–67. Available from: https://emj.bmj.com/content/41/8/459 Farhat H, Aifa K, El, Alinier G, Nashwan A, Gangaram P, Khadhraoui M, et al. Understanding patient non-transport decision theories in the pre-hospital setting: a narrative review. Int J Emerg Med. 2023;16(1):69. Coster J, O’Cathain A, Jacques R, Crum A, Siriwardena AN, Turner J. Outcomes for Patients Who Contact the Emergency Ambulance Service and Are Not Transported to the Emergency Department: A Data Linkage Study. Prehospital Emergency Care [Internet]. 2019 Jul 4 [cited 2025 Apr 15];23(4):566–77. Available from: https://www.tandfonline.com/doi/abs/ 10.1080/10903127.2018.1549628 Barnish MS, Turner S. The value of pragmatic and observational studies in health care and public health. Pragmat Obs Res [Internet]. 2017 May [cited 2025 May 9];Volume 8:49–55. Available from: https://www.tandfonline.com/doi/pdf/10.2147/POR.S137701 Black N. Why we need observational studies to evaluate the effectiveness of health care. BMJ [Internet]. 1996 May 11 [cited 2025 May 9];312(7040):1215–8. Available from: https://www.bmj.com/content/312/7040/1215 UK Government. Freedom of Information Act [Internet]. London. 2000. Available from: https://www.legislation.gov.uk/ukpga/2000/36/contents Crook MA. The Caldicott report and patient confidentiality. J Clin Pathol. 2003;56(–6):426–8. Stavseth MR, Clausen T, Røislien J. How handling missing data may impact conclusions: A comparison of six different imputation methods for categorical questionnaire data. SAGE Open Med [Internet]. 2019 Jan 1 [cited 2025 May 9];7. Available from: https://pubmed.ncbi.nlm.nih.gov/30671242/ Dami F, Golay C, Pasquier M, Fuchs V, Carron PN, Hugli O. Prehospital triage accuracy in a criteria based dispatch centre. BMC Emerg Med. 2015;15(1):32. Khorram-Manesh A, Montán KL, Hedelin A, Kihlgren M, Örtenwall P. Prehospital triage, discrepancy in priority-setting between emergency medical dispatch centre and ambulance crews. European Journal of Trauma and Emergency Surgery [Internet]. 2011 Feb 1 [cited 2025 May 28];37(1):73–8. Available from: https://pubmed.ncbi.nlm.nih.gov/26814754/ van der Sluijs R, Lokerman RD, Waalwijk JF, de Jongh MAC, Edwards MJR, den Hartog D et al. Accuracy of pre-hospital trauma triage and field triage decision rules in children (P2-T2 study): an observational study. Lancet Child Adolesc Health [Internet]. 2020 Apr 1 [cited 2025 May 28];4(4):290–8. Available from: https://www.thelancet.com/action/showFullText?pii=S2352464219304316 Rashed EA, Kodera S, Shirakami H, Kawaguchi R, Watanabe K, Hirata A. Knowledge discovery from emergency ambulance dispatch during COVID-19: A case study of Nagoya City, Japan. J Biomed Inform [Internet]. 2021 May 1 [cited 2025 May 30];117:103743. Available from: https://www.sciencedirect.com/science/article/pii/S1532046421000721 Stone EL. Clinical Decision Support Systems in the Emergency Department: Opportunities to Improve Triage Accuracy. J Emerg Nurs [Internet]. 2019 Mar 1 [cited 2025 May 30];45(2):220–2. Available from: https://www.jenonline.org/action/showFullText?pii=S0099176718307244 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 [Internet]. 2019 Jul 1 [cited 2025 May 28];74(1):140–52. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0196064418312824 Alotaibi A, Body R, Carley S, Pennington E. Towards enhanced telephone triage for chest pain: a Delphi study to define life-threatening conditions that must be identified. BMC Emerg Med [Internet]. 2021 Dec 1 [cited 2025 Apr 15];21(1):158. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC8672334/ Menshawi AM, Hassan MM. A novel triage framework for emergency department based on machine learning paradigm. Expert Syst [Internet]. 2025 Feb 1 [cited 2025 May 30];42(2):e13735. Available from: /doi/pdf/10.1111/exsy.13735 Eastwood K, Morgans A, Smith K, Hodgkinson A, Becker G, Stoelwinder J. A novel approach for managing the growing demand for ambulance services by low-acuity patients. Aust Health Rev. 2016;40(4):378. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-8314214","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":569036075,"identity":"97dcf5e0-e5c8-473d-b04e-6121924dae64","order_by":0,"name":"Konstantinos Papakonstantinou","email":"data:image/png;base64,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","orcid":"","institution":"University of Stavanger (Stavanger","correspondingAuthor":true,"prefix":"","firstName":"Konstantinos","middleName":"","lastName":"Papakonstantinou","suffix":""}],"badges":[],"createdAt":"2025-12-09 07:08:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8314214/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8314214/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":99507657,"identity":"9b0301fa-a87e-4adf-97ec-5268af6d68c8","added_by":"auto","created_at":"2026-01-05 08:55:21","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":297929,"visible":true,"origin":"","legend":"","description":"","filename":"Articlev3.docx","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/87670cd8d2a1d6de67fe618c.docx"},{"id":99507615,"identity":"76e064ec-131f-4465-a9e7-c2a5df6a5e89","added_by":"auto","created_at":"2026-01-05 08:55:15","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4419,"visible":true,"origin":"","legend":"","description":"","filename":"7152903afcc148458df7e362b9a9c60c.json","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/f531031db29ea4c5837ed49f.json"},{"id":99507652,"identity":"cd89bf7d-8fa0-4748-870c-ece9ac4bc03d","added_by":"auto","created_at":"2026-01-05 08:55:21","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":81131,"visible":true,"origin":"","legend":"","description":"","filename":"7152903afcc148458df7e362b9a9c60c1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/cadaa010dabad4fedfb3ab93.xml"},{"id":99507625,"identity":"6023fc89-90f4-42df-9354-c780baac2d74","added_by":"auto","created_at":"2026-01-05 08:55:20","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31198,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/6aeb98293707a269c2447435.png"},{"id":99507619,"identity":"132cf5f4-a323-433b-95c8-5c535b042546","added_by":"auto","created_at":"2026-01-05 08:55:16","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9133,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/3ca8f41523affd49f67680fc.png"},{"id":99507658,"identity":"a9027b57-107f-485f-b16f-7ad3779b4839","added_by":"auto","created_at":"2026-01-05 08:55:21","extension":"xml","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79402,"visible":true,"origin":"","legend":"","description":"","filename":"7152903afcc148458df7e362b9a9c60c1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/e1ccb516b4493f576e2f43a8.xml"},{"id":99507612,"identity":"71162a2c-84a0-4a32-a690-cfdc55cc6084","added_by":"auto","created_at":"2026-01-05 08:55:13","extension":"html","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87700,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/b9b605320fde25eaf6be118b.html"},{"id":99507672,"identity":"77de428e-4734-4725-891a-48d0b4751304","added_by":"auto","created_at":"2026-01-05 08:55:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":42981,"visible":true,"origin":"","legend":"\u003cp\u003eTransport mode overlay on triage priority\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/223b720ca94bd18306d22c41.png"},{"id":104429944,"identity":"5807711f-011c-45fe-a41e-8e52ccd4a380","added_by":"auto","created_at":"2026-03-11 15:27:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":471059,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8314214/v1/61f215d0-dfa7-4901-a6a9-e739252c6a04.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Emergency call triage priority predicts transport decisions and patient disposition in the London Ambulance Service","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEmergency medical services (EMS) play a pivotal role in delivering acute care to patients in prehospital settings, where rapid decision-making is essential and often guided by limited information. Central to this process is the triage system\u0026mdash;a structured method for determining the urgency of each case and allocating resources accordingly. In the United Kingdom, ambulance services rely on established protocols such as the Advanced Medical Priority Dispatch System (AMPDS) and NHS Pathways, which function under the broader framework of the Ambulance Response Programme (ARP) (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). These systems aim to match clinical need with appropriate responses, yet growing evidence highlights persistent discrepancies between triage classification and patient outcomes. Recent years have seen escalating demand on ambulance services, coupled with increasing scrutiny of resource utilisation and patient flow efficiency. The London Ambulance Service (LAS), the busiest emergency medical service in the world, embodies these challenges at scale. Serving over 8\u0026nbsp;million residents and responding to nearly two million calls annually, LAS employs ARP categories ranging from immediately life-threatening (Category 1) to non-urgent (Category 5) (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). However, concerns persist regarding both over-triage\u0026mdash;where patients receive higher priority than clinically warranted\u0026mdash;and under-triage, which risks delaying critical interventions. Previous studies have examined the accuracy of dispatch systems in identifying serious conditions, but few have investigated how triage priority correlates with downstream clinical decisions, such as transport modality, destination type, or paramedic intervention (\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Furthermore, little is known about how pharmacological treatment administered in the field may influence these relationships.\u003c/p\u003e\u003cp\u003eThis study aims to address these knowledge gaps by analysing over 850,000 LAS emergency calls from 2023. Specifically, it investigates three primary questions: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) whether higher triage categories are associated with increased use of emergency transport; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) whether lower triage categories correspond to higher rates of community discharge; and (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) whether medication administration by paramedics modifies these associations. By examining these interactions, the study seeks to evaluate the real-world effectiveness of current triage protocols and inform future refinements in EMS decision-making and resource allocation. In doing so, it contributes novel insights to the ongoing discourse on triage precision, patient safety, and the optimisation of prehospital care systems in urban environments.\u003c/p\u003e"},{"header":"Background","content":"\u003cp\u003eAmbulance services are under growing pressure to provide timely and effective emergency care in increasingly complex environments. At the core of their function lies the triage process—a mechanism designed to prioritise patient care based on perceived urgency. To manage high call volumes and diverse patient presentations, dispatch centres around the world employ structured triage protocols to categorise emergency calls. The two dominant paradigms used in prehospital settings are structured algorithmic models like the AMPDS and NHS Pathways, and more flexible criteria-based systems that allow greater discretion from experienced call handlers (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAMPDS is globally standardised and widely used in countries including the United States, Canada, the UK, and Australia. It employs a scripted series of questions based on the caller's chief complaint to generate a prioritisation code. These codes (Alpha to Echo) are mapped to levels of urgency and associated response protocols. The rigid structure of AMPDS promotes consistency and is particularly useful in environments with high call-taker turnover or limited clinical training. However, this rigidity may sacrifice sensitivity to nuance and clinical judgment. In contrast, NHS Pathways incorporates symptom-driven algorithms and real-time clinical decision support to guide non-clinical call-handlers through the triage process (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). It is used by several NHS ambulance services and NHS 111, making it integral to emergency and urgent care coordination in England (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Despite being robust and widely adopted, both systems are far from perfect. Evidence suggests that NHS Pathways identifies approximately 76% of cardiac arrests from call data, a sensitivity rate comparable to AMPDS (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Yet this also means that nearly one in four cases are under-triaged—missed or misclassified as less urgent than they are. This issue is not unique to cardiac arrests; similar performance gaps exist in identifying strokes, sepsis, and other time-critical conditions (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). The challenge is balancing over-triage, which burdens resources, with under-triage, which jeopardises patient safety. In practice, dispatch systems tend to err on the side of caution, accepting high rates of over-triage to ensure critical patients are not missed. However, the operational consequences are severe: increased ambulance workload, delayed responses for genuinely urgent cases, and higher stress on emergency departments.\u003c/p\u003e \u003cp\u003eTo improve triage accuracy, the UK introduced the ARP in 2017. This initiative restructured triage categories and response time targets, aiming to align ambulance dispatch with clinical need rather than speed alone. Crucially, ARP allowed dispatchers up to 240 seconds to triage lower-priority cases, promoting better differentiation between urgent and non-urgent calls (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). While this change improved flexibility, its real-world effect on patient outcomes remains debated. Studies report persistent over-triage, particularly within Category 2, the largest and most heterogeneous group (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). This mismatch reflects limitations in the dispatch algorithms and the difficulty of interpreting urgency through phone-based symptom descriptions alone.\u003c/p\u003e \u003cp\u003eOne of the most persistent themes in the literature is the misalignment between the triage priority assigned at the call-taking stage and the actual patient condition as assessed by ambulance clinicians. For instance, a Dutch trauma study reported that 17.7% of high-priority dispatches were downgraded on scene, while 2.3% were upgraded due to underestimated severity (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Similar findings from Finland show that only 29% of dispatches labelled “urgent” were confirmed as such by EMS providers, underscoring the system-wide challenge of predictive accuracy (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). These findings raise critical questions about the downstream effects of misclassification. High rates of over-triage not only strain ambulance availability but also contribute to unnecessary hospital admissions. Conversely, under-triage risks leaving patients with time-sensitive conditions untreated during crucial early windows. A 2022 UK study of COVID-19-related calls found that 3.5% of patients deemed not to require an ambulance experienced serious adverse outcomes (death or organ failure) within 30 days (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOnce on scene, paramedics make transport decisions based on a combination of clinical, logistical, and systemic factors. Medication administration is often used as a surrogate marker for acuity. Patients receiving interventions such as oxygen, analgesics, or intravenous drugs are more likely to be transported to hospital or specialist centres (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). However, the presence of pharmacological treatment is not always indicative of conveyance. An increasing number of patients receive care on scene and are referred to community services or advised on self-care. Several other factors influence transport decisions: patient preference, time of day, ED crowding, distance to hospital, paramedic confidence, and fear of litigation. In some cases, the default inclination is to transport “just to be safe,” especially if clinical uncertainty is high or there is a lack of clear guidance on alternatives (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Moreover, regional differences in the availability of alternative pathways—such as urgent care centres, mental health teams, or specialist referral hubs—can lead to wide variation in non-conveyance rates. In England, these rates range from 30% to over 50% depending on the service, with no consistent explanation across trusts (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e).\u003c/p\u003e "},{"header":"Methods","content":"\u003cp\u003eStudy Design\u003c/p\u003e\u003cp\u003eThis study employed a retrospective observational design to investigate the relationship between emergency call triage priority and patient outcomes in the LAS. Specifically, it assessed how initial triage classifications influenced subsequent transport decisions and examined the modifying role of medication administration by ambulance clinicians. This design was chosen for its capacity to analyse real-world operational data without introducing experimental interference or behavioural bias, allowing for the evaluation of established EMS processes across a large dataset (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eData Source and Collection\u003c/p\u003e\u003cp\u003eData were obtained from LAS through a Freedom of Information (FOI) request and encompassed all 999 emergency calls received between January 1 and December 31, 2023. The dataset included triage categories, transport modes, medication administration records, and conveyance destinations. In accordance with UK data protection legislation, all data were anonymised prior to release, with no patient-identifiable information included (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). The dataset was provided in a fully de-identified, aggregate format to ensure compliance with the General Data Protection Regulation (GDPR) and the Data Protection Act 2018. Due to the dataset’s size (over 850,000 records), additional measures were required to resolve technical challenges during transfer and processing. After initial release issues, the LAS approved a secure external file-sharing solution, and the final dataset was obtained in January 2025.\u003c/p\u003e\u003cp\u003eEthical Considerations\u003c/p\u003e\u003cp\u003eThis evaluation was classified as a service evaluation under the UK Health Research Authority guidelines, as it involved the analysis of routinely collected, anonymised operational data and did not require changes to patient care or direct patient contact. Accordingly, formal NHS Research Ethics Committee (REC) approval was not required. Data access was granted under LAS’s internal governance framework, and all principles of data minimisation, confidentiality, and responsible use were strictly observed (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eVariables and Operational Definitions\u003c/p\u003e\u003cp\u003eThe primary independent variable was the triage priority assigned at the time of the emergency call, classified using ARP categories ranging from Category 1 (life-threatening emergencies) to Category 5 (non-urgent issues). The main outcome variable was the mode of transport, grouped into three categories: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) emergency transport (using lights and sirens), (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) routine or non-emergency transport, and (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) non-conveyance (including on-scene discharge or referral to alternative services). Secondary variables included (a) the type of receiving destination—differentiated between high-acuity centres (e.g., major trauma units, hyperacute stroke units) and admission-avoidance facilities (e.g., urgent care centres, day clinics)—and (b) whether paramedics administered any medication during the encounter. Medication use was analysed as a binary variable (administered vs. not administered).\u003c/p\u003e\u003ch2\u003eData Analysis\u003c/h2\u003e\u003cp\u003eData were cleaned and prepared for statistical analysis using IBM SPSS Statistics (Version 29.0.2.0). Descriptive statistics were used to summarise categorical variables, including frequency distributions for triage priority, transport mode, hospital type, and medication administration. Chi-square tests of independence were employed to examine associations between triage priority and transport mode, as well as between triage priority and destination type. Significance was set at p \u0026lt; 0.05. Where appropriate, listwise deletion was used to handle missing data, which represented less than 2% of the dataset (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). This analytical approach enabled an evaluation of how closely triage classifications aligned with observed clinical outcomes and transport decisions. Additionally, interaction effects involving medication administration were explored to assess whether this factor modified the relationship between triage priority and transport or destination type.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eTriage Priority and Emergency Transport\u003c/p\u003e \u003cp\u003eHigher triage priority at the point of call dispatch was strongly associated with increased rates of emergency transport to hospital. Calls classified as Category 1 or 2, denoting the highest levels of urgency, resulted in hospital transport in most cases. Specifically, transport rates exceeded 70% for Category 1 and Category 2 (C1: 74.08%, n\u0026thinsp;=\u0026thinsp;100230 and C2: 72.72%, n\u0026thinsp;=\u0026thinsp;383442). In contrast, lower-priority calls, in Categories 3 and 4, were associated with significantly lower transport rates, with fewer than 6% resulting in emergency conveyance (C3: 5.9%, n\u0026thinsp;=\u0026thinsp;7634 and C4: 1.2%, n\u0026thinsp;=\u0026thinsp;55). Despite this overarching trend, notable exceptions were observed. A small but clinically significant proportion of Category 3 calls required urgent transport following paramedic assessment on scene, suggesting potential under-triage at the dispatch stage. Conversely, some Category 1 and 2 calls did not result in transport. In such cases, patient refusal, resolution of symptoms on scene, or the selection of alternative care pathways were frequently cited as reasons for non-conveyance.\u003c/p\u003e \u003cp\u003eTriage Priority and Community Discharge\u003c/p\u003e \u003cp\u003eLower triage priority was also associated with higher rates of community-based management. Approximately 40% of Category 3 calls were resolved on scene, either through discharge or referral to community or primary care services (see Table). These findings align with the expectation that lower priority calls often involve conditions of lesser acuity. In contrast, patients classified under higher triage categories were far less likely to be managed in the community. Community discharge rates for Category 1 and 2 calls were lower, reflecting the anticipated clinical severity and the corresponding need for further in-hospital evaluation. However, some higher-priority dispatches\u0026mdash;particularly those in Category 2\u0026mdash;did not result in hospital transport. These cases often involved patient preference for remaining at home or clinical reassessment by attending paramedics that led to a downgrading of urgency.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eImpact of Medication Administration\u003c/p\u003e \u003cp\u003eThe administration of medication by paramedics emerged as a significant modifier in the relationship between triage priority and patient disposition. Across all categories, patients who received medication on scene were substantially more likely to be transported to hospital. In lower-priority cases, the use of medication frequently coincided with a decision to escalate care, suggesting that clinical findings during face-to-face assessment occasionally revealed more severe pathology than initially assumed during the call triage process. Among higher-priority calls, medication administration was highly associated with emergency transport to hospital, reinforcing its role as a marker of clinical severity. These findings highlight that while triage categorisation offers an initial framework for resource allocation and response, paramedic-led clinical interventions\u0026mdash;particularly pharmacological treatment\u0026mdash;can significantly influence the ultimate patient disposition.\u003c/p\u003e \u003cp\u003eDiscrepancies and Mismatches\u003c/p\u003e \u003cp\u003eDespite the overall alignment between triage priority and patient outcome, the data revealed important mismatches. Namely, some low-priority calls required urgent intervention and hospital transport, highlighting the risk of under-triage. Conversely, a proportion of high-priority dispatches resulted in non-transport, often due to over-triage or resolution of symptoms prior to ambulance arrival. These discrepancies underscore the limitations of telephone-based triage and the critical role of paramedic assessment in determining the final patient disposition.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eKey Findings by Triage Category\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eTriage Category\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEmergency Transport Rate\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCommunity Discharge Rate\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eMedication as Modifier in Emergency Transports\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eMedication as Modifier in Routine Transports\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCategory 1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e22.7%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003e25.8%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003e78.7% of emergency transport patients received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003e36% of patients conveyed received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCategory 2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e15.2%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003e27.1%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003e65.3% of emergency transport patients received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003e29.9% of patients conveyed received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCategory 3\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e5.3%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003e43.1%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003e62.8% of emergency transport patients received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003e16.6% of patients conveyed received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCategory 4\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e1.1%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003e21.1%\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003e67.3% of emergency transport patients received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003e0.98% of patients conveyed received medications\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe results demonstrate that LAS triage system generally succeeds in aligning resources with patient need, but notable mismatches persist. Medication administration by paramedics demonstrates notable alignment with the triage priority, although its utilisation requires further research. These findings support the need for ongoing refinement of triage protocols and enhanced decision-support tools to further optimise emergency medical service delivery in complex urban environments.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study analysed over 850,000 emergency call records from the LAS to investigate the alignment between initial triage category, transport mode, medication administration, and patient disposition. The results highlight a significant degree of inconsistency between triage priority and downstream clinical decisions, offering important insights into the performance and limitations of current emergency medical dispatch systems. This section critically evaluates those findings, situates them within existing literature, and explores implications for EMS policy, clinical safety, and resource optimisation.\u003c/p\u003e \u003cp\u003eOne of the study\u0026rsquo;s central findings is the high prevalence of emergency transport among lower-priority triage categories, particularly Category 3 and 4 calls. This phenomenon reflects a broader issue identified in the literature: the systemic tendency towards over-triage in prehospital care. Over-triage\u0026mdash;dispatching high-priority responses to patients with low actual acuity\u0026mdash;has been observed in multiple jurisdictions. A Swiss study reported an over-triage rate of 78% for dispatches with lights and sirens (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e), while similar patterns were confirmed in Swedish (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e) and Dutch systems (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). This high rate of over-triage may be driven by several factors. Firstly, existing triage algorithms, such as AMPDS and NHS Pathways, prioritise sensitivity over specificity, seeking to avoid under-triage of critical cases. While this precautionary approach may be ethically justified, it results in unnecessary resource deployment, prolonged ED wait times, and reduced ambulance availability for genuine emergencies (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). In the LAS data, the tendency for Category 3 and 4 calls to be transported under emergency conditions suggests that either initial triage underestimated urgency, or that paramedics reassessed the patient on scene and overrode dispatch decisions; both of which point to structural inefficiencies.\u003c/p\u003e \u003cp\u003eThe relationship between medication administration and transport mode was also notable. Patients who received prehospital medications were significantly more likely to be conveyed via emergency transport and taken to high-acuity destinations. This finding aligns with literature suggesting that pharmacological intervention correlates with higher clinical severity (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). It further supports the use of medication administration as a proxy for patient acuity in retrospective studies where direct clinical data may be unavailable. However, medication data must be interpreted with caution. For instance, analgesics such as paracetamol or low-dose opioids may be administered for patient comfort rather than clinical necessity. Conversely, some high-risk patients, such as those in sepsis, may not receive medication until arrival at hospital due to prehospital time constraints. Thus, while useful as a heuristic, medication use cannot replace comprehensive physiological or diagnostic data in risk stratification.\u003c/p\u003e \u003cp\u003eThe study also identified substantial rates of non-conveyance in higher triage categories, especially Category 2. Approximately one in five patients initially triaged as requiring urgent response were ultimately discharged on scene. This raises concerns about \u0026ldquo;category drift\u0026rdquo;\u0026mdash;the reclassification of urgency during the care pathway\u0026mdash;which may reflect inaccuracies in initial triage. These findings are consistent with other UK-based studies. Marincowitz et al. found that 16% of patients advised to self-care by NHS Pathways experienced serious adverse outcomes within 30 days, highlighting the potential risks of mis-triage (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Similarly, Farhat et al. observed that non-conveyance decisions were often shaped by patient refusal rather than clinical stability, emphasising the need for shared decision-making frameworks (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Moreover, the literature suggests that triage errors are not random but systematically associated with certain patient groups. Elderly patients, those with vague symptoms, or with comorbidities are particularly prone to under-triage (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). Over-triage, conversely, is more common in young, otherwise healthy individuals reporting chest pain or trauma symptoms without physiological instability.\u003c/p\u003e \u003cp\u003eThe performance of NHS Pathways and AMPDS, as reflected in this study, echoes broader concerns about the accuracy of algorithm-driven triage. The NHS Pathways system has demonstrated moderate sensitivity for high-acuity conditions like cardiac arrest (76%) but limited specificity, leading to a significant proportion of low-risk patients receiving high-priority responses (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). In international comparisons, other triage systems show similar trade-offs. In the Netherlands, the Field Triage Decision Scheme achieved only 64.5% sensitivity for detecting serious trauma in children, with a 16.3% under-triage rate (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). Meanwhile, Criteria-Based Dispatch systems have shown over-triage rates exceeding 70%, calling into question their operational efficiency (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). Importantly, studies suggest that the lack of a universal gold standard for defining \u0026ldquo;correct triage\u0026rdquo; complicates efforts to benchmark system performance. Outcomes are influenced by context, comorbidities, and available alternatives to hospital transport, limiting the utility of crude metrics such as conveyance or admission alone (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOne promising avenue for improving triage accuracy is the integration of clinical decision support systems (CDSS) and machine learning (ML) tools. These systems can analyse call data, patient history, and environmental context to generate probabilistic risk scores in real time. Studies demonstrate that ML-enhanced triage models outperform conventional algorithms in both sensitivity and specificity, particularly for time-critical conditions such as sepsis and myocardial infarction (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe use of such technologies could mitigate the current binary nature of dispatch decisions and allow more nuanced risk stratification. However, this approach also requires caution. Without transparent design and continuous validation, CDSS may encode existing biases or fail in edge cases, such as rare presentations or atypical demographics (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAnother strategy gaining attention is secondary triage; reassessing low-acuity cases after initial classification to divert suitable patients to non-ambulance services. Evidence from Eastwood et al. suggests that this approach can safely reduce ambulance dispatches by up to 50% without increasing adverse outcomes (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). However, the success of such systems depends on the availability of community-based alternatives and public willingness to engage with them. In the LAS context, the use of urgent care centres, mental health crisis teams, and virtual wards remains underutilised. As the study findings indicate, even patients triaged as non-urgent are often transported to hospital, reflecting limited access to or trust in alternative care routes.\u003c/p\u003e \u003cp\u003eThe study\u0026rsquo;s results have several operational and policy implications. Firstly, the observed misalignment between triage and outcome highlights the need for feedback mechanisms linking prehospital data with hospital outcomes. Real-time outcome tracking would enable refinement of triage protocols and improve accountability. Secondly, the evidence supports expanded training for call handlers and paramedics in advanced decision-making, especially regarding discharge and non-conveyance protocols. Moreover, the findings advocate for the inclusion of structured risk scores\u0026mdash;such as NEWS2 or PRIEST\u0026mdash;into dispatch protocols. These scores have shown superior predictive value compared to symptom-based triage in studies of COVID-19 and other acute presentations (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Finally, reforming performance metrics is essential. Current targets, such as response time adherence, incentivise high-priority dispatch regardless of clinical appropriateness. A shift towards outcome-based indicators\u0026mdash;such as safe non-conveyance, reduced ED crowding, and patient satisfaction\u0026mdash;would better reflect system quality and safety.\u003c/p\u003e \u003cp\u003eWhile the present study benefits from a large sample size and operational relevance, it has limitations. Most notably, the absence of linked hospital data restricts outcome assessment to proxy indicators such as transport and medication. Future studies should aim to integrate prehospital and hospital datasets, enabling more definitive conclusions on triage validity and patient outcomes. In addition, qualitative studies exploring the decision-making processes of paramedics and dispatchers could offer valuable insights into behavioural factors influencing triage divergence. Similarly, randomised controlled trials testing revised triage algorithms, or AI-supported dispatch, would provide higher-level evidence to support system redesign.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study provides a comprehensive analysis of the relationship between emergency call triage priority and patient outcomes within the LAS, the world\u0026rsquo;s busiest ambulance service. By retrospectively examining data, the research assessed whether triage priority accurately predicts the need for emergency transport, impacts community discharge rates, and is influenced by medication administration by paramedics. The findings demonstrate that higher triage priorities are generally associated with increased rates of emergency transport and advanced interventions, reflecting greater clinical acuity. Conversely, lower priority calls more frequently result in community discharge or referral to alternative care pathways, aligning with the intended function of structured triage systems.\u003c/p\u003e \u003cp\u003eHowever, the study also identifies significant discrepancies; some low-priority cases required urgent intervention, while certain high-priority dispatches did not result in transport. Medication administration emerged as a modifier, often indicating higher clinical severity and influencing transport decisions independently of initial triage categorisation. These mismatches suggest that while current triage frameworks broadly succeed in aligning resources with patient need, they are not infallible and can impact both patient safety and system efficiency.\u003c/p\u003e \u003cp\u003eThe results underscore the complexity of prehospital decision making, shaped by clinical protocols, patient preferences, and operational pressures. They highlight the ongoing need for refinement of triage algorithms and decision-support tools to improve the accuracy of prioritisation and resource allocation. Ultimately, this research advocates for continuous evaluation and adaptation of triage systems to better match the dynamic realities of urban EMS, thereby optimising patient outcomes and enhancing the resilience of emergency medical services.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics, Consent to Participate, and Consent to Publish declarations: Not applicable – NHS Health Research Authority deem this study as service evaluation thus have waived the consent to participate and ethical approval requirements.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eClinical Trial number: Not applicable.\u003c/p\u003e\n\u003cp\u003eFunding declaration: No funding was received for the study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCompeting interests: The author declares that they have no competing interests.\u003c/p\u003e\n\u003cp\u003eAuthors' contributions: Single author paper with KP as the sole contributor.\u003c/p\u003e\n\u003cp\u003eThe datasets generated and analysed during the current study are available from the corresponding author on reasonable request. The original dataset is publicly accessible online: https://www.whatdotheyknow.com/request/ambulance_call_triage_and_outcom_2/response/2848177/attach/2/FOI%206328.xlsx\u003c/p\u003e\n\u003cp\u003eAcknowledgements: The author wishes to recognise and thank Prof. Jo Roislien for his expert input to the methodology selection and statistical analysis of this project.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eTurner J, Jacques R. Ambulance Response Programme Review [Internet]. Sheffield; 2018 May [cited 2025 Mar 10]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003c/span\u003e\u003cspan address=\"http://www.england.nhs.uk/wp-content/uploads/2018/10/ambulance-response-programme-review.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLondon Ambulance Service. About us- London Ambulance Service [Internet]. 2025 [cited 2025 Mar 10]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.londonambulance.nhs.uk/about-us/#:~:text=We%20are%20the%20world\u0026rsquo;s%20busiest,most%20dynamic%20and%20diverse%20cities\u003c/span\u003e\u003cspan address=\"https://www.londonambulance.nhs.uk/about-us/#:~:text=We%20are%20the%20world\u0026rsquo;s%20busiest,most%20dynamic%20and%20diverse%20cities\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLondon Ambulance Service. London Ambulance Service \u0026ndash; Integrated Performance Report [Internet]. 2023 [cited 2024 Mar 19]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.londonambulance.nhs.uk/wp-content/uploads/2022/03/IPR-1.pdf\u003c/span\u003e\u003cspan address=\"https://www.londonambulance.nhs.uk/wp-content/uploads/2022/03/IPR-1.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBohm K, Kurland L. The accuracy of medical dispatch - a systematic review. Scand J Trauma Resusc Emerg Med. 2018;26(1):94.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarincowitz C, Stone T, Hasan M, Campbell R, Bath PA, Turner J, et al. Accuracy of emergency medical service telephone triage of need for an ambulance response in suspected COVID-19: an observational cohort study. BMJ Open. 2022;12(5):e058628.\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\u003eDeakin CD, England S, Diffey D. Ambulance telephone triage using \u0026lsquo;NHS Pathways\u0026rsquo; to identify adult cardiac arrest. Heart. 2017;103(10):738.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeakin CD, England S, Diffey D, Maconochie I. Can ambulance telephone triage using NHS Pathways accurately identify paediatric cardiac arrest? Resuscitation. 2017;116:109\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeakin CD, Alasaad M, King P, Thompson F. Is ambulance telephone triage using advanced medical priority dispatch protocols able to identify patients with acute stroke correctly? Emerg Med J. 2009;26(6):442\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreen J, Ewings S, Wortham R, Walsh B. PP13 The NHS pathways (NHSP) medical call triage system and new \u0026lsquo;nature of call\u0026rsquo; telephone screening tool, employed prior to NHSP, can accurately identify high acuity patients: analysis of sensitivity and specificity using routine ambulance service data. Emergency Medicine Journal. 2019;36(1):e5.3-e6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBelcher J, Finn J, Whiteside A, Ball S. Association between initial presenting level of consciousness and patient acuity \u0026ndash; A potential application for secondary triage in emergency ambulance calls. Australas Emerg Care. 2023;26(3):199\u0026ndash;204.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWaalwijk JF, Lokerman RD, van der Sluijs R, Fiddelers AAA, Leenen LPH, van Heijl M, et al. Priority accuracy by dispatch centers and Emergency Medical Services professionals in trauma patients: a cohort study. Eur J Trauma Emerg Surg. 2022;48(2):1111\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalminen T, Kaartinen K, Roos M, Vaajanen V, Ekstrand A, Set\u0026auml;l\u0026auml; P et al. Differences between the dispatch priority assessments of emergency medical dispatchers and emergency medical services: a prospective register-based study in Finland. Scand J Trauma Resusc Emerg Med [Internet]. 2023 Dec 1 [cited 2025 Apr 15];31(1):1\u0026ndash;10. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://sjtrem.biomedcentral.com/articles/\u003c/span\u003e\u003cspan address=\"https://sjtrem.biomedcentral.com/articles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13049-023-01072-2\u003c/span\u003e\u003cspan address=\"10.1186/s13049-023-01072-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVillani M, Nehme E, Cox S, Anderson D, Reinders N, Nehme Z. Outcomes of adult patients discharged at scene by emergency medical services. Emergency Medicine Journal [Internet]. 2024 Aug 1 [cited 2025 May 9];41(8):459\u0026ndash;67. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://emj.bmj.com/content/41/8/459\u003c/span\u003e\u003cspan address=\"https://emj.bmj.com/content/41/8/459\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarhat H, Aifa K, El, Alinier G, Nashwan A, Gangaram P, Khadhraoui M, et al. Understanding patient non-transport decision theories in the pre-hospital setting: a narrative review. Int J Emerg Med. 2023;16(1):69.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoster J, O\u0026rsquo;Cathain A, Jacques R, Crum A, Siriwardena AN, Turner J. Outcomes for Patients Who Contact the Emergency Ambulance Service and Are Not Transported to the Emergency Department: A Data Linkage Study. Prehospital Emergency Care [Internet]. 2019 Jul 4 [cited 2025 Apr 15];23(4):566\u0026ndash;77. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.tandfonline.com/doi/abs/\u003c/span\u003e\u003cspan address=\"https://www.tandfonline.com/doi/abs/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/10903127.2018.1549628\u003c/span\u003e\u003cspan address=\"10.1080/10903127.2018.1549628\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarnish MS, Turner S. The value of pragmatic and observational studies in health care and public health. Pragmat Obs Res [Internet]. 2017 May [cited 2025 May 9];Volume 8:49\u0026ndash;55. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.tandfonline.com/doi/pdf/10.2147/POR.S137701\u003c/span\u003e\u003cspan address=\"https://www.tandfonline.com/doi/pdf/10.2147/POR.S137701\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlack N. Why we need observational studies to evaluate the effectiveness of health care. BMJ [Internet]. 1996 May 11 [cited 2025 May 9];312(7040):1215\u0026ndash;8. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.bmj.com/content/312/7040/1215\u003c/span\u003e\u003cspan address=\"https://www.bmj.com/content/312/7040/1215\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUK Government. Freedom of Information Act [Internet]. London. 2000. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.legislation.gov.uk/ukpga/2000/36/contents\u003c/span\u003e\u003cspan address=\"https://www.legislation.gov.uk/ukpga/2000/36/contents\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrook MA. The Caldicott report and patient confidentiality. J Clin Pathol. 2003;56(\u0026ndash;6):426\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStavseth MR, Clausen T, R\u0026oslash;islien J. How handling missing data may impact conclusions: A comparison of six different imputation methods for categorical questionnaire data. SAGE Open Med [Internet]. 2019 Jan 1 [cited 2025 May 9];7. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubmed.ncbi.nlm.nih.gov/30671242/\u003c/span\u003e\u003cspan address=\"https://pubmed.ncbi.nlm.nih.gov/30671242/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDami F, Golay C, Pasquier M, Fuchs V, Carron PN, Hugli O. Prehospital triage accuracy in a criteria based dispatch centre. BMC Emerg Med. 2015;15(1):32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhorram-Manesh A, Mont\u0026aacute;n KL, Hedelin A, Kihlgren M, \u0026Ouml;rtenwall P. Prehospital triage, discrepancy in priority-setting between emergency medical dispatch centre and ambulance crews. European Journal of Trauma and Emergency Surgery [Internet]. 2011 Feb 1 [cited 2025 May 28];37(1):73\u0026ndash;8. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubmed.ncbi.nlm.nih.gov/26814754/\u003c/span\u003e\u003cspan address=\"https://pubmed.ncbi.nlm.nih.gov/26814754/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Sluijs R, Lokerman RD, Waalwijk JF, de Jongh MAC, Edwards MJR, den Hartog D et al. Accuracy of pre-hospital trauma triage and field triage decision rules in children (P2-T2 study): an observational study. Lancet Child Adolesc Health [Internet]. 2020 Apr 1 [cited 2025 May 28];4(4):290\u0026ndash;8. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.thelancet.com/action/showFullText?pii=S2352464219304316\u003c/span\u003e\u003cspan address=\"https://www.thelancet.com/action/showFullText?pii=S2352464219304316\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRashed EA, Kodera S, Shirakami H, Kawaguchi R, Watanabe K, Hirata A. Knowledge discovery from emergency ambulance dispatch during COVID-19: A case study of Nagoya City, Japan. J Biomed Inform [Internet]. 2021 May 1 [cited 2025 May 30];117:103743. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.sciencedirect.com/science/article/pii/S1532046421000721\u003c/span\u003e\u003cspan address=\"https://www.sciencedirect.com/science/article/pii/S1532046421000721\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStone EL. Clinical Decision Support Systems in the Emergency Department: Opportunities to Improve Triage Accuracy. J Emerg Nurs [Internet]. 2019 Mar 1 [cited 2025 May 30];45(2):220\u0026ndash;2. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.jenonline.org/action/showFullText?pii=S0099176718307244\u003c/span\u003e\u003cspan address=\"https://www.jenonline.org/action/showFullText?pii=S0099176718307244\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\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 [Internet]. 2019 Jul 1 [cited 2025 May 28];74(1):140\u0026ndash;52. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://linkinghub.elsevier.com/retrieve/pii/S0196064418312824\u003c/span\u003e\u003cspan address=\"https://linkinghub.elsevier.com/retrieve/pii/S0196064418312824\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlotaibi A, Body R, Carley S, Pennington E. Towards enhanced telephone triage for chest pain: a Delphi study to define life-threatening conditions that must be identified. BMC Emerg Med [Internet]. 2021 Dec 1 [cited 2025 Apr 15];21(1):158. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pmc.ncbi.nlm.nih.gov/articles/PMC8672334/\u003c/span\u003e\u003cspan address=\"https://pmc.ncbi.nlm.nih.gov/articles/PMC8672334/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMenshawi AM, Hassan MM. A novel triage framework for emergency department based on machine learning paradigm. Expert Syst [Internet]. 2025 Feb 1 [cited 2025 May 30];42(2):e13735. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e/doi/pdf/10.1111/exsy.13735\u003c/span\u003e\u003cspan address=\"/doi/pdf/10.1111/exsy.13735\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEastwood K, Morgans A, Smith K, Hodgkinson A, Becker G, Stoelwinder J. A novel approach for managing the growing demand for ambulance services by low-acuity patients. Aust Health Rev. 2016;40(4):378.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"ambulance triage, EMS, call priority, call triage, resource allocation","lastPublishedDoi":"10.21203/rs.3.rs-8314214/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8314214/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThis study examines the link between emergency call triage priority and patient outcomes within the London Ambulance Service (LAS), with a focus on how triage decisions affect paramedic treatment and transport choices. As the world\u0026rsquo;s busiest ambulance service, LAS uses the Advanced Medical Priority Dispatch System under the Ambulance Response Programme to categorise incidents by urgency and allocate resources accordingly. Despite these structured frameworks, concerns remain about the accuracy of triage classifications and their impact on patient care, particularly regarding the risks of over- and under-triage.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eUsing 2023 data, this retrospective cohort study addresses three key questions: whether higher call priorities are linked to increased emergency transport rates; whether lower priorities correspond to more frequent community discharges; and whether paramedic medication administration alters these relationships. The analysis considers transport decisions, destination types (such as emergency departments, specialist units, or community care), and paramedic interventions as proxies for patient acuity.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eResults show that higher triage priorities are generally associated with greater likelihood of emergency transport and need for interventions, while lower priorities more often result in community discharge or referral to alternative care pathways. However, notable mismatches persist; some low-priority cases require urgent intervention, and some high-priority dispatches do not result in transport. Medication administration by paramedics emerged as a significant modifier, often indicating greater clinical severity and influencing transport decisions regardless of initial triage category.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe findings suggest that while current triage systems broadly align resources with patient needs, mismatches remain that can affect both patient safety and resource efficiency. These insights underscore the need for ongoing refinement of triage protocols and decision-support tools to optimise emergency medical service response and improve patient outcomes in complex urban settings like London.\u003c/p\u003e","manuscriptTitle":"Emergency call triage priority predicts transport decisions and patient disposition in the London Ambulance Service","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-05 08:54:24","doi":"10.21203/rs.3.rs-8314214/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8f470fd9-f35f-441c-81cc-3508a593d9f7","owner":[],"postedDate":"January 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-11T15:25:54+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-05 08:54:24","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8314214","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8314214","identity":"rs-8314214","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}