The ACT and the ADJUST framework: Layering structured detection and clinical judgment in emergency department triage

The ACT and the ADJUST framework: Layering structured detection and clinical judgment in emergency department triage | 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 The ACT and the ADJUST framework: Layering structured detection and clinical judgment in emergency department triage Gustav Siqueland, Jostein Singstad, Ingvild Billehaug Norum Viken, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7852065/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background Emergency department (ED) triage must balance rapid detection of critical illness with efficient use of limited resources. We evaluated the Adaptive Clinical Triage (ACT) model—a deliberately lean, nurse-led approach integrating predefined NEWS2 thresholds, symptom-urgency flags, and a discretionary concern override. We also examined ADJUST (Adaptive Judgment of Urgency and Streaming), a semi-structured physician reassessment incorporating early streaming where capacity allows. Methods We performed a retrospective cohort study of 53,645 ED encounters (1 Sept 2023–1 Apr 2025). ACT and ADJUST data were extracted from the electronic trackboard. The primary outcome was intensive care unit (ICU) admission within 24 hours, analyzed as a pragmatic proxy for high acuity. Sensitivity and specificity for ACT were calculated for Red triage assignments and stratified by transfer timing (≤ 2 h, ≤ 4 h, ≤ 6 h). A secondary analysis of paired ACT–ADJUST encounters examined how physician reassessment reclassified acuity. Results Among 46,895 encounters with ACT documentation, sensitivity for ICU admission within 24 hours was 81.2% (95% CI 76.8–85.6) and specificity 85.9% (85.6–86.3). Sensitivity reached 95.2% for ICU transfers within 2 hours and decreased gradually with longer transfer times, indicating strong early detection of critical illness. In 42,881 paired encounters, ACT and ADJUST agreed in 51%. ADJUST improved specificity (96.2%) but reduced sensitivity (64.5%), consistent with its role as a contextual refinement rather than a primary detection layer. Conclusion ACT provides a straightforward, standardized framework that maintains high early sensitivity through structured physiological and symptom-based triggers. ADJUST, when applied, enhances specificity and facilitates early streaming, reflecting its value as a contextual complement to ACT. Together they illustrate how layered triage can combine standardized safety with adaptable clinical judgment. Future studies should examine generalizability across diverse ED environments. emergency triage NEWS2 clinical judgment layered triage patient acuity operational efficiency ICU admission Background Emergency department triage is fundamental to patient safety and operational efficiency, especially amid rising demand and limited capacity. It must simultaneously enable rapid identification of critically ill patients and the efficient coordination of finite resources — two aims that often diverge as patient volumes fluctuate or operational pressures intensify. Widely used triage frameworks such as the Manchester Triage System (MTS), the Canadian Triage and Acuity Scale (CTAS), and the Rapid Emergency Triage and Treatment System (RETTS) offer comprehensive decision structures but rely on branching complaint algorithms and multiple discriminators. While such systems provide detailed guidance, they can be cognitively demanding and time-consuming to apply, particularly in departments with variable staffing, limited training, or high turnover. Their complexity can make them less adaptable in settings that require rapid, consistent decision-making under pressure, contributing to performance variability across studies [ 1 , 2 ]. Adaptive Clinical Triage (ACT) was developed as a streamlined alternative. Nurse-led by design, ACT integrates predefined NEWS2 thresholds, concise symptom-urgency flags, and a discretionary concern override, omitting complex complaint algorithms. The term adaptive reflects its design intent to combine structured thresholds with professional judgment, balancing standardization with flexibility. The model emphasizes early physiological risk detection through a limited set of symptom cues, aiming to be straightforward to learn, consistently applied, and adaptable across settings. It seeks to reduce cognitive load and promote uniformity in triage decisions, particularly in high-pressure environments. Although NEWS2 and its predecessor show good discrimination for mortality and ICU admission in ED cohorts, their role as stand-alone triage tools remains debated [ 3 – 6 ]. ACT operationalizes NEWS2 within a simplified intake structure intended to preserve clinical judgment while reducing procedural complexity. While structured triage primarily targets severity—the immediate physiological threat—emergency care also depends on acuity, a broader construct encompassing disease trajectory, comorbidity, and operational urgency. A secondary layer of physician review can therefore complement structured detection by integrating evolving information. Such reassessment enhances safety by identifying patients needing closer monitoring or escalation and improves efficiency by confirming stability or downgrading when early treatment has restored stability. To address this need, a physician-led reassessment—Adaptive Judgment of Urgency and Streaming (ADJUST)—was developed. Using the same four-tier acuity scale, it situates the initial triage decision within clinical history and risk context, and when capacity allows, may incorporate limited bedside diagnostics such as focused examination or ultrasound, and guide early patient streaming. Each ADJUST assignment is accompanied by a brief free-text rationale, allowing contextual factors and clinical reasoning to be explicitly documented. Designed as a structured yet flexible framework, ADJUST aligns acuity categorization and resource allocation with the evolving clinical picture rather than functioning as a separate triage system. This study primarily evaluates ACT as a stand-alone intake model, using ICU admission within defined time windows as a pragmatic proxy for high acuity. A secondary analysis explores how ADJUST reclassified acuity within the layered model, to illustrate how structured detection and clinical judgment can coexist within a practical, scalable triage framework. Methods Study Design and data sources This was a retrospective observational study of all emergency department (ED) encounters at a regional Norwegian hospital serving a catchment population of approximately 240,000 between 1 September 2023 and 1 April 2025. ACT and ADJUST entries were recorded prospectively in real time within the electronic ED trackboard and subsequently extracted for analysis. Supplementary File 1 provides the full English-language ACT intake criteria used in this study. The complete ACT intake chart, integrating the ABCDE urgency framework within the four-tier acuity scale, is provided in Supplementary File 3. Triage model and training The structure and rationale for ACT and ADJUST have been described above. In brief, Adaptive Clinical Triage (ACT) is a nurse-led intake model that integrates predefined physiological thresholds based on the National Early Warning Score 2 (NEWS2), symptom-urgency flags, and a discretionary concern override. This allows escalation when professional judgment indicates higher risk. Adaptive Judgment of Urgency and Streaming (ADJUST) is a physician-led reassessment using the same four-tier acuity scale (Red, Orange, Yellow, Green). It situates the initial ACT decision within the broader clinical history and risk context and, when capacity allows, incorporates bedside diagnostics such as focused examination, ECG, blood gases, or ultrasound to support early streaming. ACT was implemented as an electronic module within the ED trackboard, supported by a paper template used for training and reference. All ED nurses performed ACT regardless of seniority or prior triage experience. Training initially consisted of case-based sessions and was later standardized through a mandatory 60-minute e-learning module. Newly recruited nurses received supervised practice during orientation, including at least one mentored triage shift. ADJUST documentation was entered directly on the trackboard after initial intake. It represents a concise physician reassessment aimed at refining acuity and urgency detection rather than performing a full diagnostic evaluation. The process varied with workload and provider background. It was most often completed by the supervising physician and routinely performed by a dedicated triage physician during peak hours. In practice, it ranged from brief stability checks by non-emergency physicians to focused assessments that included limited diagnostics and early treatment initiation by emergency physicians. No formal certification was required. Full intake criteria for ACT are provided in Supplementary File 1. The ACT–ADJUST cross-classification matrix used in the secondary analysis is presented in Supplementary File 2, and the paper-based ACT intake form (v1.0), which includes both adult and pediatric criteria, is shown in Supplementary File 3. Eligibility and exclusions All ED encounters during the study period were eligible for inclusion. Encounters without ACT documentation were excluded from primary analyses; these typically reflected deferred triage entry, documentation gaps in high-acuity cases, or incomplete rollout during early implementation. Predefined alert pathways such as stroke, trauma, or resuscitation were retained, although documentation was less consistent when immediate treatment superseded formal triage. The main ED receives all adult patients with medical, surgical, and trauma-related presentations. In contrast, noncritical pediatric medical cases are managed in a dedicated pediatric emergency department and were therefore excluded by design. Minor injuries treated in a separate fast-track pathway were also excluded. Outcome definition The primary outcome was ICU admission within 24 hours of ED arrival, used as a pragmatic proxy for high acuity consistent with prior validation work [ 4 – 6 ]. Secondary analyses examined temporal alignment between triage and deterioration by stratifying ICU transfers within ≤ 2, ≤ 4, and ≤ 6 hours. Only transfers to the central ICU were included; step-down, postoperative, and procedural units were excluded to avoid misclassification of short-term monitoring as critical illness [ 1 , 2 , 8 ]. In triage research, where no definitive gold standard for “true urgency” exists, outcome measures often rely on surrogate markers—a construct-validity approach sometimes referred to as a “silver standard.” Consistent with this framework, ICU admission served as a pragmatic and auditable proxy for high acuity in this study [ 9 ]. Triage categories and analytic approach Both ACT and ADJUST assigned one of four categories (Red, Orange, Yellow, Green). Performance metrics were calculated for the Red category only, reflecting its distinct clinical mandate for immediate physician contact and continuous monitoring, whereas Orange allows up to 30 minutes to physician review with intermittent monitoring. Confidence intervals were estimated using the Wald method. Receiver operating characteristic (ROC) analysis was not performed because triage categories are predefined and ordinal; as noted in prior reviews, AUC metrics provide limited interpretive value for multi-level triage systems [ 1 ]. The main analysis evaluated ACT as a stand-alone intake tool. A secondary paired analysis examined encounters with both ACT and ADJUST documentation to assess reclassification patterns and operational impact ( Supplementary File 2 ). Ethics approval The study was approved by the internal ethics board at Vestfold Hospital Trust (10 December 2020; case number 20/03355). As a retrospective quality-improvement study using de-identified registry data, the requirement for individual informed consent was waived under national regulations. Results Cohort and documentation A total of 53,645 ED encounters were recorded during the 19 month study period. ACT documentation was available for 46,895 encounters (87.5%). Among 409 patients admitted to the ICU within 24 hours (0.8% of all encounters), 298 (72.9%) had complete ACT documentation and were included in the primary analysis. ADJUST was documented in 43,931 encounters (82.0%), of which 42,881 had paired ACT–ADJUST entries used for secondary analysis. Encounters without ACT documentation most commonly represented high-acuity presentations where immediate care superseded electronic entry or cases excluded by design (e.g., fast-track injuries and noncritical medical pediatric cases). ACT performance The sensitivity of ACT (Red category) for ICU admission within 24 hours was 81.2% (242/298; 95% CI ±4.4%), with a specificity of 85.9% (40,026/46,597; ±0.3%). Sensitivity was highest for ICU transfers within the first two hours (95.2% ±3.3%) and decreased gradually with longer transfer times—90.3% (≤ 4 h), 85.3% (≤ 6 h), and 81.2% (≤ 24 h) (Table 1). Table 1. Sensitivity of ACT (Red category) for ICU admission by time to transfer Time to ICU True Positives (TP) False negatives (FN) Sensitivity (95% CI) ≤ 2 hours 158 8 95.2% ± 3.3% ≤ 4 hours 214 23 90.3% ± 3.8% ≤ 6 hours 227 39 85.3% ± 4.3% ≤ 24 hours 242 56 81.2% ± 4.4% Legend: Sensitivity of ACT (Red category) for ICU admission by time to transfer. True positives (TP) and false negatives (FN) shown with sensitivity estimates and 95% confidence intervals. Only ICU transfers within the specified window are included. Abbreviations: TP, true positive; FN, false negative; CI, confidence interval; ICU, intensive care unit; ACT, Adaptive Clinical Triage. Trigger distribution for ACT Red Among 6,812 ACT Red designations, 53.4% were triggered by predefined symptom-urgency flags, 21.8% by NEWS2 ≥ 7, and 7.9% solely by discretionary clinical concern. The remaining cases reflected combinations of triggers (e.g., NEWS + symptom or concern) (Table 2). Table 2. Trigger distribution for ACT Red Trigger type % of Red encounters Symptom-urgency flags 53.4% NEWS2 ≥ 7 21.8% Clinical concern alone 7.9% Combinations (e.g., NEWS2 + symptom/concern) 16.9% Legend: Distribution of triggers for ACT Red assignments. Percentages reflect the proportion of all ACT Red encounters in which the trigger was present. Abbreviations: NEWS, National Early Warning Score; ACT, Adaptive Clinical Triage. Paired ACT-ADJUST encounters In the 42,881 paired encounters, ACT and ADJUST assignments matched in 51.1% of cases. Reclassification occurred in 48.9%, with 45.3% downgraded and 3.6% upgraded by ADJUST. Compared with ACT, ADJUST demonstrated higher specificity (96.2%) but lower sensitivity (64.5%) for ICU admission within 24 hours (table 3). In ICU cases with paired documentation (n = 231), ACT identified 182 (78.8%) and ADJUST identified 149 (64.5%) as Red. Among 38 ICU patients downgraded by ADJUST, 20 (52.6%) were transferred more than four hours after arrival. Table 3. Sensitivity of ADJUST (Red category) for ICU admission by time to transfer Time to ICU True positives (TP) False negatives (FN) Sensitivity (95% CI) ≤ 2 hours 136 14 90.7% ± 4.7% ≤ 4 hours 183 38 82.8% ± 5.0% ≤ 6 hours 188 60 75.8% ± 5.3% ≤ 24 hours 192 88 68.6% ± 5.4% Legend: Sensitivity of ADJUST (Red category) for ICU admission by time to transfer. True positives (TP) and false negatives (FN) shown with sensitivity estimates and 95% confidence intervals. Data limited to encounters with paired ACT–ADJUST documentation. Abbreviations: TP, true positive; FN, false negative; CI, confidence interval; ICU, intensive care unit; ACT, Adaptive Clinical Triage; ADJUST, Adaptive Judgment of Urgency and Streaming. The complete cross-classification matrix for paired ACT–ADJUST encounters is available in Supplementary File 2. Timing and nature of physician reassessment ADJUST documentation occurred within 15 minutes of arrival in 48.6% of paired encounters across all acuity categories. The mean interval between ACT and ADJUST entries was 119 minutes. These timing data encompass the full range of reassessments—from brief stability confirmations to more detailed early evaluations—rather than reflecting a single mode of physician review. Discussion Principal findings In this large single-center evaluation, the Adaptive Clinical Triage (ACT) model—a lean, nurse-led system combining predefined NEWS2 thresholds, symptom-urgency flags, and discretionary concern—achieved high sensitivity for early ICU transfer, particularly within the first two hours after ED arrival. This finding supports ACT’s role as a structured, standardized tool for early detection of critical illness. In contrast, the physician-led reassessment layer (ADJUST) improved specificity and reduced over-triage at the expected cost of sensitivity, reflecting its contextual role as a refinement rather than a replacement for structured triage. Why a layered design? Traditional complaint-based triage systems (MTS, CTAS, RETTS) depend on extensive branching algorithms that can increase cognitive load and operational variability, particularly in departments with fluctuating staffing and experience levels [1,2]. ACT was designed to minimize such variability through a limited number of structured inputs—vital-sign thresholds, key symptom cues, and professional concern—while remaining feasible across diverse ED environments. In this cohort, approximately three-quarters of Red designations arose from structured triggers (NEWS2 ≥ 7 or predefined symptom flags), while only 7.9% relied solely on clinical concern—consistent with ACT’s intent to reduce unwarranted variation while preserving professional discretion. The subsequent physician reassessment introduces a complementary layer of contextual judgment, allowing clinicians to integrate disease trajectory, comorbidity, and evolving risk. Together, the two layers balance standardized detection with clinical reasoning, addressing the inherent tension between patient safety and resource efficiency in triage. Recent work also supports the development of layered, semi-structured frameworks that integrate standardized scoring with clinical judgment [10]. This concept is reflected in findings on the NEWS family of scores, which discriminate well for mortality and ICU outcomes but less reliably for clinician-defined urgency, underscoring the value of pairing physiological scoring with contextual reassessment [6]. Together, these insights highlight that the value of a layered triage model lies not only in its structure but also in how its performance is interpreted within the temporal context of patient outcomes. Temporal alignment and endpoint interpretation Sensitivity declined progressively with longer intervals to ICU transfer, indicating that 24-hour outcomes combine immediate critical illness with later deterioration and operational influences. Stratifying outcomes within 2-, 4-, and 6-hour windows therefore provides a more accurate reflection of triage performance at the time of assessment. Although ICU admission is an imperfect proxy for clinical urgency, it remains a pragmatic and auditable endpoint that correlates with resource intensity and is widely used in triage validation studies [4–6]. Such surrogate outcomes can misclassify patients who improve after early treatment or deteriorate later for unrelated reasons, but they are consistent with a construct-validity approach—using observable, auditable markers such as admission or mortality as practical proxies for high acuity when no definitive gold standard exists [9]. Recognizing these limitations, our stratified, time-based analysis aligns with recent recommendations to evaluate triage using clinically meaningful, time-linked outcomes rather than static disposition measures [1,2,8]. This perspective highlights that triage performance must be interpreted in its temporal and operational context, rather than reduced to final disposition alone. Operational clarity of acuity tiers In this study, the Red category was analyzed separately to respect its distinct mandate for immediate physician contact and continuous monitoring, whereas Orange permits up to 30 minutes to physician review with intermittent monitoring. Pooling these tiers, although common in prior studies, risks obscuring escalation thresholds and blurring distinctions in clinical urgency [1,2,11]. This distinction is particularly relevant in the present model, where the 30-minute limit for the Orange category represents a comparatively broader interval than the corresponding time mandates in systems such as MTS and CTAS. Operational implications The ACT model demonstrates that high early sensitivity can be achieved through a straightforward, low-complexity framework requiring minimal training and infrastructure. For high-volume EDs under crowding pressure, such standardization may reduce variability in initial acuity assignment and help maintain safety despite fluctuating workload. In resource-constrained settings, ACT’s reliance on a small number of observable parameters makes it scalable without digital dependencies. ADJUST, when capacity allows, contributes operational precision through contextual downgrading and early streaming, which may improve resource alignment without compromising safety. Its application depends on staffing and workflow context, highlighting the value of embedding flexible, judgment-based reassessment within structured detection systems. Strengths and limitations This study benefits from a large real world dataset and simultaneous capture of both structured nurse triage and physician reassessment within the same electronic system. Limitations include its single-center design, incomplete documentation in some encounters, and heterogeneity in how ADJUST was applied depending on workload and physician background. The use of ICU admission as a pragmatic outcome measure provides consistency with prior validation studies but does not capture all dimensions of clinical urgency. The optional and variably timed nature of ADJUST also limits conclusions about its full effect on patient outcomes. Because ADJUST relies on contextual clinical reasoning rather than fixed criteria, traditional external validation may not be feasible; future work should instead focus on developing structured guidance to support consistent application across settings. Future directions Prospective multicenter studies could assess inter-rater reliability, calibration across subgroups, and operational outcomes such as time to treatment or boarding delays. Further work should also evaluate how structured guidance for ADJUST influences patient flow and whether specific criteria can help identify when physician reassessment adds the most value. Conclusion ACT provides a standardized, low-complexity framework that maintains high sensitivity for early detection of critical illness. ADJUST enhances specificity and operational alignment when applied, demonstrating how structured detection and clinical judgment can coexist within a scalable triage approach. Together, they offer a practical model adaptable to a range of emergency department contexts—from high-volume hospitals to resource-limited settings. Abbreviations ED Emergency Department ACT Acuity Categorization Tool ADJUST Adaptive Judgment of Urgency and Streaming ICU Intensive Care Unit NEWS National Early Warning Score CI Confidence Interval MTS, CTAS, RETTS Manchester Triage System, Canadian Triage and Acuity Scale, and Rapid Emergency Triage and Treatment System Declarations Ethics approval and consent to participate: The study was conducted in accordance with the ethical standards of the institutional and national research committees and with the 1964 Declaration of Helsinki and its later amendments. Data were obtained from the ED trackboard, an internal quality registry at Vestfold Hospital Trust approved by Sikt (the Norwegian Agency for Shared Services in Education and Research, formerly NSD; reference number 54073, valid through August 2030). The registry is approved for use in quality improvement and system performance evaluation. This study was reviewed by the institutional Data Protection Officer and approved by the Medical Director of Vestfold Hospital Trust as an additional publication purpose within the existing registry approval. Only aggregated, non-identifiable data were analyzed; therefore, individual patient consent was not required. Consent for publication: Not applicable. Availability of data and materials: De-identified datasets are available from the corresponding author upon reasonable request and with institutional approval. Competing interests: The authors declare no competing interests. Funding: Supported internally by Vestfold Hospital Trust. No external funding was received. Authors’ contributions: GS – study design, model development, data analysis, manuscript drafting; VR – study design, interpretation, critical revision; B-JS – data extraction, methodological review; VEH & RR – system integration; IBNV & THL (EM physicians), ES & SJ (ED nurses) – implementation and conceptual development. All authors reviewed and approved the final manuscript. Acknowledgements: The authors thank the triage and emergency department staff at Vestfold Hospital Trust for consistent real time documentation and the clinical informatics team for enabling system-integrated data capture. AI disclosure: Portions of this manuscript were language-edited using ChatGPT (OpenAI). No data analysis, interpretation, or content generation was performed by AI. References 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 May 28;9(5):e026471. https://doi.org/10.1136/bmjopen-2018-026471 Hinson JS, Martinez DA, Cabral S, George K, Whalen M, Hansoti B, Levin S. Triage performance in emergency medicine: a systematic review. Ann Emerg Med . https://doi.org/10.1016/j.annemergmed.2018.09.022 Royal College of Physicians. National Early Warning Score (NEWS) 2: Standardising the assessment of acute-illness severity in the NHS. London: RCP; 2017. https://www.rcp.ac.uk/media/a4ibkkbf/news2-final-report_0_0.pdf Silcock, D. J., Corfield, A. R., Gowens, P. A., & Rooney, K. D. (2015). Validation of the National Early Warning Score in the prehospital setting. Resuscitation , 89 , 31–35. https://doi.org/10.1016/j.resuscitation.2014.12.029 Covino M, Sandroni C, Della Polla D, De Matteis G, Piccioni A, De Vita A, et al. Predicting ICU admission and death in the Emergency Department: A comparison of six early warning scores. Resuscitation . 2023;190:109876. https://doi.org/10.1016/j.resuscitation.2023.109876 Zaboli A, Brigo F, Sibilio S, Massar M, Magnarelli G, Brigiari G, Turcato G. Evaluating the National Early Warning Score (NEWS) in triage: A machine learning perspective. International Emergency Nursing. 2025;80:101602. https://doi.org/10.1016/j.ienj.2025.101602 Solevåg AL, Hinna UT, Eggen EH, Engan M, Gundersen AI, Døllner H. Pediatrisk tidlig varslingsskår, triage og kommunikasjon . Akuttveileder i pediatri, Helsebiblioteket. Updated 01.01.2021. Available at: 1.13 Pediatrisk tidlig varslingsskår, triage og kommunikasjon - Helsebiblioteket Johansson A, Ekwall A, Forberg JL, Ekelund U. Development of outcomes for evaluating emergency care triage: a Delphi approach. Scand J Trauma Resusc Emerg Med . 2023;31(1):10. https://doi.org/10.1186/s13049-023-01073-1 Schinkel M, Bergsma L, Veldhuis LI, et al. Comparing complaint-based triage scales and early warning scores for emergency department triage. Emergency Medicine Journal. 2022;39(9):691–696. https://doi.org/10.1136/emermed-2021-211544 Tsiftsis D, Tasioulis A, Bampalis D. Adult triage in the emergency department: Introducing a multi-layer triage system. Healthcare . 2025;13(9):1070. https://doi.org/10.3390/healthcare13091070 Zachariasse JM, Seiger N, Rood PPM, Alves CF, Freitas P, Smit FJ, et al. Validity of the Manchester Triage System in emergency care: A prospective observational study. PLoS One . 2017;12(2):e0170811. https://doi.org/10.1371/journal.pone.0170811 Additional Declarations No competing interests reported. Supplementary Files SupplementaryFile1CoreIntakeCriteriaforAdaptiveClinicalTriageACT.docx Supplementary files: ● Supplementary File 1 → Core intake criteria (text table). SupplementaryFile2ACTADJUSTCrossClassificationMatrix.docx ● Supplementary File 2 → ACT–ADJUST cross-classification matrix. SupplementaryFile3PaperbasedACTintakeformv1.0.pdf ● Supplementary File 3 → Paper-based ACT intake form (v1.0). Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 27 Mar, 2026 Reviews received at journal 19 Nov, 2025 Reviewers agreed at journal 11 Nov, 2025 Reviewers invited by journal 11 Nov, 2025 Editor assigned by journal 15 Oct, 2025 Submission checks completed at journal 15 Oct, 2025 First submitted to journal 13 Oct, 2025 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-7852065","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":547789032,"identity":"8a6e833d-e093-415e-b9d4-00ffd529b5b8","order_by":0,"name":"Gustav Siqueland","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBElEQVRIiWNgGAWjYDADNhDxgIGBsZ+BsYGQYogKsJYEIG9mA7FaGGBaNhwgoF4+Iv35g59tDPZ87DmGDxJq7shuPn+4TYJxhw1OLYY3cgwbe9sYEtt43hgbJBx7ZrztRiJQy5k03Fpm5DA28JxhSGCTyDGTSGA7nLjtBmOzAWPbYTxa0h82/jnDYA/UYv4j4d/hxM39B/FrkZdIMGzmqWBgbAPaAnTe4cQNDImND/BpMeB5YzhbpkIC6JdnxRKJfYeNZ9wAaklsw+0X+fb0Bx/fGNjYy7cnb/zw4dth2f7+4w8OfGzDHWIGB8CUBAM4UuAgAVMlwpYGopSNglEwCkbBiAYANblZvIVP88sAAAAASUVORK5CYII=","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":true,"prefix":"","firstName":"Gustav","middleName":"","lastName":"Siqueland","suffix":""},{"id":547789033,"identity":"37af989a-fe45-454d-bb07-053f769fa87b","order_by":1,"name":"Jostein Singstad","email":"","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":false,"prefix":"","firstName":"Jostein","middleName":"","lastName":"Singstad","suffix":""},{"id":547789034,"identity":"9ed86277-1009-4f39-a933-1248bde24b0a","order_by":2,"name":"Ingvild Billehaug Norum Viken","email":"","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":false,"prefix":"","firstName":"Ingvild","middleName":"Billehaug Norum","lastName":"Viken","suffix":""},{"id":547789035,"identity":"40377baa-4298-40d5-9dfd-d27aae0df429","order_by":3,"name":"Eilin Solberg","email":"","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":false,"prefix":"","firstName":"Eilin","middleName":"","lastName":"Solberg","suffix":""},{"id":547789036,"identity":"f59b4c12-cf8c-40eb-9975-fea33cbac332","order_by":4,"name":"Siren Jess","email":"","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":false,"prefix":"","firstName":"Siren","middleName":"","lastName":"Jess","suffix":""},{"id":547789037,"identity":"0e78c489-15a9-4ab6-adad-2201c17404a9","order_by":5,"name":"Vetle Ellingsen Hauge","email":"","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":false,"prefix":"","firstName":"Vetle","middleName":"Ellingsen","lastName":"Hauge","suffix":""},{"id":547789038,"identity":"70c6def6-cac8-4e9f-91e8-d6091e38f1b7","order_by":6,"name":"Therese Hamre Leet","email":"","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":false,"prefix":"","firstName":"Therese","middleName":"Hamre","lastName":"Leet","suffix":""},{"id":547789039,"identity":"c592e660-43cd-45dd-a395-f7c0d65d1f63","order_by":7,"name":"Vidar Ruddox","email":"","orcid":"","institution":"Vestfold Hospital Trust","correspondingAuthor":false,"prefix":"","firstName":"Vidar","middleName":"","lastName":"Ruddox","suffix":""}],"badges":[],"createdAt":"2025-10-13 20:08:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7852065/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7852065/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96603262,"identity":"bcd99447-3a2d-4188-80fb-b653f33ab254","added_by":"auto","created_at":"2025-11-24 09:07:51","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":904293,"visible":true,"origin":"","legend":"","description":"","filename":"ACTADJUSTTriageSiquelandBMCEM.docx","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/33c0497d5abe28753ac7e105.docx"},{"id":96467205,"identity":"a518a4fe-feb8-4d29-9163-f4d428a74e64","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10732,"visible":true,"origin":"","legend":"","description":"","filename":"34da9382a03f436a8c290d2e0a468993.json","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/a93ed769117c177f71efcd92.json"},{"id":96467203,"identity":"a47014c0-c54b-41b4-8ee0-b82c8f8ce4b0","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12950,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFile1CoreIntakeCriteriaforAdaptiveClinicalTriageACT.docx","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/c9e89a61fb714f3cd56e5e31.docx"},{"id":96467204,"identity":"bf9fc96e-d407-464e-a9a2-943141ca726b","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10458,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFile2ACTADJUSTCrossClassificationMatrix.docx","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/77d8e3b7f8bc1953a3bbe26b.docx"},{"id":96467213,"identity":"299f6444-e7eb-4df4-a07e-c91abeb4ea86","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":640625,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFile3PaperbasedACTintakeformv1.0.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/969741f45c064529701fd752.pdf"},{"id":96603470,"identity":"b125bb72-b982-402b-b1e8-3ebdf3f368c5","added_by":"auto","created_at":"2025-11-24 09:09:26","extension":"xml","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":70418,"visible":true,"origin":"","legend":"","description":"","filename":"34da9382a03f436a8c290d2e0a4689931enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/19c23841fb023c46d432e25f.xml"},{"id":96467211,"identity":"39b0af28-cbe0-4e4e-830b-2b65c962927f","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"xml","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":67880,"visible":true,"origin":"","legend":"","description":"","filename":"34da9382a03f436a8c290d2e0a4689931structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/be194da909cd140cf6d6985a.xml"},{"id":96467214,"identity":"9555abd1-cb4a-49d0-a337-19f5653e72e4","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"html","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":77119,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/5f109d4004d3c646cc89f319.html"},{"id":96708365,"identity":"8f839aed-dd7d-4c6f-8c3b-619cb5e1d43d","added_by":"auto","created_at":"2025-11-25 10:01:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":825181,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/4b5da258-003e-41d5-8392-f1390e1f8ded.pdf"},{"id":96467207,"identity":"8e185f90-e286-47f8-b2a0-14083b2c9ef0","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":12950,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary files:\u003c/p\u003e\n\u003cp\u003e● Supplementary File 1 → \u003cem\u003eCore intake criteria\u003c/em\u003e (text table).\u003c/p\u003e","description":"","filename":"SupplementaryFile1CoreIntakeCriteriaforAdaptiveClinicalTriageACT.docx","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/3e941864aa6cf47860af78a7.docx"},{"id":96467210,"identity":"43fde1cb-3fe3-47cf-bbb0-db123e28a1ca","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":10458,"visible":true,"origin":"","legend":"\u003cp\u003e● Supplementary File 2 → \u003cem\u003eACT–ADJUST cross-classification matrix.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"SupplementaryFile2ACTADJUSTCrossClassificationMatrix.docx","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/4dd778d02556a170f4e3129f.docx"},{"id":96467212,"identity":"432eb603-4118-4693-9c3c-8e17b0d03866","added_by":"auto","created_at":"2025-11-21 11:32:51","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":640625,"visible":true,"origin":"","legend":"\u003cp\u003e● Supplementary File 3 → \u003cem\u003ePaper-based ACT intake form (v1.0).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"SupplementaryFile3PaperbasedACTintakeformv1.0.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7852065/v1/640066e87f07ea3b706426a9.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"The ACT and the ADJUST framework: Layering structured detection and clinical judgment in emergency department triage","fulltext":[{"header":"Background","content":"\u003cp\u003eEmergency department triage is fundamental to patient safety and operational efficiency, especially amid rising demand and limited capacity.\u003c/p\u003e\u003cp\u003eIt must simultaneously enable rapid identification of critically ill patients and the efficient coordination of finite resources \u0026mdash; two aims that often diverge as patient volumes fluctuate or operational pressures intensify. Widely used triage frameworks such as the Manchester Triage System (MTS), the Canadian Triage and Acuity Scale (CTAS), and the Rapid Emergency Triage and Treatment System (RETTS) offer comprehensive decision structures but rely on branching complaint algorithms and multiple discriminators. While such systems provide detailed guidance, they can be cognitively demanding and time-consuming to apply, particularly in departments with variable staffing, limited training, or high turnover. Their complexity can make them less adaptable in settings that require rapid, consistent decision-making under pressure, contributing to performance variability across studies [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAdaptive Clinical Triage (ACT) was developed as a streamlined alternative. Nurse-led by design, ACT integrates predefined NEWS2 thresholds, concise symptom-urgency flags, and a discretionary concern override, omitting complex complaint algorithms. The term adaptive reflects its design intent to combine structured thresholds with professional judgment, balancing standardization with flexibility. The model emphasizes early physiological risk detection through a limited set of symptom cues, aiming to be straightforward to learn, consistently applied, and adaptable across settings. It seeks to reduce cognitive load and promote uniformity in triage decisions, particularly in high-pressure environments. Although NEWS2 and its predecessor show good discrimination for mortality and ICU admission in ED cohorts, their role as stand-alone triage tools remains debated [\u003cspan additionalcitationids=\"CR4 CR5\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. ACT operationalizes NEWS2 within a simplified intake structure intended to preserve clinical judgment while reducing procedural complexity.\u003c/p\u003e\u003cp\u003eWhile structured triage primarily targets severity\u0026mdash;the immediate physiological threat\u0026mdash;emergency care also depends on acuity, a broader construct encompassing disease trajectory, comorbidity, and operational urgency. A secondary layer of physician review can therefore complement structured detection by integrating evolving information. Such reassessment enhances safety by identifying patients needing closer monitoring or escalation and improves efficiency by confirming stability or downgrading when early treatment has restored stability.\u003c/p\u003e\u003cp\u003eTo address this need, a physician-led reassessment\u0026mdash;Adaptive Judgment of Urgency and Streaming (ADJUST)\u0026mdash;was developed. Using the same four-tier acuity scale, it situates the initial triage decision within clinical history and risk context, and when capacity allows, may incorporate limited bedside diagnostics such as focused examination or ultrasound, and guide early patient streaming. Each ADJUST assignment is accompanied by a brief free-text rationale, allowing contextual factors and clinical reasoning to be explicitly documented. Designed as a structured yet flexible framework, ADJUST aligns acuity categorization and resource allocation with the evolving clinical picture rather than functioning as a separate triage system.\u003c/p\u003e\u003cp\u003eThis study primarily evaluates ACT as a stand-alone intake model, using ICU admission within defined time windows as a pragmatic proxy for high acuity. A secondary analysis explores how ADJUST reclassified acuity within the layered model, to illustrate how structured detection and clinical judgment can coexist within a practical, scalable triage framework.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStudy Design and data sources\u003c/h2\u003e\u003cp\u003eThis was a retrospective observational study of all emergency department (ED) encounters at a regional Norwegian hospital serving a catchment population of approximately 240,000 between 1 September 2023 and 1 April 2025. ACT and ADJUST entries were recorded prospectively in real time within the electronic ED trackboard and subsequently extracted for analysis. Supplementary File 1 provides the full English-language ACT intake criteria used in this study. The complete ACT intake chart, integrating the ABCDE urgency framework within the four-tier acuity scale, is provided in Supplementary File 3.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eTriage model and training\u003c/h3\u003e\n\u003cp\u003eThe structure and rationale for ACT and ADJUST have been described above. In brief, Adaptive Clinical Triage (ACT) is a nurse-led intake model that integrates predefined physiological thresholds based on the National Early Warning Score 2 (NEWS2), symptom-urgency flags, and a discretionary concern override. This allows escalation when professional judgment indicates higher risk. Adaptive Judgment of Urgency and Streaming (ADJUST) is a physician-led reassessment using the same four-tier acuity scale (Red, Orange, Yellow, Green). It situates the initial ACT decision within the broader clinical history and risk context and, when capacity allows, incorporates bedside diagnostics such as focused examination, ECG, blood gases, or ultrasound to support early streaming.\u003c/p\u003e\u003cp\u003eACT was implemented as an electronic module within the ED trackboard, supported by a paper template used for training and reference. All ED nurses performed ACT regardless of seniority or prior triage experience. Training initially consisted of case-based sessions and was later standardized through a mandatory 60-minute e-learning module. Newly recruited nurses received supervised practice during orientation, including at least one mentored triage shift.\u003c/p\u003e\u003cp\u003eADJUST documentation was entered directly on the trackboard after initial intake. It represents a concise physician reassessment aimed at refining acuity and urgency detection rather than performing a full diagnostic evaluation. The process varied with workload and provider background. It was most often completed by the supervising physician and routinely performed by a dedicated triage physician during peak hours. In practice, it ranged from brief stability checks by non-emergency physicians to focused assessments that included limited diagnostics and early treatment initiation by emergency physicians. No formal certification was required.\u003c/p\u003e\u003cp\u003eFull intake criteria for ACT are provided in Supplementary File 1. The ACT\u0026ndash;ADJUST cross-classification matrix used in the secondary analysis is presented in Supplementary File 2, and the paper-based ACT intake form (v1.0), which includes both adult and pediatric criteria, is shown in Supplementary File 3.\u003c/p\u003e\n\u003ch3\u003eEligibility and exclusions\u003c/h3\u003e\n\u003cp\u003eAll ED encounters during the study period were eligible for inclusion. Encounters without ACT documentation were excluded from primary analyses; these typically reflected deferred triage entry, documentation gaps in high-acuity cases, or incomplete rollout during early implementation.\u003c/p\u003e\u003cp\u003ePredefined alert pathways such as stroke, trauma, or resuscitation were retained, although documentation was less consistent when immediate treatment superseded formal triage. The main ED receives all adult patients with medical, surgical, and trauma-related presentations. In contrast, noncritical pediatric medical cases are managed in a dedicated pediatric emergency department and were therefore excluded by design. Minor injuries treated in a separate fast-track pathway were also excluded.\u003c/p\u003e\n\u003ch3\u003eOutcome definition\u003c/h3\u003e\n\u003cp\u003eThe primary outcome was ICU admission within 24 hours of ED arrival, used as a pragmatic proxy for high acuity consistent with prior validation work [\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Secondary analyses examined temporal alignment between triage and deterioration by stratifying ICU transfers within \u0026le;\u0026thinsp;2, \u0026le; 4, and \u0026le;\u0026thinsp;6 hours. Only transfers to the central ICU were included; step-down, postoperative, and procedural units were excluded to avoid misclassification of short-term monitoring as critical illness [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn triage research, where no definitive gold standard for \u0026ldquo;true urgency\u0026rdquo; exists, outcome measures often rely on surrogate markers\u0026mdash;a construct-validity approach sometimes referred to as a \u0026ldquo;silver standard.\u0026rdquo; Consistent with this framework, ICU admission served as a pragmatic and auditable proxy for high acuity in this study [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003eTriage categories and analytic approach\u003c/h3\u003e\n\u003cp\u003eBoth ACT and ADJUST assigned one of four categories (Red, Orange, Yellow, Green). Performance metrics were calculated for the Red category only, reflecting its distinct clinical mandate for immediate physician contact and continuous monitoring, whereas Orange allows up to 30 minutes to physician review with intermittent monitoring. Confidence intervals were estimated using the Wald method. Receiver operating characteristic (ROC) analysis was not performed because triage categories are predefined and ordinal; as noted in prior reviews, AUC metrics provide limited interpretive value for multi-level triage systems [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe main analysis evaluated ACT as a stand-alone intake tool. A secondary paired analysis examined encounters with both ACT and ADJUST documentation to assess reclassification patterns and operational impact (\u003cem\u003eSupplementary File 2\u003c/em\u003e).\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eEthics approval\u003c/h2\u003e\u003cp\u003e The study was approved by the internal ethics board at Vestfold Hospital Trust (10 December 2020; case number 20/03355). As a retrospective quality-improvement study using de-identified registry data, the requirement for individual informed consent was waived under national regulations.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003ch3\u003eCohort and documentation\u003c/h3\u003e\n\u003cp\u003eA total of 53,645 ED encounters were recorded during the 19 month study period. ACT documentation was available for 46,895 encounters (87.5%). Among 409 patients admitted to the ICU within 24 hours (0.8% of all encounters), 298 (72.9%) had complete ACT documentation and were included in the primary analysis. ADJUST was documented in 43,931 encounters (82.0%), of which 42,881 had paired ACT\u0026ndash;ADJUST entries used for secondary analysis.\u003c/p\u003e\n\u003cp\u003eEncounters without ACT documentation most commonly represented high-acuity presentations where immediate care superseded electronic entry or cases excluded by design (e.g., fast-track injuries and noncritical medical pediatric cases).\u003c/p\u003e\n\u003ch3\u003eACT performance\u003c/h3\u003e\n\u003cp\u003eThe sensitivity of ACT (Red category) for ICU admission within 24 hours was 81.2% (242/298; 95% CI \u0026plusmn;4.4%), with a specificity of 85.9% (40,026/46,597; \u0026plusmn;0.3%). Sensitivity was highest for ICU transfers within the first two hours (95.2% \u0026plusmn;3.3%) and decreased gradually with longer transfer times\u0026mdash;90.3% (\u0026le; 4 h), 85.3% (\u0026le; 6 h), and 81.2% (\u0026le; 24 h) (Table 1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1. Sensitivity of ACT (Red category) for ICU admission by time to transfer\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"671\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTime to ICU\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrue Positives (TP)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFalse negatives (FN)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 176px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSensitivity (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026le; 2 hours\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e158\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 176px;\"\u003e\n \u003cp\u003e95.2% \u0026plusmn; 3.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026le; 4 hours\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 176px;\"\u003e\n \u003cp\u003e90.3% \u0026plusmn; 3.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026le; 6 hours\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e227\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 176px;\"\u003e\n \u003cp\u003e85.3% \u0026plusmn; 4.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026le; 24 hours\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e242\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 165px;\"\u003e\n \u003cp\u003e56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 176px;\"\u003e\n \u003cp\u003e81.2% \u0026plusmn; 4.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eLegend: Sensitivity of ACT (Red category) for ICU admission by time to transfer. True positives (TP) and false negatives (FN) shown with sensitivity estimates and 95% confidence intervals. Only ICU transfers within the specified window are included.\u003c/p\u003e\n\u003cp\u003eAbbreviations: TP, true positive; FN, false negative; CI, confidence interval; ICU, intensive care unit; ACT, Adaptive Clinical Triage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrigger distribution for ACT Red\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong 6,812 ACT Red designations, 53.4% were triggered by predefined symptom-urgency flags, 21.8% by NEWS2 \u0026ge; 7, and 7.9% solely by discretionary clinical concern. The remaining cases reflected combinations of triggers (e.g., NEWS + symptom or concern) (Table 2).\u003c/p\u003e\n\u003cp\u003eTable 2. Trigger distribution for ACT Red\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"672\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003eTrigger type\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003e% of Red encounters\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003eSymptom-urgency flags\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003e53.4%\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003eNEWS2 \u0026ge; 7\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003e21.8%\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003eClinical concern alone\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003e7.9%\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003eCombinations (e.g., NEWS2 + symptom/concern)\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 336px;\"\u003e16.9%\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eLegend: Distribution of triggers for ACT Red assignments. Percentages reflect the proportion of all ACT Red encounters in which the trigger was present.\u003c/p\u003e\n\u003cp\u003eAbbreviations: NEWS, National Early Warning Score; ACT, Adaptive Clinical Triage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePaired ACT-ADJUST encounters\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the 42,881 paired encounters, ACT and ADJUST assignments matched in 51.1% of cases. Reclassification occurred in 48.9%, with 45.3% downgraded and 3.6% upgraded by ADJUST. Compared with ACT, ADJUST demonstrated higher specificity (96.2%) but lower sensitivity (64.5%) for ICU admission within 24 hours (table 3).\u003c/p\u003e\n\u003cp\u003eIn ICU cases with paired documentation (n = 231), ACT identified 182 (78.8%) and ADJUST identified 149 (64.5%) as Red. Among 38 ICU patients downgraded by ADJUST, 20 (52.6%) were transferred more than four hours after arrival.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3. Sensitivity of ADJUST (Red category) for ICU admission by time to transfer\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"671\" class=\"fr-table-selection-hover\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 24.4664%;\"\u003e\u003cstrong\u003eTime to ICU\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.9589%;\"\u003e\u003cstrong\u003eTrue positives (TP)\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e\u003cstrong\u003eFalse negatives (FN)\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e\u003cstrong\u003eSensitivity (95% CI)\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 24.4664%;\"\u003e\u003cstrong\u003e\u0026le; 2 hours\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.9589%;\"\u003e136\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e14\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e90.7% \u0026plusmn; 4.7%\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 24.4664%;\"\u003e\u003cstrong\u003e\u0026le; 4 hours\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.9589%;\"\u003e183\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e38\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e82.8% \u0026plusmn; 5.0%\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 24.4664%;\"\u003e\u003cstrong\u003e\u0026le; 6 hours\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.9589%;\"\u003e188\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e60\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e75.8% \u0026plusmn; 5.3%\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 24.4664%;\"\u003e\u003cstrong\u003e\u0026le; 24 hours\u003c/strong\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.9589%;\"\u003e192\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e88\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e68.6% \u0026plusmn; 5.4%\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eLegend: Sensitivity of ADJUST (Red category) for ICU admission by time to transfer. True positives (TP) and false negatives (FN) shown with sensitivity estimates and 95% confidence intervals. Data limited to encounters with paired ACT\u0026ndash;ADJUST documentation.\u003c/p\u003e\n\u003cp\u003eAbbreviations: TP, true positive; FN, false negative; CI, confidence interval; ICU, intensive care unit; ACT, Adaptive Clinical Triage; ADJUST, Adaptive Judgment of Urgency and Streaming.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe complete cross-classification matrix for paired ACT\u0026ndash;ADJUST encounters is available in Supplementary File 2.\u003c/p\u003e\n\u003cp\u003eTiming and nature of physician reassessment\u003c/p\u003e\n\u003cp\u003eADJUST documentation occurred within 15 minutes of arrival in 48.6% of paired encounters across all acuity categories. The mean interval between ACT and ADJUST entries was 119 minutes. These timing data encompass the full range of reassessments\u0026mdash;from brief stability confirmations to more detailed early evaluations\u0026mdash;rather than reflecting a single mode of physician review.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003cstrong\u003ePrincipal findings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this large single-center evaluation, the Adaptive Clinical Triage (ACT) model\u0026mdash;a lean, nurse-led system combining predefined NEWS2 thresholds, symptom-urgency flags, and discretionary concern\u0026mdash;achieved high sensitivity for early ICU transfer, particularly within the first two hours after ED arrival. This finding supports ACT\u0026rsquo;s role as a structured, standardized tool for early detection of critical illness. In contrast, the physician-led reassessment layer (ADJUST) improved specificity and reduced over-triage at the expected cost of sensitivity, reflecting its contextual role as a refinement rather than a replacement for structured triage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhy a layered design?\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTraditional complaint-based triage systems (MTS, CTAS, RETTS) depend on extensive branching algorithms that can increase cognitive load and operational variability, particularly in departments with fluctuating staffing and experience levels [1,2]. ACT was designed to minimize such variability through a limited number of structured inputs\u0026mdash;vital-sign thresholds, key symptom cues, and professional concern\u0026mdash;while remaining feasible across diverse ED environments.\u003c/p\u003e\n\u003cp\u003eIn this cohort, approximately three-quarters of Red designations arose from structured triggers (NEWS2 \u0026ge; 7 or predefined symptom flags), while only 7.9% relied solely on clinical concern\u0026mdash;consistent with ACT\u0026rsquo;s intent to reduce unwarranted variation while preserving professional discretion. The subsequent physician reassessment introduces a complementary layer of contextual judgment, allowing clinicians to integrate disease trajectory, comorbidity, and evolving risk. Together, the two layers balance standardized detection with clinical reasoning, addressing the inherent tension between patient safety and resource efficiency in triage.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRecent work also supports the development of layered, semi-structured frameworks that integrate standardized scoring with clinical judgment [10]. This concept is reflected in findings on the NEWS family of scores, which discriminate well for mortality and ICU outcomes but less reliably for clinician-defined urgency, underscoring the value of pairing physiological scoring with contextual reassessment [6]. Together, these insights highlight that the value of a layered triage model lies not only in its structure but also in how its performance is interpreted within the temporal context of patient outcomes.\u003c/p\u003e\n\u003ch3\u003eTemporal alignment and endpoint interpretation\u003c/h3\u003e\n\u003cp\u003eSensitivity declined progressively with longer intervals to ICU transfer, indicating that 24-hour outcomes combine immediate critical illness with later deterioration and operational influences. Stratifying outcomes within 2-, 4-, and 6-hour windows therefore provides a more accurate reflection of triage performance at the time of assessment.\u003c/p\u003e\n\u003cp\u003eAlthough ICU admission is an imperfect proxy for clinical urgency, it remains a pragmatic and auditable endpoint that correlates with resource intensity and is widely used in triage validation studies [4\u0026ndash;6]. Such surrogate outcomes can misclassify patients who improve after early treatment or deteriorate later for unrelated reasons, but they are consistent with a construct-validity approach\u0026mdash;using observable, auditable markers such as admission or mortality as practical proxies for high acuity when no definitive gold standard exists [9].\u003c/p\u003e\n\u003cp\u003eRecognizing these limitations, our stratified, time-based analysis aligns with recent recommendations to evaluate triage using clinically meaningful, time-linked outcomes rather than static disposition measures [1,2,8]. This perspective highlights that triage performance must be interpreted in its temporal and operational context, rather than reduced to final disposition alone.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOperational clarity of acuity tiers\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, the Red category was analyzed separately to respect its distinct mandate for immediate physician contact and continuous monitoring, whereas Orange permits up to 30 minutes to physician review with intermittent monitoring. Pooling these tiers, although common in prior studies, risks obscuring escalation thresholds and blurring distinctions in clinical urgency [1,2,11]. This distinction is particularly relevant in the present model, where the 30-minute limit for the Orange category represents a comparatively broader interval than the corresponding time mandates in systems such as MTS and CTAS.\u003c/p\u003e\n\u003ch3\u003eOperational implications\u003c/h3\u003e\n\u003cp\u003eThe ACT model demonstrates that high early sensitivity can be achieved through a straightforward, low-complexity framework requiring minimal training and infrastructure. For high-volume EDs under crowding pressure, such standardization may reduce variability in initial acuity assignment and help maintain safety despite fluctuating workload. In resource-constrained settings, ACT\u0026rsquo;s reliance on a small number of observable parameters makes it scalable without digital dependencies.\u003c/p\u003e\n\u003cp\u003eADJUST, when capacity allows, contributes operational precision through contextual downgrading and early streaming, which may improve resource alignment without compromising safety. Its application depends on staffing and workflow context, highlighting the value of embedding flexible, judgment-based reassessment within structured detection systems.\u003c/p\u003e\n\u003ch3\u003eStrengths and limitations\u003c/h3\u003e\n\u003cp\u003eThis study benefits from a large real world dataset and simultaneous capture of both structured nurse triage and physician reassessment within the same electronic system. Limitations include its single-center design, incomplete documentation in some encounters, and heterogeneity in how ADJUST was applied depending on workload and physician background. The use of ICU admission as a pragmatic outcome measure provides consistency with prior validation studies but does not capture all dimensions of clinical urgency.\u003c/p\u003e\n\u003cp\u003eThe optional and variably timed nature of ADJUST also limits conclusions about its full effect on patient outcomes. Because ADJUST relies on contextual clinical reasoning rather than fixed criteria, traditional external validation may not be feasible; future work should instead focus on developing structured guidance to support consistent application across settings.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFuture directions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eProspective multicenter studies could assess inter-rater reliability, calibration across subgroups, and operational outcomes such as time to treatment or boarding delays. Further work should also evaluate how structured guidance for ADJUST influences patient flow and whether specific criteria can help identify when physician reassessment adds the most value.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eACT provides a standardized, low-complexity framework that maintains high sensitivity for early detection of critical illness. ADJUST enhances specificity and operational alignment when applied, demonstrating how structured detection and clinical judgment can coexist within a scalable triage approach. Together, they offer a practical model adaptable to a range of emergency department contexts\u0026mdash;from high-volume hospitals to resource-limited settings.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eED Emergency Department\u003c/p\u003e\n\u003cp\u003eACT Acuity Categorization Tool\u003c/p\u003e\n\u003cp\u003eADJUST Adaptive Judgment of Urgency and Streaming\u003c/p\u003e\n\u003cp\u003eICU Intensive Care Unit\u003c/p\u003e\n\u003cp\u003eNEWS National Early Warning Score\u003c/p\u003e\n\u003cp\u003eCI Confidence Interval\u003c/p\u003e\n\u003cp\u003eMTS, CTAS, RETTS \u0026nbsp;Manchester Triage System, Canadian Triage and Acuity Scale, and Rapid Emergency Triage and Treatment System\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate:\u003c/strong\u003e The study was conducted in accordance with the ethical standards of the institutional and national research committees and with the 1964 Declaration of Helsinki and its later amendments. Data were obtained from the ED trackboard, an internal quality registry at Vestfold Hospital Trust approved by Sikt (the Norwegian Agency for Shared Services in Education and Research, formerly NSD; reference number 54073, valid through August 2030). The registry is approved for use in quality improvement and system performance evaluation. This study was reviewed by the institutional Data Protection Officer and approved by the Medical Director of Vestfold Hospital Trust as an additional publication purpose within the existing registry approval. Only aggregated, non-identifiable data were analyzed; therefore, individual patient consent was not required.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication:\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials:\u003c/strong\u003e De-identified datasets are available from the corresponding author upon reasonable request and with institutional approval.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e Supported internally by Vestfold Hospital Trust. No external funding was received.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGS \u0026ndash; study design, model development, data analysis, manuscript drafting;\u003cbr\u003eVR \u0026ndash; study design, interpretation, critical revision;\u003cbr\u003eB-JS \u0026ndash; data extraction, methodological review;\u003cbr\u003eVEH \u0026amp; RR \u0026ndash; system integration;\u003cbr\u003eIBNV \u0026amp; THL (EM physicians), ES \u0026amp; SJ (ED nurses) \u0026ndash; implementation and conceptual development.\u003c/p\u003e\n\u003cp\u003eAll authors reviewed and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e The authors thank the triage and emergency department staff at Vestfold Hospital Trust for consistent real time documentation and the clinical informatics team for enabling system-integrated data capture.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAI disclosure:\u003c/strong\u003e Portions of this manuscript were language-edited using ChatGPT (OpenAI). No data analysis, interpretation, or content generation was performed by AI.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\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 May 28;9(5):e026471. https://doi.org/10.1136/bmjopen-2018-026471 \u003c/li\u003e\n\u003cli\u003eHinson JS, Martinez DA, Cabral S, George K, Whalen M, Hansoti B, Levin S. Triage performance in emergency medicine: a systematic review. \u003cem\u003eAnn Emerg Med\u003c/em\u003e. https://doi.org/10.1016/j.annemergmed.2018.09.022 \u003c/li\u003e\n\u003cli\u003eRoyal College of Physicians. National Early Warning Score (NEWS) 2: Standardising the assessment of acute-illness severity in the NHS. London: RCP; 2017. https://www.rcp.ac.uk/media/a4ibkkbf/news2-final-report_0_0.pdf\u003c/li\u003e\n\u003cli\u003eSilcock, D. J., Corfield, A. R., Gowens, P. A., \u0026amp; Rooney, K. D. (2015). Validation of the National Early Warning Score in the prehospital setting. \u003cem\u003eResuscitation\u003c/em\u003e, \u003cem\u003e89\u003c/em\u003e, 31\u0026ndash;35. https://doi.org/10.1016/j.resuscitation.2014.12.029 \u003c/li\u003e\n\u003cli\u003eCovino M, Sandroni C, Della Polla D, De Matteis G, Piccioni A, De Vita A, et al. Predicting ICU admission and death in the Emergency Department: A comparison of six early warning scores. \u003cem\u003eResuscitation\u003c/em\u003e. 2023;190:109876. https://doi.org/10.1016/j.resuscitation.2023.109876\u003c/li\u003e\n\u003cli\u003eZaboli A, Brigo F, Sibilio S, Massar M, Magnarelli G, Brigiari G, Turcato G. Evaluating the National Early Warning Score (NEWS) in triage: A machine learning perspective. International Emergency Nursing. 2025;80:101602. https://doi.org/10.1016/j.ienj.2025.101602 \u003c/li\u003e\n\u003cli\u003eSolev\u0026aring;g AL, Hinna UT, Eggen EH, Engan M, Gundersen AI, D\u0026oslash;llner H. \u003cem\u003ePediatrisk tidlig varslingssk\u0026aring;r, triage og kommunikasjon\u003c/em\u003e. Akuttveileder i pediatri, Helsebiblioteket. Updated 01.01.2021. Available at: 1.13 Pediatrisk tidlig varslingssk\u0026aring;r, triage og kommunikasjon - Helsebiblioteket\u003c/li\u003e\n\u003cli\u003eJohansson A, Ekwall A, Forberg JL, Ekelund U. Development of outcomes for evaluating emergency care triage: a Delphi approach. \u003cem\u003eScand J Trauma Resusc Emerg Med\u003c/em\u003e. 2023;31(1):10. https://doi.org/10.1186/s13049-023-01073-1\u003c/li\u003e\n\u003cli\u003eSchinkel M, Bergsma L, Veldhuis LI, et al. Comparing complaint-based triage scales and early warning scores for emergency department triage. Emergency Medicine Journal. 2022;39(9):691\u0026ndash;696. https://doi.org/10.1136/emermed-2021-211544\u003c/li\u003e\n\u003cli\u003eTsiftsis D, Tasioulis A, Bampalis D. Adult triage in the emergency department: Introducing a multi-layer triage system. \u003cem\u003eHealthcare\u003c/em\u003e. 2025;13(9):1070. https://doi.org/10.3390/healthcare13091070\u003c/li\u003e\n\u003cli\u003eZachariasse JM, Seiger N, Rood PPM, Alves CF, Freitas P, Smit FJ, et al. Validity of the Manchester Triage System in emergency care: A prospective observational study. \u003cem\u003ePLoS One\u003c/em\u003e. 2017;12(2):e0170811. https://doi.org/10.1371/journal.pone.0170811\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-emergency-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"emmd","sideBox":"Learn more about [BMC Emergency Medicine](http://bmcemergmed.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/emmd","title":"BMC Emergency Medicine","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"emergency triage, NEWS2, clinical judgment, layered triage, patient acuity, operational efficiency, ICU admission","lastPublishedDoi":"10.21203/rs.3.rs-7852065/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7852065/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eEmergency department (ED) triage must balance rapid detection of critical illness with efficient use of limited resources. We evaluated the Adaptive Clinical Triage (ACT) model\u0026mdash;a deliberately lean, nurse-led approach integrating predefined NEWS2 thresholds, symptom-urgency flags, and a discretionary concern override. We also examined ADJUST (Adaptive Judgment of Urgency and Streaming), a semi-structured physician reassessment incorporating early streaming where capacity allows.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eWe performed a retrospective cohort study of 53,645 ED encounters (1 Sept 2023\u0026ndash;1 Apr 2025). ACT and ADJUST data were extracted from the electronic trackboard. The primary outcome was intensive care unit (ICU) admission within 24 hours, analyzed as a pragmatic proxy for high acuity. Sensitivity and specificity for ACT were calculated for Red triage assignments and stratified by transfer timing (\u0026le;\u0026thinsp;2 h, \u0026le;\u0026thinsp;4 h, \u0026le;\u0026thinsp;6 h). A secondary analysis of paired ACT\u0026ndash;ADJUST encounters examined how physician reassessment reclassified acuity.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eAmong 46,895 encounters with ACT documentation, sensitivity for ICU admission within 24 hours was 81.2% (95% CI 76.8\u0026ndash;85.6) and specificity 85.9% (85.6\u0026ndash;86.3). Sensitivity reached 95.2% for ICU transfers within 2 hours and decreased gradually with longer transfer times, indicating strong early detection of critical illness. In 42,881 paired encounters, ACT and ADJUST agreed in 51%. ADJUST improved specificity (96.2%) but reduced sensitivity (64.5%), consistent with its role as a contextual refinement rather than a primary detection layer.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eACT provides a straightforward, standardized framework that maintains high early sensitivity through structured physiological and symptom-based triggers. ADJUST, when applied, enhances specificity and facilitates early streaming, reflecting its value as a contextual complement to ACT. Together they illustrate how layered triage can combine standardized safety with adaptable clinical judgment. Future studies should examine generalizability across diverse ED environments.\u003c/p\u003e","manuscriptTitle":"The ACT and the ADJUST framework: Layering structured detection and clinical judgment in emergency department triage","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-21 11:32:46","doi":"10.21203/rs.3.rs-7852065/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"88863527531909581727545212867198077662","date":"2026-03-27T13:31:53+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-20T00:18:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"290279566187357238648075711608971552479","date":"2025-11-11T23:55:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-11T23:24:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-15T05:25:53+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-15T05:24:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Emergency Medicine","date":"2025-10-13T19:53:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-emergency-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"emmd","sideBox":"Learn more about [BMC Emergency Medicine](http://bmcemergmed.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/emmd","title":"BMC Emergency Medicine","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"42f7d1dc-ed64-4f9e-ae58-9a5caee9bb12","owner":[],"postedDate":"November 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-11-21T11:32:46+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-21 11:32:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7852065","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7852065","identity":"rs-7852065","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}