Strengths and limitations of early warning scores: A systematic review and narrative synthesis

Adult Triage in the Emergency Department: Introducing a Multi-Layer Triage System

Emergency department (ED) triage is the cornerstone of ED operations. Many different triage systems have been proposed and implemented globally. To date, an ideal triage system has not yet been identified. As the burden on EDs rises, with overcrowding being recognized as a universal problem, ED triage needs to be restructured to address this reality. Extensive and critical literature research over the years has identified the strengths and weaknesses of current ED triage implementations. A novel multi-layer triage system was introduced and implemented in Greek Eds, combining the strengths of various triage and early warning systems and scores to minimize under-triage and the adverse downstream effects it creates on patient outcomes. Acknowledging that no triage system can be universally adapted in different settings, the structural concepts of this triage system address most of the triage problems currently reported in the literature.

A comparative study of neural network architectures for vital signs monitoring based on the national early warning systems (NEWS)

Objective: The study aims to assess the efficacy of various neural network architectures in predicting the National Early Warning Systems (NEWS) score, using vital signs, to enhance early warning and monitoring in clinical settings. Methods: A comparative evaluation of 29 neural network architectures, including Discriminant Analysis, Support Vector Machines, Logistic Regression, Decision Trees, Neural Networks, and Ensemble methods, was performed. These architectures were assessed based on accuracy, sensitivity, processing speed, model size, and execution time, using synthetically generated data representing 9000 clinical scenarios. Results: The analysis revealed that Linear Discriminant Analysis, narrow and medium Neural Networks, and specific Support Vector Machine (SVM) configurations, particularly Linear SVM, Quadratic SVM, and Coarse Gaussian SVM, achieved 100% accuracy and efficiency in predicting NEWS scores, making them suitable for real-time monitoring. Other architectures exhibited varying performance, with many failing to meet the required accuracy for clinical applications. Conclusion: The study identified Linear Discriminant Analysis and narrow and medium Neural Networks, along with Linear, Quadratic, and Coarse Gaussian SVMs, as optimal for integrating machine learning with NEWS, due to their precision, speed, and suitability for deployment in healthcare environments, particularly in Intensive Care Units.

Monitoring Vital Signs With Continuous Monitoring After Major Gastrointestinal Surgical Procedures: The Patient, Nurse and Physician Perspective

BACKGROUND

Patients undergoing major abdominal surgical procedures are at risk of postoperative complications, requiring early recognition. Clinical deterioration is preceded by changes in vital sings, which are measured three times a day by a nurse. Due to the intermittent measuring, this may result in a delay in the recognition of clinical deterioration. Continuous vital sign monitoring through wireless sensors offers a potential solution for earlier recognition.

AIM

To evaluate user satisfaction of a new wireless monitoring system measuring vital signs continuously, by both patients and healthcare providers.

DESIGN

A prospective, questionnaire-based study.

METHODS

From December 2021 to November 2022, user experience questionnaires were administered to patients who underwent major abdominal surgical procedures and received the patch postoperatively. Questionnaires were administered as well to nurses and physicians working on a surgical ward with the patch. Continuous measurements of heart rate, respiratory rate, and temperature were taken using the Sensium wireless patch.

RESULTS

A total of 298 respondents completed the questionnaire, 191 patients, 88 nurses, and 19 physicians. Of the patients, 69% had a positive experience with the patch, and 74% felt safer. Sixty-three percent of the nurses were positive, and 65% had the feeling that they could monitor the patients better this way. Of the physicians, 63% were positive, 32% believed clinical deterioration could be identified earlier.

CONCLUSION

The use of the Sensium wireless patch for continuous monitoring of vital signs postoperatively was found to be feasible and well-tolerated. Patients, nurses, and physicians reported a positive experience with its use.

Continuous vital sign monitoring of patients recovering from surgery on general wards: a narrative review

Most postoperative deaths occur on general wards, often linked to complications associated with untreated changes in vital signs. Monitoring in these units is typically intermittent checks each shift or maximally every 4-6 h, which misses prolonged periods of subtle changes in physiology that can herald a critical downstream event. Continuous monitoring of vital signs is therefore intuitively necessary for patient safety. The past five decades have seen monitoring systems evolve rapidly, and today entirely wireless, wearable, and portable continuous surveillance of vital signs is possible on general wards. Introduction of this technology has the potential to modify both the sensing (afferent) and response (efferent) limbs of monitoring, and will allow earlier detection of vital signs perturbations. But this comes with challenges, including but not limited to issues with connectivity, data handling, alarm fatigue, information overload, and lack of meaningful clinical interventions. Evidence from before and after studies and retrospective propensity-matched data suggests that continuous ward monitoring decreases the risk of intensive care unit (ICU) admissions, rapid response calls, and in some instances, mortality. This review summarises the history of general ward monitoring and describes future directions, including opportunities to implement these devices using artificial intelligence, pattern detection, and user-friendly interfaces. Pragmatic, well designed and appropriately powered trials, and real-world implementation data are necessary to make continuous monitoring standard practice at every hospital bed.

Association between nurse turnover and missed nursing care in acute care hospitals in South Korea

Objectives

High nurse turnover during nursing shortages can contribute to missed nursing care. This study investigated the prevalence of missed nursing care and how nurse turnover affects missed nursing care.

Methods

A cross-sectional design was adopted to collect data from a convenience sample of nurses working in general hospitals in South Korea. Six-month turnover rates (0%, 1-14%, 15-22%, and 23-50%) and 24 missed nursing care activities were measured. A multivariate regression analysis was performed to examine the relationship between nurse turnover and missed nursing care, after controlling for nurse and work-related characteristics.

Results

The final sample was 264 nurses. The mean six-month turnover rate was 15.49%. Seven activities (turning patient every 2 h, attending interdisciplinary care conference, ambulation, patient bathing/skin care, emotional support, mouth care, full documentation) had a missed care prevalence of 30% or higher. Nurses in units with moderate turnover rates (15 and 22%) reported more missed nursing care than those in units with zero turnover.

Conclusion

Nurse turnover increases missed nursing care, highlighting the adverse effects of nurse turnover on care processes. Consequently, hospitals and governments should implement policy changes and strategies to prevent nurse turnover.

Can an increase in nursing care complexity raise the risk of intra-hospital and intensive care unit transfers in children? A retrospective observational study

INTRODUCTION

Intra-hospital patient transfers (IPTs) and transfers to intensive care units (ICUs) are high-risk events in pediatric care. Nursing care complexity, reflected by nursing diagnoses (NDs) and nursing actions (NAs), may influence the frequency of these transfers. This study explores the association between nursing care complexity and IPTs, including ICU transfers, in hospitalized children.

MATERIALS AND METHODS

A retrospective observational study was conducted at a tertiary care university hospital in Italy. Data from 1013 children aged 2 to 12 years were collected from electronic health records. Sociodemographic, clinical, and nursing data, including NDs and NAs, were analyzed. Latent Class Analysis classified nursing care complexity, while backward elimination regression and binary logistic regression identified predictors of IPTs and ICU transfers.

RESULTS

Significant positive correlations were found between IPTs and both NDs (rs = 0.326, p < 0.001) and NAs (rs = 0.428, p < 0.001). Key predictors of IPTs included Diagnosis Related Groups (DRG) weight, total comorbidities, surgical DRG, the number of medications used, and high nursing care complexity. ICU-transferred patients had significantly higher nursing care complexity (6.54 vs. 3.46 NDs, p < 0.001; 31 vs. 16 NAs, p < 0.001). High nursing care complexity increased the likelihood of ICU transfer by 18 times (OR = 18.413, p < 0.001).

CONCLUSION

Nursing care complexity strongly influences IPTs and ICU transfers. Close monitoring of patients with high nursing care complexity is essential to anticipate transfers and reduce clinical risks.

'White lies and safety nets': The perceptions of nurses on the use of early warning systems and the development of higher-order thinking skills

BACKGROUND

Algorithmic tools such as early warning systems (EWSs) have been embedded into clinical practice globally to facilitate the early recognition of patient deterioration and to guide the escalation of care. Concerns have been raised that the mandated use of these EWS tools may impact the development of nurses' higher-order thinking. However, the relationship between EWS tools and the development of higher-order thinking is poorly understood.

OBJECTIVES

This paper provides the qualitative results of a larger study that sought to explore the impact of EWS tools on the development of nurses' higher-order thinking. The objective of this component of the study was to ascertain the thoughts and perceptions of nurses on the use of EWSs and how this related to the development of higher-order thinking skills.

METHODS

A mixed-method, concurrent study design was used to explore the concept of the development of nurses' higher-order thinking in the context of EWS tools. The qualitative responses from a Qualtrics survey were thematically analysed and presented.

FINDINGS

Two major themes were uncovered: White Lies and Safety Nets. Our analysis of the data suggested that some nurses amend their documentation practice to accommodate the EWS's escalation process, uncovering a view that the tool did not account for clinical reasoning. Parallel to this, some nurses found that these systems supported clinical decision-making and helped to build confidence, thus acting as a safety net for their practice.

CONCLUSION

Reliance on EWSs can both hinder and/or support the development of higher-order thinking. Early warning systems are useful tools in ensuring patient safety but should be used in conjunction with nurses' higher-order thinking.

Machine and Deep Learning Models for Hypoxemia Severity Triage in CBRNE Emergencies

BACKGROUND/OBJECTIVES

This study develops machine learning (ML) models to predict hypoxemia severity during emergency triage, particularly in Chemical, Biological, Radiological, Nuclear, and Explosive (CBRNE) scenarios, using physiological data from medical-grade sensors.

METHODS

Tree-based models (TBMs) such as XGBoost, LightGBM, CatBoost, Random Forests (RFs), Voting Classifier ensembles, and sequential models (LSTM, GRU) were trained on the MIMIC-III and IV datasets. A preprocessing pipeline addressed missing data, class imbalances, and synthetic data flagged with masks. Models were evaluated using a 5-min prediction window with minute-level interpolations for timely interventions.

RESULTS

TBMs outperformed sequential models in speed, interpretability, and reliability, making them better suited for real-time decision-making. Feature importance analysis identified six key physiological variables from the enhanced NEWS2+ score and emphasized the value of mask and score features for transparency. Voting Classifier ensembles showed slight metric gains but did not outperform individually optimized models, facing a precision-sensitivity tradeoff and slightly lower F1-scores for key severity levels.

CONCLUSIONS

TBMs were effective for real-time hypoxemia prediction, while sequential models, though better at temporal handling, were computationally costly. This study highlights ML's potential to improve triage systems and reduce alarm fatigue, with future plans to incorporate multi-hospital datasets for broader applicability.

Performance of machine learning versus the national early warning score for predicting patient deterioration risk: a single-site study of emergency admissions

OBJECTIVES

Increasing operational pressures on emergency departments (ED) make it imperative to quickly and accurately identify patients requiring urgent clinical intervention. The widespread adoption of electronic health records (EHR) makes rich feature patient data sets more readily available. These large data stores lend themselves to use in modern machine learning (ML) models. This paper investigates the use of transformer-based models to identify critical deterioration in unplanned ED admissions, using free-text fields, such as triage notes, and tabular data, including early warning scores (EWS).

DESIGN

A retrospective ML study.

SETTING

A large ED in a UK university teaching hospital.

METHODS

We extracted rich feature sets of routine clinical data from the EHR and systematically measured the performance of tree- and transformer-based models for predicting patient mortality or admission to critical care within 24 hours of presentation to ED. We compared our proposed models to the National EWS (NEWS).

RESULTS

Models were trained on 174 393 admission records. We found that models including free-text triage notes outperform structured tabular data models, achieving an average precision of 0.92, compared with 0.75 for tree-based models and 0.12 for NEWS.

CONCLUSIONS

Our findings suggests that machine learning models using free-text data have the potential to improve clinical decision-making in the ED; our techniques significantly reduce alert rate while detecting most high-risk patients missed by NEWS.

The need for acute assessments in home healthcare - Swedish registered nurses' experiences

PURPOSE

The study aims to describe Swedish RNs' experiences of acute assessments at home. More patients with complex nursing needs are cared for at home due to an ageing population. Registered nurses (RNs) who work with home healthcare need a broad medical competence and clinical experience alongside adapted decision support systems for maintaining patient safety in acute assessments within home healthcare.

METHODS

A content analysis of qualitative survey data from RNs (n = 19) working within home healthcare in Sweden.

RESULTS

There were challenges in the acute assessments at home due to a lack of competence since several of the RNs did not have much experience working as an RN in home healthcare. Important information was missing about the patients, such as access to medical records due to organizational challenges and limited access to equipment and materials. The RNs needed support in the form of cooperation with a physician, support from colleagues, and a decision support system.

CONCLUSION

To increase the possibility of patient-safe assessments at home, skills development, collegial support, and an adapted decision support system are needed. Collaboration with primary healthcare, on-call physicians, and nursing staff, and having the opportunity to consult with someone also provide security in acute assessments.

Worry as Important "Feelers" in Clinical Anesthesia Practice: A Mixed-Methods Study

PURPOSE

Worry is an intuitive sense that goes beyond logical reasoning and is valuable in situations where patients' conditions are rapidly changing or when objective data may not fully capture the complexity of a patient's situation. Nurse anesthetists' subjective reasons for worry are quite vague as they are valued inconsistently and not accurately expressed. This study aimed to identify factors playing a role in the emergence of worry during anesthesia practice to clarify its concept.

DESIGN

Mixed-methods design consisting of quantitative online surveys followed by qualitative focus group interviews including Dutch nurse anesthetists.

METHODS

Both quantitative and qualitative thematic analyses were performed, followed by data and methodological triangulation to enhance the validity and credibility of findings and mitigate the presence of bias.

FINDINGS

Surveys (N = 102) were analyzed, and 14 nurse anesthetists participated in the focus group interviews. A total of 89% of the survey respondents reported that at least once have had the feeling of worry, of which 92% use worry during clinical anesthesia practice. Worry was mentioned to be a vital element during anesthesia practice that makes it possible to take precautionary actions to change the anesthetic care plan in a changing situation or patient deterioration.

CONCLUSIONS

While a clear definition of worry could not be given, it is a valuable element of anesthesia practice as it serves as a catalyst for critical thinking, problem-solving, clinical reasoning, and decision-making. Use of the feeling of worry alongside technological systems to make an informed decision is crucial. Technology has significantly improved the ability of health care providers to detect and respond to patient deterioration promptly, but it is crucial for nurse anesthetists to use their feeling of worry or intuition alongside technological systems and evidence-based practice to ensure quick assessments or judgments based on experience, knowledge, and observations in clinical practice.

Multifactorial fall interventions for people over 65 years in the acute hospital setting: pre-post-test design

BACKGROUND

Falls are the most reported patient safety incident for patients >65 years in acute hospital settings worldwide. While multifactorial fall interventions reduce the number of falls in subacute and rehabilitation settings, fall interventions in acute hospital settings are unknown.

AIM

To evaluate the effectiveness of multifactorial fall interventions on the number of falls using codesigned education targeting staff and the patient and review the environment in acute hospital settings in NSW, Australia for patients over 65 years of age.

METHOD

A pre-post-test design with a non-equivalent group was conducted. All acute hospital inpatient falls occurring both pre- and post-intervention within one health district were included in this study. The use of Quality Improvement methodology identified gaps in risk screening and assessment, education and information, communication of risk, and standardised fall prevention equipment. Codesigned interventions to address these gaps were undertaken.

RESULTS

The number of falls (p = 0.038) and injurious falls (p < 0.001) significantly decreased in the post-intervention group. There was a significant improvement in fall assessments (p < 0.001), delirium risk screening (p < 0.001), the provision of fall information (p < 0.001) and fall risk discussed at shift handover (p < 0.001) in the post-intervention group. Following the intervention, staff were significantly more likely to undertake fall education modules (p < 0.001) and develop a fall management plan (p < 0.001).

CONCLUSION

Falls continue to have a significant economic impact on the acute hospital setting. Our findings highlight multifactorial fall interventions that included staff and patients in the development phases reduced the number of falls. Multifactorial fall interventions targeting staff, patients and the environment may influence a reduction in the number of falls and the severity of falls in the acute hospital setting.

IMPACT STATEMENT

Multifactorial fall interventions reduce injurious falls, minor injuries, and falls resulting in serious injury and death.

Views and Uses of Sepsis Digital Alerts in National Health Service Trusts in England: Qualitative Study With Health Care Professionals

BACKGROUND

Sepsis is a common cause of serious illness and death. Sepsis management remains challenging and suboptimal. To support rapid sepsis diagnosis and treatment, screening tools have been embedded into hospital digital systems to appear as digital alerts. The implementation of digital alerts to improve the management of sepsis and deterioration is a complex intervention that has to fit with team workflow and the views and practices of hospital staff. Despite the importance of human decision-making and behavior in optimal implementation, there are limited qualitative studies that explore the views and experiences of health care professionals regarding digital alerts as sepsis or deterioration computerized clinician decision support systems (CCDSSs).

OBJECTIVE

This study aims to explore the views and experiences of health care professionals on the use of sepsis or deterioration CCDSSs and to identify barriers and facilitators to their implementation and use in National Health Service (NHS) hospitals.

METHODS

We conducted a qualitative, multisite study with unstructured observations and semistructured interviews with health care professionals from emergency departments, outreach teams, and intensive or acute units in 3 NHS hospital trusts in England. Data from both interviews and observations were analyzed together inductively using thematic analysis.

RESULTS

A total of 22 health care professionals were interviewed, and 12 observation sessions were undertaken. A total of four themes regarding digital alerts were identified: (1) support decision-making as nested in electronic health records, but never substitute professionals' knowledge and experience; (2) remind to take action according to the context, such as the hospital unit and the job role; (3) improve the alerts and their introduction, by making them more accessible, easy to use, not intrusive, more accurate, as well as integrated across the whole health care system; and (4) contextual factors affecting views and use of alerts in the NHS trusts. Digital alerts are more optimally used in general hospital units with a lower senior decision maker:patient ratio and by health care professionals with experience of a similar technology. Better use of the alerts was associated with quality improvement initiatives and continuous sepsis training. The trusts' features, such as the presence of a 24/7 emergency outreach team, good technological resources, and staffing and teamwork, favored a more optimal use.

CONCLUSIONS

Trust implementation of sepsis or deterioration CCDSSs requires support on multiple levels and at all phases of the intervention, starting from a prego-live analysis addressing organizational needs and readiness. Advancements toward minimally disruptive and smart digital alerts as sepsis or deterioration CCDSSs, which are more accurate and specific but at the same time scalable and accessible, require policy changes and investments in multidisciplinary research.

Pilot study for the development of an automatically generated and wearable-based early warning system for the detection of deterioration of hospitalized patients of an acute care hospital

BACKGROUND

Acute deteriorations of health status are common in hospitalized patients and are often preceded by changes in their vital signs. Events such as heart attacks, death or admission to the intensive care unit can be averted by early detection, therefore so-called Early Warning Scores (EWS) such as the National Early Warning Score 2 (NEWS2), including basic vital parameters such as heart rate, blood pressure, respiratory rate, temperature and level of consciousness, have been developed for a systematic approach. Although studies have shown that EWS have a positive impact on patient outcomes, they are often limited by issues such as calculation errors, time constraints, and a shortage of human resources. Therefore, development of tools for automatic calculation of EWS could help improve quality of EWS calculation and may improve patient outcomes. The aim of this study is to analyze the feasibility of wearable devices for the automatic calculation of NEWS2 compared to conventional calculation using vital signs measured by health care professionals.

METHODS

We conducted a prospective trial at a large tertiary hospital in Switzerland. Patients were given a wristband with a photoplethysmogram (PPG) sensor that continuously recorded their heart rate and respiratory rate for 3 consecutive days. Combined with data from the electronic health record (EHR), NEWS2-score was calculated and compared to NEWS2 score calculated from vital parameters in the EHR measured by medical staff. The main objective of our study was to assess the agreement between NEWS2 scores calculated using both methods. This analysis was conducted using Cohen's Kappa and Bland-Altman analysis. Secondary endpoints were compliance concerning the medical device, patient acceptance, data quality analysis and data availability and signal quality for all time stamps needed for accurate calculation.

RESULTS

Of 210 patients enrolled in our study, NEWS2 was calculated in 904 cases, with 191 cases being directly compared to conventional measurements. Thirty-three of these measurements resulted in a NEWS2 ≥ 5, 158 in a NEWS2 < 5. Comparing all 191 measurements, accordance was substantial (K = 0.76) between conventional and automated NEWS2. No adverse effects due to the device were recorded. Patient acceptance was high.

CONCLUSIONS

In conclusion, the study found strong agreement between automated and conventional NEWS2 calculations using wearable devices, with high patient acceptance despite some data quality challenges. To maximize the potential of continuous monitoring, further research into fully automated EWS calculations without relying on spot measurements is suggested, as this could provide a reliable alternative to traditional methods.

TRIAL REGISTRATION

January 26, 2023, NCT05699967.

Motif discovery in hospital ward vital signs observation networks

Vital signs observations are regular measurements used by healthcare staff to track a patient's overall health status on hospital wards. We look at the potential in re-purposing aggregated and anonymised hospital data sources surrounding vital signs recording to provide new insights into how care is managed and delivered on wards. In this paper, we conduct a retrospective longitudinal observational study of 770,720 individual vital signs recordings across 20 hospital wards in South Wales (UK) and present a network modelling framework to explore and extract behavioural patterns via analysis of the resulting network structures at a global and local level. Self-loop edges, dyad, triad, and tetrad subgraphs were extracted and evaluated against a null model to determine individual statistical significance, and then combined into ward-level feature vectors to provide the means for determining notable behaviours across wards. Modelling data as a static network, by aggregating all vital sign observation data points, resulted in high uniformity but with the loss of important information which was better captured when modelling the static-temporal network, highlighting time's crucial role as a network element. Wards mostly followed expected patterns, with chains or stand-alone supplementary observations by clinical staff. However, observation sequences that deviate from this are revealed in five identified motif subgraphs and 6 anti-motif subgraphs. External ward characteristics also showed minimal impact on the relative abundance of subgraphs, indicating a 'superfamily' phenomena that has been similarly seen in complex networks in other domains. Overall, the results show that network modelling effectively captured and exposed behaviours within vital signs observation data, and demonstrated uniformity across hospital wards in managing this practice.

A theoretical framework for identifying sociocultural factors that influence nurses' compliance with early warning systems for acute clinical deterioration: A cross-sectional survey

BACKGROUND

Systematic adoption of early warning systems in healthcare settings is dependent on the optimal and reliable application by the user. Psychosocial issues and hospital culture influence clinicians' patient safety behaviours.

OBJECTIVE

(i) To examine the sociocultural factors that influence nurses' EWS compliance behaviours, using a theory driven behavioural model and (ii) to propose a conceptual model of sociocultural factors for EWS compliance behaviour.

DESIGN

A cross-sectional survey.

SETTING

Nurses employed in public hospitals across Queensland, Australia.

PARTICIPANTS

Using convenience and snowball sampling techniques eligible nurses accessed a dedicated web site and survey containing closed and open-ended questions. 291 nurses from 60 hospitals completed the survey.

METHODS

Quantitative data were analysed using ANOVA or t-tests to test differences in means. A series of path models based on the theory were conducted to develop a new model. Directed or theory driven content analysis informed qualitative data analysis.

RESULTS

Nurses report high levels of previous compliance behaviour and strong intentions to continue complying in the future (M=4.7; SD 0.48). Individual compliance attitudes (β 0.29, p<.05), perceived value of escalation (β 0.24, p<.05) and perceived ease or difficulty complying with documentation (β -0.31, p<.05) were statistically significant, predicting 24% of variation in compliance behaviour. Positive personal charting beliefs (β 0.14, p<.05) and subjective norms both explain higher behavioural intent indirectly through personal attitudes. High ratings of peer charting beliefs indirectly explain attitudes through subjective norms (β 0.20, p<.05). Perceptions of control over one's clinical actions (β -0.24, p<.05) and early warning system training (β -0.17, p<.05) directly contributed to fewer difficulties complying with documentation requirements. Prior difficulties when escalating care (β -0.31, p<.05) directly influenced the perceived value of escalating.

CONCLUSIONS

The developed theory-based conceptual model identified sociocultural variables that inform compliance behaviour (documenting and escalation protocols). The model highlights areas of clinical judgement, education, interprofessional trust, workplace norms and cultural factors that directly or indirectly influence nurses' intention to comply with EWS protocols. Extending our understanding of the sociocultural and system wide factors that hamper nurses' use of EWSs and professional accountability has the potential to improve the compliance behaviour of staff and subsequently enhance the safety climate attitudes of hospitals.

TWEETABLE ABSTRACT

A newly developed model reports nurse's personal attitudes, peer influence, perceived difficulties encountered documenting and escalation beliefs all predict early warning system compliance behaviour.

Variation in nurses' compliance with an Early Warning Score protocol: A retrospective cohort study

Introduction

Early Warning Score (EWS) protocols are based on intermittent vital sign measurements, and aim to detect clinical deterioration in a timely manner. Despite its predictive value, its effectiveness remains suboptimal. An important limitation appears to be poor compliance with the EWS protocol and its variation between general wards. The current research does not yet provide an understanding of EWS compliance and variation in different nursing wards.

Aim

To explore the variation in nurses' compliance with the EWS protocol among patients with and without complications and between different nursing wards.

Methods

In a retrospective single-center cohort study, all patient files from three nursing wards of a tertiary teaching hospital in the Netherlands were reviewed over a 1-month period. Compliance was divided into three categories:1) calculation accuracy, 2) monitoring frequency end 3) clinical response.

Results

The cohort of 210 patients contained 5864 measurements, of which 4125 (70.6 %) included EWS. Significant differences in the measured vital signs within incomplete measurements were found among nursing wards. Compliance to monitoring frequency was higher within EWSs of 0-1 (78.4 %) than within EWSs of ≥2 (26.1 %). The proportion of correct follow-up was significantly higher in patients with complications, as was the correct clinical response to an EWS of ≥3 (84.8 % vs. 55.0; p = .011).

Conclusion

Our results suggest suboptimal compliance with the EWS protocol, with large variations between patients with and without complications and between different general care wards. Nurses tended to be more compliant with the EWS protocol for patients with complications.

External validation of the computer aided risk scoring system in predicting in-hospital mortality following emergency medical admissions

BACKGROUND

Clinical prediction models have the potential to improve the quality of care and enhance patient safety outcomes. A Computer-aided Risk Scoring system (CARSS) was previously developed to predict in-hospital mortality following emergency admissions based on routinely collected blood tests and vitals. We aimed to externally validate the CARSS model.

METHODS

In this retrospective external validation study, we considered all adult (≥18 years) emergency medical admissions discharged between 11/11/2020 and 11/11/2022 from The Rotherham Foundation Trust (TRFT), UK. We assessed the predictive performance of the CARSS model based on its discriminative (c-statistic) and calibration characteristics (calibration slope and calibration plots).

RESULTS

Out of 32,774 admissions, 20,422 (62.3 %) admissions were included. The TRFT sample had similar demographic characteristics to the development sample but had higher mortality (6.1 % versus 5.7 %). The CARSS model demonstrated good discrimination (c-statistic 0.87 [95 % CI 0.86-0.88]) and good calibration to the TRFT dataset (slope = 1.03 [95 % CI 0.98-1.08] intercept = 0 [95 % CI -0.06-0.07]) after re-calibrating for differences in baseline mortality (intercept = 0.96 [95 % CI 0.90-1.03] before re-calibration).

CONCLUSION

In summary, the CARSS model is externally validated after correcting the baseline risk of death between development and validation datasets. External validation of the CARSS model showed that it under-predicted in-hospital mortality. Re-calibration of this model showed adequate performance in the TRFT dataset.

The Effect of Modified Early Warning Score (MEWS) and Nursing Guide Application on Postoperative Patient Outcomes: A Randomized Controlled Study

PURPOSE

The aim of this study is to determine the effect of nursing guide application (NGA) on patient outcomes in patients followed up according to the modified early warning score (MEWS) in the postoperative period.

DESIGN

A randomized controlled clinical trial.

METHODS

The sample of the study consisted of 252 patients who underwent surgical intervention under general anesthesia in a university hospital between July 29, 2022, and October 31, 2022.

FINDINGS

Results showed that the development of complications was less in the study group (SG) compared to the control group (CG) during anesthesia (P = .027), in the postanesthesia care unit (PACU) (P = .017), and in the clinic (P = .001). It was found that the duration of stay in PACU in the CG was significantly shorter than in the study group (P < .001), and as the duration of stay in PACU in CG decreased, the MEWS increased (r = -0.201, P = .024). We found that there were fewer patients transferred to the intensive care unit (ICU) after PACU (P = .007), the MEWS was lower, and the number of nursing interventions applied to patients was higher (P < .05).

CONCLUSIONS

In patients followed up according to MEWS, NGA had a positive effect on preventing the development of complications and shortening the intervention time for complications, decreasing ICU admission, decreasing MEWS and increasing the number of nursing interventions. Based on the results, it may be recommended to use MEWS+NGA in the early postoperative period as it positively affects patient outcomes.

The impact of mandated use early warning system tools on the development of nurses' higher-order thinking: A systematic review

AIM

Ascertain the impact of mandated use of early warning systems (EWSs) on the development of registered nurses' higher-order thinking.

DESIGN

A systematic literature review was conducted, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and checklist (Page et al., 2021).

DATA SOURCES

CINAHL, Medline, Embase, PyscInfo.

REVIEW METHODS

Eligible articles were quality appraised using the MMAT tool. Data extraction was conducted independently by four reviewers. Three investigators thematically analysed the data.

RESULTS

Our review found that EWSs can support or suppress the development of nurses' higher-order thinking. EWS supports the development of higher-order thinking in two ways; by confirming nurses' subjective clinical assessment of patients and/or by providing a rationale for the escalation of care. Of note, more experienced nurses expressed their view that junior nurses are inhibited from developing effective higher-order thinking due to reliance on the tool.

CONCLUSION

EWSs facilitate early identification of clinical deterioration in hospitalised patients. The impact of EWSs on the development of nurses' higher-order thinking is under-explored. We found that EWSs can support and suppress nurses' higher-order thinking. EWS as a supportive factor reinforces the development of nurses' heuristics, the mental shortcuts experienced clinicians call on when interpreting their subjective clinical assessment of patients. Conversely, EWS as a suppressive factor inhibits the development of nurses' higher-order thinking and heuristics, restricting the development of muscle memory regarding similar presentations they may encounter in the future. Clinicians' ability to refine and expand on their catalogue of heuristics is important as it endorses the future provision of safe and effective care for patients who present with similar physiological signs and symptoms.

IMPACT

This research impacts health services and education providers as EWS and nurses' development of higher-order thinking skills are essential aspects of delivering safe, quality care.

NO PATIENT OR PUBLIC CONTRIBUTION

This is a systematic review, and therefore, comprises no contribution from patients or the public.

Improving diagnosis of early complications (<1 week) through continuous vital sign monitoring following oncological gastrointestinal surgical procedures

BACKGROUND

Patients undergoing major oncological abdominal surgery are prone to postoperative complications, making early recognition crucial. Clinical deterioration is often preceded by changes in vital signs, which are typically measured thrice a day by a nurse. However, intermittent measurements may delay recognizing clinical deterioration. Continuous vital parameter monitoring may lead to earlier recognition and management of complications and reduce nursing workload.

OBJECTIVE

To compare vital parameter measurements between ward nurses and a wireless continuous monitoring system (Sensium® wireless patch) and assess whether this patch can detect clinical deterioration earlier in patients with complications in the first postoperative week.

METHODS

Vital parameters (heart rate, respiratory rate, and temperature) were collected in patients undergoing an oncological resection of the liver, colorectal, or pancreas. Sensium® patch measurements were compared to nurses' measurements to assess the percentages of discordant measurements. In patients with complications in the first postoperative week, time discrepancies between nurses and Sensium® patch measurements were identified in cases of clinical deterioration (respiratory rate ≥15/min, heart rate ≥100/min, and temperature ≥38°C).

RESULTS

Among 227 patients, 22% of the patients experienced complications. Nurse and Sensium® measurements were discrepant in 586/2272 measurements (26%). In 506/586 discrepancies (86%), this was due to the respiratory rate (difference ≥4/min). Compared to nurses, the Sensium® patch detected an elevated respiratory rate 14 h earlier and heart rate 2 h earlier within complications in the first postoperative week. For temperature, no difference was observed.

CONCLUSION

Continuous monitoring with the Sensium® wireless patch holds promise for earlier recognition of complications in patients who underwent major oncological abdominal surgery.

Effect of Digital Early Warning Scores on Hospital Vital Sign Observation Protocol Adherence: Stepped-Wedge Evaluation

BACKGROUND

Early warning scores (EWS) are routinely used in hospitals to assess a patient's risk of deterioration. EWS are traditionally recorded on paper observation charts but are increasingly recorded digitally. In either case, evidence for the clinical effectiveness of such scores is mixed, and previous studies have not considered whether EWS leads to changes in how deteriorating patients are managed.

OBJECTIVE

This study aims to examine whether the introduction of a digital EWS system was associated with more frequent observation of patients with abnormal vital signs, a precursor to earlier clinical intervention.

METHODS

We conducted a 2-armed stepped-wedge study from February 2015 to December 2016, over 4 hospitals in 1 UK hospital trust. In the control arm, vital signs were recorded using paper observation charts. In the intervention arm, a digital EWS system was used. The primary outcome measure was time to next observation (TTNO), defined as the time between a patient's first elevated EWS (EWS ≥3) and subsequent observations set. Secondary outcomes were time to death in the hospital, length of stay, and time to unplanned intensive care unit admission. Differences between the 2 arms were analyzed using a mixed-effects Cox model. The usability of the system was assessed using the system usability score survey.

RESULTS

We included 12,802 admissions, 1084 in the paper (control) arm and 11,718 in the digital EWS (intervention) arm. The system usability score was 77.6, indicating good usability. The median TTNO in the control and intervention arms were 128 (IQR 73-218) minutes and 131 (IQR 73-223) minutes, respectively. The corresponding hazard ratio for TTNO was 0.99 (95% CI 0.91-1.07; P=.73).

CONCLUSIONS

We demonstrated strong clinical engagement with the system. We found no difference in any of the predefined patient outcomes, suggesting that the introduction of a highly usable electronic system can be achieved without impacting clinical care. Our findings contrast with previous claims that digital EWS systems are associated with improvement in clinical outcomes. Future research should investigate how digital EWS systems can be integrated with new clinical pathways adjusting staff behaviors to improve patient outcomes.

Emergence of Artificial Intelligence and Machine Learning Models in Sudden Cardiac Arrest: A Comprehensive Review of Predictive Performance and Clinical Decision Support

Sudden cardiac death/sudden cardiac arrest (SCD/SCA) is an increasingly prevalent cause of mortality globally, particularly in individuals with preexisting cardiac conditions. The ambiguous premortem warnings and the restricted interventional window related to SCD account for the complexity of the condition. Current reports suggest SCD to be accountable for 20% of all deaths hence accurately predicting SCD risk is an imminent concern. Traditional approaches for predicting SCA, particularly "track-and-trigger" warning systems have demonstrated considerable inadequacies, including low sensitivity, false alarms, decreased diagnostic liability, reliance on clinician involvement, and human errors. Artificial intelligence (AI) and machine learning (ML) models have demonstrated near-perfect accuracy in predicting SCA risk, allowing clinicians to intervene timely. Given the constraints of current diagnostics, exploring the benefits of AI and ML models in enhancing outcomes for SCA/SCD is imperative. This review article aims to investigate the efficacy of AI and ML models in predicting and managing SCD, particularly targeting accuracy in prediction.

A Single-Center Retrospective Evaluation of Unplanned Pediatric Critical Care Upgrades

Background  Inappropriate triage of critically ill pediatric patients can lead to poor outcomes and suboptimal resource utilization. This study aimed to determine and describe the demographic characteristics, diagnostic categories, and timing of unplanned upgrades to the pediatric intensive care unit (PICU) that required short (< 24 hours of care) and extended (≥ 24 hours of care) stays. In this article, we hypothesized that we will identify demographic characteristics, diagnostic categories, and frequent upgrade timing periods in both of these groups that may justify more optimal triage strategies. Methods  This was a single-institution retrospective study of unplanned PICU upgrades between 2012 and 2018. The cohort was divided into two groups (short and extended PICU stay). We reviewed the electronic health record and evaluated for: demographics, mortality scores, upgrade timing (7a-3p, 3p-11p, 11p-7a), lead-in time (time spent on clinical service before upgrade), patient origin, and diagnostic category. Results  Four hundred and ninety-eight patients' unplanned PICU upgrades were included. One hundred and nine patients (21.9%) required a short and 389 (78.1%) required an extended PICU stay. Lead-in time (mean, standard deviation) was significantly lower in the short group (0.65 ± 0.66 vs. 0.91 ± 0.82) ( p  = 0.0006). A higher proportion of short group patients (59, 46.1%) were upgraded during the 3p-11p shift ( p  = 0.0077). Conclusion  We found that approximately one-fifth of PICU upgrades required less than 24 hours of critical care services, were more likely to be transferred between 3p-11p, and had lower lead-in times. In institutions where ill pediatric patients can be admitted to either a PICU or a monitored step-down unit, this study highlights quality improvement opportunities, particularly in recognizing which pediatric patients truly need critical care.

Wearable wireless continuous vital signs monitoring on the general ward

PURPOSE OF REVIEW

Wearable wireless sensors for continuous vital signs monitoring (CVSM) offer the potential for early identification of patient deterioration, especially in low-intensity care settings like general wards. This study aims to review advances in wearable CVSM - with a focus on the general ward - highlighting the technological characteristics of CVSM systems, user perspectives and impact on patient outcomes by exploring recent evidence.

RECENT FINDINGS

The accuracy of wearable sensors measuring vital signs exhibits variability, especially notable in ambulatory patients within hospital settings, and standard validation protocols are lacking. Usability of CMVS systems is critical for nurses and patients, highlighting the need for easy-to-use wearable sensors, and expansion of the number of measured vital signs. Current software systems lack integration with hospital IT infrastructures and workflow automation. Imperative enhancements involve nurse-friendly, less intrusive alarm strategies, and advanced decision support systems. Despite observed reductions in ICU admissions and Rapid Response Team calls, the impact on patient outcomes lacks robust statistical significance.

SUMMARY

Widespread implementation of CVSM systems on the general ward and potentially outside the hospital seems inevitable. Despite the theoretical benefits of CVSM systems in improving clinical outcomes, and supporting nursing care by optimizing clinical workflow efficiency, the demonstrated effects in clinical practice are mixed. This review highlights the existing challenges related to data quality, usability, implementation, integration, interpretation, and user perspectives, as well as the need for robust evidence to support their impact on patient outcomes, workflow and cost-effectiveness.

Nurse by numbers: The impact of early warning systems on nurses' higher-order thinking, a quantitative study

AIM

To evaluate registered nurses' perceptions of whether the mandated use of the early warning system vital signs tool impacts the development of nurses' higher-order thinking skills.

DESIGN

A concurrent mixed methods study design.

METHOD

Using an online survey, registered nurses' perceptions were elucidated on whether early warning system algorithmic tools affected the development of their higher-order thinking. Likert-type matrix questions with additional qualitative fields were used to obtain information on nurse's perceptions of the tool's usefulness, clinical confidence in using the tool, compliance with escalation protocols, work environment and perceived compliance barriers.

RESULTS

Most of the 305 (91%) participants included in the analysis had more than 5 years of nursing experience. Most nurses supported the early warning tool and were happy to comply with escalation protocols if the early warning score concurred with their assessment of the patient (63.6%). When the score and the nurse's higher-order thinking did not align, some had the confidence to override the escalation protocol (40.0%), while others omitted (69.4%) or inaccurately documented vital signs (63.3%) to achieve the desired score. Very few nurses (3.6%) believe using early warning tools did not impede the development of higher-order thinking.

CONCLUSION

Although experienced nurses appreciate the support of early warning tools, most value patient safety above the tools and rely on their higher-order thinking. The sustained development and use of nurses' higher-order thinking should be encouraged, possibly by adding a critical thinking criterion to existing algorithmic tools.

IMPACT

The study has implications for all nurses who utilize algorithmic tools, such as early warning systems, in their practice. Relying heavily on algorithmic tools risks impeding the development of higher-order thinking. Most experienced nurses prioritize their higher-order thinking in decision-making but believe early warning tools can impede higher-order thinking.

PATIENT OR PUBLIC CONTRIBUTION

Registered nurses participated as survey respondents.

AI-enabled electrocardiography alert intervention and all-cause mortality: a pragmatic randomized clinical trial

The early identification of vulnerable patients has the potential to improve outcomes but poses a substantial challenge in clinical practice. This study evaluated the ability of an artificial intelligence (AI)-enabled electrocardiogram (ECG) to identify hospitalized patients with a high risk of mortality in a multisite randomized controlled trial involving 39 physicians and 15,965 patients. The AI-ECG alert intervention included an AI report and warning messages delivered to the physicians, flagging patients predicted to be at high risk of mortality. The trial met its primary outcome, finding that implementation of the AI-ECG alert was associated with a significant reduction in all-cause mortality within 90 days: 3.6% patients in the intervention group died within 90 days, compared to 4.3% in the control group (4.3%) (hazard ratio (HR) = 0.83, 95% confidence interval (CI) = 0.70-0.99). A prespecified analysis showed that reduction in all-cause mortality associated with the AI-ECG alert was observed primarily in patients with high-risk ECGs (HR = 0.69, 95% CI = 0.53-0.90). In analyses of secondary outcomes, patients in the intervention group with high-risk ECGs received increased levels of intensive care compared to the control group; for the high-risk ECG group of patients, implementation of the AI-ECG alert was associated with a significant reduction in the risk of cardiac death (0.2% in the intervention arm versus 2.4% in the control arm, HR = 0.07, 95% CI = 0.01-0.56). While the precise means by which implementation of the AI-ECG alert led to decreased mortality are to be fully elucidated, these results indicate that such implementation assists in the detection of high-risk patients, prompting timely clinical care and reducing mortality. ClinicalTrials.gov registration: NCT05118035 .

Wireless patient monitoring and Efficacy Safety Score in postoperative treatment at the ward: evaluation of time consumption and usability

To evaluate objective time consumption and how nurses perceived introducing wireless patient monitoring (WPM) and a validated score on patient quality and safety, the Efficacy Safety Score (ESS), at a mixed surgery ward. After fulfilling a randomised controlled trial combining the ESS and WPM, we addressed time consumption and conducted a questionnaire survey among nurses who participated in the study. The questionnaire appraised the nurses' evaluation of introducing these tools for postoperative management. Of 28 invited nurses, 24 responded to the questionnaire, and 92% reported the ESS and WPM-systems to increase patient safety and quality of care. 67% felt the intervention took extra time, but objective workload measurements revealed reduced time to 1/3 using ESS and WPM compared to standard manual assessment. Improved confidence when using the systems was reported by 83% and improved working situation by 75%. In a test situation to measure time consumption, the ESS and pre-attached WPM-systems require less time than the conventional standard of care, and may allow for more frequent clinical monitoring at the post-surgical ward. The combination of the ESS and the WPM systems was perceived as positive by participating nurses and further clinical development and research is warranted.

Undertaking a neurological assessment

Neurological observations are an essential aspect of assessment in patients with altered mental status and require the nurse to collect and analyse information using a validated assessment tool. Assessing a patient's pupil size and response is also an important element of a neurological assessment. This article summarises the pathophysiology of raised intracranial pressure and lists some of the conditions that may contribute to an alteration in a patient's mental status. The article details the use of two commonly used neurological assessment tools and the assessment of a patient's pupil size and response. The author also considers the challenges related to accurate recording of neurological observations.

Outcomes in patients receiving continuous monitoring of vital signs on general wards: A systematic review and meta-analysis of randomised controlled trials

Objective

The timely identification of deterioration on general wards is crucial to patient care with each hour of delay independently associated with increased risk of death. The introduction of continuous monitoring of patient vital signs on general wards, currently not standard care, may improve patient outcomes. Our aim was to investigate whether patients on general wards receiving continuous vital signs monitoring have better outcomes than patients receiving usual care.

Methods

Meta-analysis of randomised controlled trials comparing non-critical care patients receiving continuous monitoring of vital signs to usual care. We searched Medline, Embase, and Web of Science, and assessed risk of bias with version 2 of the Cochrane risk-of-bias tool for randomised trials. In addition to measures related to the early detection of deterioration, we planned to present all patient outcomes reported by the clinical trials included.

Results

We included seven trials involving 1284 participants. There were no statistically significant differences in the four outcomes pooled. Comparing continuously monitored to normal care, the pooled odds for hospital mortality, major event/complication, and HDU/ICU admission was 0.95 (95% CI 0.59-1.53, p = 0.84; 660 participants, 3 studies), 0.71 (95% CI 0.38-1.31, p = 0.27; 948 participants, 4 studies) and 0.82 (95% CI 0.25-2.67, p = 0.74; 655 participants, 4 studies), respectively. The mean difference for length of stay was 2.12 days lower (95% CI -5.56 to 1.32, p = 0.23; 1034 participants, 6 studies).

Conclusion

We found no significant improvements in outcomes for patients continuously monitored compared to usual care. Further research is needed to understand what modalities of continuous monitoring may influence outcomes and investigate the implications of a telepresence service and multi-parameter scoring system.

Registration

PROSPERO CRD42023458656.

Development and Validation of a Machine Learning Algorithm Using Clinical Pages to Predict Imminent Clinical Deterioration

BACKGROUND

Early detection of clinical deterioration among hospitalized patients is a clinical priority for patient safety and quality of care. Current automated approaches for identifying these patients perform poorly at identifying imminent events.

OBJECTIVE

Develop a machine learning algorithm using pager messages sent between clinical team members to predict imminent clinical deterioration.

DESIGN

We conducted a large observational study using long short-term memory machine learning models on the content and frequency of clinical pages.

PARTICIPANTS

We included all hospitalizations between January 1, 2018 and December 31, 2020 at Vanderbilt University Medical Center that included at least one page message to physicians. Exclusion criteria included patients receiving palliative care, hospitalizations with a planned intensive care stay, and hospitalizations in the top 2% longest length of stay.

MAIN MEASURES

Model classification performance to identify in-hospital cardiac arrest, transfer to intensive care, or Rapid Response activation in the next 3-, 6-, and 12-hours. We compared model performance against three common early warning scores: Modified Early Warning Score, National Early Warning Score, and the Epic Deterioration Index.

KEY RESULTS

There were 87,783 patients (mean [SD] age 54.0 [18.8] years; 45,835 [52.2%] women) who experienced 136,778 hospitalizations. 6214 hospitalized patients experienced a deterioration event. The machine learning model accurately identified 62% of deterioration events within 3-hours prior to the event and 47% of events within 12-hours. Across each time horizon, the model surpassed performance of the best early warning score including area under the receiver operating characteristic curve at 6-hours (0.856 vs. 0.781), sensitivity at 6-hours (0.590 vs. 0.505), specificity at 6-hours (0.900 vs. 0.878), and F-score at 6-hours (0.291 vs. 0.220).

CONCLUSIONS

Machine learning applied to the content and frequency of clinical pages improves prediction of imminent deterioration. Using clinical pages to monitor patient acuity supports improved detection of imminent deterioration without requiring changes to clinical workflow or nursing documentation.

Information displays for automated surveillance algorithms of in-hospital patient deterioration: a scoping review

OBJECTIVE

Surveillance algorithms that predict patient decompensation are increasingly integrated with clinical workflows to help identify patients at risk of in-hospital deterioration. This scoping review aimed to identify the design features of the information displays, the types of algorithm that drive the display, and the effect of these displays on process and patient outcomes.

MATERIALS AND METHODS

The scoping review followed Arksey and O'Malley's framework. Five databases were searched with dates between January 1, 2009 and January 26, 2022. Inclusion criteria were: participants-clinicians in inpatient settings; concepts-intervention as deterioration information displays that leveraged automated AI algorithms; comparison as usual care or alternative displays; outcomes as clinical, workflow process, and usability outcomes; and context as simulated or real-world in-hospital settings in any country. Screening, full-text review, and data extraction were reviewed independently by 2 researchers in each step. Display categories were identified inductively through consensus.

RESULTS

Of 14 575 articles, 64 were included in the review, describing 61 unique displays. Forty-one displays were designed for specific deteriorations (eg, sepsis), 24 provided simple alerts (ie, text-based prompts without relevant patient data), 48 leveraged well-accepted score-based algorithms, and 47 included nurses as the target users. Only 1 out of the 10 randomized controlled trials reported a significant effect on the primary outcome.

CONCLUSIONS

Despite significant advancements in surveillance algorithms, most information displays continue to leverage well-understood, well-accepted score-based algorithms. Users' trust, algorithmic transparency, and workflow integration are significant hurdles to adopting new algorithms into effective decision support tools.

The predictive value of the modified early warning score for admission to the intensive care unit in patients with a hematologic malignancy - A multicenter observational study

OBJECTIVES

The modified early warning score (MEWS) is used to detect clinical deterioration of hospitalized patients. We aimed to investigate the predictive value of MEWS and derived quick Sequential Organ Failure Assessment (qSOFA) scores for intensive care unit admission in patients with a hematologic malignancy admitted to the ward.

DESIGN

Retrospective, observational study in two Dutch university hospitals.

SETTING

Data from adult patients with a hematologic malignancy, admitted to the ward over a 2-year period, were extracted from electronic patient files.

MAIN OUTCOME MEASURES

Intensive care admission.

RESULTS

We included 395 patients with 736 hospital admissions; 2% (n = 15) of admissions resulted in admission to the intensive care unit. A higher MEWS (OR 1.5; 95 %CI 1.3-1.80) and qSOFA (OR 4.4; 95 %CI 2.1-9.3) were associated with admission. Using restricted cubic splines, a rise in the probability of admission for a MEWS ≥ 6 was observed. The AUC of MEWS for predicting admission was 0.830, the AUC of qSOFA was 0.752. MEWS was indicative for intensive care unit admission two days before admission.

CONCLUSIONS

MEWS was a sensitive predictor of ICU admission in patients with a hematologic malignancy, superior to qSOFA. Future studies should confirm cut-off values and identify potential additional characteristics, to further enhance identification of critically ill hemato-oncology patients.

IMPLICATIONS FOR CLINICAL PRACTICE

The Modified Early Warning Score (MEWS) can be used as a tool for healthcare providers to monitor clinical deterioration and predict the need for intensive care unit admission in patients with a hematologic malignancy. Yet, consistent application and potential reevaluation of current thresholds is crucial. This will enable bedside nurses to more effectively identify patients needing adjunctive care, facilitating timely interventions and improved outcome.

Machine Learning Models Versus the National Early Warning Score System for Predicting Deterioration: Retrospective Cohort Study in the United Arab Emirates

BACKGROUND

Early warning score systems are widely used for identifying patients who are at the highest risk of deterioration to assist clinical decision-making. This could facilitate early intervention and consequently improve patient outcomes; for example, the National Early Warning Score (NEWS) system, which is recommended by the Royal College of Physicians in the United Kingdom, uses predefined alerting thresholds to assign scores to patients based on their vital signs. However, there is limited evidence of the reliability of such scores across patient cohorts in the United Arab Emirates.

OBJECTIVE

Our aim in this study was to propose a data-driven model that accurately predicts in-hospital deterioration in an inpatient cohort in the United Arab Emirates.

METHODS

We conducted a retrospective cohort study using a real-world data set that consisted of 16,901 unique patients associated with 26,073 inpatient emergency encounters and 951,591 observation sets collected between April 2015 and August 2021 at a large multispecialty hospital in Abu Dhabi, United Arab Emirates. The observation sets included routine measurements of heart rate, respiratory rate, systolic blood pressure, level of consciousness, temperature, and oxygen saturation, as well as whether the patient was receiving supplementary oxygen. We divided the data set of 16,901 unique patients into training, validation, and test sets consisting of 11,830 (70%; 18,319/26,073, 70.26% emergency encounters), 3397 (20.1%; 5206/26,073, 19.97% emergency encounters), and 1674 (9.9%; 2548/26,073, 9.77% emergency encounters) patients, respectively. We defined an adverse event as the occurrence of admission to the intensive care unit, mortality, or both if the patient was admitted to the intensive care unit first. On the basis of 7 routine vital signs measurements, we assessed the performance of the NEWS system in detecting deterioration within 24 hours using the area under the receiver operating characteristic curve (AUROC). We also developed and evaluated several machine learning models, including logistic regression, a gradient-boosting model, and a feed-forward neural network.

RESULTS

In a holdout test set of 2548 encounters with 95,755 observation sets, the NEWS system achieved an overall AUROC value of 0.682 (95% CI 0.673-0.690). In comparison, the best-performing machine learning models, which were the gradient-boosting model and the neural network, achieved AUROC values of 0.778 (95% CI 0.770-0.785) and 0.756 (95% CI 0.749-0.764), respectively. Our interpretability results highlight the importance of temperature and respiratory rate in predicting patient deterioration.

CONCLUSIONS

Although traditional early warning score systems are the dominant form of deterioration prediction models in clinical practice today, we strongly recommend the development and use of cohort-specific machine learning models as an alternative. This is especially important in external patient cohorts that were unseen during model development.

Systems-Based Physical Assessments: Earlier Detection of Clinical Deterioration and Reduced Mortality

BACKGROUND

Despite efforts to improve early detection of deterioration in a patient's condition, delays in activating the rapid response team remain common.

OBJECTIVES

To evaluate delays in activating the rapid response team and the occurrence of serious adverse events before and after implementation of a quality improvement initiative aimed at nurses' performing systems-based physical assessments.

METHODS

A retrospective observational cohort design was used to evaluate all patients who had a rapid response team activation during the study period.

RESULTS

A total of 1080 patients were included in the analysis: 536 patients before the quality improvement initiative and 544 patients after the quality improvement initiative. The delay in activating the rapid response team decreased from 11.7 hours in the before group to 9.6 hours in the after group (P < .001). In the after group, fewer patients were transferred to the intensive care unit (36% vs 41%, P = .02) and those who were transferred had 3.58 times greater odds of death than those who stayed at the same level of care. The after group had a 44% reduction in the odds of mortality compared with the before group.

CONCLUSIONS

When nurses focus on conducting a systems-based physical assessment early in their shift, delays in recognizing a patient's deteriorating condition are reduced, fewer patients are admitted to the intensive care unit, and mortality is significantly reduced.

Early Warning Scores to Support Continuous Wireless Vital Sign Monitoring for Complication Prediction in Patients on Surgical Wards: Retrospective Observational Study

BACKGROUND

Wireless vital sign sensors are increasingly being used to monitor patients on surgical wards. Although early warning scores (EWSs) are the current standard for the identification of patient deterioration in a ward setting, their usefulness for continuous monitoring is unknown.

OBJECTIVE

This study aimed to explore the usability and predictive value of high-rate EWSs obtained from continuous vital sign recordings for early identification of postoperative complications and compares the performance of a sensor-based EWS alarm system with manual intermittent EWS measurements and threshold alarms applied to individual vital sign recordings (single-parameter alarms).

METHODS

Continuous vital sign measurements (heart rate, respiratory rate, blood oxygen saturation, and axillary temperature) collected with wireless sensors in patients on surgical wards were used for retrospective simulation of EWSs (sensor EWSs) for different time windows (1-240 min), adopting criteria similar to EWSs based on manual vital signs measurements (nurse EWSs). Hourly sensor EWS measurements were compared between patients with (event group: 14/46, 30%) and without (control group: 32/46, 70%) postoperative complications. In addition, alarms were simulated for the sensor EWSs using a range of alarm thresholds (1-9) and compared with alarms based on nurse EWSs and single-parameter alarms. Alarm performance was evaluated using the sensitivity to predict complications within 24 hours, daily alarm rate, and false discovery rate (FDR).

RESULTS

The hourly sensor EWSs of the event group (median 3.4, IQR 3.1-4.1) was significantly higher (P<.004) compared with the control group (median 2.8, IQR 2.4-3.2). The alarm sensitivity of the hourly sensor EWSs was the highest (80%-67%) for thresholds of 3 to 5, which was associated with alarm rates of 2 (FDR=85%) to 1.2 (FDR=83%) alarms per patient per day respectively. The sensitivity of sensor EWS-based alarms was higher than that of nurse EWS-based alarms (maximum=40%) but lower than that of single-parameter alarms (87%) for all thresholds. In contrast, the (false) alarm rates of sensor EWS-based alarms were higher than that of nurse EWS-based alarms (maximum=0.6 alarm/patient/d; FDR=80%) but lower than that of single-parameter alarms (2 alarms/patient/d; FDR=84%) for most thresholds. Alarm rates for sensor EWSs increased for shorter time windows, reaching 70 alarms per patient per day when calculated every minute.

CONCLUSIONS

EWSs obtained using wireless vital sign sensors may contribute to the early recognition of postoperative complications in a ward setting, with higher alarm sensitivity compared with manual EWS measurements. Although hourly sensor EWSs provide fewer alarms compared with single-parameter alarms, high false alarm rates can be expected when calculated over shorter time spans. Further studies are recommended to optimize care escalation criteria for continuous monitoring of vital signs in a ward setting and to evaluate the effects on patient outcomes.

Outlier detection of vital sign trajectories from COVID-19 patients

In this work, we present a novel trajectory comparison algorithm to identify abnormal vital sign trends, with the aim of improving recognition of deteriorating health.There is growing interest in continuous wearable vital sign sensors for monitoring patients remotely at home. These monitors are usually coupled to an alerting system, which is triggered when vital sign measurements fall outside a predefined normal range. Trends in vital signs, such as increasing heart rate, are often indicative of deteriorating health, but are rarely incorporated into alerting systems.We introduce a dynamic time warp distance-based measure to compare time series trajectories. We split each multi-variable sign time series into 180 minute, non-overlapping epochs. We then calculate the distance between all pairs of epochs. Each epoch is characterized by its mean pairwise distance (average link distance) to all other epochs, with clusters forming with nearby epochs.We demonstrate in synthetically generated data that this method can identify abnormal epochs and cluster epochs with similar trajectories. We then apply this method to a real-world data set of vital signs from 8 patients who had recently been discharged from hospital after contracting COVID-19. We show how outlier epochs correspond well with the abnormal vital signs and identify patients who were subsequently readmitted to hospital.

Development of a dynamic prediction model for unplanned ICU admission and mortality in hospitalized patients

Frequent assessment of the severity of illness for hospitalized patients is essential in clinical settings to prevent outcomes such as in-hospital mortality and unplanned admission to the intensive care unit (ICU). Classical severity scores have been developed typically using relatively few patient features. Recently, deep learning-based models demonstrated better individualized risk assessments compared to classic risk scores, thanks to the use of aggregated and more heterogeneous data sources for dynamic risk prediction. We investigated to what extent deep learning methods can capture patterns of longitudinal change in health status using time-stamped data from electronic health records. We developed a deep learning model based on embedded text from multiple data sources and recurrent neural networks to predict the risk of the composite outcome of unplanned ICU transfer and in-hospital death. The risk was assessed at regular intervals during the admission for different prediction windows. Input data included medical history, biochemical measurements, and clinical notes from a total of 852,620 patients admitted to non-intensive care units in 12 hospitals in Denmark's Capital Region and Region Zealand during 2011-2016 (with a total of 2,241,849 admissions). We subsequently explained the model using the Shapley algorithm, which provides the contribution of each feature to the model outcome. The best model used all data modalities with an assessment rate of 6 hours, a prediction window of 14 days and an area under the receiver operating characteristic curve of 0.898. The discrimination and calibration obtained with this model make it a viable clinical support tool to detect patients at higher risk of clinical deterioration, providing clinicians insights into both actionable and non-actionable patient features.

Continuous Vital Signs Monitoring with a Wireless Device on a General Ward: A Survey to Explore Nurses' Experiences in a Post-Implementation Period

BACKGROUND

Nurse engagement, perceived need and usefulness affect healthcare technology use, acceptance and improvements in quality, safety and accessibility of healthcare. Nurses' opinions regarding continuous monitoring appear to be positive. However, facilitators and barriers were little studied. This study explored nurses' post-implementation experiences of the facilitators and barriers to continuously monitoring patients' vital signs using a wireless device on general hospital wards.

METHODS

This study employed a cross-sectional survey. Vocational and registered nurses from three general wards in a Dutch tertiary university hospital participated in a survey comprising open and closed questions. The data were analysed using thematic analysis and descriptive statistics.

RESULTS

Fifty-eight nurses (51.3%) completed the survey. Barriers and facilitators were identified under four key themes: (1) timely signalling and early action, (2) time savings and time consumption, (3) patient comfort and satisfaction and (4) preconditions.

CONCLUSIONS

According to nurses, early detection and intervention for deteriorating patients facilitate the use and acceptance of continuously monitoring vital signs. Barriers primarily concern difficulties connecting patients correctly to the devices and system.

Process Evaluation of a Wireless Wearable Continuous Vital Signs Monitoring Intervention in 2 General Hospital Wards: Mixed Methods Study

BACKGROUND

Continuous monitoring of vital signs (CMVS) using wearable wireless sensors is increasingly available to patients in general wards and can improve outcomes and reduce nurse workload. To assess the potential impact of such systems, successful implementation is important. We developed a CMVS intervention and implementation strategy and evaluated its success in 2 general wards.

OBJECTIVE

We aimed to assess and compare intervention fidelity in 2 wards (internal medicine and general surgery) of a large teaching hospital.

METHODS

A mixed methods sequential explanatory design was used. After thorough training and preparation, CMVS was implemented-in parallel with the standard intermittent manual measurements-and executed for 6 months in each ward. Heart rate and respiratory rate were measured using a chest-worn wearable sensor, and vital sign trends were visualized on a digital platform. Trends were routinely assessed and reported each nursing shift without automated alarms. The primary outcome was intervention fidelity, defined as the proportion of written reports and related nurse activities in case of deviating trends comparing early (months 1-2), mid- (months 3-4), and late (months 5-6) implementation periods. Explanatory interviews with nurses were conducted.

RESULTS

The implementation strategy was executed as planned. A total of 358 patients were included, resulting in 45,113 monitored hours during 6142 nurse shifts. In total, 10.3% (37/358) of the sensors were replaced prematurely because of technical failure. Mean intervention fidelity was 70.7% (SD 20.4%) and higher in the surgical ward (73.6%, SD 18.1% vs 64.1%, SD 23.7%; P<.001). Fidelity decreased over the implementation period in the internal medicine ward (76%, 57%, and 48% at early, mid-, and late implementation, respectively; P<.001) but not significantly in the surgical ward (76% at early implementation vs 74% at midimplementation [P=.56] vs 70.7% at late implementation [P=.07]). No nursing activities were needed based on vital sign trends for 68.7% (246/358) of the patients. In 174 reports of 31.3% (112/358) of the patients, observed deviating trends led to 101 additional bedside assessments of patients and 73 consultations by physicians. The main themes that emerged during interviews (n=21) included the relative priority of CMVS in nurse work, the importance of nursing assessment, the relatively limited perceived benefits for patient care, and experienced mediocre usability of the technology.

CONCLUSIONS

We successfully implemented a system for CMVS at scale in 2 hospital wards, but our results show that intervention fidelity decreased over time, more in the internal medicine ward than in the surgical ward. This decrease appeared to depend on multiple ward-specific factors. Nurses' perceptions regarding the value and benefits of the intervention varied. Implications for optimal implementation of CMVS include engaging nurses early, seamless integration into electronic health records, and sophisticated decision support tools for vital sign trend interpretation.

Comparison of seven prehospital early warning scores to predict long-term mortality: a prospective, multicenter, ambulance-based study

The long-term predictive validity of early warning scores (EWS) has not been fully elucidated yet. The aim of the present study is to compare seven prehospital EWS to predict 1-year mortality. A prospective, multicenter, ambulance-based study of adult patients with an acute illness involving six advanced life support units and 38 basic life support units, referring to five emergency departments in Spain. The primary outcome was long-term mortality with a 1-year follow-up. The compared scores included: National Early Warning Score 2, VitalPAC early warning score, modified rapid emergency medicine score (MREMS), Sepsis-related Organ Failure Assessment, Cardiac Arrest Risk Triage Score, Rapid Acute Physiology Score, and Triage Early Warning Score. Discriminative power [area under the receiver operating characteristic curve (AUC)] and decision curve analysis (DCA) were used to compare the scores. Additionally, a Cox regression and Kaplan-Meier method were used. Between 8 October 2019, and 31 July 2021, a total of 2674 patients were selected. The MREMS presented the highest AUC of 0.77 (95% confidence interval, 0.75-0.79), significantly higher than those of the other EWS. It also exhibited the best performance in the DCA and the highest hazard ratio for 1-year mortality [3.56 (2.94-4.31) for MREMS between 9 and 18 points, and 11.71 (7.21-19.02) for MREMS > 18]. Among seven tested EWS, the use of the MREMS presented better characteristics to predict 1-year mortality; however, all these scores present moderate performances.

Is Modified Early Warning Score associated with clinical outcomes of patients admitted to a university internal medicine ward?

OBJECTIVE

To assess the MEWS association with the clinical outcomes (CO) of patients admitted to an internal medicine ward (IMW) at a Brazilian university hospital (UH).

INTRODUCTION

It is important to quickly identify patients with clinical deterioration, especially in wards. The health team must recognize and act before the situation becomes an adverse event. In Brazil, nurses' work to overcome performance myths and the application of standardized predictive scales for patients in wards is still limited.

DESIGN

An observational cohort study designed and developed by a registered nurse that followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist.

METHODS

Data were collected from the IMW of a UH located in the city of São Paulo, Brazil (2017). An ROC curve was calculated to strengthen the use of a MEWS of < or ≥ 4 as a cutoff. CO of the two subgroups were compared.

RESULTS

Three hundred patients completed the study; their vital signs were recorded consecutively throughout hospitalization in the IMW. The highest MEWS value each day was considered for analysis. Scores < 4 were significantly associated with a higher probability of hospital discharge, a lower chance of transfer to the ICU, a lower total number of days of hospitalization, and a lower risk of death. Score ≥ 4 had worse CO (orotracheal intubation and cardiac monitoring), transfer to the ICU, and increased risk of death.

CONCLUSION

Scores < 4 were associated with positive outcomes, while scores ≥ 4 were associated with negative outcomes. MEWS can help prioritize interventions, increase certainty in decision-making, and improve patient safety, especially in a teaching IMW with medical teams undergoing professional development, thereby ensuring the central role of the nursing team in Brazil.

RELEVANCE FOR CLINICAL PRACTICE

MEWS aid nurses in identifying and managing patients, prioritizing interventions through assertive decision-making.

Early warning scores and trigger recommendations must be used with care in older home nursing care patients: Results from an observational study

AIMS

To explore modified early warning scores (MEWSs) and deviating vital signs among older home nursing care patients to determine whether the MEWS trigger recommendations were adhered to in cases of where registered nurses (RNs) suspected acute functional decline.

DESIGN

Prospective observational study with a descriptive, explorative design.

METHODS

Participants were included from April 2018 to February 2019. Demographic, health-related and clinical data were collected over a 3-month period.

RESULTS

In all, 135 older patients participated. Median MEWS (n = 444) was 1 (interquartile range (IQR) 1-2). Frequently deviating vital signs were respiratory (88.8%) and heart rate (15.3%). Median habitual MEWS (n = 51) was 1 (IQR 0-1). Deviating vital signs were respiratory (72.5%) and heart rate (19.6%). A significant difference between habitual MEWS and MEWS recorded in cases of suspected functional decline was found (p = 0.002). MEWS' trigger recommendations were adhered to in 68.9% of all MEWS measurements.

Implementation of a digital early warning score (NEWS2) in a cardiac specialist and general hospital settings in the COVID-19 pandemic: a qualitative study

OBJECTIVES

To evaluate implementation of digital National Early Warning Score 2 (NEWS2) in a cardiac care setting and a general hospital setting in the COVID-19 pandemic.

DESIGN

Thematic analysis of qualitative semistructured interviews using the non-adoption, abandonment, scale-up, spread, sustainability framework with purposefully sampled nurses and managers, as well as online surveys from March to December 2021.

SETTINGS

Specialist cardiac hospital (St Bartholomew's Hospital) and general teaching hospital (University College London Hospital, UCLH).

PARTICIPANTS

Eleven nurses and managers from cardiology, cardiac surgery, oncology and intensive care wards (St Bartholomew's) and medical, haematology and intensive care wards (UCLH) were interviewed and 67 were surveyed online.

RESULTS

Three main themes emerged: (1) implementing NEWS2 challenges and supports; (2) value of NEWS2 to alarm, escalate and during the pandemic; and (3) digitalisation: electronic health record (EHR) integration and automation. The value of NEWS2 was partly positive in escalation, yet there were concerns by nurses who undervalued NEWS2 particularly in cardiac care. Challenges, like clinicians' behaviours, lack of resources and training and the perception of NEWS2 value, limit the success of this implementation. Changes in guidelines in the pandemic have led to overlooking NEWS2. EHR integration and automated monitoring are improvement solutions that are not fully employed yet.

CONCLUSION

Whether in specialist or general medical settings, the health professionals implementing early warning score in healthcare face cultural and system-related challenges to adopting NEWS2 and digital solutions. The validity of NEWS2 in specialised settings and complex conditions is not yet apparent and requires comprehensive validation. EHR integration and automation are powerful tools to facilitate NEWS2 if its principles are reviewed and rectified, and resources and training are accessible. Further examination of implementation from the cultural and automation domains is needed.

Modified Sequential Organ Failure Assessment Score vs. Early Warning Scores in Prehospital Care to Predict Major Adverse Cardiac Events in Acute Cardiovascular Disease

(1) Background: The Modified Sequential Organ Failure Assessment (mSOFA) is an Early Warning Score (EWS) that has proven to be useful in identifying patients at high risk of mortality in prehospital care. The main objective of this study was to evaluate the predictive validity of prehospital mSOFA in estimating 2- and 90-day mortality (all-cause) in patients with acute cardiovascular diseases (ACVD), and to compare this validity to that of four other widely-used EWS. (2) Methods: We conducted a prospective, observational, multicentric, ambulance-based study in adults with suspected ACVD who were transferred by ambulance to Emergency Departments (ED). The primary outcome was 2- and 90-day mortality (all-cause in- and out-hospital). The discriminative power of the predictive variable was assessed and evaluated by the area under the curve (AUC) of the receiver operating characteristic (ROC). (3) Results: A total of 1540 patients met the inclusion criteria. The 2- and 90-day mortality rates were 5.3% and 12.7%, respectively. The mSOFA showed the highest AUC of all the evaluated scores for both 2- and 90-day mortality, AUC = 0.943 (0.917-0.968) and AUC = 0.874 (0.847-0.902), respectively. (4) Conclusions: The mSOFA is a quick and easy-to-use EWS with an excellent ability to predict mortality at both 2 and 90 days in patients treated for ACVD, and has proved to be superior to the other EWS evaluated in this study.

ChroniSense National Early Warning Score Study: Comparison Study of a Wearable Wrist Device to Measure Vital Signs in Patients Who Are Hospitalized

BACKGROUND

Wearable devices could be used to continuously monitor vital signs in patients who are hospitalized, but they require validation.

OBJECTIVE

This study aimed to evaluate the clinical validity of the prototype of a semiautomated wearable wrist device (ChroniSense Polso) to measure vital signs and provide National Early Warning Scores (NEWSs).

METHODS

Vital signs and NEWSs measured using the wearable device were compared with standard, nurse-lead manual measurements. We enrolled adult patients (aged ≥18 years) who required vital sign measurements at least every 6 hours in a UK teaching district general hospital. Wearable device measurements were not used for clinical decision-making. The primary outcome was the agreement on the individual National Early Warning parameter scores and vital sign measurements: respiratory rate, oxygen saturation, body temperature, systolic blood pressure, and heart rate. Secondary outcomes were the agreement on the total NEWS, incidence of adverse events, and user acceptance. To compare the wearable device measurements with the standard measurements, we analyzed vital sign measurements by limits of agreement (Bland-Altman analysis) and conducted κ agreement analyses for NEWSs. A user experience survey was conducted with questions about comfort of the wrist device, safety, preference, and use.

RESULTS

We included 132 participants in the study, with a mean age of 62 (SD 15.81) years; most of them were men (102/132, 77.3%). The highest weighted κ values were found for heart rate (0.69, 95% CI 0.57-0.81 for all 385 measurements) and systolic blood pressure (0.39, 95% CI 0.30-0.47 for all 339 measurements). Weighted κ values were low for respiration rate (0.03, 95% CI -0.001 to 0.05 for all 445 measurements), temperature (0, 95% CI 0-0 for all 231 measurements), and oxygen saturation (-0.11, 95% CI -0.20 to -0.02 for all 187 measurements). Weighted κ using Cicchetti-Allison weights showed κ of 0.20 (95% CI 0.03-0.38) when using all 56 total NEWSs. The user acceptance survey found that approximately half (45/91, 49%) of the participants found it comfortable to wear the device and liked its appearance. Most (85/92, 92%) of them said that they would wear the device during their next hospital visit, and many (74/92, 80%) said that they would recommend it to others.

CONCLUSIONS

This study shows the promising use of a prototype wearable device to measure vital signs in a hospital setting. Agreement between the standard measurements and wearable device measurements was acceptable for systolic blood pressure and heart rate, but needed to be improved for respiration rate, temperature, and oxygen saturation. Future studies need to improve the clinical validity of this wearable device. Large studies are required to assess clinical outcomes and cost-effectiveness of wearable devices for vital sign measurement.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR2-10.1136/bmjopen-2018-028219.

The Role of Early Warning Scoring Systems NEWS and MEWS in the Acute Exacerbation of COPD

Acute exacerbations of chronic obstructive pulmonary disease (AECOPD) are the most devastating events in the course of the disease. Our aim was to investigate the value of early warning scoring systems: National Early Warning Score (NEWS) and Modified Early Warning Score (MEWS) in AECOPD. This is a prospective observational study of patients with AECOPD who were admitted at hospital. The NEWS and MEWS scores were registered at admission (NEWS-d1, MEWS-d1) and on the second day (NEWS-d2, MEWS-d2). A nasopharyngeal and sputum sample was taken for culture. Follow-up was done at 3 and 6 months after hospitalization. Any possible correlations between NEWS and MEWS and other parameters of COPD were explored. A cohort of 64 patients were included. In-hospital mortality was 4.7% while total mortality at 6 months was 26%. We did not find any significant correlation between in-hospital mortality and any of the scores but we could show a higher mortality and more frequent AECOPD at 6 months of follow-up for those with higher NEWS-d2. NEWS-d2 was associated with higher pCO₂ at presentation and a more frequent use of NIV. Higher NEWS-d1 and NEWS-d2 were predictive of a longer hospital stay. The presence of pathogens in the nasopharyngeal sample was related with a higher reduction of both scores on the second day. We therefore support the superiority of NEWS in the evaluation of hospitalized patients with AECOPD. A remaining high NEWS at the second day of hospital stay signals a high risk of hypercapnia and need of NIV but also higher mortality and more frequent exacerbations at 6 months after AECOPD.

Implementation and clinical utility of a Computer-Aided Risk Score for Mortality (CARM): a qualitative study

OBJECTIVES

The Computer-Aided Risk Score for Mortality (CARM) estimates the risk of in-hospital mortality following acute admission to the hospital by automatically amalgamating physiological measures, blood tests, gender, age and COVID-19 status. Our aims were to implement the score with a small group of practitioners and understand their first-hand experience of interacting with the score in situ.

DESIGN

Pilot implementation evaluation study involving qualitative interviews.

SETTING

This study was conducted in one of the two National Health Service hospital trusts in the North of England in which the score was developed.

PARTICIPANTS

Medical, older person and ICU/anaesthetic consultants and specialist grade registrars (n=116) and critical outreach nurses (n=7) were given access to CARM. Nine interviews were conducted in total, with eight doctors and one critical care outreach nurse.

INTERVENTIONS

Participants were given access to the CARM score, visible after login to the patients' electronic record, along with information about the development and intended use of the score.

RESULTS

Four themes and 14 subthemes emerged from reflexive thematic analysis: (1) current use (including support or challenge clinical judgement and decision making, communicating risk of mortality and professional curiosity); (2) barriers and facilitators to use (including litigation, resource needs, perception of the evidence base, strengths and limitations), (3) implementation support needs (including roll-out and integration, access, training and education); and (4) recommendations for development (including presentation and functionality and potential additional data). Barriers and facilitators to use, and recommendations for development featured highly across most interviews.

CONCLUSION

Our in situ evaluation of the pilot implementation of CARM demonstrated its scope in supporting clinical decision making and communicating risk of mortality between clinical colleagues and with service users. It suggested to us barriers to implementation of the score. Our findings may support those seeking to develop, implement or improve the adoption of risk scores.

Tracking the National Early Warning Score 2 from Prehospital Care to the Emergency Department: A Prospective, Ambulance-Based, Observational Study

Aim of the study: To assess the prognostic ability of the National Early Warning Score 2 (NEWS2) at three time points of care -at the emergency scene (NEWS2-1), just before starting the transfer by ambulance to the hospital (NEWS2- 2), and at the hospital triage box (NEWS2-3)- to estimate in-hospital mortality after two days since the index event.Methods: Prospective, multicenter, ambulance-based, cohort ongoing study in adults (>18 years) consecutively attended by advanced life support (ALS) and evacuated with high-priority to the emergency departments (ED) between October 2018 and May 2021. Vital sign measures were used to calculate the NEWS2 score at each time point, then this score was entered in a logistic regression model as the single predictor. Two outcomes were considered: first, all-cause mortality of the patients within 2 days of presentation to EMS, and second, unplanned ICU admission. The calibration and scores comparison was performed by representing the predicted vs the observed risk curves according to NEWS score value.Results: 4943 patients were enrolled. Median age was 69 years (interquartile range 53- 81). The NEWS2-3 presented the better performance for all-cause two-day in-hospital mortality with an AUC of 0.941 (95% CI: 0.917-0.964), showing statistical differences with both the NEWS2-1 (0.872 (95% CI: 0.833-0.911); p < 0.003) and with the NEWS2- 2 (0.895 (95% CI: 0.866-0.925; p < 0.05). The calibration and scores comparison results showed that the NEWS2-3 was the best predictive score followed by the NEWS2-2 and the NEWS2-1, respectively.Conclusions: The NEWS2 has an excellent predictive performance. The score showed a very consistent response over time with the difference between "at the emergency scene" and "pre-evacuation" presenting the sharpest change with decreased threshold values, thus displaying a drop in the risk of acute clinical impairment.

Utopia versus dystopia: Professional perspectives on the impact of healthcare artificial intelligence on clinical roles and skills

BACKGROUND

Alongside the promise of improving clinical work, advances in healthcare artificial intelligence (AI) raise concerns about the risk of deskilling clinicians. This purpose of this study is to examine the issue of deskilling from the perspective of diverse group of professional stakeholders with knowledge and/or experiences in the development, deployment and regulation of healthcare AI.

METHODS

We conducted qualitative, semi-structured interviews with 72 professionals with AI expertise and/or professional or clinical expertise who were involved in development, deployment and/or regulation of healthcare AI. Data analysis using combined constructivist grounded theory and framework approach was performed concurrently with data collection.

FINDINGS

Our analysis showed participants had diverse views on three contentious issues regarding AI and deskilling. The first involved competing views about the proper extent of AI-enabled automation in healthcare work, and which clinical tasks should or should not be automated. We identified a cluster of characteristics of tasks that were considered more suitable for automation. The second involved expectations about the impact of AI on clinical skills, and whether AI-enabled automation would lead to worse or better quality of healthcare. The third tension implicitly contrasted two models of healthcare work: a human-centric model and a technology-centric model. These models assumed different values and priorities for healthcare work and its relationship to AI-enabled automation.

CONCLUSION

Our study shows that a diverse group of professional stakeholders involved in healthcare AI development, acquisition, deployment and regulation are attentive to the potential impact of healthcare AI on clinical skills, but have different views about the nature and valence (positive or negative) of this impact. Detailed engagement with different types of professional stakeholders allowed us to identify relevant concepts and values that could guide decisions about AI algorithm development and deployment.