Skip to main content
Download PDF

Increased time from physiological derangement to critical care admission associates with mortality

Critical Care volume 25, Article number: 226 (2021) Cite this article

Abstract

Background

Rapid response systems aim to achieve a timely response to the deteriorating patient; however, the existing literature varies on whether timing of escalation directly affects patient outcomes. Prior studies have been limited to using ‘decision to admit’ to critical care, or arrival in the emergency department as ‘time zero’, rather than the onset of physiological deterioration. The aim of this study is to establish if duration of abnormal physiology prior to critical care admission [‘Score to Door’ (STD) time] impacts on patient outcomes.

Methods

A retrospective cross-sectional analysis of data from pooled electronic medical records from a multi-site academic hospital was performed. All unplanned adult admissions to critical care from the ward with persistent physiological derangement [defined as sustained high National Early Warning Score (NEWS) > / = 7 that did not decrease below 5] were eligible for inclusion. The primary outcome was critical care mortality. Secondary outcomes were length of critical care admission and hospital mortality. The impact of STD time was adjusted for patient factors (demographics, sickness severity, frailty, and co-morbidity) and logistic factors (timing of high NEWS, and out of hours status) utilising logistic and linear regression models.

Results

Six hundred and thirty-two patients were included over the 4-year study period, 16.3% died in critical care. STD time demonstrated a small but significant association with critical care mortality [adjusted odds ratio of 1.02 (95% CI 1.0–1.04, p = 0.01)]. It was also associated with hospital mortality (adjusted OR 1.02, 95% CI 1.0–1.04, p = 0.026), and critical care length of stay. Each hour from onset of physiological derangement increased critical care length of stay by 1.2%. STD time was influenced by the initial NEWS, but not by logistic factors such as out-of-hours status, or pre-existing patient factors such as co-morbidity or frailty.

Conclusion

In a strictly defined population of high NEWS patients, the time from onset of sustained physiological derangement to critical care admission was associated with increased critical care and hospital mortality. If corroborated in further studies, this cohort definition could be utilised alongside the ‘Score to Door’ concept as a clinical indicator within rapid response systems.

Introduction

The majority of unexpected hospital cardiac arrests, deaths, or unplanned critical care admissions are preceded by a trajectory of physiological deterioration [1,2,3,4], and it is suggested that these events can be anticipated, and potentially prevented. Whilst the development and utility of early warning scores have been criticised [5,6,7], rapid response systems have been associated with improvements in outcome [8, 9]. Specifically, the National Early Warning Score (NEWS) is an established system for detection of the deteriorating patient, (based upon deviation from normal physiology in 6 parameters; respiratory rate; SpO2; systolic blood pressure; heart rate; level of consciousness and temperature, with additional weighting for supplemental oxygen) in use across the United Kingdom (UK) National Health Service (NHS) since 2012. It was designed to standardise grading of acute illness severity using an ordinal scale from zero to 20 and has been shown to be predictive of adverse outcomes [10,11,12].

In 2011, the ‘score to door time’ [13] was proposed as a benchmarking tool to assess timeliness of the response to deterioration, and subsequent admission to critical care. The authors identified that organisational as well as clinical factors may influence the timeliness of admission; however, the relationship with clinical outcomes was not explored.

A number of studies have demonstrated that timely critical care outreach team (CCOT) response [14] and prompt admission to critical care [15,16,17] are associated with improvements in patient outcome whilst other studies have failed to show this relationship [18, 19]. There has been an imperfect definition of ‘time zero’, with studies commonly using the time of decision to admit (DTA) to critical care [15, 18, 20] rather than the onset of physiological deterioration and several studies have focused solely on the Emergency Department (ED) [15, 16, 21]. Furthermore, there are multiple confounders that may influence both time to admission and patient outcomes, including critical care capacity [17, 20], chronic health and frailty [22], surgical versus medical status [19], and acute sickness severity [13, 18] and these may account for the heterogeneity in the literature. Adjustment for differences in patient characteristics significantly alter the apparent relationship between timeliness and outcome [20]. To date, no studies have explored time from onset of physiological deterioration in ward patients to critical care admission, as a predictor of outcome.

Despite this, in 2019, the UK NHS utilised the Commissioning for Quality and Innovation (CQUIN) mechanism to focus a quality improvement effort on more timely response to deterioration on wards and admission to critical care, with the intended benefit of reducing critical care length of stay and hospital mortality [23]. For this metric, the start of the pathway was not clearly defined, which may lead to problematic heterogeneity.

Objectives

We aim to describe a derived cohort of unplanned admissions to critical care with a high level of plausibility to benefit from prompt admission, and to test the association between duration of physiological derangement with subsequent patient outcomes. Additionally, we aim to explore factors associated with the Score to Door time.

Methods

Study design and setting

A retrospective study was conducted using unplanned admissions to critical care in two hospitals at a single academic medical centre [Guys’ and St Thomas’ Foundation Trust, (GSTT)] between 1/1/2017 and 1/03/2021.

The hospitals have a fully deployed NEWS protocol, utilising observations recorded in real time into an electronic health record. Observation frequency was protocolised in accordance with NEWS [10]. The system incorporates advisory notifications to encourage adherence to the escalation policy, and a population level report that allowed oversight of all high NEWS patients across the hospital, but without specific electronic alerts of high NEWS to caregivers [24].

In accordance with the NEWS protocol, patients with a NEWS of 5 or 6 should trigger an urgent assessment by a clinician with acute care skills, whilst a NEWS of greater than, or equal to 7 should trigger an emergency response by a team with critical care skills (the CCOT).

A 24/7 CCOT was present, staffed by advanced nurse practitioners with dedicated middle and senior grade critical care doctors. Escalation to the CCOT was required for any admission to critical care and admission decisions were made by the attending critical care consultant. The critical care service compromised a 72 bedded, mixed level 2/3 service. During the COVID-19 pandemic, this service underwent expansion up to 210 beds with additional surge staffing.

Non-invasive ventilation (NIV) was only provided in critical care areas, whilst high flow nasal cannula oxygen could be delivered in selected ward areas with the support of trained ward teams and/or the CCOT. Additional protocols for the urgent management of sepsis [25] and acute kidney injury [26] on the wards were present.

Study population

We included adult patients with unplanned admissions to critical care from the ward, whom prior to admission had new and sustained physiological derangement. This was defined as the first occurrence of a NEWS greater than or equal to 7 that did not subsequently decrease below 5 in the interval prior to critical care admission. This definition was utilised as it aligns with existing NEWS trigger thresholds, whilst including a degree of tolerance in recognition that physiological parameters may naturally vary and/or there may be a partial response to treatment. Unplanned admission was defined as admission to a level 2 or 3 critical care unit from a level 1 ward, without an intervening operating theatre visit. A convenience sample was used from the onset of our study database to present. Patients were excluded if they were admitted to critical care directly from the ED or if time between initial high NEWS and critical care admission was > 7 days.

Exposures and outcomes

The primary exposure was time from initial high NEWS of greater than or equal to seven, to critical care admission (‘Score to Door’ time). Time of critical care admission was defined as the time of the first recorded heart rate, to avoid errors associated with administrative processes. The primary outcome was mortality during the first critical care admission following initial fulfilment of study inclusion criteria. Secondary outcomes were length of index critical care admission and hospital mortality, which may have included death during subsequent critical care admissions.

Data collection

Data were extracted from the GSTT Data Warehouse, which serves as an aggregate repository of data from multiple electronic sources, developed using the Health Catalyst® Data Operating System (DOS™) (Health Catalyst, Inc. Salt Lake City, Utah). Sickness severity was estimated using the sequential organ failure assessment (SOFA) [27] on day one of the critical care admission. Co-morbidity was assessed using the Charlson–Deyo co-morbidity index [28] and frailty was assessed using the Dr Foster Global Frailty Score [29]. Out-of-hours was defined as 00:00 Saturday to 00:00 Monday, and 20:00–07.59 on weekdays. Block descriptions for primary ICD-10 codes were used to categorise primary diagnoses. Cases with sepsis were identified by ICD-10 codes A40/A41 [30].

Statistical analysis

Continuous variables are presented with median and interquartile range and categorical variables with count and percentage. Between-group comparisons are made using the Mann–Whitney U test for continuous and Fisher’s Exact test for binary categorical variables. Patients with missing data in any variables of interest were excluded, resulting in no missing data.

The association between potential predictor variables and both critical care and hospital mortality were assessed using logistic regression. Logistic models were constructed in a stepwise fashion. Significant variables were determined by univariable logistic regression at p < 0.2. Significant variables were then further assessed in a multivariable logistic regression model that included Score to Door time.

Multivariable linear regression was used to explore the relationship between potential predictor variables and Score to Door time, as well as critical care length of stay. Variables were chosen based on their plausible contribution to the target variable. Dependent variables underwent natural log transformation due to their skewed distribution.

A sensitivity analysis was performed by repeating the final multivariable models on a dataset with missing data imputed rather than excluded. Imputation was performed by replacing missing values with the median of the available data for each variable.

Confidence intervals for multivariable models were calculated using the bias-corrected accelerated bootstrap method. A threshold of p < 0.05 was used to determine statistical significance in final models and comparative tests. All data manipulation and analysis was  performed using Python 3.8 [31,32,33,34].

The study was registered locally and was conducted as a service evaluation as defined by the U.K. NHS Health Research Authority (http://www.hra.nhs.uk) using anonymous, routinely collected data and therefore did not require review by the Research Ethics Committee.

Results

Characteristics and outcomes

Between 1/1/17 and 1/3/21 there were 6549 unplanned admissions to critical care. 3271 were excluded due to direct admission from ED, 2563 were excluded as they did not meet criteria for sustained high NEWS and 83 were excluded due to missing data (summary of missing data presented in Additional file 1: Table 1). This resulted in 632 patients included in the final analysis (Flowchart in Additional file 1: Fig. 1). The characteristics and outcomes of analysed patients are presented in Table 1. Baseline demographics of survivors and decedents were similar, excepting a higher frailty score in patients who survived their index critical care admission.

Table 1 Characteristics and outcomes of the study population
Full size table

Score to Door time and critical care mortality

The median Score to Door time was 6.3 [3.8–10.2] hours with most patients admitted to critical care within 24 h of the onset of physiological derangement (Fig. 1). Variables utilised for stepwise model construction are indicated in Table 1, with the univariable results reported in Additional file 1: Table 2 and final multivariable models reported in Table 2.

Fig. 1
figure 1

Distribution of time to critical care admission from initial high NEWS score (Score to Door time) (x-axis logarithmically scaled). NEWS, national early warning score

Full size image
Table 2 Multivariable logistic and log-linear regression models. Odds ratios presented for logistic models and coefficients presented for log-linear models
Full size table

Score to Door time demonstrated a small but significant association with critical care mortality with an unadjusted odds ratio of 1.02 (95% CI 1.0–1.03, p = 0.063) in a univariable model and an adjusted odds ratio of 1.02 (95% CI 1.0–1.04, p = 0.01) in the multivariable model. Age, SOFA score, need for NIV or invasive mechanical ventilation (IMV), and duration of hospital admission prior to triggering high NEWS also had a significant association with critical care mortality. Neither co-morbidity nor frailty had a statistically significant relationship.

In Fig. 2, Score to Door time is considered in 2 hourly blocks with mean critical care mortality and admission SOFA score plotted for each block. There is a trend to increasing critical care mortality with delay to admission despite a downward trend in day one SOFA score.

Fig. 2
figure 2

A Critical care mortality plotted against time from initial high NEWS score to critical care admission. B Admission SOFA score plotted against time from initial high NEWS score to critical care admission. Both plots divided into two hourly blocks with mean and 95% confidence intervals plotted. Single points represent blocks with only a single patient. Unadjusted linear regression ‘line of best fit’ with 95% confidence interval plotted in both A and B. Time from high NEWS score to critical care admission clipped at 48 h to more clearly demonstrate the majority of data. NEWS, national early warning score; SOFA, sequential organ failure assessment

Full size image

Score to Door time and hospital mortality

The association with hospital mortality was also examined. Score to Door time had a significant association with hospital mortality (adjusted OR 1.02, 95% CI 1.0–1.04, p = 0.026). Age, SOFA score, need for NIV or IMV, and time to triggering high NEWS were also significantly associated with hospital mortality. Triggering high NEWS value and elective admission type did not reach statistical significance.

Score to Door time and critical care length of stay

In patients who survived to discharge from critical care, a multivariable log-linear regression was constructed with critical care length of stay as the dependent variable (Table 2). Score to Door time was significantly associated with critical care length of stay, with each hour increasing length of stay by 1.2% (β = 0.012, 95% CI 0.01–0.02, p = 0.024). Use of NIV or IMV and sepsis status were also significantly associated. Age, frailty, co-morbidity and SOFA score were not significantly associated with critical care length of stay.

Determinants of Score to Door time

A multivariable log-linear regression was constructed with Score to Door time as the dependent variable (Table 2). The value of the triggering high NEWS was the only variable significantly associated with time to critical care admission with each point increment reducing the  time by 7.9% (β =  − 0.079, 95% CI − 0.11 to − 0.04, p = 0.007). This relationship is visualised in Fig. 3. Age, frailty, co-morbidity, time to high NEWS, out of hours triggering NEWS and peri-/cardiac arrest call prior to critical care did not demonstrate a significant relationship.

Fig. 3
figure 3

Boxplot of Score to Door time for initial high NEWS value. Whiskers 1.5 × IQR. NEWS; national early warning score; IQR; inter-quartile range

Full size image

Sensitivity analysis

Imputation of missing values did not change the level of significance of any of the covariates analysed in the multivariable logistic or linear models, with the detailed results presented in Additional file 1: Table 3.

Discussion

In this study of 632 unplanned critical care admissions, and following adjustment for patient demographics, sickness severity, co-morbidity and frailty, the time from onset of sustained physiological deterioration to critical care admission was associated with a small but important increase in both critical care and hospital mortality, as well as a longer critical care length of stay. The Score to Door time was influenced by the triggering NEWS value, but not by logistic factors such as out-of-hours status, nor by patient factors such as co-morbidity or frailty.

Comparison with the literature

Whilst it is intuitive that more rapid admission from wards to critical care following a high NEWS trigger will be beneficial, this is the first study to our knowledge to demonstrate this. The previous literature has either focused on patients in the ED [15, 16, 21], or has used the time of decision to admit [15, 18, 20]. Focussing solely on the efferent pathway ignores an important process of detection, escalation, response, and decision. Barriers to prompt recognition and escalation include a lack of timely or complete observations, poor communication, delayed response by CCOTs, or lack of senior decision maker [35], during which time the patient may suffer further deterioration [20]. We advocate for a whole-systems approach to improving both the afferent and efferent response to deterioration [36], and merely increasing critical care bed capacity will not necessarily address the challenges. Indeed, the benefit may be mediated by the prompt recognition, escalation and CCOT response, rather than critical care admission per se [14].

Whilst it is recommended to admit within 4 hours [21, 23], we found no particular inflection and it is likely that the increased risk is continuous if not linear.

Strengths and weaknesses

Our study captures the entire period from the onset of deterioration to admission for definitive care. We utilised a strict definition of the cohort to include only patients who had sustained physiological abnormalities and with plausible potential to benefit from prompt admission. By doing so, we reduced heterogeneity before then adjusting for factors that may influence the timeliness of critical care admission, most notably sickness severity. This is important because it ought to be expected that sicker patients are admitted more quickly and this may confound the relationship with outcome [18, 20]. This study goes a step further than previous publications by proposing a clear definition of an at-risk cohort based on established NEWS thresholds that can be replicated in other studies and could be used for benchmarking. By examining the effect of logistic as well as clinical factors, our data provides some insights into the question of differential standards in out-of-hours care—an area of significant contention in healthcare [37].

However, we recognise certain limitations. This relatively small, single centre and retrospective study occurred in an academic institution with a relatively high ratio of critical care to ward level beds and as such, these findings cannot necessarily be generalised to other settings. We speculate that the magnitude of the effect would be greater in hospitals that have more limited access to critical care and where delays may be greater. Whilst the strictly defined cohort limits heterogeneity, it may also reduce generalisability and the findings cannot be extrapolated to all patients with elevated NEWS. It is noted that despite the selected cohort, the effect size is small.

We utilised critical care mortality as a primary outcome due to its plausible relationship to Score to Door time but we were unable to examine patient centred outcomes following discharge from hospital.

Inherent to the retrospective design, this study does not prove causality and there may be other factors that influence clinician or system behaviour with respect to escalation and response that are themselves associated with outcome and that are not captured by our co-variates. For example, some patients may have had baseline abnormal physiology (equating to ‘false positive’ NEWS triggers) or limitation of treatment orders that affected the Score to Door time and also outcome; furthermore we were unable to control for critical care capacity and organisational stress—delays occur when the system is under increased pressure and it may be this more general phenomenon that affects care and outcome [38]. Finally, we have only used surrogates of the start and end of the pathway and there may be inaccuracies in determining the true time of deterioration or critical care admission.

Future directions

It is unknown if the relationship demonstrated exists across all strata of sickness severity or if it applies to alternative definitions of the target population.. This deserves further investigation, as patients with lower NEWS triggers are likely to be more numerous and experience longer waits to definitive treatment. The previous literature [20] has suggested that prompt admission may be less impactful for lower acuity critical care admissions and the association between timeliness of critical care admission and outcomes in a broader population of deteriorating patients should be validated before wholesale policy directives to improve care. Novel interpretations of NEWS over time, such as ‘area under the curve’ measurements, may aid in expanding the definition used in this paper to incorporate patients with lower absolute NEWS values. Finally, the impact of prolonged physiological derangement in patients ultimately not admitted to critical care also requires investigation, as this will aide in the development of future practice guidelines for managing acute deterioration.

Conclusions

In a strictly defined population of high NEWS patients, the time from onset of sustained physiological derangement to critical care admission is associated with increased critical care and hospital mortality. If corroborated in larger studies, this would support the use of this population definition and the ‘Score to Door’ concept to standardise reporting and drive quality improvement initiatives in rapid response systems.

Availability of data and materials

All data generated or analysed during this study are included in this published article and its supplementary information files. The corresponding author may provide specified analyses or fully de-identified parts of the dataset upon reasonable request.

Abbreviations

STD:

Score to door time (hours from onset of physiological derangement to critical care admission)

NEWS:

National early warning score

CCOT:

Critical care outreach team

DTA:

Decision to admit (to critical care)

CQUIN:

Commissioning for quality and innovation

GSTT:

Guys’ and St Thomas’ Foundation Trust

NIV:

Non-invasive ventilation

HFNC:

High flow nasal cannula

IMV:

Invasive mechanical ventilation

SOFA:

Sequential organ failure assessment

ED:

Emergency department

References

  1. Hillman KM, Bristow PJ, Chey T, et al. Antecedents to hospital deaths. Intern Med J. 2001;31:343–8.

    Article  CAS  Google Scholar 

  2. Kause J, Smith G, Prytherch D, et al. A comparison of antecedents to cardiac arrests, deaths and emergency intensive care admissions in Australia and New Zealand, and the United Kingdom–the ACADEMIA study. Resuscitation. 2004;62:275–82.

    Article  Google Scholar 

  3. Schein RM, Hazday N, Pena M, et al. Clinical antecedents to in-hospital cardiopulmonary arrest. Chest. 1990;98:1388–92.

    Article  CAS  Google Scholar 

  4. Goldhill DR, White SA, Sumner A. Physiological values and procedures in the 24 h before ICU admission from the ward. Anaesthesia. 1999;54:529–34.

    Article  CAS  Google Scholar 

  5. Alam N, Hobbelink EL, van Tienhoven AJ, et al. The impact of the use of the early warning score (EWS) on patient outcomes: a systematic review. Resuscitation. 2014;85:587–94.

    Article  CAS  Google Scholar 

  6. Gerry S, Bonnici T, Birks J, et al. Early warning scores for detecting deterioration in adult hospital patients: systematic review and critical appraisal of methodology. BMJ. 2020;369:m1501.

    Article  Google Scholar 

  7. Romero-Brufau S, Huddleston JM, Escobar GJ, et al. Why the C-statistic is not informative to evaluate early warning scores and what metrics to use. Crit Care. 2015;19:285.

    Article  Google Scholar 

  8. Maharaj R, Raffaele I, Wendon J. Rapid response systems: a systematic review and meta-analysis. Crit Care. 2015;19:254.

    Article  Google Scholar 

  9. Solomon RS, Corwin GS, Barclay DC, et al. Effectiveness of rapid response teams on rates of in-hospital cardiopulmonary arrest and mortality: a systematic review and meta-analysis. J Hosp Med. 2016;11:438–45.

    Article  Google Scholar 

  10. Royal College of Physicians: National Early Warning Score (NEWS) 2 [Internet]. RCP London 2017; [cited 2021 Mar 22]. https://www.rcplondon.ac.uk/projects/outputs/national-early-warning-score-news-2.

  11. Smith GB, Prytherch DR, Meredith P, et al. The ability of the National Early Warning Score (NEWS) to discriminate patients at risk of early cardiac arrest, unanticipated intensive care unit admission, and death. Resuscitation. 2013;84:465–70.

    Article  Google Scholar 

  12. Pimentel MAF, Redfern OC, Gerry S, et al. A comparison of the ability of the National Early Warning Score and the National Early Warning Score 2 to identify patients at risk of in-hospital mortality: a multi-centre database study. Resuscitation. 2019;134:147–56.

    Article  Google Scholar 

  13. Oglesby KJ, Durham L, Welch J, et al. “Score to Door Time”, a benchmarking tool for rapid response systems: a pilot multi-centre service evaluation. Crit Care. 2011;15:R180.

    Article  Google Scholar 

  14. Tillmann BW, Klingel ML, McLeod SL, et al. The impact of delayed critical care outreach team activation on in-hospital mortality and other patient outcomes: a historical cohort study. Can J Anesth/J Can Anesth. 2018;65:1210–7.

    Article  Google Scholar 

  15. Cardoso LTQ, Grion CMC, Matsuo T, et al. Impact of delayed admission to intensive care units on mortality of critically ill patients: a cohort study. Crit Care. 2011;15:R28.

    Article  Google Scholar 

  16. Hsieh C-C, Lee C-C, Hsu H-C, et al. Impact of delayed admission to intensive care units on patients with acute respiratory failure. Am J Emerg Med. 2017;35:39–44.

    Article  Google Scholar 

  17. Robert R, Reignier J, Tournoux-Facon C, et al. Refusal of intensive care unit admission due to a full unit: impact on mortality. Am J Respir Crit Care Med. 2012;185:1081–7.

    Article  Google Scholar 

  18. Stohl S, Sprung C, Lippert A, et al. Impact of triage-to-admission time on patient outcome in European intensive care units: a prospective, multi-national study. J Crit Care. 2019;53:11–7.

    Article  Google Scholar 

  19. O’Callaghan DJ, Jayia P, Vaughan-Huxley E, et al. An observational study to determine the effect of delayed admission to the intensive care unit on patient outcome. Crit Care. 2012;16:R173.

    Article  Google Scholar 

  20. Harris S, Singer M, Sanderson C, et al. Impact on mortality of prompt admission to critical care for deteriorating ward patients: an instrumental variable analysis using critical care bed strain. Intensive Care Med. 2018;44:606–15.

    Article  Google Scholar 

  21. Hung S-C, Kung C-T, Hung C-W, et al. Determining delayed admission to intensive care unit for mechanically ventilated patients in the emergency department. Crit Care. 2014;18:485.

    Article  Google Scholar 

  22. So RKL, Bannard-Smith J, Subbe CP, et al. The association of clinical frailty with outcomes of patients reviewed by rapid response teams: an international prospective observational cohort study. Crit Care. 2018;22:227.

    Article  Google Scholar 

  23. NHS England » CCG CQUIN 2020/21 Indicators Specifications [Internet]. [cited 2021 Mar 22]. https://www.england.nhs.uk/publication/ccg-cquin-2020-21-indicators-specifications/.

  24. Subbe CP, Duller B, Bellomo R. Effect of an automated notification system for deteriorating ward patients on clinical outcomes. Crit Care. 2017;21:52.

    Article  Google Scholar 

  25. Overview | Sepsis: recognition, diagnosis and early management | Guidance | NICE [Internet]. [cited 2021 Mar 22]. https://www.nice.org.uk/guidance/ng51.

  26. Think Kidneys: Minimum Care Bundle for Patients with AKI in Hospital [Internet]. Acute Kidney Injury [cited 2021 Mar 22]. https://www.thinkkidneys.nhs.uk/aki/wp-content/uploads/sites/2/2020/03/AKI-care-bundle-2020.pdf.

  27. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22:707–10.

    Article  CAS  Google Scholar 

  28. Ladha KS, Zhao K, Quraishi SA, et al. The Deyo-Charlson and Elixhauser-van Walraven Comorbidity Indices as predictors of mortality in critically ill patients. BMJ Open. 2015;5:e008990.

    Article  Google Scholar 

  29. Soong JTY, Kaubryte J, Liew D, et al. Dr Foster global frailty score: an international retrospective observational study developing and validating a risk prediction model for hospitalised older persons from administrative data sets. BMJ Open. 2019;9:e026759.

    Article  Google Scholar 

  30. Inada-Kim M, Page B, Maqsood I, et al. Defining and measuring suspicion of sepsis: an analysis of routine data. BMJ Open. 2017;7:e014885.

    Article  Google Scholar 

  31. The pandas development team: pandas-dev/pandas: Pandas [Internet]. 2020. https://doi.org/10.5281/zenodo.3509134.

  32. Seabold, Skipper, Josef Perktold: statsmodels: Econometric and statistical modeling with python. 2010.

  33. Waskom M, Gelbart M, Botvinnik O, et al.: mwaskom/seaborn: v0.11.1 (December 2020) [Internet]. Zenodo; 2020. [cited 2021 Mar 22]. https://zenodo.org/record/592845.

  34. SciPy 1.0 Contributors, Virtanen P, Gommers R, et al.: SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat. Methods 2020; 17:261–72.

  35. Difonzo M. Performance of the afferent limb of rapid response systems in managing deteriorating patients: a systematic review. Crit Care Res Pract. 2019;2019:e6902420.

    Google Scholar 

  36. Vincent J-L, Einav S, Pearse R, et al. Improving detection of patient deterioration in the general hospital ward environment. Eur J Anaesthesiol. 2018;35:325–33.

    Article  Google Scholar 

  37. Galloway M, Hegarty A, McGill S, et al. The effect of ICU out-of-hours admission on mortality: a systematic review and meta-analysis. Crit Care Med. 2018;46:290–9.

    Article  Google Scholar 

  38. Gabler NB, Ratcliffe SJ, Wagner J, et al. Mortality among patients admitted to strained intensive care units. Am J Respir Crit Care Med. 2013;188:800–6.

    Article  Google Scholar 

Download references

Acknowledgements

We wish to acknowledge the support of Leah Evans, BS, and Kathleen Merkley, DNP, APRN, ANP, of Health Catalyst, Inc. Salt Lake City, Utah for developing the Health Catalyst® Data Operating System (DOS™) utilised for this study. We would also like to acknowledge all the clinicians and support staff who cared for patients during this study.

Funding

This work received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Author information

Authors and Affiliations

  1. Department of Critical Care, Guys and St Thomas NHS Foundation Trust, Westminster Bridge Road, London, SE1 7EH, UK

    Stephen F. Whebell, Andrew T. Jones & Guy W. Glover

  2. Department of Critical Care, Kings College Hospital, Denmark Hill, London, SE5 9RS, UK

    Emma J. Prower & Joe Zhang

  3. Department of Quality and Assurance, Guy’s and St Thomas NHS Foundation Trust, Westminster Bridge Road, London, SE1 7EH, UK

    Megan Pontin

  4. Department of Clinical Informatics, Guys and St Thomas NHS Foundation Trust, Westminster Bridge Road, London, SE1 7EH, UK

    David Grant

Authors
  1. Stephen F. Whebell
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Emma J. Prower
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Joe Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Megan Pontin
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. David Grant
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Andrew T. Jones
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Guy W. Glover
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

SW contributed to formal analysis, data curation, data visualisation, and writing—original draft; EP contributed to conceptualisation, methodology, and writing—review and editing; JZ contributed to methodology and writing—review and editing; MP contributed to conceptualisation and project administration; DG contributed to conceptualisation, methodology, validation, data curation, and data visualisation; AJ contributed to conceptualisation, methodology, and writing—review and editing; GG contributed to conceptualisation, methodology, validation, supervision, and writing—original draft. DG, SW, and GG have verified the underlying data. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Guy W. Glover.

Ethics declarations

Ethics approval and consent to participate

The study was registered locally and was conducted as a service evaluation as defined by the U.K. NHS Health Research Authority (http://www.hra.nhs.uk) using anonymous, routinely collected data and therefore did not require review by the Research Ethics Committee.

Consent for publication

Not applicable.

Competing interests

All authors declare that they have no financial, ethical, or other conflicts of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

Additional figures and tables, referenced explicitly in text. Including flowchart of inclusion/exclusion and tabulated results from univariable regression models.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Whebell, S.F., Prower, E.J., Zhang, J. et al. Increased time from physiological derangement to critical care admission associates with mortality. Crit Care 25, 226 (2021). https://doi.org/10.1186/s13054-021-03650-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13054-021-03650-1

Keywords

Critical Care

ISSN: 1364-8535