Centile-based early warning scores derived from statistical distributions of vital signs

Institute of Biomedical Engineering, Old Road Campus Research Building (Off Roosevelt Drive), University of Oxford, Oxford OX3 7DQ, UK.
Resuscitation (Impact Factor: 4.17). 03/2011; 82(8):1013-8. DOI: 10.1016/j.resuscitation.2011.03.006
Source: PubMed


To develop an early warning score (EWS) system based on the statistical properties of the vital signs in at-risk hospitalised patients.
A large dataset comprising 64,622 h of vital-sign data, acquired from 863 acutely ill in-hospital patients using bedside monitors, was used to investigate the statistical properties of the four main vital signs. Normalised histograms and cumulative distribution functions were plotted for each of the four variables. A centile-based alerting system was modelled using the aggregated database.
The means and standard deviations of our population's vital signs are very similar to those published in previous studies. When compared with EWS systems based on a future outcome, the cut-off values in our system are most different for respiratory rate and systolic blood pressure. With four-hourly observations in a 12-h shift, about 1 in 8 at-risk patients would trigger our alerting system during the shift.
A centile-based EWS system will identify patients with abnormal vital signs regardless of their eventual outcome and might therefore be more likely to generate an alert when presented with patients with redeemable morbidity or avoidable mortality. We are about to start a stepped-wedge clinical trial gradually introducing an electronic version of our EWS system on the trauma wards in a teaching hospital.

Download full-text


Available from: L. Tarassenko, Dec 17, 2013
27 Reads
  • Source
    • "This means that the alarm accuracy or sensitivity and no alarm accuracy or sepcificity of the system should be as high as possible. The invention of early warning score (EWS) system marked the begining of MPMs in healthcare field [10]. Here the system assigns different scores, individually to the vital signs based on the variation from the clinically assumed normal range and the patient will be prompted for clinical review, if the score of any vital sign data or the sum of scores exceeds a predefined threshold [11]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Multi-parameter patient monitors (MPMs) are extensively used for enhancing the quality of healthcare in both intensive care units (ICU) and in-patient wards. MPMs make use of the vital signs, respiration rate, heart rate, blood pressure and oxygen saturation (SpO2) , for predicting the condition of patients. Support vector machine (SVM) is one of the most popularly used classification algorithms for developing MPMs. The kernel function, used in an SVM is a measure of similarity between any two examples, either belonging to same class or different classes. The selection of the kernel is an important aspect for the optimization of the system using SVM. If two patients have heart rates of 60 bpm and 80 bpm, intuition suggests that their heart rate similarity is 60 bpm. Extending this to n features, we may say that the total similarity is a summation of the individual similarities over n features, suggesting that intersection kernel is an ideal choice for MPM. In this paper, we explore the effectiveness of using intersection kernel SVM (IKSVM) for improving the performance of MPMs. We also compare the performance improvement of the MPM using IKSVM with the popularly used linear, polynomial and radial basis function (RBF) kernel MPMs. The results suggest that the use of intersection kernel can help enhance the performance of the MPMs significantly. Using IKSVM system, we obtained an improvement of 2.74 % absolute for overall accuracy, 1.86 % absolute for sensitivity and 3.00% absolute for specificity over the best baseline MPM using RBF kernel.
    International Journal of Applied Engineering Research 06/2015; 10(9):22703-22710.
  • Source
    • "For instance, Clifton et al. [25] compared Gaussian mixture model (GMM) and support vector machine (SVM) with HR, RR, SpO2, and SysBP as input. Tarassenko et al. [26] developed a centile-based early warning score system based on statistical properties of the vital signs (HR, RR, SpO2 and SysBP) to identify deteriorating patients. Scores were determined when the statistical value of vital sign fell into certain range of centile. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Patient monitors in modern hospitals have become ubiquitous but they generate an excessive number of false alarms causing alarm fatigue. Our previous work showed that combinations of frequently co-occurring monitor alarms, called SuperAlarm patterns, were capable of predicting in-hospital code blue events at a lower alarm frequency. In the present study, we extend the conceptual domain of a SuperAlarm to incorporate laboratory test results along with monitor alarms so as to build an integrated data set to mine SuperAlarm patterns. We propose two approaches to integrate monitor alarms with laboratory test results and use a maximal frequent itemsets mining algorithm to find SuperAlarm patterns. Under an acceptable false positive rate FPRmax, optimal parameters including the minimum support threshold and the length of time window for the algorithm to find the combinations of monitor alarms and laboratory test results are determined based on a 10-fold cross-validation set. SuperAlarm candidates are generated under these optimal parameters. The final SuperAlarm patterns are obtained by further removing the candidates with false positive rate >FPRmax. The performance of SuperAlarm patterns are assessed using an independent test data set. First, we calculate the sensitivity with respect to prediction window and the sensitivity with respect to lead time. Second, we calculate the false SuperAlarm ratio (ratio of the hourly number of SuperAlarm triggers for control patients to that of the monitor alarms, or that of regular monitor alarms plus laboratory test results if the SuperAlarm patterns contain laboratory test results) and the work-up to detection ratio, WDR (ratio of the number of patients triggering any SuperAlarm patterns to that of code blue patients triggering any SuperAlarm patterns). The experiment results demonstrate that when varying FPRmax between 0.02 - 0.15, the SuperAlarm patterns composed of monitor alarms along with the last two laboratory test results are triggered at least once for [56.7%-93.3%] of code blue patients within an 1-hour prediction window before code blue events and for [43.3% - 90.0%] of code blue patients at least 1-hour ahead of code blue events. However, the hourly number of these SuperAlarm patterns occurring in control patients is only [2.0% - 14.8%] of that of regular monitor alarms with WDR varying between 2.1 - 6.5 in a 12-hour window. For a given FPRmax threshold, the SuperAlarm set generated from the integrated data set has higher sensitivity and lower WDR than the SuperAlarm set generated from the regular monitor alarm data set. In addition, the McNemar's test also shows that the performance of the SuperAlarm set from the integrated data set is significantly different from that of the SuperAlarm set from the regular monitor alarm data set. We therefore conclude that the SuperAlarm patterns generated from the integrated data set are better at predicting code blue events.
    Journal of Biomedical Informatics 09/2014; 53. DOI:10.1016/j.jbi.2014.09.006 · 2.19 Impact Factor
  • Source
    • "The latter shows that most observations are taken at intervals of several hours, with a mean of 4.1 h between observations (but often rising to as long as eight h between observations). This current standard of care for " predictive monitoring, " involving manual observation, has a number of disadvantages. 1) The EWS assigned to each vital sign, and the thresholds against which the scores are compared, are typically heuristic [7]. 2) EWS systems are used with periodic observation of vital signs, which may be made as infrequently as once every 12 h in some wards. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The majority of patients in the hospital are ambulatory and would benefit significantly from predictive and personalized monitoring systems. Such patients are well suited to having their physiological condition monitored using low-power, minimally intrusive wearable sensors. Despite data-collection systems now being manufactured commercially, allowing physiological data to be acquired from mobile patients, little work has been undertaken on the use of the resultant data in a principled manner for robust patient care, including predictive monitoring. Most current devices generate so many false-positive alerts that devices cannot be used for routine clinical practice. This paper explores principled machine learning approaches to interpreting large quantities of continuously acquired, multivariate physiological data, using wearable patient monitors, where the goal is to provide early warning of serious physiological determination, such that a degree of predictive care may be provided. We adopt a one-class support vector machine formulation, proposing a formulation for determining the free parameters of the model using partial area under the ROC curve, a method arising from the unique requirements of performing online analysis with data from patient-worn sensors. There are few clinical evaluations of machine learning techniques in the literature, so we present results from a study at the Oxford University Hospitals NHS Trust devised to investigate the large-scale clinical use of patient-worn sensors for predictive monitoring in a ward with a high incidence of patient mortality. We show that our system can combine routine manual observations made by clinical staff with the continuous data acquired from wearable sensors. Practical considerations and recommendations based on our experiences of this clinical study are discussed, in the context of a framework for personalized monitoring.
    05/2014; 18(3):722-30. DOI:10.1109/JBHI.2013.2293059
Show more