Predicting cardiac arrest on the wards: a nested case-control study

Insights into postoperative respiration by using continuous wireless monitoring of respiratory rate on the postoperative ward: a cohort study

Change of respiratory rate (RespR) is the most powerful predictor of clinical deterioration. Brady- (RespR ≤ 8) and tachypnea (RespR ≥ 31) are associated with serious adverse events. Simultaneously, RespR is the least accurately measured vital parameter. We investigated the feasibility of continuously measuring RespR on the ward using wireless monitoring equipment, without impeding mobilization. Continuous monitoring of vital parameters using a wireless SensiumVitals® patch was installed and RespR was measured every 2 mins. We defined feasibility of adequate RespR monitoring if the system reports valid RespR measurements in at least 50% of time-points in more than 80% of patients during day- and night-time, respectively. Data from 119 patients were analysed. The patch detected in 171,151 of 227,587 measurements valid data for RespR (75.2%). During postoperative day and night four, the system still registered 68% and 78% valid measurements, respectively. 88% of the patients had more than 67% of valid RespR measurements. The RespR’s most frequently measured were 13–15; median RespR was 15 (mean 16, 25th- and 75th percentile 13 and 19). No serious complications or side effects were observed. We successfully measured electronically RespR on a surgical ward in postoperative patients continuously for up to 4 days post-operatively using a wireless monitoring system. While previous studies mentioned a digit preference of 18–22 for RespR, the most frequently measured RespR were 13–16. However, in the present study we did not validate the measurements against a reference method. Rather, we attempted to demonstrate the feasibility of achieving continuous wireless measurement in patients on surgical postoperative wards. As the technology used is based on impedance pneumography, obstructive apnoea might have been missed, namely in those patients receiving opioids post-operatively.

Minimal Impact of Implemented Early Warning Score and Best Practice Alert for Patient Deterioration

OBJECTIVES

Previous studies have looked at National Early Warning Score performance in predicting in-hospital deterioration and death, but data are lacking with respect to patient outcomes following implementation of National Early Warning Score. We sought to determine the effectiveness of National Early Warning Score implementation on predicting and preventing patient deterioration in a clinical setting.

DESIGN

Retrospective cohort study.

SETTING

Tertiary care academic facility and a community hospital.

PATIENTS

Patients 18 years old or older hospitalized from March 1, 2014, to February 28, 2015, during preimplementation of National Early Warning Score to August 1, 2015, to July 31, 2016, after National Early Warning Score was implemented.

INTERVENTIONS

Implementation of National Early Warning Score within the electronic health record and associated best practice alert.

MEASUREMENTS AND MAIN RESULTS

In this study of 85,322 patients (42,402 patients pre-National Early Warning Score and 42,920 patients post-National Early Warning Score implementation), the primary outcome of rate of ICU transfer or death did not change after National Early Warning Score implementation, with adjusted hazard ratio of 0.94 (0.84-1.05) and 0.90 (0.77-1.05) at our academic and community hospital, respectively. In total, 175,357 best practice advisories fired during the study period, with the best practice advisory performing better at the community hospital than the academic at predicting an event within 12 hours 7.4% versus 2.2% of the time, respectively. Retraining National Early Warning Score with newly generated hospital-specific coefficients improved model performance.

CONCLUSIONS

At both our academic and community hospital, National Early Warning Score had poor performance characteristics and was generally ignored by frontline nursing staff. As a result, National Early Warning Score implementation had no appreciable impact on defined clinical outcomes. Refitting of the model using site-specific data improved performance and supports validating predictive models on local data.

Comparing the predictive ability of a commercial artificial intelligence early warning system with physician judgement for clinical deterioration in hospitalised general internal medicine patients: a prospective observational study

OBJECTIVE

Our study compares physician judgement with an automated early warning system (EWS) for predicting clinical deterioration of hospitalised general internal medicine patients.

DESIGN

Prospective observational study of clinical predictions made at the end of the daytime work-shift for an academic general internal medicine floor team compared with the risk assessment from an automated EWS collected at the same time.

SETTING

Internal medicine teaching wards at a single tertiary care academic medical centre in the USA.

PARTICIPANTS

Intern physicians working on the internal medicine wards and an automated EWS (Rothman Index by PeraHealth).

OUTCOME

Clinical deterioration within 24 hours including cardiac or pulmonary arrest, rapid response team activation or unscheduled intensive care unit transfer.

RESULTS

We collected predictions for 1874 patient days and saw 35 clinical deteriorations (1.9%). The area under the receiver operating curve (AUROC) for the EWS was 0.73 vs 0.70 for physicians (p=0.571). A linear regression model combining physician and EWS predictions had an AUROC of 0.75, outperforming physicians (p=0.016) and the EWS (p=0.05).

CONCLUSIONS

There is no significant difference in the performance of the EWS and physicians in predicting clinical deterioration at 24 hours on an inpatient general medicine ward. A combined model outperformed either alone. The EWS and physicians identify partially overlapping sets of at-risk patients suggesting they rely on different cues or decision rules for their predictions.

TRIAL REGISTRATION NUMBER

NCT02648828.

Novel model for predicting inpatient mortality after emergency admission to hospital in Singapore: retrospective observational study

OBJECTIVES

To identify risk factors for inpatient mortality after patients' emergency admission and to create a novel model predicting inpatient mortality risk.

DESIGN

This was a retrospective observational study using data extracted from electronic health records (EHRs). The data were randomly split into a derivation set and a validation set. The stepwise model selection was employed. We compared our model with one of the current clinical scores, Cardiac Arrest Risk Triage (CART) score.

SETTING

A single tertiary hospital in Singapore.

PARTICIPANTS

All adult hospitalised patients, admitted via emergency department (ED) from 1 January 2008 to 31 October 2017 (n=433 187 by admission episodes).

MAIN OUTCOME MEASURE

The primary outcome of interest was inpatient mortality following this admission episode. The area under the curve (AUC) of the receiver operating characteristic curve of the predictive model with sensitivity and specificity for optimised cut-offs.

RESULTS

15 758 (3.64%) of the episodes were observed inpatient mortality. 19 variables were observed as significant predictors and were included in our final regression model. Our predictive model outperformed the CART score in terms of predictive power. The AUC of CART score and our final model was 0.705 (95% CI 0.697 to 0.714) and 0.817 (95% CI 0.810 to 0.824), respectively.

CONCLUSION

We developed and validated a model for inpatient mortality using EHR data collected in the ED. The performance of our model was more accurate than the CART score. Implementation of our model in the hospital can potentially predict imminent adverse events and institute appropriate clinical management.

Validating the Electronic Cardiac Arrest Risk Triage (eCART) Score for Risk Stratification of Surgical Inpatients in the Postoperative Setting: Retrospective Cohort Study

OBJECTIVE

Assess the accuracy of 3 early warning scores for predicting severe adverse events in postoperative inpatients.

SUMMARY OF BACKGROUND DATA

Postoperative clinical deterioration on inpatient hospital services is associated with increased morbidity, mortality, and cost. Early warning scores have been developed to detect inpatient clinical deterioration and trigger rapid response activation, but knowledge regarding the application of early warning scores to postoperative inpatients is limited.

METHODS

This was a retrospective cohort study of adult patients hospitalized on the wards after surgical procedures at an urban academic medical center from November, 2008 to January, 2016. The accuracies of the Modified Early Warning Score (MEWS), National Early Warning Score (NEWS), and the electronic cardiac arrest risk triage (eCART) score were compared in predicting severe adverse events (ICU transfer, ward cardiac arrest, or ward death) in the postoperative period using the area under the receiver operating characteristic curve (AUC).

RESULTS

Of the 32,537 patient admissions included in the study, 3.8% (n = 1243) experienced a severe adverse outcome after the procedure. The accuracy for predicting the composite outcome was highest for eCART [AUC 0.79 (95% CI: 0.78-0.81)], followed by NEWS [AUC 0.76 (95% CI: 0.75-0.78)], and MEWS [AUC 0.75 (95% CI: 0.73-0.76)]. Of the individual vital signs and labs, maximum respiratory rate was the most predictive (AUC 0.67) and maximum temperature was an inverse predictor (AUC 0.46).

CONCLUSION

Early warning scores are predictive of severe adverse events in postoperative patients. eCART is significantly more accurate in this patient population than both NEWS and MEWS.

Photoplethysmographic determination of the respiratory rate in acutely ill patients: validation of a new algorithm and implementation into a biomedical device

Background

Respiratory rate is among the first vital signs to change in deteriorating patients. The aims of this study were to evaluate the accuracy of respiratory rate measurements using a specifically dedicated reflection-mode photoplethysmographic signal analysis in a pathological condition (PPG-RR) and to validate its implementation within medical devices.

Methods

This study is derived from a data mining project, including all consecutive patients admitted to our ICU (ReaSTOC study, ClinicalTrials.gov identifier: NCT02893462). During the evaluation phase of the algorithm, PPG-RR calculations were retrospectively performed on PPG waveforms extracted from the data warehouse and compared with RR reference values. During the prospective phase, PPG-RR calculations were automatically and continuously performed using a dedicated device (FreeO₂, Oxynov, Québec, QC, Canada). In all phases, reference RR was measured continuously using electrical thoracic impedance and chronometric evaluation (Manual-RR) over a 30-s period.

Results

In total, 201 ICU patients’ recordings (SAPS II 51.7 ± 34.6) were analysed during the retrospective evaluation phase, most of them being admitted for a respiratory failure and requiring invasive mechanical ventilation. PPG-RR determination was available in 95.5% cases, similar to reference (22 ± 4 vs. 22 ± 5 c/min, respectively; p = 1), and well correlated with reference values (R = 0.952; p < 0.0001), with a low bias (0.1 b/min) and deviation (± 3.5 b/min). Prospective estimation of the PPG-RR on 30 ICU patients’ recordings was well correlated with the reference method (Manual-RR; r = 0.78; p < 0.001). Comparison of the methods depicted a low bias (0.5 b/min) and acceptable deviation (< ± 5.5 b/min).

Conclusion

According to our results, PPG-RR is an interesting approach for ventilation monitoring, as this technique would make simultaneous monitoring of respiratory rate and arterial oxygen saturation possible, thus minimizing the number of sensors attached to the patient.

Trial registry number ClinicalTrials.gov identifier NCT02893462

Artificial intelligence in healthcare

Artificial intelligence (AI) is gradually changing medical practice. With recent progress in digitized data acquisition, machine learning and computing infrastructure, AI applications are expanding into areas that were previously thought to be only the province of human experts. In this Review Article, we outline recent breakthroughs in AI technologies and their biomedical applications, identify the challenges for further progress in medical AI systems, and summarize the economic, legal and social implications of AI in healthcare.

Learning from different perspectives: Robust cardiac arrest prediction via temporal transfer learning

Predicting and preventing cardiac arrest is one of the biggest challenges of contemporary cardiology, as a patients survival depends on the effectiveness of the emergency response teams. While black-box models have shown to have better predictive accuracies for cardiac risk stratification, early warning scoring systems are more prominent in the hospital setting due to their ease of implementation and interpretability. We propose a temporal transfer learning approach to utilize information from adjacent time points to yield an early cardiac arrest prediction model that is robust in predictive accuracies as well as maintains the interpretability of the model coefficients. Our model estimates the logistic regression coefficients simultaneously for various time points to share knowledge from different observation windows. This framework can overcome small sample size issues, and result in robust estimation of the model coefficients. We find that our model consistently outperforms a logistic regression model fit only on vital sign data from a single time slice for 763 intensive care unit patients. Moreover, we find that the estimated coefficients from our model captures temporal trends in the data.

Reliability of wireless monitoring using a wearable patch sensor in high-risk surgical patients at a step-down unit in the Netherlands: a clinical validation study

BACKGROUND AND OBJECTIVES

Intermittent vital signs measurements are the current standard on hospital wards, typically recorded once every 8 hours. Early signs of deterioration may therefore be missed. Recent innovations have resulted in 'wearable' sensors, which may capture patient deterioration at an earlier stage. The objective of this study was to determine whether a wireless 'patch' sensor is able to reliably measure respiratory and heart rate continuously in high-risk surgical patients. The secondary objective was to explore the potential of the wireless sensor to serve as a safety monitor.

DESIGN

In an observational methods comparisons study, patients were measured with both the wireless sensor and bedside routine standard for at least 24 hours.

SETTING

University teaching hospital, single centre.

PARTICIPANTS

Twenty-five postoperative surgical patients admitted to a step-down unit.

OUTCOME MEASURES

Primary outcome measures were limits of agreement and bias of heart rate and respiratory rate. Secondary outcome measures were sensor reliability, defined as time until first occurrence of data loss.

RESULTS

1568 hours of vital signs data were analysed. Bias and 95% limits of agreement for heart rate were -1.1 (-8.8 to 6.5) beats per minute. For respiration rate, bias was -2.3 breaths per minute with wide limits of agreement (-15.8 to 11.2 breaths per minute). Median filtering over a 15 min period improved limits of agreement of both respiration and heart rate. 63% of the measurements were performed without data loss greater than 2 min. Overall data loss was limited (6% of time).

CONCLUSIONS

The wireless sensor is capable of accurately measuring heart rate, but accuracy for respiratory rate was outside acceptable limits. Remote monitoring has the potential to contribute to early recognition of physiological decline in high-risk patients. Future studies should focus on the ability to detect patient deterioration on low care environments and at home after discharge.

Review: Continuous Monitoring to Detect Failure to Rescue in Adult Postoperative Inpatients

Failure to rescue, or the unexpected death of a patient due to a preventable complication, is a nationally documented problem with numerous and multifaceted contributing factors. These factors include the frequency and method of collecting vital sign data, response to abnormal vital signs, and delays in the escalation of care for general ward patients who are showing signs of clinical deterioration. Patients' clinical deterioration can be complicated by concurrent secondary factors, including opioid abuse/dependence, being uninsured, or having sleep-disordered breathing. Using the Johns Hopkins Nursing Evidence-Based Practice Model, this integrative review synthesizes 43 research and nonresearch sources of evidence. Published between 2001 and 2017, these sources of evidence focus on failure to rescue, the multifaceted contributing factors to failure to rescue, and how continuous vital sign monitoring could ameliorate failure to rescue and its causes. Recommendations from the sources of evidence have been divided into system, structural, or technological categories.

Detection of physiological deterioration by the SNAP40 wearable device compared to standard monitoring devices in the emergency department: the SNAP40-ED study

BACKGROUND

In recent years, there has been increasing focus on the earlier detection of deterioration in the clinical condition of hospital patients with the aim of instigating earlier treatment to reverse this deterioration and prevent adverse outcomes. This is especially important in the ED, a dynamic environment with large volumes of undifferentiated patients, which carries inherent patient risk. SNAP40 is an innovative medical-grade device that can be worn on the upper arm that continuously monitors patients' vital signs including relative changes in systolic blood pressure, respiratory rate, heart rate, movement, blood oxygen saturation and temperature. It uses automated risk analysis to potentially allow clinical staff to easily and quickly identify high-risk patients. The aim of this study is to investigate whether the SNAP40 device is able to identify deterioration in the vital sign physiology of an ED patient earlier than current standard monitoring and observation charting techniques.

METHODS/DESIGN

Single centre, teaching hospital ED open label, prospective, observational cohort study recruiting 250 high acuity participants aged 16 years or over presenting to the ED. Participants will be approached and enrolled in the ED and after consent will have the SNAP40 wearable monitoring device attached which will be used alongside standard care monitoring. Participants will be observed throughout their time in the ED. Any SNAP40 device alarm, standard monitoring alarms or standard practice vital sign observations indicating a deterioration in a patient's vital sign physiology (defined as an increase in NEWS score) will be recorded. Primary outcome is time to detection of deterioration. Secondary outcomes include staff time spent performing observations and responding to standard monitoring alarms, clinical escalation of care when deterioration is detected and participants and staff rating of experience of both SNAP40 and current monitoring.

DISCUSSION

The SNAP40-ED study aims to recruit 250 patients. It will be the first study to compare the ability of a novel ambulatory monitoring device to detect deterioration compared to standard care in the ED. It may allow the earlier detection of deterioration in the clinical condition of ED patients and therefore earlier treatment to reverse this deterioration and prevent adverse outcomes.

TRIAL REGISTRATION

NCT03179267 ClinicalTrials.gov. Registered on June 17, 2017.

Effect of a Real-Time Electronic Dashboard on a Rapid Response System

A rapid response system (RRS) may have limited effectiveness when inpatient providers fail to recognize signs of early patient decompensation. We evaluated the impact of an electronic medical record (EMR)-based alerting dashboard on outcomes associated with RRS activation. We used a repeated treatment study in which the dashboard display was successively turned on and off each week for ten 2-week cycles over a 20-week period on the inpatient acute care wards of an academic medical center. The Rapid Response Team (RRT) dashboard displayed all hospital patients in a single view ranked by severity score, updated in real time. The dashboard could be seen within the EMR by any provider, including RRT members. The primary outcomes were the incidence rate ratio (IRR) of all RRT activations, unexpected ICU transfers, cardiopulmonary arrests and deaths on general medical-surgical wards (wards). We conducted an exploratory analysis of first RRT activations. There were 6736 eligible admissions during the 20-week study period. There was no change in overall RRT activations (IRR = 1.14, p = 0.07), but a significant increase in first RRT activations (IRR = 1.20, p = 0.04). There were no significant differences in unexpected ICU transfers (IRR = 1.15, p = 0.25), cardiopulmonary arrests on general wards (IRR = 1.46, p = 0.43), or deaths on general wards (IRR = 0.96, p = 0.89). The introduction of the RRT dashboard was associated with increased initial RRT activations but not overall activations, unexpected ICU transfers, cardiopulmonary arrests, or death. The RRT dashboard is a novel tool to help providers recognize patient decompensation and may improve initial RRT notification.

Technological aided assessment of the acutely ill patient - The case of postoperative complications

Surgical interventions come with complications and highly reported mortality after major surgery. The mortality may be a result of delayed detection of severe complications due to lower monitoring frequency in the general wards. Several studies have shown that continuous monitoring is superior to the manually intermittent recorded monitoring in terms of detecting abnormal physiological signs. Hopefully improved observations may result in earlier detection and clinical intervention. This narrative review will describe current monitoring possibilities for postoperative patients and how it may prevent complications. Several wireless systems are being developed for monitoring vital parameters, but many of these are not yet validated for critically ill patients. The ultimate goal with patient monitoring and detect of events is to prevent postoperative complications, death and costs in the health care system. A few studies indicate that monitoring systems detect deteriorating patients earlier than the nurses, and this was associated with less clinical instability. An important caveat of future devices is to assess their effect in relevant patient populations and not only in healthy test-subjects. Implementation of novel technologies is expensive although expected to be cost-effective if just few adverse events can be prevented. The future is here with promising devices and the possibility to give an unprecedented precise risk estimation of adverse post-surgical events. Next step is to integrate existing evidence based treatment algorithms to demonstrate the clinical efficacy of implementing the new technology.

Make vital signs great again - A call for action

Vital signs are the simplest, cheapest and probably the most important information gathered on patients in hospital. In this narrative review we present a large amount of evidence that vital signs are currently little valued, not regularly or accurately recorded, and frequently not acted on appropriately. It is probable that few hospitals would keep their accreditation with regulatory bodies if they collected and acted on their laboratory results in the same way that they collect and act on vital signs. Professional societies and regulatory bodies need to address this issue: if vital signs were more accurately and frequently measured, and acted on promptly and appropriately hospital care would be safer, better and cheaper.

Association Between Survival and Time of Day for Rapid Response Team Calls in a National Registry

OBJECTIVES

Decreased staffing at nighttime is associated with worse outcomes in hospitalized patients. Rapid response teams were developed to decrease preventable harm by providing additional critical care resources to patients with clinical deterioration. We sought to determine whether rapid response team call frequency suffers from decreased utilization at night and how this is associated with patient outcomes.

DESIGN

Retrospective analysis of a prospectively collected registry database.

SETTING

National registry database of inpatient rapid response team calls.

PATIENTS

Index rapid response team calls occurring on the general wards in the American Heart Association Get With The Guidelines-Medical Emergency Team database between 2005 and 2015 were analyzed.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The primary outcome was inhospital mortality. Patient and event characteristics between the hours with the highest and lowest mortality were compared, and multivariable models adjusting for patient characteristics were fit. A total of 282,710 rapid response team calls from 274 hospitals were included. The lowest frequency of calls occurred in the consecutive 1 AM to 6:59 AM period, with 266 of 274 (97%) hospitals having lower than expected call volumes during those hours. Mortality was highest during the 7 AM hour and lowest during the noon hour (18.8% vs 13.8%; adjusted odds ratio, 1.41 [1.31-1.52]; p < 0.001). Compared with calls at the noon hour, those during the 7 AM hour had more deranged vital signs, were more likely to have a respiratory trigger, and were more likely to have greater than two simultaneous triggers.

CONCLUSIONS

Rapid response team activation is less frequent during the early morning and is followed by a spike in mortality in the 7 AM hour. These findings suggest that failure to rescue deteriorating patients is more common overnight. Strategies aimed at improving rapid response team utilization during these vulnerable hours may improve patient outcomes.

Multicenter Study Validating Accuracy of a Continuous Respiratory Rate Measurement Derived From Pulse Oximetry: A Comparison With Capnography

Published ahead of print January 17, 2017.

BACKGROUND:

Intermittent measurement of respiratory rate via observation is routine in many patient care settings. This approach has several inherent limitations that diminish the clinical utility of these measurements because it is intermittent, susceptible to human error, and requires clinical resources. As an alternative, a software application that derives continuous respiratory rate measurement from a standard pulse oximeter has been developed. We sought to determine the performance characteristics of this new technology by comparison with clinician-reviewed capnography waveforms in both healthy subjects and hospitalized patients in a low-acuity care setting.

METHODS:

Two independent observational studies were conducted to validate the performance of the Medtronic Nellcor^TM Respiration Rate Software application. One study enrolled 26 healthy volunteer subjects in a clinical laboratory, and a second multicenter study enrolled 53 hospitalized patients. During a 30-minute study period taking place while participants were breathing spontaneously, pulse oximeter and nasal/oral capnography waveforms were collected. Pulse oximeter waveforms were processed to determine respiratory rate via the Medtronic Nellcor Respiration Rate Software. Capnography waveforms reviewed by a clinician were used to determine the reference respiratory rate.

RESULTS:

A total of 23,243 paired observations between the pulse oximeter-derived respiratory rate and the capnography reference method were collected and examined. The mean reference-based respiratory rate was 15.3 ± 4.3 breaths per minute with a range of 4 to 34 breaths per minute. The Pearson correlation coefficient between the Medtronic Nellcor Respiration Rate Software values and the capnography reference respiratory rate is reported as a linear correlation, R, as 0.92 ± 0.02 (P < .001), whereas Lin’s concordance correlation coefficient indicates an overall agreement of 0.85 ± 0.04 (95% confidence interval [CI] +0.76; +0.93) (healthy volunteers: 0.94 ± 0.02 [95% CI +0.91; +0.97]; hospitalized patients: 0.80 ± 0.06 [95% CI +0.68; +0.92]). The mean bias of the Medtronic Nellcor Respiration Rate Software was 0.18 breaths per minute with a precision (SD) of 1.65 breaths per minute (healthy volunteers: 0.37 ± 0.78 [95% limits of agreement: –1.16; +1.90] breaths per minute; hospitalized patients: 0.07 ± 1.99 [95% limits of agreement: –3.84; +3.97] breaths per minute). The root mean square deviation was 1.35 breaths per minute (healthy volunteers: 0.81; hospitalized patients: 1.60).

CONCLUSIONS:

These data demonstrate the performance of the Medtronic Nellcor Respiration Rate Software in healthy subjects and patients hospitalized in a low-acuity care setting when compared with clinician-reviewed capnography. The observed performance of this technology suggests that it may be a useful adjunct to continuous pulse oximetry monitoring by providing continuous respiratory rate measurements. The potential patient safety benefit of using combined continuous pulse oximetry and respiratory rate monitoring warrants assessment.

Is everyone really breathing 20 times a minute? Assessing epidemiology and variation in recorded respiratory rate in hospitalised adults

BACKGROUND

Respiratory rate (RR) is an independent predictor of adverse outcomes and an integral component of many risk prediction scores for hospitalised adults. Yet, it is unclear if RR is recorded accurately. We sought to assess the potential accuracy of RR by analysing the distribution and variation as a proxy, since RR should be normally distributed if recorded accurately.

METHODS

We conducted a descriptive observational study of electronic health record data from consecutive hospitalisations from 2009 to 2010 from six diverse hospitals. We assessed the distribution of the maximum RR on admission, using heart rate (HR) as a comparison since this is objectively measured. We assessed RR patterns among selected subgroups expected to have greater physiological variation using the coefficient of variation (CV=SD/mean).

RESULTS

Among 36 966 hospitalisations, recorded RR was not normally distributed (p<0.001), but right skewed (skewness=3.99) with values clustered at 18 and 20 (kurtosis=23.9). In contrast, HR was relatively normally distributed. Patients with a cardiopulmonary diagnosis or hypoxia only had modestly greater variation (CV increase of 2%-6%). Among 1318 patients transferred from the ward to the intensive care unit (n=1318), RR variation the day preceding transfer was similar to that observed on admission (CV 0.24 vs 0.26), even for those transferred with respiratory failure (CV 0.25).

CONCLUSIONS

The observed patterns suggest that RR is inaccurately recorded, even among those with cardiopulmonary compromise, and represents a 'spot' estimate with values of 18 and 20 breaths per minute representing 'normal.' While spot estimates may potentially be adequate to indicate clinical stability, inaccurate RR may alternatively lead to misclassification of disease severity, potentially jeopardising patient safety. Thus, we recommend greater training for hospital personnel to accurately record RR.

Association Between Opioid and Benzodiazepine Use and Clinical Deterioration in Ward Patients

BACKGROUND

Opioids and benzodiazepines are frequently used in hospitals, but little is known about outcomes among ward patients receiving these medications.

OBJECTIVE

To determine the association between opioid and benzodiazepine administration and clinical deterioration.

DESIGN

Observational cohort study.

SETTING

500-bed academic urban tertiary-care hospital.

PATIENTS

All adults hospitalized on the wards from November 2008 to January 2016 were included. Patients who were "comfort care" status, had tracheostomies, sickle-cell disease, and patients at risk for alcohol withdrawal or seizures were excluded.

MEASUREMENTS

The primary outcome was the composite of intensive care unit transfer or ward cardiac arrest. Discrete-time survival analysis was used to calculate the odds of this outcome during exposed time periods compared to unexposed time periods with respect to the medications of interest, with adjustment for patient demographics, comorbidities, severity of illness, and pain score.

RESULTS

In total, 120,518 admissions from 67,097 patients were included, with 67% of admissions involving opioids, and 21% involving benzodiazepines. After adjustment, each equivalent of 15 mg oral morphine was associated with a 1.9% increase in the odds of the primary outcome within 6 hours (odds ratio [OR], 1.019; 95% confidence interval [CI], 1.013-1.026; P < 0.001), and each 1 mg oral lorazepam equivalent was associated with a 29% increase in the odds of the composite outcome within 6 hours (OR, 1.29; CI, 1.16- 1.45; P < 0.001).

CONCLUSION

Among ward patients, opioids were associated with increased risk for clinical deterioration in the 6 hours after administration. Benzodiazepines were associated with even higher risk. These results have implications for ward-monitoring strategies. Journal of Hospital Medicine 2017;12:428-434.

Real-Time Risk Prediction on the Wards: A Feasibility Study

OBJECTIVE

Failure to detect clinical deterioration in the hospital is common and associated with poor patient outcomes and increased healthcare costs. Our objective was to evaluate the feasibility and accuracy of real-time risk stratification using the electronic Cardiac Arrest Risk Triage score, an electronic health record-based early warning score.

DESIGN

We conducted a prospective black-box validation study. Data were transmitted via HL7 feed in real time to an integration engine and database server wherein the scores were calculated and stored without visualization for clinical providers. The high-risk threshold was set a priori. Timing and sensitivity of electronic Cardiac Arrest Risk Triage score activation were compared with standard-of-care Rapid Response Team activation for patients who experienced a ward cardiac arrest or ICU transfer.

SETTING

Three general care wards at an academic medical center.

PATIENTS

A total of 3,889 adult inpatients.

MEASUREMENTS AND MAIN RESULTS

The system generated 5,925 segments during 5,751 admissions. The area under the receiver operating characteristic curve for electronic Cardiac Arrest Risk Triage score was 0.88 for cardiac arrest and 0.80 for ICU transfer, consistent with previously published derivation results. During the study period, eight of 10 patients with a cardiac arrest had high-risk electronic Cardiac Arrest Risk Triage scores, whereas the Rapid Response Team was activated on two of these patients (p < 0.05). Furthermore, electronic Cardiac Arrest Risk Triage score identified 52% (n = 201) of the ICU transfers compared with 34% (n = 129) by the current system (p < 0.001). Patients met the high-risk electronic Cardiac Arrest Risk Triage score threshold a median of 30 hours prior to cardiac arrest or ICU transfer versus 1.7 hours for standard Rapid Response Team activation.

CONCLUSIONS

Electronic Cardiac Arrest Risk Triage score identified significantly more cardiac arrests and ICU transfers than standard Rapid Response Team activation and did so many hours in advance.

Application of the National Early Warning Score (NEWS) as a stratification tool on admission in an Italian acute medical ward: A perspective study

AIM

We aimed to assess the performance of the National Early Warning Score (NEWS) as tool for patient risk stratification at admission in an acute Internal Medicine ward and to ensure patient placement in ward areas with the required and most appropriate intensity of care. As secondary objective, we considered NEWS performance in two subgroups of patients: sudden cardiac events (acute coronary syndromes and arrhythmic events), and chronic respiratory insufficiency.

METHODS

We conducted a perspective cohort single centre study on 2,677 unselected patients consecutively admitted from July 2013 to March 2015 in the Internal Medicine ward of the hospital of Trento, Italy. The NEWS was mandatory collected on ward admission. We defined three risk categories for clinical deterioration: low score (NEWS 0-4), medium score (NEWS 5-6), and high score (NEWS≥7). Following adverse outcomes were considered: total and early (<72 hours) in-hospital mortality, urgent transfers to a higher intensity of care. A logistic regression model quantified the association between outcomes and NEWS.

RESULTS

For patients with NEWS >4 vs patients with NEWS <4, the risk of early death increased from 12 to 36 times, total mortality from 3.5 to 9, and urgent transfers from 3.5 to 7. In patients with sudden cardiac events, lower scores were significantly associated with higher risk of transfer to a higher intensity of care. In patients affected by chronic hypoxaemia, adverse outcomes occurred less in medium and high score categories of NEWS.

CONCLUSIONS

National Early Warning Score assessed on ward admission may enable risk stratification of clinical deterioration and can be a good predictor of in-hospital serious adverse outcomes, although sudden cardiac events and chronic hypoxaemia could constitute some limits.

Correlation of the predictive ability of early warning metrics and mortality for cardiac arrest patients receiving in-hospital Advanced Cardiovascular Life Support

BACKGROUND

The Modified Early Warning Score (MEWS) helps identify patients experiencing a decline in physiological parameters that indicate risk for cardiac arrest (CA).

OBJECTIVES

To assess the association between MEWS values and patient survival following in-hospital CA.

METHODS

Retrospective cohort study of patients who experienced in-hospital CA. The relationship between CA survival and MEWS values as well as other risk factors such as age, gender and type of electrographic cardiac rhythms was analyzed using logistic regression.

RESULTS

Survival rate to hospital discharge was 21%. Strong predictors for survival were MEWS values at hospital admission (p < .002), younger age (p < .005), ventricular fibrillation (p < .0001), and ventricular tachycardia (p < .0001). Gender and MEWS 4 hours prior to CA were not significantly associated with survival.

CONCLUSIONS

Survival following CA was significantly associated with MEWS at hospital admission but not 4 hours prior to CA. The type of cardiac rhythm and age were also predictive of survival.

A Role for the Early Warning Score in Early Identification of Critical Postoperative Complications

OBJECTIVE

We examined whether an early warning score (EWS) could predict inpatient complications in surgical patients.

BACKGROUND

Abnormal vitals often precede in-hospital mortality. The EWS calculated using vital signs has been developed to identify patients at risk for mortality.

METHODS

Inpatient general surgery procedures with National Surgical Quality Improvement Project data from 2013 to 2014 were matched with enterprise data on vital signs and neurologic status to calculate the EWS for each postoperative vital set measured on the ward. Outcomes of major complications, unplanned intensive care unit transfer, and medical emergency team activation were classified using the Clavien-Dindo system as grade I to V. Relationship with EWS and timing of complication was assessed using Kruskal-Wallis test and linear regression accounting for clustering with generalized estimating equation.

RESULTS

Among 552 patients admitted to the ward postsurgery, 68 (12.3%) developed at least one grade I to III complication and 37 (6.7%) developed a grade IV/V complication. The mean maximum EWS was significantly higher preceding grade IV/V complications (10.1) compared with grade I to III complications (6.4) or across the hospital stay in patients without complications (5.4; P < 0.01). EWS significantly increased in the 3 days preceding grade IV/V complications (P < 0.001) and declined in patients without complications in the 3 days before discharge (P < 0.001). A threshold EWS of 8 predicted occurrence of grade IV/V complications with 81% sensitivity and 84% specificity.

CONCLUSIONS

Critical postoperative complications can be preceded by rising EWS. Interventional studies are needed to evaluate whether EWS can reduce the severity of postoperative complications and mortality for surgical patients through early identification and intervention.

Multicenter Comparison of Machine Learning Methods and Conventional Regression for Predicting Clinical Deterioration on the Wards

OBJECTIVE

Machine learning methods are flexible prediction algorithms that may be more accurate than conventional regression. We compared the accuracy of different techniques for detecting clinical deterioration on the wards in a large, multicenter database.

DESIGN

Observational cohort study.

SETTING

Five hospitals, from November 2008 until January 2013.

PATIENTS

Hospitalized ward patients

INTERVENTIONS

None

MEASUREMENTS AND MAIN RESULTS

Demographic variables, laboratory values, and vital signs were utilized in a discrete-time survival analysis framework to predict the combined outcome of cardiac arrest, intensive care unit transfer, or death. Two logistic regression models (one using linear predictor terms and a second utilizing restricted cubic splines) were compared to several different machine learning methods. The models were derived in the first 60% of the data by date and then validated in the next 40%. For model derivation, each event time window was matched to a non-event window. All models were compared to each other and to the Modified Early Warning score, a commonly cited early warning score, using the area under the receiver operating characteristic curve (AUC). A total of 269,999 patients were admitted, and 424 cardiac arrests, 13,188 intensive care unit transfers, and 2,840 deaths occurred in the study. In the validation dataset, the random forest model was the most accurate model (AUC, 0.80 [95% CI, 0.80-0.80]). The logistic regression model with spline predictors was more accurate than the model utilizing linear predictors (AUC, 0.77 vs 0.74; p < 0.01), and all models were more accurate than the MEWS (AUC, 0.70 [95% CI, 0.70-0.70]).

CONCLUSIONS

In this multicenter study, we found that several machine learning methods more accurately predicted clinical deterioration than logistic regression. Use of detection algorithms derived from these techniques may result in improved identification of critically ill patients on the wards.

Emergency care within hospitals: can it be done more efficiently?

INTRODUCTION

Cardiac Arrest Teams (CATs) are frequently activated by nurses when patients experience 'false arrests' (FAs). In those cases activation of the Rapid Response Team (RRT) might be more efficient. The authors determined the level of urgency of FAs to find a scope for improvement in efficiency within emergency care.

METHODS

CAT-activations for FAs in a university hospital from September 2009 to 2012 were retrospectively analysed and classified as urgent or less-urgent.

RESULTS

In 26% (107/405) the CAT was activated for FAs. Calls were classified as urgent in 43% (46/107). Less urgent calls comprised 57% (61/107) of the FAs, difference 14% (95%CI: 1% to 26%).

CONCLUSIONS

A significant part of the CAT-activations for FAs were less urgent and an RRT-activation might be more efficient. To minimise the CAT-activations for FAs, nurses need to recognise early patients who clinically deteriorate. Therefore, nurses should use the Modified Early Warning Score correctly.

Attitudes of nursing staff towards a Modified Early Warning System

This study aimed to understand the attitudes of qualified nursing staff on an acute medical unit concerning the Modified Early Warning System (MEWS) score chart used to monitor patients. A combination of questionnaires and a focus group was used. All respondents believed that the MEWS is beneficial in their work but the point was also raised that MEWS scores can be miscalculated and there is sometimes difficulty in getting medical staff to review the patient, even if the MEWS score is significantly high. At times a qualified nurse's seniority or the colour of his or her uniform can affect the attitude of the medical staff and can mean the difference between the patient being reviewed or not. Certain medics have a culture of dismissing a high MEWS score because they were expecting these vital physiological signs to be abnormal, owing to that particular patient's past medical history or presenting complaint. Most hospitals in the NHS now use some sort of early warning system but, at times, staff seem to be unsure of the importance of the MEWS score or what action needs to be taken. The authors agree with the view that introduction of a standard NHS-wide chart would be of benefit to staff and patients.

Survival after in-hospital cardiac arrest is highly associated with the Age-combined Charlson Co-morbidity Index in a cohort study from a two-site Swedish University hospital

BACKGROUND

In-hospital cardiac arrest (IHCA) has a poor prognosis and clinicians often write "Do-Not-Attempt-Resuscitation" (DNAR) orders based on co-morbidities.

AIM

To assess the impact of the Age-combined Charlson Co-morbidity Index (ACCI) on 30-days survival after IHCA.

MATERIAL AND METHODS

All patients suffering IHCA at Karolinska University Hospital between 1st January and 31st December 2014 were included. Data regarding patient characteristics, co-morbidities and survival were drawn from the electronic patient records. Co-morbidities were assessed prior to the IHCA as ICD-10 codes according to the ACCI. Differences in survival were assessed with adjusted logistic regression models and presented as Odds Ratios with 95% Confidence Intervals (OR, 95% CI) between patients with an ACCI of 0-4 points versus those with 5-7 points, as well as those with ≥8 points. Adjustments included hospital site, heart rhythm, ECG surveillance, witnessed status and place of IHCA.

RESULTS

In all, 174 patients suffered IHCA, of whom 41 (24%) survived at least 30 days. Patients with an ACCI of 5-7 points had a minor chance and those with an ACCI of ≥8 points had a minimal chance of surviving IHCA compared to those with an ACCI of 0-4 points (adjusted OR 0.10, 95% CI 0.04-0.26 and OR 0.04, 95% CI 0.03-0.42, respectively).

CONCLUSION

Patients with a moderate or severe burden of ACCI have a minor chance of surviving an IHCA. This information could be used as part of the decision tools during ongoing CPR, and could be an aid for clinicians in planning care and discussing DNAR orders.

Analysis of the Proportional Hazards Model with Sparse Longitudinal Covariates

Regression analysis of censored failure observations via the proportional hazards model permits time-varying covariates which are observed at death times. In practice, such longitudinal covariates are typically sparse and only measured at infrequent and irregularly spaced follow-up times. Full likelihood analyses of joint models for longitudinal and survival data impose stringent modelling assumptions which are difficult to verify in practice and which are complicated both inferentially and computationally. In this article, a simple kernel weighted score function is proposed with minimal assumptions. Two scenarios are considered: half kernel estimation in which observation ceases at the time of the event and full kernel estimation for data where observation may continue after the event, as with recurrent events data. It is established that these estimators are consistent and asymptotically normal. However, they converge at rates which are slower than the parametric rates which may be achieved with fully observed covariates, with the full kernel method achieving an optimal convergence rate which is superior to that of the half kernel method. Simulation results demonstrate that the large sample approximations are adequate for practical use and may yield improved performance relative to last value carried forward approach and joint modelling method. The analysis of the data from a cardiac arrest study demonstrates the utility of the proposed methods.

Obstructive sleep apnea and adverse outcomes in surgical and nonsurgical patients on the wards

BACKGROUND

Obstructive sleep apnea (OSA) has been associated with clinical deterioration in postoperative patients and patients hospitalized with pneumonia. Paradoxically, OSA has also been associated with decreased risk of inpatient mortality in these same populations.

OBJECTIVES

To investigate the association between OSA and in-hospital mortality in a large cohort of surgical and nonsurgical ward patients.

DESIGN

Observational cohort study.

SETTING

A 500-bed academic tertiary care hospital in the United States.

PATIENTS

A total of 93,676 ward admissions from 53,150 unique adult patients between November 1, 2008 and October 1, 2013.

INTERVENTION

None.

MEASUREMENTS

OSA diagnoses and comorbidities were identified by International Classification of Diseases, Ninth Revision, Clinical Modification codes. Logistic regression was used to control for patient characteristics, location prior to ward admission, and admission severity of illness. The primary outcome was in-hospital death. Secondary outcomes included rapid response team (RRT) activation, intensive care unit (ICU) transfer, intubation, and cardiac arrest on the wards.

MAIN RESULTS

OSA was identified in 5,625 (10.6%) patients. Patients with OSA were more likely to be older, male, and obese, and had higher rates of comorbidities. OSA patients had more frequent RRT activations (1.5% vs 1.1%) and ICU transfers (8% vs 7%) than controls (P < 0.001 for both comparisons), but a lower inpatient mortality rate (1.1% vs 1.4%, P < 0.05). OSA was associated with decreased adjusted odds for ICU transfer (odds ratio [OR]: 0.91 [0.84-0.99]), cardiac arrest (OR: 0.72 [0.55-0.95]), and in-hospital mortality (OR: 0.70 [0.58-0.85]).

CONCLUSIONS

After adjustment for important confounders, OSA was not associated with clinical deterioration on the wards and was associated with significantly decreased in-hospital mortality.

ECG changes on continuous telemetry preceding in-hospital cardiac arrests

BACKGROUND

About 200,000 patients suffer from in-hospital cardiac arrest (IHCA) annually. Identification of at-risk patients is key to improving outcomes. The use of continuous ECG monitoring in identifying patients at risk for developing IHCA has not been studied.

OBJECTIVE

To describe the profile and timing of ECG changes prior to IHCA.

DESIGN

Retrospective, observational.

SETTING

Single 520-bed tertiary care hospital.

PATIENTS

IHCA in adults between April 2010 and March 2012 with at least 3 hours of continuous telemetry data immediately prior to IHCA.

MEASUREMENTS

We evaluated up to 24 hours of telemetry data preceding IHCA for changes in PR, QRS, ST segment, arrhythmias, and QTc in ventricular tachycardia cases. We determined mechanism and likely clinical cause of the arrest by chart and telemetry review.

RESULTS

We studied 81 IHCA patients, in whom the mechanism was ventricular tachycardia/fibrillation in 14 (18%), bradyasystolic in 21 (26%), and pulseless electrical activity (PEA) in 46 (56%). Preceding ECG changes were ST segment changes (31% of cases), atrial tachyarrhythmias (21%), bradyarrhythmias (28%), P wave axis change (21%),QRS prolongation (19%), PR prolongation (17%), isorhythmic dissociation (14%), nonsustained ventricular tachycardia (6%), and PR shortening (5%). At least one of these was present in 77% of all cases, and in 89% of IHCA caused by respiratory or multiorgan failure. Bradyarrhythmias were primarily seen with IHCA in the setting of respiratory or multiorgan failure, and PR and QRS prolongation with IHCA and concomitant multiorgan failure.

LIMITATIONS

This is a retrospective study with a limited number of cases; each patient serves as their own control, and a separate control population has not yet been studied.

CONCLUSIONS

ECG changes are commonly seen preceding IHCA, and have a pathophysiologic basis. Automated detection methods for ECG changes could potentially be used to better identify patients at risk for IHCA.

Modified Early Warning Score Changes Prior to Cardiac Arrest in General Wards

Purpose

The frequency, extent, time frame, and implications of changes to the modified early warning score (MEWS) in the 24 hours prior to cardiac arrest are not known. Our aim was to determine the prevalence and trends of the MEWS prior to in-hospital cardiac arrest (IHCA) on a ward, and to evaluate the association between changes in the MEWS and in-hospital mortality.

Methods

A total of 501 consecutive adult IHCA patients who were monitored and resuscitated by a medical emergency team on the ward were enrolled in the study between March 2009 and February 2013. The MEWS was calculated at 24 hours (MEWS₂₄), 16 hours (MEWS₁₆), and 8 hours (MEWS₈) prior to cardiac arrest.

Results

Out of 380 patients, 268 (70.5%) had a return of spontaneous circulation. The survival rate to hospital discharge was 25.8%. When the MEWS was divided into three risk groups (low: ≤2, intermediate: 3–4, high: ≥5), the distribution of the low-risk MEWS group decreased at each time point before cardiac arrest. However, even 8 hours prior to cardiac arrest, 45.3% of patients were still in the low MEWS group. The MEWS was associated with in-hospital mortality at each time point. However, increasing MEWS value from MEWS₂₄ to MEWS₈ was not associated with in-hospital mortality [OR 1.24 (0.77–1.97), p = 0.38].

Conclusions

About half of patients were still in low MEWS group 8 hours prior to cardiac arrest and an increasing MEWS only occurred in 46.8% of patients, suggesting that monitoring the MEWS alone is not enough to predict cardiac arrest.

Differences in vital signs between elderly and nonelderly patients prior to ward cardiac arrest

OBJECTIVES

Vital signs and composite scores, such as the Modified Early Warning Score, are used to identify high-risk ward patients and trigger rapid response teams. Although age-related vital sign changes are known to occur, little is known about the differences in vital signs between elderly and nonelderly patients prior to ward cardiac arrest. We aimed to compare the accuracy of vital signs for detecting cardiac arrest between elderly and nonelderly patients.

DESIGN

Observational cohort study.

SETTING

Five hospitals in the United States.

PATIENTS

A total of 269,956 patient admissions to the wards with documented age, including 422 index ward cardiac arrests.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Patient characteristics and vital signs prior to cardiac arrest were compared between elderly (age, 65 yr or older) and nonelderly (age, <65 yr) patients. The area under the receiver operating characteristic curve for vital signs and the Modified Early Warning Score were also compared. Elderly patients had a higher cardiac arrest rate (2.2 vs 1.0 per 1,000 ward admissions; p<0.001) and in-hospital mortality (2.9% vs 0.7%; p<0.001) than nonelderly patients. Within 4 hours of cardiac arrest, elderly patients had significantly lower mean heart rate (88 vs 99 beats/min; p<0.001), diastolic blood pressure (60 vs 66 mm Hg; p=0.007), shock index (0.82 vs 0.93; p<0.001), and Modified Early Warning Score (2.6 vs 3.3; p<0.001) and higher pulse pressure index (0.45 vs 0.41; p<0.001) and temperature (36.4°C vs 36.3°C; p=0.047). The area under the receiver operating characteristic curves for all vital signs and the Modified Early Warning Score were higher for nonelderly patients than elderly patients (Modified Early Warning Score area under the receiver operating characteristic curve 0.85 [95% CI, 0.82-0.88] vs 0.71 [95% CI, 0.68-0.75]; p<0.001).

CONCLUSIONS

Vital signs more accurately detect cardiac arrest in nonelderly patients compared with elderly patients, which has important implications for how they are used for identifying critically ill patients. More accurate methods for risk stratification of elderly patients are necessary to decrease the occurrence of this devastating event.

The prognostic significance of respiratory rate in patients with pneumonia: a retrospective analysis of data from 705,928 hospitalized patients in Germany from 2010-2012

BACKGROUND

Measurement of the respiratory rate is an important instrument for assessing the severity of acute disease. The respiratory rate is often not measured in routine practice because its clinical utility is inadequately appreciated. In Germany, documentation of the respiratory rate is obligatory when a patient with pneumonia is hospitalized. This fact has enabled us to study the prognostic significance of the respiratory rate in reference to a large medical database.

METHOD

We retrospectively analyzed data from the external quality-assurance program for community-acquired pneumonia for the years 2010-2012. All patients aged 18 years or older who were not mechanically ventilated on admission were included in the analysis. Logistic regression was used to determine the significance of the respiratory rate as a risk factor for in-hospital mortality.

RESULTS

705,928 patients were admitted to the hospital with community-acquired pneumonia (incidence: 3.5 cases per 1000 adults per year). The in-hospital mortality of these patients was 13.1% (92 227 persons). The plot of mortality as a function of respiratory rate on admission was U-shaped and slanted to the right, with the lowest mortality at a respiratory rate of 20/min on admission. If patients with a respiratory rate of 12-20/min are used as a baseline for comparison, patients with a respiratory rate of 27-33/min had an odds ratio (OR) of 1.72 for in-hospital death, and those with a respiratory rate above 33/min had an OR of 2.55. Further independent risk factors for in-hospital death were age, admission from a nursing home, hospital, or rehabilitation facility, chronic bedridden state, disorientation, systolic blood pressure, and pulse pressure.

CONCLUSION

Respiratory rate is an independent risk marker for in-hospital mortality in community-acquired pneumonia. It should be measured when patients are admitted to the hospital with pneumonia and other acute conditions.

Modified Early Warning System improves patient safety and clinical outcomes in an academic community hospital

Background and objective

Severe adverse events such as cardiac arrest and death are often heralded by abnormal vital signs hours before the event. This necessitates an organized track and trigger approach of early recognition and response to subtle changes in a patient’s condition. The Modified Early Warning System (MEWS) is one of such systems that use temperature, blood pressure, pulse, respiratory rate, and level of consciousness with each progressive higher score triggering an action. Root cause analysis for mortalities in our institute has led to the implementation of MEWS in an effort to improve patient outcomes. Here we discuss our experience and the impact of MEWS implementation on patient care at our community academic hospital.

Methods

MEWS was implemented in a protocolized manner in June 2013. The following data were collected from non-ICU wards on a monthly basis from January 2010 to June 2014: 1) number of rapid response teams (RRTs) per 100 patient-days (100PD); 2) number of cardiopulmonary arrests ‘Code Blue’ per 100PD; and 3) result of each RRT and Code Blue (RRT progressed to Code Blue, higher level of care, ICU transfer, etc.). Overall inpatient mortality data were also analyzed.

Results

Since the implementation of MEWS, the number of RRT has increased from 0.24 per 100PD in 2011 to 0.38 per 100PD in 2013, and 0.48 per 100PD in 2014. The percentage of RRTs that progressed to Code Blue, an indicator of poor outcome of RRT, has been decreasing. In contrast, the numbers of Code Blue in non-ICU floors has been progressively decreasing from 0.05 per 100PD in 2011 to 0.02 per 100PD in 2013 and 2014. These improved clinical outcomes are associated with a decline of overall inpatient mortality rate from 2.3% in 2011 to 1.5% in 2013 and 1.2% in 2014.

Conclusions

Implementation of MEWS in our institute has led to higher rapid response system utilization but lower cardiopulmonary arrest events; this is associated with a lower mortality rate, and improved patient safety and clinical outcomes. We recommend the widespread use of MEWS to improve patient outcomes.

Reduction of unexpected serious adverse events after introducing medical emergency team

Aim

To assess the clinical benefits of introducing a medical emergency team system for early medical intervention in hospital care.

Methods

This prospective analysis included all cases of medical emergency team activation during the first year after the introduction of the medical emergency team system at Chiba University Hospital (Chiba, Japan) in February 2011. The rates of in-hospital mortality and unexpected events before and after introduction of the medical emergency team system were compared.

Results

The total number of medical emergency team activation calls was 83 (4.9 per 1,000 admissions). The activation of the medical emergency team system was requested most frequently from the general ward (56.6%) and by a physician (57.8%), with the most important reasons for activation being cardiac arrest (37.3%), breathing abnormality (33.7%), and impaired consciousness (32.5%). The most frequent medical interventions by the medical emergency team were intubation (43.3%) and oxygen inhalation (41.0%). Approximately one-half of the patients requiring activation of the medical emergency team system were critically ill and needed subsequent intensive care unit admission. Although no significant difference was observed between the pre- and post- medical emergency team in-hospital mortalities (2.1% versus 2.0%, respectively), the incidence rate of serious events significantly decreased (12.4% versus 6.8%, respectively; P = 0.015).

Conclusion

Most patients requiring activation of the medical emergency team system were critically ill and needed emergency treatment at the location of the medical emergency team activation, with subsequent critical care. Although the introduction of the medical emergency team system did not affect the in-hospital mortality rate, it reduced the incidence of unexpected serious adverse events, suggesting that it may be clinically useful.

Multicenter development and validation of a risk stratification tool for ward patients

RATIONALE

Most ward risk scores were created using subjective opinion in individual hospitals and only use vital signs.

OBJECTIVES

To develop and validate a risk score using commonly collected electronic health record data.

METHODS

All patients hospitalized on the wards in five hospitals were included in this observational cohort study. Discrete-time survival analysis was used to predict the combined outcome of cardiac arrest (CA), intensive care unit (ICU) transfer, or death on the wards. Laboratory results, vital signs, and demographics were used as predictor variables. The model was developed in the first 60% of the data at each hospital and then validated in the remaining 40%. The final model was compared with the Modified Early Warning Score (MEWS) using the area under the receiver operating characteristic curve and the net reclassification index (NRI).

MEASUREMENTS AND MAIN RESULTS

A total of 269,999 patient admissions were included, with 424 CAs, 13,188 ICU transfers, and 2,840 deaths occurring during the study period. The derived model was more accurate than the MEWS in the validation dataset for all outcomes (area under the receiver operating characteristic curve, 0.83 vs. 0.71 for CA; 0.75 vs. 0.68 for ICU transfer; 0.93 vs. 0.88 for death; and 0.77 vs. 0.70 for the combined outcome; P value < 0.01 for all comparisons). This accuracy improvement was seen across all hospitals. The NRI for the electronic Cardiac Arrest Risk Triage compared with the MEWS was 0.28 (0.18-0.38), with a positive NRI of 0.19 (0.09-0.29) and a negative NRI of 0.09 (0.09-0.09).

CONCLUSIONS

We developed an accurate ward risk stratification tool using commonly collected electronic health record variables in a large multicenter dataset. Further study is needed to determine whether implementation in real-time would improve patient outcomes.

Relationship between ICU bed availability, ICU readmission, and cardiac arrest in the general wards

OBJECTIVE

The decision to admit a patient to the ICU is complex, reflecting patient factors and available resources. Previous work has shown that ICU census does not impact mortality of patients admitted to the ICU. However, the effect of ICU bed availability on patients outside the ICU is unknown. We sought to determine the association between ICU bed availability, ICU readmissions, and ward cardiac arrests.

DESIGN

In this observational study using data collected between 2009 and 2011, rates of ICU readmission and ward cardiac arrest were determined per 12-hour shift. The relationship between these rates and the number of available ICU beds at the start of each shift (accounting for census and nursing capacity) was investigated. Grouped logistic regression was used to adjust for potential confounders.

SETTING

Five specialized adult ICUs comprising 63 adult ICU beds in an academic medical center.

PATIENTS

Any patient admitted to a non-ICU inpatient unit was counted in the ward census and considered at risk for ward cardiac arrest. Patients discharged from an ICU were considered at risk for ICU readmission.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Data were available for 2,086 of 2,190 shifts. The odds of ICU readmission increased with each decrease in the overall number of available ICU beds (odds ratio = 1.06; 95% CI, 1.00-1.12; p = 0.03), with a similar but not statistically significant association demonstrated in ward cardiac arrest rate (odds ratio = 1.06; 95% CI, 0.98-1.14; p = 0.16). In subgroup analysis, the odds of ward cardiac arrest increased with each decrease in the number of medical ICU beds available (odds ratio = 1.26; 95% CI, 1.06-1.49; p = 0.01).

CONCLUSIONS

Reduced ICU bed availability is associated with increased rates of ICU readmission and ward cardiac arrest. This suggests that systemic factors are associated with patient outcomes, and flexible critical care resources may be needed when demand is high.

Using electronic health record data to develop and validate a prediction model for adverse outcomes in the wards*

OBJECTIVE

Over 200,000 in-hospital cardiac arrests occur in the United States each year and many of these events may be preventable. Current vital sign-based risk scores for ward patients have demonstrated limited accuracy, which leads to missed opportunities to identify those patients most likely to suffer cardiac arrest and inefficient resource utilization. We derived and validated a prediction model for cardiac arrest while treating ICU transfer as a competing risk using electronic health record data.

DESIGN

A retrospective cohort study.

SETTING

An academic medical center in the United States with approximately 500 inpatient beds.

PATIENTS

Adult patients hospitalized from November 2008 until August 2011 who had documented ward vital signs.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Vital sign, demographic, location, and laboratory data were extracted from the electronic health record and investigated as potential predictor variables. A person-time multinomial logistic regression model was used to simultaneously predict cardiac arrest and ICU transfer. The prediction model was compared to the VitalPAC Early Warning Score using the area under the receiver operating characteristic curve and was validated using three-fold cross-validation. A total of 56,649 controls, 109 cardiac arrest patients, and 2,543 ICU transfers were included. The derived model more accurately detected cardiac arrest (area under the receiver operating characteristic curve, 0.88 vs 0.78; p < 0.001) and ICU transfer (area under the receiver operating characteristic curve, 0.77 vs 0.73; p < 0.001) than the VitalPAC Early Warning Score, and accuracy was similar with cross-validation. At a specificity of 93%, our model had a higher sensitivity than the VitalPAC Early Warning Score for cardiac arrest patients (65% vs 41%).

CONCLUSIONS

We developed and validated a prediction tool for ward patients that can simultaneously predict the risk of cardiac arrest and ICU transfer. Our model was more accurate than the VitalPAC Early Warning Score and could be implemented in the electronic health record to alert caregivers with real-time information regarding patient deterioration.

Respiratory rate at intensive care unit discharge after liver transplant is an independent risk factor for intensive care unit readmission within the same hospital stay: a nested case-control study

PURPOSE

Intensive care unit (ICU) readmission negatively impacts patients' outcomes. We aimed to characterize and determine risk factors for ICU readmission within the initial hospital stay after liver transplant (LT).

MATERIALS AND METHODS

The reference cohort included 369 LT recipients from a Canadian center between 2005 and 2012. One control was randomly selected per each case of ICU readmission within the initial hospital stay after LT. Survival analysis used the Kaplan-Meier method. Associations were studied by conditional logistic regression.

RESULTS

Fifty-two (14%) LT recipients were readmitted to the ICU within the initial hospital stay after LT; they had longer first hospital stay (P < .001) and lower 1-month to 2-year cumulative survival (P < .001). Respiratory failure was the major cause of ICU readmission (61%). Respiratory rate at discharge from the first ICU stay after LT was an independent risk factor for ICU readmission (odds ratio = 1.24). The cutoff point more than 20 breaths per minute prognosticated ICU readmission with a specificity of 90% and a positive predictive value of 80%.

CONCLUSIONS

Intensive care unit readmission within the initial hospital stay after LT negatively impacts LT recipients' outcomes. Monitoring respiratory rate at discharge from the first ICU stay after LT is important to prevent readmission.