Association between intensive care unit transfer delay and hospital mortality: A multicenter investigation

Real-time surveillance system for patient deterioration: a pragmatic cluster-randomized controlled trial

The COmmunicating Narrative Concerns Entered by RNs (CONCERN) early warning system (EWS) uses real-time nursing surveillance documentation patterns in its machine learning algorithm to identify deterioration risk. We conducted a 1-year, multisite, pragmatic trial with cluster-randomization of 74 clinical units (37 intervention; 37 usual care) across 2 health systems. Eligible adult hospital encounters were included. We tested if outcomes differed between patients whose care teams were and patients whose care teams were not informed by the CONCERN EWS. Coprimary outcomes were in-hospital mortality (examined as instantaneous risk) and length of stay. Secondary outcomes were cardiopulmonary arrest, sepsis, unanticipated intensive care unit transfers and 30-day hospital readmission. Among 60,893 hospital encounters (33,024 intervention; 27,869 usual care), intervention group encounters had 35.6% decreased instantaneous risk of death (adjusted hazard ratio (HR), 0.64; 95% confidence interval (CI), 0.53-0.78; P < 0.0001), 11.2% decreased length of stay (adjusted incidence rate ratio, 0.91; 95% CI, 0.90-0.93; P < 0.0001), 7.5% decreased instantaneous risk of sepsis (adjusted HR, 0.93; 95% CI, 0.86-0.99; P = 0.0317) and 24.9% increased instantaneous risk of unanticipated intensive care unit transfer (adjusted HR, 1.25; 95% CI, 1.09-1.43; P = 0.0011) compared with usual-care group encounters. No adverse events were reported. A machine learning-based EWS, modeled on nursing surveillance patterns, decreased inpatient deterioration risk with statistical significance. ClinicalTrials.gov registration: NCT03911687 .

Multicenter Development and Prospective Validation of eCARTv5: A Gradient-Boosted Machine-Learning Early Warning Score

BACKGROUND

Early detection of clinical deterioration using machine-learning early warning scores may improve outcomes. However, most implemented scores were developed using logistic regression, only underwent retrospective validation, and were not tested in important subgroups.

OBJECTIVE

The objective of our multicenter retrospective and prospective observational study was to develop and prospectively validate a gradient-boosted machine model (eCARTv5) for identifying clinical deterioration on the wards.

DERIVATION COHORT

All adult patients admitted to the inpatient medical-surgical wards at seven hospitals in three health systems for model development (2006-2022).

VALIDATION COHORT

All adult patients admitted to the inpatient medical-surgical wards and at 21 hospitals from three health systems for retrospective (2009-2023) and prospective (2023-2024) external validation.

PREDICTION MODEL

Predictor variables (demographics, vital signs, documentation, and laboratory values) were used in a gradient-boosted trees algorithm to predict ICU transfer or death in the next 24 hours. The developed model (eCARTv5) was compared with the Modified Early Warning Score (MEWS), the National Early Warning Score (NEWS), and eCARTv2 using the area under the receiver operating characteristic curve (AUROC).

RESULTS

The development cohort included 901,491 admissions, the retrospective validation cohort included 1,769,461 admissions, and the prospective validation cohort included 205,946 admissions. In retrospective validation, eCARTv5 had the highest AUROC (0.834; 95% CI, 0.834-0.835), followed by eCARTv2 (0.775 [95% CI, 0.775-0.776]), NEWS (0.766 [95% CI, 0.766-0.767]), and MEWS (0.704 [95% CI, 0.703-0.704]). eCARTv5's performance remained high (AUROC ≥0.80) across a range of patient demographics, clinical conditions, and during prospective validation.

CONCLUSION

We developed eCARTv5, which performed better than eCARTv2, NEWS, and MEWS retrospectively, prospectively, and across a range of subgroups. These results served as the foundation for Food and Drug Administration clearance for its use in identifying deterioration in hospitalized ward patients.

A proposed method for identifying Interfacility transfers in Medicare claims data

OBJECTIVE

To develop a method of consistently identifying interfacility transfers (IFTs) in Medicare Claims using patients with ST-Elevation Myocardial Infarction (STEMI) as an example.

DATA SOURCES/STUDY SETTING

100% Medicare inpatient and outpatient Standard Analytic Files and 5% Carrier Files, 2011-2020.

STUDY DESIGN

Observational, cross-sectional comparison of patient characteristics between proposed and existing methods.

DATA COLLECTION/EXTRACTION METHODS

We limited to patients aged 65+ with STEMI diagnosis using both proposed and existing methods.

PRINCIPAL FINDINGS

We identified 62,668 more IFTs using the proposed method (86,128 versus 23,460). A separately billable interfacility ambulance trip was found for more IFTs using the proposed than existing method (86% vs. 79%). Compared with the existing method, transferred patients under the proposed method were more likely to live in rural (p < 0.001) and lower income (p < 0.001) counties and were located farther away from emergency departments, trauma centers, and intensive care units (p < 0.001).

CONCLUSIONS

Identifying transferred patients based on two consecutive inpatient claims results in an undercount of IFTs and under-represents rural and low-income patients.

Differences in Characteristics of Peripartum Patients Who Did and Did Not Require an Upgrade to the Intensive Care Unit: A Propensity Score Matching Study

Background and Objectives: This study sought to identify predictors for peripartum patients admitted to non-intensive care wards who later upgraded to the Intensive Care Unit (ICU). Materials and Methods: This was a retrospective observational study of patients admitted to the Maternal Fetal Ward between 01/2017 and 12/2022, who later upgraded to the ICU. Upgraded patients were 1:1 propensity score matched with those who remained on the Maternal Fetal Ward (control). The Classification And Regression Tree, a machine learning algorithm, was used to identify significant predictors of ICU upgrade. Multivariable ordinal regression analysis was used to assess the time interval to upgrade. Results: From 1855 peripartum patients, we analyzed 37 control and 34 upgrade patients. Mean maternal age (±Standard Deviation) and gestational age for the group was 29.5 (±5.8) years and 31.5 (±7.5) weeks, respectively. The Median Sequential Organ Failure Assessment Score [Interquartile] at ward admission for the controls was 0 [0-1] versus 2 [0-3.3, p = 0.001] for upgrade patients. The Sequential Organ Failure Assessment score at Maternal Fetal Ward admission was most predictive, followed by the Acute Physiology and Chronic Health Evaluation II (APACHE II) score, and lactate dehydrogenase levels. The APACHE II score was also associated with ICU upgrade within 12 h of hospital admission (OR 1.4, 95% CI 1.08-1.91, p = 0.01). Conclusions: Compared to control patients, peripartum patients upgraded to the ICU are associated with higher physiologic scores at Maternal Fetal Ward admission. Until further studies are performed to confirm our observation, clinicians should pay attention to these physiologic scores, since they may be associated with higher-risk patients.

Mortality and Length of Stay Implications of Deterioration-Associated Transfer to the Intensive Care Unit over Different Time Frames

Background

Quality improvement initiatives in the acute care setting often target reduction of mortality and length of stay (LOS). Unplanned care escalations are associated with increased mortality risk and prolonged LOS, but may be precipitated by different factors, including appropriate triage, bed availability, and post-admission deterioration.

Objectives

This work evaluates different transfer timeframes to quantify the impact of deterioration-associated unplanned transfers to intensive care (ICU) on mortality and LOS, informing evidence-based interventions to improve patient care.

Design

This retrospective analysis examined 519 181 adult inpatients discharged from 15 hospitals in the United States. A propensity matched cohort analysis compared mortality and overall hospital LOS for patients admitted to routine and intermediate care units who did and did not have an unplanned ICU transfer within 12, 12-48, or ⩾48 hours from admission.

Methods

Population cohorts were matched on age, sex, admitting unit type, admission type, and admission acuity. Multivariable regression analysis was used to estimate the impact of unplanned transfer on mortality and LOS. Sensitivity sub-analyses compared direct ICU admissions to unplanned ICU transfers using the same transfer timeframes and endpoints.

Results

Patients with unplanned transfers in each of three timeframes had statistically higher mortality rates and longer LOS than matched cohorts without unplanned transfer. Differences between cohorts was greatest in patients transferring ⩾48 hours post-admission for both mortality (25.1% vs 1.9%, P < .0001) and LOS (x¯ = 14.7 vs 5.3, P < .0001). Multivariate analysis showed unplanned ICU transfer significantly increased odds of mortality and prolonged LOS, with later transfers having the most profound influence (19-fold increase in mortality and 2-fold increase in LOS). Sensitivity analyses found a statistically significant increase in mortality and LOS associated with unplanned ICU transfer across all three timeframes.

Conclusion

The association of later transfers with elevated mortality and LOS underscores the importance of timely intervention on patient deterioration.

Effectiveness of Remote Patient Monitoring Equipped With an Early Warning System in Tertiary Care Hospital Wards: Retrospective Cohort Study

BACKGROUND

Monitoring vital signs in hospitalized patients is crucial for evaluating their clinical condition. While early warning scores like the modified early warning score (MEWS) are typically calculated 3 to 4 times daily through spot checks, they might not promptly identify early deterioration. Leveraging technologies that provide continuous monitoring of vital signs, combined with an early warning system, has the potential to identify clinical deterioration sooner. This approach empowers health care providers to intervene promptly and effectively.

OBJECTIVE

This study aimed to assess the impact of a Remote Patient Monitoring System (RPMS) with an automated early warning system (R-EWS) on patient safety in noncritical care at a tertiary hospital. R-EWS performance was compared with a simulated Modified Early Warning System (S-MEWS) and a simulated threshold-based alert system (S-Threshold).

METHODS

Patient outcomes, including intensive care unit (ICU) transfers due to deterioration and discharges for nondeteriorating cases, were analyzed in Ramaiah Memorial Hospital's general wards with RPMS. Sensitivity, specificity, chi-square test for alert frequency distribution equality, and the average time from the first alert to ICU transfer in the last 24 hours was determined. Alert and patient distribution by tiers and vitals in R-EWS groups were examined.

RESULTS

Analyzing 905 patients, including 38 with deteriorations, R-EWS, S-Threshold, and S-MEWS generated more alerts for deteriorating cases. R-EWS showed high sensitivity (97.37%) and low specificity (23.41%), S-Threshold had perfect sensitivity (100%) but low specificity (0.46%), and S-MEWS demonstrated moderate sensitivity (47.37%) and high specificity (81.31%). The average time from initial alert to clinical deterioration was at least 18 hours for RPMS and S-Threshold in deteriorating participants. R-EWS had increased alert frequency and a higher proportion of critical alerts for deteriorating cases.

CONCLUSIONS

This study underscores R-EWS role in early deterioration detection, emphasizing timely interventions for improved patient outcomes. Continuous monitoring enhances patient safety and optimizes care quality.

Multicenter Development and Prospective Validation of eCARTv5: A Gradient Boosted Machine Learning Early Warning Score

OBJECTIVE

Early detection of clinical deterioration using machine learning early warning scores may improve outcomes. However, most implemented scores were developed using logistic regression, only underwent retrospective validation, and were not tested in important subgroups. Our objective was to develop and prospectively validate a gradient boosted machine model (eCARTv5) for identifying clinical deterioration on the wards.

DESIGN

Multicenter retrospective and prospective observational study.

SETTING

Inpatient admissions to the medical-surgical wards at seven hospitals in three health systems for model development (2006-2022) and at 21 hospitals from three health systems for retrospective (2009-2023) and prospective (2023-2024) external validation.

PATIENTS

All adult patients hospitalized at each participating health system during the study years.

INTERVENTIONS

None MEASUREMENTS AND MAIN RESULTS: Predictor variables (demographics, vital signs, documentation, and laboratory values) were used in a gradient boosted trees algorithm to predict intensive care unit transfer or death in the next 24 hours. The developed model (eCART) was compared to the Modified Early Warning Score (MEWS) and the National Early Warning Score (NEWS) using the area under the receiver operating characteristic curve (AUROC). The development cohort included 901,491 admissions, the retrospective validation cohort included 1,769,461 admissions, and the prospective validation cohort included 205,946 admissions. In retrospective validation, eCART had the highest AUROC (0.835; 95%CI 0.834, 0.835), followed by NEWS (0.766 (95%CI 0.766, 0.767)), and MEWS (0.704 (95%CI 0.703, 0.704)). eCART's performance remained high (AUROC ≥0.80) across a range of patient demographics, clinical conditions, and during prospective validation.

CONCLUSIONS

We developed eCART, which performed better than the NEWS and MEWS retrospectively, prospectively, and across a range of subgroups. These results served as the foundation for Food and Drug Administration clearance for its use in identifying deterioration in hospitalized ward patients.

Early Warning Scores With and Without Artificial Intelligence

Importance

Early warning decision support tools to identify clinical deterioration in the hospital are widely used, but there is little information on their comparative performance.

Objective

To compare 3 proprietary artificial intelligence (AI) early warning scores and 3 publicly available simple aggregated weighted scores.

Design, Setting, and Participants

This retrospective cohort study was performed at 7 hospitals in the Yale New Haven Health System. All consecutive adult medical-surgical ward hospital encounters between March 9, 2019, and November 9, 2023, were included.

Exposures

Simultaneous Epic Deterioration Index (EDI), Rothman Index (RI), eCARTv5 (eCART), Modified Early Warning Score (MEWS), National Early Warning Score (NEWS), and NEWS2 scores.

Main Outcomes and Measures

Clinical deterioration, defined as a transfer from ward to intensive care unit or death within 24 hours of an observation.

Results

Of the 362 926 patient encounters (median patient age, 64 [IQR, 47-77] years; 200 642 [55.3%] female), 16 693 (4.6%) experienced a clinical deterioration event. eCART had the highest area under the receiver operating characteristic curve at 0.895 (95% CI, 0.891-0.900), followed by NEWS2 at 0.831 (95% CI, 0.826-0.836), NEWS at 0.829 (95% CI, 0.824-0.835), RI at 0.828 (95% CI, 0.823-0.834), EDI at 0.808 (95% CI, 0.802-0.812), and MEWS at 0.757 (95% CI, 0.750-0.764). After matching scores at the moderate-risk sensitivity level for a NEWS score of 5, overall positive predictive values (PPVs) ranged from a low of 6.3% (95% CI, 6.1%-6.4%) for an EDI score of 41 to a high of 17.3% (95% CI, 16.9%-17.8%) for an eCART score of 94. Matching scores at the high-risk specificity of a NEWS score of 7 yielded overall PPVs ranging from a low of 14.5% (95% CI, 14.0%-15.2%) for an EDI score of 54 to a high of 23.3% (95% CI, 22.7%-24.2%) for an eCART score of 97. The moderate-risk thresholds provided a median of at least 20 hours of lead time for all the scores. Median lead time at the high-risk threshold was 11 (IQR, 0-69) hours for eCART, 8 (IQR, 0-63) hours for NEWS, 6 (IQR, 0-62) hours for NEWS2, 5 (IQR, 0-56) hours for MEWS, 1 (IQR, 0-39) hour for EDI, and 0 (IQR, 0-42) hours for RI.

Conclusions and Relevance

In this cohort study of inpatient encounters, eCART outperformed the other AI and non-AI scores, identifying more deteriorating patients with fewer false alarms and sufficient time to intervene. NEWS, a non-AI, publicly available early warning score, significantly outperformed EDI. Given the wide variation in accuracy, additional transparency and oversight of early warning tools may be warranted.

Efficacy of Early Warning Scores as the Prediction Tool for Detecting Patients With Acute Deterioration in a High Dependent Unit

Background Early detection and response to patient deterioration are essential to prevent serious outcomes such as unplanned intensive care unit (ICU) transfers and cardiac arrests. Rapid response systems (RRS) have been implemented globally, leading to a reduction in in-hospital mortality. In high-dependency units (HDUs), where patient monitoring is more frequent than in general wards, but staffing levels are lower than in intensive care units (ICUs), the challenge of identifying deteriorating patients persists. The Early Warning Score System (EWSS), including the National EWS (NEWS) and the artificial intelligence (AI)-based Visensia Safety Index (VSI), offer tools for early detection of patients' exacerbations. This study evaluates the effectiveness of NEWS and VSI in predicting patient deterioration in an HDU in Japan. Methods This single-center retrospective cohort study was conducted at a 344-bed acute care hospital. The study population included patients admitted to the HDU between September 27, 2019, and December 31, 2019. Study outcomes were unexpected ICU admission, endotracheal intubation, de novo use of vasopressors, de novo use of non-invasive positive pressure ventilation (NPPV), cardiac arrest, and the composite outcome (any of these outcomes). The predictive accuracy of the National Early Warning Score (NEWS) and the Visensia Safety Index (VSI) for detecting outcomes within 24 hours from each time point was analyzed. Results A total of 356 patients were included, with a median age of 76 years old (interquartile range (IQR): 64-84) and a median HDU stay of 2.0 days (IQR: 2.0-3.0). Among the 2648 analyzed vital sign sets, the median NEWS score was 5.0 (IQR: 3.0-7.0) and the median VSI score was 1.1 (IQR: 0.7-1.9). Twenty-six patients (7.3%) experienced outcomes. Among these, 13 (3.7%) required unexpected ICU transfers, 6 (1.7%) required endotracheal intubation, 15 (4.2%) required de novo vasopressor use, and 7 (2.0%) required de novo NPPV. NEWS could predict unexpected ICU admissions with good accuracy (the area under the curve (AUC) 0.834), as well as de novo vasopressor use (AUC 0.765) and the composite outcome (AUC 0.701). VSI also demonstrated modest predictive efficacy for unexpected ICU transfers (AUC 0.767), endotracheal intubation (AUC 0.712), and de novo vasopressor use (AUC 0.733). Based on ROC analysis, the appropriate threshold for NEWS ranged between 6.0 and 8.0, and the optimal threshold for VSI was found to be 1.5. Chronologically, NEWS remained stable before the occurrence of outcomes, whereas VSI showed a significant upward trend, suggesting that VSI may be more sensitive to detecting early deterioration. Conclusions In this study, both NEWS and VSI demonstrated modest accuracy in predicting unexpected ICU admission and vasopressor use in an HDU. While NEWS high risk (7≥) is an appropriate cutoff value for detecting adverse outcomes, particularly lowering the VSI cutoff to 1.5, may enhance the identification of high-risk patients.

Causes, Diagnostic Testing, and Treatments Related to Clinical Deterioration Events Among High-Risk Ward Patients

IMPORTANCE

Timely intervention for clinically deteriorating ward patients requires that care teams accurately diagnose and treat their underlying medical conditions. However, the most common diagnoses leading to deterioration and the relevant therapies provided are poorly characterized.

OBJECTIVES

We aimed to determine the diagnoses responsible for clinical deterioration, the relevant diagnostic tests ordered, and the treatments administered among high-risk ward patients using manual chart review.

DESIGN, SETTING, AND PARTICIPANTS

This was a multicenter retrospective observational study in inpatient medical-surgical wards at four health systems from 2006 to 2020. Randomly selected patients (1000 from each health system) with clinical deterioration, defined by reaching the 95th percentile of a validated early warning score, electronic Cardiac Arrest Risk Triage, were included.

MAIN OUTCOMES AND MEASURES

Clinical deterioration was confirmed by a trained reviewer or marked as a false alarm if no deterioration occurred for each patient. For true deterioration events, the condition causing deterioration, relevant diagnostic tests ordered, and treatments provided were collected.

RESULTS

Of the 4000 included patients, 2484 (62%) had clinical deterioration confirmed by chart review. Sepsis was the most common cause of deterioration (41%; n = 1021), followed by arrhythmia (19%; n = 473), while liver failure had the highest in-hospital mortality (41%). The most common diagnostic tests ordered were complete blood counts (47% of events), followed by chest radiographs (42%) and cultures (40%), while the most common medication orders were antimicrobials (46%), followed by fluid boluses (34%) and antiarrhythmics (19%).

CONCLUSIONS AND RELEVANCE

We found that sepsis was the most common cause of deterioration, while liver failure had the highest mortality. Complete blood counts and chest radiographs were the most common diagnostic tests ordered, and antimicrobials and fluid boluses were the most common medication interventions. These results provide important insights for clinical decision-making at the bedside, training of rapid response teams, and the development of institutional treatment pathways for clinical deterioration.

Beyond Unplanned ICU Transfers: Linking a Revised Definition of Deterioration to Patient Outcomes

OBJECTIVES

To develop an electronic descriptor of clinical deterioration for hospitalized patients that predicts short-term mortality and identifies patient deterioration earlier than current standard definitions.

DESIGN

A retrospective study using exploratory record review, quantitative analysis, and regression analyses.

SETTING

Twelve-hospital community-academic health system.

PATIENTS

All adult patients with an acute hospital encounter between January 1, 2018, and December 31, 2022.

INTERVENTIONS

Not applicable.

MEASUREMENTS AND MAIN RESULTS

Clinical trigger events were selected and used to create a revised electronic definition of deterioration, encompassing signals of respiratory failure, bleeding, and hypotension occurring in proximity to ICU transfer. Patients meeting the revised definition were 12.5 times more likely to die within 7 days (adjusted odds ratio 12.5; 95% CI, 8.9-17.4) and had a 95.3% longer length of stay (95% CI, 88.6-102.3%) compared with those who were transferred to the ICU or died regardless of meeting the revised definition. Among the 1812 patients who met the revised definition of deterioration before ICU transfer (52.4%), the median detection time was 157.0 min earlier (interquartile range 64.0-363.5 min).

CONCLUSIONS

The revised definition of deterioration establishes an electronic descriptor of clinical deterioration that is strongly associated with short-term mortality and length of stay and identifies deterioration over 2.5 hours earlier than ICU transfer. Incorporating the revised definition of deterioration into the training and validation of early warning system algorithms may enhance their timeliness and clinical accuracy.

Development and Validation of a Deep Learning Model for Prediction of Adult Physiological Deterioration

BACKGROUND

Prediction-based strategies for physiologic deterioration offer the potential for earlier clinical interventions that improve patient outcomes. Current strategies are limited because they operate on inconsistent definitions of deterioration, attempt to dichotomize a dynamic and progressive phenomenon, and offer poor performance.

OBJECTIVE

Can a deep learning deterioration prediction model (Deep Learning Enhanced Triage and Emergency Response for Inpatient Optimization [DETERIO]) based on a consensus definition of deterioration (the Adult Inpatient Decompensation Event [AIDE] criteria) and that approaches deterioration as a state "value-estimation" problem outperform a commercially available deterioration score?

DERIVATION COHORT

The derivation cohort contained retrospective patient data collected from both inpatient services (inpatient) and emergency departments (EDs) of two hospitals within the University of California San Diego Health System. There were 330,729 total patients; 71,735 were inpatient and 258,994 were ED. Of these data, 20% were randomly sampled as a retrospective "testing set."

VALIDATION COHORT

The validation cohort contained temporal patient data. There were 65,898 total patients; 13,750 were inpatient and 52,148 were ED.

PREDICTION MODEL

DETERIO was developed and validated on these data, using the AIDE criteria to generate a composite score. DETERIO's architecture builds upon previous work. DETERIO's prediction performance up to 12 hours before T0 was compared against Epic Deterioration Index (EDI).

RESULTS

In the retrospective testing set, DETERIO's area under the receiver operating characteristic curve (AUC) was 0.797 and 0.874 for inpatient and ED subsets, respectively. In the temporal validation cohort, the corresponding AUC were 0.775 and 0.856, respectively. DETERIO outperformed EDI in the inpatient validation cohort (AUC, 0.775 vs. 0.721; p < 0.01) while maintaining superior sensitivity and a comparable rate of false alarms (sensitivity, 45.50% vs. 30.00%; positive predictive value, 20.50% vs. 16.11%).

CONCLUSIONS

DETERIO demonstrates promise in the viability of a state value-estimation approach for predicting adult physiologic deterioration. It may outperform EDI while offering additional clinical utility in triage and clinician interaction with prediction confidence and explanations. Additional studies are needed to assess generalizability and real-world clinical impact.

Intensive care unit admission criteria: a scoping review

Background

Effectively identifying deteriorated patients is vital to the development and validation of automated systems designed to predict clinical deterioration. Existing outcome measures used for this purpose have significant limitations. Published criteria for admission to high acuity inpatient areas may represent markers of patient deterioration and could inform the development of alternate outcome measures.

Objectives

In this scoping review, we aimed to characterise published criteria for admission of adult inpatients to high acuity inpatient areas including intensive care units. A secondary aim was to identify variables that are extractable from electronic health records (EHRs).

Data sources

Electronic databases PubMed and ProQuest EBook Central were searched to identify papers published from 1999 to date of search. We included publications which described prescriptive criteria for admission of adult inpatients to a clinical area with a higher level of care than a general hospital ward.

Charting methods

Data was extracted from each publication using a standardised data-charting form. Admission criteria characteristics were summarised and cross-tabulated for each criterion by population group.

Results

Five domains were identified: diagnosis-based criteria, clinical parameter criteria, organ-support criteria, organ-monitoring criteria and patient baseline criteria. Six clinical parameter-based criteria and five needs-based criteria were frequently proposed and represent variables extractable from EHRs. Thresholds for objective clinical parameter criteria varied across publications, and by disease subgroup, and universal cut-offs for criteria could not be elucidated.

Conclusions

This study identified multiple criteria which may represent markers of deterioration. Many of the criteria are extractable from the EHR, making them potential candidates for future automated systems. Variability in admission criteria and associated thresholds across the literature suggests clinical deterioration is a heterogeneous phenomenon which may resist being defined as a single entity via a consensus-driven process.

Tele-Critical Care Support Outside the Intensive Care Unit

Tele-intensive care unit (ICU), or Tele Critical Care (TCC), has been in active use for 25 years and has expanded beyond the original model to support critically ill patients beyond the confines of the ICU. Here, the author reviews the role of TCC in supporting rapid response events, critical care in emergency departments, and disaster and pandemic responses. The ability to rapidly expand critical care services has important capacity and care quality implications. Moreover, as TCC infrastructure becomes less expensive, the opportunities to leverage this care modality also have potentially important financial benefits.

Development and Validation of a Machine Learning COVID-19 Veteran (COVet) Deterioration Risk Score

BACKGROUND AND OBJECTIVE

To develop the COVid Veteran (COVet) score for clinical deterioration in Veterans hospitalized with COVID-19 and further validate this model in both Veteran and non-Veteran samples. No such score has been derived and validated while incorporating a Veteran sample.

DERIVATION COHORT

Adults (age ≥ 18 yr) hospitalized outside the ICU with a diagnosis of COVID-19 for model development to the Veterans Health Administration (VHA) (n = 80 hospitals).

VALIDATION COHORT

External validation occurred in a VHA cohort of 34 hospitals, as well as six non-Veteran health systems for further external validation (n = 21 hospitals) between 2020 and 2023.

PREDICTION MODEL

eXtreme Gradient Boosting machine learning methods were used, and performance was assessed using the area under the receiver operating characteristic curve and compared with the National Early Warning Score (NEWS). The primary outcome was transfer to the ICU or death within 24 hours of each new variable observation. Model predictor variables included demographics, vital signs, structured flowsheet data, and laboratory values.

RESULTS

A total of 96,908 admissions occurred during the study period, of which 59,897 were in the Veteran sample and 37,011 were in the non-Veteran sample. During external validation in the Veteran sample, the model demonstrated excellent discrimination, with an area under the receiver operating characteristic curve of 0.88. This was significantly higher than NEWS (0.79; p < 0.01). In the non-Veteran sample, the model also demonstrated excellent discrimination (0.86 vs. 0.79 for NEWS; p < 0.01). The top three variables of importance were eosinophil percentage, mean oxygen saturation in the prior 24-hour period, and worst mental status in the prior 24-hour period.

CONCLUSIONS

We used machine learning methods to develop and validate a highly accurate early warning score in both Veterans and non-Veterans hospitalized with COVID-19. The model could lead to earlier identification and therapy, which may improve outcomes.

Multisite Pragmatic Cluster-Randomized Controlled Trial of the CONCERN Early Warning System

Importance

Late predictions of hospitalized patient deterioration, resulting from early warning systems (EWS) with limited data sources and/or a care team's lack of shared situational awareness, contribute to delays in clinical interventions. The COmmunicating Narrative Concerns Entered by RNs (CONCERN) Early Warning System (EWS) uses real-time nursing surveillance documentation patterns in its machine learning algorithm to identify patients' deterioration risk up to 42 hours earlier than other EWSs.

Objective

To test our a priori hypothesis that patients with care teams informed by the CONCERN EWS intervention have a lower mortality rate and shorter length of stay (LOS) than the patients with teams not informed by CONCERN EWS.

Design

One-year multisite, pragmatic controlled clinical trial with cluster-randomization of acute and intensive care units to intervention or usual-care groups.

Setting

Two large U.S. health systems.

Participants

Adult patients admitted to acute and intensive care units, excluding those on hospice/palliative/comfort care, or with Do Not Resuscitate/Do Not Intubate orders.

Intervention

The CONCERN EWS intervention calculates patient deterioration risk based on nurses' concern levels measured by surveillance documentation patterns, and it displays the categorical risk score (low, increased, high) in the electronic health record (EHR) for care team members.

Main Outcomes and Measures

Primary outcomes: in-hospital mortality, LOS; survival analysis was used. Secondary outcomes: cardiopulmonary arrest, sepsis, unanticipated ICU transfers, 30-day hospital readmission.

Results

A total of 60 893 hospital encounters (33 024 intervention and 27 869 usual-care) were included. Both groups had similar patient age, race, ethnicity, and illness severity distributions. Patients in the intervention group had a 35.6% decreased risk of death (adjusted hazard ratio [HR], 0.644; 95% confidence interval [CI], 0.532-0.778; P<.0001), 11.2% decreased LOS (adjusted incidence rate ratio, 0.914; 95% CI, 0.902-0.926; P<.0001), 7.5% decreased risk of sepsis (adjusted HR, 0.925; 95% CI, 0.861-0.993; P=.0317), and 24.9% increased risk of unanticipated ICU transfer (adjusted HR, 1.249; 95% CI, 1.093-1.426; P=.0011) compared with patients in the usual-care group.

Conclusions and Relevance

A hospital-wide EWS based on nursing surveillance patterns decreased in-hospital mortality, sepsis, and LOS when integrated into the care team's EHR workflow.

Trial Registration

ClinicalTrials.gov Identifier: NCT03911687.

Effect of a fall within three months of admission on delirium in critically Ill elderly patients: a population-based cohort study

BACKGROUND

Delirium is common among elderly patients in the intensive care unit (ICU) and is associated with prolonged hospitalization, increased healthcare costs, and increased risk of death. Understanding the potential risk factors and early prevention of delirium is critical to facilitate timely intervention that may reverse or mitigate the harmful consequences of delirium.

AIM

To clarify the effects of pre-admission falls on ICU outcomes, primarily delirium, and secondarily pressure injuries and urinary tract infections.

METHODS

The study relied on data sourced from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Statistical tests (Wilcoxon rank-sum or chi-squared) compared cohort characteristics. Logistic regression was employed to investigate the association between a history of falls and delirium, as well as secondary outcomes, while Kaplan-Meier survival curves were used to assess short-term survival in delirium and non-delirium patients.

RESULTS

Study encompassed 22,547 participants. Delirium incidence was 40%, significantly higher in patients with a history of falls (54.4% vs. 34.5%, p < 0.001). Logistic regression, controlling for confounders, not only confirmed that a history of falls elevates the odds of delirium (OR: 2.11; 95% CI: 1.97-2.26; p < 0.001) but also showed it increases the incidence of urinary tract infections (OR:1.50; 95% CI:1.40-1.62; p < 0.001) and pressure injuries (OR:1.36; 95% CI:1.26-1.47; p < 0.001). Elderly delirium patients exhibited lower 30-, 180-, and 360-day survival rates than non-delirium counterparts (all p < 0.001).

CONCLUSIONS

The study reveals that history of falls significantly heighten the risk of delirium and other adverse outcomes in elderly ICU patients, leading to decreased short-term survival rates. This emphasizes the critical need for early interventions and could inform future strategies to manage and prevent these conditions in ICU settings.

Implementation of an Interdisciplinary Transfer Huddle Intervention for Prolonged Wait Times During Inter-ICU Transfer

BACKGROUND

ICU transfers from a regional to a tertiary-level hospital are initiated typically for a higher level of care. Extended transfer wait times can negatively affect survival, length of stay (LOS), and cost.

METHODS

In this prospective single-center study, the subjects were adult ICU patients admitted to regional hospitals between January and October 2022, for whom a request was made to transfer to a tertiary-level medical ICU. The authors developed and implemented an interdisciplinary transfer huddle intervention (THI) with the goal of reducing wait times by providing a consistent channel of communication between key stakeholders. The primary outcome was the number of hours elapsed between transfer request and the time of transfer to the tertiary hospital. Secondary outcomes included in-hospital mortality, discharge to home, ICU LOS, and hospital LOS. Data were abstracted from electronic health records and periods before (January to June 2022) and after (June to October 2022) the intervention were compared. Data were analyzed using logistic regression or negative binomial regression, adjusting for patient demographic and clinical characteristics. ICU fellows also completed a daily survey about barriers they perceived to the THI application.

RESULTS

During the study period, 76 patients were transferred. The THI was completed 75.0% of the time. There were no statistically significant differences in the primary and secondary outcomes before and after the intervention. The top perceived barriers to transfer were lack of physical beds (50.0%) and staffing limitations (37.5%).

CONCLUSION

The authors successfully developed and implemented a transfer huddle to ensure consistent interdisciplinary communication for patients being transferred between ICUs and identified barriers to such transfer. However, transfer times and patient outcomes were not significantly different after the change. Future studies should consider staffing challenges, hospital capacity, and the role of dedicated transfer teams in in decreasing inter-ICU transfer wait times.

Multi-modal learning for inpatient length of stay prediction

Predicting inpatient length of stay (LoS) is important for hospitals aiming to improve service efficiency and enhance management capabilities. Patient medical records are strongly associated with LoS. However, due to diverse modalities, heterogeneity, and complexity of data, it becomes challenging to effectively leverage these heterogeneous data to put forth a predictive model that can accurately predict LoS. To address the challenge, this study aims to establish a novel data-fusion model, termed as DF-Mdl, to integrate heterogeneous clinical data for predicting the LoS of inpatients between hospital discharge and admission. Multi-modal data such as demographic data, clinical notes, laboratory test results, and medical images are utilized in our proposed methodology with individual "basic" sub-models separately applied to each different data modality. Specifically, a convolutional neural network (CNN) model, which we termed CRXMDL, is designed for chest X-ray (CXR) image data, two long short-term memory networks are used to extract features from long text data, and a novel attention-embedded 1D convolutional neural network is developed to extract useful information from numerical data. Finally, these basic models are integrated to form a new data-fusion model (DF-Mdl) for inpatient LoS prediction. The proposed method attains the best R2 and EVAR values of 0.6039 and 0.6042 among competitors for the LoS prediction on the Medical Information Mart for Intensive Care (MIMIC)-IV test dataset. Empirical evidence suggests better performance compared with other state-of-the-art (SOTA) methods, which demonstrates the effectiveness and feasibility of the proposed approach.

Causes, Diagnostic Testing, and Treatments Related to Clinical Deterioration Events among High-Risk Ward Patients

OBJECTIVE

Timely intervention for clinically deteriorating ward patients requires that care teams accurately diagnose and treat their underlying medical conditions. However, the most common diagnoses leading to deterioration and the relevant therapies provided are poorly characterized. Therefore, we aimed to determine the diagnoses responsible for clinical deterioration, the relevant diagnostic tests ordered, and the treatments administered among high-risk ward patients using manual chart review.

DESIGN

Multicenter retrospective observational study.

SETTING

Inpatient medical-surgical wards at four health systems from 2006-2020 PATIENTS: Randomly selected patients (1,000 from each health system) with clinical deterioration, defined by reaching the 95th percentile of a validated early warning score, electronic Cardiac Arrest Risk Triage (eCART), were included.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Clinical deterioration was confirmed by a trained reviewer or marked as a false alarm if no deterioration occurred for each patient. For true deterioration events, the condition causing deterioration, relevant diagnostic tests ordered, and treatments provided were collected. Of the 4,000 included patients, 2,484 (62%) had clinical deterioration confirmed by chart review. Sepsis was the most common cause of deterioration (41%; n=1,021), followed by arrhythmia (19%; n=473), while liver failure had the highest in-hospital mortality (41%). The most common diagnostic tests ordered were complete blood counts (47% of events), followed by chest x-rays (42%), and cultures (40%), while the most common medication orders were antimicrobials (46%), followed by fluid boluses (34%), and antiarrhythmics (19%).

CONCLUSIONS

We found that sepsis was the most common cause of deterioration, while liver failure had the highest mortality. Complete blood counts and chest x-rays were the most common diagnostic tests ordered, and antimicrobials and fluid boluses were the most common medication interventions. These results provide important insights for clinical decision-making at the bedside, training of rapid response teams, and the development of institutional treatment pathways for clinical deterioration.

KEY POINTS

Question: What are the most common diagnoses, diagnostic test orders, and treatments for ward patients experiencing clinical deterioration? Findings: In manual chart review of 2,484 encounters with deterioration across four health systems, we found that sepsis was the most common cause of clinical deterioration, followed by arrythmias, while liver failure had the highest mortality. Complete blood counts and chest x-rays were the most common diagnostic test orders, while antimicrobials and fluid boluses were the most common treatments. Meaning: Our results provide new insights into clinical deterioration events, which can inform institutional treatment pathways, rapid response team training, and patient care.

Epidemiology and outcomes of previously healthy critically ill patients with COVID-19: A population-based cohort

Comorbid conditions represent a major risk for severe illness among persons with COVID-19. Previously healthy people with COVID-19 can also develop severe illness, but are expected to have better outcomes than those with comorbid conditions. Nevertheless, recent data suggest that the former may have, counterintuitively, higher risk of death among those with non-COVID sepsis. However, the epidemiology and outcomes of previously healthy people among critically ill patients with COVID-19 are unknown. We used statewide data to identify intensive care unit (ICU) admissions aged ≥18 years in Texas with COVID-19 in 2020. Multilevel logistic regression was used to estimate the association of comorbid state with short-term mortality (defined as in-hospital mortality or discharge to hospice) overall and with higher illness severity among ICU admissions. Among 52,776 ICU admissions with COVID-19, 6373 (12.1%) were previously healthy. Short-term mortality among previously healthy ICU admissions and those with comorbidities was 16.9% versus 34.6%. On adjusted analyses, the odds of short-term mortality were lower among the previously healthy compared to those with comorbidities overall (adjusted odds ratio (aOR) 0.84 (95% CI: 0.73-0.98)), but did not differ among those with ≥3 organ dysfunctions (aOR 1.11 (95% CI: 0.84-1.46)) and the mechanically ventilated (aOR 0.87 (95% CI: 0.68-1.12)), while being higher among those with do-not-resuscitate status (aOR 1.40 (95% CI: 1.04-1.89)). Over one in eight ICU admissions with COVID-19 were previously healthy. Although being previously healthy was associated with lower risk of death compared to those with comorbidities overall, it had no prognostic advantage among the more severely ill.

Risk Prediction Models for Hospital Mortality in General Medical Patients: A Systematic Review

Objective

To systematically review contemporary prediction models for hospital mortality developed or validated in general medical patients.

Methods

We screened articles in five databases, from January 1, 2010, through April 7, 2022, and the bibliography of articles selected for final inclusion. We assessed the quality for risk of bias and applicability using the Prediction Model Risk of Bias Assessment Tool (PROBAST) and extracted data using the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) checklist. Two investigators independently screened each article, assessed quality, and extracted data.

Results

From 20,424 unique articles, we identified 15 models in 8 studies across 10 countries. The studies included 280,793 general medical patients and 19,923 hospital deaths. Models included 7 early warning scores, 2 comorbidities indices, and 6 combination models. Ten models were studied in all general medical patients (general models) and 7 in general medical patients with infection (infection models). Of the 15 models, 13 were developed using logistic or Poisson regression and 2 using machine learning methods. Also, 4 of 15 models reported on handling of missing values. None of the infection models had high discrimination, whereas 4 of 10 general models had high discrimination (area under curve >0.8). Only 1 model appropriately assessed calibration. All models had high risk of bias; 4 of 10 general models and 5 of 7 infection models had low concern for applicability for general medical patients.

Conclusion

Mortality prediction models for general medical patients were sparse and differed in quality, applicability, and discrimination. These models require hospital-level validation and/or recalibration in general medical patients to guide mortality reduction interventions.

A deep learning approach for inpatient length of stay and mortality prediction

PURPOSE

Accurate prediction of the Length of Stay (LoS) and mortality in the Intensive Care Unit (ICU) is crucial for effective hospital management, and it can assist clinicians for real-time demand capacity (RTDC) administration, thereby improving healthcare quality and service levels.

METHODS

This paper proposes a novel one-dimensional (1D) multi-scale convolutional neural network architecture, namely 1D-MSNet, to predict inpatients' LoS and mortality in ICU. First, a 1D multi-scale convolution framework is proposed to enlarge the convolutional receptive fields and enhance the richness of the convolutional features. Following the convolutional layers, an atrous causal spatial pyramid pooling (SPP) module is incorporated into the networks to extract high-level features. The optimized Focal Loss (FL) function is combined with the synthetic minority over-sampling technique (SMOTE) to mitigate the imbalanced-class issue.

RESULTS

On the MIMIC-IV v1.0 benchmark dataset, the proposed approach achieves the optimum R-Square and RMSE values of 0.57 and 3.61 for the LoS prediction, and the highest test accuracy of 97.73% for the mortality prediction.

CONCLUSION

The proposed approach presents a superior performance in comparison with other state-of-the-art, and it can effectively perform the LoS and mortality prediction tasks.

Predicting Lead-Time RSV-Related Pediatric Hospitalizations From Historic Google Trend Search

OBJECTIVES

Respiratory syncytial virus (RSV) causes seasonal outbreaks of respiratory tract infections in children, leading to increased emergency department visits and hospitalizations. Although the risk of severe illnesses difficult to predict, the sudden surge in RSV may strain the health care system. Therefore, the objective of this study was to examine the utility of Google Trends search activity on RSV to predict changes in RSV-related hospitalizations in children in the United States in 2019.

METHODS

A retrospective cross-sectional analysis of pediatric hospitalization was conducted using the 2019 HCUP-Kids Inpatient Database. Google Trends search activity for "RSV" was abstracted as a monthly relative interest score for 2019. RSV-related hospitalizations were identified using International Classification of Diseases 9/10 codes. We applied finite distributed lag models to estimate the causal effect over time of historical relative search activity and long-run propensity to calculate the cumulative effect of changes in relative search activity on admission rate.

RESULTS

Of the total 102 127 RSV-related pediatric hospitalizations, 90% were in those aged ≤2 years. Admissions were common in males (55%), non-Hispanic Whites (50%), and South region (39%). Across 2o successive months, the cumulative effect of a 1-unit score increase in relative interest was associated with an increase of 140.7 (95% confidence interval, 96.2-185.2; P < .05) RSV-related admissions.

CONCLUSIONS

Historic Google Trends search activity for RSV predicts lead-time RSV-related pediatric hospitalization. Further studies are needed to validate these findings using regional health systems.

Delayed Transfer of Critically Ill Patients from Emergency Department to Intensive Care Unit

Background and aim

Delay in the transfer of critically ill patients from the emergency department (ED) to intensive care units (ICUs) may worsen clinical outcomes. This prospective, observational study was done to find the incidence of delayed transfer.

Materials and methods

After approval from the institute ethics committee and written informed consent, all patients admitted to ICU from ED over 6 months were divided into groups I and II as patients getting transferred to ICU within 30 minutes of the decision or not, respectively. The factors affecting the immediate transfer and clinical outcome of all patients were noted. Monthly feedback was given to the ED team.

Results

Out of 52 ICU admissions from ED, 35 (67.3%) patients were not transferred within 30 minutes, and the most frequent factor preventing immediate transfer was ED-related (54%). A statistically significant difference was found in acute physiology and chronic health evaluation (APACHE II) score, clinical deterioration during transfer, longer duration of mechanical ventilation and length of stay, and higher mortality with patients transferred immediately to ICU. A reduction of 42.6% was noted in transfer time from the first month to the last month of study.

Conclusion

The incidence of delayed transfer of patients from ED to ICU was 67.3% with ED-related factors being the most frequent cause of delay (54.2%).

How to cite this article

Bosco S, Sahni N, Jain A, Arora P, Raj V, Yaddanapudi L. Delayed Transfer of Critically Ill Patients from Emergency Department to Intensive Care Unit. Indian J Crit Care Med 2023;27(8):580-582.

Preferred Language Mediates Association Between Race, Ethnicity, and Delayed Presentation in Critically Ill Patients With COVID-19

Which social factors explain racial and ethnic disparities in COVID-19 access to care and outcomes remain unclear.

OBJECTIVES

We hypothesized that preferred language mediates the association between race, ethnicity and delays to care.

DESIGN SETTING AND PARTICIPANTS

Multicenter, retrospective cohort study of adults with COVID-19 consecutively admitted to the ICU in three Massachusetts hospitals in 2020.

MAIN OUTCOME AND MEASURES

Causal mediation analysis was performed to evaluate potential mediators including preferred language, insurance status, and neighborhood characteristics.

RESULTS

Non-Hispanic White (NHW) patients (157/442, 36%) were more likely to speak English as their preferred language (78% vs. 13%), were less likely to be un- or under-insured (1% vs. 28%), lived in neighborhoods with lower social vulnerability index (SVI) than patients from racial and ethnic minority groups (SVI percentile 59 [28] vs. 74 [21]) but had more comorbidities (Charlson comorbidity index 4.6 [2.5] vs. 3.0 [2.5]), and were older (70 [13.2] vs. 58 [15.1] years). From symptom onset, NHW patients were admitted 1.67 [0.71-2.63] days earlier than patients from racial and ethnic minority groups (p < 0.01). Non-English preferred language was associated with delay to admission of 1.29 [0.40-2.18] days (p < 0.01). Preferred language mediated 63% of the total effect (p = 0.02) between race, ethnicity and days from symptom onset to hospital admission. Insurance status, social vulnerability, and distance to the hospital were not on the causal pathway between race, ethnicity and delay to admission.

CONCLUSIONS AND RELEVANCE

Preferred language mediates the association between race, ethnicity and delays to presentation for critically ill patients with COVID-19, although our results are limited by possible collider stratification bias. Effective COVID-19 treatments require early diagnosis, and delays are associated with increased mortality. Further research on the role preferred language plays in racial and ethnic disparities may identify effective solutions for equitable care.

Towards Predicting Length of Stay and Identification of Cohort Risk Factors Using Self-Attention-Based Transformers and Association Mining: COVID-19 as a Phenotype

Predicting length of stay (LoS) and understanding its underlying factors is essential to minimizing the risk of hospital-acquired conditions, improving financial, operational, and clinical outcomes, and better managing future pandemics. The purpose of this study was to forecast patients' LoS using a deep learning model and to analyze cohorts of risk factors reducing or prolonging LoS. We employed various preprocessing techniques, SMOTE-N to balance data, and a TabTransformer model to forecast LoS. Finally, the Apriori algorithm was applied to analyze cohorts of risk factors influencing hospital LoS. The TabTransformer outperformed the base machine learning models in terms of F1 score (0.92), precision (0.83), recall (0.93), and accuracy (0.73) for the discharged dataset and F1 score (0.84), precision (0.75), recall (0.98), and accuracy (0.77) for the deceased dataset. The association mining algorithm was able to identify significant risk factors/indicators belonging to laboratory, X-ray, and clinical data, such as elevated LDH and D-dimer levels, lymphocyte count, and comorbidities such as hypertension and diabetes. It also reveals what treatments have reduced the symptoms of COVID-19 patients, leading to a reduction in LoS, particularly when no vaccines or medication, such as Paxlovid, were available.

National Early Warning Score Deployment in a Veterans Affairs Facility: A Quality Improvement Initiative and Analysis

Early warning scores are algorithms designed to identify clinical deterioration. Current literature is predominantly in non-Veteran populations. Studies in Veterans are lacking. This study was a prospective quality improvement project deploying and assessing the National Early Warning Score (NEWS) at Kansas City VA Medical Center. Performance of NEWS was assessed as follows: discrimination for predicting a composite outcome of intensive care unit transfer or mortality within 24 hours via area under the receiver operating curve. A total of 4781 Veterans with 142 375 NEWS values were included. The NEWS area under the receiver operating curve for the composite outcome was 0.72 (95% CI, 0.71-0.74), indicating acceptable predictive accuracy. A NEWS of ≥7 was more likely associated with the composite outcome versus <7 (13.6% vs 0.8%; P < 0.001). This is one of the first studies to demonstrate successful deployment of NEWS in a Veteran population, with resultant important implications across the Veterans Health Administration.

Patient and Operative Factors Predict Risk of Discretionary Prolonged Postoperative Mechanical Ventilation in a Broad Surgical Cohort

BACKGROUND

Patients undergoing surgery with general anesthesia and endotracheal intubation are ideally extubated upon case completion, as prolonged postoperative mechanical ventilation (PPMV) has been associated with poor outcomes. However, some patients require PPMV for surgical reasons, such as airway compromise, while others remain intubated at the discretion of the anesthesia provider. Incidence and risk factors for discretionary PPMV (DPPMV) have been described in individual surgical subspecialties and intensive care unit (ICU) populations, but are relatively understudied in a broad surgical cohort. The present study seeks to fill this gap and identify the perioperative risk factors that predict DPPMV.

METHODS

After obtaining institutional review board (IRB) exemption, existing electronic health record databases at our large referral center were retrospectively queried for adult surgeries performed between January 2018 and December 2020 with general anesthesia, endotracheal intubation, and by surgical services that do not routinely leave patients intubated for surgical reasons. Patients who arrived to the ICU intubated after surgery were identified as experiencing DPPMV. Selection of candidate risk factors was performed with LASSO-regularized logistic regression, and surviving variables were used to generate a multivariable logistic regression model of DPPMV risk.

RESULTS

A total of 32,915 cases met inclusion criteria, of which 415 (1.26%) experienced DPPMV. Compared to extubated patients, those with DPPMV were more likely to have undergone emergency surgery (42.9% versus 3.4%; P < .001), surgery during an existing ICU stay (30.8% versus 2.8%; P < 0.001), and have 20 of the 31 elixhauser comorbidities ( P < .05 for each comparison), among other differences. A risk model with 12 variables, including American Society of Anesthesiologists (ASA) physical classification status, emergency surgery designation, four Elixhauser comorbidities, surgery during an existing ICU stay, surgery duration, estimated number of intraoperative handoffs, and vasopressor, sodium bicarbonate, and albuterol administration, yielded an area under the receiver operating characteristic curve of 0.97 (95% confidence interval, 0.96-0.97) for prediction of DPPMV.

CONCLUSIONS

DPPMV was uncommon in this broad surgical cohort but could be accurately predicted using readily available patient-specific and operative factors. These results may be useful for preoperative risk stratification, postoperative resource allocation, and clinical trial planning.

Using EHR Data to Identify Patient Frailty and Risk for ICU Transfer

The predictive properties of four definitions of a frailty risk score (FRS) constructed using combinations of nursing flowsheet data, laboratory tests, and ICD-10 codes were examined for time to first intensive care unit (ICU) transfer in medical-surgical inpatients ≥50 years of age. Cox regression modeled time to first ICU transfer and Schemper-Henderson explained variance summarized predictive accuracy of FRS combinations. Modeling by age group and controlling for sex, all FRS measures significantly predicted time to first ICU transfer. Further multivariable modeling controlling for clinical characteristics substantially improved predictive accuracy. The effect of frailty on time to first ICU transfer depended on age, with highest risk in 50 to <60 years and ≥80 years age groups. Frailty prevalence ranged from 25.1% to 56.4%. Findings indicate that FRS-based frailty is a risk factor for time to first ICU transfer and should be considered in assessment and care-planning to address frailty in high-risk patients.Frailty prevalence was highest med-surg pts 60 to <70 years (56%); highest risk for time to first ICU transfer was in younger (50 to <60 years) and older (≥80 years) groups.

Beyond prediction: Off-target uses of artificial intelligence-based predictive analytics in a learning health system

Introduction

Artificial-intelligence (AI)-based predictive analytics provide new opportunities to leverage rich sources of continuous data to improve patient care through early warning of the risk of clinical deterioration and improved situational awareness.Part of the success of predictive analytic implementation relies on integration of the analytic within complex clinical workflows. Pharmaceutical interventions have off-target uses where a drug indication has not been formally studied for a different indication but has potential for clinical benefit. An analog has not been described in the context of AI-based predictive analytics, that is, when a predictive analytic has been trained on one outcome of interest but is used for additional applications in clinical practice.

Methods

In this manuscript we present three clinical vignettes describing off-target use of AI-based predictive analytics that evolved organically through real-world practice.

Results

Off-target uses included:real-time feedback about treatment effectiveness, indication of readiness to discharge, and indication of the acuity of a hospital unit.

Conclusion

Such practice fits well with the learning health system goals to continuously integrate data and experience to provide.

Impact on outcomes of measuring lactates prior to ICU in unselected heterogeneous critically ill patients: A propensity score analysis

BACKGROUND

Elevated blood lactate levels were reported as effective predictors of clinical outcome and mortality in ICU. However, there have been no studies simply comparing the timing of measuring lactates before vs. after ICU admission.

METHODS

A total of 19,226 patients with transfer time ≤ 24 hr were extracted from the Medical Information Mart for Intensive Care IV database (MIMIC-IV). After 1:1 propensity score matching, the patients were divided into two groups: measuring lactates within 3 hr before (BICU group, n = 4,755) and measuring lactate within 3 hr after ICU admission(AICU group, n = 4,755). The primary and secondary outcomes were hospital mortality, hospital 28-day mortality, ICU mortality, ICU length of stay (LOS), hospital LOS, and restricted mean survival time (RMST).

RESULTS

Hospital, hospital 28-day, and ICU mortality were significantly higher in AICU group (7.0% vs.9.8%, 6.7% vs. 9.4%, and 4.6% vs.6.7%, respectively, p<0.001 for all) Hospital LOS and ICU LOS were significantly longer in AICU group (8.4 days vs. 9.0 days and 3.0 days vs. 3.5 days, respectively, p<0.001 for both). After adjustment for predefined covariates, a significant association between the timing of measuring lactate and hospital mortality was observed in inverse probability treatment weight (IPTW) multivariate regression, doubly robust multivariate regression, and multivariate regression models (OR, 0.96 [95%CI, 0.95-0.97], OR 0.52 [95%CI, 0.46-0.60], OR 0.66 [95%CI, 0.56-0.78], respectively, p<0.001 for all), indicating the timing as a significant risk-adjusted factor for lower hospital mortality. The difference (BICU-AICU) of RMST at 28- days after ICU admission was 0.531 days (95%CI, 0.002-1.059, p<0.05). Placement of A-line and PA-catheter, administration of intravenous antibiotics, and bolus fluid infusion during the first 24-hr in ICU were significantly more frequent and faster in the BICU vs AICU group (67.6% vs. 51.3% and 126min vs.197min for A-line, 19.6% vs.13.2% and 182min vs. 274min for PA-catheter, 77.5% vs.67.6% and 109min vs.168min for antibiotics, and 57.6% vs.51.6% and 224min vs.278min for bolus fluid infusion, respectively, p<0.001 for all). Additionally, a significant indirect effect was observed in frequency (0.19879 [95% CI, 0.14061-0.25697] p<0.001) and time (0.07714 [95% CI, 0.22600-0.13168], p<0.01) of A-line replacement, frequency of placement of PA-catheter (0.05614 [95% CI, 0.04088-0.07140], p<0.001) and frequency of bolus fluid infusion (0.02193 [95%CI, 0.00303-0.04083], p<0.05).

CONCLUSIONS

Measuring lactates within 3 hr prior to ICU might be associated with lower hospital mortality in unselected heterogeneous critically ill patients with transfer time to ICU ≤ 24hr, presumably due to more frequent and faster therapeutic interventions.

Association of Time to Rapid Response Team Activation With Patient Outcomes Using a Range of Physiologic Deterioration Thresholds

Clinical deterioration of hospitalized patients is common and can lead to critical illness and death. Rapid response teams (RRTs) assess and treat high-risk patients with signs of clinical deterioration to prevent further worsening and subsequent adverse outcomes. Whether activation of the RRT early in the course of clinical deterioration impacts outcomes, however, remains unclear. We sought to characterize the relationship between increasing time to RRT activation after physiologic deterioration and short-term patient outcomes.

DESIGN

Retrospective multicenter cohort study.

SETTING

Three academic hospitals in Pennsylvania.

PATIENTS

We included the RRT activation of a hospitalization for non-ICU inpatients greater than or equal to 18 years old.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The primary exposure was time to RRT activation after physiologic deterioration. We selected four Cardiac Arrest Risk Triage (CART) score thresholds a priori from which to measure time to RRT activation (CART score ≥ 12, ≥ 16, ≥ 20, and ≥ 24). The primary outcome was 7-day mortality-death or discharge to hospice care within 7 days of RRT activation. For each CART threshold, we modeled the association of time to RRT activation duration with 7-day mortality using multivariable fractional polynomial regression. Increased time from clinical decompensation to RRT activation was associated with higher risk of 7-day mortality. This relationship was nonlinear, with odds of mortality increasing rapidly as time to RRT activation increased from 0 to 4 hours and then plateauing. This pattern was observed across several thresholds of physiologic derangement.

CONCLUSIONS

Increasing time to RRT activation was associated in a nonlinear fashion with increased 7-day mortality. This relationship appeared most marked when using a CART score greater than 20 threshold from which to measure time to RRT activation. We suggest that these empirical findings could be used to inform RRT delay definitions in further studies to determine the clinical impact of interventions focused on timely RRT activation.

Time-to-event modeling for hospital length of stay prediction for COVID-19 patients

Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.

Early Prediction of All-Cause Clinical Deterioration in General Wards Patients: Development and Validation of a Biomarker-Based Machine Learning Model Derived From Rapid Response Team Activations

OBJECTIVE

The aim of the study is to evaluate the performance of a biomarker-based machine learning (ML) model (not including vital signs) derived from reviewed rapid response team (RRT) activations in predicting all-cause deterioration in general wards patients.

DESIGN

This is a retrospective single-institution study. All consecutive adult patients' cases on noncritical wards identified by RRT calls occurring at least 24 hours after patient admission, between April 2018 and June 2020, were included. The cases were reviewed and labeled for clinical deterioration by a multidisciplinary expert consensus panel. A supervised learning approach was adopted based on a set of biomarkers and demographic data available in the patient's electronic medical record (EMR).

SETTING

The setting is a 250-bed tertiary university hospital with a basic EMR, with adult (>18 y) patients on general wards.

PATIENTS

The study analyzed the cases of 514 patients for which the RRT was activated. Rapid response teams were extracted from the hospital telephone log data. Two hundred eighteen clinical deterioration cases were identified in these patients after expert chart review and complemented by 146 "nonevent" cases to build the training and validation data set.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The best performance was achieved with the random forests algorithm, with a maximal area under the receiver operating curve of 0.90 and F1 score of 0.85 obtained at prediction time T0-6h, slightly decreasing but still acceptable (area under the receiver operating curve, >0.8; F1 score, >0.75) at T0-42h. The system outperformed most classical track-and-trigger systems both in terms of prediction performance and prediction horizon.

CONCLUSIONS

In hospitals with a basic EMR, a biomarker-based ML model could be used to predict clinical deterioration in general wards patients earlier than classical track-and-trigger systems, thus enabling appropriate clinical interventions for patient safety and improved outcomes.

The Impact of a Machine Learning Early Warning Score on Hospital Mortality: A Multicenter Clinical Intervention Trial

OBJECTIVES

To determine the impact of a machine learning early warning risk score, electronic Cardiac Arrest Risk Triage (eCART), on mortality for elevated-risk adult inpatients.

DESIGN

A pragmatic pre- and post-intervention study conducted over the same 10-month period in 2 consecutive years.

SETTING

Four-hospital community-academic health system.

PATIENTS

All adult patients admitted to a medical-surgical ward.

INTERVENTIONS

During the baseline period, clinicians were blinded to eCART scores. During the intervention period, scores were presented to providers. Scores greater than or equal to 95th percentile were designated high risk prompting a physician assessment for ICU admission. Scores between the 89th and 95th percentiles were designated intermediate risk, triggering a nurse-directed workflow that included measuring vital signs every 2 hours and contacting a physician to review the treatment plan.

MEASUREMENTS AND MAIN RESULTS

The primary outcome was all-cause inhospital mortality. Secondary measures included vital sign assessment within 2 hours, ICU transfer rate, and time to ICU transfer. A total of 60,261 patients were admitted during the study period, of which 6,681 (11.1%) met inclusion criteria (baseline period n = 3,191, intervention period n = 3,490). The intervention period was associated with a significant decrease in hospital mortality for the main cohort (8.8% vs 13.9%; p < 0.0001; adjusted odds ratio [OR], 0.60 [95% CI, 0.52-0.71]). A significant decrease in mortality was also seen for the average-risk cohort not subject to the intervention (0.49% vs 0.26%; p < 0.05; adjusted OR, 0.53 [95% CI, 0.41-0.74]). In subgroup analysis, the benefit was seen in both high- (17.9% vs 23.9%; p = 0.001) and intermediate-risk (2.0% vs 4.0 %; p = 0.005) patients. The intervention period was also associated with a significant increase in ICU transfers, decrease in time to ICU transfer, and increase in vital sign reassessment within 2 hours.

CONCLUSIONS

Implementation of a machine learning early warning score-driven protocol was associated with reduced inhospital mortality, likely driven by earlier and more frequent ICU transfer.

Is the Critical Care Resuscitation Unit Sustainable: A 5-Year Experience of a Beneficial and Novel Model

Background. The 6-bed critical care resuscitation unit (CCRU) is a unique and specialized intensive care unit (ICU) that streamlines the interhospital transfer (IHT—transfer between different hospitals) process for a wide range of patients with critical illness or time-sensitive disease. Previous studies showed the unit successfully increased the number of ICU admissions while reducing the time of transfer in the first year of its establishment. However, its sustainability is unknown. Methods. This was a descriptive retrospective analysis of adult, non-trauma patients who were transferred to an 800-bed quaternary medical center. Patients transferred to our medical center between January 1, 2014 and December 31, 2018 were eligible. We used interrupted time series (ITS) and descriptive analyses to describe the trend and compare the transfer process between patients who were transferred to the CCRU versus those transferred to other adult inpatient units. Results. From 2014 to 2018, 50,599 patients were transferred to our medical center; 31,582 (62%) were non-trauma adults. Compared with the year prior to the opening of the CCRU, ITS showed a significant increase in IHT after the establishment of the CCRU. The CCRU received a total of 7,788 (25%) IHTs during this period or approximately 20% of total transfers per year. Most transfers (41%) occurred via ground. Median and interquartile range [IQR] of transfer times to other ICUs (156 [65–1027] minutes) were longer than the CCRU (46 [22–139] minutes, $P<0.001$ ). For the CCRU, the most common accepting services were cardiac surgery (16%), neurosurgery (11%), and emergency general surgery (10%). Conclusions. The CCRU increases the overall number of transfers to our institution, improves patient access to specialty care while decreasing transfer time, and continues to be a sustainable model over time. Additional research is needed to determine if transferring patients to the CCRU would continue to improve patients’ outcomes and hospital revenue.

Validation of Dozee, a Ballistocardiography-based Device, for Contactless and Continuous Heart Rate and Respiratory Rate Measurement

Long-term acquisition of respiratory and heart signals is useful in a variety of applications, including sleep analysis, monitoring of respiratory and heart disorders, and so on. Ballistocardiography (BCG), a non-invasive technique that measures micro-body vibrations caused by cardiac contractions as well as motion caused by breathing, snoring, and body movements, would be ideal for long-term vital parameter acquisition. Turtle Shell Technologies Pvt. Ltd.'s Dozee device, which is based on BCG, is a contactless continuous vital parameters monitoring system. It is designed to measure Heart Rate (HR) and Respiratory Rate (RR) continuously and without contact in a hospital setting or at home. A validation study for HR and RR was conducted using Dozee by comparing it to the vitals obtained from the FDA-approved Patient Monitor. This was done in a sleep laboratory setting over 110 nights in 51 subjects to evaluate HR and over 20 nights in 17 subjects to evaluate RR at the National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore, India. Approximately 789 hours data for HR and approximately 112 hours data for RR was collected. Dozee was able to achieve a mean absolute error of 1.72 bpm for HR compared to the gold standard ECG. A mean absolute error of ∼1.24 breaths/min was obtained in determining RR compared to currently used methods. Dozee is ideal for long-term contactless monitoring of vital parameters due to its low mean absolute errors in measuring both HR and RR. Clinical Relevance- Continuous and long-term vitals monitoring is known to enable early screening of clinical deterioration, improve patient outcomes and reduce mortality. Current methods of continuous monitoring are overly complex, costly, and rely heavily on patient compliance. The proposed remote vitals monitoring solution based on BCG was found to be at par with gold standard methods of recording HR and RR. As a result, clinicians can use it to effectively monitor patients in both the hospital and at home.

Emergency Department Boarding of Mechanically Ventilated Patients

Background and objective The boarding of critically ill patients in the emergency department (ED) has been on the rise over the past few years. Emergency physicians now frequently encounter critically ill patients who require rapid resuscitation and stabilization and they provide extended care in the ED. This study aimed to evaluate the association between the boarding duration of mechanically ventilated patients in the ED and outcomes in such patients. Methods This was a retrospective study conducted during the period 2018-2019 at an academic institution; it included adult patients who were mechanically ventilated, requiring and awaiting admission to the ICU from the ED. Results We included a total of 388 out of 537 patients in the analysis. Patients were stratified into three groups as follows: 93 (24%) were admitted to the ICU within six hours; 126 (32.5%) were admitted to the ICU within 6-24 hours; and 169 (43.6%) were admitted to the ICU after 24 hours. Patients admitted to the ICU within six hours were significantly younger; the mean age of the patients was 55 ± 16.30 years in group 1, 61.96 ± 17.73 years in group 2, and 62.65 ± 16.62 years in group 3 (p=0.001). The ICU mortality in group 1 was lower than in other groups, and mortality increased with increasing boarding time [28 (30.1%), 51 (40.5%), 79 (46.7%), respectively, p=0.032]. Boarding time in the ED was associated with an increased risk of ICU mortality in group 3 compared with group 1 (0.1664 ± 0.063, p=0.009). The logistic regression analysis showed higher mortality rates in groups 2 [adjusted odds ratio: 3.29; 95% confidence interval (CI): 1.95-5.55, p<0.01] and 3 (adjusted odds ratio: 1.98; 95% CI: 1.17-3.35, p=0.01). Conclusion Based on our findings from this small-sample, single-center study, ED boarding of mechanically ventilated patients is associated with higher ICU mortality rates.

Delayed Admission to the Intensive Care Unit and Mortality of Critically Ill Adults: Systematic Review and Meta-analysis

Delayed admission of patients to the intensive care unit (ICU) is increasing worldwide and can be followed by adverse outcomes when critical care treatment is not provided timely. This systematic review and meta-analysis appraised and synthesized the published literature about the association between delayed ICU admission and mortality of adult patients. Articles published from inception up to August 2021 in English-language, peer-reviewed journals indexed in CINAHL, PubMed, Scopus, Cochrane Library, and Web of Science were searched by using key terms. Delayed ICU admission constituted the intervention, while mortality for any predefined time period was the outcome. Risk for bias was evaluated with the Newcastle-Ottawa Scale and additional criteria. Study findings were synthesized qualitatively, while the odds ratios (ORs) for mortality with 95% confidence intervals (CIs) were combined quantitatively. Thirty-four observational studies met inclusion criteria. Risk for bias was low in most studies. Unadjusted mortality was reported in 33 studies and was significantly higher in the delayed ICU admission group in 23 studies. Adjusted mortality was reported in 18 studies, and delayed ICU admission was independently associated with significantly higher mortality in 13 studies. Overall, pooled OR for mortality in case of delayed ICU admission was 1.61 (95% CI 1.44-1.81). Interstudy heterogeneity was high (I² = 66.96%). According to subgroup analysis, OR for mortality was remarkably higher in postoperative patients (OR, 2.44, 95% CI 1.49-4.01). These findings indicate that delayed ICU admission is significantly associated with mortality of critically ill adults and highlight the importance of providing timely critical care in non-ICU settings.

The PaCO2/FiO2 ratio as outcome predictor in SARS-COV-2 related pneumonia: a retrospective study

BACKGROUND AND AIM

Respiratory failure in SARS-CoV-2 patients is characterized by the presence of hypoxemia and hypocapnia without relevant dyspnea. To date, the use of respiratory parameters other than PaO2/FiO2 ratio to stratify the risk of worsening of these patients has not been sufficiently studied.  Aim of this work was to evaluate whether the ratio between partial pressure levels of carbon dioxide (PaCO2) and the fraction of inspired oxygen (FiO2) measured at emergency department (ED) admission is predictive of the clinical course of patients suffering from SARS-CoV-2 pneumonia.

METHODS

We retrospectively studied 236 patients with SARS-CoV-2 pneumonia evaluated at the ED of the Perugia Hospital. The end-points were: in-hospital mortality, need for invasive mechanical ventilation (IMV) and length of in-hospital stay (LOS). Clinical, blood gas and laboratory data were collected at ED admission.

RESULTS

Of the 236 patients 157 were male, the mean age was 64 ± 16. Thirtythree patients (14%) needed IMV, 49 died (21%). In the univariate analysis, the PaCO2/FiO2 ratio was inversely associated with the need for IMV (p <0.001), mortality (p <0.001) and LOS (p = 0.005). At the multivariate analysis the PaCO2/FiO2 ratio was found to be predictive of the need for IMV, independently from age, gender, number of comorbidities, neutrophils, lymphocytes, glomerular filtrate, d-dimer, LDH and CRP.

CONCLUSIONS

the PaCO2/FiO2 ratio is predictive of the risk of respiratory failure worsening in patients with SARS-CoV-2 pneumonia, independently from other several confounding factors.

Peripheral oxygen saturation to inspiratory oxygen fraction ratio-based identification of critically ill coronavirus disease patients for early therapeutic interventions

BACKGROUND

Early identification of critically ill coronavirus disease (COVID-19) patients in clinical settings is crucial in reducing the mortality rate. Therefore, this study aimed to determine whether the saturation of peripheral oxygen (SpO2) to fraction of inspiratory oxygen (FiO2) ratio (SF ratio) at admission is useful for the early identification of severe COVID-19.

METHODS

This single-center, retrospective, observational study conducted at the University Hospital, Kyoto, Japan, included 26 patients diagnosed with COVID-19 between January 24 and May 6, 2020. COVID-19 severity was classified into two groups based on the SF ratio: ≤ 235 (moderate to severe disease: low group) and  > 235 (normal to mild disease: high group). The characteristics, laboratory data, and outcomes of the patients were examined retrospectively and compared between the groups.

RESULTS

Of the 26 patients [median age 51.5 years, interquartile range 35.8-67.0], 6 were in the low group (23%) and 20 in the high group (77%). The low group had a higher respiratory rate than the high group (p < 0.05). Blood tests immediately after admission showed that the low group had significantly lower albumin (p < 0.01), and higher lactate dehydrogenase (p < 0.01), C-reactive protein (p < 0.01), and D-dimer (p < 0.01) levels than the high group. Moreover, all patients received antiviral agents; four received continuous renal replacement therapy and invasive positive pressure ventilation, one received extracorporeal membrane oxygenation, and two died in the low group.

CONCLUSION

SF ratio measurement at admission could assist clinicians in the early identification of severe COVID-19, which in turn can lead to early therapeutic interventions.

Performance in mortality prediction of SAPS 3 And MPM-III scores among adult patients admitted to the ICU of a private tertiary referral hospital in Tanzania: a retrospective cohort study

Background

Illness predictive scoring systems are significant and meaningful adjuncts of patient management in the Intensive Care Unit (ICU). They assist in predicting patient outcomes, improve clinical decision making and provide insight into the effectiveness of care and management of patients while optimizing the use of hospital resources. We evaluated mortality predictive performance of Simplified Acute Physiology Score (SAPS 3) and Mortality Probability Models (MPM₀-III) and compared their performance in predicting outcome as well as identifying disease pattern and factors associated with increased mortality.

Methods

This was a retrospective cohort study of adult patients admitted to the ICU of the Aga Khan Hospital, Dar- es- Salaam, Tanzania between August 2018 and April 2020. Demographics, clinical characteristics, outcomes, source of admission, primary admission category, length of stay and the support provided with the worst physiological data within the first hour of ICU admission were extracted. SAPS 3 and MPM₀-III scores were calculated using an online web-based calculator. The performance of each model was assessed by discrimination and calibration. Discrimination between survivors and non-survivors was assessed by the area under the receiver operator characteristic curve (ROC) and calibration was estimated using the Hosmer-Lemeshow goodness-of-fit test.

Results

A total of 331 patients were enrolled in the study with a median age of 58 years (IQR 43-71), most of whom were male (n = 208, 62.8%), of African origin (n = 178, 53.8%) and admitted from the emergency department (n = 306, 92.4%). In- hospital mortality of critically ill patients was 16.1%. Discrimination was very good for all models, the area under the receiver-operating characteristic (ROC) curve for SAPS 3 and MPM₀-III was 0.89 (95% CI [0.844-0.935]) and 0.90 (95% CI [0.864-0.944]) respectively. Calibration as calculated by Hosmer-Lemeshow goodness-of-fit test showed good calibration for SAPS 3 and MPM₀-III with Chi- square values of 4.61 and 5.08 respectively and P-Value > 0.05.

Conclusion

Both SAPS 3 and MPM₀-III performed well in predicting mortality and outcome in our cohort of patients admitted to the intensive care unit of a private tertiary hospital. The in-hospital mortality of critically ill patients was lower compared to studies done in other intensive care units in tertiary referral hospitals within Tanzania.

Nursing and precision predictive analytics monitoring in the acute and intensive care setting: An emerging role for responding to COVID-19 and beyond

As the global response to COVID-19 continues, nurses will be tasked with appropriately triaging patients, responding to events of clinical deterioration, and developing family-centered plans of care within a healthcare system exceeding capacity. Predictive analytics monitoring, an artificial intelligence (AI)-based tool that translates streaming clinical data into a real-time visual estimation of patient risks, allows for evolving acuity assessments and detection of clinical deterioration while the patient is in pre-symptomatic states. While nurses are on the frontline for the COVID-19 pandemic, the use of AI-based predictive analytics monitoring may help cognitively complex clinical decision-making tasks and pave a pathway for early detection of patients at risk for decompensation. We must develop strategies and techniques to study the impact of AI-based technologies on patient care outcomes and the clinical workflow. This paper outlines key concepts for the intersection of nursing and precision predictive analytics monitoring.

Demand for Intensive Care beds and patient classification according to the priority criterion

OBJECTIVE

to assess the demand for Intensive Care Unit beds as well as the classification of the patients for admission, according to the priority system.

METHOD

a retrospective and cross-sectional study, developed from January2014 to December2018 in two Intensive Care Units for adults of a university hospital. The sample consisted of the requests for vacancies according to the priority system(scale from 1 to 4, where 1 is the highest priority and 4 is no priority), registered in the institution's electronic system.

RESULTS

a total of 8,483 vacancies were requested, of which 4,389(51.7%) were from unitB. The highest percentage in unitA was of Priority2 patients(32.6%); and Priority1 was prevalent in unitB(45.4%). The median lead time between request and admission to unitA presented a lower value for priority1 patients(2h57) and a higher value for priority4 patients(11h24); in unitB, priority4 patients presented shorter time(5h54) and priority3 had longer time(11h54). 40.5% of the requests made to unitA and 48.5% of those made to unitB were fulfilled, with 50.7% and 48.5% of these patients being discharged from the units, respectively.

CONCLUSION

it is concluded that the demand for intensive care beds was greater than their availability. Most of the patients assisted were priorities1 and2, although a considerable percentage of those classified as priorities3 and4 is observed.

The Japanese Clinical Practice Guidelines for Management of Sepsis and Septic Shock 2020 (J-SSCG 2020)

The Japanese Clinical Practice Guidelines for Management of Sepsis and Septic Shock 2020 (J-SSCG 2020), a Japanese-specific set of clinical practice guidelines for sepsis and septic shock created as revised from J-SSCG 2016 jointly by the Japanese Society of Intensive Care Medicine and the Japanese Association for Acute Medicine, was first released in September 2020 and published in February 2021. An English-language version of these guidelines was created based on the contents of the original Japanese-language version. The purpose of this guideline is to assist medical staff in making appropriate decisions to improve the prognosis of patients undergoing treatment for sepsis and septic shock. We aimed to provide high-quality guidelines that are easy to use and understand for specialists, general clinicians, and multidisciplinary medical professionals. J-SSCG 2016 took up new subjects that were not present in SSCG 2016 (e.g., ICU-acquired weakness [ICU-AW], post-intensive care syndrome [PICS], and body temperature management). The J-SSCG 2020 covered a total of 22 areas with four additional new areas (patient- and family-centered care, sepsis treatment system, neuro-intensive treatment, and stress ulcers). A total of 118 important clinical issues (clinical questions, CQs) were extracted regardless of the presence or absence of evidence. These CQs also include those that have been given particular focus within Japan. This is a large-scale guideline covering multiple fields; thus, in addition to the 25 committee members, we had the participation and support of a total of 226 members who are professionals (physicians, nurses, physiotherapists, clinical engineers, and pharmacists) and medical workers with a history of sepsis or critical illness. The GRADE method was adopted for making recommendations, and the modified Delphi method was used to determine recommendations by voting from all committee members.As a result, 79 GRADE-based recommendations, 5 Good Practice Statements (GPS), 18 expert consensuses, 27 answers to background questions (BQs), and summaries of definitions and diagnosis of sepsis were created as responses to 118 CQs. We also incorporated visual information for each CQ according to the time course of treatment, and we will also distribute this as an app. The J-SSCG 2020 is expected to be widely used as a useful bedside guideline in the field of sepsis treatment both in Japan and overseas involving multiple disciplines.

Clinical Characteristics and In-Hospital Mortality of Cardiac Arrest Survivors in Brazil: A Large Retrospective Multicenter Cohort Study

Supplemental Digital Content is available in the text.

Data on cardiac arrest survivors from developing countries are scarce. This study investigated clinical characteristics associated with in-hospital mortality in resuscitated patients following cardiac arrest in Brazil.

The effectiveness of continuous respiratory rate monitoring in predicting hypoxic and pyrexic events: a retrospective cohort study

Respiratory rate (RR) is routinely used to monitor patients with infectious, cardiac and respiratory diseases and is a component of early warning scores used to predict patient deterioration. However, it is often measured visually with considerable bias and inaccuracy.Objectives. Firstly, to compare distribution and accuracy of electronically measured RR (EMRR) and visually measured RR (VMRR). Secondly, to determine whether, and how far in advance, continuous electronic RR monitoring can predict oncoming hypoxic and pyrexic episodes in infectious respiratory disease.Approach.A retrospective cohort study analysing the difference between EMRR and VMRR was conducted using patient data from a large tertiary hospital. Cox proportional hazards models were used to determine whether continuous, EMRR measurements could predict oncoming hypoxic (SpO₂ < 92%) and pyrexic (temperature >38 °C) episodes.Main results.Data were gathered from 34 COVID-19 patients, from which a total of 3445 observations of VMRR (independent of Hawthorne effect), peripheral oxygen saturation and temperature and 729 117 observations of EMRR were collected. VMRR had peaks in distribution at 18 and 20 breaths per minute. 70.9% of patients would have had a change of treatment during their admission based on the UK's National Early Warning System if EMRR was used in place of VMRR. An elevated EMRR was predictive of hypoxic (hazard ratio: 1.8 (1.05-3.07)) and pyrexic (hazard ratio: 9.7 (3.8-25)) episodes over the following 12 h.Significance.Continuous EMRR values are systematically different to VMRR values, and results suggest it is a better indicator of true RR as it has lower kurtosis, higher variance, a lack of peaks at expected values (18 and 20) and it measures a physiological component of breathing directly (abdominal movement). Results suggest EMRR is a strong marker of oncoming hypoxia and is highly predictive of oncoming pyrexic events in the following 12 h. In many diseases, this could provide an early window to escalate care prior to deterioration, potentially preventing morbidity and mortality.