A critical analysis of unplanned ICU transfer within 48 hours from ED admission as a quality measure

Patient outcomes after interhospital transfer: the impact of early intensive care unit upgrade

BACKGROUND

There is little known about the prevalence and outcomes of medical patients requiring early intensive care unit upgrade (EIU) following interhospital transfer, and previous studies of EIU focus on patients admitted through the emergency room. We aimed to examine the characteristics and risk factors for poor outcome among medical patients undergoing EIU after interhospital transfer.

MATERIALS AND METHODS

The publicly available Medical Information Mart for Intensive Care (MIMIC) IV database (2008-2019) was queried to identify non-surgical patients undergoing interhospital transfer. Patients who subsequently underwent EIU, defined as ICU admission within 24 hours of arrival after interhospital transfer, were compared to those who did not experience EIU for differences in mortality and length-of-stay (LOS.) We used multivariate logistic regression to identify risk factors for hospital death in this population and negative binomial regression to estimate the impact of EIU on hospital LOS.

RESULTS

We identified 5,619 patients who underwent interhospital transfer, of which 339 (6.0%) experienced EIU and 5280 (94.0%) did not. Patients undergoing EIU after interhospital transfer were significantly older (median age 69 vs. 64 years; p = 0.001,) but there was no difference in sex. After risk-adjustment, we found an association between EIU and a higher risk of mortality (aOR 6.9, 95%CI 5.24-9.08). Increased comorbidity burden as measured by Charlson Comorbidity Index (CCI) was linked to higher odds of death (aOR 1.26, 95% CI 1.22-1.31,) as was nonwhite race (aOR 1.69, 95% CI 1.34-2.14). EIU was associated with a longer hospital LOS (IRR 1.40, 95%CI 1.28-1.54).

CONCLUSION

EIU after interhospital transfer is associated with higher mortality and longer LOS. Further study will help identify process features of transfer and patient characteristics contributing to poor outcome after arrival from an outlying facility and guide efforts to mitigate risk and provide equitable care across the transfer continuum.

Machine Learning Models for the Early Real-Time Prediction of Deterioration in Intensive Care Units-A Novel Approach to the Early Identification of High-Risk Patients

Background Predictive machine learning models have made use of a variety of scoring systems to identify clinical deterioration in ICU patients. However, most of these scores include variables that are dependent on medical staff examining the patient. We present the development of a real-time prediction model using clinical variables that are digital and automatically generated for the early detection of patients at risk of deterioration. Methods Routine monitoring data were used in this analysis. ICU patients with at least 24 h of vital sign recordings were included. Deterioration was defined as qSOFA ≥ 2. Model development and validation were performed internally by splitting the cohort into training and test datasets and validating the results on the test dataset. Five different models were trained, tested, and compared against each other. The models were an artificial neural network (ANN), a random forest (RF), a support vector machine (SVM), a linear discriminant analysis (LDA), and a logistic regression (LR). Results In total, 7156 ICU patients were screened for inclusion in the study, which resulted in models trained from a total of 28,348 longitudinal measurements. The artificial neural network showed a superior predictive performance for deterioration, with an area under the curve of 0.81 over 0.78 (RF), 0.78 (SVM), 0.77 (LDA), and 0.76 (LR), by using only four vital parameters. The sensitivity was higher than the specificity for the artificial neural network. Conclusions The artificial neural network, only using four automatically recorded vital signs, was best able to predict deterioration, 10 h before documentation in clinical records. This real-time prediction model has the potential to flag at-risk patients to the healthcare providers treating them, for closer monitoring and further investigation.

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.

Risk factors predicting the need for intensive care unit admission within forty-eight hours of emergency department presentation: A case-control study

BACKGROUND

Patients admitted from the emergency department to the wards, who progress to a critically unwell state, may require expeditious admission to the intensive care unit. It can be argued that earlier recognition of such patients, to facilitate prompt transfer to intensive care, could be linked to more favourable clinical outcomes. Nevertheless, this can be clinically challenging, and there are currently no established evidence-based methods for predicting the need for intensive care in the future.

OBJECTIVES

We aimed to analyse the emergency department data to describe the characteristics of patients who required an intensive care admission within 48 h of presentation. Secondly, we planned to test the feasibility of using this data to identify the associated risk factors for developing a predictive model.

METHODS

We designed a retrospective case-control study. Cases were patients admitted to intensive care within 48 h of their emergency department presentation. Controls were patients who did not need an intensive care admission. Groups were matched based on age, gender, admission calendar month, and diagnosis. To identify the associated variables, we used a conditional logistic regression model.

RESULTS

Compared to controls, cases were more likely to be obese, and smokers and had a higher prevalence of cardiovascular (39 [35.1%] vs 20 [18%], p = 0.004) and respiratory diagnoses (45 [40.5%] vs 25 [22.5%], p = 0.004). They received more medical emergency team reviews (53 [47.8%] vs 24 [21.6%], p < 0.001), and more patients had an acute resuscitation plan (31 [27.9%] vs 15 [13.5%], p = 0.008). The predictive model showed that having acute resuscitation plans, cardiovascular and respiratory diagnoses, and receiving medical emergency team reviews were strongly associated with having an intensive care admission within 48 h of presentation.

CONCLUSIONS

Our study used emergency department data to provide a detailed description of patients who had an intensive care unit admission within 48 h of their presentation. It demonstrated the feasibility of using such data to identify the associated risk factors to develop a predictive model.

Association between emergency department disposition and mortality in patients with COVID-19 acute respiratory distress syndrome

Objectives

Patients hospitalized for COVID-19 frequently develop hypoxemia and acute respiratory distress syndrome (ARDS) after admission. In non-COVID-19 ARDS studies, admission to hospital wards with subsequent transfer to intensive care unit (ICU) is associated with worse outcomes. We hypothesized that initial admission to the ward may affect outcomes in patient with COVID-19 ARDS.

Methods

This was a retrospective study of consecutive adults admitted for COVID-19 ARDS between March 2020 and March 2021 at Stanford Health Care. Mortality scores at hospital admission (Coronavirus Clinical Characterization Consortium Mortality Score [4C score]) and ICU admission (Simplified Acute Physiology Score III [SAPS-III]) were calculated, as well as ROX index for patients on high flow nasal oxygen. Patients were classified by emergency department (ED) disposition (ward-first vs. ICU-direct), and 28- and 60-day mortality and highest level of respiratory support within 1 day of ICU admission were compared. A second cohort (April 2021‒July 2022, n = 129) was phenotyped to validate mortality outcome.

Results

A total of 157 patients were included, 48% of whom were first admitted to the ward (n = 75). Ward-first patients had more comorbidities, including lung disease. Ward-first patients had lower 4C and similar SAPS-III score, yet increased mortality at 28 days (32% vs. 17%, hazard ratio [HR] 2.0, 95% confidence interval [95% CI] 1.0‒3.7, p = 0.039) and 60 days (39% vs. 23%, HR 1.83, 95% CI 1.04‒3.22, p = 0.037) compared to ICU-direct patients. More ward-first patients escalated to mechanical ventilation on day 1 of ICU admission (36% vs. 14%, p = 0.002) despite similar ROX index. Ward-first patients who upgraded to ICU within 48 h of ED presentation had the highest mortality. Mortality findings were replicated in a sensitivity analysis.

Conclusion

Despite similar baseline risk scores, ward-first patients with COVID-19 ARDS had increased mortality and escalation to mechanical ventilation compared to ICU-direct patients. Ward-first patients requiring ICU upgrade within 48 h were at highest risk, highlighting a need for improved identification of this group at ED admission.

Temporal Clustering of Critical Illness Events on Medical Wards

Importance

Recognizing and preventing patient deterioration is important for hospital safety.

Objective

To investigate whether critical illness events (in-hospital death or intensive care unit [ICU] transfer) are associated with greater risk of subsequent critical illness events for other patients on the same medical ward.

Design, Setting, and Participants

Retrospective cohort study in 5 hospitals in Toronto, Canada, including 118 529 hospitalizations. Patients were admitted to general internal medicine wards between April 1, 2010, and October 31, 2017. Data were analyzed between January 1, 2020, and April 10, 2023.

Exposures

Critical illness events (in-hospital death or ICU transfer).

Main Outcomes and Measures

The primary outcome was the composite of in-hospital death or ICU transfer. The association between critical illness events on the same ward across 6-hour intervals was studied using discrete-time survival analysis, adjusting for patient and situational factors. The association between critical illness events on different comparable wards in the same hospital was measured as a negative control.

Results

The cohort included 118 529 hospitalizations (median age, 72 years [IQR, 56-83 years]; 50.7% male). Death or ICU transfer occurred in 8785 hospitalizations (7.4%). Patients were more likely to experience the primary outcome after exposure to 1 prior event (adjusted odds ratio [AOR], 1.39; 95% CI, 1.30-1.48) and more than 1 prior event (AOR, 1.49; 95% CI, 1.33-1.68) in the prior 6-hour interval compared with no exposure. The exposure was associated with increased odds of subsequent ICU transfer (1 event: AOR, 1.67; 95% CI, 1.54-1.81; >1 event: AOR, 2.05; 95% CI, 1.79-2.36) but not death alone (1 event: AOR, 1.08; 95% CI, 0.97-1.19; >1 event: AOR, 0.88; 95% CI, 0.71-1.09). There was no significant association between critical illness events on different wards within the same hospital.

Conclusions and Relevance

Findings of this cohort study suggest that patients are more likely to be transferred to the ICU in the hours after another patient's critical illness event on the same ward. This phenomenon could have several explanations, including increased recognition of critical illness and preemptive ICU transfers, resource diversion to the first event, or fluctuations in ward or ICU capacity. Patient safety may be improved by better understanding the clustering of ICU transfers on medical wards.

Predicting Unplanned Intensive Care Unit Admission for Trauma Patients: The CRASH Score

INTRODUCTION

Unplanned transfer of trauma patients to the intensive care unit (ICU) carries an associated increase in mortality, hospital length of stay, and cost. Trauma teams need to determine which patients necessitate ICU admission on presentation rather than waiting to intervene on deteriorating patients. This study sought to develop a novel Clinical Risk of Acute ICU Status during Hospitalization (CRASH) score to predict the risk of unplanned ICU admission.

METHODS

The 2017 Trauma Quality Improvement Program database was queried for patients admitted to nonICU locations. The group was randomly divided into two equal sets (derivation and validation). Multiple logistic regression models were created to determine the risk of unplanned ICU admission using patient demographics, comorbidities, and injuries. The weighted average and relative impact of each independent predictor were used to derive a CRASH score. The score was validated using area under the curve.

RESULTS

A total of 624,786 trauma patients were admitted to nonICU locations. From 312,393 patients in the derivation-set, 3769 (1.2%) had an unplanned ICU admission. A total of 24 independent predictors of unplanned ICU admission were identified and the CRASH score was derived with scores ranging from 0 to 32. The unplanned ICU admission rate increased steadily from 0.1% to 3.9% then 12.9% at scores of 0, 6, and 14, respectively. The area under the curve for was 0.78.

CONCLUSIONS

The CRASH score is a novel and validated tool to predict unplanned ICU admission for trauma patients. This tool may help providers admit patients to the appropriate level of care or identify patients at-risk for decompensation.

Characteristics of Children Who Deteriorate After Transport and Associated Preadmission Factors

OBJECTIVE

The incidence of deterioration and associated characteristics are largely unknown for children transported for admission from referring emergency departments (EDs) to general inpatient units. This study describes this population and identifies associated preadmission characteristics.

METHODS

This single-center cohort study included children ≤ 18 years old transferred from an ED and directly admitted to general inpatient units from 2016 to 2019. Deterioration was defined as 1 or more of the following occurring within 24 hours of admission: rapid response team activation, transfer to the intensive care unit (ICU), or cardiac or respiratory arrest. ICU transfer was the secondary outcome. Logistic regression was performed.

RESULTS

One thousand nine hundred eighty-eight patients were included; the median age was 4.2 years, 53.9% were male, and 44.1% had respiratory diagnoses. Deterioration occurred in 135 (6.8%) children overall and in 10.1% of children with respiratory complaints. Deterioration was associated with ≥ 2 complex chronic conditions (adjusted odds ratio [aOR] = 2.09; 95% confidence interval [CI], 1.04-4.19) and a longer stabilization time (per 10 minutes) (aOR = 1.17; 95% CI, 1.01-1.36). ICU transfer was associated with ≥ 2 complex chronic conditions (aOR = 2.33; 95% CI, 1.13-4.80), supplemental oxygen via nasal cannula (aOR = 2.13; 95% CI, 1.18-3.85), and nebulizer treatment (aOR = 2.77; 95% CI, 1.21-6.35).

CONCLUSION

Deterioration was experienced by 7% of children admitted to a general unit, with the majority having respiratory complaints. Transport teams should consider the potential for increased risk of deterioration among children with respiratory disease, multiple complex chronic conditions, and a nasal cannula or nebulizer therapy. The clinical significance of marginally longer stabilization times is unclear and warrants further investigation.

Leveraging EHR Data to Evaluate the Association of Late Recognition of Deterioration With Outcomes

OBJECTIVES

Emergency transfers (ETs), deterioration events with late recognition requiring ICU interventions within 1 hour of transfer, are associated with adverse outcomes. We leveraged electronic health record (EHR) data to assess the association between ETs and outcomes. We also evaluated the association between intervention timing (urgency) and outcomes.

METHODS

We conducted a propensity-score-matched study of hospitalized children requiring ICU transfer between 2015 and 2019 at a single institution. The primary exposure was ET, automatically classified using Epic Clarity Data stored in our enterprise data warehouse endotracheal tube in lines/drains/airway flowsheet, vasopressor in medication administration record, and/or ≥60 ml/kg intravenous fluids in intake/output flowsheets recorded within 1 hour of transfer. Urgent intervention was defined as interventions within 12 hours of transfer.

RESULTS

Of 2037 index transfers, 129 (6.3%) met ET criteria. In the propensity-score-matched cohort (127 ET, 374 matched controls), ET was associated with higher in-hospital mortality (13% vs 6.1%; odds ratio, 2.47; 95% confidence interval [95% CI], 1.24-4.9, P = .01), longer ICU length of stay (subdistribution hazard ratio of ICU discharge 0.74; 95% CI, 0.61-0.91, P < .01), and longer posttransfer length of stay (SHR of hospital discharge 0.71; 95% CI, 0.56-0.90, P < .01). Increased intervention urgency was associated with increased mortality risk: 4.1% no intervention, 6.4% urgent intervention, and 10% emergent intervention.

CONCLUSIONS

An EHR measure of deterioration with late recognition is associated with increased mortality and length of stay. Mortality risk increased with intervention urgency. Leveraging EHR automation facilitates generalizability, multicenter collaboratives, and metric consistency.

Mortality of Patients Requiring Escalation to Intensive Care within 24 Hours of Admission in a Mixed Medical-Surgical Population

OBJECTIVE

Delayed intensive care unit (ICU) admissions are associated with increased mortality. We present a retrospective study looking at whether indirect admissions to the ICU within 24 hours of hospital admission were associated with increased mortality.

DESIGN

Retrospective cohort study SETTING: Mixed medical-surgical ICU at a large tertiary United States Veterans Affairs (VA) Hospital System POPULATION: The patients were a mix of medical and surgical patients. Patients included both those directly admitted from the operating room as well as those escalated to the ICU after initial admission to the ward (indirect admission).

METHODS

All admissions to a medical-surgical ICU from 2008 to 2013 were included in the study. The database was queried for time and location where the admission originated. Separate lists were created for patients with severe sepsis, patients who transferred to the ICU within the first 24 hours, and patients who had rapid response or code team activations. Analysis was applied to the whole group and to medical and surgical subpopulations.

RESULTS

A total of 3,862 ICU admissions were studied. Univariate analysis indicated an impact of delayed admission on whole group and surgical patients; however, multivariate analysis indicated a significant effect of delayed admission on 1-year surgical mortality. Multivariate analysis also showed a consistent effect of age, ICU length of stay, and cardiac arrest on mortality of both medical and surgical ICU patients.

CONCLUSION

In a large retrospective study, surgical patients had increased 1-year mortality if they required escalation to the ICU within 24 hours of hospital admission. This result was not replicated in medical patients, possibly related to a burden of illness that could not be altered by earlier care.

Quality Assurance in the Emergency Department

Quality assurance (QA) of care in the emergency department encompasses activities ensuring that the care provided meets applicable standards. Health care delivery is complex and many factors affect quality of care. Thus, quantification of health care quality is challenging, especially with regard to attribution of outcomes to various factors contributing to such care. A critical component of the process of QA is determination of quality health care and the concept of (unjustified) deviation from the reference applicable standard of care.

Early Detection of In-Patient Deterioration: One Prediction Model Does Not Fit All

Supplemental Digital Content is available in the text.

Objectives:

Early detection of subacute potentially catastrophic illnesses using available data is a clinical imperative, and scores that report risk of imminent events in real time abound. Patients deteriorate for a variety of reasons, and it is unlikely that a single predictor such as an abnormal National Early Warning Score will detect all of them equally well. The objective of this study was to test the idea that the diversity of reasons for clinical deterioration leading to ICU transfer mandates multiple targeted predictive models.

Design:

Individual chart review to determine the clinical reason for ICU transfer; determination of relative risks of individual vital signs, laboratory tests and cardiorespiratory monitoring measures for prediction of each clinical reason for ICU transfer; and logistic regression modeling for the outcome of ICU transfer for a specific clinical reason.

Setting:

Cardiac medical-surgical ward; tertiary care academic hospital.

Patients:

Eight-thousand one-hundred eleven adult patients, 457 of whom were transferred to an ICU for clinical deterioration.

Interventions:

None.

Measurements and Main Results:

We calculated the contributing relative risks of individual vital signs, laboratory tests and cardiorespiratory monitoring measures for prediction of each clinical reason for ICU transfer, and used logistic regression modeling to calculate receiver operating characteristic areas and relative risks for the outcome of ICU transfer for a specific clinical reason. The reasons for clinical deterioration leading to ICU transfer were varied, as were their predictors. For example, the three most common reasons—respiratory instability, infection and suspected sepsis, and heart failure requiring escalated therapy—had distinct signatures of illness. Statistical models trained to target-specific reasons for ICU transfer performed better than one model targeting combined events.

Conclusions:

A single predictive model for clinical deterioration does not perform as well as having multiple models trained for the individual specific clinical events leading to ICU transfer.

Mixed-integer optimization approach to learning association rules for unplanned ICU transfer

After admission to emergency department (ED), patients with critical illnesses are transferred to intensive care unit (ICU) due to unexpected clinical deterioration occurrence. Identifying such unplanned ICU transfers is urgently needed for medical physicians to achieve two-fold goals: improving critical care quality and preventing mortality. A priority task is to understand the crucial rationale behind diagnosis results of individual patients during stay in ED, which helps prepare for an early transfer to ICU. Most existing prediction studies were based on univariate analysis or multiple logistic regression to provide one-size-fit-all results. However, patient condition varying from case to case may not be accurately examined by such a simplistic judgment. In this study, we present a new decision tool using a mathematical optimization approach aiming to automatically discover rules associating diagnostic features with high-risk outcome (i.e., unplanned transfers) in different deterioration scenarios. We consider four mutually exclusive patient subgroups based on the principal reasons of ED visits: infections, cardiovascular/respiratory diseases, gastrointestinal diseases, and neurological/other diseases at a suburban teaching hospital. The analysis results demonstrate significant rules associated with unplanned transfer outcome for each subgroups and also show comparable prediction accuracy (>70%) compared to state-of-the-art machine learning methods while providing easy-to-interpret symptom-outcome information.

Predictive factors for secondary intensive care unit admission within 48 hours after hospitalization in a medical ward from the emergency room

BACKGROUND

Unplanned transfer to an ICU within 48 hours of admission from the emergency department (ED) can be considered an adverse event. Screening at risk for such an event is a challenge for ED staff. Our purpose was to identify the clinical and biological variables which may be identified in the ED setting and can predict short-term unplanned secondary transfer to the intensive care setting.

METHODS

This was a three-year retrospective case controlled monocentric study. The cases were patients transferred to a medical ICU within 48 hours of admission to the general wards from the ED. Each case was matched to two controls (patients not transferred to the ICU) based on age, gender, year of admission, and hospital unit. A conditional logistic regression was performed.

RESULTS

Three hundred nineteen patients, including 107 cases and 212 controls, were studied. Community-acquired pneumonia (CAP) was the most frequent diagnosis (23% of cases) followed by sepsis (16%). We identified six predictive factors of an unplanned short-term transfer to the ICU. Former smoking status, fever between 38°C and 40°C, dyspnea as the chief complaint in the ED, a lower MEDS score, an elevated acute physiology age chronic health evaluation score, and the ordering of an arterial blood gas each correlate with secondary transfer to an intensive care setting.

CONCLUSION

We report a higher risk of short-term unscheduled ICU transfer in patients meeting these criteria. These patients should be closely monitored and frequently re-evaluated before being transferred to a general ward.

Early in-hospital clinical deterioration is not predicted by severity of illness, functional status, or comorbidity

BACKGROUND

Prior studies concentrated on unplanned intensive care unit (ICU) transfer to gauge deterioration occurring shortly following hospital admission. However, examining only ICU transfers is not ideal since patients could stabilize with treatment, refuse ICU admission, or not require ICU evaluation. To further explore etiologies of early clinical deterioration, we used rapid response team (RRT) activation within 48 hours of admission as an index of early clinical worsening.

METHODS

A retrospective analysis of prospectively gathered admissions from the emergency department in an academic medical center was done. Data were reviewed independently by two physicians. We assessed severity of illness, functional status, comorbidity, the frequency of ICU and palliative care consultations, and changes in advance health care directives.

RESULTS

Of 655 rapid responses (RRs) within the study period, 24.6% occurred within 48 hours of admission. Disease trajectory was the most frequent perceived reason for RRs (55.6% and 58.9%, reviewer 1 and 2, respectively) followed by medical error (15.6% and 15.2%). Acute physiology and chronic health evaluation II (APACHE-II) and modified early warning scores (MEWS) were higher at the time of RR compared to admission (p<0.0001). However, admission APACHE-II, MEWS, functional status, and comorbidity scores did not predict early RRs. One third of RRs resulted in ICU consultation and 95% were accepted. Palliative care consults were requested for 15%, the majority (65%) after RR and all resulting in advance directive change.

CONCLUSION

Disease trajectory accounted for most clinical deterioration and medical error contributed to 15%. Our data suggest that it is difficult to predict early clinical deterioration as none of the measured parameters were associated with RRT activation.