Dataset Information

Using machine learning tools to predict outcomes for emergency department intensive care unit patients.

ABSTRACT: The number of critically ill patients has increased globally along with the rise in emergency visits. Mortality prediction for critical patients is vital for emergency care, which affects the distribution of emergency resources. Traditional scoring systems are designed for all emergency patients using a classic mathematical method, but risk factors in critically ill patients have complex interactions, so traditional scoring cannot as readily apply to them. As an accurate model for predicting the mortality of emergency department critically ill patients is lacking, this study's objective was to develop a scoring system using machine learning optimized for the unique case of critical patients in emergency departments. We conducted a retrospective cohort study in a tertiary medical center in Beijing, China. Patients over 16 years old were included if they were alive when they entered the emergency department intensive care unit system from February 2015 and December 2015. Mortality up to 7 days after admission into the emergency department was considered as the primary outcome, and 1624 cases were included to derive the models. Prospective factors included previous diseases, physiologic parameters, and laboratory results. Several machine learning tools were built for 7-day mortality using these factors, for which their predictive accuracy (sensitivity and specificity) was evaluated by area under the curve (AUC). The AUCs were 0.794, 0.840, 0.849 and 0.822 respectively, for the SVM, GBDT, XGBoost and logistic regression model. In comparison with the SAPS 3 model (AUC?=?0.826), the discriminatory capability of the newer machine learning methods, XGBoost in particular, is demonstrated to be more reliable for predicting outcomes for emergency department intensive care unit patients.

SUBMITTER: Zhai Q

PROVIDER: S-EPMC7708467 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Using machine learning tools to predict outcomes for emergency department intensive care unit patients.

Zhai Qiangrong Q Lin Zi Z Ge Hongxia H Liang Yang Y Li Nan N Ma Qingbian Q Ye Chuyang C

Scientific reports 20201201 1

The number of critically ill patients has increased globally along with the rise in emergency visits. Mortality prediction for critical patients is vital for emergency care, which affects the distribution of emergency resources. Traditional scoring systems are designed for all emergency patients using a classic mathematical method, but risk factors in critically ill patients have complex interactions, so traditional scoring cannot as readily apply to them. As an accurate model for predicting the ...[more]

PMID: 33262471

Similar Datasets

Project description:Rationale: A small but growing number of hospitals are experimenting with emergency department-embedded critical care units (CCUs) in an effort to improve the quality of care for critically ill patients with sepsis and acute respiratory failure (ARF).Objectives: To evaluate the potential impact of an emergency department-embedded CCU at the Hospital of the University of Pennsylvania among patients with sepsis and ARF admitted from the emergency department to a medical ward or intensive care unit (ICU) from January 2016 to December 2017.Methods: The exposure was eligibility for admission to the emergency department-embedded CCU, which was defined as meeting a clinical definition for sepsis or ARF and admission to the emergency department during the intervention period on a weekday. The primary outcome was hospital length of stay (LOS); secondary outcomes included total emergency department plus ICU LOS, hospital survival, direct admission to the ICU, and unplanned ICU admission. Primary interrupted time series analyses were performed using ordinary least squares regression comparing monthly means. Secondary retrospective cohort and before-after analyses used multivariable Cox proportional hazard and logistic regression.Results: In the baseline and intervention periods, 3,897 patients met the inclusion criteria for sepsis and 1,865 patients met the criteria for ARF. Among patients admitted with sepsis, opening of the emergency department-embedded CCU was not associated with hospital LOS (β = -1.82 d; 95% confidence interval [CI], -4.50 to 0.87; P = 0.17 for the first month after emergency department-embedded CCU opening compared with baseline; β = -0.26 d; 95% CI, -0.58 to 0.06; P = 0.10 for subsequent months). Among patients admitted with ARF, the emergency department-embedded CCU was not associated with a significant change in hospital LOS for the first month after emergency department-embedded CCU opening (β = -3.25 d; 95% CI, -7.86 to 1.36; P = 0.15) but was associated with a 0.64 d/mo shorter hospital LOS for subsequent months (β = -0.64 d; 95% CI, -1.12 to -0.17; P = 0.01). This result persisted among higher acuity patients requiring ventilatory support but was not supported by alternative analytic approaches. Among patients admitted with sepsis who did not require mechanical ventilation or vasopressors in the emergency department, the emergency department-embedded CCU was associated with an initial 9.9% reduction in direct ICU admissions in the first month (β = -0.099; 95% CI, -0.153 to -0.044; P = 0.002), followed by a 1.1% per month increase back toward baseline in subsequent months (β = 0.011; 95% CI, 0.003-0.019; P = 0.009). This relationship was supported by alternative analytic approaches and was not seen in ARF. No associations with emergency department plus ICU LOS, hospital survival, or unplanned ICU admission were observed among patients with sepsis or ARF.Conclusions: The emergency department-embedded CCU was not associated with clinical outcomes among patients admitted with sepsis or ARF. Among less sick patients with sepsis, the emergency department-embedded CCU was initially associated with reduced rates of direct ICU admission from the emergency department. Additional research is necessary to further evaluate the impact and utility of the emergency department-embedded CCU model.

Project description:ImportanceIncreased patient acuity, decreased intensive care unit (ICU) bed availability, and a shortage of intensivist physicians have led to strained ICU capacity. The resulting increase in emergency department (ED) boarding time for patients requiring ICU-level care has been associated with worse outcomes.ObjectiveTo determine the association of a novel ED-based ICU, the Emergency Critical Care Center (EC3), with 30-day mortality and inpatient ICU admission.Design, setting, and participantsThis retrospective cohort study used electronic health records of all ED visits between September 1, 2012, and July 31, 2017, with a documented clinician encounter at a large academic medical center in the United States with approximately 75 000 adult ED visits per year. The pre-EC3 cohort included ED patients from September 2, 2012, to February 15, 2015, when the EC3 opened, and the post-EC3 cohort included ED patients from February 16, 2015, to July 31, 2017. Data analyses were conducted from March 2, 2018, to May 28, 2019.ExposuresImplementation of EC3, an ED-based ICU designed to provide rapid initiation of ICU-level care in the ED setting and seamless transition to inpatient ICUs.Main outcomes and measuresThe main outcomes were 30-day mortality among ED patients and rate of ED to ICU admission.ResultsA total of 349 310 visits from a consecutive sample of ED patients (mean [SD] age, 48.5 [19.7] years; 189 709 [54.3%] women) were examined; the pre-EC3 cohort included 168 877 visits and the post-EC3 cohort included 180 433 visits. Implementation of EC3 was associated with a statistically significant reduction in risk-adjusted 30-day mortality among all ED patients (pre-EC3, 2.13%; post-EC3, 1.83%; adjusted odds ratio, 0.85; 95% CI, 0.80-0.90; number needed to treat, 333 patient encounters; 95% CI, 256-476). The risk-adjusted rate of ED admission to ICU decreased with implementation of EC3 (pre-EC3, 3.2%; post-EC3, 2.7%; adjusted odds ratio, 0.80; 95% CI, 0.76-0.83; number needed to treat, 179 patient encounters; 95% CI, 149-217).Conclusions and relevanceImplementation of a novel ED-based ICU was associated with improved 30-day survival and reduced inpatient ICU admission. Additional research is warranted to further explore the value of this novel care delivery model in various health care systems.

Project description:BackgroundEarly prediction model of hemodynamic instability has the potential to improve the critical care, whereas limited external validation on the generalizability. We aimed to independently validate the Hemodynamic Stability Index (HSI), a multi-parameter machine learning model, in predicting hemodynamic instability in Asian patients.MethodHemodynamic instability was marked by using inotropic, vasopressor, significant fluid therapy, and/or blood transfusions. This retrospective study included among 15,967 ICU patients who aged 20 years or older (not included 20 years) and stayed in ICU for more than 6 h admitted to Taipei Veteran General Hospital (TPEVGH) between January 1, 2010, and March 31, 2020, of whom hemodynamic instability occurred in 3053 patients (prevalence = 19%). These patients in unstable group received at least one intervention during their ICU stays, and the HSI score of both stable and unstable group was calculated in every hour before intervention. The model performance was assessed using the area under the receiver operating characteristic curve (AUROC) and was compared to single indicators like systolic blood pressure (SBP) and shock index. The hemodynamic instability alarm was set by selecting optimal threshold with high sensitivity, acceptable specificity, and lead time before intervention was calculated to indicate when patients were firstly identified as high risk of hemodynamic instability.ResultsThe AUROC of HSI was 0.76 (95% CI, 0.75-0.77), which performed significantly better than shock Index (0.7; 95% CI, 0.69-0.71) and SBP (0.69; 95% CI, 0.68-0.70). By selecting 0.7 as a threshold, HSI predicted 72% of all 3053 patients who received hemodynamic interventions with 67% in specificity. Time-varying results also showed that HSI score significantly outperformed single indicators even up to 24 h before intervention. And 95% unstable patients can be identified more than 5 h in advance.ConclusionsThe HSI has acceptable discrimination but underestimates the risk of stable patients in predicting the onset of hemodynamic instability in an external cohort.

Project description:BackgroundDevelopment of emergency department (ED) triage systems that accurately differentiate and prioritize critically ill from stable patients remains challenging. We used machine learning models to predict clinical outcomes, and then compared their performance with that of a conventional approach-the Emergency Severity Index (ESI).MethodsUsing National Hospital and Ambulatory Medical Care Survey (NHAMCS) ED data, from 2007 through 2015, we identified all adult patients (aged ??18?years). In the randomly sampled training set (70%), using routinely available triage data as predictors (e.g., demographics, triage vital signs, chief complaints, comorbidities), we developed four machine learning models: Lasso regression, random forest, gradient boosted decision tree, and deep neural network. As the reference model, we constructed a logistic regression model using the five-level ESI data. The clinical outcomes were critical care (admission to intensive care unit or in-hospital death) and hospitalization (direct hospital admission or transfer). In the test set (the remaining 30%), we measured the predictive performance, including area under the receiver-operating-characteristics curve (AUC) and net benefit (decision curves) for each model.ResultsOf 135,470 eligible ED visits, 2.1% had critical care outcome and 16.2% had hospitalization outcome. In the critical care outcome prediction, all four machine learning models outperformed the reference model (e.g., AUC, 0.86 [95%CI 0.85-0.87] in the deep neural network vs 0.74 [95%CI 0.72-0.75] in the reference model), with less under-triaged patients in ESI triage levels 3 to 5 (urgent to non-urgent). Likewise, in the hospitalization outcome prediction, all machine learning models outperformed the reference model (e.g., AUC, 0.82 [95%CI 0.82-0.83] in the deep neural network vs 0.69 [95%CI 0.68-0.69] in the reference model) with less over-triages in ESI triage levels 1 to 3 (immediate to urgent). In the decision curve analysis, all machine learning models consistently achieved a greater net benefit-a larger number of appropriate triages considering a trade-off with over-triages-across the range of clinical thresholds.ConclusionsCompared to the conventional approach, the machine learning models demonstrated a superior performance to predict critical care and hospitalization outcomes. The application of modern machine learning models may enhance clinicians' triage decision making, thereby achieving better clinical care and optimal resource utilization.

Dataset Information

Using machine learning tools to predict outcomes for emergency department intensive care unit patients.

Publications

Using machine learning tools to predict outcomes for emergency department intensive care unit patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets