Dataset Information

Dynamic prediction of mortality in COVID-19 patients in the intensive care unit: A retrospective multi-center cohort study.

ABSTRACT:

Background

The COVID-19 pandemic continues to overwhelm intensive care units (ICUs) worldwide, and improved prediction of mortality among COVID-19 patients could assist decision making in the ICU setting. In this work, we report on the development and validation of a dynamic mortality model specifically for critically ill COVID-19 patients and discuss its potential utility in the ICU.

Methods

We collected electronic medical record (EMR) data from 3222 ICU admissions with a COVID-19 infection from 25 different ICUs in the Netherlands. We extracted daily observations of each patient and fitted both a linear (logistic regression) and non-linear (random forest) model to predict mortality within 24 h from the moment of prediction. Isotonic regression was used to re-calibrate the predictions of the fitted models. We evaluated the models in a leave-one-ICU-out (LOIO) cross-validation procedure.

Results

The logistic regression and random forest model yielded an area under the receiver operating characteristic curve of 0.87 [0.85; 0.88] and 0.86 [0.84; 0.88], respectively. The recalibrated model predictions showed a calibration intercept of -0.04 [-0.12; 0.04] and slope of 0.90 [0.85; 0.95] for logistic regression model and a calibration intercept of -0.19 [-0.27; -0.10] and slope of 0.89 [0.84; 0.94] for the random forest model.

Discussion

We presented a model for dynamic mortality prediction, specifically for critically ill COVID-19 patients, which predicts near-term mortality rather than in-ICU mortality. The potential clinical utility of dynamic mortality models such as benchmarking, improving resource allocation and informing family members, as well as the development of models with more causal structure, should be topics for future research.

SUBMITTER: Smit JM

PROVIDER: S-EPMC9356569 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Dynamic prediction of mortality in COVID-19 patients in the intensive care unit: A retrospective multi-center cohort study.

Smit J M JM Krijthe J H JH Endeman H H Tintu A N AN de Rijke Y B YB Gommers D A M P J DAMPJ Cremer O L OL Bosman R J RJ Rigter S S Wils E-J EJ Frenzel T T Dongelmans D A DA De Jong R R Peters M A A MAA Kamps M J A MJA Ramnarain D D Nowitzky R R Nooteboom F G C A FGCA De Ruijter W W Urlings-Strop L C LC Smit E G M EGM Mehagnoul-Schipper D J DJ Dormans T T De Jager C P C CPC Hendriks S H A SHA Achterberg S S Oostdijk E E Reidinga A C AC Festen-Spanjer B B Brunnekreef G B GB Cornet A D AD Van den Tempel W W Boelens A D AD Koetsier P P Lens J A JA Faber H J HJ Karakus A A Entjes R R De Jong P P Rettig T C D TCD Arbous M S MS Lalisang R C A RCA Tonutti M M De Bruin D P DP Elbers P W G PWG Van Bommel J J Reinders M J T MJT

Intelligence-based medicine 20220806

<h4>Background</h4>The COVID-19 pandemic continues to overwhelm intensive care units (ICUs) worldwide, and improved prediction of mortality among COVID-19 patients could assist decision making in the ICU setting. In this work, we report on the development and validation of a dynamic mortality model specifically for critically ill COVID-19 patients and discuss its potential utility in the ICU.<h4>Methods</h4>We collected electronic medical record (EMR) data from 3222 ICU admissions with a COVID-1 ...[more]

PMID: 35958674

Similar Datasets

Project description:IntroductionRecent data have suggested that patient admission during intensive care unit (ICU) morning bedside rounds is associated with less favorable outcome. We undertook the present study to explore the association between morning round-time ICU admissions and hospital mortality in a large Canadian health region.MethodsA multi-center retrospective cohort study was performed at five hospitals in Edmonton, Canada, between July 2002 and December 2009. Round-time ICU admission was defined as occurring between 8 and 11:59 a.m. Multivariable logistic regression analysis was used to explore the association between round-time admission and outcome.ResultsOf 18,857 unique ICU admissions, 2,055 (10.9%) occurred during round time. Round-time admissions were more frequent in community hospitals compared with tertiary hospitals (12.0% vs. 10.5%; odds ratio [OR] 1.16; 95% CI, 1.05-1.29, P < 0.004) and from the ward compared with the emergency department (ED) or operating theater (17.5% vs. 9.2%; OR 2.1; 95% CI, 1.9-2.3, P < 0.0001). Round-time admissions were more often medical than surgical (12.6% vs. 6.6%; OR 2.06; 95% CI, 1.83-2.31, P < 0.0001), had more comorbid illness (11.9% vs. 10.5%; OR 1.15; 95% CI, 1.04-1.27, P < 0.008) and higher APACHE II score (22.2 vs. 21.3, P < 0.001), and were more likely to have a primary diagnosis of respiratory failure (37.0% vs. 31.3%, P < 0.001) or sepsis (11.1% vs. 9.0%, P = 0.002). Crude ICU mortality (15.3% vs. 11.6%; OR 1.38; 95% CI, 1.21-1.57, P < 0.0001) and hospital mortality (23.9% vs. 20.6%; OR 1.21; 95% CI, 1.09-1.35, P < 0.001) were higher for round-time compared with non-round-time admissions. In multi-variable analysis, round-time admission was associated with increased ICU mortality (OR 1.19, 95% CI, 1.03-1.38, P = 0.017) but was not significantly associated with hospital mortality (OR 1.02; 95% CI, 0.90-1.16, P = 0.700). In the subgroup admitted from the ED, round-time admission showed significantly higher ICU mortality (OR 1.54; 95% CI, 1.21-1.95; P < 0.001) and a trend for higher hospital mortality (OR 1.22; 95% CI, 0.99-1.51, P = 0.057).ConclusionsApproximately 1 in 10 patients is admitted during morning rounds. These patients are more commonly admitted from the ward and are burdened by comorbidities, are non-operative, and have higher illness severity. These patients admitted during morning rounds have higher observed ICU mortality but no difference in hospital mortality.

Project description:BackgroundCoronavirus disease 2019 (COVID-19) mortality has waned significantly over time; however, factors contributing towards this reduction largely remain unidentified. The purpose of this study was to evaluate the trend in mortality at our large tertiary academic health system and factors contributing to this trend.MethodsThis is a retrospective cohort study of intensive care unit (ICU) patients diagnosed with COVID-19 between March and August 2020 admitted across 14 hospitals in the Philadelphia area. Collected data included demographics, comorbidities, admission risk of mortality score, laboratory values, medical interventions, survival outcomes, hospital and ICU length of stay (LOS) and discharge disposition. Chi-square (χ2) test, Fisher exact test, Cochran-Mantel-Haenszel method, multinomial logistic regression models, independent sample t-test, Mann-Whitney U test and one-way analysis of variance (ANOVA) were used.ResultsA total of 1,204 patients were included. Overall mortality was 39%. Mortality declined significantly from 46% in March to 14% in August 2020 (P < 0.05). The most common underlying comorbidities were hypertension (60.2%), diabetes mellitus (44.7%), dyslipidemia (31.6%) and congestive heart failure (14.7%). Hydroxychloroquine (HCQ) use was more commonly associated with the patients who died, while the use of remdesivir, tocilizumab, steroids and duration of these medications were not significantly different. Peak values of ferritin, lactate dehydrogenase (LDH), C-reactive protein (CRP) and D-dimer levels were significantly higher in patients who died (P < 0.05). The mean hospital LOS was significantly longer in the patients who survived compared to the patients who died (18 vs. 12, P < 0.05).ConclusionsThe mortality of patients admitted to our ICU system significantly decreased over time. Factors that may have contributed to this may be the result of a better understanding of COVID-19 pathophysiology and treatments. Further research is needed to elucidate the factors contributing to a reduction in the mortality rate for this patient population.

Project description:Objective Congestive heart failure (CHF) is a clinical syndrome in which the heart disease progresses to a severe stage. Early diagnosis and risk assessment of death of patients with CHF are critical to prognosis and treatment. The purpose of this study was to establish a nomogram that predicts the in-hospital death of patients with CHF in the intensive care unit (ICU). Design A retrospective observational cohort study. Setting and participants Data for the study were from 30 411 patients with CHF in the Medical Information Mart for Intensive Care database and the eICU Collaborative Research Database (eICU-CRD). Primary outcome In-hospital mortality. Methods Univariate logistic regression analysis was used to select risk factors associated with in-hospital mortality of patients with CHF, and multivariate logistic regression was used to build the prediction model. Discrimination, calibration and clinical validity of the model were evaluated by AUC, calibration curve, Hosmer-Lemeshow χ2 test and decision curve analysis, respectively. Finally, data from 15 503 patients with CHF in the multicentre eICU-CRD were used for external validation of the established nomogram. Results The inclusion criteria were met by 15 983 subjects, whose in-hospital mortality rate was 12.4%. Multivariate analysis determined that the independent risk factors were age, race, norepinephrine, dopamine, phenylephrine, vasopressin, mechanical ventilation, intubation, hepatic failure (HepF), heart rate, respiratory rate, temperature, systolic blood pressure (SBP), anion gap (AG), blood urea nitrogen (BUN), creatinine, chloride, mean corpuscular volume (MCV), red blood cell distribution width (RDW) and white cell count (WCC). The C-index of the nomogram (0.767, 95% CI 0.759 to 0.779) was superior to that of the traditional Sequential Organ Failure Assessment, Acute Physiology Score III and Get With The Guidelines Heart Failure scores, indicating its discrimination power. Calibration plots demonstrated that the predicted results are in good agreement with the observed results. The decision curves of the derivation and validation sets both had net benefits. Conclusion The 20 independent risk factors for in-hospital mortality of patients with CHF were age, race, norepinephrine, dopamine, phenylephrine, vasopressin, mechanical ventilation, intubation, HepF, heart rate, respiratory rate, temperature, SBP, AG, BUN, creatinine, chloride, MCV, RDW and WCC. The nomogram, which included these factors, accurately predicted the in-hospital mortality of patients with CHF. The novel nomogram has the potential for use in clinical practice as a tool to predict and assess mortality of patients with CHF in the ICU.

Project description:IntroductionPatients with idiopathic inflammatory myopathies (IIMs) are sometimes complicated with life-threatening conditions requiring intensive care unit (ICU) admission. In the past, owing to the low incidence of IIM, little was known about such patients. Our aim was to investigate the clinical features and outcomes of these patients and identify their risk factors for mortality.MethodsA retrospective study was performed of IIM patients admitted over an 8-year period to the medical ICU of a tertiary referral center in China. We collected data regarding demographic features, IIM-related clinical characteristics, reasons for admission, organ dysfunction, and outcomes. Independent predictors of ICU mortality were identified through multivariate logistic regression analysis.ResultsOf the 102 patients in our cohort, polymyositis (PM), dermatomyositis (DM), and clinically amyopathic dermatomyositis (CADM) accounted for 23.5%, 64.7%, and 11.7% respectively. The median duration from the onset of IIM to ICU admission was 4.3 months (interquartile range [IQR], 2.6-9.4 months). Reasons for ICU admission were infection alone (39.2%), acute exacerbation of IIM alone (27.5%), the coexistence of both (27.5%), or other reasons (5.8%). Pneumonia accounted for 97% of the infections; 63.2% of infections with documented pathogens were caused by opportunistic agents. Rapid progressive interstitial lung disease (RP-ILD) was responsible for 87.5% of acute exacerbation of IIM. The median Acute Physiology and Chronic Health Evaluation II (APACHE II) score on ICU day 1 was 17 (IQR 14-20). On ICU admission, acute respiratory failure (ARF) was the most common type (80.4%) of organ failure. The mortality rate in the ICU was 79.4%. Factors associated with increased ICU mortality included a diagnosis of DM (including CADM), a high APACHE II score, the presence of ARF, a decreased PaO2/FiO2 ratio, and a low lymphocyte count at the time of ICU admission.ConclusionsThe outcome of IIM patients admitted to the ICU was extremely poor. A diagnosis of DM/CADM, the presence and severity of ARF, and the lymphocyte counts at ICU admission were shown to be valuable for predicting outcome. Opportunistic infections and rapidly progressive interstitial lung disease warrant concern in treating these patients.

Dataset Information

Dynamic prediction of mortality in COVID-19 patients in the intensive care unit: A retrospective multi-center cohort study.

Background

Methods

Results

Discussion

Publications

Dynamic prediction of mortality in COVID-19 patients in the intensive care unit: A retrospective multi-center cohort study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets