A machine learning-based model for predicting intensive care unit length of stay in patients with cardiac surgery (Preprint)

2021 ◽  
Author(s):  
Nianyue Wu ◽  
Siru Liu ◽  
Haotian Zhang ◽  
Xiaomin Hou ◽  
Ping Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care ◽  
Length Of Stay ◽  
Confidence Interval ◽  
Medical Information ◽  
Lasso Regression ◽  
Icu Length Of Stay ◽  
Data Source ◽  
The Mean ◽  
Selection Operator

BACKGROUND The intensive care unit (ICU) length of stay is significant to evaluate the effect of cardiac surgical treatment inpatient. OBJECTIVE This research aims to accurately predict the ICU length of stay in patients with cardiac surgery. Methods: We used machine learning methods to construct the model, and the medical information mart for intensive care (MIMIC IV) database was used as the data source. A total of 7,567 patients were enrolled and the mean length of stay in the ICU was 3.12 days. A total of 126 predictors were included, and 44 important predictors were screened by least absolute shrinkage and selection operator (Lasso) regression. METHODS We used machine learning methods to construct the model, and the medical information mart for intensive care (MIMIC IV) database was used as the data source. A total of 7,567 patients were enrolled and the mean length of stay in the ICU was 3.12 days. A total of 126 predictors were included, and 44 important predictors were screened by least absolute shrinkage and selection operator (Lasso) regression. RESULTS The mean accuracy are 0.603 (95% confidence interval (CI): [0.602-0.604]), 0.687 (95% confidence interval (CI): [0.687-0.688]) and 0.688 (95% confidence interval (CI): [0.687-0.689]) for the logistic regression (LR) with all variables, the gradient boosted decision tree (GBDT) with important variables and the GBDT with all variables respectively. CONCLUSIONS The GBDT model with important predictors partly overestimated patients whose length of stay was less than 3 days and underestimated patients whose length of stay was longer than 3 days. But the better prediction performance of GBDT facilitates early intervention of ICU patients with a long period of hospitalization.

scholarly journals The Effects of Early Rehabilitation in the Intensive Care Unit for Patients with Severe COVID-19 Pneumonia: A Retrospective Cohort Study

2022 ◽  
Vol 11 (2) ◽  
pp. 357
Author(s):  
Tokio Kinoshita ◽  
Yukihide Nishimura ◽  
Yasunori Umemoto ◽  
Yasuhisa Fujita ◽  
Ken Kouda ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Cohort Study ◽  
Intensive Care ◽  
Length Of Stay ◽  
Retrospective Cohort Study ◽  
Retrospective Cohort ◽  
Small Sample ◽  
Icu Length Of Stay ◽  
The Mean ◽  
Mean Time

This retrospective cohort study aimed to examine the rehabilitation effect of patients with coronavirus disease 2019 (COVID-19) in the intensive care unit (ICU) under mechanical ventilation and included ICU patients from a university hospital who received rehabilitation under ventilator control until 31 May 2021. Seven patients were included, and three of them died; thus, the results of the four survivors were examined. The rehabilitation program comprised the extremity range-of-motion training and sitting on the bed’s edge. The Sequential Organ Failure Assessment score (median (25–75th percentiles)) at admission was 7.5 (5.75–8.5), and the activities of daily living (ADLs) were bedridden, the lowest in the Functional Independence Measure (FIM) and Barthel Index (BI) surveys. Data on the mean time to extubation, ICU length of stay, and ADLs improvement (FIM and BI) during ICU admission were obtained. Inferential analyses were not performed considering the small sample size. The mean time to extubation was 4.9 ± 1.1 days, and the ICU length of stay was 11.8 ± 5.0 days. ΔFIM was 36.5 (28.0–40.5), and the ΔBI was 22.5 (3.75–40.0). Moreover, no serious adverse events occurred in the patients during rehabilitation. Early mobilization of patients with COVID-19 may be useful in ADLs improvement during ICU stay.

scholarly journals Outcomes of non-COVID-19 critically ill patients during the COVID-19 pandemic

2021 ◽  
Author(s):  
Răzvan Bologheanu ◽  
Mathias Maleczek ◽  
Daniel Laxar ◽  
Oliver Kimberger
Keyword(s):  
Intensive Care ◽  
Length Of Stay ◽  
Critically Ill ◽  
Critically Ill Patients ◽  
Medical Specialty ◽  
Routine Care ◽  
Treatment Pathways ◽  
Icu Length Of Stay ◽  
Matched Study ◽  
The Impact

Summary Background Coronavirus disease 2019 (COVID-19) disrupts routine care and alters treatment pathways in every medical specialty, including intensive care medicine, which has been at the core of the pandemic response. The impact of the pandemic is inevitably not limited to patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and their outcomes; however, the impact of COVID-19 on intensive care has not yet been analyzed. Methods The objective of this propensity score-matched study was to compare the clinical outcomes of non-COVID-19 critically ill patients with the outcomes of prepandemic patients. Critically ill, non-COVID-19 patients admitted to the intensive care unit (ICU) during the first wave of the pandemic were matched with patients admitted in the previous year. Mortality, length of stay, and rate of readmission were compared between the two groups after matching. Results A total of 211 critically ill SARS-CoV‑2 negative patients admitted between 13 March 2020 and 16 May 2020 were matched to 211 controls, selected from a matching pool of 1421 eligible patients admitted to the ICU in 2019. After matching, the outcomes were not significantly different between the two groups: ICU mortality was 5.2% in 2019 and 8.5% in 2020, p = 0.248, while intrahospital mortality was 10.9% in 2019 and 14.2% in 2020, p = 0.378. The median ICU length of stay was similar in 2019: 4 days (IQR 2–6) compared to 2020: 4 days (IQR 2–7), p = 0.196. The rate of ICU readmission was 15.6% in 2019 and 10.9% in 2020, p = 0.344. Conclusion In this retrospective single center study, mortality, ICU length of stay, and rate of ICU readmission did not differ significantly between patients admitted to the ICU during the implementation of hospital-wide COVID-19 contingency planning and patients admitted to the ICU before the pandemic.

scholarly journals 1412. The utility of respiratory viral panel testing for nosocomial fever in the neonatal intensive care unit

2020 ◽  
Vol 7 (Supplement_1) ◽  
pp. S713-S713
Author(s):  
Carlo Fopiano Palacios ◽  
Eric Lemmon ◽  
James Campbell
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Linear Regression ◽  
Length Of Stay ◽  
Neonatal Intensive Care Unit ◽  
Neonatal Intensive Care ◽  
P Value ◽  
Fisher Exact Test ◽  
The Mean ◽  
Pcr Test

Abstract Background Patients in the neonatal intensive care unit (NICU) often develop fevers during their inpatient stay. Many neonates are empirically started on antibiotics due to their fragile clinical status. We sought to evaluate whether the respiratory viral panel (RVP) PCR test is associated with use of antibiotics in patients who develop a fever in the NICU. Methods We conducted a retrospective chart review on patients admitted to the Level 4 NICU of the University of Maryland Medical Center from November 2015 to June 2018. We included all neonates who developed a fever 48 hours into their admission. We collected demographic information and data on length of stay, fever work-up and diagnostics (including labs, cultures, RVP), and antibiotic use. Descriptive statistics, Fisher exact test, linear regression, and Welch’s ANOVA were performed. Results Among 347 fever episodes, the mean age of neonates was 72.8 ± 21.6 days, and 45.2% were female. Out of 30 total RVP samples analyzed, 2 were positive (6.7%). The most common causes of fever were post-procedural (5.7%), pneumonia (4.8%), urinary tract infection (3.5%), meningitis (2.6%), bacteremia (2.3%), or due to a viral infection (2.0%). Antibiotics were started in 208 patients (60%), while 61 neonates (17.6%) were already on antibiotics. The mean length of antibiotics was 7.5 ± 0.5 days. Neonates were more likely to get started on antibiotics if they had a negative RVP compared to those without a negative RVP (89% vs. 11%, p-value < 0.0001). Patients with a positive RVP had a decreased length of stay compared to those without a positive RVP (30.3 ± 8.7 vs. 96.8 ± 71.3, p-value 0.01). On multivariate linear regression, a positive RVP was not associated with length of stay. Conclusion Neonates with a negative respiratory viral PCR test were more likely to be started on antibiotics for fevers. Respiratory viral PCR testing can be used as a tool to promote antibiotic stewardship in the NICU. Disclosures All Authors: No reported disclosures

scholarly journals Attributable Cost and Length of Stay for Patients With Central Venous Catheter—Associated Bloodstream Infection in Mexico City Intensive Care Units A Prospective, Matched Analysis

2007 ◽  
Vol 28 (1) ◽  
pp. 31-35 ◽  
Author(s):  
Francisco Higuera ◽  
Manuel Sigfrido Rangel-Frausto ◽  
Victor Daniel Rosenthal ◽  
Jose Martinez Soto ◽  
Jorge Castañon ◽  
...  
Keyword(s):  
Intensive Care ◽  
Length Of Stay ◽  
Central Venous Catheter ◽  
Bloodstream Infection ◽  
Mexico City ◽  
Venous Catheter ◽  
Central Venous ◽  
Icu Stay ◽  
The Mean ◽  
Length Of Icu Stay

Background.No information is available about the financial impact of central venous catheter (CVC)-associated bloodstream infection (BSI) in Mexico.Objective.To calculate the costs associated with BSI in intensive care units (ICUs) in Mexico City.Design.An 18-month (June 2002 through November 2003), prospective, nested case-control study of patients with and patients without BSI.Setting.Adult ICUs in 3 hospitals in Mexico City.Patients and Methods.A total of 55 patients with BSI (case patients) and 55 patients without BSI (control patients) were compared with respect to hospital, type of ICU, year of hospital admission, length of ICU stay, sex, age, and mean severity of illness score. Information about the length of ICU stay was obtained prospectively during daily rounds. The daily cost of ICU stay was provided by the finance department of each hospital. The cost of antibiotics prescribed for BSI was provided by the hospitals' pharmacy departments.Results.For case patients, the mean extra length of stay was 6.1 days, the mean extra cost of antibiotics was $598, the mean extra hospital cost was $11,591, and the attributable extra mortality was 20%.Conclusions.In this study, the duration of ICU stay for patients with central venous catheter-associated BSI was significantly longer than that for control patients, resulting in increased healthcare costs and a higher attributable mortality. These conclusions support the need to implement preventive measures for hospitalized patients with central venous catheters in Mexico.

scholarly journals Tracheostomy in a Neuro-Intensive Care Setting: Indications and Timing

1997 ◽  
Vol 25 (4) ◽  
pp. 365-368 ◽  
Author(s):  
W. Y. Koh ◽  
T. W. K. Lew ◽  
N. M. Chin ◽  
M. F. M. Wong
Keyword(s):  
Intensive Care ◽  
Length Of Stay ◽  
Care Setting ◽  
Ventilatory Support ◽  
Upper Airway ◽  
Ventilator Associated Pneumonia ◽  
Intensive Care Setting ◽  
Mechanically Ventilated ◽  
Icu Length Of Stay ◽  
Elective Tracheostomy

A retrospective review was made of 49 survivors who were mechanically ventilated for more than 48 hours in the neurosurgical ICU. Thirty-two patients (Gp I) were successfully extubated, 9 patients (Gp II) underwent tracheostomy after one or more failed extubations, and 8 patients (Gp III) underwent elective tracheostomy. Glasgow Coma Scale (GCS) scores at extubation were 11.3±2.8 (mean (SD) for Gp I vs 7.8±2.7 for Gp II (P= n.s.) and at elective tracheostomy (Gp III) was 5.4±2.3. Incidence of ventilator-associated pneumonia were 35% in Gp I vs 100% of patients in Gp II and III (P<0.05). Reasons for reintubation in 7 of 9 patients (Gp II) were upper airway obstruction and tenacious tracheal secretions while 14 of 17 patients were weaned off the ventilator within 48 hours of tracheostomy. The length of stay in ICU was 16.8±7.1 days in Gp II vs 11.7±2.9 days in Gp III (P<0.05). In our study, elective tracheostomy for selected patients with poor GCS scores and nosocomial pneumonia has resulted in shortened ICU length of stay and rapid weaning from ventilatory support.

The implementation of an Intensive Care Information System allows shortening the ICU length of stay

2014 ◽  
Vol 29 (2) ◽  
pp. 263-269 ◽  
Author(s):  
Eric Levesque ◽  
Emir Hoti ◽  
Daniel Azoulay ◽  
Philippe Ichai ◽  
Didier Samuel ◽  
...  
Keyword(s):  
Information System ◽  
Intensive Care ◽  
Length Of Stay ◽  
Icu Length Of Stay ◽  
Care Information

scholarly journals (P2-94) How One Group of Physicians Helped a Busy Emergency Department

2011 ◽  
Vol 26 (S1) ◽  
pp. s167-s167
Author(s):  
J. Hu ◽  
J. Xu ◽  
J. Botler ◽  
S. Haydar
Keyword(s):  
Intensive Care Unit ◽  
Emergency Department ◽  
Intensive Care ◽  
Length Of Stay ◽  
Medical Center ◽  
Hospital Medicine ◽  
Intermediate Care ◽  
Physician Time ◽  
The Mean ◽  
Medical Telemetry

A pilot admission leadership physician (ALP) program was experimented within a 693-bed, tertiary medical center with a 60-bed emergency department. This trial was intended to investigate whether having a physician triage potential patients would shorten patients' length-of-stay in the emergency department. After a emergency physician evaluated patients, ALP triaged them. The ALP ordered the appropriate bed for the patients if they qualified for the inpatient criteria, choosing among medical, medical telemetry, cardiac telemetry, intermediate care, or intensive care bed. The mean patient door-to-bed order time (time between patients reaching the emergency department to time to bed ordered by ALP) is 330.7 minutes (n = 234, SD = 151.68, 95% CI = 310.21–351.28) with ALP involvement. Compared with the mean door-to-bed order time of 337.8 minutes (n = 827, SD = 149.71, 95%CI = 326.98–348.57) without ALP, ALP shortened the waiting time by 7.09 minutes. During the same period, the door-to-physician time was 41.38 minutes (SD = 38.87 95%CI = 36.38–46.39), compared with 39.52 minutes (SD = 40.32, 95%CI = 36.77–42.27) before ALP. The time for patients waiting in the emergency department for other services such as surgery, psychiatry, and pediatrics also have decreased accordingly. Incorrect medical admissions such as scrambling to get the patient to the intensive care unit right after seeing patients has decreased (data not provided). Identifying physicians as physicians in the emergency department who triage potential admissions also has improved efficiencies within the hospital medicine group and bonding with ER physicians.

scholarly journals The assessment of clinical evaluation and treatment results of high-energy blunt polytrauma patients

Medicina ◽  
2007 ◽  
Vol 43 (2) ◽  
pp. 137 ◽  
Author(s):  
Algimantas Pamerneckas ◽  
Andrei Pijadin ◽  
Giedrius Pilipavičius ◽  
Gintaras Tamulaitis ◽  
Vytautas Toliušis ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Length Of Stay ◽  
Hospital Stay ◽  
Injury Severity ◽  
Motor Vehicle ◽  
High Energy ◽  
Ventilation Time ◽  
Polytrauma Patients ◽  
The Mean

The aim of this study was to evaluate the mechanism of high-energy blunt trauma, age and gender of patients, severity of regional and multiple injury, ventilation time, length of stay in intensive care unit and in-hospital stay, in-hospital complications, and treatment outcome. Materials and methods. Data on 159 patients with severe multiple injuries, meeting inclusion criteria, were collected prospectively and evaluated retrospectively. Results. The mean age of multiple trauma patients was 43.9±1.4 years; males were injured 2.5 times more often than females (P<0.001). More than half (66.7%) of patients were 17–64-year-old males. Majority (83%) of all patients were injured in motor vehicle crashes, and 52.2% of these patients were pedestrians. The mean Injury Severity Score was 29.5±0.8, and severe (Abbreviated Injury Scale score of 3 and more) injuries of extremities, head, and chest made up 69.1% of all injuries. The mean ventilation time, mean length of stay in intensive care unit, and mean in-hospital stay were 5.5±0.7, 7.0±0.8, and 23.6±1.6 days, respectively. Acute lung complications were the most common (25.2%). Systemic inflammatory response syndrome developed in 7.5% of patients, and sepsis in 3.8% of patients. More than one-fifth (20.8%) of polytrauma patients died. Conclusions. Working-age male pedestrians (17–64 years old) made up two-thirds of all polytrauma patients. Severe injuries of extremities, head, and chest were present in 69.1% of all cases. Lung complications were the most common.

Economic Evaluation of Propofol for Sedation of Patients Admitted to Intensive Care Units

2002 ◽  
Vol 96 (1) ◽  
pp. 196-201 ◽  
Author(s):  
Aslam H. Anis ◽  
Xiao-hua Wang ◽  
Hector Leon ◽  
Richard Hall ◽  
Keyword(s):  
Sensitivity Analysis ◽  
Intensive Care ◽  
Length Of Stay ◽  
Resource Utilization ◽  
Health Resources ◽  
Sensitivity Analyses ◽  
Health Resource ◽  
Health Resource Utilization ◽  
Icu Length Of Stay ◽  
Extubation Time

Background The goal of the current study was to evaluate the economic impact of propofol as compared with midazolam for sedating patients in the intensive care unit (ICU). Methods A randomized, unblinded, multicenter pharmacoeconomic trial captured health resource utilization and outcome measurements associated with sedation and treatment of patients in four ICUs across Canada. Statistical analysis was performed to investigate the difference in sedation quality, ICU length of stay, and other health resources used. The authors compared the costs (1997 Canadian dollars) associated with the two treatments. Two types of sensitivity analyses were performed. Results Although overall sedation duration was similar, propofol patients spent more time at adequately sedated status (60.2% vs. 44%; P = 0.01) and were extubated faster (median extubation time, 2.5 vs. 7.1 h; P = 0.001). The ICU length of stay and health resource utilization did not differ. The total cost per patient, including drug cost and ICU stay cost, did not differ between groups (median, $5,718 for propofol vs. $5,950 for midazolam; P = 0.94). The first sensitivity analysis suggested that the incremental cost (per patient) of propofol varies from an extra cost of $114 to a savings of $2,709. Based on a hypothetical model, the second sensitivity analysis showed a potential saving of $479 per patient as a result of improved discharge planning. Conclusion The analysis demonstrated that using propofol resulted in a reduction of time to extubation and higher sedative regimen costs. There was no difference in intensity of resource use or ICU length of stay and hence in costs. Issues regarding discharge delay among propofol-treated patients remain to be explored.

Sign in / Sign up

Export Citation Format

Share Document