scholarly journals Missing data methods for intensive care unit SOFA scores in electronic health records studies: results from a Monte Carlo simulation

2021 ◽  
Author(s):  
Daniel L Brinton ◽  
Dee W Ford ◽  
Renee H Martin ◽  
Kit N Simpson ◽  
Andrew J Goodwin ◽  
...  
Keyword(s):  
Intensive Care ◽  
Missing Data ◽  
Sample Size ◽  
Sofa Score ◽  
Electronic Health Record Data ◽  
Complete Dataset ◽  
Failure Assessment ◽  
Electronic Health ◽  
And Performance ◽  
Score Component

Aim: Missing data cause problems through decreasing sample size and the potential for introducing bias. We tested four missing data methods on the Sequential Organ Failure Assessment (SOFA) score, an intensive care research severity adjuster. Methods: Simulation study using 2015–2017 electronic health record data, where the complete dataset was sampled, missing SOFA score elements imposed and performance examined of four missing data methods – complete case analysis, median imputation, zero imputation (recommended by SOFA score creators) and multiple imputation (MI) – on the outcome of in-hospital mortality. Results: MI performed well, whereas other methods introduced varying amounts of bias or decreased sample size. Conclusion: We recommend using MI in analyses where SOFA score component values are missing in administrative data research.

scholarly journals Establishment of a prognostic model based on the Sequential Organ Failure Assessment score for patients with first-time acute myocardial infarction

2021 ◽  
Vol 49 (5) ◽  
pp. 030006052110119
Author(s):  
Shuai Zheng ◽  
Jun Lyu ◽  
Didi Han ◽  
Fengshuo Xu ◽  
Chengzhuo Li ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Sequential Organ Failure Assessment ◽  
Organ Failure ◽  
Sofa Score ◽  
Cardiovascular Events ◽  
Failure Assessment ◽  
First Time

Objective This study aimed to identify the prognostic factors of patients with first-time acute myocardial infarction (AMI) and to establish a nomogram for prognostic modeling. Methods We studied 985 patients with first-time AMI using data from the Multi-parameter Intelligent Monitoring for Intensive Care database and extracted their demographic data. Cox proportional hazards regression was used to examine outcome-related variables. We also tested a new predictive model that includes the Sequential Organ Failure Assessment (SOFA) score and compared it with the SOFA-only model. Results An older age, higher SOFA score, and higher Acute Physiology III score were risk factors for the prognosis of AMI. The risk of further cardiovascular events was 1.54-fold higher in women than in men. Patients in the cardiac surgery intensive care unit had a better prognosis than those in the coronary heart disease intensive care unit. Pressurized drug use was a protective factor and the risk of further cardiovascular events was 1.36-fold higher in nonusers. Conclusion The prognosis of AMI is affected by age, the SOFA score, the Acute Physiology III score, sex, admission location, type of care unit, and vasopressin use. Our new predictive model for AMI has better performance than the SOFA model alone.

Abstract 303: Inpatient Sleep Interruptions Among Patients With Congestive Heart Failure

2020 ◽  
Vol 13 (Suppl_1) ◽  
Author(s):  
Evan Claggett ◽  
Rachel H Krallman ◽  
Delaney Feldeisen ◽  
Daniel G Montgomery ◽  
Kim Eagle ◽  
...  
Keyword(s):  
Heart Failure ◽  
Congestive Heart Failure ◽  
Electronic Health Record ◽  
Sample Size ◽  
Health Record ◽  
Post Discharge ◽  
Electronic Health Record Data ◽  
Care Clinic ◽  
Electronic Health ◽  
Poor Outcomes

Background: The effects of sleep deprivation are vast, ranging from increased stress responses, to lowered immunity and delayed wound healing. However, sleep disruptions are common in the inpatient setting. This study sought to quantify the number and frequency of inpatient sleep disturbances and analyze post-discharge outcomes (emergency department visit, readmission, death) among congestive heart failure (CHF) patients. Methods: Data were collected retrospectively from 30 randomly selected patients admitted for CHF and referred to a cardiac transitional care clinic from 2014 to 2017. Each night over the course of the hospitalization was broken into 12 one-hour intervals (1900-0659 hours), and the electronic health record was examined for 20 variables indicative of sleep disruption (e.g. vitals taken, medications dispensed, wound care) (Figure 1). Demographics and outcomes were compared between high (above median) and low (below median) groups for average number of nightly interval interruptions and average longest uninterrupted sleep interval (LUSI). Results: On average, patients had a length of admission of 5.4 nights, a LUSI of 2.9 hours (range: 1-4), and 6.3 disruptions between 1900-0659 hours (range: 3-8). The readmission rates for the total population were 23% at 30 days and 63% at 180 days. No significant differences were seen in demographics or outcomes up to 180 days post-discharge when comparing high and low patient groups in either average nightly interval interruptions or average LUSI. Conclusion: Although no differences were seen between groups, the majority of patients had poor outcomes (23% were readmitted at 30 days; 63% at 180 days) as well as poor sleep during their admission. The lack of sleep across the entire patient population may be contributing to the poor outcomes observed. Many of the variables reviewed (e.g. vitals taken, medications dispensed, etc.) had potentially elective timing, which suggests actionable changes to the inpatient process may be possible to improve sleep quantity and quality. This was an exploratory pilot study to determine the ability to use electronic health record data for this purpose. As such, the sample size was too small to detect differences. A larger sample size is needed to better understand the extent to which sleep disruptions impact patient outcomes.

scholarly journals Utilization of Deep Learning for Subphenotype Identification in Sepsis-Associated Acute Kidney Injury

2020 ◽  
Vol 15 (11) ◽  
pp. 1557-1565 ◽  
Author(s):  
Kumardeep Chaudhary ◽  
Akhil Vaid ◽  
Áine Duffy ◽  
Ishan Paranjpe ◽  
Suraj Jaladanki ◽  
...  
Keyword(s):  
Intensive Care ◽  
Deep Learning ◽  
Medical Information ◽  
Tertiary Care ◽  
Vital Signs ◽  
Kidney Injury ◽  
The United States ◽  
Care Hospital ◽  
Electronic Health Record Data ◽  
Electronic Health

Background and objectivesSepsis-associated AKI is a heterogeneous clinical entity. We aimed to agnostically identify sepsis-associated AKI subphenotypes using deep learning on routinely collected data in electronic health records.Design, setting, participants, & measurementsWe used the Medical Information Mart for Intensive Care III database, which consists of electronic health record data from intensive care units in a tertiary care hospital in the United States. We included patients ≥18 years with sepsis who developed AKI within 48 hours of intensive care unit admission. We then used deep learning to utilize all available vital signs, laboratory measurements, and comorbidities to identify subphenotypes. Outcomes were mortality 28 days after AKI and dialysis requirement.ResultsWe identified 4001 patients with sepsis-associated AKI. We utilized 2546 combined features for K-means clustering, identifying three subphenotypes. Subphenotype 1 had 1443 patients, and subphenotype 2 had 1898 patients, whereas subphenotype 3 had 660 patients. Subphenotype 1 had the lowest proportion of liver disease and lowest Simplified Acute Physiology Score II scores compared with subphenotypes 2 and 3. The proportions of patients with CKD were similar between subphenotypes 1 and 3 (15%) but highest in subphenotype 2 (21%). Subphenotype 1 had lower median bilirubin levels, aspartate aminotransferase, and alanine aminotransferase compared with subphenotypes 2 and 3. Patients in subphenotype 1 also had lower median lactate, lactate dehydrogenase, and white blood cell count than patients in subphenotypes 2 and 3. Subphenotype 1 also had lower creatinine and BUN than subphenotypes 2 and 3. Dialysis requirement was lowest in subphenotype 1 (4% versus 7% [subphenotype 2] versus 26% [subphenotype 3]). The mortality 28 days after AKI was lowest in subphenotype 1 (23% versus 35% [subphenotype 2] versus 49% [subphenotype 3]). After adjustment, the adjusted odds ratio for mortality for subphenotype 3, with subphenotype 1 as a reference, was 1.9 (95% confidence interval, 1.5 to 2.4).ConclusionsUtilizing routinely collected laboratory variables, vital signs, and comorbidities, we were able to identify three distinct subphenotypes of sepsis-associated AKI with differing outcomes.

Sequential Organ Failure Assessment (SOFA) Score As a Predictor of Outcome in Sepsis in Pediatric Intensive Care Unit

2020 ◽  
Author(s):  
A.V. Lalitha ◽  
J.K. Satish ◽  
Mounika Reddy ◽  
Santu Ghosh ◽  
Jiny George ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Hospital Mortality ◽  
Pediatric Intensive Care Unit ◽  
Sofa Score ◽  
Severity Of Illness ◽  
Pediatric Intensive Care ◽  
Inotropic Support ◽  
Discriminative Power ◽  
Failure Assessment

AbstractSequential organ failure assessment (SOFA) score is used as a predictor of outcome of sepsis in the pediatric intensive care unit. The aim of the study is to determine the application of SOFA scores as a predictor of outcome in children admitted to the pediatric intensive care unit with a diagnosis of sepsis. The design involved is prospective observational study. The study took place at the multidisciplinary pediatric intensive care unit (PICU), tertiary care hospital, South India. The patients included are children, aged 1 month to 18 years admitted with a diagnosis of sepsis (suspected/proven) to a single center PICU in India from November 2017 to November 2019. Data collected included the demographic, clinical, laboratory, and outcome-related variables. Severity of illness scores was calculated to include SOFA score day 1 (SF1) and day 3 (SF3) using a pediatric version (pediatric SOFA score or pSOFA) with age-adjusted cutoff variables for organ dysfunction, pediatric risk of mortality III (PRISM III; within 24 hours of admission), and pediatric logistic organ dysfunction-2 or PELOD-2 (days 1, 3, and 5). No intervention was observed during the period of study. A total of 240 patients were admitted to the PICU with septic shock during the study period. The overall mortality rate was 42 of 240 patients (17.5%). The majority (59%) required mechanical ventilation, while only 19% required renal replacement therapy. The PRISM III, PELOD-2, and pSOFA scores correlated well with mortality. All three severity of illness scores were higher among nonsurvivors as compared with survivors (p < 0.001). pSOFA scores on both day 1 (area under the curve or AUC 0.84) and day 3 (AUC 0.87) demonstrated significantly higher discriminative power for in-hospital mortality as compared with PRISM III (AUC, 0.7), and PELOD-2 (day 1, [AUC, 0.73]), and PELOD-2 (day 3, [AUC, 0.81]). Utilizing a cutoff SOFA score of >8, the relative risk of prolonged duration of mechanical ventilation, requirement for vasoactive infusions (vasoactive infusion score), and PICU length of stay were all significantly increased (p < 0.05), on both days 1 and 3. On multiple logistic regression, adjusted odds ratio of mortality was elevated at 8.65 (95% CI: 3.48–21.52) on day 1 and 16.77 (95% confidence interval or CI: 4.7–59.89) on day 3 (p < 0.001) utilizing the same SOFA score cutoff of 8. A positive association was found between the delta SOFA ([Δ] SOFA) from day 1 to day 3 (SF1–SF3) and in-hospital mortality (chi-square for linear trend, p < 0.001). Subjects with a ΔSOFA of ≥2 points had an exponential mortality rate to 50%. Similar association was—observed between ΔSOFA of ≥2 and—longer duration of inotropic support (p = 0.0006) with correlation co-efficient 0.2 (95% CI: 0.15–0.35; p = 0.01). Among children admitted to the PICU with septic shock, SOFA scores on both days 1 and 3, have a greater discriminative power for predicting in-hospital mortality than either PRISM III score (within 24 hours of admission) or PELOD-2 score (days 1 and 3). An increase in ΔSOFA of >2 adds additional prognostic accuracy in determining not only mortality risk but also duration of inotropic support as well.

scholarly journals A Modified Sequential Organ Failure Assessment Score for Critical Care Triage

2010 ◽  
Vol 4 (4) ◽  
pp. 277-284 ◽  
Author(s):  
Colin K. Grissom ◽  
Samuel M. Brown ◽  
Kathryn G. Kuttler ◽  
Jonathan P. Boltax ◽  
Jason Jones ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Mechanical Ventilation ◽  
Intensive Care ◽  
Confidence Interval ◽  
Sequential Organ Failure Assessment ◽  
Organ Failure ◽  
Sofa Score ◽  
Laboratory Measurement ◽  
Receiver Operating Curve ◽  
Failure Assessment

ABSTRACTObjective: The Sequential Organ Failure Assessment (SOFA) score has been recommended for triage during a mass influx of critically ill patients, but it requires laboratory measurement of 4 parameters, which may be impractical with constrained resources. We hypothesized that a modified SOFA (MSOFA) score that requires only 1 laboratory measurement would predict patient outcome as effectively as the SOFA score.Methods: After a retrospective derivation in a prospective observational study in a 24-bed medical, surgical, and trauma intensive care unit, we determined serial SOFA and MSOFA scores on all patients admitted during the 2008 calendar year and compared the ability to predict mortality and the need for mechanical ventilation.Results: A total of 1770 patients (56% male patients) with a 30-day mortality of 10.5% were included in the study. Day 1 SOFA and MSOFA scores performed equally well at predicting mortality with an area under the receiver operating curve (AUC) of 0.83 (95% confidence interval 0.81-.85) and 0.84 (95% confidence interval 0.82-.85), respectively (P = .33 for comparison). Day 3 SOFA and MSOFA predicted mortality for the 828 patients remaining in the intensive care unit with an AUC of 0.78 and 0.79, respectively. Day 5 scores performed less well at predicting mortality. Day 1 SOFA and MSOFA predicted the need for mechanical ventilation on day 3, with an AUC of 0.83 and 0.82, respectively. Mortality for the highest category of SOFA and MSOFA score (>11 points) was 53% and 58%, respectively.Conclusions: The MSOFA predicts mortality as well as the SOFA and is easier to implement in resource-constrained settings, but using either score as a triage tool would exclude many patients who would otherwise survive.(Disaster Med Public Health Preparedness. 2010;4:277-284)

Sign in / Sign up

Export Citation Format

Share Document