MIMIC-IV, a freely accessible electronic health record dataset

Johnson, Alistair E. W.; Bulgarelli, Lucas; Shen, Lu; Gayles, Alvin; Shammout, Ayad; Horng, Steven; Pollard, Tom; Hao, Sicheng; Moody, Benjamin; Gow, Brian J.; Lehman, Li-wei H.; Celi, Leo Anthony; Mark, Roger G.

doi:10.1038/s41597-022-01899-x

Cited by 579 publications

(456 citation statements)

References 21 publications

Supporting

Mentioning

318

Contrasting

Order By: Relevance

“…Data were collected from two large open critical care databases, the Medical Information Mart for Intensive Care III (MIMIC III) and MIMIC IV. MIMIC IV, which includes ICU patients treated at Beth Israel Deaconess Medical Center from 2008 to 2019, was employed to examine the relationship between LAR at admission and mortality in ICU patients with AKI [ 17 ]. MIMIC III, which contains data from 2001 to 2012, was carried out to validate the results [ 17 ].…”

Section: Methodsmentioning

confidence: 99%

“…MIMIC IV, which includes ICU patients treated at Beth Israel Deaconess Medical Center from 2008 to 2019, was employed to examine the relationship between LAR at admission and mortality in ICU patients with AKI [ 17 ]. MIMIC III, which contains data from 2001 to 2012, was carried out to validate the results [ 17 ]. The authors were entitled to extract data from the two databases after completion of the course "Protecting Human Research Participants".…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Prognostic implication of lactic dehydrogenase-to-albumin ratio in critically ill patients with acute kidney injury

Deng

Lai

et al. 2023

Clin Exp Nephrol

View full text Add to dashboard Cite

Background No studies have been published on the correlation between lactic dehydrogenase-to-albumin ratio (LAR) and poor prognosis of acute kidney injury (AKI) patients, warranting further research. This analysis sought to investigate the prognostic implication of LAR in critically ill patients with AKI. Methods The present study enrolled 11,046 and 5180 adults with AKI from the Medical Information Mart for Intensive Care III (MIMIC III) and MIMIC IV, respectively. Data from MIMIC IV were identified as the training cohort, and those from MIMIC III were identified as the validation cohort. We applied multivariate regression analysis to identify the link between LAR and all-cause mortality. Restricted cubic spline (RCS) was conducted to figure out the correlation between LAR and in-hospital mortality. Furthermore, we carried out stratification analyses to examine if the effects of LAR on in-hospital mortality were consistent across various subclasses. Results The level of LAR was remarkably higher in the in-hospital non-survivor group ( p < 0.001). Furthermore, the increased LAR group presented a remarkably higher rate of in-hospital mortality at AKI stages 1, 2, and 3 compared with the decreased LAR group (all p < 0.001). Multivariate regression analyses exhibited the independent prognostic significance of LAR for all-cause mortality (all p < 0.001). MIMIC III observed concordant results. RCS indicated a non-linear correlation between LAR and in-hospital death ( P for non-linearity < 0.001). The relationship between LAR and in-hospital mortality was still significant in patients with various subclasses. Conclusions Elevated LAR at admission is a prognostic risk factor for critically ill patients with AKI. Supplementary Information The online version contains supplementary material available at 10.1007/s10157-023-02321-5.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Prognostic implication of lactic dehydrogenase-to-albumin ratio in critically ill patients with acute kidney injury

Deng

Lai

et al. 2023

Clin Exp Nephrol

View full text Add to dashboard Cite

show abstract

“…The data in this retrospective study were extracted from the MIMIC IV (Johnson et al, 2021), which is an extensive, freelyavailable database comprising de-identified health-related data from patients who were admitted to a tertiary academic medical center in Boston, MA, USA. It contains information between 2008-2019 for each patient while they were in the hospital: laboratory measurements, medications administered, vital signs documented, and so on.…”

Section: Data Sourcementioning

confidence: 99%

Protective efficacy of statins in patients with Klebsiella pneumoniae bloodstream infection

Zheng

Shen

et al. 2023

Front. Cell. Infect. Microbiol.

View full text Add to dashboard Cite

BackgroundPatients with bloodstream infection of Klebsiella pneumoniae (BSI-KP) have a high risk of death and septic shock. This study aims to identify the risk factors for mortality and severity in patients of BSI-KP.MethodsData of BSI-KP patients were extracted from the MIMIC IV (Medical Information Mart for Intensive Care IV) database, and patients infected with only K. pneumoniae in blood were included in this study. The risk factors of 28-day mortality and septic shock in BSI-KP patients were analyzed, respectively.ResultsA total of 279 patients enrolled and the all-cause 28-day mortality rate was 11.8%. The use of statins (OR 0.220, 95% CI 0.060-0.801, p = 0.022) and quinolones (OR 0.356, 95% CI 0.143-0.887, p = 0.027) were both independent protective factors for death within 28 days, while the use of vasoactive drugs (OR 7.377, 95% CI 1.775-30.651, p = 0.006) was a risk factor. Besides, pulmonary disease (OR 2.348, 95% CI 1.126-4.897, p = 0.023), bleeding and coagulation disorders (OR 3.626, 95% CI 1.783-7.372, p < 0.001), respiratory failure (OR 2.823, 95% CI 0.178-6.767, p = 0.020) and kidney dysfunction (OR 2.450, 95% CI 1.189-5.047, p = 0.015) were independent risk factors for patients suffered from septic shock while hypertension was a protective one. The receiver operating characteristic (ROC) curves could well predict the risk of death within 28-day (area under ROC = 0.855, 95% CI = 0.796–0.914, p < 0.001) and septic shock (AUROC = 0.815, 95% CI = 0.755–0.874, p < 0.001) in patients with BSI-KP.ConclusionThe use of statins could decrease the risk of 28-day mortality in patients of BSI-KP. The risk factor-based prediction model provided evidence for drug treatment in BSI-KP patients. Paying more attention to the strategy of drug treatment will be an optimal way to improve patient’s outcome in clinical practice.

show abstract

“…All subjects' data were extracted from two large critical-care medical databases, which are free and open to researchers from all over the world. Data of the training cohort were collected from the Multiparameter Intelligent Monitoring in Intensive Care IV (MIMIC-IV; version 2.0) database (18,19), which is jointly developed and run by the Massachusetts Institute of Technology (Cambridge, MA, USA) and the Beth Israel Deaconess Medical Center (BIDM, Boston, MA, USA). This database contains the detailed and comprehensive clinical data of about 250,000 patients admitted to the intensive care unit (ICU) and emergency unit of the Beth Israel Deaconess Medical Center from 2008 to 2019, including demography, laboratory tests, documented vital signs, medications administered, and so on.…”

Section: Data Sourcesmentioning

confidence: 99%

Association between prothrombin time-international normalized ratio and prognosis of post-cardiac arrest patients: A retrospective cohort study

Tang

Sun²,

Yu³

et al. 2023

Front. Public Health

View full text Add to dashboard Cite

BackgroundCardiac arrest (CA) can activate blood coagulation. This study aimed to explore the potential prognostic value of prothrombin time–international normalized ratio (INR) in post-CA patients.MethodsThe clinical data of eligible subjects diagnosed with CA was extracted from the MIMIC-IV database as the training cohort. Restricted cubic spline (RCS), Kaplan–Meier (K-M) survival curve, and Cox regression analyses were conducted to elucidate the association between the INR and all-cause mortality of post-CA patients. Subgroup analysis, propensity score matching (PSM), and inverse probability of treatment (IPTW) were also conducted to improve stability and reliability. Data of the validation cohort were collected from the eICU database, and logistic-regression analyses were performed to verify the findings of the training cohort.ResultsA total of 1,324 subjects were included in the training cohort. A linear correlation existed between INR and the risk of all-cause death of post-CA patients, as shown in RCS analysis, with a hazard ratio (HR) >1 when INR exceeded 1.2. K-M survival curve preliminarily indicated that subjects with INR ≥ 1.2 presented lower survival rate and shorter survival time, and the high level of INR was independently associated with 30-day, 90-day, 1-year, and in-hospital mortalities, with multivariate-adjusted HR of 1.44 (1.20, 1.73), 1.46 (1.23, 1.74), 1.44 (1.23, 1.69), and 1.37 (1.14, 1.64), respectively. These findings were consistent and robust across the subgroup analysis, PSM and IPTW analyses, and validation cohort.ConclusionsWe systematically and comprehensively demonstrated that elevated INR was associated with increased short- and long-term all-cause mortality of post-CA patients. Therefore, elevated INR may be a promising biomarker with prognosis significance.

show abstract

MIMIC-IV, a freely accessible electronic health record dataset

Cited by 579 publications

References 21 publications

Prognostic implication of lactic dehydrogenase-to-albumin ratio in critically ill patients with acute kidney injury

Prognostic implication of lactic dehydrogenase-to-albumin ratio in critically ill patients with acute kidney injury

Protective efficacy of statins in patients with Klebsiella pneumoniae bloodstream infection

Association between prothrombin time-international normalized ratio and prognosis of post-cardiac arrest patients: A retrospective cohort study

Contact Info

Product

Resources

About