--- title: "MIMIC-III, a freely accessible critical care — Johnson 2016" description: "Hospitals have moved quickly to electronic records, but systems rarely interoperate." authors: ["Alistair E. W. Johnson", "Tom Pollard", "Lu Shen", "Li-wei H. Lehman", "Mengling Feng", "Mohammad M. Ghassemi", "Benjamin Moody", "Peter Szolovits", "Leo Anthony Celi", "Roger G. Mark"] year: 2016 venue: "Scientific Data" source: "https://doi.org/10.1038/sdata.2016.35" doi: "10.1038/sdata.2016.35" licence: "CC BY 4.0 (publisher, via OpenAlex)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/johnson-2016-mimic-iii-freely-accessible-critical/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/johnson-2016-mimic-iii-freely-accessible-critical/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2016](https://md.research.elamur.ai/free/en/papers/2016/) / MIMIC-III, a freely accessible critical care database # MIMIC-III, a freely accessible critical care database **Authors:** [Alistair E. W. Johnson](https://md.research.elamur.ai/free/en/authors/alistair-e-w-johnson/), [Tom Pollard](https://md.research.elamur.ai/free/en/authors/tom-pollard/), [Lu Shen](https://md.research.elamur.ai/free/en/authors/lu-shen/), [Li-wei H. Lehman](https://md.research.elamur.ai/free/en/authors/li-wei-h-lehman/), Mengling Feng, Mohammad M. Ghassemi, [Benjamin Moody](https://md.research.elamur.ai/free/en/authors/benjamin-moody/), [Peter Szolovits](https://md.research.elamur.ai/free/en/authors/peter-szolovits/), [Leo Anthony Celi](https://md.research.elamur.ai/free/en/authors/leo-anthony-celi/), Roger G. Mark · **Year:** 2016 · **Venue:** Scientific Data · **Licence:** [CC BY 4.0 (publisher, via OpenAlex)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://doi.org/10.1038/sdata.2016.35) · [PDF](https://www.nature.com/articles/sdata201635.pdf) · [DOI 10.1038/sdata.2016.35](https://doi.org/10.1038/sdata.2016.35) · [OpenAlex](https://openalex.org/W2396881363) **Citations:** 7999 citations (OpenAlex, 2026-10-14) · [531](https://md.research.elamur.ai/free/en/papers/johnson-2016-mimic-iii-freely-accessible-critical/cited-by/) in our library, of the 140884 papers whose bibliographies we read ## Summary Hospitals have moved quickly to electronic records, but systems rarely interoperate. As a result, clinical data seldom become usable for research, and closed datasets make studies hard to reproduce. This data descriptor presents MIMIC-III (Medical Information Mart for Intensive Care), the successor to MIMIC-II. The MIT Laboratory for Computational Physiology built it from routine care records at Beth Israel Deaconess Medical Center in Boston. The team gathered data from three kinds of sources: - two ICU information systems, Philips CareVue and iMDsoft MetaVision; - the hospital's electronic health record databases; - the Social Security death file. They deidentified the records and organized them into relational tables. These cover vital signs, lab results, medications, fluid balance, billing codes, and free-text notes and reports. The result is a critical care resource spanning over ten years. Any qualified researcher can obtain it after completing training and signing a data use agreement. Its value is that it supports reproducible clinical studies, teaching and method development at a scale otherwise unavailable. This includes NLP on clinical notes. A shared community code repository supports this work. ## Key points - **Scale:** 38,597 adult patients, 49,785 hospital admissions and 53,423 ICU stays (2001–2012), plus 7,870 neonates (2001–2008). Adults have a median age of 65.8 years, 55.9% are male, and in-hospital mortality is 11.5%. - **Detail per admission:** each admission averages about 4,579 charted observations (CHARTEVENTS) and 380 lab measurements (LABEVENTS). Free text is stored in NOTEEVENTS. Fluid intake data could not be merged across the two ICU systems, so it is kept in separate tables (INPUTEVENTS_CV and INPUTEVENTS_MV). - **Access:** the data come as CSV files with import scripts for PostgreSQL, MySQL and MonetDB. Downloads go through PhysioNetWorks after the researcher completes a human-subjects course that covers HIPAA and signs a data use agreement. Approval takes at least a week. - **Community:** the GitHub repository MIT-LCP/mimic-code provides code for severity scores, comorbidity scores, and ventilation and vasopressor durations. The project also runs datathons and supports mapping to OMOP-CDM. MIMIC is already used in courses at Stanford, MIT, Georgia Tech, UT Austin and Columbia. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Other](https://md.research.elamur.ai/free/en/shelves/other/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{johnson2016mimic, title = {{MIMIC-III, a freely accessible critical care database}}, author = {Johnson, Alistair E. W. and Pollard, Tom and Shen, Lu and Lehman, Li-wei H. and Feng, Mengling and Ghassemi, Mohammad M. and Moody, Benjamin and Szolovits, Peter and Celi, Leo Anthony and Mark, Roger G.}, booktitle = {Scientific Data}, year = {2016}, doi = {10.1038/sdata.2016.35}, url = {https://doi.org/10.1038/sdata.2016.35} } ``` - [← previous Semi-Supervised Classification with Graph Convolutional Networks](https://md.research.elamur.ai/free/en/papers/kipf-2016-semi-supervised-classification-graph-convolutional/) - [next → Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://md.research.elamur.ai/free/en/papers/selvaraju-2016-grad-cam-visual-explanations-deep/) - [↑ back to papers of 2016](https://md.research.elamur.ai/free/en/papers/2016/) ## References 14 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Charles, D., King, J., Patel, V. & Furukawa, M. Adoption of Electronic Health record Systems among U.S. Non-federal Acute Care Hospitals. ONC Data Brief No. 9, 1–9 (2013). [find](https://scholar.google.com/scholar?q=Charles%2C+D%2E%2C+King%2C+J%2E%2C+Patel%2C+V%2E+%26+Furukawa%2C+M%2E+Adoption+of+Electronic+Health+record+Systems+among+U%2ES%2E+Non%2Dfederal+Acute+Care+Hospitals%2E+ONC+Data+Brief+No%2E+9%2C+1%E2%80%939+%282013%29%2E) 2. Collins, F. S. & Tabak, L. A. NIH plans to enhance reproducibility. Nature 505, 612–613 (2014). [find](https://scholar.google.com/scholar?q=Collins%2C+F%2E+S%2E+%26+Tabak%2C+L%2E+A%2E+NIH+plans+to+enhance+reproducibility%2E+Nature+505%2C+612%E2%80%93613+%282014%29%2E) 3. Saeed, M. et al. Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II): A public-access intensive care unit database. Critical Care Medicine 39, 952–960 (2011). [find](https://scholar.google.com/scholar?q=Saeed%2C+M%2E+et+al%2E+Multiparameter+Intelligent+Monitoring+in+Intensive+Care+II+%28MIMIC%2DII%29%3A+A+public%2Daccess+intensive+care+unit+database%2E+Critical+Care+Medicine+39%2C+952%E2%80%93960+%282011%29%2E) 4. Abhyankar, S., Demner-Fushman, D. & McDonald, C. J. Standardizing clinical laboratory data for secondary use. J Biomed Inform 45, 642–650 (2012). [find](https://scholar.google.com/scholar?q=Abhyankar%2C+S%2E%2C+Demner%2DFushman%2C+D%2E+%26+McDonald%2C+C%2E+J%2E+Standardizing+clinical+laboratory+data+for+secondary+use%2E+J+Biomed+Inform+45%2C+642%E2%80%93650+%282012%29%2E) 5. Observational Medical Outcomes Partnership Common Data Model. Website [http://www.ohdsi.org/data-standardization/the-common-data-model/(Accessed:](http://www.ohdsi.org/data-standardization/the-common-data-model/(Accessed:) March 2016). [↗](http://www.ohdsi.org/data-standardization/the-common-data-model/(Accessed:) 6. Neamatullah, I. et al. Automated de-identification of free-text medical records. BMC Medical Informatics and Decision Making 8, 1–32 (2008). [find](https://scholar.google.com/scholar?q=Neamatullah%2C+I%2E+et+al%2E+Automated+de%2Didentification+of+free%2Dtext+medical+records%2E+BMC+Medical+Informatics+and+Decision+Making+8%2C+1%E2%80%9332+%282008%29%2E) 7. Wilson, G. et al. Best practices for scientific computing. PLOS Biology 12, e1001745 (2014). [find](https://scholar.google.com/scholar?q=Wilson%2C+G%2E+et+al%2E+Best+practices+for+scientific+computing%2E+PLOS+Biology+12%2C+e1001745+%282014%29%2E) 8. MIMIC-III Critical Care Database: Documentation and Website [http://mimic.physionet.org](http://mimic.physionet.org) (Accessed: March 2016). [↗](http://mimic.physionet.org) 9. Goldberger, A. L. et al. PhysioBank, PhysioToolkit, and PhysioNet. Circulation 101, e215–e220 (2000). [find](https://scholar.google.com/scholar?q=Goldberger%2C+A%2E+L%2E+et+al%2E+PhysioBank%2C+PhysioToolkit%2C+and+PhysioNet%2E+Circulation+101%2C+e215%E2%80%93e220+%282000%29%2E) 10. Mayaud, L. et al. Dynamic data during hypotensive episode improves mortality predictions among patients with sepsis and hypotension. Critical Care Medicine 41, 954–962 (2014). [find](https://scholar.google.com/scholar?q=Mayaud%2C+L%2E+et+al%2E+Dynamic+data+during+hypotensive+episode+improves+mortality+predictions+among+patients+with+sepsis+and+hypotension%2E+Critical+Care+Medicine+41%2C+954%E2%80%93962+%282014%29%2E) 11. Lehman, L. H., Saeed, M., Talmor, D., Mark, R. G. & Malhotra, A. Methods of Blood Pressure Measurement in the ICU. Critical Care Medicine 41, 34–40 (2013). [find](https://scholar.google.com/scholar?q=Lehman%2C+L%2E+H%2E%2C+Saeed%2C+M%2E%2C+Talmor%2C+D%2E%2C+Mark%2C+R%2E+G%2E+%26+Malhotra%2C+A%2E+Methods+of+Blood+Pressure+Measurement+in+the+ICU%2E+Critical+Care+Medicine+41%2C+34%E2%80%9340+%282013%29%2E) 12. Velupillai, S., Mowery, D., South, B. R., Kvist, M. & Dalianis, H. Recent Advances in Clinical Natural Language Processing in Support of Semantic Analysis. Yearbook of Medical Informatics 10, 183–193 (2015). [find](https://scholar.google.com/scholar?q=Velupillai%2C+S%2E%2C+Mowery%2C+D%2E%2C+South%2C+B%2E+R%2E%2C+Kvist%2C+M%2E+%26+Dalianis%2C+H%2E+Recent+Advances+in+Clinical+Natural+Language+Processing+in+Support+of+Semantic+Analysis%2E+Yearbook+of+Medical+Informatics+10%2C+183%E2%80%93193+) 13. Abhyankar, S., Demner-Fushman, D., Callaghan, F. M. & McDonald, C. J. Combining structured and unstructured data to identify a cohort of ICU patients who received dialysis. J. Am. Med. Inform. Assoc. 21, 801–807 (2014). [find](https://scholar.google.com/scholar?q=Abhyankar%2C+S%2E%2C+Demner%2DFushman%2C+D%2E%2C+Callaghan%2C+F%2E+M%2E+%26+McDonald%2C+C%2E+J%2E+Combining+structured+and+unstructured+data+to+identify+a+cohort+of+ICU+patients+who+received+dialysis%2E+J%2E+Am%2E+Med%2E+Inform%2E+Assoc%2E+) 14. Aboab, J. et al. A ‘datathon’ model to support cross-disciplinary collaboration. Science Translational Medicine 8, 333–ps8 (2016). Data Citation 1. Pollard, T. J. & Johnson, A. E. W. The MIMIC-III Clinical Database http://dx.doi.org/10.13026/C2XW26 (2016). \#\# Acknowledgements This research and devel… — [doi:10.13026/c2xw26](https://doi.org/10.13026/c2xw26) [↗](http://dx.doi.org/10.13026/C2XW26)