Using electronic health record collected clinical variables to predict medical intensive care unit mortality

Jacob Calvert, Qingqing Mao, Jana L Hoffman, Melissa Jay, Thomas Desautels, Hamid Mohamadlou, Uli Chettipally, Ritankar Das, Jacob Calvert, Qingqing Mao, Jana L Hoffman, Melissa Jay, Thomas Desautels, Hamid Mohamadlou, Uli Chettipally, Ritankar Das

Abstract

Background: Clinical decision support systems are used to help predict patient stability and mortality in the Intensive Care Unit (ICU). Accurate patient information can assist clinicians with patient management and in allocating finite resources. However, systems currently in common use have limited predictive value in the clinical setting. The increasing availability of Electronic Health Records (EHR) provides an opportunity to use medical information for more accurate patient stability and mortality prediction in the ICU.

Objective: Develop and evaluate an algorithm which more accurately predicts patient mortality in the ICU, using the correlations between widely available clinical variables from the EHR.

Methods: We have developed an algorithm, AutoTriage, which uses eight common clinical variables from the EHR to assign patient mortality risk scores. Each clinical variable produces a subscore, and combinations of two or three discretized clinical variables also produce subscores. A combination of weighted subscores produces the overall score. We validated the performance of this algorithm in a retrospective study on the MIMIC III medical ICU dataset.

Results: AutoTriage 12 h mortality prediction yields an Area Under Receiver Operating Characteristic value of 0.88 (95% confidence interval 0.86 to 0.88). At a sensitivity of 80%, AutoTriage maintains a specificity of 81% with a diagnostic odds ratio of 16.26.

Conclusions: Through the multidimensional analysis of the correlations between eight common clinical variables, AutoTriage provides an improvement in the specificity and sensitivity of patient mortality prediction over existing prediction methods.

Keywords: Clinical decision support systems; Electronic health records; Medical informatics; Mortality prediction.

Figures

Fig. 1
Fig. 1
Patient inclusion flowchart.
Fig. 2
Fig. 2
Receiver Operating Characteristic (ROC) curves for 12-h mortality prediction in the Medical Intensive Care Unit for AutoTriage, Modified Early Warning Score (MEWS), Sequential Organ Failure Assessment (SOFA), and Simplified Acute Physiology Score (SAPS II). A MEWS of at least 3 has a specificity of 74% and a sensitivity of 66%, whereas an AutoTriage threshold of −2 at a similar specificity of 81% has a sensitivity of 80%.
Fig. 3
Fig. 3
Patient distribution across AutoTriage score for survivors and non-survivors. The vertical line represents an AutoTriage score of −2.
Fig. 4
Fig. 4
Distribution of consecutive hours of threshold breach prior to death for AutoTriage ≥ −2 in black, and Modified Early Warning Score (MEWS) ≥ 3 in red.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5
Fig. 5
Area under receiver operating characteristic for AutoTriage as a function of time preceding in-hospital death in the Medical Intensive Care Unit.

References

    1. Karandeep S., Wright A. Clinical decision support. In: Finnell J.T., Dixon B.E., editors. Clinical Informatics Study Guide. Springer International Publishing; 2016. pp. 111–133. ISBN:978-3-319-22752-8 (Print) 978-3-319-22753-5 (Online)
    1. Siontis G.C., Tzoulaki I., Ioannidis J.P. Predicting death: an empirical evaluation of predictive tools for mortality. Arch. Intern. Med. 2011;171(19):1721–1726. PMID:21788535.
    1. Reini K., Fredrikson M., Oscarsson A. The prognostic value of the Modified Early Warning Score in critically ill patients: a prospective, observational study. Eur. J. Anaesthesiol. 2012;29(3):152–157. PMID:22273831.
    1. Khazaei H., McGregor C., Eklund J.M., El-Khatib K. Real-time and retrospective health-analytics-as-a-service: a novel framework. JMIR Med. Inf. 2015;3(4) PMID:26582268.
    1. Subbe C., Slater A., Menon D., Gemmell L. Validation of physiological scoring systems in the accident and emergency department. Emerg. Med. J. 2006;23(11):841–845. PMID:17057134.
    1. Vincent J., Moreno R., Takala J., Willatts S., De Mendonça A., Bruining H., Reinhart C.K., Suter P.M., Thijs L.G. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Med. 1996;22(7):707–710. PMID:8844239.
    1. Le Gall J.R., Lemeshow S., Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA. 1993;270(24):2957–2963. PMID:8254858.
    1. Celi LA, Zimolzak AJ, Stone DJ. Dynamic clinical data mining: search engine-based decision support. JMIR Med. Inf., 2(1), e13. PMID:25600664.
    1. Lehman L.W., Adams R.P., Mayaud L., Moody G.B., Malhotra A., Mark R.G., Nemati S. A physiological time series dynamics-based approach to patient monitoring and outcome prediction. IEEE J. Biomed. Health Inf. 2015;19(3):1068–1076. PMID:25014976.
    1. Ghose S, Mitra J, Khanna S, Dowling J. An Improved Patient-Specific Mortality Risk Prediction in ICU in a Random Forest Classification Framework. Driving Reform: Digital health is Everyone's business: Selected Papers from the 23rd Australian National Health Informatics Conference (HIC 2015). IOS Press. PMID:26210418.
    1. Pirracchio R., Petersen M.L., Carone M., Rigon M.R., Chevret S., van der Laan M.J. Mortality prediction in intensive care units with the Super ICU Learner Algorithm (SICULA): a population-based study. Lancet Respir. Med. 2015;3(1):42–52. PMID:25466337.
    1. Moskovitch R., Walsh C., Wang F., Hripcsak G., Tatonetti N. IEEE International Conference on Data Mining (ICDM) 2015. Outcomes prediction via time intervals related patterns; pp. 919–924.
    1. Johnson A.E., Pollard T.J., Shen L., Lehman L.W., Feng M., Ghassemi M., Moody B., Szolovits P., Celi L.A., Mark R.G. MIMIC-III, a freely accessible critical care database. Sci. Data. 2016 May 24;3:160035. PubMed PMID: 27219127; PubMed Central PMCID: PMC4878278.
    1. Calvert J.S., Price D.A., Barton C.W., Chettipally U.K., Das R. Discharge recommendation based on a novel technique of homeostatic analysis. J. Am. Med. Inf. Assoc. 2016:ocw014. PMID: 27026611.
    1. Calvert J.S., Mao Q., Rogers A.J., Barton C., Jay M., Desautels T., Mohamadlou H., Jan J., Das R. A computational approach to mortality prediction of alcohol use disorder inpatients. Comput. Biol. Med. 2016 Aug 1;75:74–79. PMID: 27253619.
    1. . Archived at: .
    1. Rapsang A.G., Shyam D.C. Scoring systems in the intensive care unit: a compendium. Indian J. Crit. Care Med. Peer-reviewed, Off. Publ. Indian Soc. Crit. Care Med. 2014;18(4):220–228. PMC. Web. 3 Aug. 2016. PMID: 24872651.
    1. Sekulic A.D., Trpkovic S.V., Pavlovic A.P., Marinkovic O.M., Ilic A.N. Scoring systems in assessing survival of critically ill ICU patients. Med. Sci. Monit. Int. Med. J. Exp. Clin. Res. 2015;21:2621. PMID: 26336861.
    1. Calvert J.S., Price D.A., Chettipally U.K., Barton C.W., Feldman M.D., Hoffman J.L. A computational approach to early sepsis detection. Comput. Biol. Med. 2016;74:69–73. PMID: 27208704.
    1. Calvert J., Desautels T., Chettipally U., Barton C., Hoffman J., Jay M., Mao Q., Mohamadlou H., Das R. High-performance detection and early prediction of septic shock for alcohol-use disorder patients. Ann. Med. Surg. 2016 May 10;8:50–55. eCollection 2016 Jun. PubMed PMID: 27489621; PubMed Central PMCID: PMC4960347.
    1. Roshanov P.S., Fernandes N., Wilczynski J.M., Hemens B.J., You J.J., Handler S.M., Nieuwlaat R., Souza N.M., Beyene J., Van Spall H.G., Garg A.X., Haynes R.B. Features of effective computerised clinical decision support systems: meta-regression of 162 randomised trials. BMJ. 2013;346:f657. PMID:23412440.

Source: PubMed

Sponsorit ja yhteistyökumppanit

Sairaudet

Huumeiden interventiot

3
Tilaa