Annals of Operations Research

, Volume 263, Issue 1–2, pp 429–451 | Cite as

Classifying readmissions to a cardiac intensive care unit

  • Yazan F. Roumani
  • Yaman Roumani
  • Joseph K. Nwankpa
  • Mohan Tanniru
Data Mining and Analytics


Research has associated intensive care unit (ICU) readmissions with increased risk of morbidity and mortality. Readmitted patients are also exposed to complications as they are transferred between hospital units. Moreover, due to their unexpected nature, readmissions increase ICU costs and the complexity of managing ICUs. Existing studies on ICU readmissions have mainly used logistic regression for identifying patients who are more likely to be readmitted. However, such studies do not account for the imbalanced nature of the data where the class of interest (readmitted patients) is the minority group. This paper empirically compares three approaches for handling the imbalanced ICU readmissions data: misclassification cost ratio, synthetic minority oversampling technique (SMOTE), and random under-sampling. We used three classification techniques for identifying patients who are more likely to be readmitted to the ICU within the same hospital stay: support vector machines, C5.0, and logistic regression. We evaluated the classification performance of the three methods using recall, specificity, accuracy, F-measure, G-mean, confusion entropy, and area under the receiver operating characteristic curve. Our results showed that SMOTE is the best approach for addressing the imbalanced nature of the data. The sensitivity analysis identified prolonged ventilation, renal failure, and pneumonia as the top three predictors of ICU readmissions. Our findings can be used to develop a decision support tool to help ICU clinicians and administrators in identifying patients who are more likely to be readmitted and hence provide the patients with the appropriate care to minimize their risk of readmission.


Imbalanced data Data mining Intensive care unit Readmission 



This research was partially supported by a 2015 Oakland University School of Business Administration Spring/Summer Research Fellowship.


  1. Basti, E., Kuzey, C., & Delen, D. (2015). Analyzing initial public offerings’ short-term performance using decision trees and SVMs. Decision Support Systems, 73, 15–27.CrossRefGoogle Scholar
  2. Beckmann, U., Gillies, D. M., Berenholtz, S. M., Wu, A. W., & Pronovost, P. (2004). Incidents relating to the intra-hospital transfer of critically ill patients. Intensive Care Medicine, 30(8), 1579–1585.CrossRefGoogle Scholar
  3. Braxton, C. C., Reilly, P. M., & Schwab, C. W. (2000). The traveling intensive care unit patient: Road trips. Surgical Clinics of North America, 80(3), 949–956.CrossRefGoogle Scholar
  4. Bujlow, T., Riaz, T., & Pedersen, J. M. (2012). A method for classification of network traffic based on C5. 0 machine learning algorithm. In 2012 international conference on computing, networking and communications (ICNC), (pp. 237–241).Google Scholar
  5. Campbell, A. J., Cook, J. A., Adey, G., & Cuthberston, B. H. (2008). Predicting death and readmission after intensive care discharge. British Journal of Anesthesia, 100(5), 656–662.CrossRefGoogle Scholar
  6. Chang, C. C., & Lin, C. J. (2011). LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology (TIST), 2(3), 27.Google Scholar
  7. Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357.Google Scholar
  8. Chawla, N. V., Japkowicz, N., & Kolcz, A. (2004). Editorial: Special issue on learning from imbalanced data sets. SIGKDD Explor, 6(1), 1–6.CrossRefGoogle Scholar
  9. Cooper, G. S., Sirio, C. A., Rotondi, A. J., Shepardson, L. B., & Rosenthal, G. E. (1999). Are readmissions to the intensive care unit a useful measure of hospital performance? Medical care, 37(4), 399–408.CrossRefGoogle Scholar
  10. Elliott, M., Worral-Carter, L., & Page, K. (2014). Intensive care readmission: A contemporary review of the literature. Intensive and Critical Care Nursing, 30(3), 121–137.CrossRefGoogle Scholar
  11. Fakhry, S. M., Leon, S., Derderian, C., Al-Harakeh, H., & Ferguson, P. L. (2013). Intensive care unit bounce back in trauma patients: An analysis of unplanned returns to the intensive care unit. Journal of Trauma and Acute Care Surgery, 74(6), 1528–1533.CrossRefGoogle Scholar
  12. Fan, R.-E., Chang, K.-W., Hsieh, C.-J., Wang, X.-R., & Lin, C.-J. (2008). LIBLINEAR: A library for large linear classification. The Journal of Machine Learning Research, 9, 1871–1874.Google Scholar
  13. Fialho, A. S., Kaymak, U., Cismondi, F., Vieira, S. M., Reti, S. R., Sousa, J. M., & Finkelstein, S. N. (2013). Predicting intensive care unit readmissions using probabilistic fuzzy systems. In 2013 IEEE international conference on fuzzy systems (FUZZ), , (pp. 1–7).Google Scholar
  14. Fialho, A. S., Cismondi, F., Vieira, S. M., Reti, S. R., Sousa, J. M., & Finkelstein, S. N. (2012). Data mining using clinical physiology at discharge to predict ICU readmissions. Expert Systems with Applications, 39(18), 13158–13165.CrossRefGoogle Scholar
  15. Garcia, M. N. M., Robledo, J. G., Gonzalez, F. M., Hernandez, F. S., & Barba, M. S. (2014). Machine learning methods for mortality prediction of polytraumatized patients in intensive care units—Dealing with imbalanced and high-dimensional data. In E. Corchado, J. A. Lozano, H. Quintián & H. Yin (Eds.), Intelligent data engineering and automated learning–IDEAL 2014 (pp. 309–317). Springer International Publishing.Google Scholar
  16. Gorunescu, F. (2013). Data mining: Concepts, models and techniques. Berlin: Springer.Google Scholar
  17. Gruenberg, D. A., Shelton, W., Rose, S. L., Rutter, A. E., Socaris, S., & McGee, G. (2006). Factors influencing length of stay in the intensive care unit. American Journal of Critical Care, 15(5), 502–509.Google Scholar
  18. Hastie, T., Tibshirani, R., Friedman, J., Hastie, T., & Friedman, J. (2009). The elements of statistical learning, (Vol. 2, No. 1). New York: Springer.CrossRefGoogle Scholar
  19. He, H., & Garcia, E. A. (2009). Learning from imbalanced data. IEEE Transactions on Knowledge and Data Engineering, 21(9), 1263–1284.CrossRefGoogle Scholar
  20. Japkwoicz, N., & Stephen, S. (2002). The class imbalance problem: A systematic study. Intelligent Data Analysis, 6(5), 429–449.Google Scholar
  21. Jarden, R. J., & Quirke, S. (2010). Improving safety and documentation in intrahospital transport: Development of an intrahospital transport tool for critically ill patients. Intensive and Critical Care Nursing, 26(2), 101–107.CrossRefGoogle Scholar
  22. Kantardzic, M. (2011). Data mining: Concepts, models, methods, and algorithms. New York: Wiley.CrossRefGoogle Scholar
  23. Kleinbaum, D., Kupper, L., Nizam, A., & Rosenberg, E. (2013). Applied regression analysis and other multivariable methods (5th ed.). Boston: Cengage Learning.Google Scholar
  24. Lemnaru, C., & Potolea, R. (2012). Imbalanced classification problems: Systematic study, issues and best practices. In R. Zhang, J. Zhang, Z. Zhang, J. Filipe & J. Cordeiro (Eds.), Enterprise information systems (pp. 35–50). Berlin: Springer.Google Scholar
  25. Litmathe, J., Kurt, M., Feindt, P., Gams, E., & Boeken, U. (2009). Predictors and outcome of ICU readmission after cardiac surgery. The Thoracic and Cardiovascular Surgeon, 57(07), 391–394.CrossRefGoogle Scholar
  26. Liu, Y., Loh, H. T., & Sun, A. (2009). Imbalanced text classification: A term weighting approach. Expert systems with Applications, 36(1), 690–701.CrossRefGoogle Scholar
  27. López, V., Fernández, A., Moreno-Torres, J. G., & Herrera, F. (2012). Analysis of preprocessing vs. cost-sensitive learning for imbalanced classification. Open problems on intrinsic data characteristics. Expert Systems with Applications, 39(7), 6585–6608.CrossRefGoogle Scholar
  28. Ouanes, I., Schwebel, C., Français, A., Bruel, C., Philippart, F., Vesin, A., et al. (2012). A model to predict short-term death or readmission after intensive care unit discharge. Journal of Critical Care, 27(4), 10.CrossRefGoogle Scholar
  29. Oztekin, A., Delen, D., Turkyilmaz, A., & Zaim, S. (2013). A machine learning-based usability evaluation method for eLearning systems. Decision Support Systems, 56, 63–73.CrossRefGoogle Scholar
  30. Pronovost, P. J., Needham, D. M., Waters, H., Birkmeyer, C. M., Calinawan, J. R., Birkmeyer, J. D., et al. (2004). Intensive care unit physician staffing: Financial modeling of the Leapfrog standard. Critical Care Medicine, 32(6), 1247–1253.CrossRefGoogle Scholar
  31. Renton, J., Pilcher, D. V., Santamaria, J. D., Stow, P., Bailey, M., Hart, G., et al. (2011). Factors associated with increased risk of readmission to intensive care in Australia. Intensive Care Medicine, 37(11), 1800–1808.CrossRefGoogle Scholar
  32. Roumani, Y. F., May, J. H., Strum, D. P., & Vargas, L. G. (2013). Classifying highly imbalanced ICU data. Healthcare Management Science, 16(2), 119–128.CrossRefGoogle Scholar
  33. Schorr, C. A. (2012). Fishing for answers to avoid intensive care unit readmissions: Are we reeling in a “catch 22”? Critical Care Medicine, 40(1), 295–296.CrossRefGoogle Scholar
  34. Seref, O., Razzaghi, T., & Xanthopoulos, P. (2014). Weighted relaxed support vector machines. Annals of Operations Research, 1, 1–37.Google Scholar
  35. Sevim, C., Oztekin, A., Bali, O., Gumus, S., & Guresen, E. (2014). Developing an early warning system to predict currency crises. European Journal of Operational Research, 237(3), 1095–1104.CrossRefGoogle Scholar
  36. Shrivastava, H., Huddar, V., Bhattacharya, S., & Rajan, V. (2015). Classification with imbalance: A similarity-based method for predicting respiratory failure. In 2015 IEEE international conference on bioinformatics and biomedicine (BIBM), (pp. 707–714).Google Scholar
  37. Thongkam, J., Xu, G., Zhang, Y., & Huang, F. (2009). Toward breast cancer survivability prediction models through improving training space. Expert Systems with Applications, 36(10), 12200–12209.CrossRefGoogle Scholar
  38. Ting, K. M. (2002). An instance-weighting method to induce cost-sensitive trees. IEEE Transactions on Knowledge and Data Engineering, 14(3), 659–665.CrossRefGoogle Scholar
  39. Visa, S., & Ralescu, A. (2005). Issues in mining imbalanced data sets—A review paper. Proceedings of the sixteenth midwest artificial intelligence and cognitive science conference, (pp. 67–73).Google Scholar
  40. Voigt, L. P., Pastores, S. M., Raoof, N. D., Thaler, H. T., & Halpern, N. A. (2009). Review of a large clinical series: Intrahospital transport of critically ill patients: Outcomes, timing, and patterns. Journal of Intensive Care Medicine, 24(2), 108–115.CrossRefGoogle Scholar
  41. Wang, S., Jian, W., & Tsui, K.-L. (2010). Adjusted support vector machines based on a new loss function. Annals of Operations Research, 174(1), 83–101.CrossRefGoogle Scholar
  42. Wei, J. M., Yuan, X. J., Hu, Q. H., & Wang, S. Q. (2010). A novel measure for evaluating classifiers. Expert Systems with Applications, 37(5), 3799–3809.CrossRefGoogle Scholar
  43. Wu, X., Kumar, V., Quinlan, J. R., Ghosh, J., Yang, Q., Motoda, H., et al. (2008). Top 10 algorithms in data mining. Knowledge and Information Systems, 14(1), 1–37.CrossRefGoogle Scholar
  44. Xanthopoulos, P., Pardalos, P., & Trafalis, T. B. (2012). Robust data mining. New York: Springer.Google Scholar
  45. Yen, S., & Lee, Y. (2009). Cluster-based under-sampling approaches for imbalanced data distributions. Expert Systems with Applications, 36(3), 5718–5727.CrossRefGoogle Scholar
  46. Zolfaghar, K., Verbiest, N., Agarwal, J., Meadem, N., Chin, S. C., Roy, S. B., & Reed, L. (2013). Predicting risk-of-readmission for congestive heart failure patients: A multi-layer approach. Preprint.Google Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  • Yazan F. Roumani
    • 1
  • Yaman Roumani
    • 2
  • Joseph K. Nwankpa
    • 3
  • Mohan Tanniru
    • 4
  1. 1.Department of Decision and Information SciencesOakland UniversityRochesterUSA
  2. 2.Department of Computer Information SystemsEastern Michigan UniversityYpsilantiUSA
  3. 3.Department of Information SystemsUniversity of Texas Rio Grande ValleyEdinburgUSA
  4. 4.Department of Decision and Information SciencesOakland UniversityRochesterUSA

Personalised recommendations