Abstract
Length of stay (LOS) and risk of readmission of patients are critical indicators of the quality and operation efficiency of hospitals. Various machine learning (ML) approaches have been applied to predict a patient’s hospital LOS and risk of readmission, but those with more accurate predictions are often the so called ‘black-box’ approaches. This study aims to add interpretability in predicting LOS and the risk of readmission in 30-day among all-cause patients admitted through the emergency department (ED) while improving the accuracy and parsimony of the ML approach. Several state-of-the-art ML models were applied to our prediction tasks and their predictive power reported and compared. The CatBoost model outperformed the rest, hence is chosen as the baseline for this study. For interpretability, we introduced Shapley values and analyzed, at both aggregated and individual levels, the prediction results from the CatBoost model. Lower dimension models were further developed following the guidance of Shapley values. Our results show that the lower dimension model can robustly predict hospital LOS and risk of readmission, indicating that Shapley values are not only useful for adding model interpretability, but also effective for creating a lower-dimensional model amenable to implementation.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Morley, C., Unwin, M., Peterson, G.M., Stankovich, J., Kinsman, L.: Emergency department crowding: a systematic review of causes, consequences and solutions. PloS one 13(8), e0203316 (2018)
Jo, S., et al.: Emergency department crowding is associated with 28-day mortality in community-acquired pneumonia patients. J. Infect. 64(3), 268–275 (2012)
Chaou, C.H., et al.: Predicting length of stay among patients discharged from the emergency department-using an accelerated failure time model. PloS one 12(1), e0165756 (2017)
Baek, H., Cho, M., Kim, S., Hwang, H., Song, M., Yoo, S.: Analysis of length of hospital stay using electronic health records: a statistical and data mining approach. PloS one 13(4), e0195901 (2018)
Upadhyay, S., Stephenson, A.L., Smith, D.G.: Readmission rates and their impact on hospital financial performance: a study of Washington hospitals. INQUIRY J. Health Care Organ. Provision Finan. 56, 0046958019860386 (2019)
Authority, N.H.P.: Hospital performance: length of stay in public hospitals in 2011–12 (2013)
CMS: Hospital readmissions reduction program (HRRP). https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/Readmissions-Reduction-Program, Accessed 4 July 2021
Wiens, J., Shenoy, E.S.: Machine learning for healthcare: on the verge of a major shift in healthcare epidemiology. Clin. Infect. Dis. 66(1), 149–153 (2017). https://doi.org/10.1093/cid/cix731
Peck, J.S., Benneyan, J.C., Nightingale, D.J., Gaehde, S.A.: Predicting emergency department inpatient admissions to improve same-day patient flow. Acad. Emerg. Med. 19(9), E1045–E1054 (2012)
Combes, C., Kadri, F., Chaabane, S.: Predicting hospital length of stay using regression models: application to emergency department. In: 10ème Conférence Francophone de Modélisation, Optimisation et Simulation-MOSIM’14 (2014)
Sun, Y., Heng, B.H., Tay, S.Y., Seow, E.: Predicting hospital admissions at emergency department triage using routine administrative data. Acad. Emerg. Med. 18(8), 844–850 (2011)
Leegon, J., Jones, I., Lanaghan, K., Aronsky, D.: Predicting hospital admission for emergency department patients using a bayesian network. In: AMIA Annual Symposium Proceedings, vol. 2005, p. 1022. American Medical Informatics Association (2005)
Hilton, C.B., et al.: Personalized predictions of patient outcomes during and after hospitalization using artificial intelligence. NPJ Dig. Med. 3(1), 1–8 (2020)
Artetxe, A., Beristain, A., Graña, M., Besga, A.: Predicting 30-day emergency readmission risk. In: Graña, M., López-Guede, J.M., Etxaniz, O., Herrero, Á., Quintián, H., Corchado, E. (eds.) SOCO/CISIS/ICEUTE -2016. AISC, vol. 527, pp. 3–12. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-47364-2_1
Baig, M.M., et al.: Machine learning-based risk of hospital readmissions: predicting acute readmissions within 30 days of discharge. In: 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 2178–2181. IEEE (2019)
Morel, D., Kalvin, C.Y., Liu-Ferrara, A., Caceres-Suriel, A.J., Kurtz, S.G., Tabak, Y.P.: Predicting hospital readmission in patients with mental or substance use disorders: a machine learning approach. Int. J. Med. Inf. 139, 104136 (2020)
Roquette, B.P., Nagano, H., Marujo, E.C., Maiorano, A.C.: Prediction of admission in pediatric emergency department with deep neural networks and triage textual data. Neural Netw. 126, 170–177 (2020)
Hong, W.S., Haimovich, A.D., Taylor, R.A.: Predicting hospital admission at emergency department triage using machine learning. PloS one 13(7), e0201016 (2018)
Dorogush, A.V., Ershov, V., Gulin, A.: Catboost: gradient boosting with categorical features support. arXiv preprint arXiv:1810.11363 (2018)
Lundberg, S.M., et al.: From local explanations to global understanding with explainable AI for trees. Nature Mach. Intell. 2(1), 2522–5839 (2020)
Lemaître, G., Nogueira, F., Aridas, C.K.: Imbalanced-learn: a python toolbox to tackle the curse of imbalanced datasets in machine learning. J. Mach. Learn. Res. 18(1), 559–563 (2017)
Authority, I.H.P.: Australian refined diagnosis related groups version 6.x addendum, https://www.ihpa.gov.au/publications/australian-refined-diagnosis-related-groups-version-6x-addendum, Accessed 10 July 2021
Pereira, M., Singh, V., Hon, C.P., McKelvey, T.G., Sushmita, S., De Cock, M.: Predicting future frequent users of emergency departments in California state. In: Proceedings of the 7th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics, pp. 603–610 (2016)
Guyon, I., Elisseeff, A.: An introduction to variable and feature selection. J. Mach. Learn. Res. 3(Mar), 1157–1182 (2003)
Bergstra, J., Yamins, D., Cox, D.D., et al.: Hyperopt: a python library for optimizing the hyperparameters of machine learning algorithms. In: Proceedings of the 12th Python in Science Conference, vol. 13, p. 20. Citeseer (2013)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, Y., Qin, S. (2022). An Interpretable Machine Learning Approach for Predicting Hospital Length of Stay and Readmission. In: Li, B., et al. Advanced Data Mining and Applications. ADMA 2022. Lecture Notes in Computer Science(), vol 13087. Springer, Cham. https://doi.org/10.1007/978-3-030-95405-5_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-95405-5_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-95404-8
Online ISBN: 978-3-030-95405-5
eBook Packages: Computer ScienceComputer Science (R0)