Dynamic Predictive Models for Hospital Readmission Risk
DOI:
https://doi.org/10.63503/j.ijcma.2024.26Keywords:
Hospital readmission, predictive models, logistic regression, neural networks, machine learning, healthcare analytics, patient outcomesAbstract
Healthcare readmission rates in hospitals are sensitive indicators of quality and clients’ outcomes. Proper models for prediction are crucial in helping avoid cases of unnecessary readmission and the correct allocation of re-sources. In this paper, dynamic predictive models for risk evaluation of hospital readmission are developed, and machine learning strategies are applied to enhance the model's accuracy. In this paper, two types of models are considered and built using logistic regression and a neural network model. Incident data sources include clinical and demographic measures from a hospital database and accuracy, precision, recall, and F1- score are used to assess the performance of the models. Comparing the predictive accuracy of the results, the neural network model performs better than the selected logistic regression model when identifying high-risk patients for readmission. This paper also presents a comparative analysis depending on the model parameters and data divisions, which shows that the neural network can more flexibly approximate nonlinear trends in the data distribution. This paper ends with a discourse on applying high-dependent predictive models to decreasing hospital readmission rates in clinical practice. The results of the research are directed to the need for proper risk assessment to enhance the results of the treatments and the quality of the healthcare services.
References
Bonetti AF, Reis WC, Mendes AM, et al. Impact of pharmacist-led discharge counseling on hospital readmission and emergency department visits: a systematic review and meta-analysis. J Hosp Med. 2020;15:52–59.
Watson DE, Marashi-Pour S, Tran B, Witchard A. Patient-reported experiences and outcomes following hospital care are asso-ciated with risk of readmission among adults with chronic health conditions. PLoS One. 2022;17, e0276812.
Bureau of Health Information. Technical Supplement – Adult Admitted Patient Survey 2019. Sydney (NSW): BHI; 2020.
Sugiura A, Kavsur R, Spieker M, et al. Recurrent mitral regurgitation after MitraClip: predictive factors, Morphology, and clinical Implication. Circ Cardiovasc Interv. 2022; 15(3):e010895.
MIMIC-IV, a freely accessible electronic health record dataset. Scientific Data 10(1), 1 (2023). https://doi.org/10.1038/s41597-022-01899-x
Allam, A., Nagy, M., Thoma, G., Krauthammer, M.: Neural networks versus Logistic regression for 30 days all-cause readmis-sion prediction. Sci. Rep. 9, 9277 (2019). https://doi.org/10.1038/s41598-019-45685-z
DeLong, E.R., DeLong, D.M., Clarke-Pearson, D.L.: Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics, pp. 837–845 (1988)
Gabeur, V., Sun, C., Alahari, K., Schmid, C.: Multi-modal transformer for video retrieval. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12349, pp. 214–229. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58548-8_13
Hamilton, W., Ying, Z., Leskovec, J.: Inductive representation learning on large graphs. Advances in neural information processing systems 30 (2017)
He, K., Fan, H., Wu, Y., Xie, S., Girshick, R.: Momentum contrast for unsupervised visual representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9729–9738 (2020)
Huang, K., Altosaar, J., Ranganath, R.: Clinicalbert: Modeling clinical notes and predicting hospital readmission. arXiv preprint arXiv:1904.05342 (2019)
Johnson, A., P.T.H.S.C.L.A.M.R.: Mimic-iv-note: Deidentified free-text clinical notes (version 2.2). PhysioNet (2023). https://doi.org/10.13026/1n74-ne17
Li, L.H., Yatskar, M., Yin, D., Hsieh, C.J., Chang, K.W.: Visualbert: A simple and performant baseline for vision and language. arXiv preprint arXiv:1908.03557 (2019)
Li, Y., et al.: Behrt: transformer for electronic health records. Sci. Rep. 10(1), 7155 (2020)
Nagrani, A., Yang, S., Arnab, A., Jansen, A., Schmid, C., Sun, C.: Attention bottlenecks for multimodal fusion. Adv. Neural. Inf. Process. Syst. 34, 14200–14213 (2021)
