Development of a predictive model for in-hospital mortality using logistic regression: A retrospective study of 16,632 records from a university hospital
DOI:
https://doi.org/10.33448/rsd-v14i9.49466Keywords:
Hospital Mortality, Logistic Models, Epidemiological Monitoring.Abstract
Introduction: Hospital mortality is an indicator used to measure the effectiveness of healthcare delivery, and its prediction may support regional strategies and interventions. Objective: To develop a predictive model of hospital mortality based on data available from the Hospital Information System of the Brazilian Unified Health System. Method: A retrospective cohort study was conducted with 16,632 hospitalizations from a regional university hospital in Paraná, Brazil, between May 2022 and May 2024. Secondary data were obtained from Hospitalization Reports and Patient Death Reports. Predictive variables were identified using a multivariate logistic regression model, adjusted through the stepwise backward elimination method. Results: Two predictive models were developed: the first adjusted for all analyzed variables, and the second adjusted for interactions between the variables "Hospital Ward" and "Length of Stay." Both models identified the following as predictors of in-hospital death: age groups of 60 to 79 years and 80 years or older, admission to the ICU, and a length of stay of 15 days or more. Conclusion: Two predictive models of hospital mortality were developed, which may contribute to improving clinical and care management.
References
Aguiar, L. M. M., Martins, G. de S., Valduga, R., Gerez, A. P., Carmo, E. C. do, Cunha, K. da C., Cipriano, G. F. B., & Silva, M. L. da. (2022). Perfil de unidades de terapia intensiva adulto no Brasil: revisão sistemática de estudos observacionais. Revista Brasileira de Terapia Intensiva, 33, 624–634. https://doi.org/10.5935/0103-507X.20210088
Aragon, T. J., Fay, M. P., Wollschlaeger, D., & Omidpanah, A. (2020). Package ‘epitools’. https://doi.org/10.32614/CRAN.package.epitools
Blanes-Selva, V., Ruiz-García, V., Tortajada, S., Benedí, J.-M., Valdivieso, B., & García-Gómez, J. M. (2021). Design of 1-year mortality forecast at hospital admission: A machine learning approach. Health Informatics Journal, 27(1). https://doi.org/10.1177/1460458220987580
Brasil. (2022). Ministério da Saúde. Portaria GM/MS nº 913, de 22 de abril de 2022. Declara o encerramento da Emergência em Saúde Pública de Importância Nacional (ESPIN) em decorrência da infecção humana pelo novo coronavírus (2019-nCoV) e revoga a Portaria GM/MS nº 188, de 3 de fevereiro de 2020. Diário Oficial da União. 2022; Seção 1:1.
Calazans, J. A., Guimarães, R., & Nepomuceno, M. R. (2023). Diferenciais regionais da mortalidade no Brasil: Contribuição dos grupos etários e de causas de óbito sobre a variação da esperança de vida e da dispersão da idade à morte entre 2008 e 2018. Revista Brasileira de Estudos de População, 40, e0244. https://doi.org/10.20947/S0102-3098a0244
Chen, Z., Li, T., Guo, S., Zeng, D., & Wang, K. (2023). Machine learning-based in-hospital mortality risk prediction tool for intensive care unit patients with heart failure. Frontiers in Cardiovascular Medicine, 10. https://doi.org/10.3389/fcvm.2023.1119699
Cordeiro, P., & Martins, M. (2018). Mortalidade hospitalar em pacientes idosos no Sistema Único de Saúde, região Sudeste. Revista de Saúde Pública, 52, 69. https://doi.org/10.11606/s1518-8787.2018052000146
Costa, M.L.M., Mafra, A. C. C. N., Cendoroglo, M. S., Rodrigues, P. S., Ferreira, M. S., Studenski, S. A., & Franco, F. G. de M. (2022). Development and validation of predictive model for long-term hospitalization, readmission, and in-hospital death of patients over 60 years old. Einstein, 20:eAO8012. https://doi.org/10.31744/einstein_journal/2022AO8012
Cowling, T. E., Cromwell, D. A., Bellot, A., Sharples, L. D., & van der Meulen, J. (2021). Logistic regression and machine learning predicted patient mortality from large sets of diagnosis codes comparably. Journal of Clinical Epidemiology, 133, 43–52. https://doi.org/10.1016/j.jclinepi.2020.12.018
Cowling, T. E., Cromwell, D. A., Sharples, L. D., & van der Meulen, J. (2020). A novel approach selected small sets of diagnosis codes with high prediction performance in large healthcare datasets. J Clin Epidemiol, 128:20-28. https://doi.org/10.1016/j.jclinepi.2020.08.001
Ding, N., Nath, T., Mahendra Damarla, Gao, L., & Hassoun, P. M. (2024). Early predictive values of clinical assessments for ARDS mortality: a machine-learning approach. Scientific Reports, 14(1). https://doi.org/10.1038/s41598-024-68653-8
Dutt-Ross, S. (2020). Manual de Análise de Dados. https://livro.metodosquantitativos.com/docs/bookdown.pdf
Faro, A. R. M. C., Andrade, A. M., Guimarães, F. P. F., Belasco, A. G. S., Grandi, J. L., & Barbosa, D. A. (2021). Causas de óbitos em hospital de ensino da Amazônia Ocidental brasileira. Acta paul enferm, 34:eAPE002515. https://doi.org/10.37689/acta-ape/2021AO002515
Geskey, J. M., Foreman, J. K., Witkowski, M. E., Huerta, S. M., Berkland, D., Hohmann, S. F., & Meurer, S. (2022). Improving Mortality Through a Multihospital, Collaborative Quality Improvement Project. American Journal of Medical Quality, 37(4), 321–326. https://doi.org/10.1097/jmq.0000000000000039
Hydoub, Y. M., Walker, A. P., Kirchoff, R. W., Alzu’bi, H. M., Chipi, P. Y., Gerberi, D. J., M. Caroline Burton, M. Hassan Murad, & Dugani, S. B. (2023). Risk Prediction Models for Hospital Mortality in General Medical Patients: A Systematic Review. American Journal of Medicine Open, 10, 100044–100044. https://doi.org/10.1016/j.ajmo.2023.100044
Klug, M., Barash, Y., Bechler, S., Resheff, Y. S., Tron, T., Ironi, A., Soffer, S., Zimlichman, E., & Klang, E. (2019). A Gradient Boosting Machine Learning Model for Predicting Early Mortality in the Emergency Department Triage: Devising a Nine-Point Triage Score. Journal of General Internal Medicine, 35(1), 220–227. https://doi.org/10.1007/s11606-019-05512-7
Li, K., Shi, Q., Liu, S., Xie, Y., & Liu, J. (2021). Predicting in-hospital mortality in ICU patients with sepsis using gradient boosting decision tree. Medicine, 100(19), e25813. https://doi.org/10.1097/md.0000000000025813
Macedo, L. R., Borges, C., Freire, L., Christiane, J., Paiva, N. S., & Joseé, A. (2024). Excesso de mortalidade geral e mortalidade por COVID-19 no Brasil e regiões no ano de 2020. Cadernos de Saúde Pública, 40(11). https://doi.org/10.1590/0102-311xpt217323
Maluangnon, C., Kanogpotjananont, P., & Tongyoo S. (2023). Comparing Outcomes of Critically Ill Patients in Intensive Care Units and General Wards: A Comprehensive Analysis. Int J Gen Med. 16:3779-3787. https://doi.org/10.2147/IJGM.S422791
Okoroiwu, H. U., Uchendu, K. I., & Essien, R. A. (2020). Causes of morbidity and mortality among patients admitted in a tertiary hospital in southern Nigeria: A 6 year evaluation. PLOS ONE, 15(8), e0237313. https://doi.org/10.1371/journal.pone.0237313
Pereira, A. S. et al. (2018). Metodologia da pesquisa científica. Santa Maria. Editora da UFSM.
Rai, S., Brace, C. L., Ross, P., Jai Darvall, Haines, K., Mitchell, I., Frank Van Haren, & Pilcher, D. (2023). Characteristics and Outcomes of Very Elderly Patients Admitted to Intensive Care: A Retrospective Multicenter Cohort Analysis. Critical Care Medicine, 51(10):1328-1338. https://doi.org/10.1097/ccm.0000000000005943
Rodrigues, A. R., Oliveira, A., Vieira, T., Assis, R., Lume, C., Gonçalves-Pereira, J., & Fernandes SM. (2024). A prolonged intensive care unit stay defines a worse long-term prognosis Insights from the critically ill mortality by age (Cimba) study. Aust Crit Care, 37(5):734-739. https://doi.org/10.1016/j.aucc.2024.03.001
Schilling, M. P. R., Portela, M. C., & Martins, M. (2024). Razão de mortalidade hospitalar padronizada: limites e potencialidades do indicador para a avaliação do desempenho hospitalar no Sistema Único de Saúde, Brasil. Cadernos de Saúde Pública, 40(2). https://doi.org/10.1590/0102-311XPT080723
Seki, T., Kawazoe, Y., & Ohe, K. (2021). Machine learning-based prediction of in-hospital mortality using admission laboratory data: A retrospective, single-site study using electronic health record data. PLOS ONE. 16(2):e0246640. https://doi.org/10.1371/journal.pone.0246640
Shitsuka, R.etal. (2014). Matemática fundamental para tecnologia. (2ed). Editora Érica.
Stoessel, D., Fa, R., Artemova, S., von Schenck, U., Nowparast Rostami, H., Madiot, P.-E., Landelle, C., Olive, F., Foote, A., Moreau-Gaudry, A., & Bosson, J.-L. (2023). Early prediction of in-hospital mortality utilizing multivariate predictive modelling of electronic medical records and socio-determinants of health of the first day of hospitalization. BMC Medical Informatics and Decision Making, 23(1). https://doi.org/10.1186/s12911-023-02356-4
TRIPOD. (2015). TRIPOD Checklist: Prediction Model Developmenth. https://www.tripod-statement.org/resources/
van Valburg, M. K. (2024). Predicting 30-day mortality in intensive care unit patients with ischaemic stroke or intracerebral haemorrhage. Eur J Anaesthesiol, 41(2):136-145. https://doi.org/10.1097/EJA.0000000000001920
Walicka, M., Chlebus, M., Śliwczyński, A., Brzozowska, M., Rutkowski, D., Czech, M., Tuszyńska, A., Jacyna, A., Puzianowska-Kuźnicka, M., & Franek, E. (2020). Non-surgical in-hospital mortality predictors: a multivariable regression analysis of 2,855,029 hospitalizations. Polish Archives of Internal Medicine. https://doi.org/10.20452/pamw.15185
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Adriele da Mota de França Pinto, Camila Marinelli Martins
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
1) Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
2) Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
3) Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
