A case study in hospitals: Comorbidity modeling on Covid-19 mortality risk using logistic regression
1 Department of Medical records and health information systems, Dian Nuswantoro University Semarang Indonesia.
2 Department of Environmental Health Dian Nuswantoro University, Semarang Indonesia.
Review
International Journal of Scientific Research Updates, 2022, 03(01), 001–009.
Article DOI: 10.53430/ijsru.2022.3.1.0043
Publication history:
Received on 23 November 2021; revised on 28 December 2021; accepted on 30 December 2021
Abstract:
Background: Comorbidities may have a role in Covid-19 patients' deaths. The goal of this research was to determine the impact of comorbidities on the risk of death from COVID-19.
Methods: The research took place at one of the Covid-19 patients' referral hospitals, from March 2020 to February 2021, data on inpatients was collected, and a logistic regression analysis was used to find co-morbid COVID-19 in patients at high risk of death.
Result: Model 1 predicts the risk of death from covid-19 based on age, sex, and the top ten co-morbid covid-19 with an accuracy of 85.3% and an R2 of 62.4%. With a p-value>0.05, sex (SEX), hypertension (HP), diabetes mellitus (DM), cardiac arrest (CA), anemia (AN), hyperuricemia (HR), and urinary tract infection (UTR) do not affect the probability of death in Covid-19 patients in model 1. Model 2 is based on the variables that affect the risk of death (p-value 0.05) are age (AGE), moderate malnutrition (MM), unspecified malnutrition (MU), obesity (OB), and chronic renal disease (CKD). Model 2 is Ln [ þ / (1-þ) ] = -3,295 +0,046*AGE +0,667*OB +0,611*MM +1,53*MU +1,248*CKD showed an accuracy of 69.4% with an R2 of 17.1%.
Conclusion: Model 2 is more effective since it only considers diseases with high mortality risk. The results of the model 2 simulation show that the higher the age and comorbid complications, the greater the risk of death for Covid-19 patients.
Keywords:
Comorbidity modeling; Mortality risk; Comorbid; Logistic regression; Covid-19; Logistic Regression
Full text article in PDF:
Copyright information:
Copyright © 2022 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0