Application of Cox Proportional Hazards Modeling for Length of Stay in Anemia Patients

Anemia is a health problem that can increase the risk of clinical complications and prolong the length of hospitalization. Length of hospitalization is an important indicator in evaluating the quality of hospital services and the efficiency of health resource use. This study aims to identify clinical, hematological, and demographic variables that influence the length of hospitalization of anemia patients using the Cox Proportional Hazards model. The data used were 1,098 observations of anemia patients' hospitalizations at RSUD Haji Surabaya in 2025. The research stages included data preprocessing, exploratory analysis using the Kaplan-Meier curve and the log-rank test, and Cox Proportional Hazards modeling which included variable selection, testing the proportional hazards
assumption, and model evaluation. The results showed that age, gender, and hemoglobin levels significantly influenced the length of hospitalization of anemia patients with Hazard Ratio values of 0.993, 0.815, and 0.947, respectively (p-value < 0.05). The final model met the proportional hazards assumption based on the Schoenfeld residual test and produced a log-likelihood value of -6022.375, an AIC of 12050.749, and a C-index of 0.565, indicating that the model has limited discriminatory ability in differentiating risks between patients. Overall, the Cox Proportional Hazards model can be used to identify factors that influence the length of hospitalization of anemia patients and support data-driven clinical analysis.

Keywords: Anemia, Survival Analysis, Cox Proportional Hazards, Length of Stay (LOS)

Copyright (c) 2026 Journal of Dinda : Data Science, Information Technology, and Data Analytics

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.