Pemodelan Survival Lama Rawat Inap Pasien Anemia Menggunakan Cox Proportional Hazards

Rabbani, Amandasari Dinda (2026) Pemodelan Survival Lama Rawat Inap Pasien Anemia Menggunakan Cox Proportional Hazards. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 22083010093-Cover.pdf Download (1MB)
	Text (BAB I) 22083010093-BAB I.pdf Download (256kB)
	Text (BAB II) 22083010093-BAB II.pdf Restricted to Repository staff only until 24 September 2029. Download (734kB)
	Text (BAB III) 22083010093-BAB III.pdf Restricted to Repository staff only until 24 September 2029. Download (773kB)
	Text (BAB IV) 22083010093-BAB IV.pdf Restricted to Repository staff only until 24 September 2029. Download (2MB)
	Text (BAB V) 22083010093-BAB V.pdf Download (235kB)
	Text (DAFTAR PUSTAKA) 22083010093-DAFTAR PUSTAKA.pdf Download (188kB)
	Text (DAFTAR LAMPIRAN) 22083010093-DAFTAR LAMPIRAN.pdf Restricted to Repository staff only until 24 September 2029. Download (575kB)

Abstract

Anemia is a health issue that can extend the length of stay (LOS), thereby impacting service quality, hospital resource utilization, and treatment costs. Analyzing LOS requires methods capable of handling time-to-event and censored data. However, research on the LOS of anemia patients using the Cox Proportional Hazards metho specifically incorporating hematological parameters remains limited. Therefore, this study aims to identify variables influencing the LOS of anemia patients, construct a Cox Proportional Hazards model, test the Proportional Hazards assumption, stratify patient risk, and implement the model within a Streamlit-based application. The study utilized medical record data from anemia patients at RSUD Haji Surabaya, following a process that included data preprocessing, Exploratory Data Analysis (EDA), Kaplan–Meier analysis, the Log-Rank test, variable selection, Cox Proportional Hazards model construction, Proportional Hazards assumption testing, model evaluation (using Log-Likelihood, Akaike Information Criterion [AIC], and Concordance Index [C-index]), and patient risk stratification. The results indicate that Age, Gender, and Hemoglobin (Hb) levels significantly influence the LOS of anemia patients, forming a final model that satisfies the Proportional Hazards assumption. The resulting model successfully categorized patients into high, medium, and low-risk groups and was implemented in a Streamlit-based application to serve as a tool for prediction and visualization of analysis results. The findings demonstrate that the developed model supports the analysis of anemia patient LOS and provides risk information that can facilitate data-driven clinical decision-making. It should be emphasized that the resulting model estimates the hazard rate relative to the length of hospital stay, rather than directly predicting the number of days of hospitalization.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Hindrayani, Kartika Maulida	NIDN199209092022032009	kartika.maulida.ds@upnjatim.ac.id
Thesis advisor	Nasrudin, Muhammad	NIDN199609092024061002	nasrudin.fasilkom@upnjatim.ac.id

Subjects:

Q Science > Q Science (General)
Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Amandasari Dinda Rabbani

Date Deposited:

24 Sep 2026 03:11

Last Modified:

24 Sep 2026 03:15

URI:

https://repository.upnjatim.ac.id/id/eprint/60341