Analisis Survival Kejadian Berulang Strok Menggunakan Pendekatan Counting Process (Studi Kasus: Pasien di RSUD Haji Provinsi Jawa Timur)

Fadila, Defara Herena Fitri (2025) Analisis Survival Kejadian Berulang Strok Menggunakan Pendekatan Counting Process (Studi Kasus: Pasien di RSUD Haji Provinsi Jawa Timur). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003211060-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB)

Abstract

Strok merupakan penyebab disabilitas nomor satu dan penyebab kematian utama di Indonesia, dengan prevalensi tinggi mencapai 10,9 per 1.000 penduduk. Seorang pasien yang berhasil melalui serangan strok pertama diketahui mengalami peningkatan risiko untuk mengalami kejadian strok berikutnya jika dibandingkan dengan populasi umum. Kejadian berulang strok memiliki dampak lebih berat dibandingkan serangan pertama, baik dari segi tingkat keparahan gejala maupun kecacatan sehingga meningkatkan beban ekonomi dan sosial. Pencegahan kejadian berulang strok memerlukan pemahaman mendalam terhadap faktor risiko yang dapat dimodifikasi seperti hipertensi, diabetes, dan gaya hidup tidak sehat. Penelitian ini bertujuan untuk memahami pola waktu kejadian berulang strok di RSUD Haji Provinsi Jawa Timur dengan menggunakan model regresi Cox proportional hazard yang dikombinasikan dengan pendekatan counting process. Model ini memungkinkan analisis kejadian berulang sebagai kejadian independen sehingga memberikan informasi rinci mengenai waktu terjadinya kejadian berulang strok. Hasil penelitian menunjukkan bahwa sebanyak 89,02% pasien strok di RSUD Haji Provinsi Jawa Timur mengalami serangan strok berulang sebanyak satu kali. Sebagian besar pasien berusia sekitar 61 tahun, dengan tekanan darah diastolik sekitar 89 mm/Hg dan tekanan darah sistolik sekitar 150 mm/Hg. Mayoritas pasien adalah laki-laki, tipe strok adalah tipe iskemik, tidak memiliki riwayat diabetes melitus, dan bukan merupakan perokok. Median waktu survival pasien adalah 528 hari. Kaplan-Meier dan uji log-rank menunjukkan bahwa variabel usia, tekanan darah diastolik, tekanan darah sistolik, jenis kelamin, tipe strok, status diabetes melitus, dan status merokok tidak memiliki perbedaan yang signifikan dalam kurva survival antar kategorinya. Model regresi Cox proportional hazard dengan pendekatan counting process menunjukkan bahwa tidak terdapat pengaruh yang signifikan dari variabel-variabel prediktor terhadap waktu pasien mengalami serangan strok berulang, baik secara serentak maupun secara parsial pada masing-masing variabel prediktor. Setelah dilakukan pemilihan model terbaik melalui eliminasi backward, didapatkan variabel yang berpengaruh signifikan terhadap waktu pasien mengalami strok berulang di RSUD Haji Provinsi Jawa Timur adalah status diabetes melitus.
=========================================================================================================================================
Stroke is the leading cause of disability and a major cause of death in Indonesia with a high prevalence of 10.9 per 1000 population. A patient who survives an initial stroke is known to have an increased risk of experiencing a recurrent stroke compared to the general population. Recurrent stroke events have more severe impacts than the first occurrence, both in terms of symptom severity and disability that increasing economic and social burdens. Preventing recurrent stroke requires a deep understanding of modifiable risk factors such as hypertension, diabetes, and unhealthy lifestyles. This study aims to understand the time patterns of recurrent stroke events at RSUD Haji East Java using the Cox proportional hazard regression model combined with the counting process approach. This model allows the analysis of recurrent events as independent occurrences, providing detailed information on the timing of recurrent strokes. The results showed that 89.02% of stroke patients at RSUD Haji East Java Province experienced a recurrent stroke once. Most patients were around 61 years old, with a diastolic blood pressure of approximately 88 mm/Hg and a systolic blood pressure of around 150 mm/Hg. The majority of patients were male, had ischemic-type strokes, did not have a history of diabetes mellitus, and were non-smokers. The median survival time was 528 days. Kaplan-Meier curves and log-rank tests indicated that variables such as age, diastolic blood pressure, systolic blood pressure, gender, stroke type, diabetes mellitus status, and smoking status did not show significant differences in survival curves between categories. The Cox proportional hazards regression model with a counting process approach revealed no significant effect of the independent variables on the time to recurrent stroke, either jointly or partially for each predictor variable. However, after selecting the best model through backward elimination, it was found that the variable with a significant effect on the time to recurrent stroke in RSUD Haji East Java Province was diabetes mellitus status.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Analisis Survival, Pendekatan Counting Process, Regresi Cox Proportional Hazard, RSUD Haji Jawa Timur, Strok Berulang, Counting Process Approach, Cox Proportional Hazard Regression, Recurrent Strokes, RSUD Haji East Jawa, Survival Analysis
Subjects:	Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Defara Herena Fitri Fadila
Date Deposited:	01 Aug 2025 06:12
Last Modified:	01 Aug 2025 06:12
URI:	http://repository.its.ac.id/id/eprint/125972

Actions (login required)

View Item