Rachman, Zelika Anindita (2026) Analisis Survival Kejadian Berulang Pada Pasien Kanker Serviks Di Rumah Sakit Universitas Airlangga Menggunakan Model Prentice-Williams-Peterson Total Time. Other thesis, Institut Teknologi Sepuluh Nopember.
|
Text
5003221144-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy |
Abstract
Kanker serviks merupakan salah satu keganasan dengan angka kejadian tinggi pada perempuan, dan penanganannya kerap dihadapkan pada persoalan kekambuhan yang dapat terjadi lebih dari satu kali pada pasien yang sama. Karakteristik data yang bersifat kejadian berulang tersebut tidak dapat dianalisis secara tepat menggunakan analisis survival satu kejadian, sehingga diperlukan pendekatan yang mampu mengakomodasi urutan kejadian. Penelitian ini bertujuan mengkaji karakteristik dan pola kejadian berulang serta mengidentifikasi faktor-faktor yang memengaruhi intensitas kekambuhan pasien kanker serviks menggunakan model Prentice-Williams-Peterson Total Time (PWP-TT). Data yang digunakan merupakan data sekunder rekam medis 53 pasien kanker serviks di Rumah Sakit Universitas Airlangga yang dibentuk dalam struktur counting process dan dibatasi hingga kekambuhan kedua, sehingga terbentuk 106 interval pengamatan dengan 64 kejadian. Tujuh variabel prediktor yang dianalisis adalah usia, stadium, operasi, kemoterapi, radioterapi, komorbid, dan komplikasi. Pola kejadian berulang digambarkan melalui Mean Cumulative Function (MCF), sedangkan parameter model diestimasi dengan maximum partial likelihood estimation disertai robust sandwich variance per pasien, dan model terbaik dipilih menggunakan backward elimination berdasarkan AIC. Hasil analisis menunjukkan nilai MCF keseluruhan sebesar 1,682, yang berarti secara rata-rata setiap pasien mengalami sekitar 1,7 kali kekambuhan. Model terbaik memuat variabel stadium, operasi, dan radioterapi. Pasien stadium lanjut memiliki risiko kekambuhan 1,872 kali lebih tinggi dan bersifat konsisten pada kedua urutan kejadian. Operasi dan radioterapi bersifat protektif pada kekambuhan pertama dengan hazard ratio berturut-turut 0,422 dan 0,177, namun berbalik arah pada kekambuhan kedua menjadi 3,562 dan 2,926. Temuan ini menunjukkan bahwa pengaruh terapi terhadap kekambuhan bergantung pada urutan kejadian, sehingga model PWP-TT bermanfaat untuk mendukung stratifikasi risiko dan pemantauan pascaterapi pasien kanker serviks.
=================================================================================================================================
Cervical cancer is one of the most prevalent malignancies among women, and its management is frequently challenged by recurrences that may occur more than once in the same patient. Such recurrent-event data cannot be properly analyzed using single-event survival analysis, thus requiring an approach that accounts for the event ordering. This study aims to examine the characteristics and patterns of recurrent events and to identify the factors influencing the recurrence intensity of cervical cancer patients using the Prentice-Williams-Peterson Total Time (PWP-TT) model. The data were secondary medical records of 53 cervical cancer patients at Universitas Airlangga Hospital, structured as a counting process and restricted to the second recurrence, resulting in 106 observation intervals with 64 events. Seven predictors were analyzed: age, stage, surgery, chemotherapy, radiotherapy, comorbidity, and complication. The recurrence pattern was described using the Mean Cumulative Function (MCF), while the model parameters were estimated through maximum partial likelihood estimation with a per-patient robust sandwich variance, and the best model was selected via backward elimination based on AIC. The results show an overall MCF of 1.682, indicating that on average each patient experienced about 1.7 recurrences. The best model contained stage, surgery, and radiotherapy. Advanced-stage patients had a 1.872 times higher recurrence risk, consistent across both event orders. Surgery and radiotherapy were protective at the first recurrence with hazard ratios of 0.422 and 0.177, respectively, but reversed at the second recurrence to 3.562 and 2.926. These findings indicate that the effect of therapy on recurrence depends on the event order, so that the PWP-TT model is useful for supporting risk stratification and post-treatment monitoring of cervical cancer patients.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Analisis Survival, Kanker Serviks, Kejadian Berulang, Mean Cumulative Function, PWP-TT, Cervical Cancer, Mean Cumulative Function, PWP-TT, Recurrent Event, Survival Analysis |
| Subjects: | Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Zelika Anindita Rachman |
| Date Deposited: | 05 Aug 2026 02:40 |
| Last Modified: | 05 Aug 2026 02:40 |
| URI: | http://repository.its.ac.id/id/eprint/143838 |
Actions (login required)
![]() |
View Item |
