Puranti, Naomi Puspita Happy (2019) Analisis Propensity Score Stratification Menggunakan Klasifikasi Pohon Pada Penderita Diabetes Melitus Tipe 2. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211540000062_Undergraduate_Theses.pdf - Published Version Download (1MB) | Preview |
Abstract
Diabetes Melitus adalah penyakit kronis, yang disebabkan karena pankreas tidak menghasilkan cukup insulin. Laporan dari Kementerian Kesehatan atau Kemenkes (2013) menunjukan data untuk provinsi Jawa Timur menyumbang persentase prevalensi penderita DM sebesar 2,1% yang hampir 80% prevalensi kejadian DM adalah tipe 2. Penelitian ini akan menggunakan objek penelitian pasien DM tipe 2 di Rumah Sakit Umum Daerah (RSUD) kabupaten Pasuruan, Jawa Timur. Faktor-faktor yang diamati adalah Usia, Jenis Kelamin, Riwayat Genetik, Lama Menderita, Obesitas, Hipertensi dan Aktivitas Olahraga. Untuk mengatasi masalah bias akibat observasi yang tidak random dengan propensity score, didapatkan variabel confounding adalah aktivitas olahraga. Jenis propensity score yang digunakan adalah propensity score stratification dan didapati berdasarkan klasifikasi pohon terbentuk strata sejumlah 4, semua kovariat setelah dilakukan stratifikasi adalah balance, namun variabel yang dijadikan terminal node nilainya akan konstan dan tidak dapat dilakukan pengujian balance covariate pada variabel yang bersangkutan tersebut. Kinerja propensity score dievaluasi dengan Average Treatment Effect (ATE) dan didapati efek perlakuan Z terhdap Y adalah sebesar 0,573. Pendekatan PSS dalam penelitian ini berhasil mereduksi bias (PBR) sebesar 28,04%.
==============================================================================================================================
Diabetes mellitus is a chronic disease, which is caused by the pancreas not producing enough insulin. Reports from the Ministry of Health (2013) that data for East Java province accounted for the percentage prevalence of DM patients at 2.1%, which is almost 80%. The prevalence of DM is type 2. This study will use the object of research in DM type 2 patients at the Regional General Hospital (RSUD) in Pasuruan, East Java. The factors observed were Age, Gender, Genetic History, Duration of Suffering, Obesity, Hypertension and Sports Activities. The confounding variable was carried out with a method to overcome the problem of bias due to non-random observations is propensity score. It was found that the confounding variable was Sports Activity. The type of propensity score used is Propensity Score Stratification (PSS) and found that tree classification form 4 strata for the data, all covariates after doing this stratification are balanced, but the variables which used as terminal nodes are constant and cannot be tested for the balance of the covariates concerned. The performance of the propensity score stratification was evaluated by Average Treatment Effect (ATE) and found the effect of Z on Y was 0.573. The PSS approach in this study managed to reduce bias (PBR) by 28.04%.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.536 Pur a-1 2019 |
| Uncontrolled Keywords: | Diabetes Melitus tipe 2, Variabel Confounding, PSS, Klasifikasi Pohon, Uji Balance Covariate, ATE, PBR, Confounding Variable, Tree Classification |
| Subjects: | Q Science Q Science > QA Mathematics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics) |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Naomi Puspita Happy Puranti |
| Date Deposited: | 02 Jan 2023 08:38 |
| Last Modified: | 02 Jan 2023 08:38 |
| URI: | http://repository.its.ac.id/id/eprint/63417 |
Actions (login required)
 |
View Item |