Analisis Propensity Score Matching Menggunakan Regresi Logistik Biner Pada Kasus Kejadian Katarak Senilis di RSUD Muhammad Sani Provinsi Kepulauan Riau

Simanjuntak, Debby Septien (2021) Analisis Propensity Score Matching Menggunakan Regresi Logistik Biner Pada Kasus Kejadian Katarak Senilis di RSUD Muhammad Sani Provinsi Kepulauan Riau. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06211940005016-Undergraduate_Thesis.pdf]
Preview
Text
06211940005016-Undergraduate_Thesis.pdf - Accepted Version

Download (4MB) | Preview

Abstract

Katarak adalah kekeruhan pada lensa mata yang menyebabkan gangguan penglihatan dan penyebab utama terjadinya kebutaan. Laporan hasil Riset Kesehatan Dasar menunjukkan prevalensi buta katarak di Provinsi Kepulauan Riau sebesar 1.4%, melebihi batas prevalensi kebutaan oleh WHO sebesar 0.5%. Salah satu jenis penyakit katarak dengan jumlah penderita terbesar adalah katarak senilis. Berdasarkan hasil kunjungan rawat inap pada tahun 2017 di RSUD Muhammad Sani Kepulauan Riau, penyakit katarak senilis merupakan kasus dengan kunjungan tertinggi. Melihat tingginya kasus katarak senilis, dibutuhkan tindakan preventif untuk mengendalikan faktor-faktor penyebab terjadinya katarak senilis. Penelitian ini bertujuan untuk mengetahui kombinasi faktor-faktor penyebab katarak senilis untuk mengatasi bias yang disebabkan oleh variabel confounding menggunakan Propensity Score Matching (PSM) dengan metode estimasi nilai propensity score yaitu regresi logistik biner. Faktor-faktor yang diduga berpengaruh adalah jenis kelamin, usia, pendidikan, pekerjaan yang berhubungan dengan paparan sinar ultraviolet, riwayat diabetes mellitus, hipertensi, dan trauma mata. Hasil analisis balance kovariat menunjukkan usia, pendidikan terakhir dan riwayat hipertensi merupakan variabel yang berpengaruh secara langsung terhadap katarak senilis. Bias yang dapat direduksi dari hasil propensity score matching dengan regresi logistik biner mencapai 98%.
================================================================================================
Cataract is cloudiness in the lens of the eye that can cause visual disturbances and is a major cause of blindness. The report from the results of the Basic Health Research shows that the prevalence of cataract blindness in Riau Islands Province is 1.4%, while the blindness prevalence limit by WHO is only 0.5%. One type of cataract with the largest number of sufferers is senile cataract. Based on the results of inpatient visits in 2017 at Muhammad Sani Hospital, Karimun Regency, Riau Islands, senile cataract was the case with the highest visits. Seeing the high cases of senile cataracts, preventive measures are needed to control the factors that cause senile cataracts. This study aims to determine the combination of factors causing senile cataracts to overcome bias caused by confounding variables using Propensity Score Matching (PSM) with the method of estimating the value of the propensity score, namely binary logistic regression. The factors that are thought to influence are gender, age, education, occupation related to exposure to ultraviolet light, history of diabetes Mellitus, hypertension, and eye trauma. The results of the covariate balance analysis showed that age, last education and history of hypertension were variables that directly affected senile cataract. The bias that can be reduced from the results of the propensity score matching with binary logistic regression is 98%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Confounding, Katarak Senilis, Propensity Score Matching, Senile Cataracts, Propensity Score Matching
Subjects: Q Science > Q Science (General)
Q Science > QA Mathematics
Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Debby Septien Simanjuntak
Date Deposited: 28 Aug 2021 05:45
Last Modified: 03 Oct 2024 05:10
URI: http://repository.its.ac.id/id/eprint/90219

Actions (login required)

View Item View Item