TB Research

Pemodelan Waktu Survival Pasien Tuberkulosis menggunakan Regresi Cox Proportional Hazard dengan Data Tersensor

Elsa Oktaviani, Nonong Amalita, Atus Amadi Putra, Dony Permana

UNP Journal of Statistics and Data Science · 2023-08

Abstract

Cox proportional hazard regression is a type of survival analysis that can be applied to tuberculosis cases. This study aims to determine the Cox proportional hazard regression model and the factors that influence the survival time of tuberculosis patients at RSUP Dr. M. Djamil Padang. The survival period used is the time when the patient is taking TB treatment at RSUP Dr. M. Djamil Padang in 2021 until the patient is declared dead. The method used in the Cox Proportional Hazard Regression analysis is the Maximum Partial Likelihood Estimation Method. By using the cox proportional hazard regression model, the factors that influence the survival time of tuberculosis patients at RSUP Dr. M. Djamil's is BMI (X3) , leukocytes (X5) , fever (X9) , shortness of breath (X11) , and decreased appetite (X12) . The Cox Proportional Hazard Regression Model obtained is hi(t) = h0(t) exp(1,315X3 + 1,224X5 + 1,138X9 +1,623X11 + 1,251X12).

MeSH terms

  • Proportional hazards model
  • Medicine
  • Survival analysis
  • Hazard ratio
  • Regression analysis
  • Hazard
  • Tuberculosis
  • Internal medicine
  • Statistics