ANALISIS SURVIVAL TERHADAP PASIEN PENYAKIT GINJAL KRONIS DENGAN MENGGUNAKAN COX REGRESSION
Abstract
Chronic Kidney Disease (CKD) is a clinical condition characterized by decreased kidney function requiring renal replacement therapy were fixed, in the form of dialysis. The mortality rate caused by CKD increases significantly each year and needs to be reduced, so that the research aimed to identify factors that influence the survival time of patients with CKD at Arifin Achmad Hospital in Pekanbaru, Riau on 2015-2017. One of the tools to analyze the longer survival of patients with CKD is cox proportional hazards regression. In the long survival data of patients with CKD there is a possibility of ties, so there are several methods in determining the partial likelihood estimation parameters, ie breslow, efron, and exact. Based on the analysis results obtained the factors that most influence with the best parameter estimation is the exact approach are diastolic blood pressure, respiratory rate, urea levels, and action. There is a variable that does not meet the proportional hazard assumption, ie the action variable. The best way to overcome its assumption, namely Regression Cox Stratified Without Interaction and Cox Stratified Regression with Interaction. Based on the results of the analysis that meets the proportional hazard assumption is the method of Cox Stratified Regression Without Interaction and obtained the factors that affect the length of survival of CKD patients are diastolic blood pressure, respiratory rate, and urea levels. Interpretation of equation model of Cox Stratified Regression Without Interpretation of equation model of Cox Stratified Regression Without Interaction obtained result of increase of diastolic blood pressure patient has chance to die smaller equal to 2.639158 times, increase of respiration rate patient have chance to die bigger equal to 1.05902 times, and increase of ureum level patient have chance to die bigger of 1.005884 times.
Collections
- Statistics [900]