Survival analysis based on ischemic time

The analysis is based on heart transplanation data. This study gives the potential to extract significant information about the survival probability depending on ischemic time. So, a variation in ischemic time can influence the adaptation of the transplant in years; and what is the survival probabilty if the ischemic time is 1 hour;

Info: Ischemic time is the time that an organ is outside the body when the heart is not beating or supplied with O2 by the coronary arteries.

Installation & run project

These are the packages used in the analysis:

library(survminer)
library(survival)

git clone https://github.com/askandali/heart-transplant-analysis.git
cd heart-transplant-analysis

install.packages("survival", "survminer")
Rscript bio.R

Data

The file SurvData6.txt consists of 884 observations and 3 variables. These variables give us the following information about a patient:

• Survival time
• Censor
• Ischemic time

The ischemic time (IT) in a barplot: In most of the cases the ischemic time is between 2.87 and 5.27 hours.

Kaplan - Meier curve

In the Kaplan - Meier curve it is obvious that at the survival probability instantly reduces to about 80% without the passing of the time. This shows that there are cases that the tranplant didn't adapt in the patient organism from the first time.

Cox Model

The Cox proportional-hazards model is essentially a regression model commonly used statistical in medical research for investigating the association between the survival time of patients and one or more predictor variables.

sCox1 <- coxph(s ~ IT,data=b)
summary(sCox1)

Results

1. By applying the Cox model, the ischemic time is not statistically significant and doesn't influence a lot the survival time.

2. If the ischemic time is 1 hour, the average survival probability is 77%.