1. Section 7

Table 4 SECTION 7 Hospital course contains variable: Discharge.day, In-Hospital.mortality, Bleeding, Blood.transfusion,
Stroke,MI, Stent.thrombosis, Vascular.complications, Re-cath, CABG. Looking for statistical relationship between risk factor to PCI

1.1 Upload library

library(dplyr)
library(survival)
library(survminer)
library(lubridate)
library(gtsummary)
#library(TMB)
library(broom)
library(kableExtra)
library(SemiCompRisks)

1.2 Upload data

d <- read.csv("d:/UPWORK-SAAD/PCI-01.csv")
#---tb1
#Secsion 1-4
d1 <- d[,2:19]
d2 <- d[,21:31]
d01 <- cbind(d1,d2)
d3 <- d[,32:40]
d02 <- cbind(d01,d3)
#---Section 7
d4 <- d[,54:63]
d04 <- cbind(d02,d4)
d04$Valvular.heart.disease <- as.numeric(d04$Valvular.heart.disease)
## Warning: NAs introduced by coercion

3. Model

3.1 Cox regression

lm_dat <- 
  d04 %>% 
  filter(Age >= 30) 
lm_dat <- 
  lm_dat %>% 
  mutate(
    lm_T1 = Age - 30
    )
    lm_fit <- survfit(Surv(Age, PCI)~
          HTN + Dyslipidemia + 
          HF + Valvular.heart.disease + CKD + Malignancy + 
          Stroke + MI + atrial.fib + COPD + CABG + DM+
          Clopidogrel+ Insulin+Diuretics+
          Normal.BP.1.high.BP..more.than.140.90..2+
          Bleeding+ Stroke.1+CABG.1, 
            data = lm_dat)

3.2 Test Model

coxph(
  Surv(Age, PCI)~ 
          HTN + Dyslipidemia + 
          HF + Valvular.heart.disease + CKD + Malignancy + 
          Stroke + MI + atrial.fib + COPD + CABG + DM+
          Clopidogrel+ Insulin+Diuretics+
          Normal.BP.1.high.BP..more.than.140.90..2+
          Bleeding+ Stroke.1+CABG.1,  
  subset = Age >= 30, 
  data = d04
  ) %>% 
  gtsummary::tbl_regression(exp = TRUE)
Characteristic HR1 95% CI1 p-value
HTN 0.65 0.45, 0.93 0.020
Dyslipidemia 1.48 1.10, 1.99 0.009
HF 0.92 0.60, 1.41 0.7
Valvular.heart.disease 0.31 0.17, 0.56 <0.001
CKD 0.68 0.46, 1.01 0.054
Malignancy 0.63 0.35, 1.13 0.12
Stroke 1.06 0.59, 1.92 0.8
MI 2.03 1.49, 2.77 <0.001
atrial.fib 0.42 0.17, 1.05 0.064
COPD 0.51 0.24, 1.11 0.089
CABG 0.97 0.58, 1.61 >0.9
DM 0.78 0.55, 1.09 0.14
Clopidogrel 1.38 1.06, 1.79 0.016
Insulin 1.53 1.12, 2.10 0.008
Diuretics 0.81 0.60, 1.09 0.2
Normal.BP.1.high.BP..more.than.140.90..2 0.79 0.60, 1.02 0.072
Bleeding 1.85 0.84, 4.08 0.13
Stroke.1 1.65 0.22, 12.3 0.6
CABG.1 0.68 0.33, 1.40 0.3

1 HR = Hazard Ratio, CI = Confidence Interval

4. Plot Survival Probability

lm_fit1 <- survfit(Surv(lm_T1, PCI) ~ Stroke.1+Stent.thrombosis+
                     Vascular.complications+CABG,
              data = lm_dat)       
ggsurvplot(lm_fit1)

5. Analysis

From the above table found the variables important influence the PCI outcome indicated by Pvalue < 0.05 :HTN, Dislipidea,Valvular hert desease, MI,Clopidogril and Insulin. From the Plot survival probability can be seen that CABG has survival probability Age 40-50.