Cox 风险比例模型绘制森林图

coxph函数构建公式，公式包括生存对象和纳入多因素cox的变量，注意变量可以是分类变量 需要Surv函数构建生存对象 多因素分析得到HR, 95%CI，Pvalue 还可一句简单的命令绘制森林图

require("survival")
require(survminer)
library(ggplot2)
head(colon)
##   id study      rx sex age obstruct perfor adhere nodes status differ
## 1  1     1 Lev+5FU   1  43        0      0      0     5      1      2
## 2  1     1 Lev+5FU   1  43        0      0      0     5      1      2
## 3  2     1 Lev+5FU   1  63        0      0      0     1      0      2
## 4  2     1 Lev+5FU   1  63        0      0      0     1      0      2
## 5  3     1     Obs   0  71        0      0      1     7      1      2
## 6  3     1     Obs   0  71        0      0      1     7      1      2
##   extent surg node4 time etype
## 1      3    0     1 1521     2
## 2      3    0     1  968     1
## 3      3    0     0 3087     2
## 4      3    0     0 3087     1
## 5      2    0     1  963     2
## 6      2    0     1  542     1
## 多因素cox回归
# coef回归系数
model <- coxph( Surv(time, status) ~ sex + rx + adhere,
                data = colon )
model
## Call:
## coxph(formula = Surv(time, status) ~ sex + rx + adhere, data = colon)
## 
##               coef exp(coef) se(coef)      z        p
## sex       -0.04615   0.95490  0.06609 -0.698 0.484994
## rxLev     -0.02724   0.97313  0.07690 -0.354 0.723211
## rxLev+5FU -0.43723   0.64582  0.08395 -5.208 1.91e-07
## adhere     0.29355   1.34118  0.08696  3.376 0.000736
## 
## Likelihood ratio test=46.51  on 4 df, p=1.925e-09
## n= 1858, number of events= 920
ggforest(model)

对变量因子化进行命名标签，因子分类变量

再一次说明变量可以有多个分类，不一定是二分类变量 这样就能清楚的指定参考的分类变量标签

colon <- within(colon, {
  sex <- factor(sex, labels = c("female", "male"))
  differ <- factor(differ, labels = c("well", "moderate", "poor"))
  extent <- factor(extent, labels = c("submuc.", "muscle", "serosa", "contig."))
})
bigmodel <-
  coxph(Surv(time, status) ~ sex + rx + adhere + differ + extent + node4,
    data = colon )
ggforest(bigmodel)
## Warning in .get_data(model, data = data): The `data` argument is not
## provided. Data will be extracted from model fit.
## Warning: Removed 4 rows containing missing values (geom_errorbar).