如何修正timereg与riskRegression代码以匹配tidycmprsk竞争风险分析结果?
竞争风险分析结果不一致的问题修正请求
我正在使用{MASS}包中的Melanoma数据集开展竞争风险分析,但在tidycmprsk、timereg和riskRegression三个包中得到的结果不一致。
我已按照教程修改了数据集中的status编码,修改代码如下:
Melanoma <- Melanoma %>% mutate( status = as.factor(recode(status, `2` = 0, `1` = 1, `3` = 2)) )
我尝试的代码均可运行但结果不同,现询问如何修正comp.risk和riskRegression部分的代码,使其结果与crr部分一致:
1. {tidycmprsk}包的crr()函数
crr(Surv(time, status) ~ sex + age, data = Melanoma)
运行结果:
## ## Variable Coef SE HR 95% CI p-value ## sex 0.588 0.272 1.80 1.06, 3.07 0.030 ## age 0.013 0.009 1.01 0.99, 1.03 0.18
2. {timereg}包的comp.risk()函数
fit_comprisk <- comp.risk(Event(time, status, cens.code=0) ~ const(sex) + const(age), data = Melanoma, cause = 1, model = "prop") summary(fit_comprisk)
运行结果:
## Competing risks Model ## ## No test for non-parametric terms ## Parametric terms : ## Coef. SE Robust SE z P-val lower2.5% upper97.5% ## const(sex) -0.394000 0.20200 0.20200 -1.950 0.051 -0.7900 0.00191 ## const(age) 0.000604 0.00557 0.00557 0.108 0.914 -0.0103 0.01150
3. {riskRegression}包的riskRegression()函数
fit_rskreg <- riskRegression(Hist(time, status, cens.code=0) ~ sex + age, data = Melanoma, cause = 1, link = "prop") fit_rskreg
运行结果:
## Competing risks regression model ## ## IPCW weights: marginal Kaplan-Meier for the censoring distribution. ## Link: 'cloglog' yielding sub-hazard ratios (Fine & Gray 1999) ## No covariates with time-varying coefficient specified. ## ## Time constant regression coefficients: ## Variable Levels Coef Lower Upper Pvalue ## sex 1.94 1.115 3.37 0.0189 ## age 1.01 0.991 1.03 0.3306 ## ## ## Note: The coefficients (Coef) are sub-hazard ratios (Fine & Gray 1999)
内容的提问来源于stack exchange,提问作者Toshi
相关产品推荐
相关产品推荐

