5.1 R包介绍

scitb是一个专门用于基线表开发的包，相较于tableone scitb具有更好使用，导出方便，快速设置的优点

CRAN库

作者教程

5.2 R包安装

install.packages("scitb")

5.3 R包使用

5.3.1 导入示例数据

导入数据prematurity，：low 是否是小于2500g早产低体重儿，age 母亲的年龄，lwt 末次月经体重，race 种族，smoke孕期抽烟，ptl 早产史（计数），ht 有高血压病史，ui 子宫过敏，ftv 早孕时看医生的次数bwt 新生儿体重数值

library(scitb)
library(bruceR)
data <- prematurity
head(data,5)
##   id low age lwt  race     smoke ptl ht ui ftv  bwt
## 1 85   0  19 182 black nonsmoker   0  0  1   0 2523
## 2 86   0  33 155 other nonsmoker   0  0  0   3 2551
## 3 87   0  20 105 white    smoker   0  0  0   1 2557
## 4 88   0  21 108 white    smoker   0  0  1   2 2594
## 5 89   0  18 107 white    smoker   0  0  1   0 2600

5.3.2 生成描述性表1

5.3.2.1 首先定义变量

为什么这里要定义变量，而不在基线表函数里直接引入字符向量，这是为了有良好的编程习惯。先从构建函数框架开始，最后函数调用，后续无论是修改变量还是增删变量都会很清晰

allVars<-c("low","age","lwt","race","smoke","ptl","ht","ui","ftv","bwt")
fvars<-c("low","race","smoke","ptl","ht","ui","ftv")
strata<-c("bwt")

5.3.2.2 生成基线表

data_neat <- data |> 
  mutate(
    race = as.factor(race),
    smoke = as.factor(smoke)
  )

tab0<-scitb1(vars=allVars,fvars=fvars,strata=strata,data=data_neat)
##       Min 33.33333% 66.66667%       Max 
##   709.000  2598.000  3283.667  4990.000
## Strata is treated as a continuous variable.
tab0
##       Characteristic G.0                      G.1                     
## scinn "N"            "63"                     "63"                    
## d1    "age"          "22.00 (19.50-25.00)"    "23.00 (19.00-28.00)"   
## d1    "lwt"          "120.00 (105.00-131.00)" "120.00 (109.50-132.50)"
## d1    "bwt"          "2126.24±394.60"         "2969.94±190.20"        
## d2    "low"          " "                      " "                     
##       "0"            " 4 (6.35%)"             "63 (100.00%)"          
##       "1"            "59 (93.65%)"            " 0 (0.00%)"            
## d2    "race"         " "                      " "                     
##       "black"        "12 (19.05%)"            " 8 (12.70%)"           
##       "other"        "26 (41.27%)"            "25 (39.68%)"           
##       "white"        "25 (39.68%)"            "30 (47.62%)"           
## d2    "smoke"        " "                      " "                     
##       "nonsmoker"    "31 (49.21%)"            "39 (61.90%)"           
##       "smoker"       "32 (50.79%)"            "24 (38.10%)"           
## d2    "ptl"          " "                      " "                     
##       "0"            "45 (71.43%)"            "58 (92.06%)"           
##       "1"            "16 (25.40%)"            " 4 (6.35%)"            
##       "2"            " 2 (3.17%)"             " 1 (1.59%)"            
##       "3"            " 0 (0.00%)"             " 0 (0.00%)"            
## d2    "ht"           " "                      " "                     
##       "0"            "56 (88.89%)"            "61 (96.83%)"           
##       "1"            " 7 (11.11%)"            " 2 (3.17%)"            
## d2    "ui"           " "                      " "                     
##       "0"            "47 (74.60%)"            "54 (85.71%)"           
##       "1"            "16 (25.40%)"            " 9 (14.29%)"           
## d2    "ftv"          " "                      " "                     
##       "0"            "37 (58.73%)"            "32 (50.79%)"           
##       "1"            "12 (19.05%)"            "15 (23.81%)"           
##       "2"            " 8 (12.70%)"            "13 (20.63%)"           
##       "3"            " 5 (7.94%)"             " 2 (3.17%)"            
##       "4"            " 1 (1.59%)"             " 1 (1.59%)"            
##       "6"            " 0 (0.00%)"             " 0 (0.00%)"            
##       G.2                      p value 
## scinn "63"                     ""      
## d1    "23.00 (19.50-28.00)"    "0.483" 
## d1    "131.00 (120.00-156.00)" "<0.001"
## d1    "3736.68±317.76"         "<0.001"
## d2    " "                      "<0.001"
##       "63 (100.00%)"           " "     
##       " 0 (0.00%)"             " "     
## d2    " "                      "0.060" 
##       " 6 (9.52%)"             " "     
##       "16 (25.40%)"            " "     
##       "41 (65.08%)"            " "     
## d2    " "                      "0.037" 
##       "45 (71.43%)"            " "     
##       "18 (28.57%)"            " "     
## d2    " "                      "0.012" 
##       "56 (88.89%)"            " "     
##       " 4 (6.35%)"             " "     
##       " 2 (3.17%)"             " "     
##       " 1 (1.59%)"             " "     
## d2    " "                      "0.154" 
##       "60 (95.24%)"            " "     
##       " 3 (4.76%)"             " "     
## d2    " "                      "0.005" 
##       "60 (95.24%)"            " "     
##       " 3 (4.76%)"             " "     
## d2    " "                      "0.283" 
##       "31 (49.21%)"            " "     
##       "20 (31.75%)"            " "     
##       " 9 (14.29%)"            " "     
##       " 0 (0.00%)"             " "     
##       " 2 (3.17%)"             " "     
##       " 1 (1.59%)"             " "

5.3.2.3 设置有非正态资料

设置非正态后可以看到lwt变量是用百分位数进行描述，后边的pvalue也调用非参检验

tab2 <-scitb1(vars=allVars,fvars=fvars,strata=strata,data=data_neat, nonnormal=c("lwt"))
##       Min 33.33333% 66.66667%       Max 
##   709.000  2598.000  3283.667  4990.000
## Strata is treated as a continuous variable.
tab2
##       Characteristic G.0                      G.1                     
## scinn "N"            "63"                     "63"                    
## d1    "age"          "22.00 (19.50-25.00)"    "23.00 (19.00-28.00)"   
## d1    "lwt"          "120.00 (105.00-131.00)" "120.00 (109.50-132.50)"
## d1    "bwt"          "2126.24±394.60"         "2969.94±190.20"        
## d2    "low"          " "                      " "                     
##       "0"            " 4 (6.35%)"             "63 (100.00%)"          
##       "1"            "59 (93.65%)"            " 0 (0.00%)"            
## d2    "race"         " "                      " "                     
##       "black"        "12 (19.05%)"            " 8 (12.70%)"           
##       "other"        "26 (41.27%)"            "25 (39.68%)"           
##       "white"        "25 (39.68%)"            "30 (47.62%)"           
## d2    "smoke"        " "                      " "                     
##       "nonsmoker"    "31 (49.21%)"            "39 (61.90%)"           
##       "smoker"       "32 (50.79%)"            "24 (38.10%)"           
## d2    "ptl"          " "                      " "                     
##       "0"            "45 (71.43%)"            "58 (92.06%)"           
##       "1"            "16 (25.40%)"            " 4 (6.35%)"            
##       "2"            " 2 (3.17%)"             " 1 (1.59%)"            
##       "3"            " 0 (0.00%)"             " 0 (0.00%)"            
## d2    "ht"           " "                      " "                     
##       "0"            "56 (88.89%)"            "61 (96.83%)"           
##       "1"            " 7 (11.11%)"            " 2 (3.17%)"            
## d2    "ui"           " "                      " "                     
##       "0"            "47 (74.60%)"            "54 (85.71%)"           
##       "1"            "16 (25.40%)"            " 9 (14.29%)"           
## d2    "ftv"          " "                      " "                     
##       "0"            "37 (58.73%)"            "32 (50.79%)"           
##       "1"            "12 (19.05%)"            "15 (23.81%)"           
##       "2"            " 8 (12.70%)"            "13 (20.63%)"           
##       "3"            " 5 (7.94%)"             " 2 (3.17%)"            
##       "4"            " 1 (1.59%)"             " 1 (1.59%)"            
##       "6"            " 0 (0.00%)"             " 0 (0.00%)"            
##       G.2                      p value 
## scinn "63"                     ""      
## d1    "23.00 (19.50-28.00)"    "0.483" 
## d1    "131.00 (120.00-156.00)" "<0.001"
## d1    "3736.68±317.76"         "<0.001"
## d2    " "                      "<0.001"
##       "63 (100.00%)"           " "     
##       " 0 (0.00%)"             " "     
## d2    " "                      "0.060" 
##       " 6 (9.52%)"             " "     
##       "16 (25.40%)"            " "     
##       "41 (65.08%)"            " "     
## d2    " "                      "0.037" 
##       "45 (71.43%)"            " "     
##       "18 (28.57%)"            " "     
## d2    " "                      "0.012" 
##       "56 (88.89%)"            " "     
##       " 4 (6.35%)"             " "     
##       " 2 (3.17%)"             " "     
##       " 1 (1.59%)"             " "     
## d2    " "                      "0.154" 
##       "60 (95.24%)"            " "     
##       " 3 (4.76%)"             " "     
## d2    " "                      "0.005" 
##       "60 (95.24%)"            " "     
##       " 3 (4.76%)"             " "     
## d2    " "                      "0.283" 
##       "31 (49.21%)"            " "     
##       "20 (31.75%)"            " "     
##       " 9 (14.29%)"            " "     
##       " 0 (0.00%)"             " "     
##       " 2 (3.17%)"             " "     
##       " 1 (1.59%)"             " "

5.3.3 连续变量快速分层描述

接下来我们来根据age变量设置分层，制作表1。

输出结果可以看出年龄的自动分层记录，表明从14到20到25到45分了三层。

如果想要生成其它分层数，设置参数 num = n

# allvar删不删除年龄都可以

allVars3 <-c("race", "lwt","smoke", "ptl", "ht", "ui","ftv", "bwt")
fvars3 <-c("smoke","ht","ui","race")
strata3 <- "age"

tab3 <-scitb1(vars=allVars3,fvars=fvars3,strata=strata3,data=data_neat)
##       Min 33.33333% 66.66667%       Max 
##        14        20        25        45
## Strata is treated as a continuous variable.
tab3
##       Characteristic G.0                      G.1                     
## scinn "N"            "51"                     "69"                    
## d1    "lwt"          "119.00 (106.00-135.00)" "121.00 (110.00-132.00)"
## d1    "ptl"          "0.00 (0.00-0.00)"       "0.00 (0.00-0.00)"      
## d1    "ftv"          "0.00 (0.00-1.00)"       "0.00 (0.00-1.00)"      
## d1    "bwt"          "2974.08±612.51"         "2921.42±697.85"        
## d2    "smoke"        " "                      " "                     
##       "nonsmoker"    "28 (54.90%)"            "44 (63.77%)"           
##       "smoker"       "23 (45.10%)"            "25 (36.23%)"           
## d2    "ht"           " "                      " "                     
##       "0"            "48 (94.12%)"            "64 (92.75%)"           
##       "1"            " 3 (5.88%)"             " 5 (7.25%)"            
## d2    "ui"           " "                      " "                     
##       "0"            "43 (84.31%)"            "58 (84.06%)"           
##       "1"            " 8 (15.69%)"            "11 (15.94%)"           
## d2    "race"         " "                      " "                     
##       "black"        "10 (19.61%)"            "10 (14.49%)"           
##       "other"        "18 (35.29%)"            "27 (39.13%)"           
##       "white"        "23 (45.10%)"            "32 (46.38%)"           
##       G.2                      p value
## scinn "69"                     ""     
## d1    "130.00 (115.00-153.00)" "0.086"
## d1    "0.00 (0.00-0.00)"       "0.171"
## d1    "1.00 (0.00-2.00)"       "0.011"
## d1    "2945.13±839.82"         "0.927"
## d2    " "                      "0.587"
##       "43 (62.32%)"            " "    
##       "26 (37.68%)"            " "    
## d2    " "                      "0.929"
##       "65 (94.20%)"            " "    
##       " 4 (5.80%)"             " "    
## d2    " "                      "0.873"
##       "60 (86.96%)"            " "    
##       " 9 (13.04%)"            " "    
## d2    " "                      "0.320"
##       " 6 (8.70%)"             " "    
##       "22 (31.88%)"            " "    
##       "41 (59.42%)"            " "

5.3.4 保存结果

write.csv(tab0,file= "01-attch\\10\\data.csv",row.names = F, 
    fileEncoding = "gbk")

保存出来是下图，复制到word改改就能用

图

5.4 总结

使用体验：自定义度，结果展示等不如comparegroups，但是胜在参数简单，在期刊没那么高要求和平时展示可以用