R 函数可以跨多个列搜索多个字符串并创建二进制输出变量

以下是使用示例数据的一种方法.您可以创建诊断向量的命名列表:

anxiety_codes <- c("F400", "F401", "F402")
asthma_codes <- c("J450", "J451", "J458", "J459", "J46X")
chronic_lyme_codes <- c("A692", "G630", "M012")

dx_list <- mget(ls(pattern = "(\\w+)_codes"))
names(dx_list) <- sub("_codes$", "", names(dx_list))

dx_list

它看起来像这样:

$anxiety
[1] "F400" "F401" "F402"

$asthma
[1] "J450" "J451" "J458" "J459" "J46X"

$chronic_lyme
[1] "A692" "G630" "M012"

我制作了一个包含id、主要诊断代码和几个次要诊断代码的示例数据框.

df <- data.frame(
  id = 1:3,
  diag_icd10 = c("F400", "J458", "M012"),
  sec_diag_0 = c("J450", NA, "J46X"),
  sec_diag_1 = c(NA, NA, "F401")
)

以下是数据.框架:

  id diag_icd10 sec_diag_0 sec_diag_1
1  1       F400       J450       <NA>
2  2       J458       <NA>       <NA>
3  3       M012       J46X       F401

您可以在诊断列表中使用sapply，并首先判断主要诊断列，并将列名更改为包括"Prim"作为主要诊断.对于二次诊断，在多列中使用Reduce和lapply，然后为二次诊断加上后缀"sec".

df_result <- cbind(
  df[1],
  setNames(
    as.data.frame(sapply(dx_list, \(x) +(df[,2] %in% x))),
    paste(names(dx_list), "prim", sep = "_")
  ),
  setNames(
    as.data.frame(sapply(dx_list, \(x) +Reduce(`|`, lapply(df[,3:4], `%in%`, x)))),
    paste(names(dx_list), "sec", sep = "_")
  )
)

结果看起来是这样的:

  id anxiety_prim asthma_prim chronic_lyme_prim anxiety_sec asthma_sec chronic_lyme_sec
1  1            1           0                 0           0          1                0
2  2            0           1                 0           0          0                0
3  3            0           0                 1           1          1                0

您可以组合LIKE列并添加其他摘要(一次或二次诊断):

sapply(split.default(df_result[-1], sub("_(prim|sec)$", "", names(df_result[-1]))),
       \(x) +(rowSums(x) > 0))

其输出如下(如果需要，可以添加到df_result):

     anxiety asthma chronic_lyme
[1,]       1      1            0
[2,]       0      1            0
[3,]       1      1            1

或者，如果您只想一次处理所有列，请try :

cbind(
  df[1],
  setNames(
    as.data.frame(sapply(dx_list, \(x) +(df[,2] %in% x))),
    paste(names(dx_list), "prim", sep = "_")
  ),
  setNames(
    as.data.frame(sapply(dx_list, \(x) +Reduce(`|`, lapply(df[,3:4], `%in%`, x)))),
    paste(names(dx_list), "sec", sep = "_")
  ),
  sapply(dx_list, \(x) +Reduce(`|`, lapply(df[,2:4], `%in%`, x)))
)