如何在R中同时评估和分析不同的临床症状

问题描述 投票：0回答：1

我目前正在进行一项研究，旨在比较接受特定类型脑部手术的患者的术后并发症。在我们的一项分析中，我们希望比较患者（使用 Createtableone 功能）在脑部手术后经历的症状。

我们面临的问题是我们收集了不同类别变量中的症状。每个患者最多可能有四种不同的症状，但有些患者的症状少于四种。我们将症状归类为 FND（局灶性神经功能缺损），产生四个变量：FNDtype、FNDtype2、FNDtype3 和 FNDtype4。现在，我想将这些变量合并为一个，以便我可以进行描述性分析，显示每个分层组的每种症状的频率。

以下是 Rscript 的相关部分：

# Postcraniectomia Infection
data$postcraniotomy_infection <- factor(data$`Postcraniotomy infection (y=1, n=0)`, levels = c(0,1), labels = c("no", "yes"))

# FND Type 
data$FNDtype <- factor(data$`Type of FND (1)`, levels = c(0,1,2,3,4,5,6,7,8,9), labels = c("none", "hemiparesis", "hypoesthesia", "aphasia", "visual field defects", "coordiantion disturbance", "cranial nerve dysfunction", "neglect","dysarthria","other"))

# FND Type 2 
data$FNDtype2 <- factor(data$`Type of FND (2)`, levels = c(0,1,2,3,4,5,6,7,8,9), labels = c("none", "hemiparesis", "hypoesthesia", "aphasia", "visual field defects", "coordiantion disturbance", "cranial nerve dysfunction", "neglect","dysarthria","other"))
                        
# FND Type 3 
data$FNDtype3 <- factor(data$`Type of FND (3)`, levels = c(0,1,2,3,4,5,6,7,8,9), labels = c("none", "hemiparesis", "hypoesthesia", "aphasia", "visual field defects", "coordiantion disturbance", "cranial nerve dysfunction", "neglect","dysarthria","other"))
                       
# FND Type 4 
data$FNDtype4 <- factor(data$`Type of FND (4)`, levels = c(0,1,2,3,4,5,6,7,8,9), labels = c("none", "hemiparesis", "hypoesthesia", "aphasia", "visual field defects", "coordiantion disturbance", "cranial nerve dysfunction", "neglect","dysarthria","other"))

# Trying to merge FND Types into one variable
data$fndtype <- paste(data$FNDtype, data$FNDtype2, data$FNDtype3, data$FNDtype4, sep = ",") 

vars <- c("fndtype")

table <- CreateTableOne(vars = vars, data = data, strata = c("primary_infection"), testNonNormal = kruskal.test, addOverall = TRUE)

kableone(table, nonnormal = vars)

这是 dput 数据集，包括分层变量 postcraniotomy_infection 和我们的患者经历的四种不同症状：

list(postcraniotomy_infection = structure(c(2L, 1L, 2L, 2L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 1L, 2L, 2L, 1L, 
2L, 2L, 2L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 2L, 2L, 2L, 1L, 2L, 1L, 
2L, 2L, 1L, 2L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 2L, 1L, 1L, 1L, 2L, 
2L, 1L, 2L, 2L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 2L, 
2L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 2L, 2L, 1L), levels = c("no", 
"yes"), class = "factor"), FNDtype = structure(c(1L, 2L, 2L, 
1L, 2L, 1L, 8L, 2L, 4L, 4L, 1L, 3L, 4L, 2L, 2L, 4L, 2L, 3L, 1L, 
1L, 1L, 3L, 1L, 9L, 7L, 1L, 2L, 2L, 1L, 1L, 5L, 1L, 3L, 1L, 2L, 
3L, 3L, 2L, 4L, 2L, 1L, 10L, 2L, 1L, 3L, 6L, 1L, 1L, 1L, 1L, 
2L, 2L, 1L, 1L, 6L, 1L, 5L, 10L, 1L, 1L, 2L, 10L, 1L, 1L, 1L, 
2L, 1L, 1L, 1L, 5L, 5L, 1L, 1L, 2L, 2L, 1L, 2L, 2L, 2L, 5L, 3L, 
1L, 3L, 1L), levels = c("none", "hemiparesis", "hypoesthesia", 
"aphasia", "visual field defects", "coordiantion disturbance", 
"cranial nerve dysfunction", "neglect", "dysarthria", "other"
), class = "factor"), FNDtype2 = structure(c(1L, 1L, 1L, 1L, 
3L, 1L, 1L, 5L, 1L, 8L, 1L, 7L, 1L, 4L, 1L, 7L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 8L, 7L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 7L, 
1L, 6L, 6L, 3L, 1L, 1L, 6L, 1L, 9L, 10L, 1L, 1L, 1L, 1L, 1L, 
3L, 1L, 1L, 8L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 3L, 4L, 1L, 4L, 9L, 6L, 1L, 4L, 1L, 6L, 
1L), levels = c("none", "hemiparesis", "hypoesthesia", "aphasia", 
"visual field defects", "coordiantion disturbance", "cranial nerve dysfunction", 
"neglect", "dysarthria", "other"), class = "factor"), FNDtype3 = structure(c(1L, 
1L, 1L, 1L, 6L, 1L, 1L, 8L, 1L, 1L, 1L, 8L, 1L, 7L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 9L, 8L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 6L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 4L, 1L, 1L, 10L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 4L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 8L, 1L, 6L, 1L, 9L, 1L, 
1L, 1L, 1L, 1L), levels = c("none", "hemiparesis", "hypoesthesia", 
"aphasia", "visual field defects", "coordiantion disturbance", 
"cranial nerve dysfunction", "neglect", "dysarthria", "other"
), class = "factor"), FNDtype4 = structure(c(1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 9L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 8L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 10L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L
), levels = c("none", "hemiparesis", "hypoesthesia", "aphasia", 
"visual field defects", "coordiantion disturbance", "cranial nerve dysfunction", 
"neglect", "dysarthria", "other"), class = "factor"))

我尝试使用带有参数 

paste()
的 
sep=","
函数，它确实创建了一个变量，其中不同的值用逗号分隔。然而，当我尝试使用 CreateTableOne 创建描述性分析时，结果向我显示了有关症状组合的统计数据（例如，偏瘫、感觉减退、协调障碍 n=xx），而不是有关单一症状的统计数据（例如，偏瘫 n= XX，感觉减退 n=XX，协调障碍 n=XX)。

我真的希望 SO 社区中的有人可以帮助我，因为我坚信这个问题一定有解决方案。如果您需要更多信息，请告诉我！

提前感谢您的支持。

最诚挚的问候，马西莫

r merge statistics categorical-data
1个回答
0
投票

您可以使用不同的分类变量来缓解问题。对于每个症状，我们创建一个以全零开头的新列。然后，我们检查每位患者，看看他们在四个 FND 条目中是否有这种症状。如果有，请针对此症状标记为 1。 最后我们创建一个 df ，其中 row 是不同的症状。这些列显示了每种症状在感染/未感染患者中出现的频率。

#install.packages("tableone")
library(tableone)    

analyze_fnd_symptoms <- function(data_list) {
  # Print diagnostic information
  cat("Number of observations:", length(data_list$postcraniotomy_infection), "\n")
  cat("Available symptoms:", levels(data_list$FNDtype)[-1], "\n")
  
  # Get number of observations
  n <- length(data_list$postcraniotomy_infection)
  
  # Get all symptom levels (excluding "none") from FNDtype
  all_symptoms <- levels(data_list$FNDtype)[-1]  # Remove "none" which is the first level
  
  # Initialize result dataframe
  result_df <- data.frame(
    postcraniotomy_infection = data_list$postcraniotomy_infection
  )
  
  # Create binary columns for each symptom with explicit comparison
  for(symptom in all_symptoms) {
    result_df[[symptom]] <- 0  # Initialize with zeros
    
    # Check each FND type and update if symptom is present
    for(i in 1:n) {
      if(data_list$FNDtype[i] == symptom || 
         data_list$FNDtype2[i] == symptom ||
         data_list$FNDtype3[i] == symptom ||
         data_list$FNDtype4[i] == symptom) {
        result_df[i, symptom] <- 1
      }
    }
    
    # Print diagnostic information for each symptom
    cat("Symptom:", symptom, "- Count:", sum(result_df[[symptom]]), "\n")
  }
  
  # Verify we have valid columns
  valid_cols <- sapply(result_df[,-1], function(x) sum(x) > 0)
  valid_symptoms <- names(valid_cols)[valid_cols]
  
  if(length(valid_symptoms) == 0) {
    stop("No symptoms found with non-zero counts")
  }
  
  # Create table comparing symptoms between infection groups
  table <- CreateTableOne(vars = valid_symptoms, 
                          data = result_df, 
                          strata = "postcraniotomy_infection",
                          addOverall = TRUE)
  
  return(table)
}
# use it

vars <- c("fndtype")

table <- analyze_fnd_symptoms(data)
kableone(table, nonnormal = vars)

输出：

> kableone(table, nonnormal = vars)
整体 不 是的 p 测试
n 84 37 47
偏瘫（平均值（SD）） 0.26 (0.44) 0.41 (0.50) 0.15 (0.36) 0.008
感觉减退（平均值（SD）） 0.18 (0.39) 0.22 (0.42) 0.15 (0.36) 0.430
失语症（平均（SD）） 0.13 (0.34) 0.19 (0.40) 0.09 (0.28) 0.164
视野缺损（平均值（SD）） 0.07 (0.26) 0.11 (0.31) 0.04 (0.20) 0.252
协调障碍（平均值（SD）） 0.12 (0.33) 0.11 (0.31) 0.13 (0.34) 0.787
脑神经功能障碍（平均值（SD）） 0.07 (0.26) 0.08 (0.28) 0.06 (0.25) 0.764
忽视（平均值（SD）） 0.11 (0.31) 0.19 (0.40) 0.04 (0.20) 0.031
构音障碍（平均值（SD）） 0.07 (0.26) 0.14 (0.35) 0.02 (0.15) 0.045
其他（平均值（SD）） 0.07 (0.26) 0.16 (0.37) 0.00 (0.00) 0.004
最新问题
© www.soinside.com 2019 - 2025. All rights reserved.