医学生用gtsummary包遇'tbl_summary'函数未找到错误求助

先搞定tbl_summary找不到的问题 #

出现这个报错，哪怕你以为加载了包，大概率是这几个原因：

• 包未正确安装：gtsummary依赖不少tidyverse相关包，安装时可能因网络或依赖缺失没装全。重新运行：

  install.packages("gtsummary", dependencies = TRUE)
  library(gtsummary)
  

• 加载包时拼写错误：确认你写的是library(gtsummary)，不是library(gt_summary)或其他错漏。
• 函数名冲突：如果同时加载了其他包，可能有重名函数。直接指定包名调用即可解决：

  data %>% 
    select(age_years, gender, weight, height, pregnant, breastfeeding, income, work_intensity, religion, marital, education) %>%
    gtsummary::tbl_summary()  # 明确调用gtsummary包的函数
  

更易上手的定制汇总表方案 #

1. 基于gtsummary定制（适配医学研究需求） #

你需要的年龄分组可以这么实现：先对年龄变量分组，再生成带分组对比的汇总表，示例如下：

data %>%
  mutate(age_group = cut(age_years, breaks = c(17, 30, 50, Inf), labels = c("18-30", "31-50", "51+"))) %>%
  select(age_group, gender, weight, height, pregnant, breastfeeding, income, work_intensity, religion, marital, education) %>%
  tbl_summary(
    by = age_group,  # 按年龄分组展示对比
    type = list(weight ~ "continuous", height ~ "continuous"),  # 指定变量类型
    statistic = list(continuous ~ "{mean} ({sd})", categorical ~ "{n} ({p}%)")  # 自定义统计量格式
  ) %>%
  add_p()  # 添加组间比较P值（医学研究常用）

2. 用tableone包（医学领域轻量化工具） #

tableone是医学研究做基线表的常用包，语法更直白：

# 安装加载包
install.packages("tableone")
library(tableone)

# 定义变量列表与分类变量
vars <- c("age_years", "gender", "weight", "height", "pregnant", "breastfeeding", "income", "work_intensity", "religion", "marital", "education")
catVars <- c("gender", "pregnant", "breastfeeding", "income", "work_intensity", "religion", "marital", "education")

# 年龄分组
data$age_group <- cut(data$age_years, breaks = c(17, 30, 50, Inf), labels = c("18-30", "31-50", "51+"))

# 生成分组汇总表
tableOne <- CreateTableOne(vars = vars, factorVars = catVars, data = data, strata = "age_group")
# 输出格式化表格
print(tableOne, showAllLevels = TRUE, printToggle = FALSE)

3. dplyr+gt组合（完全自定义，灵活度最高） #

如果需要极致自定义表格样式，可以用dplyr做数据汇总，再用gt包美化：

library(dplyr)
library(gt)

data %>%
  mutate(age_group = cut(age_years, breaks = c(17, 30, 50, Inf), labels = c("18-30", "31-50", "51+"))) %>%
  group_by(age_group) %>%
  summarise(
    年龄均值 = mean(age_years, na.rm = TRUE),
    年龄标准差 = sd(age_years, na.rm = TRUE),
    男性占比 = sum(gender == "男", na.rm = TRUE)/n()*100,
    体重均值 = mean(weight, na.rm = TRUE)
    # 按需添加其他变量的统计量
  ) %>%
  gt() %>%
  tab_header(title = "研究人群基线特征汇总") %>%
  fmt_number(columns = c(年龄均值, 年龄标准差, 体重均值), decimals = 1) %>%
  fmt_percent(columns = 男性占比, decimals = 0)

内容的提问来源于stack exchange，提问作者Aran Kartal