R语言绘制两种条件共患人群占比柱状图及可拓展函数实现

需提前安装加载tidyverse套件，包含数据处理和绘图所需的dplyr、ggplot2等工具。

plot_cross_cond_pct <- function(df, 
                                base_var = "A",
                                target_vars = c("B", "C", "D"),
                                group_by_sex = FALSE,
                                filter_age_gt55 = FALSE) {
  # 年龄条件筛选
  if (filter_age_gt55) {
    df <- df %>% filter(Age > 55)
  }
  # 生成分组变量
  group_cols <- c()
  if (group_by_sex) {
    group_cols <- c(group_cols, "sex")
  }
  # 交叉条件占比计算
  stat_df <- df %>%
    group_by(across(all_of(group_cols))) %>%
    summarise(
      total = n(),
      across(all_of(target_vars), ~ sum(.data[[base_var]] == 1 & .x == 1, na.rm = T)/total * 100,
             .names = "pct_{.col}")
    ) %>%
    pivot_longer(cols = starts_with("pct_"), 
                 names_to = "dimension", 
                 values_to = "percentage",
                 names_prefix = "pct_")
  # 绘制柱状图
  plot <- ggplot(stat_df, aes(x = dimension, y = percentage)) +
    geom_col(position = position_dodge2(preserve = "single"), fill = "#2c7fb8", alpha = 0.8) +
    labs(x = "维度", y = "占比(%)") +
    theme_bw()
  # 按性别分组时调整填充映射
  if (group_by_sex) {
    plot <- plot + aes(fill = sex) + scale_fill_brewer(palette = "Set2")
  }
  return(plot)
}

第一步：导入示例数据 #

Lines <- "id sex Age A B C D
1       male   90 1 1 0 1
2       female  87 0 1 1 0
3       male  50 1 1 0 0
4       female  54 0 1 0 1"
df <- read.table(text = Lines, header = T)

第二步：不同场景调用函数 #

• 基础场景：无筛选无分组，输出B/C/D三个维度的A&条件占比

plot_cross_cond_pct(df)

• 按性别拆分分组展示

plot_cross_cond_pct(df, group_by_sex = TRUE)

• 仅统计年龄>55的样本

plot_cross_cond_pct(df, filter_age_gt55 = TRUE)

• 扩展多变量场景：如果新增E、F等变量，只需修改target_vars参数即可

# 新增变量调用示例
# plot_cross_cond_pct(df, target_vars = c("B","C","D","E","F"))

函数特性说明 #

• 支持任意数量的目标交叉变量，仅需调整target_vars参数传入对应的变量名向量即可
• 两个分组/筛选参数独立控制，可自由组合使用
• 返回值为标准ggplot对象，可后续自行调整配色、标签、主题等样式

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