如何用ggplot2复现带分组比值比的森林图及侧边统计表格？

这种图是森林图+分组统计表格的组合，我们可以用ggplot2绘制森林图主体，再结合patchwork（或gridExtra）将统计表格与图拼接，也可以直接在ggplot图层中添加文本实现表格效果。以下是完整的复现步骤：

1. 准备模拟数据 #

先构造符合场景的数据集，包含结局变量、分组的效应估计值、置信区间、P值等信息：

library(tidyverse)
library(patchwork)

# 模拟数据：结局变量、分组（干预/对照）的均值/率、OR值及CI、P值
df_plot <- tibble(
  outcome = factor(c("结局1", "结局2", "结局3", "结局4", "结局5"), 
                   levels = c("结局5", "结局4", "结局3", "结局2", "结局1")), # 反转顺序对应y轴
  group1_est = c("2.3 (1.1-3.5)", "45.2%", "12.8", "0.76", "89.1"), # 分组1的估计值
  group2_est = c("1.8 (0.9-2.7)", "32.1%", "9.6", "0.52", "76.3"), # 分组2的估计值
  or = c(1.32, 1.78, 1.33, 1.46, 1.68),
  ci_low = c(0.91, 1.22, 0.87, 1.01, 1.15),
  ci_high = c(1.92, 2.59, 2.02, 2.11, 2.47),
  p_value = c("0.14", "<0.01", "0.18", "0.04", "<0.01")
)

2. 绘制森林图主体 #

用ggplot2绘制核心的森林图，包含点估计和置信区间：

p_forest <- ggplot(df_plot, aes(x = or, y = outcome)) +
  # 添加置信区间
  geom_errorbarh(aes(xmin = ci_low, xmax = ci_high), height = 0.2, color = "#636363") +
  # 添加点估计
  geom_point(size = 3, color = "#2c3e50") +
  # 添加OR=1的参考线
  geom_vline(xintercept = 1, linetype = "dashed", color = "#e74c3c") +
  # 设置x轴范围和标签
  scale_x_continuous(limits = c(0.5, 3), breaks = seq(0.5, 3, 0.5)) +
  labs(x = "比值比 (OR)", y = "") +
  # 调整主题，去除多余元素
  theme_minimal() +
  theme(
    panel.grid.major.y = element_blank(),
    panel.grid.minor = element_blank(),
    axis.text.y = element_text(size = 10),
    axis.title.x = element_text(size = 11, face = "bold")
  )

3. 制作侧边统计表格 #

我们用ggplot2生成表格样式的图，方便和森林图拼接：

p_table <- ggplot(df_plot, aes(y = outcome)) +
  # 添加分组1的估计值文本
  geom_text(aes(x = 1, label = group1_est), hjust = 0.5, size = 3.5) +
  # 添加分组2的估计值文本
  geom_text(aes(x = 2, label = group2_est), hjust = 0.5, size = 3.5) +
  # 添加P值文本
  geom_text(aes(x = 3, label = p_value), hjust = 0.5, size = 3.5) +
  # 添加表格标题
  annotate("text", x = 1, y = length(df_plot$outcome)+1, label = "干预组", size = 4, fontface = "bold") +
  annotate("text", x = 2, y = length(df_plot$outcome)+1, label = "对照组", size = 4, fontface = "bold") +
  annotate("text", x = 3, y = length(df_plot$outcome)+1, label = "P值", size = 4, fontface = "bold") +
  # 设置坐标轴和主题
  scale_x_continuous(limits = c(0.5, 3.5), breaks = NULL) +
  scale_y_discrete(limits = rev(levels(df_plot$outcome))) + # 和森林图y轴顺序一致
  labs(x = "", y = "") +
  theme_minimal() +
  theme(
    panel.grid = element_blank(),
    axis.text = element_blank(),
    plot.margin = margin(5, 10, 5, 5)
  )

4. 拼接森林图和表格 #

用patchwork将两个图组合，调整宽度比例：

# 拼接表格和森林图，表格宽度占比小，森林图占比大
p_table + p_forest + plot_layout(widths = c(1, 2))

关键调整说明 #

• 如果需要更贴合示例图的样式，可以修改颜色、字体大小、文本对齐方式
• 若你的数据是其他效应量（如RR、均值差），只需替换or列和x轴标签即可
• 如果要在森林图右侧添加表格，只需调整拼接顺序和布局

内容的提问来源于stack exchange，提问作者Nao