如何将患者治疗首尾日期行扩展为连续日期序列？

原始数据准备 #

首先修正原始数据的类型问题（Day列需转为数值型才能生成连续序列）：

a <- as.data.frame(matrix(c("1", "1", "1", "1", "2", "2", "A first day", "A last day", "B first day", "B last day", "A first day", "A last day", 3, 5, 10, 14, 2, 5), ncol = 3))
colnames(a) = c("Patient", "Treatment", "Day")
a$Day <- as.numeric(a$Day) # 转换Day列为数值型

原始数据结构：

方法一：使用tidyverse工具链（推荐） #

通过dplyr做数据清洗，tidyr做格式转换，步骤清晰易读：

library(tidyverse)

# 1. 提取治疗标识和日期类型，转换为宽格式
a_clean <- a %>%
  mutate(
    Treatment = str_sub(Treatment, 1, 1), # 提取A/B治疗标识
    Date_type = ifelse(str_detect(Treatment, "first"), "first", "last") # 区分起始/结束日
  ) %>%
  pivot_wider(
    idvars = c("Patient", "Treatment"),
    names_from = Date_type,
    values_from = Day
  )

# 2. 生成连续日期序列并展开为行
final_df <- a_clean %>%
  rowwise() %>%
  mutate(Day = list(seq(first, last, by = 1))) %>% # 生成周期内的所有日期
  unnest(Day) %>% # 将列表展开为单独行
  select(Patient, Treatment, Day) %>% # 保留目标字段
  arrange(Patient, Treatment, Day) # 按患者、治疗类型、日期排序

print(final_df)

方法二：使用Base R实现 #

无需额外安装包，纯基础R代码完成：

# 1. 提取关键信息
a$Treatment <- substr(a$Treatment, 1, 1) # 提取A/B治疗标识
a$Date_type <- ifelse(grepl("first", a$Treatment), "first", "last") # 标记日期类型

# 2. 转换为宽格式，将起始/结束日放在同一行
wide_df <- reshape(
  a,
  idvar = c("Patient", "Treatment"),
  timevar = "Date_type",
  direction = "wide"
)
colnames(wide_df) <- c("Patient", "Treatment", "first_day", "last_day")

# 3. 遍历每行生成日期序列，合并为最终数据框
final_list <- lapply(1:nrow(wide_df), function(row_idx) {
  current_row <- wide_df[row_idx, ]
  days_seq <- seq(current_row$first_day, current_row$last_day, by = 1)
  data.frame(
    Patient = current_row$Patient,
    Treatment = current_row$Treatment,
    Day = days_seq
  )
})

final_df <- do.call(rbind, final_list)
final_df <- final_df[order(final_df$Patient, final_df$Treatment, final_df$Day), ]

print(final_df)

最终输出结果 #

两种方法都会生成目标格式的数据框：

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