数据帧中的子集，以获得r中小于给定日期的行数

Question

我下面有一个例子，我想保留他们的癌症日期(最后一列)发生在10 years or less(即最多10年)之前的行(个人)。

d <- structure(list(ind = c(1, 2, 3, 4, 5), `inclusion date0` = structure(c(1414800000,NA, NA, NA, NA), class = c("POSIXct", "POSIXt"), tzone = "UTC"), 
`inclusion date2` = structure(c(NA, 1453075200, 1453075200,NA, NA), class = c("POSIXct", "POSIXt"), tzone = "UTC"), 
`inclusion dat4` = structure(c(NA, NA, NA, NA, 1544486400), 
class = c("POSIXct", "POSIXt"), tzone = "UTC"), 
`inclusion date6` = structure(c(NA,NA, NA, 1594425600, NA), 
class = c("POSIXct", "POSIXt"), tzone = "UTC"),
`cancer date` = structure(c(1099267200, 965174400, 1294963200,1458086400, 1230854400), 
class = c("POSIXct", "POSIXt"), tzone = "UTC")), 
class = c("tbl_df", "tbl", "data.frame"), row.names = c(NA, -5L))
d$`inclusion date0` <- as.Date(d$`inclusion date0`)
d$`inclusion date2` <- as.Date(d$`inclusion date2`)
d$`inclusion date2` <- as.Date(d$`inclusion date2`)
d$`inclusion dat4` <- as.Date(d$`inclusion dat4`)
d$`inclusion date6` <- as.Date(d$`inclusion date6`)
d$`cancer date` <- as.Date(d$`cancer date`)

我很感激你的帮助。

Answer 1

library(tidyverse)
library(lubridate)

一些额外的数据清理

d %>% 
janitor::clean_names() %>%  
as_tibble() %>% 
select(ind, cancer_date, everything()) %>%  
mutate(across(2:6, ~ as.Date(.x))) %>% 
gather(-c(ind, cancer_date), key = "inclusion_id", value = "inclusion_date") %>%  
drop_na() %>%  
mutate(diff = interval(cancer_date, inclusion_date) %>% 
as.numeric('years'))

计算癌症日期与纳入日期的时间差，以年为单位

# A tibble: 5 x 5
ind cancer_date inclusion_id    inclusion_date  diff
<dbl> <date>      <chr>           <date>         <dbl>
1     1 2004-11-01  inclusion_date0 2014-11-01     10.0 
2     2 2000-08-02  inclusion_date2 2016-01-18     15.5 
3     3 2011-01-14  inclusion_date2 2016-01-18      5.01
4     5 2009-01-02  inclusion_dat4  2018-12-11      9.94
5     4 2016-03-16  inclusion_date6 2020-07-11      4.32

过滤

d %>%  
filter(diff <= 10)
# A tibble: 4 x 5
ind cancer_date inclusion_id    inclusion_date  diff
<dbl> <date>      <chr>           <date>         <dbl>
1     1 2004-11-01  inclusion_date0 2014-11-01     10.0 
2     3 2011-01-14  inclusion_date2 2016-01-18      5.01
3     5 2009-01-02  inclusion_dat4  2018-12-11      9.94
4     4 2016-03-16  inclusion_date6 2020-07-11      4.32