delete line under condition R dplyr

I have a df like this one:

With in the first column the patient identifier (ID), in the 2nd the follow-up number, in the 3rd column the occurrence of a recurrence (yes/no) and in the last column the date of recurrence (if recurrence = yes).

There are therefore repeated lines for several patients, if they have several follow-ups.

I’d like to obtain a df with only 1 line per patient.

• If the patient has had several recurrences, I’d like to keep the line with the oldest recurrence (date of first recurrence).
• If the patient has had several follow-ups with recurrence and without recurrence, I’d like to keep the line with the (oldest) recurrence.
• If the patient has several follow-ups without recurrence, I’d like to keep follow-up 1.

In the end, I’d like to get this df here:

Here’s a reproducible example :

df <- data.frame(ID = c("15", "15", "16","29", "29", "34","34", "34", "36","36"),
                 FOLLOW_UP = c("1", "2", "1","1", "2", "1","2", "3", "1","2"),
                 RECURRENCE = c("no", "yes", "no","yes", "yes", "yes","no", "yes", "no","no"),
                 RECURRENCE_DATE = c("", "2003-05-15", "","2002-06-10", "2002-12-16", "2009-11-15","", "2012-05-03", "","")
)

I thought of using the “group by” and “slice” functions but I don’t know how to apply them with conditions.

Thanks in advance

>Solution :

something like this:

df$RECURRENCE_DATE <- as.Date(df$RECURRENCE_DATE, format = "%Y-%m-%d")
df_filtered <- df %>%
  group_by(ID) %>%
  arrange(desc(RECURRENCE), RECURRENCE_DATE, FOLLOW_UP) %>%  
  slice_head(n = 1) %>%
  ungroup()