creating a dummy variable based on dates for longitudinal data in R

Question

I have longitudinal patient data in R. I would like to create the new_dummy variable just like I demonstrated in the table below. I would like the dummy_variable value to be 0 if the consultation_date variable is smaller than the registration_date variable and to be 1 if the consultation_date variable is bigger than the registration_date variable.

patid	consultation_date	registration_date	new_dummy
1	07/07/2016	07/07/2018	0
1	07/07/2019	07/07/2018	1
2	14/08/2016	07/09/2016	0
3	07/05/2015	19/02/2016	0
3	02/12/2016	19/02/2016	1

Answer 1

If your columns are in the right class (Date), you can simply use

df$new_dummy <- as.numeric(df$consultation_date < df$registration_date)

To convert them in the right format, it's:

df$consultation_date <- as.Date(df$consultation_date, "%d/%m/%Y")
df$registration_date <- as.Date(df$registration_date, "%d/%m/%Y")

(thanks @akrun for the date format)