I noticed the new update has 2050 symptomatic patients, whereas the old latestdata file had 3205 symptomatic patients. I merged the two datasets to see if I could recover the missing symptoms when I found out there are duplicate patients with different IDs.
Is it possible to add unique patient IDs? Unique patient IDs would be a great help to keep track of new updates.
Here are example patients that are duplicated. Some of them have updated dates of death/discharge, like 000-1-40992.
I noticed the new update has 2050 symptomatic patients, whereas the old latestdata file had 3205 symptomatic patients. I merged the two datasets to see if I could recover the missing symptoms when I found out there are duplicate patients with different IDs.
Is it possible to add unique patient IDs? Unique patient IDs would be a great help to keep track of new updates.
Here are example patients that are duplicated. Some of them have updated dates of death/discharge, like 000-1-40992.
debugging_beoutbreakprepared_data.xlsx
Best, Bruno