Observations where the event of interest (e.g., patient death) hasn't occurred by the end of the study period.