A control group that is chosen from a group of patients who were observed at some previous time.