Figure 1.

A hypothetical example of target trial emulation applying cloning and censoring techniques. Notes: In observed data (A), patients with identification (ID) numbers 1 to 6 received treatment at some time after hospital admission. Patients with ID 7-10 were never treated during their hospital admission. Patients with ID 1, 3, 5, 7, and 9 died, and 2, 4, 6, 8, and 10 were discharged alive. Two clones of the observed data were created for each patient, with one clone assigned to each treatment strategy. (B) All patients were assigned to the X-treated arm in cloned sample B, and patients were censored when they deviated from the planned protocol. Patients with IDs 5 to 8 who were never treated were censored at the end of the grace period. (C) All patients were assigned to the non-X-treated arm in cloned sample C, and patients were censored when they deviated from the planned protocol. Patients with IDs 1 to 4 treated within the grace period were censored at the time they received treatment.