To reduce the chance of mis-dosing

To reduce bias

To reduce the cost of the study

To ensure the randomization schedule is followed