Family medicine
-
Randomization in randomized controlled trials involves more than generation of a random sequence by which to assign subjects. For randomization to be successfully implemented, the randomization sequence must be adequately protected (concealed) so that investigators, involved health care providers, and subjects are not aware of the upcoming assignment. The absence of adequate allocation concealment can lead to selection bias, one of the very problems that randomization was supposed to eliminate. ⋯ The goal of masking is to prevent ascertainment bias. In contrast to allocation concealment, masking cannot always be incorporated into a randomized controlled trial. Both allocation concealment and masking add to the elimination of bias in randomized controlled trials.