ABSTRACT
At the heart of a randomized controlled trial is the comparison between intervention and control groups. ‘Intention to treat’ implies that the analysis of the two groups is on the basis of randomization, so that any difference in outcome is attributable solely to the intervention and no other factors. From the point of randomization onward, an individual is part of the trial, even if in retrospect he/she did not satisfy entry criteria. At randomization it is the investigators’ intention that an individual receive treatment/ placebo. Although some patients may not receive treatment/placebo, or complete the protocol, they have entered the trial, and must be included in the analysis. Failure to do so, compromises the study design, and biases the findings.
