The Zelen's randomized consent designs are modifications to a standard randomized controlled trial [Page 1658](RCT), in which participants are asked for consent to take part in the trial after randomization has taken place. Marvin Zelen proposed two modifications: In the single-consent design only those allocated to the intervention group are asked to give consent, and in the double consent design, those in both the intervention and comparison arms are asked for consent.

Randomized consent designs are widely used in the context of cluster randomized trials. With respect to individually randomized trials, they have found little favor with trial methodologists partly for ethical reasons and partly for scientific issues. Crossover is the main scientific problem with Zelen's approach, and even modest crossover rates will result in biased ...