In Reply: Drs Cantor and Malani claim that
placebo effects can be demonstrated by comparing a pair of 2-arm trials of
active treatment vs placebo that have different proportions of participants
randomized to receive active treatment. This is an interesting speculation;
however, we are unaware of any data to support this claim or to indicate that
this design is just as able, or more able, to discriminate placebo effects
than would be a 3-arm trial comparing active treatment, placebo, and no treatment.
Their alternative design may suggest placebo effects but still leaves open
the critical question of whether either the active treatment or placebo-control
is better than no treatment.
Miller FG, Emanuel EJ, Rosenstein DL, Straus SE. Ethical Issues in Research in Complementary and Alternative Medicine—Reply. JAMA. 2004;291(18):2192–2194. doi:10.1001/jama.291.18.2193-d