Compare two studies of a medicine’s effectiveness.  In the first study there was a placebo control group.  Subjects who actually got the medicine believed with 50% probability that they were taking a sugar pill.  In the second study there was no placebo control.  Those who got the medicine knew it.

Those who actually got the medicine had better outcomes when they knew it than when they were unsure.

Our group at Columbia has completed preliminary work involving metaanalyses of randomized controlled trials comparing antidepressant medications to a placebo or active comparator in geriatric outpatients with Major Depressive Disorder (Sneed et al. 2006). In placebo controlled trials, the medication response rate was 48% and the remission rate 33%, compared to a response rate of 62% and remission rate of 43% in the comparator trials (p < .05). The effect size for the comparison of response rate to medications in the comparator and placebo controlled trials was large (Cohen’s d = 1.2).