Placebo effect in clinical trials

Abstract

It is well known that placebos evoke a variety of effects in the human body. It is less known that placebo medication can produce adverse drug reactions (ADRs). The efficacy and, as the main topic, the tolerability were investigated from an international clinical data pool on placebo-controlled randomized multicenter studies of 5 different groups of indications. The therapeutic areas analyzed were neuropsychiatry (nimodipine, ipsapirone), cardiology (nisoldipine), metabolism (acarbose) and gastroenterology (hydrotalcit). The placebo efficacy was evident and varied not only in comparison to 5 indication groups analyzed, but also within them. Placebo showed for example hardly an effect on the symptomatology of stroke patients with a severe neurological deficit in comparison to verum while there was a 50% improvement in symptoms following placebo application in patients with a mild stroke. In angina patients exercise tolerance increased under placebo by 10% (time to onset of ST segment depression and symptoms of angina) while verum showed a 22% improvement. In diabetes placebo had no effect as compared to baseline and to a verum on blood glucose and HbA1C. ADRs could be observed under placebo treatment. The frequency and kind of placebo-induced ADRs varied also between the indication groups, i.e. typical CV side-effects were observed in CV controls. The placebo ADR profile was similar to the reference drug also. Some ADRs were reported even more frequently under placebo than under verum, like "dry mouth" in patients with general anxiety status. Placebo therapy is frequently effective and is not a "non-therapy". Placebo effects can be shown only by direct comparison with a "non-therapy". A placebo therapy is frequently accompanied by adverse drug reactions (ADRs) just as a verum therapy. Placebo ADRs are frequently illness- and verum-specific. Effects and ADRs of placebo therapy must be known, to judge the effect of verum medication in controlled clinical trials. The mechanisms of placebo effects are manifold (e.g. endorphine release, conditioned learning). Since placebo therapy outside of evidence-based medicine (use of drugs without proven efficacy = pseudoplacebos) may potentially also cause serious side effects, the use may not only be useless but also harmful.