Assessment of scientific and ethical issues in two randomized clinical trial designs for patients with Tourette's syndrome: a model for studies of multiple neuropsychiatric diagnoses.

Most clinical trials in Tourette's syndrome (TS) involve fewer than 60 patients. This is partially due to difficulties recruiting patients who have multiple neuropsychiatric diagnoses. Few studies permit treatment of comorbid diagnoses or compare active treatments. As a result, standard clinical practice requires choices between multiple agents shown to be superior to placebo agents in highly selected samples but never compare the two. Clinical practice also requires use of untested medication combinations. The authors review scientific and ethical shortcomings of placebo-controlled, monotherapy trials in TS, proposing specific conditions and methods and discussing the scientific, ethical, and economic implications of an alternative design: randomized, active-comparator, rater-only blinded clinical trials.