From trial and error to trial simulation. Part 2: an appraisal of current beliefs in the design and analysis of clinical trials for antidepressant drugs.

Study design factors are partly to blame for the high failure rate in trials with antidepressant drugs. Clinical trial simulation (CTS) allows the investigation of the influence of design characteristics on important aspects of clinical trials such as power and type I error. Using CTS scenarios, we evaluated the impact of population size, randomization ratio, frequency of assessments, dropout mechanisms, clinical end point, and statistical method on the outcome of clinical trials with antidepressant drugs. The results reveal that (i) an increase in the frequency of visits does not increase statistical power, (ii) a skewed randomization for a placebo or comparator arm may decrease statistical power, and (iii) analysis of the percentage of responders should be avoided. CTS should become best practice in the optimization of study design. To date, no other statistical approach has enabled such comprehensive evaluation of the factors contributing to study failure in depression.

Duke Scholars

Author Joseph Patrick Horrigan Psychiatry, Child & Family Mental Health & Community Psychia ...