Type II Error and Antidepressants

Letter to the Editor

Sir: We read with interest the study conducted by Weihs et al., which concluded that bupropion sustained release and paroxetine were similarly effective in elderly, depressed patients treated for 6 weeks. However, the authors’ failure to observe statistical significance between interventions does not preclude the possibility of a clinically meaningful difference between the 2 agents. We suggest consideration of the possibility of a type II (β) error or a falsely negative trial.