A worldwide database was employed to assess the effect of paroxetine on symptoms of anxiety and agitation associated with depression. Data derived from the use of paroxetine (n = 2963), placebo (n = 554) and active control (n = 1151) in short-term clinical trials were compared. Paroxetine and active control were significantly better than placebo in reducing psychic anxiety. However, paroxetine was superior to active control from week 2 onwards. Both paroxetine and active control had a beneficial effect on somatic anxiety, but this effect was seen earlier with active control. Neither paroxetine nor active control caused emergent (new) anxiety symptoms. Paroxetine had a more robust effect in reducing baseline symptoms of agitation when compared with active control, while both therapies protected from emergent (new) agitation, when compared with placebo. There was no difference between the three groups in the spontaneously reported adverse events indicative of increased arousal. The use of major and minor tranquillizers in the paroxetine and active control groups was similar.