The enhanced sensitivity of the elderly to the side effects produced by tricyclic antidepressants (TCAs), and the frequency and type of adverse events, have made the treatment of depression in this group difficult. The selective serotonin reuptake inhibitors (SSRIs) have been reported to produce significantly fewer undesirable side effects and display better tolerance than TCAs. We compared the therapeutic actions and side effects produced by citalopram, the most selective SSRI available, with amitriptyline in a group of elderly patients (aged 65 and older) diagnosed with major depression. In a double-blind, double-dummy, parallel-group, multicenter comparison of citalopram (20 or 40 mg/day) and amitriptyline (50 or 100 mg/day), patients who did not respond to placebo during a 1-week single-blind phase were randomly assigned to receive citalopram or amitriptyline for 8 weeks. Efficacy measures included the Montgomery-Asberg Depression Rating Scale (MADRS), the Hamilton Depression Scale (HAMD), and Clinical Global Impressions. Both drug treatments produced equivalent time-related declines in severity of depression, so that by 8 weeks slightly more than 50% of the patients in each group experienced marked recovery, defined as MADRS scores < or = 12. Amitriptyline produced a greater overall incidence of adverse events, including a significantly higher (P < 0.001) percentage of patients reporting dry mouth (34% vs. 7%), as well as a significantly higher (P < 0.02) incidence of somnolence. Constipation and fatigue also occurred more frequently in the amitriptyline than in the citalopram group. For only one event (nausea) did the citalopram group report a significantly greater (P = 0.012) incidence (12.8% vs. 4.8%). On the basis of these results, it was concluded that citalopram is as effective an antidepressant as amitriptyline in the treatment of the depressed elderly. Because of its low incidence and low magnitude of side effects, citalopram seems especially useful in private practice.