Evaluating the tolerability of the newer antidepressants.

Given their equal efficacy, the choice of a specific antidepressant is largely influenced by side effect (SE) profiles. A number of new agents have recently become available. However, data directly comparing the side effects of these agents are scarce. As suggested by AHCPR guidelines, we used the 1998 Physicians' Desk Reference (PDR) to construct a comparison table using treatment emergent, placebo-adjusted incidence rates for the major (gastrointestinal, central nervous system, and sexual) side effects caused by nine antidepressants (fluoxetine, paroxetine, sertraline, fluvoxamine, nefazodone, bupropion SR, mirtazapine, venlafaxine XR, and citalopram). The results were tabulated to show the relative propensity of each drug to cause a particular side effect. Bupropion SR had the most favorable overall side-effect profile, and fluvoxamine the least favorable. However, there are several limitations in using the PDR to compare the newer antidepressants. Clinical studies directly comparing SEs of newer antidepressants are needed. Sexual SEs substantially affected total SE liability. A simplified summary table, with its advantages and some limitations, is not simple to construct. Pitfalls in this process are discussed.