Serotonin-norepinephrine reuptake inhibitors (SNRIs) may be used as an alternative treatment for depressed patients who do not tolerate or respond adequately to treatment with a conventional antidepressant. This randomized, open-label, multicenter study compared the effectiveness of the SNRI venlafaxine extended release (VXR) with that of conventional antidepressants (CA) in patients who were referred to an outpatient psychiatric specialty care setting for treatment after failure to tolerate or respond to at least 4 weeks of treatment with a CA in a primary care setting. Patients with a Hamilton Depression Rating Scale (HAM-D17) score > or =17 were randomly assigned to treatment with an alternative CA or VXR. Remission was defined as a score < or =7 on the HAM-D17. Efficacy analyses were carried out on 3,097 patients from the intent-to-treat (ITT) population (1,632 VXR; 1,465 CA). The antidepressants prescribed most frequently in the CA group were paroxetine (21.3%), citalopram (20.1%), sertraline (19.1%), fluoxetine (17.0%), and mirtazapine (7.9%). After 24 weeks of treatment, the VXR group demonstrated a significantly higher remission rate than did the CA group (59.3% VXR; 51.5% CA; P<.0001; odds ratio: 1.37; 95% CI: 1.19-1.58; P<.01). Despite the limitations of the open design, the results of this study suggest that venlafaxine extended release may be more effective than the conventional antidepressants used in this study when treating depressed patients who do not tolerate or respond adequately to treatment with a conventional antidepressant.