Although depression is the most common and costly mental health problem managed in primary care, treatment recommendations are based predominantly on studies of patients in subspecialty centers. Research indicates that primary care patients with major depression may have a different disease etiology and progression. This has led to concern about the relevance of current recommendations, particularly those based on drug efficacy. MacGillivray and colleagues reviewed the evidence for efficacy and tolerability of selective serotonin reuptake inhibitors (SSRIs) and tricyclic anti-depressants in the treatment of depression in patients managed in primary care.

The authors searched the Cochrane Collaboration database, reviewed reference lists of previously identified studies, and contacted experts to identify all studies comparing SSRIs with tricyclic antidepressants in adult primary care patients with depression. The primary outcomes were differences in depression scores and the proportion of patients who responded to treatment. Secondary outcomes were the total number of patients who withdrew from treatment and the number who withdrew because of side effects.

The authors identified 284 studies, of which only 11, involving a total of 2,954 patients, met criteria for inclusion in the review. Study participants were predominantly white Europeans averaging 40 to 45 years of age. About three fourths of the participants were women. The studies varied considerably in quality, and four failed to meet minimum criteria on at least one key methodologic component. All of the studies had some form of commercial sponsorship.

Only six studies met criteria for inclusion in the efficacy analysis based on changes in depression scores. Of these studies, only three reported data in an unambiguous format. Overall, the two classes of antidepressants did not differ significantly in efficacy. The slightly better performance of tricyclics that emerged when all six studies were analyzed disappeared when only the three unambiguous studies were analyzed. Three studies (totaling 740 patients) that reported clinical global impression as the measure of improvement also failed to show a statistically significant difference between the two classes of antidepressants.

Assessment of tolerability was based on six studies (2,375 patients). A significantly lower proportion of patients withdrew from treatment with an SSRI (20.7 percent) than from treatment with a tricyclic antidepressant (27.9 percent). The relative risk of withdrawal was calculated as 0.78 in favor of SSRIs. Seven studies provided data specifically on withdrawal because of adverse events. A significantly lower proportion of patients withdrew because of adverse events during treatment with SSRIs (11.6 percent) than with tricyclics (17 percent).

The authors emphasize that although only limited high-quality data are currently available, SSRIs and tricyclics appear to be comparable in short-term efficacy in primary care, but SSRIs appear to be better tolerated by patients. The authors call for much more high-quality research on the management of depression in primary care.