Efficacy and Tolerability of Tricyclic Antidepressants and SSRIs Compared With Placebo for Treatment of Depression in Primary Care: A Meta-Analysis

Arroll and colleagues provided a meta-analysis of placebo-controlled trials of TCAs and SSRIs for depression, with a focus on studies carried out in the primary care setting. The paper focuses on “efficacy” studies and is useful in calling attention to the idea that antidepressants work in primary care; at the same time, I would like to raise several issues related to primary care and depression that were not addressed.

The notion of generating evidence from primary care settings compared to specialty mental health settings is important. In effectiveness research (in contrast to efficacy research) we may assume medications have an effect but we need to study the patient, family, and practice factors that relate to the effectiveness of treatment [1]. Such factors may contribute to how well medication works or even if medicine is accepted. In many effectiveness studies, “usual care” is often considered as a “placebo” or “control” group, but clearly usual care means different things in different practices or to different physicians. There is room for improvement in that Arroll and colleagues concluded that only 56 to 60% respond to active treatment.

Recognizing that the study focused on whether medications had any effect on depression outcomes as assessed by standardized instruments, our work has drawn attention to other processes in primary care, such as processes involved in identification of depression and acceptance of diagnosis across ethnic groups [2]. Testing in placebo-controlled fashion whether a drug has an effect is one thing, providing the services to accompany the prescription is another. Since physicians, especially primary care physicians, prefer to have depression care integrated into the primary care practice [3], how to design practice to improve depression care will be a critical future need [4].

1. Bogner HR, Cary MS, Bruce ML, Reynolds CF, Mulsant B, Ten Have T, Alexopoulos GS, and the Prospect Group. The role of medical comorbidity on outcomes of major depression in primary care: The Prospect Study. American Journal of Geriatric Psychiatry. 2005, October.

2. Gallo JJ, Bogner HR, Morales KH, Ford DE. Patient ethnicity and the identification and active management of depression in late life. Archives of Internal Medicine. 2005; Sept 26; 1962-1968.

3. Gallo JJ, Zubritsky, C, Maxwell J, Nazar M, Bogner HR, Quijano LM, Syropoulos H, Cheal KL, Chen H, Sanchez H, Dodson J, Levkoff S, and PRISM-E Investigators. Primary care providers evaluate integrated and referral models of behavioral health care for older adults: results from a multisite effectiveness trial (PRISM-E). Annals of Family Medicine. 2004 July/August;2(4):305-309.

4. Dietrich AJ, Oxman TE, Williams JW Jr, Kroenke K, Schulberg HC, Bruce M, Barry SL. Going to scale: re-engineering systems for primary care treatment of depression. Annals of Family Medicine 2004 Jul- Aug;2(4):301-4.

Competing interests:   None declared

Arroll et al are to be congratulated for doing a Cochrane style review of the efficacy of antidepressants (TCAs or SSRIs) for depressed primary care patients. As the authors point out, there is evidence of qualitative differences between depressed patients in primary care and those seen in specialty settings, and one can not assume that the effects of recommended treatments will be the same. The authors' data synthesis thus has particular clinical significance for primary care practice.

Refreshingly, their systematic review finds evidence for the effectiveness of both TCAs and SSRIs in primary care depressed patients. The findings support what for many years has been recommended practice (treat with antidepressants) to primary care providers. The findings also lay to rest the notion that this medication treatment must be provided by a psychiatrist to be effective, as treatment effects were present when the medication was provided by primary care providers.

All of the above is good news. However, to this reader, the worrisome - or at least thought-provoking - news is that the effect size, although statistically significant, is relatively modest : 55-60% improvement for active medication vs 42-47% for placebo. That 40-45% do not improve on active treatment raises important questions, both for future research and for clinical practice. I for one would accept that the evidence is in that antidepressants work in primary care depressed patients; an additional important question is for whom? What are factors which predict which patients will respond (? particular symptom patterns; ? depression symptom severity; ? family history; ? what else). And, for that matter, what factors relate to improvement for those 42-47% who respond to placebo? Such research, and its subsequent findings, would have considerable clinical utility for primary care practitioners in helping them decide which management would be helpful for which patient. This review highlights the need for such research.

A minor point, but puzzling to me: not included in the review was a randomized placebo controlled trial of an SSRI (paroxetine) in primary care patients age 18-59 with dysthymia or minor depression carried out by myself and John Williams and reported in the Journal of Family Practice in 2001. Our results were generally consistent with those in this review - effectiveness of the SSRI for remission in primary care patients with dysthymia and a high remission rate (61-66%) with all treatments for minor depression.

Barrett,JE, Williams, JW, Oxman, TE, Frank, E, Katon, W, Sullivan, M, Hegel, MT, Cornell, JE, Sengtupa, AS.Treatment of dysthymia and minor depression in primary care. J Fam Practice 2001; 50: 405-412.

Competing interests:   None declared