Statistics from Altmetric.com
QUESTION: How do anti-depressants and psychotherapy compare in generating full remission in major depressive disorder?
Systematic review with meta-analysis.
Studies were identified using Medline and PsychInfo (to November 2000).
Eligible studies were randomised controlled trials with blinded participants and assessors that (1) directly compared medications, psychotherapy and controls for well defined major depressive disorder; (2) included remission data. 6 studies were included with 883 outpatients with mild to moderate non-psychotic major depressive disorder. Mean age ranged from 32–40 years; 77% women (range 68–85%). Median treatment duration was 16 weeks (range 10–34).
Data were extracted on baseline measurements, sample size, duration of treatment, remission, follow up and dropout rates. The main outcome was full remission.
Tricyclic antidepressants, phenelzine and psychotherapy (primarily cognitive behaviour and interpersonal therapies) were more effective at promoting remission than control conditions. There was no difference in remission between antidepressants and psychotherapy regimens (table).
Antidepressants and psychotherapy may both be effective first-line treatments for outpatients with mild to moderate depression.
This review addresses several concerns about previous meta-analyses, including having a control condition as an internal validity check and a well-defined diagnosis of major depressive disorder (instead of including people with minor depression or subsyndromal depressive conditions). It also focuses on clinical remission, usually defined as a score on a depression scale within the normal range. There is evidence that residual depressive symptoms, even after initial response to treatment, are associated with poor psychosocial function and quality of life and higher risk of relapse, chronicity and suicide.1–2 Many trials now report remission rates in addition to depression scores and response rates. Newer evidence-based guidelines list full remission as an objective of acute phase treatment.3
Most previous meta-analyses have found no difference between psychotherapy and pharmacotherapy for reducing depression scores among people with mild to moderately severe depression. This meta-analysis reinforces that pharmacotherapy and psychotherapy are similarly effective in achieving symptom remission.
Casacalenda et al acknowledge the limitations of their review. As multi-arm studies are difficult to conduct and fund, only 6 studies were identified, all of which involved tricyclic antidepressants or a monoamine oxidase inhibitor. These medications are no longer considered first-line treatments for depression, especially of mild to moderate severity. Although previous meta-analyses have found no differences between these older medications and the newer antidepressants, those studies did not focus on remission. There is evidence that remission rates may differ across classes of the newer antidepressants.4 This may impact on comparisons with psychotherapy.
Consistent with past trials and reviews, the average remission rate of the monotherapies was about 46%. While the remission rates may be higher in clinical practice, this suggests that up to half of those treated for major depressive disorder may not achieve remission with monotherapy. We need to move away from simple comparisons of psychotherapy versus pharmacotherapy to consider combined therapy. The advantages of combination treatment may be more apparent using remission as an outcome measure. There is preliminary evidence that combining antidepressant medications with psychotherapy may increase remission rates over monotherapy in severe depression5 and in chronic depression.6
In most studies, remission is defined only at a single point in time, such as at the final treatment visit. Sustained remission or recovery may be more important than the point prevalence of remission. Quality of life should also be assessed in trials since the ultimate goal of treatment is remission of symptoms and restoration of premorbid psychosocial function.
For correspondence: N Casacalenda, Institute of Community and Family Psychiatry, Jewish General Hospital, Montreal, Canada.
Source of funding: None specified.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.