notes by dr Claudio Italiano
Systematic reviews have documented that antidepressants are effective in the
acute treatment of depressive disorders of any severity, at different levels of
care and regardless of the presence or absence of concomitant physical diseases.
We found no evidence of clinically significant differences in the positive
effects of various antidepressant drugs, although there are differences in
adverse effects. A systematic review of randomized trials of variable
methodological quality found that the use of hypericum is an effective treatment
for mild or moderate depression.
Two systematic reviews found that electroconvulsive therapy is effective in
treating depression. Interpersonal therapy improves symptoms of mild or moderate
depression. Randomized trials of smaller size found that problem solving
interventions are more effective than placebo.
We have not found reliable evidence that one type of treatment (pharmacological
or not) is superior to another. Limited evidence indicates that the combination
of pharmacological and psychological treatments may have significant additional
positive effects in severe depression, but not in mild or moderate depression.
We found insufficient evidence to support the effectiveness of other treatments,
including exercise, bibliotherapy, socialization therapy and non-directive
counseling.
Among all the interventions examined, antidepressant drugs are the only
treatment whose efficacy has been proven in severe depressive disorders and
psychotic elements.
We did not find randomized trials comparing pharmacological and
non-pharmacological treatments in severe depressive disorders. A systematic
review and subsequent randomized trials have found that the continuation of
antidepressant drug therapy for a period of 4 to 6 months after healing reduces
the risk of relapse. We found no evidence of differences between treatments in
terms of long-term positive effects.
The studies compared cognitive therapy and antidepressants in subjects with mild
to moderate depressive disorders. Globally, 30% of subjects treated with
cognitive therapy and 60% of those treated with antidepressants had relapses.
However, the number of subjects included in these studies was too much.
We found another small randomized trial (40 subjects) that compared cognitive
therapy with standard clinical management for residual depressive symptoms in
subjects who responded to antidepressant therapy. Also this study reported that,
after 2 years, in the group treated with cognitive therapy the number of
subjects who had relapses was lower than in the group treated with
antidepressants.
Antidepressant drugs are effective in the acute treatment of depressive
disorders of any severity level. We found no clinically significant difference
in the efficacy of different antidepressant drugs.
However, the drugs differ due to the adverse effects they cause. On average,
selective serotonin reuptake inhibitors (SSRIs) appear to be better tolerated
than traditional antidepressants, but the differences are not marked. We have
not found convincing evidence that the use of fluoxetine is related to an
increased risk of suicide.
The abrupt cessation of treatment with SSRI is associated, in some cases, with
symptoms such as vertigo and rhinitis; in part this happens more easily (and is
probably more serious) with short-term drugs such as paroxetine. A systematic
review found that heterocyclic antidepressants and SSRIs are effective in mild
to moderate depression.
However, the overall effects of treatment are modest. The limitations of this
review include the heterogeneity of the observed populations and the short
duration of numerous studies. For ethical reasons, high quality randomized
trials are unlikely in the future.
Antidepressant drugs in young people vs placebo
All studies compared 2 antidepressant drugs and included a placebo-controlled
control group.
The review found an average therapeutic effect size of 0.50 for antidepressants
compared to placebo: this means that in 69% of patients treated with placebo the
results were worse than in the average patient on antidepressants.
Antidepressant drugs were more effective in subjects with depressive disorders
diagnosed using standard criteria (mainly those defined in the review of the
third edition of the Diagnostic and Statistical Manual of Mental Disorders,
DSM-III-R). Other studies have compared antidepressants and placebo in patients
with dysthymia (mild, chronic depressive disorders). The response to treatment
was approximately double in the antidepressant group.
We have identified 3 systematic reviews that compared SSRIs and tricyclics. The
reviews revealed no significant difference in overall effectiveness. Overall,
SSRIs appear to be slightly better tolerated on the basis of the number of
subjects who have dropped out of studies.
Other studies have compared the effectiveness, in the context of basic care, of
the newer antidepressants compared to placebo and compared to more traditional
antidepressants. The mean response rate was 63% for the most recent compounds,
35% for placebo and 60% for tricyclics (relative risk with SSRI vs placebo 1.6,
95% confidence limits from 1.2 to 2.1) . A small subsequent randomized study
(152 subjects), which compared adherence to treatment with dotiepine and
fluoxetine over a 12-week period, found no significant difference between the 2
drugs.
Monoamine oxidase inhibitors (IMAOs) vs tricyclics: We found a systematic review
(date of research not reported, 55 randomized trials that compared IMAO and
tricyclics in different subgroups of subjects with depression).
The review found that MAOIs were less effective in individuals with severe
depressive disorders, but may be more effective in atypical depressive disorders
(depressive disorders with opposite characteristics, such as increased sleep and
appetite, reactivity, sensitivity to rejection). Significant positive effects
were also reported with fluoxetine, trazodone and phenelzine.
Association of antidepressants and benzodiazepines vs monotherapy with
antidepressants
We found a systematic review comparing combined treatment with antidepressants
and benzodiazepines compared to antidepressant-based monotherapy.
After one week, the combination therapy was more likely to produce a response
than the use of antidepressants alone, but after 6 weeks there were no
differences between the two treatments. In subjects suffering from both
depression and physical illness vs placebo:
A systematic review reported that antidepressants were more effective than
placebo in subjects with depression and physical illness (number of cases to be
treated 4, 95% confidence limits from 3 to 7). Subjects treated with
antidepressants were more likely to leave the study (number of cases to be
treated to have an individual who left the study compared to the placebo group
10, 95% confidence limits from 5 to 43).
In severe depression, but not in mild or moderate depression, the addition of
pharmacological treatment to interpersonal or cognitive therapy is more
effective than psychological therapy alone.
A non-systematic meta-analysis of 6 randomized trials (595 subjects) did not
identify any benefit in the association of specific pharmacological and
psychological treatments in mild to moderate depressive disorders, but found
that in the more severe forms of depression the combination of therapy
pharmacology with interpersonal therapy or cognitive therapy was more effective
than interpersonal or cognitive therapies alone.
A recent randomized study (681 adults with chronic depressive disorders)
compared nefazodone and cognitive-behavioral analytical psychotherapy, either
alone or in combination.
Insufficient evidence from a small randomized study indicates that socialization
interventions can reduce symptoms of depression.
We found a small randomized study (86 women with chronic depression) on
socialization interventions conducted in London. The initial identification was
made through a postal selection of women enrolled in basic care, but not
actively followed by their doctor. The study found that women treated with
social interventions were more likely to experience symptom relief after a
period of 13 months (65% with socialization vs. 39% in the control group.