Focus on Alternative and Complementary Therapies
www.pharmpress.com/fact
Focus Alternat Complement Ther©2005 Pharmaceutical Press
Focus Altern Complement Ther 1999; 4: 111–4
Depression is a common psychiatric disorder, with one-third of the population estimated to have experienced an episode during their lifetime.1 It is also among the 10 most frequent indications for which complementary therapies are used, with herbal remedies, relaxation and exercise being the most popular treatments.2 A review of the scientific literature suggests that these particular therapies are also those for which the most convincing evidence of efficacy exists.
The herb St John’s wort (Hypericum perforatum) has been investigated in a number of randomised controlled trials (RCTs) involving depressed patients. A meta-analysis3 of 23 trials involving 1757 patients with mild to moderate depression found hypericum to be superior to placebo (odds ratio 2.67; 95% confidence interval 1.78–4.01) and to have a similar efficacy to conventional antidepressants (1.10; 0.93–1.31). There were methodological shortcomings with most of the trials included. Several more recent and rigorously designed placebo-controlled trials456–7 have provided further evidence of the efficacy of hypericum in mild to moderate depression. However, two recent RCTs comparing hypericum with tricyclic antidepressants8,9 did not provide compelling evidence that hypericum is as effective as synthetic treatments because of the possible underdosing of patients in the comparison group. One was the first such study9 to be undertaken in severely depressed patients and, although its findings were generally positive, they were not sufficiently conclusive to support the use of hypericum in patients with severe depression. Hypericum is associated with few adverse effects and has demonstrated a more favourable safety profile than conventional antidepressants.10 No rigorous research into other herbal treatments has been conducted.
There is a large body of evidence relating to exercise and depression, although many studies lack methodological rigour. A meta-analysis11 of 80 studies yielded an overall effect size of –0.53 (SD = 0.85), indicating that regular exercise decreased depression scores by half a standard deviation more than did comparison interventions. However, some of the studies in this meta-analysis involved normal subjects rather than patients with depression, so generalisations to the clinical population are difficult. A review12 of nine experimental exercise intervention studies (seven RCTs, all involving clinically depressed patients, reported positive results in all trials. Depression scores were reduced more in patients undertaking aerobic exercise than in no-exercise control groups, the reduction being as great as that achieved with psychotherapy. Unfortunately, findings in some trials may have been confounded by concomitant treatments. More recently, the results of a rigorous meta-analysis13 that included only trials involving clinically depressed patients and examined the effects of possible moderating variables were published. Thirty studies were included, producing an overall effect size of –0.72 (SE = 0.10). Exercise was found to be as effective as psychotherapy and behavioural interventions, and the antidepressant effect was the same for aerobic exercise as for non-aerobic types of activity. Larger reductions in depression were achieved in more severely depressed patients, which contrasts with reviews suggesting that exercise is of benefit only in mild depression. Several non-specific effects associated with exercise intervention, such as attention from the therapist, being part of a group and exercising in different venues, did not influence the antidepressant effect. However, because such a large number of physical, cognitive and social factors are potentially involved in the relationship between exercise and depression, with no specific mechanisms yet established, all that can be concluded from the evidence is that exercise programmes are associated with reductions in depression scores.
Three RCTs have investigated relaxation therapies in depression. The first study14 involved 30 depressed adolescents whose scores on the Beck Depression Inventory (BDI) decreased significantly more following 5 weeks of relaxation training or cognitive behavioural therapy compared with a waiting list control group. Another trial,15 involving 30 depressed outpatients, found that progressive muscle relaxation and a relaxation therapy based on yoga and auto-suggestion improved symptoms significantly more over 3 days than did a control condition in which patients talked about their troubles. In the final study,16 progressive muscle relaxation and cognitive behavioural therapy reduced BDI scores in 37 moderately depressed patients significantly more than did tricyclic antidepressants over a 16-week period. However, non-compliance with the pharmacological treatment was substantial. All these trials suffered from small sample sizes and failed to control for non-specific effects. The second one also had a very short treatment period and used a non-validated outcome measure.
Preliminary evidence for acupuncture in the treatment of depression comes from case series and uncontrolled trials. Three RCTs have compared electro-acupuncture with tricyclic antidepressants. The first trial17 included 47 depressed patients and found that the treatments significantly reduced scores on the Hamilton depression scale (HAMD) to a similar degree after 5 weeks. The second trial18 lasted 6 weeks and involved 241 patients. Again, HAMD scores were significantly reduced by the treatments to the same extent, with the rate of recurrence at 2–4 years’ follow-up not differing significantly between the groups. Similarly, in the third trial,19 involving 41 depressed patients, there was no significant difference between the treatments regarding the improvement in HAMD scores after 6 weeks. The preliminary results of a trial20 comparing standard electro-acupuncture with computer-controlled electro-acupuncture suggested that greater clinical improvements were found with the computer-controlled method. None of these studies controlled for the non-specific effects of acupuncture treatment, which may have novelty value compared with taking tablets.
There have been two RCTs relating to massage and depression. One trial21 involved 72 hospitalised children and adolescents with depression or adjustment disorder who either received back massages or watched a relaxing video for 30 minutes a day for 5 days. Profile of mood states (POMS) depression scores were significantly lower immediately after massage compared with pretreatment values and significantly lower after 5 days than in the group watching videos. The other trial22 involved 122 intensive care patients rather than people suffering from depression. The participants were randomised to either 30 minutes a day of massage, massage with lavender oil or rest for 3 days. There were no differences in physiological stress indices, but subjective ratings of mood on a four-point scale showed similar improvements in all groups, with the lavender oil massage initially demonstrating superiority in anxiety reduction compared with rest, although this effect was not sustained. Both these trials had very short treatment periods and the second one used a crude outcome measure. The only other study23 of aroma-therapy in depression was a small pilot trial involving 20 depressed patients receiving antidepressant medication. Twelve patients were exposed to citrus fragrance in the air for 11 weeks, and their use of antidepressants was substantially reduced (and in nine cases dispensed with) compared with that of the eight patients not exposed to the fragrance. This study was not randomised, had a very small sample and was confounded by the drug treatment.
Dance and movement therapy has been investigated in two trials. In the first one,24 12 depressed inpatients receiving antidepressant drugs and psychotherapy were randomised to undergo movement therapy on 7 out of 14 days. Compared with days without therapy, depression scores on the Depression Adjective Checklist were reduced following movement therapy in five patients. The other study25 involved 20 psychiatric inpatients and 20 control subjects, half of whom were randomised to a dance and movement therapy session while the other half received no treatment. Only the psychiatric patients who received therapy showed a significant decrease in anxiety and depression according to the Multiple Affect Adjective Checklist. In both these studies, the samples were small, treatment periods short and the non-specific effects (such as attention from the therapist) were not controlled for.
One RCT26 has investigated the effects of music therapy on depression. Thirty elderly depressed patients were allocated to either a therapist-administered music therapy programme, a self-administered music therapy programme or a waiting list control group for 8 weeks. Scores on the Geriatric Depression Scale, along with other measures, improved to significantly greater extent in both music therapy groups than in control subjects. This was a small trial that did not control for all non-specific effects of the therapy.
There are numerous anecdotal and case reports relating to the use of homoeopathic treatments for depression. However, a comprehensive review of homoeopathy trials27 uncovered only one unpublished trial of homoeopathy in depression.28 A positive result was reported in favour of homoeopathic treatment compared with dia-zepam in patients with mixed anxiety and depressive states in an open, randomised study. Without further details it is impossible to evaluate this study, but the lack of blinding is one obvious source of bias.
Most complementary therapies have not been thoroughly investigated in depressed patients, preventing conclusions about their efficacy as treatments for depression. There is, however, reasonably compelling evidence for a beneficial effect of Hypericum perforatum and to a lesser extent of exercise, although it remains to be determined whether exercise per se or related factors account for the antidepressant effect. Therapies such as relaxation, acupuncture and massage look promising but require further investigation in rigorous trials.
