Blog

Fonagy, P., Rost, F., Carlyle, J., McPherson, S.,… Taylor, D. (2015). Pragmatic randomized controlled trial of long-term psychoanalytic psychotherapy for treatment-resistant depression: The Tavistock adult depression study (TADS). World Psychiatry, 14, 312-321.

Usually I do not write about individual studies, mainly because meta-analyses and systematic reviews are much more reliable. But occasionally a unique study is published that is important enough to report. This is a rare trial that focuses on “treatment-resistant” depression defined as long-standing depression that has not responded to at least two previous evidence-based interventions. Depression is known to have a relapsing chronic course for about 12% to 20% of patients. And not responding to treatment is highly predictive of non-response to future treatment for depression. Fonagy and colleagues argued that in order to be effective, treatments for chronic and resistant depression need to be longer and more complex than current time-limited evidence-based approaches. Further, they argued that follow ups should be of longer duration. The authors tested a manualized long term psychoanalytic psychotherapy (LTPP). The treatment involved 60 sessions over 18 months provided by 22 trained therapists. In this trial, the “control” condition was treatment as usual (TAU) as defined by the National Institute for Clinical Excellence in the United Kingdom. TAU was made up of short term evidence-based interventions like antidepressant medications or CBT provided by licensed trained professionals. LTPP plus TAU was compared to TAU alone for 129 patients randomly assigned to one of the conditions. At pre-treatment, the majority of patients scored in the severe range on the Beck Depression Inventory (BDI) or the Hamilton Depression Rating Scale (HDRS). The average patient had 4 previous unsuccessful treatments for depression. No differences were found between LTPP and TAU at post treatment, but differences began to emerge after 24 months. Complete remission was infrequent in both conditions after 42 months (14.9% LTPP vs 4.4% TAU). However, partial remission at 42 months was significantly more likely in LTPP (30.0%) than TAU (4.4%). Patients were significantly more likely not to meet DSM-IV criteria for depression at 42 months in LTPP (44%) than in TAU (10%). The differences between conditions in mean BDI and HDRS scores were significant and medium sized indicating greater improvement with LTPP.

Practice Implications

This is the first study of its kind to test a manualized LTPP for treatment resistant depression. Patients in LTPP were more likely to maintain gains whereas those receiving evidence-based TAU were more likely to relapse. Although this is only one study and should be interpreted cautiously, it does suggest that chronic treatment-resistant depression is more likely to respond to longer and more complex treatment, and that outcomes of such treatment tend to be maintained in the longer term.

Lemmens, L.H.J.M., Arntz, A., Peeters, F., Hollon, S.D., Roefs, A., & Huibers, M.J.H. (2015). Clinical effectiveness of cognitive therapy v. interpersonal psychotherapy for depression: Results of a randomized controlled trial. Psychological Medicine, doi:10.1017/S0033291715000033

Generally, I prefer to report on meta analyses rather than individual studies mainly because findings from meta analyses are based on a larger number of studies and so are more reliable (see my November, 2013 blog). However, this study by Lemmens and colleagues represents a large clinical trial of 182 depressed patients who were randomized to cognitive therapy (CT), interpersonal psychotherapy (IPT), or a no-treatment control condition. The size of the trial provided the study with enough statistical power to test a hypothesis of non-inferiority of treatments. (A statistical note: A study finding of “non-inferiority” between treatments is sometimes unreliable because it is easier to detect such a finding with a small or poorly designed study. Studies with larger sample sizes provide greater statistical power, which in part makes a non-inferiority finding more reliable). A previous meta analysis showed both CT and IPT to be equally effective interventions for major depression. However, none of the studies in that meta analysis had sufficiently large sample sizes to reliably detect non-inferiority of interventions, none reported outcomes after post-treatment, and none of the studies used a no-treatment comparison condition. In their study, Lemmens and colleagues provided 16 to 20 sessions of individual therapy (45 minutes in length) to participants who met criteria for major depressive disorder. CT was based on Beck’s model and focused on identifying and altering cognitions, schemas, and attitudes associated with negative affect. IPT seeks to understand the social and interpersonal context of a patient’s depressive symptoms, and helps the patient to solve the interpersonal problem or change their relation to the problem, which may result in a resolution of the depressive symptoms. The study by Lemmens and colleagues was well designed in which: patients were randomized to conditions (CT, IPT, wait-list), 10 licensed therapists were expertly trained (5 CT therapists, 5 IPT therapists), and the therapies were competently delivered. Depressive symptoms significantly decreased for patients in both CT and IPT conditions with large effects, and these findings remained stable to 5 months post treatment. There were no differences between CT and IPT at post treatments and follow up, and both treatments were more effective than the waitlist control condition. Half of the sample had clinical improvements in symptoms, and 37% of patients were without depressive symptoms at 1 year follow up.

Practice Implications

CT and IPT did not differ in the treatment of depression in the short (post-treatment) and long term (follow up). The study does not address why two very different treatments led to similar positive outcomes. The authors suggest two possible reasons: (1) different specific treatment pathways led to similar results, or (2) change was driven by factors common to both treatments like motivation and therapeutic alliance.

Driessen, E., Van, H.L., Don, F.J., Peen, J., Kool, S. ....Dekker, J.J. (2013). The efficacy of cognitive-behavioral therapy and psychodynamic therapy in the outpatient treatment of major depression: A randomized clinical trial. American Journal of Psychiatry, 170, 1041-1050.

Psychotherapy is one of the most widely used treatments for major depression. Unfortunately there is no commercial entity like the pharmaceutical industry to support research and development of psychotherapy. As a result, researchers have limited ability to conduct larger-scale studies of comparative treatment effectiveness, of which there are only a handful. Although psychodynamic therapy (PDT) has been used to treat depressed patients for decades, randomized controlled trials of its efficacy are relatively infrequent. A concurrent problem with outcome research in psychotherapy is that sample sizes tend to be too small to actually test if two treatments are equivalent in what is called an “equivalency trial”. Without large samples, all one can conclude is that two treatments are “not significantly different” (a statistical note: an equivalency trial is planned from the outset to have a large enough sample to test the hypothesis that, with 95% certainty, the effect of one treatment falls within a narrow, predetermined margin of the effect of another treatment). The study by Driessen and colleagues was conducted in several sites in Amsterdam, in which 341 patients seeking outpatient psychotherapy for depression in psychiatric clinics were randomized to PDT or cognitive behavioural therapy (CBT). This is largest trial ever of PDT. Participants received 16 weeks of therapy and then were followed up for 1 year. About 40% of patients started with severe depression. Therapists were 93 experienced and well trained therapists who provided one of the two treatments. The main outcome was remission from depression, defined by achieving a low score on a validated observer rating scale. Post treatment remission rates were 21% for CBT and 24% for PDT, indicating that the treatments were equivalent.

Practice Implications

Cognitive-behavioral therapy (CBT) and short-term PDT provided similar outcomes for patients with a major depressive episode, but remission rates at the end of treatment were low for both treatments. Lower remission rates were likely due to the greater level of severity for these patients compared to those seen in primary care settings. The results highlight that even the best available psychological (and pharmacological) treatments yield modest outcomes for more severely depressed patients. Nevertheless, this rare equivalency trial found that both CBT and PDT were equivalent in terms of outcomes for these patients.

Author email: e.driessen@vu.nl