According a metanalysis study by Robinson L. & al. (Psychol. Bull., 1990), the average Effect Size of psychotherapy (mainly CBT) of depression as measured against no treatment controls is 0,73 at the end of treatment and 0,68 at follow-up. In my metanalysis, the ES of CBT against no tretment controls (for all conditions) was 0,69 (Sibilia L. Metanalysis of Psychotherapy outcomes: Limits and Results. Invited keynote lecture, 3rd European Congress of Psychology, Tampere, 1993, July 5-9. See also: Sibilia L. (1994). Meta-analysis in the assessment of psychotherapy outcomes: limits and results. Psychologie Europe, 1, 4: 45-60).

This recent paper reviews several different types of treatment and reports that in blind studies all interventions are better than doing nothing but none are more effective than placebo

Khan A, Faucett J, Lichtenberg P, Kirsch I, Brown WA (2012) A Systematic Review of Comparative Efficacy of Treatments and Controls for Depression. PLoS ONE 7(7): e41778. doi:10.1371/journal.pone.0041778

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0041778

The Kahn paper concludes that all depression treatments are equally effective. It seems not logical to only reimburse antidepressants. This is the case for Belgium, don't know for other countries. Maybe a cost-effectiveness analysis could shed some light? Who can point me a best available evidence report? Thank you.

check out Steve Hollon (of Vanderbilt Univ)

To Colin: The Kahn study does not report that none of the interventions is more effective than placebo! Their conclusion is: "These data suggest that type of treatment offered is less important than getting depressed patients involved in an active therapeutic program. Thus, treatment type might best be chosen on the basis of differences in the clinical presentations, risks and patient preferences and acceptance." An active therapeutic program cannot be called a placebo.

To Frederik: You may like another recent paper in Plos|Medicine:

Barth J, Munder T, Gerger H, Nuesch E, Trelle S, et al. (2013). Comparative Efficacy of Seven Psychotherapeutic Interventions for Patients with Depression: A Network Meta-Analysis.

http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.1001454;jsessionid=6264179C1C8EFD4192146A7D9539F51D

Hi Peter - sorry for not getting back to you about our other correspondence - I have been offline for a while and it seemed that too much time had passed to make a reply worthwhile - as I say, apologies.

Re the Kahn paper - Figure 2 (http://www.plosone.org/article/fetchObject.action?uri=info:doi/10.1371/journal.pone.0041778.g002&representation=PNG_M) is the crucial one - it shows that doing nothing has the worst outcome ('waiting list') and that anything else produces about a 40-50% improvement.

'Anything else' includes placebo and no intervention is more effective than that (I discount the 'non-blind' studies reported since these cannot be considered reliable) and although some of the bars are slightly higher than others, these differences would not be statistically significant. The only conclusion therefore is that every intervention that we currently have is based on a placebo effect.

Placebo isn't the same as 'no effect' - in this case it produces a reliable 40-50% remission but these findings do mean that if we're looking to improve our treatment of depression we have to look at this problem in a different way - and in order to do that a paradigm shift is required.

I would say what is perhaps obvious: Placebo isn't the same as no treatment; it is a treatment whose essential ingredient is hope, with any neurophysiological sequelae that follow from hope. But one question has always dogged me: What does a placebo treatment consist of in psychotherapy? Perhaps simple placement on a wait list, since this act can instil hope. Any other type of placebo treatment?

Thanks Terry.

Keith Dobson and Steven Hollon report on an RCT that includes a cost-effectiveness analysis. For depressions lasting longer than 9 months, psychotherapy (20 sessions) has the best cost-effectiveness. Anyone knows a meta-analysis on the cost to heal depression?

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2648513/

Thanks Peter.

the Barth paper again confirms that it doesn't matter what type of psychotherapy one uses. Their placebo effect seems smaller than the Kahn paper, probably due to unblinding?

Collin,

I agree that we must make a paradigm shift, if we want to 'understand' placebo. I think it points towards the unconscious, which is by its very definition 'that what cannot be conscious'. Hard to grasp under conventional scientific methods. What paradigm shift do you think would be needed?

As I suggested above, placebo turns around hope and expectancy. I once read an article about the neurophysiological correlates of hope but cannot remember the title. As for the unconscious, all emotions can be unconscious, including the emotions associated with hope. I have recently dealth with a client who, on the SURFACE, is devasted and deeply and chronically depressed by the premature death of her son 1 year ago. I can assure you that this person, despite her insistance that all his hopeless and death would be a welcome release from her suffering, this person shows various signs of hope, ranging from scrpulous care of her long-term health to planning and looking forward to a vacation. Believe it or not this person has not the least consciousness that she is exhibting hope, and I am convinced that she would be disturbed by the idea that, despite the tragedy, she is hopeful and believes in (her) life. She would be disturbed because, to put it briefly, the more she is conscious of a good feeling (like hope), the worse she feels, i.e. she MUST remain unconscious of her hope to avoid the pain of GUILT, for what right does she, at the age of 63, have to survive her son, who died at the age of 38, especially given that she suggested that he undergo a routine operation for obesity that went wrong due to (proven) medical negligence and killed him. The worst thing I could do would be to try to give her hope. The best thing I can do is never to contradict her assertions that she is hopeless. By the way Collin, an unconscious idea or emotion is not an idea that necessarily cannot become unconscious; it is an idea or emotion that under certain circumstances can and should become conscious, and under other circumstances, such as those of my client, can not and should not become conscious.

To Colin: I agree that placebo pills leads to substantial symptom reduction (38%), according to blinded raters, which is significantly better than waiting list (13%), but psychotherapy, antidepressants and their combination do result in significantly higher percentages of symptom reduction, resp. 47%, 46% and 52%. Perhaps these differences do not impress one as clinically significant, but your statement that "although some of the bars are slightly higher than others, these differences would not be statistically significant" is not in agreement with table 3 of the authors.

What I am missing in this study is the average number of subjects considered "substantially improved", because that figure may show that the combination of anti-depressants and psychotherapy affects a higher percentage of patients positively, rather than effectuating a higher degree of symptom reduction within the same percentage of patients. But I suppose that figure is missing in most of the original studies reviewed.

Do read my APT papers on placebo and they will explain why these findings keep happening consistently.Advances in psychiatric treatment (2013), vol. 19, 171–180 doi: 10.1192/apt.bp.112.010405Rethinking placebo in psychiatry:

how and why placebo effects occur†

Daniel McQueen, Sarah Cohen, Paul St John-Smith & Hagen Rampes

Rethinking Placebo in Psychiatry I The range of placebo effects.

Rethinking Placebo in Psychiatry

IIHow and Why Placebo effects ultimately occur?

Advances in Psychiatric Treatment APT 2013Dr Daniel McQueen.Dr Sarah Cohen Dr Paul St John-Smith Dr Hagen Rampes.

Paul st John-Smith

Dear Peter - there's a need to look at the distributions as well - the normal rule of thumb is if the confidence intervals overlap then the groups are probably not different - I spent some minutes trying to describe what I'm seeing in words but that didn't work so I hope you don't mind that I've done a quick edit of the figure to try and show what I mean (http://img.photobucket.com/albums/v462/chendrie/Fig2.jpg).

The added light blue line shows the top limit of the confidence interval for the placebo group - and as you can see, this is above or crosses through the confidence intervals of every group except combination (as in my other post I'm ignoring the non-blind studies - the red bars) meaning that none of these are likely to be significant (or if they are, the difference is only slight and probably not clinically meaningful - which is what Table 3 shows).

That might lead to the conclusion that the combination therapy IS significant but a single placebo is not a true control for the combination group since they had two interventions - hence the control needs to have two placebos before conclusions can be drawn.

I think my main point is that whatever the just noticeable differences that can be seen or not seen, none of the treatments produce a complete relief from depression which is what we really want.

By analogy, no-one would be satisfied with anaesthetics that were only capable of reducing severe pain by 50% or so (whether this was statistically significant or not) but that is the situation we have found ourselves in with antidepressant therapies

Hi Colin, still the difference is statistically significant (which you denied), even if not clinically impressive.

Apart from that, I agree that about 50% symptom reduction is but a meager result, although I don't think therapies need to end in complete recovery to consider them effective. Depression is often related to problems in life, but these are lifelong challenges for all people, so a treated patient needs not be a very happy guy. Just functioning reasonably well is good enough, methinks. And what is wrong with being a little bit melancholy. We owe beautiful music to that.

Another question is whether symptom reduction is a good recovery criterion in the first place. To a certain extent yes, of course, but why not use a measure of level of functioning in general? Like GAF in DSM-IV, but then operationalized thoroughly and measured by a valid instrument.

As you say Peter but part of the problem with clinical psychopharmacology has been that they have been so desperate to find things that really aren't there they've been willing to accept just noticeable statistical significance instead of clinical improvement that can be seen with the naked eye.

This blog by Jim Coyne outlining one of his papers is a great example - he argues that antidepressant drugs ARE effective because there is one less suicide in every 650 at risk patients that are given antidepressants (http://blogs.plos.org/mindthebrain/2013/06/20/do-rising-rates-of-antidepressant-prescription-translate-into-lower-rates-of-suicide-evidence-from-29-countries/). That one life is of course important for that individual and their family but many others are dying because they are deriving no benefit from these drugs.

Hence, my argument is that the more we try to hang on to the idea that current antidepressants are in any way fit for purpose the longer it will be before we start to develop better ones - and that will be long enough given that most drug companies are no longer researching in this area.

By the same token, the differences in the Khan paper may be statistically significant using the approach they have used but many would argue that more stringent tests should be applied - the 'signal', such as there is is very weak and would be filtered out with only a slightly higher p value - so, as you say, statistically significant without being clinically significant.

For me, given the almost complete lack of progress over the past 60 years, clinical significance should now be the only test that is allowed to count.

One thing that might help would be to stop considering depressive patients as a homogeneous group. Perhaps a minority would profit most from a SSRI or another drug, but most of them would profit from psychotherapy, adjusted to the ins and outs of their problems in living. And do not measure the effectiveness of it by mere symptom reduction. There is more to recovery than that.

I agree Peter - and think the issue here is for patients to find the treatment that works for them - drugs of any kind, psychotherapy of this sort or that - in practical terms it makes no difference if this is a placebo effect or not, just so long as it works for them.

However, that is a long way from knowing the answer and why I say for research purposes we must tear up what we think we know and start again

In the meantime we can only work with what we've got ...

Frederik, I would suggest the following paper, where they present a very good benchmark for what is an efficacious psychotherapy treatment and what are the values obtained by "natural history of depression" (i.e., no treatment condition). I think that this is what you looking for, right? Minami, T., Wampold, B. E., Serlin, R. C., Kircher, J. C., & Brown, G. S. (Jeb). (2007). Benchmarks for psychotherapy efficacy in adult major depression. Journal of Consulting and Clinical Psychology, 75(2), 232–243. doi:10.1037/0022-006X.75.2.232

Hi Collin,

Re psychopharmacological trials, isn't it a matter of confusing statistical with CLINICAL significance? Failure to make this distinction can have a sinister motive, namely to mislead the reader. Several years ago the editorial boards of respectable journals took pharmaceutical companies to task, warning them that they would no longer accept such chicanery.

Colin - I like your statistical rigour. The Kirsch meta-analysis was problematic, which is perhaps why it took so long to get published. I am persuaded by antidepressant medication for mod-severe illness from my day to day clinical perspective. However, meds alone are not enough - a human influence is also req'd.

Yanal - That is quite a statement to make. Which type(s) of psychotherapy? Where is your evidence?

I also do not agree that psychotherapy is about finding joy. It is partly about turning pathological depression into "ordinary unhappiness" (see Freud), that is, the aim is not to give a false version of reality. Life is not all about "joy"! As for your statement that psychotherapy prolongs recovery - you need to define "recovery", as well as "psychotherapy" and give evidence to support your statement. Psychotherapy is not just for mood disorders, eg depression and anxiety, but for people with personality disorders and serious mental health problems. Psychodynamic psychotherapy can be short term or long term, depending on the circumstances. hence the importance of a psychodynamic formulation or assessment.

Here is one paper on the efficacy of psychodynamic psychotherapy by Dr Jonathan Sheddler: http://www.apsa.org/portals/1/docs/news/jonathanshedlerstudy20100202.pdf

This paper is interesting because he looks at manualized approaches as in CBT, as well as meta-analyses and includes references to studies involving antidepressants Sheddler gives a Table to illustrate the effect sizes from meta-analyses of treatment outcomes. Unfortunately, due to a number of reasons, there exists some very unfortunate misunderstandings regarding psychodynamic psychotherapy. At the centre of psychodynamic psychotherapy are interpersonal relationships, including the relationship between patient and therapist. It also involves the patient's fantasy life.

I think that some information you can get from published lectures from the MAP conference in Rome (Managing Anxiety in Practice). A lot of studies used waiting list because you need to exclude hope in psychotherapy and they used this method.

What are the effects of psychotherapy for adults with depression?

The issue

A high prevalence of psychiatric morbidity has been reported internationally. Estimates indicate that 20–30% of the general public in several countries – Australia, Canada, the Netherlands, New Zealand, the United States of America and the Caribbean and Scandinavian countries – satisfy the criteria for a psychiatric diagnosis. Where a diagnosis is provided, depression is the most common. The symptoms vary, but often include loss of energy and feelings of hopelessness and worthlessness. Depression can appear as a single episode or follow a pattern of recurring episodes. It also varies in severity. Depression has extensive consequences for the individual’s self-perceived functioning and reduced workplace productivity and increased disability pensions often consequently affect society.

The most common treatments for depression involve antidepressant medications and various forms of conversational therapy, used separately or in combination. Psychotherapy is a group of different types of conversational therapy, and its effects are debated.

Findings

The results show that patients significantly improved after psychotherapy, so that they were no longer considered clinically depressed or experienced fewer symptoms. The effects were sustained over several months. Findings suggest that psychotherapy is effective in reducing patients’ depression symptomatology compared to treatment as usual (such as guidance by a physician). Improvements from psychotherapy are similar to those observed for antidepressant medications. When comparing different variants of psychotherapy – such as psychodynamic psychotherapy, cognitive behavioural therapy, behavioural therapy and support therapy – no specific variant appears to be superior.

According to the best available evidence, psychotherapy appears to improve depression symptomatology. The evidence base was heterogeneous, however, so the results must be viewed as tentative.

Policy considerations

According to this systematic review, psychotherapy appears to be beneficial in the treatment of people with depression. Nevertheless, the evidence base is not solid. To establish the optimal treatment for adults, further high-quality research is needed to gain more knowledge that can be the basis for sound decision-making.

Type of evidence

This systematic review included and summarized the effects of psychotherapy for adults with depression from nine studies. Most were considered of high quality. The studies included patients from Finland, Ireland, Norway, Spain and the United Kingdom (Great Britain).

The views expressed in this summary are based on a publication of a HEN Network member agency and do not necessarily represent the decisions or stated policy of WHO/Europe.

Source

--------------------------------------------------------------------------------

Effekten av psykoterapi for mennesker med depressive lidelser [Effects of psychotherapy for adults with depression]

Oslo, Norwegian Knowledge Centre for the Health Services, 2009. Accessed 14 December 2009

We reviewed short-term psychodynamic models for depression. Large effects and better than control treatments and individual format is equal to other individual bona fide models like CBT. There are several publications since than that will add more N with the same overall effects when we redo this meta-analysis in a year or so. I agree that this such a mixed group of problems in the bag of "depression". Good news is talking approaches have a good reach. In 2011 we also published in "Psychiatry" a review of all the STPP studies for depression in the setting of personality disorders and the effects were again large and sustained. see www.istdp.ca for these publications.

Allan

Good morning

Sorry to join late this interesting conversation.

I will point out to references about comparing treatment (med or therapy) with control group (such as placebo or waiting list) but randomized comparison with no treatment (as you seem to ask) are not done any more today. In fact, it's not easy to have a placebo controlled randomized trial approved those days. I also think that waiting list are different from no Treatment. Anyway, so my point is, as most people who have done placebo controlled trial can attest, that placebo is far from being no treatment.

I hope this helps. You can also see my RCT but it is a placebo controlled trial not a NO TREATMENT.

This is one paper that I have done where in the discussion I touched (briefly) on those issues.

First, I fully agree with Dr. Barber that placebo is different or "better" than no treatment. Placebo-effects in depression are up to 40% in the first 3 weeks of treatment. Regarding the specific question posed here, one must first define which kind of psychotherapy we are talking about (Cognitive? Psychodinamic?), and then which kind of effect and outcome variables (remission of acute symptoms? rate of relapses after remission?).

Hi, there are several experiences published. I have an article where the levels of anxiety and depression was compared among two groups, the control goup that has not psychological treatment compared with the treatment group. the diferences among groups was statistically significant during the followup: " Resultado de un programa de intervención psicoterapéutica en pacientes con enfermedad pulmonar obstructiva crónica (EPOC). Cuadernos de Medicina psicosomática y psiquiatría de enlace" 100. 2012.

However, the NICE (2007) outcomes reveals that the cognitivebehavioral therapy is the more likely approache to treath the depresives and anxiety disorder. Only recommended drug treatment only in severe cases and always attached to psychotherapy.

Schema therapy—a form of psychotherapy combining cognitive-behavior therapy and psychodynamic concepts—was found to be superior to other approaches in treatment of personality disorders, according to a report of the study "Results of a Multicenter Randomized Controlled Trial of the Clinical Effectiveness of Schema Therapy for Personality Disorders" in AJP in Advance.

Dutch researchers conducted a multicenter randomized controlled trial from 2006 to 2011 at 12 mental health institutes. A total of 323 patients with personality disorders were randomly assigned to different forms of therapy—147 received schema therapy, 135 received “treatment as usual,” and 41 received “clarification-oriented psychotherapy.” (In treatment-as-usual, clinicians were free to use whatever approach seemed to fit the patient's needs.) The primary outcome was recovery from personality disorder three years after treatment started (assessed by blinded interviewers). Secondary outcomes were dropout rates and measures of personality disorder traits, depressive and anxiety disorders, general psychological complaints, general and social functioning, self-ideal discrepancy, and quality of life.

Researchers found that a significantly greater proportion of patients recovered in schema therapy than in treatment as usual or clarification-oriented psychotherapy. Findings did not vary with specific personality disorder diagnosis. Schema therapy patients had less depressive disorder and higher general and social functioning at follow-up.

I like the concept of Schema therapy, but would like to see more comparative studies with DBT (both standard model and RO model of Lynch - no relation by the way!) and other therapy modalities. You raise an important point on the co morbidity of Axis 2 with the depression diagnostic construct and what is recovery. I think this is the group that needs further study both from a psychotherapy and psychopharmacology cal viewpoint.

Please ignore last typo - psychopharmacological - the iPad I am using auto-corrects too much and I suspect has an Axis 2 problem!

I can only answer with the results of my personal experiences; no empirical data or research findings. Participation in NO treatment for my clinical depression resulted in the depression deepening and broadening (I cannot explain those terms as they relate to me. I can only say they fit). My quality of life degraded, my isolation increased and my suffering became almost unbearable. Medications helped but I feel they didn't address the root cause(s), they only alleviated the symptoms. To be able to deal with the underlying causes of my disorder I needed psychotherapy. Then I was able to make changes that helped me deal with those causes. Honestly, while medication improved my life, I still felt incomplete and ignorant of my true self. It was working with wonderful mental health professionals that made real progress possible.

What more can be said? Maybe neither medication or therapy is always the answer sometimes I feel we underestimate the social network and human touch.

Very nice to see what you wrote Sean. I tend to agree about interaction and human touch. Of course, in my case I was on the meds when I started psychotherapy so I am sure they contributed to my improvement. But I am not sure either alone would work as well as the two of them together.

I keep thinking about how I can improve this reply. Depression can be strictly caused by neurotransmitter imbalances and it can be caused by psychological events, and it can be a combination of the two. Dependence of medication corrects the chemical imbalance but most likely won't deal with psychological issues. Psychotherapy can work alone but rarely. The combination of therapy and medication has been proven to be most effective. The medical community is quick to prescribe but not as quick to encourage the therapy. I am sure money and patient loads are the primary reasons for this. Patients can also be at fault for thinking that because the mediaction may make them feel better, they don't need therapy or for failing to engage in therapy because they are 'too busy'. Any attempt to get control over depression without trying therapy and medication together, I think, is going to have less than optimal results.

I am a patient receiving treatment through the Veterans Health Care System. I suffer from depression and anxiety, as well as several physical injuries caused by having been in combat. Unfortunately I do not have the funds to pay for outside therapy, therefore I have go to the VA for all of my healthcare needs. The problem I have with this is the therapists are overloaded with patients and are not able to see patients on a weekly basis. At best a patient may be able to see their therapist on a bimonthly basis. This may be okay for some patients who feel that they "don't need therapy" however, for patients who really do need the therapy this is not enough. I have been prescribed several anti-depressants and anti-anxiety medications as well but I know from experience when I had received weekly counseling with a therapist the two combined worked much better.

Jason, as a fellow veteran I say "Hello Brother!" The VA's problems are legendary and often due to D.C., not the local facility. To be brutally honest, if you had a substance abuse issue comorbid with your depression and anxiety and you were enrolled in the VA SRS program, you would be able to see a counselor weekly and attend group therapy sessions on top of that. If you are on prescription painkillers and wish to get off of them, they could prescribe you Suboxone and begin psychotherapy. (Suboxone can be very effective in relieving pain also). This is the case for larger VA hospitals, If you have access to a VA hospital (not clinic), and you are on prescription painkillers you may want to explore that option. Best of luck my brother. Please feel free to contact me if you have any questions.

I would look at Bergin and Garfield's Handbook of Psychotherapy and Behavior Change, Edition: 6 (2013), Editor: MJ Lambert, Publisher: John Wiley & Sons, New York, NY, for excellent reviews & meta-analyses from the various models (DYN, CBT, HUM) & regarding specific disorders, including depression.

CAn you read the NICE 2007. Improve the acces to psychological therapies . Published By the national Institute of clinical excellence . U.K.

Have a Happy new year. 

To add to the discussion, here are three articles spread over nearly two decades of work in this area

Article Effectiveness of psychotherapy for adults with depression: A...

Article Effectiveness of Psychotherapy and Combination Treatment for...

Article Psychological Treatment of Depression in Primary Care: Recen...

The evidence appears to be that psychotherapies are as effective as pharmacotherapy in the management of depression in primary care settings and mild to moderate depression, positive effects are more durable than pharmacotherapy and, most importantly, are preferred by majority of patients. No particular psychotherapy appears to be superior to others. In chronic major depression, combined therapy outperforms drugs or psychotherapy alone.