Depressive disorders (DDs) can be highly disabling and are ranked third in terms of disease burden as defined by the World Health Organization, and first among all psychiatric disorders in terms of disability adjusted life years. In addition, DDs seem to be rising globally, and a 20% annual increase in its incidence has been predicted. Improvements in treatment methods and prevention measures, and the availability of community psychiatric services are, therefore, as important as ever before.
So begins, without much controversy, this week’s Reading – which happens to be one of the most controversial papers of the year.
This meta-analysis has been mentioned in newspapers and blogs. No wonder – in 22 pages, it raises questions about the effectiveness of a major psychiatric treatment: cognitive behavioural therapy.
The Reading: “The Effects of Cognitive Behavioral Therapy as an Anti-Depressive Treatment is Falling: A Meta-Analysis” by Tom J. Johnsen and Oddgeir Friborg, recently published in the prestigious Psychological Bulletin.
The full paper can be found here:
A quick summary: analyzing data from 70 studies over nearly four decades, Johnsen and Fribourg find CBT to have become less effective at reducing depressive symptoms.
It’s not surprising that such a conclusion would cause a bit of a stir. Noting the drop in efficacy, the Pacific Standard magazine writes: “psychotherapy’s go-to method isn’t what it used to be.” The Guardian weighs in too, noting the study, and beginning the article: “everybody loves cognitive behavioural therapy.” The paper was even covered by Incidental Economist – a blog dedicated to… economics.
The Johnsen and Friborg paper opens by noting the advancement of the practice in other areas. “Old procedures and methods have been replaced with safer and more effective solutions.” Coronary angioplasty is a good example. Once a complicated procedure, it’s now “the technique is quick and presently requires minimal rehabilitation… it represents a huge improvement.” They ask: does cognitive behavioural therapy follow the trend?
Here’s what they did:
· The researchers began by searching several databases with broad search terms (“treatment effectiveness evaluation,” “depression and treatment,” etc.), yielding over 29,000 titles. After reviewing the titles and then abstracts, the list was down to 489 papers, which were read.
· Exclusion criteria were applied, including that the implemented therapy wasn’t pure CBT; the primary diagnosis wasn’t depressive disorder; participants weren’t adults; treatment wasn’t implemented with face-to-face therapy.
· Using this approach, the final study pool was 70. These studies stretched from the 1970s (5 studies) to 2010s (12 studies)
· Data from the studies were coded for several things, including year of implementation, duration, type of therapist, and special characteristics of the population (like diabetes, alcohol abuse, MS).
· The researchers investigated more than a dozen moderator variables, including gender, age, and proportion of patients using medications.
· Effect sizes were calculated based on Beck Depression Inventory (BDI) and Hamilton Rating Scale for Depression (HRSD), and considered prepost study design and a controlled trial design.
· An effort was made to ensure interrater reliability (based on a random sample of 20 studies).
Here’s what they found:
· 70 studies were included; 52 were randomized controlled trials; 5 were controlled trials without randomization; 2 were uncontrolled; 11 were clinical field studies.
· The total number of patients was 2,436.
· The biggest finding involved time and effect size. See below.
· That said, this paper looked at many aspects of CBT. Age wasn’t significantly related to variation in treatment effect, gender was (women responded better). The number of therapy sessions wasn’t related to better treatment effect.
The main finding was that the treatment effect of CBT showed a declining trend across time and across both measures of depression (the BDI and the HRSD). Contemporary clinical treatment trials therefore, seem to be less effective than the therapies conducted decades ago. (!)
The plot portrays the change in Beck Depression Inventory effect sizes across time.
The plot portrays the change in Hamilton Rating Scale for Depression effect sizes across time.
The paper goes on to explore the reason for the trend. The authors forward a couple of theories. First, with increased popularity, there has been an influx of therapists – often with limited experience and poor training.
The ostensibly simple treatment objective of CBT (i.e., changing maladaptive cognitions to alleviate emotional disorders), has made it particularly attractive and has created a misconception of being easy to learn. However, proper training, considerable practice, and competent supervision are very important to provide CBT in an efficacious manner.
Second, they suggest that the placebo effect plays a role.
In the initial phase of the cognitive era, CBT was frequently portrayed as the gold standard for the treatment of many disorders. In recent times, however, an increasing number of studies have not found this method to be superior to other techniques. Coupled with the increasing availability of such information to the public, including the Internet, it is not inconceivable that patients’ hope and faith in the efficacy of CBT has decreased somewhat, in recent decades.
Reaction to the Paper
Dr. Scott Patten, editor of the Canadian Journal of Psychiatry, writes a lively blog where he considers another explanation:
In past decades, few people outside of the mental health world were even aware of CBT and often even the first few, psychoeducationally oriented, CBT sessions were real eye openers for them. The basic concepts of CBT sometimes led to a big change in peoples’ ways of looking at the world. In this day and age, however, everyone has heard of CBT, many have read books about it or encountered similar ideas about the relationship between thoughts and emotions in popular media. Skillfully delivered CBT has a lot to offer people who only have a cursory knowledge of it…[N]owadays, some proportion of the gains may already have been made before people enter therapy.
Babes-Bolyai Univerity’s Prof. Ioana Cristea, a cognitive behavioural therapist herself, hits the Johnsen-Friborg paper hard at a blog dedicated to mental health, “The Mental Elf:”
But for anyone familiar with the methodological aspects of meta-analyses, this particular one does not engender a loss of confidence in CBT. It does, however, elicit considerable loss of faith in meta-analyses and in their reliability and usefulness.
(Cristea’s comments have been toned down since the original posting, and you can see Johnsen’s response to Cristea’s critique in the comments section.)
1. This is an interesting paper.
2. This is also a thoughtful paper, covering a vast literature and doing so well.
3. Even with a decline over the years, CBT has a good effect size.
4. Like both the study authors and Dr. Patten, I wonder if CBT has become a victim of its own success: highly popular but watered down and poorly delivered.
5. Some – like Prof. Cristea – seem to think that this is a criticism of CBT itself. I see this paper as something of a wake-up call.
Dr. Patten’s blog “Clinical Depression in Canada” can be found here (and I recommend it):
“The Mental Elf” can be found here (and I recommend this blog too):
Thanks to Dr. Patten for helping me understand this paper and its criticism better.
Reading of the Week. Every week I pick a reading — often an article or a paper — from the world of Psychiatry.