Sweden opens up CBT monopoly, gives nod to psychodynamic therapies

Psychodynamic therapists of the world, rejoice! After years of complaining that CBT sucks up all the public funding, it seems that psychodynamic therapists may be about to get a break – in Sweden at least.

For the last four years or so, Sweden’s government has put substantial funds (around £200 million according to one source) into CBT provision and CBT training. Now, it looks like the government’s National Board of Health and Welfare, Socialstyrelsen, has accepted that psychodynamic therapies are as effective as CBT at treating depression – which experts say is likely to lead to the introduction of government support for psychodynamic therapies.

This is significant for UK mental health policy, as our government has also put substantial funds into CBT, and is facing a similar dispute from psychodynamic therapists who claim that practice-based research shows that all therapies work equally well in the field – therefore they should all get funding, not just CBT.

Rolf Holmqvist

The shift in Swedish policy is in part due to the work of Rolf Holmqvist, professor of clinical psychology at Linköping University, whose research suggests that just about every form of talking therapy is equally effective when used in the field. He’s written an article in the new issue of Socionomen, the journal for social workers in Sweden, in which he presents his latest research. Rolf agreed to be interviewed to explain his findings and their implications. I should say at the beginning that I’m a big supporter of CBT and the UK government’s funding for it, but don’t want to be blindly defending my own preferences.

JE: Sweden’s government is a big supporter of CBT, isn’t it?

RH: Yes, it’s a pretty similar situation to the UK. In Sweden, the government has put a lot of money into training therapists to do CBT.

JE: I read it has spent 2 billion kronor (£200 million) on it in the last four years or so.

RH: I’m not sure of the exact figures, but it’s a lot of money. Several hundred therapists and social workers have been trained in CBT. Unfortunately, at some places therapists do not really do CBT, they just call it that to get public money. The government sponsors CBT treatments for depression and anxiety, up to around £1,000 per person.

JE: So therapists must ‘convert’ to CBT?

RH: They’re not obliged to. But if they want government funding, they must either provide CBT or Interpersonal Psychotherapy (IPT)

JE: So tell me about the new issue of Socionomen, and how Swedish mental health policy is changing.

RH: In our study we used the CORE-OM system for rating therapy outcomes [as opposed to the Beck Depression Index, designed by Aaron Beck, who’s also the founder of Cognitive Behavioural Therapy]. We started by examining outcomes in primary care centres. In Sweden, there is perhaps one such centre for every 10,000 people. And at every centre, there is one or two people providing psychological treatment. We asked therapists to ask their patients to rate their state on the CORE-OM outcome measure, so we could follow the progress of their treatment, which was typically rather short – on the average six sessions. We compared a number of things, particularly how different treatment orientations succeeded – particularly CBT and psychodynamic,. We found exactly the same results, for both depression and anxiety. They all got good results, with about half of patients recovering. Even supportive therapy, which is the Cinderella of therapies because it seems too simple, got quite good results.

Effect Size for All Treatments

  CORE-OM   Function   Symptoms   N
Supportive .68 .56 .68 108
Dynamic 1.04 .82 1.0 84
CBT 1.05 .85 1.09 99
Cognitive 1.72 1.43 1.67 41
Crisis intervention 1.18 .85 1.34 49
Behavioral .91 .73 .81 21
Relational 1.25 .95 1.57 12
Client-centered .48 .35 .27 10
Systemic  .64 .48 .66 17
Counselling 1.0 .53 .85 10
Directive  1.16 .97 1.14 173
Reflective 1.07 .85 1.06 99


JE: Can you briefly describe the difference between CBT and psychodynamic therapies?

RH: CBT is directive. It’s educational, and it helps people to train themselves to get better. Psychodynamic therapy is reflective. It helps people reflect on their feelings.

'Everybody has won, and all must have prizes.'

JE: So does the research show the famous Dodo effect – all talking therapies seem to have the same impact.

RH: Yes, on many psychiatric states. And we also found that, in practice, therapists don’t always follow only one therapeutic approach. In practice, therapists and patients together tend to negotiate and find a treatment that works for the patient. By the way, there was a parallel study in the UK recently that found exactly the same results: Stiles at al (2008) [for a response from David M. Clark, the chief champion of the government’s support for CBT, to Stiles at al, see this paper].

JE: So your study found that all these different therapies showed some beneficial results? Because I saw a write-up of the Socionomen report which suggested it says the government’s CBT programme has had no impact whatsoever, or even a negative impact.

RH: That was another report by professors in health economy from the Karolinska Institute. They were looking at whether CBT was helping people to get off benefits and go back to work. In that respect, they couldn’t see any effect of CBT treatment. But I wouldn’t say there was no effect – we were able to show a good effect.

JE: So is it true the Swedish government is changing its approach and broadening the range of therapies that it might support?

RH: It’s true that the National Board of Health and Welfare, Socialstyrelsen, said a few months ago that it feels as if psychodynamic therapies are as good as CBT for depression. It still insists CBT is the best for anxiety, although our practice-based findings suggest psychodynamic therapies are also just as good for anxiety.

JE: Is that likely to mean a broadening of financial support for training in and provision of other therapies?

RH: Yes, it’s likely.

JE: What are the other implications of your research?

RH: I think the main implication is to recognise that there are two types of valid research paradigms: firstly, randomised controlled trials (RCTs), where you compare clearly defined treatments. Secondly, practice-based studies, where you don’t compare narrowly-defined treatments for selected patients, but instead look at how therapies are provided within real settings. The problem with RCTs is they are not as clean as they claim to be – a lot of noise gets in to them, through researchers’ allegiance and therapists’ expectations and so on. When governments in Sweden and the UK looked at which therapies to support, they decided there must be accountability. So they looked at the field of therapies, and they found lots of RCT studies for CBT, and few for psychodynamic therapies. But practice-based studies better show the successful outcomes for psychodynamic therapies. Practice-based studies are becoming more accepted now. For example, in the new edition of the Handbook of Psychotherapy and Behavioural Change, there will be a new chapter on practice-based studies.

JE: What I don’t understand about the Dodo effect is that these different therapies often have very different and conflicting conceptual underpinnings. Different theories about what emotions are and how to change them, for example. So they can’t all be right, can they? I mean, either emotions are connected to beliefs, and you can change them by changing your beliefs, or they’re not.

RH: Well, what you often find is what therapists say is the mechanism of change is usually not. So in cognitive therapy, for example, Aaron Beck thought that cognitive restructuring of beliefs is the way to change people’s mood. In fact, some research suggests that the depression changes first, then the thinking. [It also seems that, with anxiety disorders, the behavioural component of CBT is as important or more important in recovery than cognitive restructuring – see Clark et al (2008)]

We’ve lived now for some decades with this big debate between psychodynamic therapy and CBT. And in 15 years, there will be other kinds of division between them. Even now, people use lots of combinations of the two.But, in general, it seems that talking therapies, when they work, enhance the possibility to stand and accept strong emotions. They help people explore affects and try to stand them.

I can think of critiques to Rolf’s findings – if, by his own admission, therapists in the field are using a jumble of all kinds of different therapies (while often calling it CBT), then how can he compare the outcomes for CBT to psychodynamic therapies? The Dodo effect also has worrying implications for government support for mental health policy. If all therapies work the same (and I’m not sure they do, for specific conditions like social anxiety for example), then should government finance everything from maracas-shaking shamans to aromatherapists?  There is also, clearly, a difference between passing episodes of stress, which might naturally clear up on their own no matter what therapy a person receives, and more chronic conditions – a point made in Clark’s rebuttal to Stiles et al, which is linked to above. I will discuss these issues, and the problem of the Dodo effect, further in my newsletter tomorrow. In the meantime, feel free to leave comments below.


  • Where to begin? First, the table of effect sizes above seems to conflict with his conclusions in the interview, so I’m guessing there’s some information missing. Perhaps the differences between effect sizes were found not to be statistically significant. However, you can see that the mean effect sizes above range from 1.72 for cognitive therapy to 0.48 for client-centred counselling – psychodynamic therapy is 1.04, in the middle. On the face of it, that data appears to show that cognitive therapy is much more effective than other treatments and particularly client-centred approaches, which fare very badly on this measure.

  • Okay, I’ve now figured out what’s going on with those data… The “cognitive”, “CBT” and “cognitive-behavioural” categories are being lumped into the “directive” super-category at the bottom, at which point the individual differences in effect size disappear and the mean turns out the same to be only slightly higher (not a statistically significant difference?) than the other super-category for “reflective” therapies. That obviously raises the question as to why the “cognitive” sub-category performed so much better than the “CBT” sub-category, particularly as “pure cognitive therapy” is rare and most cognitive therapy includes behavioural components (the “B” of CBT). I suspect there might be some problems of definition here in terms of the therapy modalities used for classification. Also, I note from the presentation that these therapies were delivered in a brief format, mostly apparently 1-6 sessions. Especially as there’s no control condition to provide a measure of spontaneous remission, regression to the mean, or non-specific treatment factors, it’s possible that all we’re observing here is that when treatment is delivered in such an abbreviated format there’s not much difference between broad categories of therapy approaches because improvement is mainly due to other factors. Perhaps only whatever is being labelled as “cognitive” rather than “CBT” works particularly well in such a short time frame.

  • As a therapist, I find that the opinion of most therapists I know (and most of them are not eclectic), is that all mainstream therapies can be equally effective, in the right hands. It’s a shame that logical positivism rather than the reflective wisdom of experienced practitioners repeatedly holds a disproportionate hold on public policy.

    To understand how different therapies can be at least equally effective, and how this fact should not lead to us to funding shamans, it’s important to take a Wittgensteinian turn away from a mechanistically causal concept of change. Therapies work best, I believe, not by operating on causes and effects, even if they assert that they do. Rather Rolf’s closing statement captures the matter.

    We need to put on our philosopher hats rather than empiricist scientific ones. What helps a person to cope with pain, adversity, change, and so forth? What helps them to get a handle on things and regain their balance? My suspicion: The therapies help people to become philosophical, as it were. They help people to stand back, put things into a bigger context, give them a narrative (remember Epictetus: men find pain acceptable they find it rational), practice assenting and willing to tolerate discomfort; which means getting some air, some calm, getting a handle on things, making acceptable sense of things, not feeling so overwhelmed and so…anxious by virtue of a new context or interpretation. Not just seeing the dark picture, but coming to increasingly see some light too. Viktor Frankl, quoting Nietzsche, wrote that “He who can find a *why* can find a *how*”. Therapy enacts standing back, finding meaning, taking a breath.

    In the same way that the value of philosophy, *at its deepest*, cannot be empirically validated, so too therapy cannot be validated according to the measures which many CBT researchers use and boast about. But that doesn’t place philosophy in equivalence with shamanism, and neither does it land therapy in that place.

    To add one final point, philosophy is not just a technique. Epictetus’ students said they felt he was talking to them, that his way of being present itself motivated them. Regarding modern therapy, what matters less is the therapeutic mode/techniques than the quality of the practitioner in any mode (their skills and case conceptualisation, but also their relational qualities, what the client can perceive in them), and how well that practitioner can engage the client in taking a step back (and forward).

  • Tim Sharp says:

    Its interesting that the discussion still gets drawn into therapy as ‘doing something to somebody’ or even worse as being a simile for medication that can be applied to discrete disorders. In the UK CBT is talked about within the NHS as if it is a set of techniques that once learnt can be applied – the behemoth lurching in this in this direction pays little heed to those of us who are more guarded in their enthusiasm for monotherapy.

  • Miek P says:

    The Dodo bird effect tells something about BONAFIDE therapies. That means therapies with scientific research. The work of Shamans is not based on scientific research, so The Dodo bird verdict can’t be used for it.

    Why can different therapies have the same results? Look for the famous study of Lambert & Barley (2001): They found that 40% of the variance in therapeutic outcome was due to outside factors, 15% to expectancy effects, 15% to specific therapy techniques, and 30% of variance was predicted by the therapeutic relationship/common factors. Lambert and Barley concluded that, “Improvement in psychotherapy may best be accomplished by learning to improve ones ability to relate to clients and tailoring that relationship to individual clients.” So the technique is not very important, but the therapeutic relationship is the most important factor that the therapist can learn.

    Another remark is that there are few client-centered therapist (CCT) nowadays. Most of them practice newer forms of CCT, like experiential therapy. A recent therapy based on CCT is emotion focused therapy (EFT, not the tapping!) that is doing a lot of research. With good results.

  • Phil Harrison says:

    I would like to thank Matthew Bishop for his response, it had an immediate calming effect on my blood pressure….

    I read with interest Rolf Holmqvist’s gentle assertions, whilst being very aware of the difficulty his interviewer was having in hearing them.

    I wonder why anyone would imagine that their should be an immutable unidirectional thread running through our brains such that emotions can be changed by beliefs and that this must be provable as either true or not true. Anyone who does to agree with this is surely just too preposterous!

    Welcome to my world, and to the wonderful world of the Preposteri!

  • As a philosophy student before becoming a psychologist I am delighted to find this web site. I am an American and a member of the American Academy of Psychotherapists. Note “Psychotherapists” rather than “Psychotherapy”. This difference is in recognition that the person of the psychotherapist is probably the most significant single variable in the effectiveness, in distinction to the product of RCT research “efficacy.” This is consistent with the research that has shown that regardless of what theoretical orientation a therapist comes out of, the longer they have been in practice the more alike what they actually do becomes. Most of the theories recognize something that works in relieving human suffering and the longer we practice the more able we are in recognizing and using what works. I call myself a psychodynamic therapist because I regard it as the most broadly inclusive of all the labels. I was trained in a program with a psychoanalytic theory of personality, a Rogerian therapeutic technique and an existential value system and therefore Wittgensteinian epistemology. What therapist coming out of a philosophical tradition doesn’t pay attention to and work with the meanings of words and he conceptional framework in which they are used.

  • […] for Life”, enthusiast of the politics of well-being, ancient Greek philosophy and CBT, interviews Rolf Holmqvist. Prof Holmqvist is a clinical psychologist at Linköping University whose research suggests that […]

  • Sean Manning says:

    Can you give me the precise reference to Holmqvist’s article in Socionomen?

  • Farhad Dalal says:

    Could you post information about this major conference on your website please?
    November 1-2 2014 in Dartington Devon-
    Challenging the Cognitive Behavioural Therapies: The Overselling of CBT’s Evidence Base


  • Right here is the perfect site for anyone who hopes to understand this topic.
    You realize so much its almost hard to argue with
    you (not that I really will need to…HaHa). You certainly put a
    new spin on a topic that’s been discussed for ages. Excellent stuff, just great!

Leave a Reply

Your email address will not be published. Required fields are marked *