Inner Actions Header Logo

PLEASE NOTE: I retired in September and am no longer seeing clients.

 wise woman smiling


This section was written in conjunction with Allan Brownsdon in 2015. It remains current and useful.

What works in therapy

The trouble with models of therapy is that there are so many of them: CBT, Psychodynamic, Gestalt, Rogerian, Jungian, Psychoanalysis, Transactional Analysis, Acceptance Commitment Therapy; to name but a few. There are, depending on your definitions, potentially hundreds of different models. How is the average person able to assess what is appropriate for them?

We are accepting the challenge to attempt to help you do just that. Rest assured this article is not about pushing any particular model. Rather we are presenting some interesting information that has caused some to suggest that All have won, so all deserve prizes [Alice in Wonderland, Lewis Carrol] and also explains why and what that means for you.

If you're in a hurry you can jump to “What we think” at the end and avoid the explanation.

Measuring what works

There have been many studies of the effectiveness of various techniques and approaches to therapy. There is obvious benefit for a profession that genuinely wants to assist people, to determine what works, with whom and under what conditions.

A complicating factor in assessing this can be the degree of complexity of a model. A model that says “how we think about something influences how we feel and behave” is clearly easier to test than a model that says “thoughts influence feelings at the same time as feelings are generating thoughts about that feeling, and how we act in relation to those thoughts and feelings can modify both”.

Given the relative simplicity of the first compared to the second, it's not difficult to imagine which types of models might have more research applied to them and therefore a greater number of “effectiveness studies”.

Another problem is that it can be difficult to know what is actually being measured. In any research related to human beings where two people sit in a room talking, there are variables that are difficult to control. Is it the technique or model that the therapist is applying that makes a difference? What about the personality of the therapist? How accurately does one therapist follow the model compared to another? Do additional personal factors about the client influence the outcomes?

Some researchers have asked the same questions and what they discovered is very interesting - very interesting indeed.

The Research Findings

It is beyond this brief article to fully explore all of the research, so we will point to just a few of the relevant studies and their key findings.

In the early 2000's, Peter Fonagy [What Works for Whom? A Critical Review of Treatments for Children and Adolescents, Peter Fonagy et al, 1st Ed, 2004] and a group of researchers in the United Kingdom examined the literature to try and determine what therapies worked for what diagnosis and for whom. Their findings were somewhat controversial in that they clearly outlined a series of problems in researching for evidence based outcomes. They pointed out the distinction between the “efficacy of a therapy” in a research setting versus its “clinical effectiveness” in routine practice. While they did create a list of therapies based on their research they also suggested caution in overly relying on the specifics of the list. In the 2nd Edition of their book, they have been even clearer stating:

“The ability of the individual practitioner to deliver a specific therapeutic intervention tailored to the needs of the individual client is as important, if not more important, as matching the type of therapy to the presenting mental health problems of the client.”
— Peter Fonagy et al, 2nd Ed, 2015, What Works for Whom? A Critical Review of Treatments for Children and Adolescents


Michael Lambert [Prevention of Treatment Failure: The Use of Measuring, Monitoring, and Feedback in Clinical Practice, Michael J Lambert, 2010] is a researcher from the United States whose interest is in finding ways to improve therapeutic outcomes. He has a particular interest in working out how to detect when therapy is failing and identifying the measures of success. However in undertaking this research he has also necessarily looked at the question of what works in therapy. He comes to some interesting conclusions. Lambert presented some of these at a conference [Combined ACA & PACFA Conference, Canberra, 2009] in Canberra in 2009. He stated that “patient outcome” can be determined by four factors:

Factor Effect Meaning
Client Life 40% Factors in the patients life that are not related to the therapy
Common Factors 35% A list of common factors applying to all therapies
Therapist Effects 20% Specific factors relating to the nature of the therapist and how they work
Techniques 5% The specific techniques or models the therapist is using

Lambert provides a long list of the ”common factors” he considers apply. These include: empathy, a positive relationship with the therapist, a structured environment, acceptance, insight, modeling, taking risks, facing fears, “corrective emotional experience”, and appropriate advice and feedback.

He further claims that when comparisons between two different theory-based approaches are researched, they seldom show differences in outcome.

Lambert's research places significant importance on the ability of the therapist to create an environment in which the “patient” can deal with the issues they confront and on the ability of the therapist to be with the “patient”. That is, the ability of the therapist to effectively attune to the patients individual needs and respond to those needs.

Jonathon Shedler is another researcher who is concerned with what actually works in therapy. Shedler has a particular interest in demonstrating the effectiveness of Psychodynamic approaches to therapy and comparing the outcomes to those of other therapy approaches including CBT and medication alone. Shedler uses a complex meta-analysis of research findings and statistical analysis to look at the “effect size” of approaches. His research suggests some interesting issues in relation how to really measure the impact of individual therapies or treatments on outcome. What is interesting for us here is perhaps best summed up in the following quote where Shedler is comparing so called evidence based research methods with what happens in practice:

“There are also differences in the way therapists practice, even therapists ostensibly providing the same treatment. What takes place in the individual consulting room reflects the qualities and style of the individual therapist, the individual patient, and the unique patterns of interaction that develop between them.”
— J Shedler, The Efficacy of Psychodynamic Psychotherapy, Jonathan Shedler, American Psychologist in Press, 2010


What we think

Without wishing to sound flippant, we think that the research is, in effect, showing us what we should have known all along: “Different strokes for different folks”. Fundamentally therapy is an activity which takes place between two people, you the client and your therapist. As such the therapy must necessarily be designed around your needs and your personality, and respond to you as an individual person.

This is not to say that the therapist is in any way neutral or infinitely adaptable. For some people, it will be as important to find the right therapist as it will be to find the therapist using a model of therapy with which they are comfortable.

As therapists we need a model of therapy that works for us. This is important, both for us as therapists and for you as the client. Having an overall guiding model of what works and how is essential for the therapist to be able to map the issues you present and to understand how to respond to you. Without a guiding model, the therapist will be jumping from one approach to the other and you as the client will experience the therapy as chaotic and uncontained; it will not feel safe. A sense of safety and trust in the relationship with your therapist is essential to allow you to explore, open up and take the risk of changing aspects of your life that are not working.

In summary we think:

  • The therapist needs a model of therapy with which he or she is comfortable to provide a guiding map for the journey;
  • The models being used by the therapist need some reasonable level of research and evidence to avoid open ended eclecticism or “flaky” approaches. This is best tested by ensuring that your therapist has suitable qualifications;
  • You as the client, need to be able to trust that the therapist understands you and your issues and has some experience and skill assisting in those areas;
  • The therapist must be able to attune to and adapt to your needs as a client, or if not to be able to refer you to a therapist who may better match your needs;
  • The focus of therapy needs to be on you the client. Your therapist needs to be responding to you first and their model of therapy second. Otherwise, you are not going to feel heard and understood; and
  • The healing aspect of any successful therapy is the result of two people working together towards the common goal of making your life more meaningful and fulfilling. In other words, it's all about the professional relationship between you and your therapist.

We hope this brief overview of some very complex research gives you some sense of what might be important for you in selecting and working with a therapist.