The intrigues of psychotherapy research: what works and for whom?

Depression. Anxiety. Anger. Grief. Trauma. Stress. Eating Disorders. Weight Loss, Addictions. People have all kinds of needs for psychotherapy. For the consumer, it seems a daunting task to choose the right therapy. Needless to say, most people would want to engage in a therapeutic model that has “scientific evidence” that it works. This, however, also proves to be not as simple as it sounds.

Consider the well-researched Cognitive Behavioural Therapy (CBT), the most empirically supported treatment (observable outcomes) in psychotherapy today. To the lay person, this may sound like a therapy of choice. After all, if so many researchers have conducted randomised controlled trials (RCTs[1]) and CBT has been found to produce better outcomes than other therapies, shouldn’t we all head for it? The answer is: not necessarily.

Efficacy and effectiveness: “same, same but different”.

What the lay person does not understand is that there are two kinds of research studies: efficacy studies and effectiveness studies. In clinical psychology, the term efficacy refers to the “internal validity[2]” of a controlled, randomised, clinical trial of an intervention, performed in rigid settings, with treatment delivered according to prescribed manualised procedures, usually in university research clinics, often recruiting students as clients. Effectiveness studies seek to establish whether efficacy protocols derived from laboratory settings in efficacy studies can be successfully transported to the wider clinical settings to which real life people present for treatment. These studies are usually conducted in the real world, with real clients and real therapists in clinical settings.

In laymen’s terms, efficacy is concerned with the controlled testing of an intervention (defined by the exact procedures detailed in the manual) while effectiveness measures the outcome experienced in the real world through real interactions between therapists and clients.

The idea of manualized psychotherapy depicted in literature often gives one an impression that it has a “technological precision”. But such is not the case, even though the idea of making psychotherapy dummy-proof, through manualised treatment is highly seductive. It is simply not possible to treat the users of manualised treatment, both the client and the therapist, as if they do not count, in arriving at conclusions about treatments.

Although CBT has been the subject of many efficacy studies, these were conducted in controlled environments that rarely, if ever, represent what happens in a real clinic. Another item in these efficacy studies being challenged is the idea of therapeutic placebos (such as using supportive counselling without any other “interventions”). Placebos in psychological treatments are actually logically and practically impossible, because they have the same psychological effects as the treatment they are replacing, thus the placebo becomes the treatment. Furthermore, CBT efficacy studies have also been challenged by a group of researchers who hold that there are “common factors” in therapy that contribute to therapeutic outcomes, not the specific interventions. The earliest such researcher is Jerome Frank (1909-2005) who developed a common factors model that included four factors: 1) the “emotionally charged, confiding relationship” with the therapist; 2) the “healing setting”; 3) the rational or myth; and 4) the believable treatment or ritual . He proposed that if hope and positive expectations are experienced, the demoralised state of the client improves. These hopes and expectations are brought about through the four factors, whether or not they are scientifically valid “treatments” or just “rituals”[3].

Another myth of efficacy studies is that therapists can practice “pure” therapy strictly according to a manual. In a relational interchange of conversations, it is nigh on impossible to “direct” therapy according to a manual, making these very artificial studies quite meaningless. Jonathan Shedler, a researcher from the University of Colorado Denver School of Medicine, points out that the mechanisms of change in CBT, for example,  are often not related to the prototype supposedly used, but elements from the Psychodynamic therapy model instead. In fact, two studies have found that if therapists actually adhered to their supposed models, CBT fared worse than psychodynamic therapy[4].

While efficacy tests are being challenged for their relevance in clinical settings, there is a move towards more effectiveness studies. In a 2004 book on psychotherapy research, 279 process-outcome studies and a further 42 empirical reviews and meta-analysis (studies analysing and synthesizing research findings) were counted, nearly a 50% increase as cited in 1994[5]. However, even these effectiveness studies are not without their own challenges. One of which is the problem of ‘common factors’. The question is raised whether therapy works because of the person of the therapist, or because of the particular model of therapy used.

Common Factors in Therapy

There are almost ninety common factors in psychotherapy[6]. These are factors other than the models and techniques employed during therapy. Of the greatest significance is the therapeutic alliance, or the relationship between therapist and client. Then there are factors that are extra-therapeutic such the client winning the lottery and depression is “cured”, or from assessment tools being administered, causing a recognition of the problem and therefore therapeutic change, such as an when an alcoholic gets asked to complete a questionnaire on alcohol consumption, or simply through the passage of time, such as healing from grief over time, regardless of the therapy. Some therapist factors are particularly significant for building the therapeutic relationship: strong working alliance, group cohesion (in the case of group therapy) and a high degree of collaboration, as well as some personality traits such as open-mindedness, creativity and patience[7].

The Medical Model vs. the Contextual Model

The medical model views the client as suffering from a maladaptive psychological disorder or mental illness and the therapist as someone who can ‘dispense’ treatment, by the use of certain techniques of ‘intervention’, as a doctor dispenses medication for physical illnesses, to help the client to alleviate symptoms and to recover. In this model, the therapist is the expert and the client the non-expert.

In the Medical Model, clients are labelled as suffering from ‘disorders’. In fact, the Diagnostic and Statistical Manual of Mental Disorders (4th Edition) is nearly a thousand pages thick and lists nearly 300 disorders, seven times as thick and listing almost three times as many disorders than when it was first published in 1952. Note also that at one time, homosexuality was a mental disorder! Currently, there is a huge debate in the psychotherapy circle concerning the impending publication of the 5th edition, which may include some troubling new categories of mental disorders. Although it is not clear if some of the debated new disorders (such as bitterness, apathy, extreme shopping or overuse of the internet) will be classified as mental illnesses, the fact that the American Psychological Association (APA) has assigned a task force to decide on these conditions sounds like it’s a case of diagnoses gone mad. As one commentator posts on the internet, “If you spend hours online, have sex more frequently than aging psychiatrists, and moan incessantly… [about] the federal government…, take heed, you may soon be classed among the 48 million Americans the APA already considers mentally ill[8].”

The Contextual Model, sometimes called the Common Factors Model, however, views techniques and interventions less significant in producing therapeutic change than the therapeutic relationship. In this model, techniques are viewed as necessary only “to construct a treatment that therapists have faith in and that provides a convincing rationale to clients” but that these techniques are only part of a bigger picture in which the six elements common to all psychotherapies exist:

-          The therapist develops a relationship with he/she helps the client to combat feelings of demoralisation;

-          The therapist maintains the client’s expectation of being helped and instils a hope of improvement through the therapeutic processes;

-          The therapist provides new learning experiences;

-          The client’s feelings are aroused because of the therapy;

-          The therapist enhances the client’s sense of self mastery;

-          The therapist provides opportunities for practice[9].

The Contextual Model, as can be seen, is a more humanistic model in which the client is the expert of his/her own life and the therapist works collaboratively with the client for therapeutic change, rather than dispensing treatment techniques to “cure” the client.

Collaborative Therapy

Collaborative therapy is not another model of therapy. Rather, it is a philosophical stance, a “way of being”, in which the client is the expert (of his/her life) and the therapist is the non-expert (of the client’s life) who starts from a not knowing position, rather than a power position. They are conversational partners joining together for mutual inquiry in a relational-responsive way of understanding so that client and therapist transform together. The therapist places trust in uncertainty whilst maintaining confidence and views therapy as everyday, ordinary life[11].

A collaborative approach of inquiry challenges traditional empiricism which typically separates the researcher from the subjects of research. Instead, the investigator joins with the subject to work towards mutual goals while retaining participants’ agency[12]. This is a more respectful way of doing therapy.

Brief Intervention vs Intensive Therapy

There is also evidence, indeed overwhelming evidence, that the length of therapy, is not necessarily a good predictor of therapy outcomes. In a study called Project MATCH, testing three forms of therapy, it was found that 12 sessions of CBT did not produce better results than just 4 sessions of Motivational Enhancement Therapy, for people with alcohol problems.[10]


The long and short of psychotherapy research is that what is “scientifically-proven” may not necessarily be more effective in the real world than therapies that have received fewer controlled trials. Also, intensive therapies are not necessarily more effective than shorter therapies. Models of therapy are not the determinants of therapy outcomes, as there are factors common to all therapies. Therapy can either be viewed from a medical, problem-oriented perspective, or from a relational, motivational perspective of change.

As a therapist who combines collaborative counselling using a multi-modal approach, with clinical hypnosis, I consider therapy both an art and a science. I hope this article has helped you understand more about psychotherapy research and that you will be able to make an informed choice of a therapist, rather than just relying on “scientific evidence” which may not reflect real life outcomes.

If you have any questions about this and other topics, or if you have some therapy stories you feel will help others, please feel free to leave a comment.







[1] RCTs are studies in which clients are randomly assigned therapies and therapists without knowing what therapy they’re receiving, with results compared with controlled group(s) being treated with an alternative therapy and/or with no therapy (such as being put on wait list). The idea is to scientifically test whether it is the particular form of therapy that produces the outcomes, which are measured by various “instruments” or tests that clients take at intervals during and/or after therapy duration, which give indications of their wellbeing, degree of problems being experienced (or not) and any number of factors that indicate therapeutic outcomes.


[2] Internal validity is achieved in a clinical setting when all other variables are controlled except the one that is being trialled so that the results can be shown to be caused by that particular variable.

[3] Wampold, B. & Weinberger, J. (2010). Jerome D Frank: psychotherapy researcher and humanitarian. In L. G. Castonguay, J. C. Muran, L. Angus, J. A. Hayes, N. Ladany, & A. Anderson. Bringing psychotherapy research to life. (pp. 29-38). Washington: APA.


[4] Shedler, J. (2010). The efficacy of psychodynamic psychotherapy. Feb-Mar 2010 Vol 65. No. Vol. 2. pp. 98-109.  American Psychologist.


[5] Orlinsky, d. E., Ronnstad, M. H., Willutzki, U. (2004). Fifty years of psychotherapy process-outcome research: continuity and change. In M. J. Lambert. Bergin and Garfield’s handbook of psychotherapy and behaviour change. (pp.307-389). New York: Wiley.


[6] Gastonguay, L. G., Reid, J. J., Halperin, G. S., Goldfried, M. R. (2003) Psychotherapy Integration. In G. Stricker & T. A. Widiger (Vol Eds), I. B, Weiner (Ed.). Handbook of psychology: volume 8: clinical psychology.  (pp. 327-345). New Jersey: Wiley.


[7] Gastonguay, L. G. & Beutler, L. E. (2006). Common and unique principles of therapeutic change: what do we know and what do we need to know? In L. G. Gastonguay & L. E. Beutler. (Eds.). Principles of therapeutic change that work. (pp.353-370) Oxford: OUP. p.358.


[9] Frank & Frank (1991, as cited in Wampold, B. E. (2001). The great psychotherapy debate: models, methods and findings. Mahwah, NJ: Lawrence Erlbaum. p.25.


[10] Andreasson, S. & Ojehagen. (2003). Psychosocial treatment for alcohol dependence. In M. Berglund, s. Thelander & E. Jonsson. (Eds.). Treating alcohol and drug abuse. (pp.43-188). Weinheim: Wiley-VCH.

[11] Anderson, Harlene. (2007). The heart and spirit of collaborative therapy: the philosophical stance – a “way of being” in relationship and conversation. In Anderson, H; Gehart, D. (Eds.) Collaborative therapy: relationships and conversations that make a difference. New York: Routledge. Pp. 43-59.


[12] Gergen, K. (2009). An invitation to social construction. London: Sage.


This entry was posted in Alcoholism, Anger Management, Anxiety Attacks, Bulimia/Anorexia, Chronic Pain Management, Depression, Drug Addiction, Drug/Substance Abuse, Fear of flying, Gambling, Grief, OCD Obsessive Compulsive Disorder, Panic Attacks, Parenting Issues, Performance Anxiety, Phobia, PTSD, Public Speaking, Quit Smoking, Relationship Problems, Self Harm, Self-Esteem, Sexual Dysfunction, Sports Perfomance, Uncategorized, Weight Loss. Bookmark the permalink.

Leave a Reply

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


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>