After more than thirty years, researchers still do not agree whether Cognitive Therapy (CT) is an efficacious treatment for severe depression (Cuijpers, van Straten, Andersson & Oppen, 2008). The discussion and the research is still relevant as the Treatment of Depression Collaborative Research Program (TDCRP) in 1989 (cited in: Cuijpers et al. 2008; DeRubeis, Gelfand, Tang & Simons, 1999; DeRubeis, Hollon, Amsterdam, Shelton, Young, Salomon et al., 2005; Dimidjian, Hollon, Dobson, Schmaling, Kohlenberg, Addis et al., 2006; Jacobson, Dobson, Truax, Addis, Koerner, Gollan et al., 1996; Jacobson & Hollon, 1996; Parker, Roy & Eyers, 2003; Wampold, Minami, Baskin & Callen Tierney, 2002) a large treatment results/outcomes study, is still used to determine treatment strategies in the United States, thus research in this area is still important and has far-reaching implications and applications (DeRubeis et al., 1999; Parker et al., 2003).
In the TDCRP, severe depression was diagnosed by using the Structured Clinical Interviews for Diagnosis (SCID) of the Diagnostic and Statistical Manual of Mental Disorders – third edition revised (DSM -111), together with a score of 14 or higher on the first seventeen questions of the Hamilton Depression Rating Scale (HDRS) (Hollon et al., 2005). It appears that these measurement scales are extensively and routinely used, though researchers now use a later version of the DSM. In some research, the Beck Depression Inventory (BDI) was used in tandem with the HDRS, as a measurement for severity of depression with a score of at least 20 on the BDI being the indicator of severe/major depression (Jacobson et al., 1996).
Though it does appear that a number of researchers do at least agree on a definition and measurement of depression, one of the difficulties in assessing the effect of therapy on depression is that researchers cannot agree on other terms and research approaches. The basis of this essay is to demonstrate the difficulty in truly assessing the efficaciousness of cognitive therapy because of the inconsistency in definitions, efficacy-based research, measurement and reporting.
In examining whether cognitive therapy is an efficacious treatment for severe depression it is important to not only agree on what constitutes depression but also determine what cognitive therapy is. This is not as easy as it might seem. In their 2008 meta-analysis, Cuijpers, van Straten, Andersson and Oppen include definitions of seven major types of psychological therapies used to treat depression. They put CT under the heading of Cognitive Behaviour Therapy (CBT). Cuijpers et al. (2008) believe there are two main types of CBT, the cognitive therapy (CT) part, that mainly involves teaching the client skills to recognise, understand and change their dysfunctional thought patterns (cognitive restructuring); and the CT plus behavioural activation (BA) part, where clients learn ways to relax, change their thoughts and behaviour, plus cope and respond to stressors more effectively.
To facilitate their meta-analysis, Cuijpers et al. (2008) also further divide the CT part of CBT into two subheadings: those treatments based on the work of Beck et al. and those treatments that do not refer to Beck et al.’s manual (Cuijpers et al., 2008). Though Cuijpers et al. (2008) do define their understanding of what constitutes CT, every research team has an individual understanding of what CT is and each research team uses varying levels of efficacy within their research.
What is Meant by Efficacy?
The term ‘efficacy’ is used to describe effectiveness within a highly controlled research setting (Bower, 2003). Think of it as a more stringent form of ‘effective’, which is tested under controlled conditions. We understand the word ‘effective’ to mean a treatment works, it is useful and someone has tested it or observed it to work. The term ‘clinical utility’ means that a treatment has been utilised in a clinical setting and observed to be effective. When we hear ‘efficacy’ we know that strict research conditions have been observed and that as many confounds as possible have been excluded from a randomised control trial (RCT), giving as true a finding as possible (Bower, 2003).
One of the key points of efficacy is that a RCT is involved. A randomised control trial is a type of research where participants are randomly assigned to each of the treatment conditions and strict controls are instigated to reduce confounds. A confound is something not accounted for, that might affect the result. That is, any possible cause of effectiveness has to be eliminated, except for that which can only be attributed to the treatment being investigated (Bower, 2003).
In therapy versus drug-treatment trials, researchers strictly control the recruitment of participants, assess the current health of each participant in a standardised, formal and reportable way; only include participants of a particular type (eg. participants who score above a particular score on a test); randomly assign each participant to one research condition; train and monitor therapists to make sure they are all doing the same thing and assess each participant in the same way at the same intervals (Bower, 2003). Researchers use RCTs to achieve a high internal validity for their drug and therapy studies because it demonstrates their research is unbiased (Bower, 2003).
High internal validity means that not only do the research findings come from the specific variables tested but the research can be replicated by someone else and produce comparable results. It is important to note however, that just because research is successfully replicated often, that does not make the effect/findings true (Jacobson & Hollon, 1996). If the research is carried out in the same way each time and there are confounding variables in play, then the findings are not true no matter how many times the research is done.
Efficacy research is not easy to implement, especially when it involves the use of therapists. Therapists vary in their skills and in personality (Jacobson et al. 1996). It would seem that no matter how well a script is followed it would be very difficult for a therapist to rehearse every possible response they would make to every possible client response. Thus the results of the trial can only be truly representative of real life if those exact conditions are met all the time. As it is highly unlikely that the research conditions are going to occur naturally (in real life), the research results cannot be generalised or applied outside the research setting (Bower, 2003; Jacobson & Hollon, 1996).
This difficulty with results being applicable in the wider community – outside the research setting – is called external validity. In efficacy-based research, high internal validity and all the controls that are involved means that the findings cannot be truly representative of what happens in real life, without all the controls in place. So, the findings are very limited and it is a contentious issue as to whether efficacy research is worthwhile at all (Bower, 2003; Jacobson & Hollon, 1996).
Efficaciousness of Cognitive Therapy, Particularly in Severe Depression
The findings of the TDCRP have been revisited many times but their results have not been replicated – not even close. Jacobson and Hollon (1996) believe that although the TDCRP demonstrates that drug therapy is more efficacious in the treatment of the severely depressed, in the acute phase of their illness, the results should be viewed with caution and not used as a basis for treatment strategies. They came to this conclusion because the authors of the TDCRP acknowledge that therapy results varied significantly across their data-collection-sites and, within that study, no definitive conclusion re: effectiveness could be drawn (cited in: Jacobson & Hollon, 1996). The authors of the TDCRP also acknowledge that in some sites, CBT was efficacious in treating depression (cited in: Jacobson & Hollon, 1996) but they only tested CBT. As stated earlier, CBT has different components and is administered in varying degrees of competence and consistency. So, once again, back to the problem of definition and consistency.
In 1996, Jacobson et al. used three forms of CT/CBT to conduct research to see if there was any difference in the efficacy of each form of therapy. They found that although some of their research participants were more severely depressed than those in the TDCRP (they had higher pre-test BDI scores) their findings indicated that CT and BA and Automatic Thoughts (AT) are equally efficacious. Strangely, whilst Jacobson et al. (1996) think BA and CBT are part of the same therapy base, Dimidjian et al. (2006) believe they are different therapies and find BA to be significantly better than CT in the treatment of severely depressed participants.
If one research team thinks there are different types of CT, how many variations are there and what do other research teams think CT is? In 1996, Jacobson et al.’s Component Analysis of CBT, they report that, “... the treatment [CT] is so multifaceted that a number of alternative accounts for its efficacy are possible”. If there is no consistency of definition then how can there be any conclusive evidence for the efficacy of the treatment?
In committing to efficacy research, more recent studies have involved strict controls on therapist training and treatment consistency. In their 2006 study, Dimidjian et al. assessed tapes of therapy consultations and used the Cognitive Therapy Scale (CTS) to rate competence and compliance of therapists. They seem to have made every attempt to bring consistency of therapy to their study and they did find that BA was more efficacious than CT, with severely depressed participants.
However, the crossover between BA and CT appears to have defeated them. In the Treatment section of their report, Dimidjian et al. (2006) describe their two variables, BA and CT, then acknowledge that they, “... share certain elements ...” and (confusingly) go on to describe how they are intertwined. They shoot themselves in the foot, when describing their CT treatment condition – they assert that in their study the therapists were free to “...use the full range of BA strategies outlined in the CT texts ...”. This does not inspire confidence that there are two separate treatment conditions being investigated. Whilst they inspire confidence in their thoroughness, the interconnectedness of BA and CT make their results somewhat spurious.
Relapse rates after CT are shown to be lower than after ending drug treatment (Hollon et al., 2005) except in the TDCRP, where relapse rates after CT were shown to be higher than in other studies (Cited in: Jacobson & Hollon, 1996) but this is considered inconclusive because of the CT-response inconsistency across sites in that study (Jacobson & Hollon, 1996).
My understanding of what CT/CBT is, leads me to believe that the behavioural activation (BA) component of CT is an efficacious support to drug treatment in the acute phase of severe depression. Once drug treatment has ceased (as quickly as possible and only after the threat of suicide has passed) then BA/CT is invaluable as a support and development tool for the on-going treatment and prevention of relapse of depression. However, though better than drug treatment (after the acute phase of severe depression) I do not believe that BA is any more efficacious than some other forms of psychotherapy but that has a lot to do with the classification of BA and CT and the consistency of its implementation.
With a multitude of ethical, consistency, methodology and efficacy issues involved in studying CT and depression, is it the case that the inconclusiveness in support of CT is not about the therapy itself but about the difficulty in using efficacy-based research for this type of study? Parker et al. (2003) believes that in these highly controlled studies, where “... a psychotherapy that is variably defined and administered” is compared with any other treatment, or even to versions of itself, then comparisons of findings across studies is difficult. So, is it worthwhile? If it is not worthwhile and is just an academic exercise, should the research results really be used to determine treatments (Bower, 2003; Jacobson & Hollon, 1996; Parker et al., 2003)?
As of 2003, Parker et al. (2003) found little reported research on the efficacy of CT when used in combination with drug treatment. I would be interested to see the results of research using drug treatment for the first six weeks after diagnosis, alongside CT, then CT alone for the next twelve weeks. However, there are a number of issues concerned with efficacy in research and these are not adequately addressed in the reviewed literature. Bower (2003) alluded to the problems of using a control group in therapy research but the ethical issues this presents were not adequately addressed. It is beyond the scope of this essay to examine the ethics of withholding treatment for research purposes, other than to say that if lives are in danger (there was a suicide reported) only observational research can be permitted.
Bower, P. (2003). Efficacy in evidence-based practice. Clinical Psychology and Psychotherapy, 10, 328 – 336.
Cuijpers, P., van Straten, A., Andersson, G., & van Oppen, P. (2008). Psychotherapy for depression in adults: A meta-analysis of comparative outcome studies. Journal of Consulting and Clinical Psychology, 76, 909-922.
DeRubeis, R.J., Gelfand, L.A., Tang, T.Z., & Simons, A.D. (1999). Medications versus cognitive behaviour therapy for severely depressed outpatients: Mega-analysis of four randomised comparisons. American Journal of Psychiatry, 156, 1007 – 1013.
DeRubeis, R.J., Hollon, S.D., Amsterdam, J.D., Shelton, R.C., Young, P.R., Salomon, R. M. et al. (2005). Cognitive therapy versus medication in the treatment of moderate to severe depression. Archives of General Psychiatry, 62, 409 – 416.
Dimidjian, S., Hollon, S., Dobson, K.S., Schmaling, K.B., Kohlenberg, R. J., Addis, M. E. et al. (2006). Randomised trial of behavioural activation, cognitive therapy and antidepressant medication in the acute treatment of adults with major depression. Journal of Consulting & Clinical Psychology, 74(4), 658 – 670.
Hollon, S.D., DeRubeis, R.J., Shelton, R.C., Amsterdam, J.D., Salomon, R. M., O’Reardon, J. P. et al. (2005). Prevention of relapse following cognitive therapy versus medication in moderate to severe depression. Archives of General Psychiatry, 62, 417 – 422.
Jacobson, N.S., Dobson, K.S., Trux., P.A., Addis, M. E., Koerner, K., Gollan, J.K. et al. (1996). A component analysis of cognitive-behavioural treatment for depression. Journal of Consulting & Clinical Psychology, 64(2), 295 – 304.
Jacobson, N.S., & Hollon, S.D. (1996). Cognitive-behaviour therapy versus pharmacotherapy: Now that the jury’s returned its verdict, it’s time to present the rest of the evidence. Journal of Consulting & Clinical Psychology, 64(1), 74 – 80.
Parker, G., Roy, K., & Eyers, K. (2003). Cognitive behaviour therapy for depression? Choose horses for courses. American Journal of Psychiatry, 160(5), 825 – 834.
Wampold, B.E., Minami, T., Baskin, T.W. Callen Tierney, S.(2002). A meta-(re)analysis of the effects of cognitive therapy versus other therapies for depression. Journal of Affective Disorders, 68, 159 – 165.