Isolating therapeutic procedures to investigate mechanisms of change in cognitive behavioral therapy for depression
By Sanne Bruijniks, a PhD Candidate working at Vrije Universiteit Amsterdam.
Although cognitive behavioral therapy (CBT) is the most extensively studied and evidence-based treatment for depression, 47% of the individuals suffering from depression show no response to CBT, and improvement is highly necessary. One way to improve treatments is to find out how treatments work. However, finding out how treatment works exactly is difficult. Most studies have attempted to increase insight on CBT’s mechanisms of change by measuring what processes (also called mediators) change during CBT, and by investigating whether change in the mediators are responsible for subsequent change in the symptoms. In CBT for depression, changing negative thought patterns, or acquiring the ability to identify and change negative thought patterns, has been hypothesized as the major mediator involved in the reduction of depression. Unfortunately, so far it has been very difficult, if not impossible, to establish a causal relationship between change in the mediator and subsequent change in the outcome.
Another way to investigate causal relationships is by manipulating the mediator and investigating its direct effects on the outcome. Although this method is common in studies that investigate how treatments for anxiety work, they have been used less to investigate treatments for depression. Our study demonstrates an experimental approach to investigate mechanisms of change in CBT for depression. We suggest that the most straightforward way to investigate the direct role of processes in CBT for depression might be to isolate a cognitive procedure that is expected to lead to immediate changes in the mediator and the subsequent outcome. Our experiment leads to clear recommendations for future experimental research on CBT for depression (see the table below) and provides a blueprint for other researchers interested in the underlying mechanisms of change in CBT for depression.
We suggested the following factors that will be important for future experimental depression research:
Because in a happy sample it will be very difficult to investigate change in dysfunctional thinking or sad mood, we suggest the use of a (sub) clinically depressed sample or the induction of sad mood and/or dysfunctional thinking.
Instead of investigating the overall concept of mechanisms of change, future studies should disentangle procedures (the application of the technique by the therapist) from processes (the process that is hypothesized to be affected by the procedure).
Instead of using standardized questionnaires to measure processes as dysfunctional thinking and cognitive therapy skills, process measures should be personalized to enable the measurement of clinically relevant change on these processes for the specific individual.
Not all levels of dysfunctional thinking are considered equally accessible and different levels of dysfunctional thinking should be distinguished in future research.
Studies should focus on what dose of the treatment procedure is necessary for what change in the process.
Comparing the effects of a non-cognitive change procedure versus a cognitive procedure on cognitive change (one of CBT’s major hypothesized processes of change) will enable studies to investigate whether the effects of a cognitive change procedure on cognitive change are specific to cognitive change procedures.
Sanne J. E. Bruijniks, Marit Sijbrandij, Caroline Schlinkert, Marcus J. H. Huibers
Sanne Bruijniks is a PhD Candidate working at the Vrije Universiteit Amsterdam.