Empirically Supported Treatments: An Agenda for 2014
By David F. Tolin, Ph.D., ABPP
I’m delighted to write this first column during my term as President of Division 12, and hope to follow the exemplary leadership example of my predecessor, Past-President Dr. Mark Sobell. I would like to extend my thanks to this year’s Program Chairs, Drs. Bunmi Olatunji and Lisa Elwood, who have worked tirelessly to put together an outstanding lineup for this year’s Convention.
One of my 2014 presidential initiatives concerns empirically supported treatments (ESTs), a topic that is near and dear to the hearts of many of our members, but which has also sparked a great deal of controversy. Many scholars, even those committed to the integration of science and practice, have raised thoughtful critiques of ESTs and the EST movement. Some, for example, have suggested that the success of psychological therapy has little to do with specific therapeutic techniques, and much to do with common factors such as a strong therapeutic relationship. They argue, therefore, that our main thrust of scientific inquiry should be to investigate the principles of empirically-supported relationships, rather than empirically-supported treatments.
Other authors have argued that the overreliance on randomized controlled trials (RCTs) is fundamentally flawed. The reliance on DSM diagnostic criteria has been one common sticking point; critics have pointed out that many clinical patients either meet criteria for multiple diagnoses, or do not neatly conform to any specific diagnosis. Others have pointed out the high exclusion rates in many RCTs, which raise important questions about the degree to which the results of these studies will generalize to clinic patients who would not have been excluded. Still others have noted that the “bar” is rather low in many RCTs of psychological treatments, suggesting that it is not terribly difficult for any treatment, sufficiently studied, to prove superior to no treatment, and that the absence of double-blind placebo control makes these studies prone to inflated outcomes and researcher allegiance effects.
Many authors have objected to the association between the EST movement and manualized treatment. They point out that a practicing clinician would need to possess and master an exorbitant number of treatment manuals—a point compounded by the fact that many patients do not respond adequately to the first treatment provided. Manuals are perceived by many as being unnecessarily stifling and lacking in flexibility.
Finally, some critics have pointed to significant problems of interpretation under the current EST structure. Many treatments on the current EST list either overlap substantially with one another, or contain ingredients that have been demonstrated to be inert. The “box score” approach to determining efficacy has also been criticized, with some authors noting that there is no means of determining relative strength of treatment effects, degree of scientific support, or cost-effectiveness. The absence of long-term outcomes and the relative inattention to outcomes such as quality of life or degree of functional impairment have also been noted.
For 2014, the Division 12 Committee on Science and Practice, chaired by Dr. Evan Forman, has been tasked with updating and revising our list of ESTs (www.psychologicaltreatments.org). As we go forward, we will have to grapple with a number of thorny issues, including:
- Should relationship and techniques be considered different domains of psychological treatment? Or is it time to lump them together under the category of “stuff the psychologist should do?”
- Should we stop applying the “empirically supported” label to multi-component treatment packages, without understanding what the active and inactive ingredients are? Or should we turn our focus to empirically supported principles of change?
- Should we continue to focus on ESTs, and manuals thereof, for discrete DSM diagnoses? Or should we instead focus on treatments for syndromes of psychopathology that cut across diagnoses?
- Should we continue to call a treatment “empirically supported” if it has not been demonstrated to be superior to placebo or an alternative treatment? Or is it time to raise the bar?
- Should we continue to rely on the “box score” methodology of counting successful trials? Or should we adopt a more sophisticated rating system, such as using pooled effect size or number needed to treat?
- Is symptom reduction a satisfactory outcome for determining whether a treatment is empirically supported? Or should the treatment also be required to have demonstrable benefit on functioning or quality of life?