How “Supportive” Is Internet-Based Supportive Psychotherapy?

To the Editor: In their article, published in the November 2007 issue of the Journal, Brett T. Litz, Ph.D., et al. presented thought-provoking preliminary data on Internet-assisted, cognitive behavioral self-management of posttraumatic stress disorder (PTSD) symptoms (1) . In a report that emphasized technology and downplayed human contact, however, it might have been helpful to clarify certain details pertaining to the control intervention. A randomized study is only as credible as its control intervention, which raises conundrums. What exactly is Internet supportive counseling—the control condition—in this trial? Furthermore, how much therapist contact did subjects actually receive?

One imagines that supportive counseling would require affective mirroring and interpersonal warmth. Although the study design included a 2-hour initial meeting between the subject and therapist and allowed “periodic and ad lib study therapist contact via e-mail and telephone” ( 1, p. 1677), it was not clear how much direct human contact and loving kindness the supportive counseling patients received. Although therapists were “instructed to be empathic and validating” ( 1, p. 1681), e-mail in particular can obscure affect. The fact that patients read about stress and its management and wrote about “daily nontrauma-related concerns and hassles” ( 1, p. 1681) does not actually explain how the treatment was supportive. The authors described data on the frequency of Internet sessions but not on the background e-mail and phone contacts. It may have been helpful if they had commented further on how frequent, how long, and how supportive the interpersonal contacts were in each cell.

Training good supportive therapists requires a great deal of work (2) . Although the article emphasized the study web site, it omitted any description of the training and prior experience of the therapists involved. Did these same therapists back up both the cognitive and supportive web sites? If so, could this have introduced allegiance bias (3) into the study? Were attempts made to monitor therapist adherence to the respective treatments?

Finally, the authors described their cognitive web site at length, but relatively little about its supportive counterpart was mentioned. What features of the latter make it “supportive”?