iCBTs refer to a broad area of Cognitive Behavioral Therapy programs that are delivered through the internet. Most of these interventions are accompanied with some extent of human support. This Symposium aims to discuss how iCBTs’ design and human therapist's involvement affect the therapeutic process and clinical outcomes. Through the discussion of intervention’s design impact on relevant outcomes, we focus on "evidence based therapeutic process". Evidence based therapeutic process regards the path the intervention leads the user through in order to reach desired results. This includes interactions with human therapist, monitoring of user state, and providing just-in-time feedback. Four presenters will present studies which shed light on the connection between the therapeutic process – which is nurtured by intervention design and therapist's protocol –and relevant outcomes. Prof. Carlbring will present preliminary results from a randomized study utilizing a factorial design that examines the importance of content tailoring (Unified protocol/Diagnosis specific treatment) in the context of available therapist support (With/Without). Since the therapist may provide more adaptive and persuasive intervention, the study design enables to discuss more broadly when a human therapist matters most, and why. Dr. Baumel will present a study in which he investigated the correlation between online activities of users of 30 web-based behavioral interventions (collected from a proprietary data set of anonymized logs) and interventions’ design quality ratings obtained by trained raters. Findings suggest the importance of persuasive design and behavior change techniques incorporation in the development of digital behavioral interventions. In Prof. Huppert presentation, the theory and practice of measuring therapist contribution to iCBT via existing measures and a new measure will be discussed. Preliminary data from two trials: one on iCBT for panic and another on iCBT for depression will be presented, and results will be discussed in the context of other findings about alliance and outcome in iCBT and beyond. Dr. Aderka will present data from a randomized controlled trial of iCBT for social anxiety disorder (SAD). The study examines patterns of change in usage of the program (e.g., time spent in the program, forms completed) and patterns of change in engagement with the human therapist (number of exchanges, number of words in correspondence). We will present data on the association between the two over time as well as on the relationship between the two patterns and treatment outcome.
Amit Baumel (presenter and discussant). 1*. Per Carlbring. 2*, Jonathan Huppert. 3*, Jonathan G. Shalom. 4*
1. Department of Community Mental Health, University of Haifa, Haifa, Israel;
2. Department of Psychology, Stockholm University, Stockholm, Sweden;
3. Department of Psychology, Hebrew University, Jerusalem, Israel;
4. Department of Psychology, University of Haifa, Haifa, Israel.
1st PAPER TITLE:
iCBT for Social Anxiety Disorder: Does Engagement with human therapists predict outcome more than engagement with non-human elements?
AUTHORS: Idan Aderka, Jonathan Shalom, Jonathan Huppert, Gerhard Andersson
We will present data from a randomized controlled trial of iCBT for social anxiety disorder (SAD). Treatment in this study included components embedded in the interface such as reading text, and completing forms and questionnaires. In addition, treatment included weekly correspondence with a human therapist. The study examines patterns of change in usage of the non-human components of therapy (e.g., time spent in the program, forms completed) and patterns of change in engagement with the human therapist (number of exchanges, number of words in correspondence). We will present data on the association between the two over time as well as on the relationship between the two patterns and treatment outcome. Finally, clinical and research implications of the study for iCBT will be discussed
2nd PAPER TITLE:
Unified protocol vs. diagnostic specific treatment of social anxiety - a randomized treatment study with a factorial design
Per Carlbring, Matilda Enoh, Philip Lindner, Gerhard Andersson
Traditionally, Cognitive Behavior Therapy researchers have developed and tested diagnostic specific treatment manuals. Most of which are efficacious. However, since comorbidity between anxiety and mood disorders is common and as the choice of the best suited treatment manual for each patient can be hard when delivering the treatment via the internet, transdiagnostic interventions have been put forward.
The purpose of this study was to compare the effect of unified protocol vs. diagnostic specific treatment in a group of 147 individuals with social anxiety. The score on Liebowitz Social Anxiety Scale at pre-treatment was M=52.7.
The interventions were both delivered via the internet during 9 weeks. In addition, the relative effect of no guidance vs. guidance on demand was investigated in a factorial design with 25 % of the patients randomized into each of the four groups. The preliminary results at week 9 show large within group effects (average Cohen’s d = 1,03), but no significant between-group differences as measured with the self-report MINI-SPIN-R and OASIS/ODSIS.
At the time of the conference 6-month follow-up data will be available in addition to the already collected post-assessment data (analyzed according to the intention-to-treat principle).
3rd PAPER TITLE:
Can we disaggregate therapist and program’s contributions to the alliance and outcomes in ICBT?
Jonathan D. Huppert, Asher Strauss, Dina Zalaznik, Asala Halaj, Isaac Fradkin, Elad Zlotnik, David Ebert & Gerhard Andersson
In the current talk, I will present a number of issues that we have considered in terms of what the role of the therapist is in ICBT. Data suggest that the presence of a therapist assists effectiveness in terms of adherence, treatment completion, and symptom reduction. However, less is known about how active the therapist should be, how to measure the therapist’s contribution, and how to disaggregate the therapist’s contribution from the program’s contribution. Can we distinguish the alliance with the program and the alliance with the therapist? Interviews with patients post-treatment suggests that some see the therapist as a built in part of the treatment whereas others make a sharp distinction between the internet program and the therapist. In addition, some patients describe their relationship with their online therapist as being similar to the relationship that they had in past face-to-face treatment whereas others describe a large discrepancy in ICBT compared to previous face-to-face treatment. I will present two ways we have attempted to separately assess the relationship with the therapist and computer and some preliminary data on how these relate to outcomes. Then, I will raise a number of questions for future research.
4th PAPER TITLE:
The impact of product design on user engagement with iCBTs in real world use
Amit Baumel 1, John M. Kane 2,
1 University of Haifa, Israel
2 The Zucker Hillside Hospital, NY, USA
The quality of iCBT product design may affect user engagement. This presentation provides several analyses that shed light on this relationship.
First, using two independent expert raters we rated the quality of off-the-shelf iCBTs on different aspects of product design (e.g., Usability, Visual Design, Content, Therapeutic Persuasiveness, and Therapeutic Alliance), and examined the correlations between these quality aspects. Second, using several data sets we examined which quality aspects predict user engagement in the real world.
Findings indicate that most self-help iCBTs that have high quality of content do not have high quality of persuasive design and do not well incorporate behavior change techniques (i.e. therapeutic persuasiveness) within the digital product. Moreover, findings indicate that several quality aspects predict which iCBTs are more engaging. At the time of the conference analysis of several data sets – in which we examined which aspects of product quality predict user engagement – will be available and presented.