Ava Schulz

Predicting Social Anxiety Treatment Outcome Based on Therapeutic Email Conversations

Predicting therapeutic outcome in the mental health domain is of utmost importance to enable therapists to provide the most effective treatment to a patient. Using information from the writings of a patient can potentially be a valuable source of information, especially now that more and more treatments involve computer-based exercises or electronic conversations between patient and therapist. In this paper, we study predictive modeling using writings of patients under treatment for a social anxiety disorder. We extract a wealth of information from the text written by patients including their usage of words, the topics they talk about, the sentiment of the messages, and the style of writing. In addition, we study trends over time with respect to those measures. We then apply machine learning algorithms to generate the predictive models. Based on a dataset of 69 patients, we are able to show that we can predict therapy outcome with an area under the curve of 0.83 halfway through the therapy and with a precision of 0.78 when using the full data (i.e., the entire treatment period). Due to the limited number of participants, it is hard to generalize the results, but they do show great potential in this type of information.

A mobile app for social anxiety disorder: A three-arm randomized controlled trial comparing mobile and PC-based guided self-help interventions.

Objective: Internet-based cognitive–behavioral treatments (ICBT) have shown promise for various mental disorders, including social anxiety disorder (SAD). Most of these treatments have been delivered on desktop computers. However, the use of smartphones is becoming ubiquitous and could extend the reach of ICBT into users’ everyday life. Only a few studies have empirically examined the efficacy of ICBT delivered through a smartphone app and there is no published study on mobile app delivered ICBT for SAD. This three-arm randomized-controlled trial (RCT) is the first to compare the efficacy of guided ICBT for smartphones (app) and conventional computers (PC) with a wait list control group (WL). Method: A total of 150 individuals meeting the diagnostic criteria for SAD were randomly assigned to one of the three conditions. Primary endpoints were self-report measures and diagnostic status of SAD. Results: After 12 weeks of treatment, both active conditions showed superior outcome on the composite of all SAD measures (PC vs. WL: d = 0.74; App vs. WL: d = 0.89) and promising diagnostic response rates (NNTPC = 3.33; NNTApp = 6.00) compared to the WL. No significant between-groups effects were found between the two active conditions on the composite score (Cohen’s d = 0.07). Treatment gains were maintained at 3-month follow-up. Program use was more evenly spread throughout the day in the mobile condition, indicating an integration of the program into daily routines. Conclusions: ICBT can be delivered effectively using smartphones.

An Internet-Based Compassion-Focused Intervention for Increased Self-Criticism: A Randomized Controlled Trial

Increased levels of self-criticism and a lack of self-compassion have been associated with the development and maintenance of a range of psychological disorders. In the current study, we tested the efficacy of an online version of a compassion-focused intervention, mindfulness-based compassionate living (MBCL), with guidance on request. A total of 122 self-referred participants with increased levels of self-criticism were randomly assigned to care as usual (CAU) or the intervention group (CAU + online intervention). Primary endpoints were self-reported depressive, anxiety and distress symptoms (DASS-21) and self-compassion (SCS) at 8 weeks. Secondary endpoints were self-criticism, mindfulness, satisfaction with life, fear of self-compassion, self-esteem, and existential shame. At posttreatment, the intervention group showed significant changes with medium to large effect sizes compared to the control group regarding primary outcomes (Cohens d: 0.79 [DASS] and -1.21 [SCS]) and secondary outcomes (Cohens ds: between 0.40 and 0.94 in favor of the intervention group). The effects in the intervention group were maintained at 6-months postrandomization. Adherence measures (number of completed modules, self-reported number of completed exercises per week) predicted postintervention scores for self-compassion but not for depressive, anxiety, and distress symptoms in the intervention group. The current study shows the efficacy of an online intervention with a transdiagnostic intervention target on a broad range of measures, including depressive and anxiety symptoms and self-compassion.