Data-Driven Implications for Translating Evidence-Based Psychotherapies into Technology-Delivered Interventions

Abstract

Mobile mental health interventions have the potential to reduce barriers and increase engagement in psychotherapy. However, most current tools fail to meet evidence-based principles. In this paper, we describe data-driven design implications for translating evidence-based interventions into mobile apps. To develop these design implications, we analyzed data from a month-long field study of an app designed to support dialectical behavioral therapy, a psychotherapy that aims to teach concrete coping skills to help people better manage their mental health. We investigated whether particular skills are more or less effective in reducing distress or emotional intensity. We also characterized how an individual's disorders, characteristics, and preferences may correlate with skill effectiveness, as well as how skill-level improvements correlate with study-wide changes in depressive symptoms. We then developed a model to predict skill effectiveness. Based on our findings, we present design implications that emphasize the importance of considering different environmental, emotional, and personal contexts. Finally, we discuss promising future opportunities for mobile apps to better support evidence-based psychotherapies, including using machine learning algorithms to develop personalized and context-aware skill recommendations.

Keywords: Data Science; Dialectical Behavioral Therapy; Mental Health; Mobile Health Interventions.

Figures

Figure 1:

Pocket Skills helps people learn and practice skills. (a) Pocket Skills includes four modules, each focusing on different types of skills. (b) Each skill walks people through DBT content via a conversational interface.

Figure 2:

Distribution of the total number of skills practiced over the course of the Pocket Skills feasibility study, showing skills with pre- and post-ratings and skills with only post-ratings.

Figure 3:

Average skill improvement on the 5-point scales of all skill uses (purple), Emotion Regulation skills (blue), and Distress Tolerance skills (red), with standard error bars. People improved more using Emotion Regulation skills than Distress Tolerance skills.

Figure 4:

Average skill improvement (top) and post-ratings (bottom) across age subgroups (left), education subgroups (middle), and disorder types (right), with standard error bars. Higher improvement indicates more improvement (i.e., is better), while lower post-ratings indicate more positive ratings (i.e., is better). Some subgroups varied more, on average, than others. Section 4.3.2 discusses significant differences between subgroups.

Figure 5:

Accuracy of model prediction across individual skills, with accuracy of a single model across all skills displayed on the right. The varying accuracy of different models for different skills, together with an ablation study highlighting the importance of skill ID, reveals a need for skill-specific models that can account for different context relevant to each skill.