Levin, M. E. (2013). Evaluating a prototype acceptance and commitment training web-based prevention program for depression and anxiety in college students (Unpublished doctoral dissertation). University of Nevada, Reno, NV.
Depression and anxiety disorders are prevalent among college students. Innovative, cost effective approaches are needed to help prevent the incidence and worsening of these problems. Initial feasibility research with a prototype web-based program based on Acceptance and Commitment Therapy called ACT on College Life (ACT-CL), which currently includes two modules targeting values and acceptance, indicated it may be a promising approach for preventing depression and anxiety among college students. The current feasibility study compared the ACT-CL prototype to an active control website providing basic education on depression and anxiety disorders in a randomized controlled trial with 228 undergraduate college students using a universal prevention design. Assessments were conducted at baseline, post intervention and 1- and 3-month follow up. Results indicated significantly lower user engagement and satisfaction ratings with the ACT-CL prototype relative to the control website and initial pilot trial. Analyses with both the intent to treat sample and program completers only generally found no differences between condition over time on outcome and process measures, although in a few instances the control condition outperformed ACT-CL. Subgroup analyses did not identify any consistent results indicating ACT-CL was more or less effective or usable in any subgroup of students. Psychological flexibility processes were predictive of outcomes at baseline and of changes in outcomes over time. A pattern of results suggested that the more users engaged in the ACT-CL program the more they improved on psychological flexibility processes. Possible explanations for low program engagement and satisfaction ratings for the ACT-CL prototype and lack of treatment effects are explored including website design features, sample characteristics, and other methodological factors. Overall, these results suggest that the existing components of a prototype version of ACT-CL do not outperform an active control website and that some website features may reduce user engagement and satisfaction.