Internet-delivered therapist-assisted cognitive behavioral therapy for gambling disorder: a randomized controlled trial.

Abstract

Cognitive behavioral therapy (CBT) is the most promising treatment for gambling disorder (GD) but only 21% of those with problematic gambling seek treatment. CBT over the Internet might be one way to reach a larger population. The aim of this study was to assess the effectiveness of Internet-delivered CBT with therapist guidance compared to an active control treatment. Using a single-blinded design, 71 treatment-seeking gamblers (18-75 years) diagnosed with GD were randomized to 8 weeks of Internet-delivered CBT guided by telephone support, or 8 weeks of Internet-delivered motivational enhancement paired with motivational interviewing via telephone (IMI). The primary outcome was gambling symptoms measured at a first face-to-face assessment, baseline (treatment start), every 2 weeks, post-treatment, and 6-month follow-up. Gambling expenditures, time spent gambling, depression, anxiety, cognitive distortions, and quality of life were assessed as secondary outcomes. Analysis was performed on the full analysis sample (n = 60), with intention-to-treat sensitivity analyses (n = 69). In the CBT group, 80% stayed in treatment until the final week, compared to 67% in IMI. Post-treatment and at 6-month follow-up, no differences were found between CBT and IMI for any outcome measure. An exploratory analysis of the total sample (n = 60) showed a significant effect of time during treatment on gambling symptoms (d, [95% CI] 0.52, [-1.15, 2.02]) and all secondary outcomes except the gambling diary (depression (0.89, [-1.07-2.65]); anxiety (0.69, [-1.20-2.38])); cognitive distortions (0.84, [-0.73-2.29]); quality of life (0.60, [-0.61-1.70])). Post-treatment, there were no clinical gambling symptoms in either group. Some deterioration was seen between post-treatment and 6-month follow-up on gambling symptoms (0.42, [-1.74-2.43]), depression (0.59, [-0.82-1.86]), and anxiety (0.30, [-0.99-1.48]). Additionally, it was observed that the largest reduction in gambling symptoms was between the first assessment and baseline. Both treatments offered in this study were effective at reducing gambling symptoms. It is also possible that the process of change started before treatment, which gives promise to low-intensity interventions for GD. Additional research is needed as this approach could be both cost-effective and has the potential to reach more patients in need of treatment than is currently possible. https://www.isrctn.com/, identifier ISRCTN38692394.