REVIEW OF: Sagoe D et al, J Behav Addict 2021;10(3):546–565

STUDY TYPE: Systematic review and meta-analysis of randomized sham-controlled trials

Problematic gambling is common, affecting up to 6.5% of all adults. But only 10% of them get treated, at least in part due to limited provider availability and stigma. The easy access and anonymity offered by internet-based gambling programs address both barriers. But how well do they work?

Researchers conducted a systematic review and meta-analysis, compiling 13 randomized trials that enrolled over 2,000 participants. Eight of the studies had control groups, the specifics of which varied between trials. Overall, 22 treatments were tested; 13 of them were based on cognitive behavioral therapy (CBT), while the others were based on a range of other psychotherapeutic interventions (motivational interviewing, couples therapy, and brief advice, among others). The number of sessions ranged from one to 28 (mean 9.9 sessions), and four of the protocols included therapist support. The outcomes were gambling frequency, amount of money lost, and score on a gambling severity scale (different studies used different scales), which the authors called general gambling symptoms.

Results showed that internet-based treatments were associated with improved outcomes, particularly for general gambling symptoms. At the conclusion of treatment, the effect size (Hedge’s g) for improvement of general gambling symptoms was 0.729, indicating a medium to large effect. The effect sizes for gambling frequency and amount of money lost were more modest (0.291 and 0.190, respectively). Ten of the studies included assessments after a follow-up period (ranging from one to 36 months, mean 8.3 months), and these showed that benefits persisted over time; effect sizes for general gambling symptoms, gambling frequency, and amount of money lost at follow-up were g=1.197, g=0.361, and g=0.202, respectively. The effect sizes being higher at follow-up than at the conclusion of treatment is a statistical consequence of the fact that not all studies included follow-up periods.

Patients with more severe symptoms tended to show a greater degree of improvement, and interventions that included therapist support were associated with greater benefit. Unsurprisingly, studies that included a control group typically had smaller effect sizes. The authors compared these results with previous findings and determined that internet-based treatments did not work quite as well as in-person therapy but did work better than self-guided interventions (Goslar M et al, J Behav Addict 2017;6(2):142–162).

Carlat Take

This meta-analysis suggests that internet-based treatments for problematic gambling may be helpful. Consider seeking them out, especially if your patient prefers virtual treatment or if face-to-face counseling is unavailable.