Background and aims There is evidence that cognitive behavioral therapy (CBT) is effective for treating problem gambling (PG). Some research points to the possible benefits of involving concerned significant others (CSOs) in treatment. This study compared the efficacy of behavioral couples therapy (BCT) and CBT for both the gambler and the CSO. Design Two parallel-group randomized controlled study comparing two different internet-based treatments for PG. Follow-up measures were conducted at treatment finish, and at 3-, 6- and 12-month post-treatment. Setting Stockholm, Sweden. Participants A total of 136 problem gamblers and 136 CSOs were included in the study: 68 gamblers and 68 CSOs for each treatment condition. The gamblers were on average 35.6 years old and 18.4% were female. CSOs were on average 45.3 years old and 75.7% were women. Interventions A treatment based on BCT was compared with a CBT intervention. Both treatments were internet-based, with 10 therapist-guided self-help modules accompanied by weekly telephone and e-mail support from a therapist. CSOs were given treatment in the BCT condition, but not in the CBT condition. Measurements The primary outcome measures were time-line follow-back for gambling (TLFB-G) and the NORC Diagnostic Screen for Gambling Problems (NODS) for problem gamblers, corresponding to DSM-IV criteria for pathological gambling. Secondary outcomes measures were the Patient Health Questionnaire-9 (PHQ-9), the Generalized Anxiety Disorder seven-item scale (GAD-7), the Relation Assessment Scale Generic (RAS-G), the Alcohol Use Disorders Identification Test (AUDIT), the Inventory of Consequences of Gambling for the Gambler and CSO (ICS) and adherence to treatment for both the problem gambler and the CSO. Findings The outcomes of both gambler groups improved, and differences between the groups were not statistically significant: TLFB-G: multiplicative effect = 1.13, 95% confidence interval (CI) = 0.30;4.31); NODS: multiplicative effect = 0.80, 95%, 95% CI = 0.24;2.36. BCT gamblers began treatment to a higher proportion than CBT gamblers: P = 0.002. Conclusions Differences in the efficacy of internet-based behavioral couples therapy and cognitive behavioral therapy for treatment of problem gambling were not significant, but more gamblers commenced treatment in the behavioral couples therapy group.