BACKGROUND: This study aimed to identify moderators of treatment effect (i.e. depressive symptoms and well-being) of Mindfulness-Based Cognitive Therapy (MBCT) and Positive Psychology Intervention (PPI) in patients with bipolar disorder. METHODS: Data were drawn from two multicenter randomized controlled trials investigating effectiveness of MBCT vs treatment as usual (TAU; n = 144) and PPI vs TAU (n = 97) in bipolar disorder. Outcomes were assessed at baseline, posttreatment, and 12 months after baseline. Data were analyzed using separate linear regression models, comparing the pooled MBCT or PPI outcomes to TAU, and comparing MBCT to PPI. RESULTS: The exploratory analyses not corrected for multiple comparisons showed a number of variables that were associated with stronger response to the interventions, including higher baseline anxiety, lower well-being, and lower levels of self-focused positive rumination, well-being, and self-compassion, and variables associated with a stronger response to either MBCT (higher levels of depression and anxiety and being married) or PPI (being male). After correcting for multiple testing, depressive symptoms appeared to be the most robust variable associated with better response to MBCT than PPI. LIMITATIONS: The RCTs handled slightly different enrollment criteria and outcome measures. CONCLUSIONS: The most robust finding is that patients with more severe symptomatology seem to benefit more from MBCT than PPI. CLINICAL IMPLICATIONS: This is a first step to improve personalized assignment of third-wave CBT interventions for patients with bipolar disorder. However, before definite treatment assignment criteria can be formulated and implemented in clinical practice, these findings should be replicated.