BACKGROUND: Mindfulness-based cognitive therapy (MBCT) is a promising alternative treatment for generalized anxiety disorder (GAD). The objective of this study was to examine whether the efficacy of group MBCT adapted for treating GAD (MBCT-A) was noninferior to group cognitive behavioural therapy (CBT) designed to treat GAD (CBT-A), which was considered one of first-line treatments for GAD patients. We also explored the efficacy of MBCT-A in symptomatic GAD patients compared with CBT-A for a variety of outcomes of anxiety symptoms, as well as depressive symptoms, overall illness severity, quality of life and mindfulness. METHODS: This was a randomized, controlled, noninferiority trial with two arms involving symptomatic GAD patients. Adult patients with GAD (n = 138) were randomized to MBCT-A or CBT-A in addition to treatment as usual (TAU). The primary outcome was the anxiety response rate assessed at 8 weeks after treatment as measured using the Hamilton Anxiety Scale (HAMA). Secondary outcomes included anxiety remission rates, scores on the HAMA, the state-trait anxiety inventory (STAI), the Hamilton Depression Scale (HAMD), the Severity Subscale of the Clinical Global Impression Scale (CGI-S), and the 12-item Short-Form Health Survey (SF-12), as well as mindfulness, which was measured by the Five Facet Mindfulness Questionnaire (FFMQ). Assessments were performed at baseline, 8 weeks after treatment, and 3 months after treatment. Both intention-to-treat (ITT) and per-protocol (PP) analyses were performed for primary analyses. The χ2 test and separate two-way mixed ANOVAs were used for the secondary analyses. RESULTS: ITT and PP analyses showed noninferiority of MBCT-A compared with CBT-A for response rate [ITT rate difference = 7.25% (95% CI: -8.16, 22.65); PP rate difference = 5.85% (95% CI: - 7.83, 19.53)]. The anxiety remission rate, overall illness severity and mindfulness were significantly different between the two groups at 8 weeks. There were no significant differences between the two groups at the 3-month follow-up. No severe adverse events were identified. CONCLUSIONS: Our data indicate that MBCT-A was noninferior to CBT-A in reducing anxiety symptoms in GAD patients. Both interventions appeared to be effective for long-term benefits. TRIAL REGISTRATION: Registered at chictr.org.cn (registration number: ChiCTR1800019150 , registration date: 27/10/2018).