Increased tendencies towards ruminative responses to negative mood and anxious worry are important vulnerability factors for relapse to depression. In this study, we investigated the trajectories of change in rumination and anxious worry over the course of an eight-week programme of mindfulness-based cognitive therapy (MBCT) for relapse prevention in patients with a history of recurrent depression. One hundred and four participants from the MBCT-arm of a randomized-controlled trial provided weekly ratings. Mixed linear models indicated that changes in rumination and worry over the course of the programme followed a general linear trend, with considerable variation around this trend as indicated by significant increases in model fit following inclusion of random slopes. Exploration of individual trajectories showed that, despite considerable fluctuation, there is little evidence to suggest that sudden gains are a common occurrence. The findings are in line with the general notion that, in MBCT, reductions in vulnerability are driven mainly through regular and consistent practice, and that sudden cognitive insights alone are unlikely to lead into lasting effects.