Abstract Background Home practice is considered a key element in increasing treatment effectiveness of Mindfulness-Based Cognitive Therapy (MBCT) for depression. However, long-term longitudinal research into the associations between home practice and depression outcomes is scarce. The current study examined the prospective associations between the extent of formal home practice and subsequent depression severity during 15 months of follow-up. Methods Data from two randomized-controlled trials on MBCT for recurrent depression were used (n = 200). Depressive symptoms were assessed at 3-month intervals: 0 (baseline), 3 (posttreatment), 6, 9, 12, and 15 months. Formal home practice frequency was calculated for each 3-month period. Autoregressive latent trajectory (ALT) modelling was applied. Results Participants practiced formal exercises on 57% (SD = 0.22, range 0–1) of the days during MBCT, equivalent to an average of 4 days per week, which showed a rapid decline after MBCT. The level of depressive symptoms did not change over the full study period. A small positive association was found between formal home practice frequency during each three-month period on subsequent depressive symptoms, but sensitivity analyses did not confirm this. More robust, a small negative association was found between levels of depressive symptoms at each measurement point and formal home practice frequency during the subsequent three-month periods. Conclusions The hypothesis that more frequent home practice would lead to reductions in depressive symptoms was not confirmed in the current study. Rather, it seems that patients with higher levels of depression may subsequently practice less frequently at home. The interplay between home practice and outcome might not be as straightforward as expected. However, these results are preliminary and should be replicated first before recommendations for clinical practice can be formulated.