Mindfulness-based interventions for perinatal depression, anxiety, and stress: a systematic review and meta-analysis of randomized controlled trials, with an exploratory subgroup signal by delivery mode

Background: Perinatal depression is a highly common clinical complication. While expert panels advocate for non-pharmacological treatments, the relative benefits of mindfulness-based interventions delivered face-to-face versus through emerging digital platforms have not been examined within a single integrated synthesis that distinguishes the contribution of delivery mode from that of trial era and methodology.

Methods: We conducted a systematic review and meta-analysis of parallel-group randomized controlled trials evaluating mindfulness-based interventions for perinatal women. The synthesis relied exclusively on directly verifiable, publicly accessible study-level data, obtained from primary full-text reports and, for seven of the ten depression trials, from the open controlled-data table of a prior peer-reviewed meta-analysis (Lever Taylor et al., 2016). The primary outcome was post-intervention depressive symptoms, with anxiety and stress as secondary outcomes. A pre-specified subgroup analysis examined delivery mode (digital versus face-to-face) as an exploratory, hypothesis-generating moderator rather than as a head-to-head comparison.

Results: Ten RCTs (591 perinatal women) contributed depression data. MBIs significantly reduced depressive symptoms (Hedges g = -0.31, 95% CI -0.48 to -0.14; P < .001; I² = 0%), with a 95% prediction interval excluding the null. In an exploratory, pre-specified subgroup analysis, the digital subgroup showed a larger pooled effect than the face-to-face subgroup (g = -0.40 vs. -0.13; test for subgroup differences, P = .04). Because delivery mode was completely confounded with publication year, trial size, and risk of bias, and because all three digital trials enrolled antenatal samples, this difference is reported as hypothesis-generating rather than as evidence of modality superiority. Pooled effects favored MBIs for anxiety (g = -0.29; k = 8) but were not significant for stress (g = -0.11; k = 8). Sensitivity analyses supported the robustness of the depression finding.

Conclusion: MBIs produce a small-to-moderate reduction in perinatal depressive and anxiety symptoms, with a numerically larger depression effect in the digital subgroup. Because delivery mode is completely confounded with publication year, trial size, and risk of bias in the present evidence base, this subgroup difference is hypothesis-generating, and no delivery mode can yet be considered superior. Confirmatory trials using active comparators, longer postpartum follow-up, and standardized outcome reporting are needed before any delivery mode can be considered superior.