Revealing Subgroup-Specific Mechanisms of Change via Moderated Mediation: A Meditation Intervention Example

Objective: Effective psychosocial interventions exist for numerous mental health conditions. However, despite decades of research, limited progress has been made in clarifying the mechanisms that account for their beneficial effects. We know that many treatments work, but we know relatively little about why they work. Mechanisms of change may be obscured due to prior research collapsing across heterogeneous subgroups of patients with differing underlying mechanisms of response. Studies identifying baseline individual characteristics that predict differential response (i.e., moderation) may inform research on why (i.e., mediation) a particular subgroup has better outcomes to an intervention via tests of moderated mediation. Method: In a recent randomized controlled trial comparing a 4-week meditation app with a control condition in school system employees (N = 662), we previously developed a "Personalized Advantage Index" (PAI) using baseline characteristics, which identified a subgroup of individuals who derived relatively greater benefit from meditation training. Here, we tested whether the effect of mindfulness acquisition in mediating group differences in outcome was moderated by PAI scores. Results: A significant index of moderated mediation (IMM = 1.22, 95% CI [0.30, 2.33]) revealed that the effect of mindfulness acquisition in mediating group differences in outcome was only significant among those individuals with PAI scores predicting relatively greater benefit from the meditation app. Conclusions: Subgroups of individuals may differ meaningfully in the mechanisms that mediate their response to an intervention. Considering subgroup-specific mediators may accelerate progress on clarifying mechanisms of change underlying psychosocial interventions and may help inform which specific interventions are most beneficial for whom. What is the public health significance of this article? Individuals receiving psychosocial interventions likely differ substantially in which intervention elements they respond to and in their causal pathways of change. This study demonstrates the utility of considering subgroup-specific mediators, which may accelerate progress on clarifying mechanisms of change underlying psychosocial interventions and may help inform which specific interventions are most beneficial for whom.