Bi-Directional and Time-Lagged Associations between Engagement and Mental Health Symptoms in a Group Mindfulness-Based Mental Health Intervention

Abstract

There is a high need for accessible avenues for improving mental health among emerging adults, particularly on college campuses. Mindfulness-based intervention (MBI) is a promising avenue for reducing mental health symptoms, but initial discomforts associated with MBI may cause symptoms to fluctuate before decreasing, which presents a barrier to engagement with mindfulness on a daily basis. Consistent mindfulness practice is key for forming habits related to MBI, and engagement with mindfulness at home, including between intervention sessions, is an important predictor of mental health outcomes. Research suggests that mental health symptoms may serve as barriers to their own treatment. Thus, it is important to understand how mental health symptom levels impact adherence to treatment protocols. To improve understanding of symptom-specific barriers to treatment and engagement with mindfulness, the present study collected daily diary surveys about engagement with mindfulness and mental health symptoms from a sample of 62 adults recruited to participate in a six-week mindfulness intervention. We explored mental health symptoms as a predictor of engagement with MBI at the mean level and whether within-person variability in symptoms predicted same-day or time-lagged changes in engagement via mixed-effects associations. Using heterogeneous location scale models, we further explored whether erraticism in either mental health symptoms or engagement with mindfulness predicted the other and if outcomes of the mindfulness intervention were homogeneous among subjects. Results showed that bi-directional and time-lagged associations exist between symptoms and engagement, indicating that there is a nuanced temporal and reciprocal relationship between engagement with mindfulness and mental health symptoms. Daily within-person elevations in engagement with mindfulness were associated with concurrent improvements in mental health but prospective increases in mental health symptoms. We also found that higher engagement (over personal averages) was not consistently associated with improvements in mental health across the sample but was instead associated with greater heterogeneity in outcomes. We also found that increases in mental health symptoms (over personal averages), as well as higher average levels of mental health symptoms, were both associated with lower levels of engagement in the mindfulness treatment protocol.