Proximal Effects of a Transdiagnostic, Ecological Momentary Intervention for Enhancing Resilience in Help-Seeking Young People (EMIcompass): Findings From a Secondary Analysis of an Exploratory Randomized Controlled Trial.

BACKGROUND: Ecological momentary interventions (EMIs) offer a promising strategy for targeting putative mechanisms of mental health problems by delivering real-time, tailored intervention components that adapt to person, moment, and context based on data collected using ecological momentary assessment (EMA). However, most research to date focuses on effects on distal outcomes, at the person level, whereas exploration of processes at the microlevel, that is, proximal effects of EMI components on putative momentary mechanisms and outcomes, remains very limited.

OBJECTIVE: This study aimed to explore the proximal effects of completing EMI components on momentary mental health mechanisms and outcomes (ie, negative affect, stress, and stress reactivity) at the subsequent EMA among youths with early mental health problems.

METHODS: EMIcompass is a novel, transdiagnostic, hybrid, compassion-focused EMI for prevention and early intervention of severe mental health conditions in help-seeking youths. Data for our secondary analysis were derived from the experimental condition of the EMIcompass exploratory randomized controlled trial, in which youths aged 14-25 years with early mental health problems were randomly allocated to the EMIcompass intervention+treatment as usual (experimental condition) or treatment as usual only (control condition). We used self-reported within-subject EMA data collected as part of the EMIcompass intervention in the experimental condition, in which participants were offered EMI tasks and voluntarily decided whether or not to complete them.

RESULTS: Data on proximal effects of EMI interactive task completion on negative affect, stress, and stress reactivity at the subsequent EMA were available in a small subsample of 37 participants. The effect sizes of within-subject proximal effects on momentary negative affect (β=-0.07, 95% CI -0.18 to 0.04; t472.52=-1.21; d-type=-0.12), stress (β=-0.12, 95% CI -0.31 to 0.06; t467.81=-1.31; d-type=-0.14), and stress reactivity (β=-0.04, 95% CI -0.15 to 0.06; t464.70=-0.78; d-type=-0.05) were small to moderate and, at the within-subject level, clinically meaningful, but 95% CIs included 0; hence, these effects were not statistically distinguishable from 0 in this small subsample. Moreover, effect sizes of cumulative training frequency on momentary negative affect, stress, and stress reactivity were very small, and 95% CIs included 0 in the small subsample of 37. Effect sizes for proximal effects for individual task types were small (d-type=-0.12 to 0.16), but 95% CIs included, again, 0 in an even smaller subsample of 27 participants.

CONCLUSIONS: This secondary analysis implemented a novel approach to advancing our understanding of proximal effects within an EMI framework. While our findings illustrate the feasibility of investigating microlevel processes within an EMI framework, they remain equivocal with regard to proximal effects, due to limited power in this small sample. Further research with larger samples and appropriate study designs, such as microrandomized or N-of-1 designs, is needed to clarify how EMI components impact momentary mental health mechanisms and outcomes.