The MARIGOLD study: Feasibility and enhancement of an online intervention to improve emotion regulation in people with elevated depressive symptoms
Abstract Background This manuscript describes the first two phases of pilot testing MARIGOLD, an online self-guided positive emotion skills intervention for adults with elevated depressive symptoms, along with enhancements to overcome retention and adherence problems reported in previous research. Methods Adults with elevated depressive symptoms were recruited online and assessed at baseline, post-intervention, 1- and 3-month follow-up. Phase 1 participants (n=58) were randomized to MARIGOLD, daily emotion reporting, or waitlist. Phase 2 participants (n=79) were randomized to MARIGOLD plus one enhancement: online discussion board (ODB), virtual badges (VB), or facilitator contact (FC). Post-intervention interviews assessed acceptability. Intention-to-treat analyses examined retention, adherence, and preliminary efficacy. Results In both phases, retention and adherence did not differ between groups. MARIGOLD skills were highly acceptable, but qualitative results indicate web-based features (e.g., log-in, ODB, VB) require refinement prior to larger testing. Neither phase demonstrated between-group differences in preliminary efficacy. In Phase 1 within-group analyses, MARIGOLD and emotion reporting control demonstrated a similar pattern of findings (stable depressive symptoms, increased positive emotion, decreased negative emotion and stress), whereas the waitlist group significantly increased in depressive mood. Most Phase 2 within-group analyses demonstrated the expected pattern of results (i.e., decreases in PHQ-8 and negative emotion, increases in positive emotion). However, CES-D scores were stable in FC; perceived stress was stable in FC and ODB. Limitations This pilot study is not powered to evaluate efficacy. Conclusion Positive emotion skills, plus enhancements for web-based, self-guided delivery, warrant additional study in people with elevated depressive symptoms.