Factors influencing implementation of a digital compassion training program: Findings from a Randomized Waitlist-Controlled Trial (Preprint)

Authors:
Preprint
Publisher website
Google Scholar
Abstract
BACKGROUND Digital mental health interventions are increasingly used to address stress, burnout, and well-being among working adults because they offer scalable and accessible approaches to psychological support. Although digitally delivered compassion training programs show promise, implementation outcomes such as feasibility, acceptability, and participant-perceived fit remain underexamined. In particular, little is known about the extent to which digital compassion interventions help participants achieve the outcomes they most hope to gain through participation. OBJECTIVE This study evaluated the feasibility, acceptability, and expectation-outcome alignment of Compassion U, a blended care adaptation of Cognitively Based Compassion Training (CBCT®) program tailored for healthcare, education, and business professionals. Exploratory analyses examined potential signals of change in compassion-related and well-being outcomes. METHODS A pilot randomized waitlist-controlled trial was conducted with 137 participants from healthcare (n=46), business (n=54), and education (n=37) professions. Participants were randomized to either Compassion U or a waitlist control condition. The intervention combined asynchronous mobile-accessible CBCT modules with weekly live instructor-facilitated sessions over 10 weeks. Primary implementation outcomes included feasibility, acceptability assessed using the Theoretical Framework of Acceptability (TFA), and expectation-outcome alignment. Expectation fulfillment was quantified using a novel Expectation Fulfillment Index (EFI), representing the proportion of participants’ desired baseline outcomes subsequently perceived as having benefited from the intervention. Exploratory outcome analyses examined compassion, well-being, loneliness, stress, emotion regulation, and workplace-related outcomes. RESULTS Feasibility was supported by successful recruitment across all three professional sectors and retention rates comparable to those reported in digital mental health intervention studies. Participants demonstrated high baseline readiness to engage with the intervention, with no significant readiness differences observed across sectors. Acceptability was high at postintervention (overall acceptability score: mean 4.33/5) and remained favorable at follow-up. Participants reported broadly positive perceived benefits across domains. The mean EFI was 0.58 (SD 0.41); however, expectation fulfillment was highly heterogeneous, with 30.6% of participants reporting high benefits to desired outcomes and 27.8% reporting no benefits. Personal growth, self-compassion, and mental health were among the most frequently endorsed desired outcomes, whereas professional development demonstrated the largest expectation-benefit gap. Expectation fulfillment was most strongly associated with participant motivation and commitment to continued practice, suggesting that participant-defined goal attainment may represent an important but often overlooked dimension of success in digital mental health interventions. Consistent with the pilot nature of the study, exploratory effectiveness analyses identified 2 notable outcome signals favoring the intervention: increased beliefs in the malleability of compassion (Hedges g=0.70, 95% CI 0.26-1.13) and reduced loneliness (Hedges g=−0.55, 95% CI −0.97 to −0.11). Most other outcomes demonstrated small or negligible effects, and observed effects generally attenuated by follow-up. CONCLUSIONS Compassion U demonstrated strong feasibility and acceptability across healthcare, education, and business professionals and was perceived as beneficial in several domains prioritized by participants. Findings suggest that expectation-outcome alignment and expectation fulfillment provide useful participant-centered indicators of intervention fit and participant-defined goal attainment that complement traditional implementation and effectiveness outcomes. The association between expectation fulfillment and participant motivation further suggests that intervention success may be shaped not only by program use, but also by the extent to which participants remain invested in personally meaningful goals. Future fully powered trials should further evaluate expectation fulfillment as a participant-centered lens for understanding how individuals experience and evaluate digital mental health interventions.

This publication has 28 references indexed in Scilit: