Effects of Acceptance and Commitment Therapy on Self‐Compassion, Self‐Criticism, and Emotional Well‐Being in Adults With Mental Health Concerns: A Systematic Review

ABSTRACT Background and Aims Self‐compassion is a supportive self‐relating style linked to emotional well‐being. Acceptance and Commitment Therapy (ACT) targets psychological flexibility and may influence self‐compassion, yet evidence is inconsistently quantified. This systematic review investigated changes in self‐compassion following ACT or ACT‐based interventions in adults with mental health concerns and summarized findings for self‐criticism and emotional well‐being. Methods We followed PRISMA 2020 and preregistered the protocol on the Open Science Framework (10.17605/OSF.IO/8u7e9). PubMed, Embase, Web of Science, and the Cochrane Library were searched to August 5, 2024. Two reviewers (A.C., D.L.) independently screened studies and extracted data. C.F. adjudicated discrepancies, with substantial agreement at title and abstract screening ( κ = 0.72) and full‐text eligibility ( κ = 0.78). RoB 2 and ROBINS‐I were used for randomized and non‐randomized studies. We synthesized results narratively following SWiM and reported effect estimates with uncertainty when available. Meta‐analysis was not performed. Results Ten studies were included (five randomized controlled trials and five non‐randomized, experimental, or qualitative studies). Self‐compassion was most often measured with the self‐compassion scale. In trials, within‐group improvements were common, but between‐group effects were mixed, ranging from negligible contrasts to large gains in selected samples (one trial reported a large within‐group change, d ≈ 1.5, versus a smaller change in the waitlist). A pre–post chronic pain evaluation reported a small improvement in self‐compassion ( d ≈ 0.21). Self‐criticism was reported in one trial with minimal post‐treatment separation, and emotional well‐being outcomes were variably defined and mixed. Most randomized controlled trials had some concerns of bias (one high risk), and most non‐randomized studies were at serious risk or provided insufficient information. Conclusion ACT may improve self‐compassion, but comparative effects are inconsistent and certainty is limited. Standardized measurement and complete reporting are needed to clarify durability and clinical relevance.