Purpose <p>One major function of non-suicidal self-injury (NSSI) is emotion regulation, and several interventions include emotion regulation as a core component. However, evidence on the effectiveness of these interventions for NSSI outcomes is mixed. This study examines whether interventions that primarily focus on enhancing emotion regulation (DBT, ERGT, ERITA, and UP) are effective in reducing the frequency of NSSI.</p> Methods <p>We selected both randomized controlled and non-randomized studies of emotion-focused interventions for non-suicidal self-injury through systematic searches in English and Korean. The databases included Web of Science, PubMed, PsycINFO, and PsycArticles for English, and DBpia, SCHOLAR, KISS, and RISS for Korean. We used a random effects model and conducted subgroup analysis. This meta-analysis followed PRISMA guidelines and was registered in PROSPERO (registration number: CRD420261304636).</p> Results <p>This study analyzed 22 effect sizes from 20 studies with 1,580 participants. Emotion regulation-focused interventions had a significant medium effect on the frequency of NSSI (<i>g</i> = 0.61, 95% confidence interval [0.40, 0.83], <i>I</i>² = 85.94%), with high heterogeneity. Subgroup analyses found no moderators influencing this effect. The interventions also had a significant large effect on Difficulties in Emotion Regulation Scale (DERS) scores (<i>g</i> = 0.81, 95% confidence interval [0.64, 0.98], <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({I}^{2}\)</EquationSource> </InlineEquation> = 22.92%), with low heterogeneity. Therapist training was an important moderator in subgroup analyses.</p> Conclusions <p>This meta-analysis identifies a moderate effect of emotion-focused interventions on NSSI frequency and a large effect on DERS scores, highlighting their clinical importance.</p>