Table 1.

Data extraction table of included studies

Author	Participants	Details of intervention	Control	Primary outcome measures	Main findings
Fletcher (2011)	T: 72, I: 36, C: 36 Population type: Current or past patients of structured weight-loss programme Gender: 60 female, 12 male Age: 40–64 years, M = 52.6 (no SD reported) Ethnicity: 64 Caucasian, 5 Hispanic/Latino, 2 African American, 1 unknown BMI: M = 35.45 (no SD reported)	Setting: University laboratory hosting structured weight-loss programme Intervention: 1 × 6 h ACT workshop to increase physical activity: introduced ACT concepts covered barriers to physical activity, focused on weight stigma. Facilitated by 3 × Doctoral level graduate students with > 2 years’ experience of delivering ACT	Waitlist control (completed workshop after 3 month follow-up)	Outcome measure: 30-item Weight Self-Stigma Questionnaire (WSSQ) (Lillis et al.; 2009). No process measures Timing of administration Pre- and directly post-workshop, 3 month follow-up	No significant differences in WSQ scores between the ACT and waitlist control conditions at post-workshop assessment or 3-month follow-up Attrition: 59 of 72 dropped out (18.9%). ACT group: 5 did not attend workshop, 2 dropped out after workshop attendance, 29 of 36 (80.6%) completed all 3 assessments. Control group: 1 dropped before completing any assessments, 2 dropped out after first assessment, 30 of 36 (83.3%) completed all 3 assessments
Levin et al. (2017)	T: 13, I: 13, No control Population type: Individuals self-reporting weight self-stigma with a BMI > 27.5 Gender: 12 female and 1 male Age: 18–60, M = 35.10, SD = 12.63 Ethnicity: 9 Caucasian, 1 Asian American BMI: M = 34.11, SD = 5.21	Setting: Participant’s’ home Intervention: ACT self-guided workbook for weight concerns and weight self-stigma; 7 weekly sessions plus weekly 5–10 min coaching. Facilitated by 2 ACT-trained Clinical Psychology Doctoral students	No control	Outcome measure: 12-item Weight Self-Stigma Questionnaire (WSSQ; Lillis et al. 2010) Process measures: Acceptance and Action Questionnaire for Weight (AAQW; Lillis and Hayes 2008);Valuing Questionnaire (VQ, Smout et al. 2014) Timing of administration: pre- and directly post-intervention, 3-month follow up	Compared to baseline: ACT condition significantly lowered weight self-stigma at 3 month follow up with large effect size (M difference = 15.20, SE = 2.63, p < 0.001, Cohen’s d = 2.63) 3 month follow up: ACT condition significantly increased weight-related psychological flexibility with large effect size (M difference = 42.70, SE = 5.77, p < 0.001, Cohen’s d = 2.41); significantly reduced obstruction with large effect size (i.e. how much barriers get in the way of valued action) (M difference = 9.95, SE = 1.16, p < 0.001, Cohen’s d = 2.80); significantly increased progress in valued action with large effect size (i.e. engaging in meaningful patterns of activity) (M difference = 7.23, SE = 1.41, p < 0.001, Cohen’s d = 1.41). However no mediation analysis was conducted on the process variables on weight self-stigma Attrition: 3 dropped out within 3 weeks of using the self-help book
Lillis et al. (2009)	T: 84, I: 44, C: 40 Population type: Completed at least 6 months of a weight loss program in the past 2 years Gender: 76 females and 8 males Age: 39–64.4, M = 50.75, (no SD reported) Ethnicity: 78 Caucasian, 4 Hispanic, 2 African American BMI: M = 33.04	Setting: university laboratory Intervention: 1 day (6 h) group ACT workshop to improve the quality of life of obese people. Focused on obesity stigma, increasing acceptance, mindfulness and values based action Facilitated by experienced ACT facilitator and ACT-trained clinical psychologist trainee	Waiting list control (completed workshop after 3 month follow-up)	Outcome measure: 30-item Weight Self-Stigma Questionnaire (WSSQ) (not validated). Designed for this study as no relevant measure of weight-related self-stigma existed Process measures: Acceptance and Action Questionnaire for Weight (AAQ-W; Lillis and Hayes 2008); Acceptance and Action Questionnaire (Bond and Bunce 2003) Timing of administration: pre-intervention, 3 months post-intervention	3 month follow up: compared to control ACT condition significantly lowered levels of weight-related self-stigma with large effect size (F (1, 83.00) = 24.34, p < 0.001, partial η2 = .23, d = 1.07); improved weight-related psychological flexibility (p = 0.00, CI 4.55–15.01) and general psychological flexibility (AAQ) (p < 0.01, CI 0.82–7.69); significantly mediated reductions in weight related self-stigma; resulted in greater weight loss (post analyses showed improvements in weight related self-stigma not due to weight loss in ACT group) Attrition: 3 in ACT condition completed assessment procedures but dropped out before attending workshop
Palmeira et al. (2017a, b)	T: 73, I: 36, C: 37 Population type: Overweight or obese women self reporting weight self-stigma Gender: all female Age: 18–55, M = 42.35, SD = 8.58 BMI: M = 34.24, SD = 5.05 Ethnicity: not reported	Setting: hospital Intervention: ACT and self-compassion 12 session group intervention designed to reduce weight self-stigma Facilitator: ACT-trained clinical psychologist and one clinical psychology masters student	Treatment as Usual: maintained medical and nutritional appointments	Outcome measure: 12-item Weight Self-Stigma Questionnaire (Portuguese Version) (WSSQ; Lillis et al. 2010; Palmeira et al. 2017a, b) Process measures: Acceptance and Action Questionnaire for Weight-Related Difficulties-Revised (AAQW-R, Palmeira, et al. 2016); Other as Shamer Scale (OAS; Goss et al. 1994; Matos et al. 2016); Self-Compassion Scale (SCS; Castilho et al. 2015; Neff 2003) (contains one positive and one negative subscale: self-compassion and self-judgement) Five Facet Mindfulness Questionnaire—15 (FFMQ-15, Baer et al. 2006) Timing of administration: Pre- and directly post-intervention or within 2-weeks post intervention, 3-month follow up	Post intervention: compared to control ACT condition significantly reduced weight self-stigma with large effect size (F(1, 57) 16.94, p < 0.001, η2 = 0.24).; Reductions in weight self-stigma were mediated by decreased levels of weight-related psychological inflexibility with large effect size (p < 0.01, d = 3.00), shame with large effect size (p < 0.001, d = 1.96,a), self-judgment patterns with large effect size (p < 0.01, d = 1.80) and increased self-compassion with large effect size (p < 0.05, d = 0.96) and mindfulness with large effect size (p < 0.001, d = 1.53) 3-month follow up: Reductions in weight self-stigma from baseline to post-treatment were maintained in the intervention group. However no comparison of the ACT intervention with the control group was included. ACT resulted in greater weight loss (BMI) compared to control. However post analyses showed that improvements in weight related self-stigma not due to weight loss in ACT group Attrition: 9 of 36 in ACT group and 5 of 37 in TAU group dropped out, 2 more dropped out by 3 month follow up
Pearson et al. (2012)	T: 73, I: 39, C: 34 Population type: Women with body dissatisfaction responding to advert Gender: all female Age: 18–68, M = 43.4, SD = 14.7 BMI: M = 29.3, SD = 6.0 Ethnicity: not reported	Setting: university Intervention: 1 day (8 h) group ACT workshop including: creative hopelessness, control as the problem/willingness as the solution, mindfulness and acceptance, values clarification, barriers to values, and committed action. Facilitated by 2 ACT-trained clinical psychologist trainees	Waiting list control condition (completed ACT workshop 2 weeks after intervention group)	Outcome measures: The Physical Appearance State and Trait Anxiety Scale (PASTAS-S) (Reed et al. 1991) The Preoccupation with Eating, Weight and Shape Scale (PEWS; Niemeier et al. 2002); Process measures: Acceptance and Action Questionnaire for Weight (AAQW; Lillis and Hayes, 2008); Acceptance and Action Questionnaire has good reliability and validity (Hayes et al. 2004) Timing of administration: Pre- and directly post-intervention, 1 week post intervention, 2 weeks post intervention	Over time: Compared to control ACT condition significantly reduced body anxiety with medium effect size (t(58.82) = 2.31, p = 0.025, d = 0.55); reduced preoccupation with eating, weight and shape with large effect size t(105.01) = − 2.71, p = 0.01, d = − 1.40. No significant difference between changes in scores between the ACT and control conditions Effect of ACT condition was significantly mediated by: changes in weight-related psychological flexibility (AAQ-W) with large effect size (F(1, 51.16) = 21.05, p = 0.00, d = 1.28); general psychological flexibility (AAQ) with medium effect size F(1, 60.21) = 6.91, p = 0.01 (d = .68) Attrition: 3 of 73 did not attend workshop. 12 out of 34 in control group (35% attrition) and 10 out of 39 ACT group did not complete 1st follow up (26% attrition). 9 out of 34 in control group 26% attritrion and 13 out of 39 (33% attrition) in ACT group did not complete 2nd follow up. Overall 30% attrition
Weineland et al. (2012a); Weineland, et al. (2012b) (6 follow up data)	T: 39, I: 19, C: 20 Population type: Participants 6 months post bariatric surgery Gender: 35 females and 4 males Age: 25–59 years, M = 43.08 (no SD reported) BMI: M = 27.19 (range 20.76—38.01) Ethnicity: not reported	Setting: local surgery centre and home (via internet) Intervention ACT programme for patients who had received bariatric surgery. 8 weekly sessions 2 face-to-face, 6 via Internet, plus 30 min weekly telephone support with trained ACT facilitator Included concepts of acceptance, committed action, values, defusion, and contact with the present moment; acceptance of body image; identified health behaviours values including exercise and eating	Treatment as usual participants received follow-up sessions with bariatric surgery team	Outcome measures: Body shape questionnaire short version (BSQ) Cooper et al. (1987); Eating disorders examination questionnaire (EDEQ)—weight and shape concern subscales (Fairburn and Beglin 1994) Process measures: Acceptance and action questionnaire for weight (AAQ-W; Lillis and Hayes 2008) Timing of administration: Pre- and directly post-intervention, 6 months follow up (see Weineland et al. 2012b)	Post intervention: Compared to control ACT significantly reduced body shape concerns (BSQ) with medium effect size (F(1,37) = 5.65, p = 0.02, η2 = 0.13); significantly reduced shape concerns using EDEQ with large effect (F(1,37) = 7.59, p = 0.009, η2 = 0.17); significantly reduced weight concerns with medium effect (F(1,37) = 5.06, p = 0.03, η2 = 0.12) 6 month follow up: Significant difference in BSQ scores between conditions was maintained with a medium effect size (t [58.65] = 2.09, P = 0.041, d = 0.77), differences in shape concerns (EDEQ), were maintained with a significant medium effect size (t (59.01) = 1.84, P = 0.07, d = 0.68); no significant difference between conditions for weight concerns was identified Improvements in body dissatisfaction were significantly mediated by improvements in weight-related psychological flexibility (P < 0.05) (point estimate = − 8.16, SE = 4.00, 95% CI: -22.75, -2.67) Attrition: 6 of 39 dropped out (15.3%). Out of 3 of 15 (20% attrition) in ACT group and 1 of 18 in control group (5.5% attrition) dropped out at 6 months

NB T total number of participants, I number of participants in ACT intervention group, C number of participants in the control group