Kg-Free: an Acceptance, Mindfulness and Compassionate-based Intervention for Women Struggling With Their Weight

This project aims to develop and test the efficacy of an integrated intervention (Kg-FREE), based on different and complementary approaches from the 3rd generation therapies (Acceptance and commitment therapies - ACT, Mindfulness and Compassion Focused Therapy) that have been showing promising results individually, but that were never tested simultaneously in intervention for people struggling with their eating and weight. Moreover, the intervention aims to act specifically on patients' self-criticism and experiential avoidance patterns in order to reduce weight-self-stigma and promote healthy behaviours and quality-of-life. Indeed, several studies have been suggesting the crucial role of these emotional regulation processes on behavioural modification. This study will investigate the impact of the changes in several psychological processes in different obesity dimensions, such as body mass and biochemical parameters associated with health risks, eating behaviours, stigma, health and quality of life.

The Kg-FREE intervention encloses 10 weekly group sessions (2, 5 hours/ each), highly experiential followed by two 2 booster fortnightly sessions. The intervention is divided into several modules, that are applied in a structured way: a) psycho education related to eating and physical exercise healthy habits; b) how our brain works and how does this relate to our eating and weight; c) Life values identifications to enhance intrinsic motivation; d) cognitive fusion and experiential avoidance as maladaptive strategies that underlie the psychological suffering, the negative affect (e.g. shame) and self-criticism; e) mindfulness exercises as a more adaptive strategy for emotional regulation; f) development of self-compassion as an antidote for shame and self-criticism, activating the soothing system and deactivating the threat system.

Kg-Free sessions were carried out by a clinical psychologist with previous training in contextual-behavioral therapies and one clinical psychology master student. The intervention protocol was built upon pre-existent ACT and mindfulness-based protocols for people with eating and weight issues (e.g., Forman et al., 2013; Kristeller \& Wolever, 2011; Lillis et al., 2009; Tapper et al., 2009). Additionally, and because it has been recognized the importance of integrating compassion in behavioural interventions (e.g., Luoma \& Platt, 2015), specially to deal with shame and self-criticism (Gilbert, 2010) a compassion component was added. This component aimed at training participants to develop a more accepting and kind relationship with themselves, particularly when things go wrong or when facing errors or failures.

All session followed the same basic structure, starting with 30 minutes of shared experience, followed by a five-minute mindfulness practice. Then the session content was delivered, followed by an eating mindfulness practice to train the ability to pay attention to food and eating physical sensations. Finally, the session content was briefly revised and practices for the week were established (e.g., mindfulness or compassion exercises). A participants' manual was provided, which included the targeted constructs, examples and exercise sheets. Audio files were given to the participants to ensure the practice of mindfulness and compassion exercises between sessions. Moreover, a Portuguese therapist manual was also assembled and is available upon request.

Previously to data collection, ethical approval was obtained from Clinical Trials Unit from Coimbra's University Hospital. Participants were adult women, aged between 18 and 55 years old, with overweight or obesity (BMI ≥ 25) without binge-eating, enrolled in nutritional treatment for weight loss in primary care units and Hospitals from Coimbra's district, Portugal. Participants were recruited directly at the medical care units in the day of their appointment. Participants were informed about the voluntary and confidential nature of the data. Also, a brief overview of the treatment program was presented and participants were required to sign the informed consent. In order to assure confidentiality a numerical unique code was assigned to each participant.

After eligibility assessment, participants that meet inclusion criteria were randomly assigned into one of the two study conditions - experimental group that will receive the Kg-Free intervention and a control group that will maintain Treatment as usual (which consisted in dietary and physical activity support), using a using a computer-based random allocation. The study followed the principle of treatment received.

The study comprised four different assessment moments (M0, M1, M2 e M3), namely before intervention (M0), at the end of the intervention (M1) and two follow-up moments at three (M2) and six months (M3) after the intervention. In all assessments data regarding participants' biochemical (lipid profile) and body mass parameters, eating behaviours, health, stigma and quality of life as well as psychological processes was collected. Regarding biochemical data, the blood samples were collected and analysed by the clinical analysis laboratory from the Pharmacy department. Confidentiality was assured hence only the research code for each participant was provided to the laboratory. All participants were weighted with their street clothes (without shoes) using the same Body Composition Analyzer (Tanita TBF-300) accurate to 0.1kg. Waist circumference was measured using a tape measure at the umbilicus. The remaining data collection was carried out by clinical psychologists from the research team (blinded to participant's treatment condition).

Statistical analysis All data analyses were performed using Statistical Package for the Social Sciences (SPSS, V.20) and alpha level was set at .05. Independent sample t tests were used in order to compare intervention and control groups at baseline. Power analysis for ANCOVA was calculated at priori. Using a significance level of p<.05 and a power of 80%, a sample size of 31 per group (N = 62) was needed to detect a standardized effect size of f = 0.40.

To test between-group differences at post-treatment ANCOVA with baseline as covariate and condition as a fixed factor were performed. Paired samples t-tests were performed to explore within-group differences from pre to post test, for each group separately. All effect sizes were calculated using Cohen's d.

In a second study, Repeated Measures ANOVA was performed to test differences between pre-treatment, post- treatment and 3month follow-up assessments for all participant's that completed Kg-Free intervention (N = 53). Whenever sphericity assumption was violated, the Greenhouse-Geisser correction was used to produce a valid F-ratio. Post-Hoc analyses using Bonferroni adjustment for multiple comparisons was used to explore pairwise differences (pre-to-post treatment; pre-to-3 months and post-to-3 months). Effect sizes were calculated using partial eta square (ƞ2). Finally, to explore the psychological processes that underlie the changes in health and weight and eating-related outcomes after the Kg-Free intervention MEMORE macro (Mediation and Moderation analysis for Repeated measures designs) was used. MEMORE allows for the estimation of total, direct, and indirect effects of independent variable (X) on dependent variable (Y) through one or more mediators (M) simultaneously two-occasion within-subjects design.