The Effects of Acceptance and Commitment Therapy (ACT) Intervention on Inflammation and Stress Biomarkers : a Randomized Controlled Trial

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ The Effects of Acceptance and Commitment Therapy (ACT) Intervention on Inflammation and Stress Biomarkers : a Randomized Controlled Trial © The Authors 2020 Published version Järvelä-Reijonen, Elina; Puttonen, Sampsa; Karhunen, Leila; Sairanen, Essi; Laitinen, Jaana; Kolehmainen, Mikko; Pihlajamäki, Jussi; Kujala, Urho M.; Korpela, Riitta; Ermes, Miikka; Lappalainen, Raimo; Kolehmainen, Marjukka Järvelä-Reijonen, E., Puttonen, S., Karhunen, L., Sairanen, E., Laitinen, J., Kolehmainen, M., Pihlajamäki, J., Kujala, U. M., Korpela, R., Ermes, M., Lappalainen, R., & Kolehmainen, M. (2020). The Effects of Acceptance and Commitment Therapy (ACT) Intervention on Inflammation and Stress Biomarkers : a Randomized Controlled Trial. International Journal of Behavioral Medicine, 25(7), 539-555. https://doi.org/10.1007/s12529-020-09891-8 2020 FULL LENGTH MANUSCRIPT The Effects of Acceptance and Commitment Therapy (ACT) Intervention on Inflammation and Stress Biomarkers: a Randomized Controlled Trial Elina Järvelä-Reijonen 1 &Sampsa Puttonen 2 &Leila Karhunen 1,3 &Essi Sairanen 4,5 &Jaana Laitinen 2 & Mikko Kolehmainen 6 &Jussi Pihlajamäki 1,3 &Urho M Kujala 7 &Riitta Korpela 8 &Miikka Ermes 1,9 &Raimo Lappalainen 4 & Marjukka Kolehmainen 1,3 #The Author(s) 2020 Abstract Background Psychological processes can be manifested in physiological health. We investigated whether acceptance and commitment therapy (ACT), targeted on psychological flexibility (PF), influences inflammation and stress biomarkers among working-age adults with psychological distress and overweight/obesity. Method Participants were randomized into three parallel groups: (1) ACT-based face-to-face (n= 65; six group sessions led by a psychologist), (2) ACT-based mobile (n= 73; one group session and mobile app), and (3) control (n= 66; only the measurements). Systemic inflammation and stress markers were analyzed at baseline, at 10 weeks after the baseline (post-intervention), and at 36 weeks after the baseline (follow-up). General PF and weight-related PF were measured with questionnaires (Acceptance and Action Questionnaire, Acceptance and Action Questionnaire for Weight-Related Difficulties). Results A group × time interaction (p= .012) was detected in the high-sensitivity C-reactive protein (hsCRP) level but not in other inflammation and stress biomarkers. hsCRP decreased significantly in the face-to-face group from week 0 to week 36, and at week 36, hsCRP was lower among the participants in the face-to-face group than in the mobile group (p= .035, post hoc test). Age and sex were stronger predictors of biomarker levels at follow-up than the post-intervention PF. Conclusion The results suggest that ACT delivered in group sessions may exert beneficial effects on low-grade systemic inflammation. More research is needed on how to best apply psychological interventions for the health of both mind and body among people with overweight/obesity and psychological distress. Trial Registration ClinicalTrials.gov Identifier: NCT01738256, Registered 17 August, 2012 Keywords Psychological flexibility .ACT .Mindfulness .Obesity .Low-grade inflammation .Stress *Elina Järvelä-Reijonen [email protected] Sampsa Puttonen [email protected] Leila Karhunen [email protected] Essi Sairanen [email protected] Jaana Laitinen [email protected] Mikko Kolehmainen [email protected] Jussi Pihlajamäki [email protected] Urho M Kujala [email protected] Riitta Korpela [email protected] Miikka Ermes [email protected] Raimo Lappalainen [email protected] Marjukka Kolehmainen [email protected] Extended author information available on the last page of the article International Journal of Behavioral Medicine https://doi.org/10.1007/s12529-020-09891-8 Introduction Chronic low-grade systemic inflammation is a minor activation of the inflammatory system, without actual infection or tissue injury [1], which is present in several metabolic dysfunctional states such as obesity, type 2 diabetes, and cardiovascular disease [2,3]. In low-grade systemic inflammation, the circulating levels of C-reactive protein (CRP) and interleukin-1 receptor antagonist (IL-1Ra) are increased whereas the levels of anti-inflammatory markers, such as adiponectin, are decreased [4]. Elevated high-sensitivity Creactive protein (hsCRP) and IL-1Ra levels and low adiponectin levels predict a persistence of the metabolic syndrome [5]. A chronic low-grade systemic inflammation also often co-occurs with a chronic stress response [6]. Stress is a complex concept with no uniform definition [7]. The response to a stressful situation is usually both psychological and physiological [8,9]. The physiological stress response, i.e., the activation of the hypothalamic-pituitary-adrenocortical (HPA) axis and sympathetic nervous system [10], leads to altered secretion of cortisol and dehydroepiandrosterone sulfate (DHEAS) [11,12]. Cortisol and DHEAS are markers of HPA axis activation [12]and,thusinthisstudy, are exploited as markers of either acute or chronic stress reactions [11–14]. Altered cortisol secretion is associated with, for example, immune and inflammatory outcomes, obesity, cancer, and increased risk for mortality [15]. DHEAS is known to counteract several of the effects of cortisol and to, for example, attenuate inflammatory process and reduce mortality [12, 16,17]. Allostatic load is a holistic concept of how chronic stress can lead to negative health outcomes [18,19]. Markers of inflammation and HPA axis activity are suggested to mediate the effects of chronic stress underpinning disturbed health [14,20,21]. Thus, finding ways to improve the circulating levels of the inflammatory and stress biomarkers would be one approach to improving long-term health. In addition to chronic stress, also obesity and lifestyle factors such as diet and physical activity are related to low-grade systemic inflammation [22].Adiposetissueisamajorsourceofstressrelated pro-inflammatory markers [22]. Conversely, a diet consistent with nutrition recommendations, such as a healthy Nordic diet, is associated with better inflammatory status [23,24]. Obesity, psychosocial distress, a diet not following nutritional recommendations, and a sedentary lifestyle are major health challenges in our modern society. For an individual, these challenges are difficult to overcome—it is not easy to make long-term lifestyle changes. For example, the majority of individuals tend to regain weight after weight loss [25]. Often, these difficulties arise from our problematic inner experiences and psychological processes. For example, individuals who gain weight after they have lost weight tend to escape or avoid their problems [26] and adopt an inflexible, dichotomous thinking style [27,28]. Psychological inflexibility means that the person is not able to change or maintain a behavior which would lead to personally valued outcomes and is not able to fully contact the present moment [29]. The non-adaptive psychological processes, such as psychological inflexibility [29], are thought to underlie a wide variety of forms of human suffering [30], irrespective of diagnoses [31]. Psychological inflexibility is also associated with higher uncontrolled and emotional eating [32] which are features of eating behavior associated with poorer dietary choices and higher BMI [33–35]. Thus, increasing psychological flexibility, i.e., the ability to contact the present moment and to change or persist in behavior according to personal values [29], could be one possible way to promote an individual’s health and well-being. One of the so-called process-based therapies aimed at improving psychological flexibility in general is acceptance and commitment therapy (ACT) [36]. ACT seeks to increase psychological flexibility and consists of six interrelated core processes: (1) clarification of own values, (2) commitment to act based on those values, (3) being in contact with the present moment (i.e., mindfulness), (4) having self as context (i.e., being aware of thoughts, feelings, etc., without attaching to them), (5) defusion (i.e., altering the way to interact with or relate to thoughts, feelings, etc.), and (6) acceptance [29]. Previous studies have investigated the effects of ACT or value clarification interventions on the cortisol response in individuals performing an experimental psychosocial stress test [37,38]. To the best of our knowledge, there are no previous studies about the effectiveness of an ACT intervention on inflammation and stress biomarkers in real-world non-clinical, randomized controlled settings. However, interventions based on one part of ACT, namely mindfulness, have shown positive effects on some inflammation and stress biomarkers such as hsCRP [39,40] and cortisol [41]. However, the reviews of mindfulness-based randomized controlled interventions have concluded that the effects on circulating inflammation markers [42–44] and cortisol [44] are contradictory, and warrant further research. In summary, based on the literature, it can be hypothesized that problems in psychological processes (i.e., psychological inflexibility) are closely related to behavior (e.g., poor diet and sedentary lifestyle leading to obesity) and subjective psychosocial distress. All these factors are linked to chronic lowgrade systemic inflammation and a chronic physiological stress response. These physiological processes, in turn, are a risk of future ill health and major chronic diseases. The aim of this study was to investigate how the effects of an ACT intervention, targeting psychological inflexibility, can alter inflammation and stress biomarkers among adults with psychological distress and overweight. In addition, because the psychological and physiological factors are intertwined, we conducted post hoc, exploratory analyses to explore how psychological flexibility (general or weight-related) and the biomarkers Int.J. Behav. Med. are clustered within this study population or whether they cluster more strongly with other psychological, anthropometric, or lifestyle (diet, physical activity) measures. Methods Study Design The present study is a secondary analysis of a parallel-arm randomized controlled trial, in which three different psychological interventions were studied and compared to a non-treatment control group [45]. The randomization process and CONSORT flow chart have been published previously [45]. The present study investigates the effects of the two ACT-based intervention arms compared to control. The study participants were recruited by advertisements in local newspapers and screened for eligibility via telephone inquiry and an online questionnaire from August 2012 until January 2013. The participants had to be 25–60 years old and have a self-reported body mass index (BMI) of 27–34.9 kg/ m 2 . This range for self-reported BMI was used because we aimed to have participants with overweight and obesity and anticipated that the range of laboratory-measured BMI would be most likely wider. The participants also had to be psychologically distressed (≥3/12 points from the General Health Questionnaire (GHQ-12) [46]) and have computer and Internet access. There were several exclusion criteria, such as diagnosed severe chronic illness (including symptomatic cardiovascular disease, type 1 or 2 diabetes, kidney disease requiring dialysis, and eating disorder), disabilities/illnesses substantially affecting physiological or mental health, medical surgery within the past 6 months, heart attack or stroke within the past 6 months, pacemaker, regular oral cortisone medication, pregnancy or breastfeeding within the past 6 months, psychotherapy or other psychological or mental treatments at least twice a month, disability pension for psychological reasons, and participation in some other intervention trial during the present study. An additional exclusion criterion for the present analyses was a hsCRP value > 10 mg/L indicating acute infection or inflammation [47]. The study was conducted in three cities in Finland (Jyväskylä, Kuopio, and Helsinki), and the recruitment was conducted in two phases. The participants in the first phase started in autumn, and the second phase started in spring. The participants filled in electronic questionnaires, visited the local study center for clinical and biochemical measurements, and reported their food consumption in a 48-h dietary recall by telephone. Measurements were conducted before the intervention (baseline, study week 0), after the 8-week intensive intervention period (post-intervention, study week 10), and 36 weeks after baseline measurements (follow-up, study week 36). The measurements were collected from August 2012 until December 2013. The sample size of the current study is based on the power calculation (for depression symptoms) of the randomized controlled trial [45], resulting in a sample size of n=80–85 per group. Study Groups The face-to-face and mobile interventions were based on the same ACT program constructed by the same research group. Thus, only the delivery method of the intervention differed. The two interventions included the following main components: value clarification, acting according to one’s own values, mindfulness skills, the observing self (e.g., observing thoughts without being caught up in them), and acceptance skills (e.g., making room for unpleasant feelings and urges, allowing them to come and go). The main focus was on ACT skills, but minor parts of mindful eating, relaxation, and everyday physical activity were also included. However, the intervention did not include nutrition education. Only a hyperlink to a public nutritional website was provided to the participants in the intervention groups, which was to be utilized if the dietary changes were according to one’s own values. Lappalainen et al. [45]fora more detailed description of the intervention. The face-to-face group had six group sessions led by a psychologist during the 8-week intervention period. Each session took approximately 90 min, and each group consisted of 6–12 participants. The sessions included exercises, pair and group discussions, and homework for which the participants received a workbook. The treatment adherence has been described previously and was found to be good [48]. On average, the participants attended five group sessions (SD = 1). The mobile group had one group session in which participants learned about the principles of ACT and received smartphones with the pre-installed Oiva mobile app [49]. The Oiva app contains 46 exercises in text and audio formats and introductory videos about the ACT skills. The user experience results of the app were positive [49]. The participants were free to choose exercises and videos in any order and to do them as many times as the participants wanted during the 8-week intervention period. The participants returned the smartphones during the post-intervention laboratory study visit. The participants’usage of the mobile app has been reported in detail previously [50]. Briefly, among the study population of the present analyses, the median total duration of the mobile app use was 280 min (interquartile range 198–423, total range 52–2001). Participants randomized to the control group took part in all of the measurements but did not receive any intervention. After the follow-up measurements, the participants in the control group had an opportunity to attend one group session in which the principles of ACT were presented and they were told that they could utilize an Internet-based lifestyle coaching program [45]. Int.J. Behav. Med. Measures Inflammation and Stress Biomarkers An antecubital venous blood sample was taken after a 12-h overnight fast in the study laboratory between 7 and 10 a.m. The plasma samples were collected into EDTA tubes and centrifuged as soon as possible. The serum samples were centrifuged after the blood had clotted. The samples were stored at −80 °C until analyzed. Inflammation markers known to be associated with metabolic syndrome (MetS) components (i.e., high hsCRP and IL-1Ra levels and low adiponectin level) [51] were analyzed. Plasma hsCRP concentration was determined with a photometric immunoturbidimetric method (Konelab; Thermo Fisher Scientific, Vantaa, Finland), with a measurement range from 0.1 to 10 mg/L, and extended range using automatic dilution from 0.1 to 40 mg/L. The level of plasma IL-1Ra was measured with an enzyme immunoassay (Quantikine ® ELISA Kits; R&D Systems, Inc., Minneapolis, USA) with a measurement range from 31.2 to 2000 pg/mL. Serum high molecular weight (HMW) adiponectin was measured with enzyme immunoassay (Quantikine ® ELISA for Human HMW Adiponectin/Acrp Immunoassay; R&D Systems, Inc., Minneapolis, USA) with a measurement range from 0.39 to 25 μg/mL. Plasma total cortisol concentrations were measured with chemiluminescent immunoassay (LIAISON ® Cortisol; DiaSorin, Saluggia, Italy) with a quantitation limit of 4.1 nmol/L and a dilution threshold of 2208 nmol/L. Plasma DHEAS was determined with chemiluminescent immunoassay (LIAISON ® DHEA-S; DiaSorin, Saluggia, Italy) with a quantitation limit of 0.027 μmol/L and a dilution threshold of 20.3 μmol/L. The cortisol/DHEAS ratio was used as a more sensitive index of the HPA axis activation [12]andthe catabolic/anabolic balance [14,52] under stress conditions. The cortisol/DHEAS ratio was calculated by dividing the raw value of cortisol (ng/mL) by the raw value of DHEAS (ng/mL) [52]. The ratio between cortisol and DHEAS, i.e., the balance of these catabolic and anabolic stress hormones, is suggested to be more informative of psychiatric and general health status than either of the hormone levels alone [14]. Although the DHEAS concentration increases as a part of an acute stress response [14], its levels decline during chronic stress conditions [12,14]. DHEAS has been related to several positive health effects [17] and has been shown to counteract several of cortisol’seffects[12,16]. Thus, a higher cortisol/ DHEAS ratio is proposed to indicate higher chronic stress and to contribute to ill health [14]. Anthropometric Measurements Weight and height were measured with calibrated instruments in the study laboratory in the morning after a 12-h overnight fast. BMI was calculated as kilograms per meters squared. Waist circumference was measured halfway between the lowest rib and the iliac crest. Body composition (% of body fat) is based on multifrequency bioelectrical impedance analysis using an In-Body 720 device (Mega Electronics, Kuopio, Finland) or Tanita BC-418 MA device (Tanita, Japan). Psychological Flexibility Psychological flexibility was measured with two questionnaires. The Acceptance and Action Questionnaire (AAQ-II) [53] measures general psychological flexibility. The 7 items (e.g., “I worry about not being able to control my worries and feelings”) are answered with 7-point Likert scale from “never true”(1) to “always true”(7). The possible score range is 7–49 with a lower score reflecting more psychological flexibility. Cronbach’salphawas0.91. The Acceptance and Action Questionnaire for WeightRelated Difficulties (AAQW) [54] measures weight-related psychological flexibility. The 22 items (e.g., “I try hard to avoid feeling bad about my weight or how I look”) are answered with a 7-point Likert scale from “never true/not at all believable”(1) to “always true/completely believable”(7).Thepossiblescore range is 22–154. Here too, a lower score reflects more psychological flexibility related to difficult weight-related thoughts and feelings. Cronbach’salphawas0.90. Psychological Stress and Symptoms of Depression The 12-item General Health Questionnaire (GHQ-12) [46] was used to screen the volunteers for psychological distress. The GHQ-12 has been found to be a valid screening tool for common mental health problems in the Finnish population [55]. The respondents were asked, considering the past few weeks, to answer questions such as “Have you recently felt constantly under strain?”In the screening, a bimodal scoring system was used (“not at all”(0 point), “same as usual”(0), “rather more than usual”(1), and “much more than usual”(1), with the total sum score ranging from 0 to 12). Cronbach’s alpha was 0.73. In the present analyses, Likert scoring (0, 1, 2, and 3 points; possible range 0–36) was used in the statistical analyses to achieve a larger variation in the GHQ-12 scores. Cronbach’s alpha using Likert scoring was 0.82. The 14-item Perceived Stress Scale (PSS-14) [56] was used to assess the degree to which a person perceives life as being stressful. The questionnaire has demonstrated acceptable psychometric properties worldwide [57]. Questions concern how often a person has experienced certain feelings and thoughts during the previous month, e.g., “In the last month, how often have you found that you could not cope with all the things that you had to do?”The 5-point Likert scale ranged from “never”(0) to “very often”(4) and is summed for the total score (possible range 0–56). Cronbach’s alpha was 0.88. Symptoms of Int.J. Behav. Med. depression were measured by the 21-item Beck Depression Inventory-II (BDI-II) [58]. The 4-point Likert scale is scored from 0 to 3, and the scores are summed to calculate the total score (possible range 0–63). Cronbach’s alpha was 0.87. For all these measures, higher scores indicate higher psychological distress and depressive symptoms. Diet and Physical Activity Index of Diet Quality (IDQ) questionnaire [59] measured adherence to Nordic and Finnish nutrition recommendations. Apparently unrealistic answers (e.g., 27 slices of bread per day) were verified from the participant when possible or coded as missing. Alcohol consumption during the previous 6 months was assessed using Alcohol Use Disorders Identification Test Consumption (AUDIT-C) questionnaire [60]. A 48-h dietary recall was conducted on the telephone between Tuesday and Friday. The time for the telephone call was pre-scheduled for practical reasons, and the participants were informed that the interviewer would ask about their diet. However, it was not mentioned that the interview will consider their diet during the past 48 h. The interview protocol was planned based on the protocol used in national FINDIET 2012 Survey [61]. We placed a special emphasis on the protocol to assist the participant to remember and give accurate information of the foods and drinks consumed. The protocol has been described in more detail previously [62]. Nutrient intake was calculated based on the 48-h dietary recall using AivoDiet software (v 2.0.2.2; Aivo Ltd., Turku, Finland), utilizing the Fineli ® Finnish Food Composition Database (National Institute for Health and Welfare, Nutrition Unit, Helsinki, Finland). Leisure time physical activity and commuting activity were assessed by a questionnaire [63,64]. Leisure time metabolic equivalent (MET) index (MET h/day) was calculated as a sum score of the different activities multiplied by the intensity (MET), duration (h), and frequency of the activity [63,64]. Statistical Methods Statistical analyses were conducted with IBM SPSS Statistics (version 23) and MATLAB R2017b. A pvalue < .05 was considered as statistically significant. Baseline differences between the groups were analyzed using Pearson chi-square test for categorical variables and one-way ANOVA with the Tukey HSD post hoc test for continuous variables. The normality assumption was assessed by the histograms of the residuals. If the normality assumption was not met, the nonparametric Kruskal-Wallis test was used. In the results, descriptive values of the normally distributed variables are presented as means ± SD and of the non-normally distributed variables as medians (interquartile range). General linear mixed model was used to analyze the differences between the three study groups using all three time points (group × time interaction) and the main effect of time on outcome variables. Participants were included as random effects and intercept, group, time, interaction term, and covariates as fixed effects. In case of a statistically significant group × time interaction, the Sidak post hoc test was conducted for pairwise comparisons. The analysis utilizes all of the available data with the assumption of missing data as “missing at random.”The missing data also seemed to be “missing at random”as there were no differences in the study group, starting time of the study, study center (chi-square test, p= .303, p= .460, and p= .496, respectively), gender (Fisher’s exact test, p= .302), age, GHQ-12 score, and baseline BMI (ttest, p=.548,p= .374, and p= .141, respectively) between the participants who provided all data as compared to the participants with missing data. Participants considered as outliers (over mean ± 5 SDs) at any of the three time points were excluded. The distributions of the values were so wide that using the cutoff of mean ± 5 SDs enabled to exclude the true, explicit outliers. The normality assumption was assessed by the residual histograms. The non-normally distributed outcome variables hsCRP, IL-1Ra, HMW adiponectin, and cortisol/ DHEAS ratio were log-transformed. All the analyses were adjusted for study center and starting time of the study (i.e., basic adjusted model) and, in a fully adjusted model, also for age, sex, and baseline BMI. Furthermore, to study the effect of baseline BMI on the intervention effects, a group × time × baseline BMI interaction term was added into the model. Linear regression analysis was used to investigate whether psychological flexibility (general or weight-related) after the intervention (week 10) predicted the levels of inflammation and stress biomarkers after the follow-up period (week 36) among the participants in the ACT intervention groups. At first, psychological flexibility, age, sex, study center, and starting time of the study were included in the model as independent variables. Because study center and starting time of the study were non-significant in each model, they were excluded from the final models. The assumptions were evaluated with the Durbin-Watson test (between 1 and 3), tolerance (0.1–1), VIF (1–10), studentized residuals (about 95% between −2 and 2), Cook’s distance < 1, linearity in scatter plots, and normality in residual histograms [65]. All the models met the assumptions. Exploratory principal component analyses (PCAs) were conducted for psychological, physiological, anthropometric, and lifestyle measures. These factors are often interrelated, but we wanted to explore if the measures were clustered in this sample. Two analyses were performed: (a) with values at baseline and (b) for changes from baseline to week 36 (Δ36 = week 36 value −week 0 value). The PCAs were intended to be descriptive and hypothesis generating instead of being formal hypothesis testing. The intention of these post hoc PCAs was Int.J. Behav. Med. to explore how psychological flexibility (general or weightrelated) and the biomarkers would cluster among the study participants: do they correlate with each other or correlate more strongly with other psychological, anthropometric, and lifestyle measures. In the PCAwith the changes from baseline to week 36, missing data was handled by excluding the cases with any missing data. Of the 200 participants with data from both time points, 11% (n= 21) had to be excluded because of hsCRP > 10 and 3% (n=5)wereexcludedbecauseofamissing value in any of the variables resulting in 174 participants providing full data for this analysis. An outlier with a value over mean ± 5 SDs was excluded (n= 1 because of a high increase in HMW adiponectin value) resulting in n=173. PCAs were conducted with standardized values. The Humphreys-Ilgen parallel analysis was used to determine the number of components [66,67]. The components were rotated using varimax rotation [68]. Loadings of the rotated components are reported. A threshold of 0.3 was used for interpreting the baseline components’loadings. A threshold of 0.25 was used for interpreting the change (Δ36) components’loadings in order to include the inflammation markers into the components. Results Participants Of the 254 individuals randomized to the face-to-face (n=84), mobile (n= 85), and control (n= 85) groups, 219 participated in baseline measurements (Table 1). At baseline, 13 participants (6%) had a hsCRP value > 10 mg/L indicating acute Table 1 Characteristics of the participants in each group at baseline (n=204) Face-to-face Mobile Control p a Number of participants (n) Randomized 84 85 85 Participated in baseline measurements 70 78 71 Included in the present analyses 65 73 66 Starting time of the study (n) Autumn b 32 34 29 .832 Spring c 33 39 37 Study center (n) Jyväskylä 18 20 16 .985 Kuopio 21 25 22 Helsinki 26 28 28 Sex (n) Female 56 61 54 .795 Male 9 12 12 Age (years) 51.0 ± 6.5 48.8 ± 7.7 49.0 ± 7.5 .148 Weight (kg) 85.9 ± 10.4 88.3 ± 10.4 88.5 ± 11.5 .308 BMI (kg/m 2 ) 30.8 ± 3.0 31.5 ± 2.7 31.3 ± 2.9 .405 Waist circumference (cm) 101.6 ± 9.0 103.3 ± 7.9 103.5 ± 9.1 .407 Psychological distress (GHQ-12 score) 7.0 (4.5–9.5) 6.0 (5.0–9.0) 7.0 (5.0–10.0) .564 d Perceived stress (PSS score) 25.8 ± 8.0 26.9 ± 7.7 27.1 ± 7.6 .558 General psychological flexibility (AAQ-II score) 19.9 ± 8.6 20.2 ± 9.2 21.5 ± 9.3 .527 Weight-related psychological flexibility (AAQW score) 84.3 ± 19.3 87.7 ± 21.1 87.8 ± 21.2 .540 Diet quality (IDQ score) 10.4 ± 2.1 10.5 ± 1.9 10.2 ± 2.1 .620 Leisure time physical activity (MET index) 3.6 ± 2.9 3.3 ± 3.2 2.9 ± 3.0 .435 Values are n/ mean ± SD / median (interquartile range) BMI body mass index, GHQ-12 General Health Questionnaire-12, PSS Perceived Stress Scale, AAQ-II Acceptance and Action Questionnaire, AAQW Acceptance and Action Questionnaire for Weight-Related Difficulties, IDQ Index of Diet Quality, MET metabolic equivalent a pvalue for differences between the study groups (Pearson chi-square for categorical variables and one-way ANOVA for continuous variables unless otherwise noted) b September–October 2012 c January–February 2013 d Non-parametric Kruskal-Wallis test Int.J. Behav. Med. infection or inflammation [47] and were excluded from the present analyses. For the same reason, nine participants (4%) were excluded from the week 10 and 15 participants (8%) from week 36 assessments. In addition, one participant was excluded from the analyses at each measurement time point (different persons each time) because of a high IL-1Ra value considered to be an outlier (over 5 SDs above the mean) and one participant (same person each time) because of a very high adiponectin value (over 5 SDs above the mean). The numbers of participants included in the present analyses are shown in Tables 1and 2. The measured BMI of the participants at the baseline laboratory visit (n= 204, 84% females) ranged between 25.7 and 40.1 kg/m 2 (mean ± SD 31.2 ± 2.9 kg/m 2 ). Age ranged between 27 and 61 years (50 ± 7 years). Less than a third (28%) had a high relative risk for cardiovascular disease (hsCRP > 3.0 mg/L), whereas 36% had hsCRP levels reflecting low risk (< 1 mg/L) and 36% reflecting average (1.0 to 3.0 mg/L) risk [69]. There were no differences in the baseline demographic, clinical, psychological, and lifestyle characteristics between the three study groups (Table 1). Intervention Effects on Inflammation and Stress Biomarkers During the entire study period, a group × time interaction was found in hsCRP but not in the other biomarkers (Table 2). In the basic adjusted model, the level of hsCRP decreased significantly in the face-to-face group from week 0 to week 36 (p= .045 for the post hoc test) and from week 10 to week 36 (p= .014, post hoc test). The only difference between the groups was that the hsCRP was lower among the participants in the face-to-face group as compared to the mobile group in the week 36 (p= .035, post hoc test) with a small-to-medium effect size (r=−.196, p= .035) (Fig. 1). Thus, the ACT groups did not differ from the control group significantly. In the fully adjusted model, the difference between the face-toface and mobile groups was no longer apparent in week 36, with merely a trend (p= .062, post hoc test) being detected. In this model, baseline BMI was significant (estimate .054, p< .001). However, the decrease in hsCRP from week 0 to week 36 (p= .048, post hoc test) and from week 10 to week 36 (p= .014, post hoc test) remained significant in the face-toface group. The hsCRP values presented in Table 2are the actual unestimated values (i.e., without adjustments) for clinical interpretation. Figure 1shows the logarithmically transformed estimated (adjusted for study center and starting time of the study) values from the statistical analysis. The estimated marginal mean ± SE hsCRP values in the face-to-face group for weeks 0, 10, and 36 without logarithmic transformation were 1.955 ± 0.253 mg/L, 2.150 ± 0.261 mg/L, and 1.759 ± 0.264 mg/L, respectively. During the entire study period, a significant decrease (p< .001) in IL-1Ra over time was evident, in all of the study groups (Table 2). The logarithmically transformed estimated marginal mean ± SE IL-1Ra values for weeks 0, 10, and 36 of the fully adjusted model were 2.588 ± .014, 2.597 ± .014, and 2.555 ± .014, respectively. In addition, a trend (p= .056) towards an increase in the cortisol/DHEAS ratio was observed (log-estimated marginal mean ± SE values for weeks 0, 10, and 36 of the fully adjusted model −.887 ± .019, −.893 ± .019, and −.864 ± .019, respectively). There were no statistically significant differences (pvalues > .186) in the baseline values of the inflammation and stress biomarkers between the groups (Table 2). The intervention effect was not dependent on baseline BMI (group × time × baseline BMI interaction non-significant for all inflammation and stress biomarkers). Association of Psychological Flexibility with Inflammation and Stress Biomarkers We further analyzed whether the main target of the ACT intervention, psychological flexibility, was associated with the levels of inflammation and stress biomarkers among the participants in the ACT intervention groups (face-to-face and mobile) over the long term. During the intensive intervention period (from baseline to week 10), general psychological flexibility increased in most of the participants (54%), but it decreased in 31%, while remaining the same in 15% of the participants in the ACT groups. Weight-related psychological flexibility increased in 66%, decreased in 33%, and remained the same in 1% of the participants in the ACT groups. The effects of the intervention on psychological flexibility have been reported previously [70]. The level of psychological flexibility (general or weightrelated) after the intervention was not a strong predictor of the levels of inflammation and stress biomarkers at the end of the follow-up period (Table 3). Only general psychological flexibility was a significant predictor (p= .041) such that a higher AAQ-II score, i.e., lower psychological flexibility, predicted a higher DHEAS level. In addition, lower age (p< .001) and male sex (p= .002) were also predictors of a higher DHEAS concentration in the model. Explorative Analysis of Factors Associated with Inflammation and Stress Biomarkers To explore the associations between psychological well-being with inflammation and stress biomarkers, considering also clinical and lifestyle factors related to low-grade systemic inflammation, we conducted post hoc, exploratory PCAs (a) of the variables measured at baseline and (b) of the changes during the entire study period. Int.J. Behav. Med. Table 2 The effects of the ACT-based face-to-face and mobile interventions on inflammation and stress biomarkers (n= 192) analyzed with the linear mixed model Face-to-face Mobile Control p a (group ×time) p a (time) 0 week (n= 64) 10 weeks (n= 57) 36 weeks (n= 55) 0week (n= 67) 10 weeks (n= 64) 36 weeks (n= 62) 0week (n= 61) 10 weeks (n= 58) 36 weeks (n=57) hsCRP (mg/L) b 1.2 (0.6–2.3) 1.5 (0.5–2.6) 1.2 (0.4–2.0) 1.6 (0.5–3.7) 1.6 (0.9–2.7) 1.7 (0.8–3.3) 1.5 (0.5–3.0) 1.4 (0.6–2.9) 1.5 (0.6–3.3) .012 .391 .012 .406 IL-1Ra (pg/mL) b 356.5 (258.8–528.1) 383.6 (263.3–522.3) 301.1 (248.5–425.5) 366.6 (265.9–607.8) 366.1 (258.3–559.4) 333.1 (246.5–482.1) 417.1 (277.2–572.8) 387.0 (281.7–594.8) 367.1 (292.5–526.3) .883 <.001 .880 <.001 HMW adiponectin (μg/mL) b 4.9 (2.8–7.3) 4.9 (3.2–6.7) 5.1 (2.8–7.8) 5.0 (3.3–7.8) 4.7 (3.3–7.1) 4.8 (3.0–7.4) 4.8 (3.1–6.7) 4.7 (3.0–6.9) 4.8 (3.0–7.1) .353 .848 .353 .877 Cortisol (ng/mL) 115.6 ± 45.9 109.7 ± 39.2 117.3 ± 42.6 123.1 ± 47.9 117.3 ± 42.5 118.5 ± 39.0 108.9 ± 35.8 116.1 ± 41.4 119.2 ± 40.3 .272 .467 .287 .466 DHEAS (ng/mL) 982.3 ± 518.4 969.1 ± 468.5 952.7 ± 495.0 1013.7 ± 540.0 1015.1 ± 547.2 997.3 ± 493.4 939.4 ± 424.6 951.7 ± 440.6 908.3 ± 440.6 .887 .093 .884 .098 Cortisol /DHEAS ratio b 0.12 (0.09–0.19) 0.12 (0.08–0.17) 0.13 (0.07–0.20) 0.12 (0.08–0.19) 0.12 (0.08–0.18) 0.13 (0.08–0.20) 0.12 (0.09–0.16) 0.13 (0.09–0.16) 0.13 (0.10–0.21) .129 .053 .119 .056 The values are unestimated medians (interquartile range) / means ± SD (i.e., without adjustments). There were no statistically significant differences (pvalues > .186) in the baseline values between the groups hsCRP high-sensitivity C-reactive protein, IL-1Ra interleukin-1 receptor antagonist, HMW high molecular weight, DHEAS dehydroepiandrosterone sulfate a pvalue for differences between the three study groups using all measured time points (study weeks 0, 10, and 36) analyzed with the linear mixed model. Thepvalue above is adjusted for study center and starting time of the study, and the pvalue below is adjusted for study center, starting time of the study, age, sex, and baseline BMI. Values in italics indicate a significant pvalue < 0.05 b Analyzed with logarithmically transformed values. Values presented in the table are non-transformed Int.J. Behav. Med. 10. Stratakis CA, Chrousos GP. Neuroendocrinology and pathophysiology of the stress system. Ann N Y Acad Sci. 1995;771:1–18. 11. Miller GE, Chen E, Zhou ES. If it goes up, must it come down? Chronic stress and the hypothalamic-pituitary-adrenocortical axis in humans. Psychol Bull. 2007;133(1):25–45. 12. Kamin HS, Kertes DA. Cortisol and DHEA in development and psychopathology. Horm Behav. 2017;89:69–85. 13. Tomiyama AJ, Dallman MF, Epel ES. Comfort food is comforting to those most stressed: evidence of the chronic stress response network in high stress women. Psychoneuroendocrinology. 2011;36(10):1513–9. 14. Wolkowitz OM, Epel ES, Reus VI. Stress hormone-related psychopathology: pathophysiological and treatment implications. World J Biol Psychiatry. 2001;2(3):115–43. 15. Adam EK, Quinn ME, Tavernier R, McQuillan MT, Dahlke KA, Gilbert KE. Diurnal cortisol slopes and mental and physical health outcomes: a systematic review and meta-analysis. Psychoneuroendocrinology. 2017;83:25–41. 16. Maninger N, Wolkowitz OM, Reus VI, Epel ES, Mellon SH. Neurobiological and neuropsychiatric effects of dehydroepiandrosterone (DHEA) and DHEA sulfate (DHEAS). Front Neuroendocrinol. 2009;30(1):65–91. 17. Traish AM, Kang HP, Saad F, Guay AT. Dehydroepiandrosterone (DHEA)–a precursor steroid or an active hormone in human physiology. J Sex Med. 2011;8(11):2960–82. 18. McEwen BS, Stellar E. Stress and the individual: mechanisms leading to disease. Arch Intern Med. 1993;153:2093–101. 19. McEwen BS. Protective and damaging effects of stress mediators. N Engl J Med. 1998;338(3):171–9. 20. McEwen BS, Seeman T. Protective and damaging effects of mediators of stress. Elaborating and testing the concepts of allostasis and allostatic load. Ann N Y Acad Sci. 1999;896:30–47. 21. McEwen BS. Biomarkers for assessing population and individual health and disease related to stress and adaptation. Metabolism. 2015;64(3 Suppl 1):S2–S10. 22. Calder PC, Ahluwalia N, Brouns F, Buetler T, Clement K, Cunningham K, et al. Dietary factors and low-grade inflammation in relation to overweight and obesity. Br J Nutr. 2011;106(Suppl 3): S5–S78. 23. Kanerva N, Kaartinen NE, Rissanen H, Knekt P, Eriksson JG, Sääksjärvi K, et al. Associations of the Baltic Sea diet with cardiometabolic risk factors–a meta-analysis of three Finnish studies. Br J Nutr. 2014;112(4):616–26. 24. Kolehmainen M, Ulven SM, Paananen J, de Mello V, Schwab U, Carlberg C, et al. Healthy Nordic diet downregulates the expression of genes involved in inflammation in subcutaneous adipose tissue in individuals with features of the metabolic syndrome. Am J Clin Nutr. 2015;101(1):228–39. 25. Langeveld M, DeVries JH. The long-term effect of energy restricted diets for treating obesity. Obesity (Silver Spring). 2015;23(8):1529–38. 26. Kayman S, Bruvold W, Stern JS. Maintenance and relapse after weight loss in women: behavioral aspects. Am J Clin Nutr. 1990;52(5):800–7. 27. Byrne S, Cooper Z, Fairburn C. Weight maintenance and relapse in obesity: a qualitative study. Int J Obes Relat Metab Disord. 2003;27(8):955–62. 28. Byrne SM, Cooper Z, Fairburn CG. Psychological predictors of weight regain in obesity. Behav Res Ther. 2004;42(11):1341–56. 29. Hayes SC, Luoma JB, Bond FW, Masuda A, Lillis J. Acceptance and commitment therapy: model, processes and outcomes. Behav Res Ther. 2006;44(1):1–25. 30. Kashdan TB, Rottenberg J. Psychological flexibility as a fundamental aspect of health. Clin Psychol Rev. 2010;30(7):865–78. 31. Garland EL, Howard MO. A transdiagnostic perspective on cognitive, affective, and neurobiological processes underlying human suffering. Res Soc Work Pract. 2014;24(1):142–51. 32. Nevanperä N, Lappalainen R, Kuosma E, Hopsu L, Uitti J, Laitinen J. Psychological flexibility, occupational burnout and eating behavior among working women. Open J Prev Med. 2013;3(4):355–61. 33. de Lauzon B, Romon M, Deschamps V, Lafay L, Borys JM, Karlsson J, et al. The Three-Factor Eating Questionnaire-R18 is able to distinguish among different eating patterns in a general population. J Nutr. 2004;134(9):2372–80. 34. Keskitalo K, Tuorila H, Spector TD, Cherkas LF, Knaapila A, Kaprio J, et al. The Three-Factor Eating Questionnaire, body mass index, and responses to sweet and salty fatty foods: a twin study of genetic and environmental associations. Am J Clin Nutr. 2008;88(2):263–71. 35. Nevanperä NJ, Hopsu L, Kuosma E, Ukkola O, Uitti J, Laitinen JH. Occupational burnout, eating behavior, and weight among working women. Am J Clin Nutr. 2012;95(4):934–43. 36. Hayes SC, Hofmann SG. The third wave of cognitive behavioral therapy and the rise of process-based care. World Psychiatry. 2017;16(3):245–6. 37. Gregg JA, Namekata MS, Louie WA, Chancellor-Freeland C. Impact of values clarification on cortisol reactivity to an acute stressor. J Contextual Behav Sci. 2014;3:299–304. 38. Gloster AT, Klotsche J, Aggeler T, et al. Psychoneuroendocrine evaluation of an acceptance and commitment based stress management training. Psychother Res. 2017:1–11. https://doi.org/10.1080/ 10503307.2017.1380862. 39. Dalen J, Smith BW, Shelley BM, Sloan AL, Leahigh L, Begay D. Pilot study: Mindful Eating and Living (MEAL): weight, eating behavior, and psychological outcomes associated with a mindfulness-based intervention for people with obesity. Complement Ther Med. 2010;18(6):260–4. 40. Fang CY, Reibel DK, Longacre ML, Rosenzweig S, Campbell DE, Douglas SD. Enhanced psychosocial well-being following participation in a mindfulness-based stress reduction program is associated with increased natural killer cell activity. J Altern Complement Med. 2010;16(5):531–8. 41. Jensen CG, Vangkilde S, Frokjaer V, Hasselbalch SG. Mindfulness training affects attention–or is it attentional effort? J Exp Psychol Gen. 2012;141(1):106–23. 42. Bower JE, Irwin MR. Mind-body therapies and control of inflammatory biology: a descriptive review. Brain Behav Immun. 2016;51:1–11. 43. Black DS, Slavich GM. Mindfulness meditation and the immune system: a systematic review of randomized controlled trials. Ann N Y Acad Sci. 2016;1373(1):13–24. 44. Pascoe MC, Thompson DR, Jenkins ZM, Ski CF. Mindfulness mediates the physiological markers of stress: systematic review and meta-analysis. J Psychiatr Res. 2017;95:156–78. 45. Lappalainen R, Sairanen E, Järvelä E, et al. The effectiveness and applicability of different lifestyle interventions for enhancing wellbeing: the study design for a randomized controlled trial for persons with metabolic syndrome risk factors and psychological distress. BMC Public Health. 2014;14:310. https://doi.org/10. 1186/1471-2458-14-310. 46. Goldberg D. The detection of psychiatric illness by questionnaire. Maudsley monograph no. 21. Oxford: Oxford University Press; 1972. 47. Myers GL, Rifai N, Tracy RP, Roberts WL, Alexander RW, Biasucci LM, et al. CDC., AHA. CDC/AHA workshop on markers of inflammation and cardiovascular disease: application to clinical and public health practice: report from the laboratory science discussion group. Circulation. 2004;110(25):e545–9. 48. Järvelä-Reijonen E, Karhunen L, Sairanen E, et al. The effects of acceptance and commitment therapy on eating behavior and diet delivered through face-to-face contact and a mobile app: a randomized controlled trial. Int J Behav Nutr Phys Act. 2018;15:22. https:// doi.org/10.1186/s12966-018-0654-8. Int.J. Behav. Med. 49. Ahtinen A, Mattila E, Välkkynen P, et al. Mobile mental wellness training for stress management: feasibility and design implications based on a one-month field study. JMIR Mhealth Uhealth. 2013;1(2):e11. https://doi.org/10.2196/mhealth.2596. 50. Mattila E, Lappalainen R, Välkkynen P, et al. Usage and dose response of a mobile acceptance and commitment therapy app: secondary analysis of the intervention arm of a randomized controlled trial. JMIR Mhealth Uhealth. 2016;4:e90. https://doi.org/10.2196/ mhealth.5241. 51. Saltevo J, Vanhala M, Kautiainen H, Kumpusalo E, Laakso M. Association of C-reactive protein, interleukin-1 receptor antagonist and adiponectin with the metabolic syndrome. Mediators Inflamm. 2007;2007:93573. https://doi.org/10.1155/2007/93573. 52. Sollberger S, Ehlert U. How to use and interpret hormone ratios. Psychoneuroendocrinology. 2016;63:385–97. 53. Bond FW, Hayes SC, Baer RA, Carpenter KM, Guenole N, Orcutt HK, et al. Preliminary psychometric properties of the Acceptance and Action Questionnaire-II: a revised measure of psychological inflexibility and experiential avoidance. Behav Ther. 2011;42(4): 676–88. 54. Lillis J, Hayes SC. Measuring avoidance and inflexibility in weight related problems. Int J Behav Consult Ther. 2008;4(4):348–54. 55. Holi MM, Marttunen M, Aalberg V. Comparison of the GHQ-36, the GHQ-12 and the SCL-90 as psychiatric screening instruments in the Finnish population. Nord J Psychiatry. 2003;57(3):233–8. 56. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav. 1983;24(4):385–96. 57. Lee EH. Review of the psychometric evidence of the perceived stress scale. Asian Nurs Res (Korean Soc Nurs Sci). 2012;6(4): 121–7. 58. Beck AT, Steer RA, Ball R, Ranieri W. Comparison of Beck Depression Inventories -IA and -II in psychiatric outpatients. J Pers Assess. 1996;67(3):588–97. 59. Leppälä J, Lagström H, Kaljonen A, Laitinen K. Construction and evaluation of a self-contained index for assessment of diet quality. Scand J Public Health. 2010;38(8):794–802. 60. Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA. The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Ambulatory Care Quality Improvement Project (ACQUIP). Alcohol Use Disorders Identification Test. Arch Intern Med. 1998;158(16):1789–95. 61. Helldán A, Raulio S, Kosola M, Tapanainen H, Ovaskainen M, Virtanen S. Finravinto 2012-tutkimus—the national FINDIET 2012 Survey. Helsinki: THL Report. 2013;16. 62. Järvelä-Reijonen E, Karhunen L, Sairanen E, et al. High perceived stress is associated with unfavorable eating behavior in overweight and obese Finns of working age. Appetite. 2016;103:249–58. 63. Kujala UM, Kaprio J, Sarna S, Koskenvuo M. Relationship of leisure-time physical activity and mortality: the Finnish twin cohort. JAMA. 1998;279(6):440–4. 64. Waller K, Kaprio J, Kujala UM. Associations between long-term physical activity, waist circumference and weight gain: a 30-year longitudinal twin study. Int J Obes. 2008;32(2):353–61. 65. Field A. Discovering statistics using IBM SPSS statistics: and sex and drugs and rock ‘n’roll. 4th ed. Los Angeles; London: Sage cop.; 2013. 66. Montanelli RG, Humphreys LG. Latent roots of random data correlation matrices with squared multiple correlations on the diagonal: a Monte Carlo study. Psychometrika. 1976;41(3):341–8. 67. Lance CE, Butts MM, Michels LC. The sources of four commonly reported cutoff criteria: what did they really say? Organ Res Methods. 2006;9(2):202–20. 68. Jolliffe IT. Principal component analysis. 2nd ed. New York: Springer; 2002. 69. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO III, Criqui M, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107(3):499–511. 70. Sairanen E, Tolvanen A, Karhunen L, Kolehmainen M, JärveläReijonen E, Lindroos S, et al. Psychological flexibility mediates change in intuitive eating regulation in acceptance and commitment therapy interventions. Public Health Nutr. 2017;20:1681–91. 71. Heckenberg RA, Eddy P, Kent S, Wright BJ. Do workplace-based mindfulness meditation programs improve physiological indices of stress? A systematic review and meta-analysis. J Psychosom Res. 2018;114:62–71. 72. Ridker PM, Cook N. Clinical usefulness of very high and very low levels of C-reactive protein across the full range of Framingham risk scores. Circulation. 2004;109(16):1955–9. 73. Malarkey WB, Jarjoura D, Klatt M. Workplace based mindfulness practice and inflammation: a randomized trial. Brain Behav Immun. 2013;27(1):145–54. 74. Daubenmier J, Moran PJ, Kristeller J, Acree M, Bacchetti P, Kemeny ME, et al. Effects of a mindfulness-based weight loss intervention in adults with obesity: a randomized clinical trial. Obesity (Silver Spring). 2016;24(4):794–804. 75. Creswell JD, Irwin MR, Burklund LJ, Lieberman MD, Arevalo JMG, Ma J, et al. Mindfulness-based stress reduction training reduces loneliness and pro-inflammatory gene expression in older adults: a small randomized controlled trial. Brain Behav Immun. 2012;26(7):1095–101. 76. Uusitupa M, Schwab U. Diet, inflammation and prediabetes-impact of quality of diet. Can J Diabetes. 2013;37(5):327–31. 77. Bjermo H, Iggman D, Kullberg J, Dahlman I, Johansson L, Persson L, et al. Effects of n-6 PUFAs compared with SFAs on liver fat, lipoproteins, and inflammation in abdominal obesity: a randomized controlled trial. Am J Clin Nutr. 2012;95(5):1003–12. 78. Uusitupa M, Hermansen K, Savolainen MJ, Schwab U, Kolehmainen M, Brader L, et al. Effects of an isocaloric healthy Nordic diet on insulin sensitivity, lipid profile and inflammation markers in metabolic syndrome –a randomized study (SYSDIET). J Intern Med. 2013;274(1):52–66. 79. Chacko SA, Yeh GY, Davis RB, Wee CC. A mindfulness-based intervention to control weight after bariatric surgery: preliminary results from a randomized controlled pilot trial. Complement Ther Med. 2016;28:13–21. 80. Daubenmier J, Kristeller J, Hecht FM, et al. Mindfulness intervention for stress eating to reduce cortisol and abdominal fat among overweight and obese women: an exploratory randomized controlled study. J Obes. 2011;2011:651936. https://doi.org/10.1155/ 2011/651936. 81. Basso JC, McHale A, Ende V, Oberlin DJ, Suzuki WA. Brief, daily meditation enhances attention, memory, mood, and emotional regulation in non-experienced meditators. Behav Brain Res. 2019;356: 208–20. 82. Roeser RW, Schonert-Reichl KA, Jha A, Cullen M, Wallace L, Wilensky R, et al. Mindfulness training and reductions in teacher stress and burnout: results from two randomized, waitlist-control field trials. J Educ Psychol. 2013;105(3):787–804. 83. Carlson LE, Speca M, Patel KD, Goodey E. Mindfulness-based stress reduction in relation to quality of life, mood, symptoms of stress and levels of cortisol, dehydroepiandrosterone sulfate (DHEAS) and melatonin in breast and prostate cancer outpatients. Psychoneuroendocrinology. 2004;29(4):448–74. 84. Kroboth PD, Salek FS, Pittenger AL, Fabian TJ, Frye RF. DHEA and DHEA-S: a review. J Clin Pharmacol. 1999;39(4):327–48. 85. Edwards S, Clow A, Evans P, Hucklebridge F. Exploration of the awakening cortisol response in relation to diurnal cortisol secretory activity. Life Sci. 2001;68(18):2093–103. Int.J. Behav. Med. 86. Golden SH, Wand GS, Malhotra S, Kamel I, Horton K. Reliability of hypothalamic-pituitary-adrenal axis assessment methods for use in population-based studies. Eur J Epidemiol. 2011;26(7):511–25. 87. Wolgast M. What does the Acceptance and Action Questionnaire (AAQ-II) really measure? Behav Ther. 2014;45(6):831–9. 88. Rochefort C, Baldwin AS, Chmielewski M. Experiential avoidance: an examination of the construct validity of the AAQ-II and MEAQ. Behav Ther. 2018;49(3):435–49. 89. O'Connor MF, Bower JE, Cho HJ, et al. To assess, to control, to exclude: effects of biobehavioral factors on circulating inflammatory markers. Brain Behav Immun. 2009;23(7):887–97. 90. MacIntyre UE, Wenhold FAM. Measuring dietary intake. In: Lanham-New SA, Hill TR, Gallagher AM, Vorster HH, editors. Introduction to human nutrition. 3rd ed. Hoboken: Wiley Blackwell; 2020. p. 13–55. 91. Hofmann SG, Hayes SC. The future of intervention science: process-based therapy. Clin Psychol Sci. 2018;7:1–14. https://doi. org/10.1177/2167702618772296. 92. Tomiyama AJ. Stress and obesity. Annu Rev Psychol. 2019;70:5.1– 5.16. https://doi.org/10.1146/annurev-psych-010418-102936. Publisher’sNoteSpringer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Affiliations Elina Järvelä-Reijonen 1 &Sampsa Puttonen 2 &Leila Karhunen 1,3 &Essi Sairanen 4,5 &Jaana Laitinen 2 & Mikko Kolehmainen 6 &Jussi Pihlajamäki 1,3 &Urho M Kujala 7 &Riitta Korpela 8 &Miikka Ermes 1,9 &Raimo Lappalainen 4 & Marjukka Kolehmainen 1,3 1 Institute of Public Health and Clinical Nutrition, Clinical Nutrition, University of Eastern Finland, P.O. Box 1627, FI70211 Kuopio, Finland 2 Finnish Institute of Occupational Health, P.O. Box 40, FI00251 Helsinki, Finland 3 Institute of Clinical Medicine and Clinical Nutrition, Kuopio University Hospital, P.O. Box 100, FI-70029 KYS Kuopio, Finland 4 Department of Psychology, University of Jyväskylä, P.O. Box 35, FI40014 Jyväskylä, Finland 5 Department of Social and Psychological Studies, Karlstad University, SE-651 88 Karlstad, Sweden 6 Department of Environmental and Biological Sciences, University of Eastern Finland, P.O. Box 1627, FI-70211 Kuopio, Finland 7 Faculty of Sport and Health Sciences, University of Jyväskylä, P.O. Box 35, FI-40014 Jyväskylä, Finland 8 Medical Faculty, Pharmacology, Medical Nutrition Physiology and Human Microbe Research Program, University of Helsinki, P.O. Box 63, FI-00014 Helsinki, Finland 9 VTT Technical Research Centre of Finland, P.O. Box 1300, FI33101 Tampere, Finland Int.J. Behav. Med.