Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jan 1.
Published in final edited form as: Behav Ther. 2023 May 26;55(1):26–41. doi: 10.1016/j.beth.2023.05.007

The Relationship Between Psychological Inflexibility and Well-Being in Adults: A Meta-Analysis of the Acceptance and Action Questionnaire (AAQ)

Clarissa W Ong a,b, Abigail L Barthel a, Stefan G Hofmann c
PMCID: PMC10787153  NIHMSID: NIHMS1905483  PMID: 38216235

Abstract

Psychological inflexibility is defined as the rigid responding to stimuli (e.g., unpleasant thoughts and feelings) that interferes with well-being and valued actions. It is the treatment target in acceptance and commitment therapy (ACT). Despite the centrality of the link between inflexibility and well-being to ACT theory, an empirical review clarifying the nature of this relationship has not been conducted. As such, the current meta-analysis examined the meta-correlation between psychological inflexibility, measured by the Acceptance and Action Questionnaire (AAQ) and its variants, and well-being. A systematic review yielded 151 studies, including 25 versions of the AAQ and 43 well-being measures. Consistent with ACT theory, higher psychological inflexibility was associated with worse well-being (r = −.47, 95% CI[−.49, −.45]). In addition, sample diagnosis, type of AAQ, and type of well-being measure significantly moderated this relationship. Overall, our findings support the hypothesized link between psychological inflexibility and worse well-being. Limitations include reliance on cross-sectional data, precluding causal interpretation.

Keywords: psychological inflexibility, acceptance and action questionnaire, well-being, valued action, life satisfaction, meta-analysis

Psychological inflexibility is defined as rigid responding to inner experiences (e.g., thoughts, feelings) in ways that interfere with well-being and pursuit of valued actions (Hayes et al., 2011; Kashdan et al., 2020). It is the key target of change in acceptance and commitment therapy (ACT), a cognitive-behavioral treatment focused on reducing suffering and enhancing well-being through decreasing psychological inflexibility and increasing psychological flexibility (Hayes et al., 2011). In fact, from an ACT perspective, psychological inflexibility is a primary cause of suffering, such that individuals who rigidly avoid difficult thoughts or emotions and important or meaningful activities are hypothesized to experience greater distress and impairment, compromising well-being and adaptive coping in daily life. Psychological inflexibility tends to manifest topographically as struggles with emotion regulation, avoidance behaviors, and maladaptive cognitions, and it has been consistently associated with various forms of psychopathology (Levin et al., 2014).

Research using different study designs supports the role of psychological inflexibility as a process of change related to meaningful outcomes. First, reductions in psychological inflexibility across evidence-based treatments (EBTs) have been associated with greater well-being over the course of treatment (Arch et al., 2012; Arch & Craske, 2008; Gloster, Klotsche, et al., 2017; Hayes et al., 2019). Second, meta-analytic and literature reviews have shown that ACT and other psychological flexibility-promoting interventions are effective for improving outcomes in various domains, including chronic pain (e.g., Gentili et al., 2019), cancer (e.g., Zhao et al., 2021), diabetes (e.g., Kioskli et al., 2019), and obesity (e.g., Ciarrochi et al., 2014). Third, psychological inflexibility itself is associated with maladaptation in varied contexts, including cancer (e.g., Brown et al., 2020), general stigma and prejudice (e.g., Krafft et al., 2017), diabetes (e.g., Kılıç et al., 2022), and weight problems (e.g., Finger et al., 2020).

Conversely, psychological flexibility has been linked to healthy coping and well-being in medical populations (McCracken & Morley, 2014) and serves a protective function in the general population (Gloster, Meyer, et al., 2017). Furthermore, research on human flourishing supports well-being as a malleable target linked to psychological flexibility and values (Dahl, Wilson-Mendenhall, & Davidson, 2020). Such research suggests that focusing on flourishing and well-being is distinct from symptom or problem-focused psychotherapies (Jankowski et al., 2020) and has positive health benefits (VanderWeele et al., 2019), underscoring the utility of understanding the relationship between psychological inflexibility and well-being, not merely that between psychological inflexibility and symptoms.

Despite its relevance to a range of psychological problems and treatments (Hayes et al., 2022), psychological inflexibility has been notoriously difficult to operationalize and measure because it is functionally rather than topographically defined, requiring some insight on the respondent’s part to discern the function of their own behavior. Nonetheless, numerous validated measures can be found in the literature, with the most popular measure of psychological inflexibility being the Acceptance and Action Questionnaire—II (AAQ-II; Bond et al., 2011; Cherry et al., 2021). The AAQ has been updated several times, and its current form, the AAQ-II, is a seven-item revised version of its original iteration. There are also at least 19 domain-specific variants based on the AAQ, which typically assess psychological inflexibility in the context of a specific problem (e.g., chronic pain, trichotillomania; see Ong et al., 2019 for a review). The AAQ and its variants have been used in healthy, subclinical, medical, and psychiatric samples across numerous languages and translations (Ong, Lee, et al., 2019). Psychometric studies on the AAQ-II have demonstrated test-retest reliability, incremental validity, convergent validity, and measurement invariance across demographically diverse samples (Bond et al., 2011; Correa-Fernández et al., 2020; Gloster et al., 2011), showing the AAQ-II to be a well-rounded measure of psychological inflexibility. Furthermore, these properties appear to hold for its translations (e.g., Eisenbeck & Szabó-Bartha, 2018; Paladines-Costa et al., 2021). In addition, a systematic review found that the AAQ was the most consistently replicated mediator of treatment outcomes, supporting its empirical status as a process of change measure (Hayes et al., 2022).

At the same time, the AAQ-II has been criticized for its poor discriminant validity with respect to general distress (e.g., Rochefort et al., 2018; Tyndall et al., 2019; Wolgast, 2014), lack of measurement invariance in clinical vs. non-clinical samples (Ong, Pierce, et al., 2019), and overemphasis on the experiential avoidance facet of psychological inflexibility (Francis et al., 2016). Accordingly, other validated measures of psychological inflexibility that aim to address some of these limitations have been developed. They include the Comprehensive assessment of ACT processes (CompACT; Francis et al., 2016), a 23-item self-report measure that assesses all six dimensions of psychological flexibility, and the Multidimensional Psychological Flexibility Inventory (MPFI; Rolffs et al., 2016), a 60-item self-report measure that assesses the 12 dimensions of psychological flexibility and inflexibility. In addition, some researchers have critiqued the empirical status of psychological flexibility as a process of change in ACT in the first place, underscoring the need for stronger measures and more robust study designs (Arch et al., 2022). Nonetheless, given the ubiquity of the AAQ and the vast amount of data associated with it relative to more recently developed measures, the current review focuses on the AAQ and its variants as primary measures of psychological inflexibility.

Well-being is broadly defined and may include such constructs as life satisfaction, quality of life, vitality, and health or psychosocial health. Like the AAQ, well-being measures are sometimes domain-specific, which means they assess well-being as it pertains to a specific diagnosis (e.g., ability to do activities of daily living for chronic pain) as opposed to global well-being. Other well-being measures parse the different components of well-being which may include emotional, physical, social, psychological, or general health, in addition to providing a total sum score. A more ACT-consistent version of well-being may conceptualize it as a measure of valued action, meaning, fulfillment, or vitality; that is, is the individual acting in a way that is in line with their values or maximizes meaningful life satisfaction?

Study Aims and Impact

With the rising popularity of ACT, research on psychological inflexibility and related concepts has exponentially increased over the past several decades, including several meta-analyses and systematic reviews on ACT processes and/or psychological inflexibility. For example, Daks and Rogge (2020) summarized the relationship between psychological (in)flexibility and family and romantic relationship functioning, whereas Garner and Golijani-Moghaddam (2021) reviewed the relationship between psychological flexibility and work-related quality of life in healthcare professionals. Both meta-analyses found that psychological flexibility was generally associated with positive outcomes (e.g., compassion satisfaction in healthcare professionals), whereas psychological inflexibility was associated with worse outcomes (e.g., parenting stress). In addition, Kashdan and Rottenberg (2010) conducted a narrative review of the construct of psychological (in)flexibility, also supporting the relationship between psychological flexibility and well-being. Although each of these reviews relate, in part, to the overall relationship between psychological inflexibility and well-being, previous reviews either did not use meta-analytic methods to quantify the magnitude of this relationship or focused on specific samples rather than a wide range of samples. To our knowledge, no study has used a meta-analysis to quantify the association between these two constructs broadly, which is striking given the key roles they play in the theory underlying ACT.

Relatedly, psychological research and interventions have shifted toward a focus on cross-cutting processes of change (Hayes et al., 2019), which highlights the specific need for a meta-analytic understanding of the relationship between a transdiagnostic process, such as psychological inflexibility, and a personally relevant outcome, like well-being. Further, ACT and other EBTs base much of their frameworks on the assumption that psychological inflexibility and well-being are associated, so greater understanding of the nature of this relationship, including moderators, is warranted to empirically validate this theory. Thus, the primary aims of this study are to: (1) examine the meta-correlation between psychological inflexibility—as measured by the AAQ and its variants—and well-being, and (2) identify moderators of this relationship, including study characteristics (e.g., sample size, mean age, sample diagnosis) and measure characteristics (e.g., type of AAQ, aspect of well-being).

Method

The meta-analytic protocol for this research was registered with the International Prospective Register of Systematic Reviews (PROSPERO) and assigned the following registration identifier: CRD42020192925. This meta-analysis followed the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Figure 1 shows a PRISMA figure for this study. Data, code, and appendices for this meta-analysis can be found at https://osf.io/n8usm/.

Figure 1.

Figure 1.

Open in a new tab

PRISMA flow diagram including searches of databases and other sources.

Literature Search

To identify eligible articles for the current meta-analysis, we entered predetermined search terms into the following online databases: PsycINFO, Web of Science, and PubMed. Depending on the settings of the database, we used the “Filter” option whenever possible to narrow results down to articles or academic journals, adult populations (at least 18 years old), and articles written or translated into English. The initial literature search was conducted in August 2021, and an updated search was done in November 2022. PubMed and PsycINFO searches were set to a “Title/Abstract” search, while Web of Science searched “All” due to inability to specify a title/abstract focus.

The following search string was entered into PsycINFO, Web of Science, and PubMed: ((“psychological flexibility”[title/abstract]) or (“psychological inflexibility”[title/abstract]) or (“acceptance and action questionnaire”[title/abstract]) or (“acceptance and action diabetes questionnaire”[title/abstract]) or (“acceptance and action epilepsy questionnaire”[title/abstract]) or (“avoidance and inflexibility scale”[title/abstract]) or (“body image psychological inflexibility scale”[title/abstract]) or (“chronic pain acceptance questionnaire”[title/abstract]) or (“experiential avoidance in caregiving questionnaire”[title/abstract]) or (“multiple sclerosis acceptance questionnaire”[title/abstract]) or (“parenting-specific psychological flexibility”[title/abstract]) or (“parental acceptance questionnaire”[title/abstract]) or (“parental psychological flexibility”[title/abstract]) or (“psychological inflexibility in pain scale”[title/abstract]) or (“sleep problem acceptance questionnaire”[title/abstract]) or (“tinnitus acceptance questionnaire”[title/abstract]) or (“voices acceptance and action scale”[title/abstract])) and ((“functioning”[title/abstract]) or (“health”[title/abstract]) or (“outcome*”[title/abstract])or (“quality of life”[title/abstract]) or (“satisfaction”[title/abstract]) or (“valu*”[title/abstract]) or (“well-being”[title/abstract]) or (“wellbeing”[title/abstract])).

We focused on studies that measured psychological inflexibility using the AAQ and its variants, because they are the most frequently used measures of psychological inflexibility and assess psychological inflexibility in a similar way (Cherry et al., 2021; Ong, Lee, et al., 2019). Well-being measures were included in our analyses if they purported or were known to assess global—as opposed to domain-specific—well-being, because we were interested in the relationship between psychological inflexibility and overall well-being, whereas domain-specific well-being tended to be more problem-focused. In addition, when delineating well-being measures, we included measures of adaptive rather than maladaptive functioning (e.g., life satisfaction rather than burnout) in the spirit of orienting to flourishing/prosperity as an outcome rather than languishing/suffering (Hofmann & Hayes, 2019; Keyes, 2005). Based on literature and theory about the areas of life functioning psychological inflexibility is expected to be associated with, we considered the following constructs part of well-being: quality of life, well-being, functioning, valued action, life satisfaction, and flourishing.

Study Selection

The first and second authors (CWO and ALB) assessed article eligibility along with trained research assistants. Articles were included if they: (1) were written in or translated into English; (2) studied human participants; (3) were published in a peer-reviewed journal; (4) used an adult sample (age 18 and older); (5) simultaneously used the AAQ or its variant and a self-report measure of well-being or a related construct; and (6) reported a correlation coefficient between psychological inflexibility and well-being. Articles were excluded if they: (1) measured psychological inflexibility using a non-AAQ instrument; (2) studied a sample that met criteria for a neurodegenerative disorder (e.g., Alzheimer’s disease, Huntington’s disease, multiple sclerosis); (3) were non-empirical (e.g., theoretical discussion); (4) were systematic reviews or meta-analyses; or (5) used a single-subject design (e.g., multiple baseline).

In the first stage of screening, article abstracts were evaluated against the eligibility criteria outlined above. Eighty percent of abstracts were screened by two reviewers as an additional measure of quality control, and any discrepancies were resolved via discussion until a consensus was reached. In the next stage, we reviewed full-text articles. Consistent with PRISMA guidelines, each full-text article was assessed by both reviewers (CWO and ALB); similar to the abstract screening, any discrepancies were resolved via discussion until a consensus was reached. Inter-rater reliability statistics were calculated to quantify agreement between reviewers on full-text screening.

Data Extraction

The first two authors (CWO and ALB) completed data extraction independently, compared data to check consistency, and resolved discrepancies through discussion as needed. Data for the following variables were extracted: (1) correlation coefficients representing the association between psychological inflexibility and well-being, (2) Cronbach’s α (internal consistency of measures), (3) sample size, (4) sample demographic variables (i.e., age, percent female, percent identifying as White/majority ethnicity), (5) sample characteristics (e.g., diagnosis), (6) language of the measure, (7) country of study origin, and (8) mean and standard deviation statistics for each of the measures extracted in a given article. Coding for categorical variables is described in Table 1.

Table 1.

Variable Coding for Proposed Moderators

Variable Levels Examples
Sample diagnosis Psychological Anxiety, autism spectrum disorder, depression, eating disorder, mixed psychiatric diagnoses, obsessive-compulsive and related disorders, posttraumatic stress disorder, serious mental illness, substance use disorder
Physical/Medical Cancer, cardiovascular disease
Chronic condition Chronic pain, lipedema, obesity, tinnitus
Subclinical Elevated self-report symptoms without established diagnoses
None Community, unscreened, students
Type of AAQ AAQ-II (7 items)
AAQ with > 7 items
Domain-specific AAQ AAQ-Weight, Body Image-AAQ, Chronic Pain Acceptance Questionnaire, Voices Acceptance and Action Scale
Type of well-being Overall quality of life General Health Questionnaire (12 items), Short Form Health Survey (36 items), Short Warwick Edinburgh Mental Well-Being Scale, WHOQOL
Life satisfaction/flourishing Meaning in Life Questionnaire Presence of Meaning subscale, Mental Health Continuum-Short Form, Psychological Wellbeing Scale, Satisfaction with Life Scale, Quality of Life Enjoyment and Satisfaction Questionnaire
Valued action Chronic Pain Values Inventory-Success, Engagement in Life Scale, Valuing Questionnaire-Progress
Open in a new tab

Note. AAQ = Acceptance and Action Questionnaire.

If studies reported correlations from multiple independent samples within the same paper, we extracted all data. If studies had overlapping samples, we only extracted data from the largest sample to maximize representativeness and avoid “double counting.” For AAQ-related measures of psychological inflexibility, subscale scores reported without total scores (e.g., experiential avoidance subscale) were excluded. For well-being outcomes, we included total scores and certain subscale scores that better represented aspects of well-being hypothesized to be related to psychological inflexibility (e.g., mental health subscale of a quality-of-life measure but not physical health subscale). Finally, we excluded studies that only measured stress or perceived stress as an outcome as they assess maladaptive rather than adaptive functioning.

Study Quality Assessment

The Appraisal tool for Cross-Sectional Studies (AXIS; Downes et al., 2016) was used to rate the quality of each study included in our review. The AXIS is a tool that was developed to evaluate overall risk of bias in observational cross-sectional studies specifically. The AXIS contains 20 items that assess various aspects of study design and reporting quality, including clarity of study aims, sample size justification, non-response bias, and discussion of study limitations. Criteria for each item are detailed in the original study (see Downes et al., 2016).

Each item is coded as “yes,” “no,” or “don’t know.” To quantify study quality ratings, “yes” counted as 1, “no” and “don’t know” were considered 0, except for items 13 (“Does the response rate raise concerns about non-response bias?”) and 19 (“Were there any funding sources or conflicts of interest that may affect the authors’ interpretation of the results?”), which were reverse-coded. Item scores were summed to create an overall quality rating variable (possible range = 0 to 20). CWO and ALB first practiced rating 10 articles independently and discussed any discrepancies in scores to clarify application of the rating rubric. The remaining articles were independently scored by each rater, and uncertainties in scoring were resolved through discussion between the raters.

Statistical Analyses

Interrater reliability.

Interrater reliability analyses were conducted using AgreeStat360 (agreestat.com), a cloud-based inter-rater reliability program, based on Gwet (2014). Traditionally, inter-rater reliability coefficient interpretation is based on the Altman’s Kappa Benchmark Scale (Altman, 1991), with the following benchmarks: <0.2 (poor), 0.21–0.40 (fair), 0.41–0.60 (moderate), 0.61–0.80 (good), and 0.81–1.0 (very good). However, to account for often misleading interpretations of traditional kappa coefficients for inter-rater reliability (i.e., assuming that coefficients directly translate to magnitude of agreement irrespective of study sample sizes instead of degree of “true” certainty alongside unknowns; Gwet, 2014), we adopted the benchmarking procedure outlined by Gwet (2014). This procedure computes the probability that a given agreement coefficient falls into a range of certainty values between <0.20–1.0, based on the Altman’s Kappa Benchmark Scale (Altman, 1991). For example, if the probability of a coefficient falling within this range is small, we would interpret the likelihood that “true” agreement at a given value is small and vice versa. Cumulative membership probabilities were then computed for each kappa coefficient using procedures outlined by Gwet (2014) and compared to a cumulative probability cut-off point of 0.95 or higher (i.e., coefficients with a probability of 0.95 or higher are likely to fall within that range or benchmark values as a value of “true” certainty). Thus, this procedure allows for interpretation of the coefficient and its strength while accounting for probability. We were primarily concerned with achieving “good” or “very good” agreement values and probabilities for Kappa and AC1.

Meta-analyses.

Meta-correlation and meta-regression analyses were conducted using the metafor package (Viechtbauer, 2010) in R (R Core Team, 2022). For the meta-correlation analysis, in the first step, effect size estimates and their corresponding sample variances were calculated based on raw correlation coefficients and sample sizes. Unbiased estimates of the correlation coefficient were determined by correcting negative bias in the raw correlation coefficient (Olkin & Pratt, 1958), and approximately unbiased estimates of the sampling variances were obtained. In the second step, a meta-analytic model was fitted using the unbiased estimates calculated in step one to obtain a meta-correlation coefficient. Given the differences in sample characteristics and methodology across studies, assumed heterogeneity was estimated with the Hunter-Schmidt estimator (Field, 2001, 2005).

For the mixed-effects meta-regression, proposed moderators were added to the meta-analytic model to test whether they explained heterogeneity. The Knapp and Hartung adjustment was used to account for uncertainty in the estimate of heterogeneity among true effects, based on a non-normal reference distribution (Knapp & Hartung, 2003). Moderator effects were tested in two separate meta-regression models. The first used predictors representing study characteristics: year of publication, mean age, sample size, percent female, sample diagnosis (psychological, physical/medical, chronic, subclinical, nonclinical), and study quality. The second was based on measure characteristics: type of AAQ (AAQ-II, old AAQ, domain-specific AAQ) and aspect of well-being (quality of life, life satisfaction/flourishing, valued action). Table 1 describes how moderator variables were coded with examples. The reason for using two meta-regression models was to reduce listwise deletion of studies to maintain power and representativeness of the sample. However, given risk of Type I error, we used the conservative Bonferroni correction (p = .05/2 = .025) to interpret significance.

Publication bias.

Publication bias was assessed using a funnel plot and regression test for funnel plot asymmetry (Egger et al., 1997; Sterne et al., 2001). A funnel plot provides helpful visual representation for detecting heterogeneity and certain types of publication bias, whereas the regression test permits statistical evaluation of the relationship between observed outcomes and a specific predictor that is hypothesized to be related to the observed outcomes when there is publication bias (e.g., standard error, sampling variance, sample size). Standard error was used as the predictor in our regression (Sterne & Egger, 2005).

Results

Interrater Reliability

Percent agreement between the first two authors’ coding was 92.3% in the article screening phase (SE = 0.010; 95% CI = 0.903, 0.944; p < 0.01). Appendix A presents tables with the traditional kappa benchmarking values (Altman, 1991), qualitative descriptions based on cumulative membership probability, and five agreement coefficients of interrater reliability including: Kappa, AC1, Scott’s Pi, Krippendorff, and Brennan-Prediger. Based on the Altman procedure, agreement based on Kappa was “good.”

Study Characteristics

The 151 studies included in this meta-analysis represented 70,738 participants, at least 35 countries, 22 languages, 25 versions of the AAQ, 43 different well-being scales, and 262 correlation analyses (details are presented in Tables 2 and 3). A range of samples were represented in the current review, including adults with chronic pain, women with lipedema, undergraduate students, people seeking psychological treatment, and customer service workers. The mean quality rating out of a possible total of 20 was 13.2 (SD = 1.7, range = 9 to 17). Detailed descriptions of each study are presented in Appendix B and references for articles included in the meta-analysis are in Appendix C.

Table 2.

Measures of Psychological Inflexibility and Well-Being Included in Meta-Analysis

Psychological Inflexibility Well-Being
Acronym Measure Acronym Measure
6-PAQ Parental Acceptance Questionnaire BIEPS-A Psychological Wellbeing Scale for Adults (Translated)
AADQ-6 Acceptance and Action Diabetes Questionnaire Bull’s Eye
AAQ-CVD Acceptance and Action Questionnaire-Cardiovascular Disease CAQ-8 Committed Action Questionnaire (8 items)
AAQ-9 Acceptance and Action Questionnaire (9 items) CAQ-18 Committed Action Questionnaire (18 items)
AAQ-10 Acceptance and Action Questionnaire (10 items) CompACT-18-Valued Action Comprehensive Assessment of Acceptance and Commitment Therapy processes (18 items)
AAQ-16 Acceptance and Action Questionnaire (16 items) CompACT-Valued Action Comprehensive Assessment of Acceptance and Commitment Therapy processes (23 items)
AAQ-H Acceptance and Action Questionnaire-Hoarding CPVI-Success Chronic Pain Values Inventory
AAQ-II Acceptance and Action Questionnaire-II ELS Engaged Living Scale
AAQ-II-6 Acceptance and Action Questionnaire-II (6 items) EQ-5D EuroQol (5 Dimensions)
AAQ-US Acceptance and Action Questionnaire-University Students EUROHIS-QOL -
AAQ-W Acceptance and Action Questionnaire-Weight Related Difficulties FACT-G Functional Assessment of Cancer Therapy - General
AAQ-W-R Acceptance and Action Questionnaire for Weight-Related Difficulties – Revised FS The Flourishing Scale
BI-AAQ Body Image Acceptance and Action Questionnaire GB-CORE Clinical Outcomes in routine Evaluation-General Population
CPAQ-8 Chronic Pain Acceptance Questionnaire (8 items) GHQ-8 General Health Questionnaire (8 items)
CPAQ-20 Chronic Pain Acceptance Questionnaire (20 items) GHQ-12 General Health Questionnaire (12 items)
CPAQ-34 Chronic Pain Acceptance Questionnaire (34 items) GWBS General Wellbeing Schedule
CTAQ Chronic Tinnitus Acceptance Questionnaire ICECAP-O ICEpop CAPability measure for Older people
IBSAAQ Irritable Bowel Syndrome Acceptance and Action Questionnaire IWB Index of Wellbeing
PIPS Psychological Inflexibility in Pain Scale LET Life Engagement Test
PIPS-II Psychological Inflexibility in Pain Scale-II Life Satisfaction Item
SA-AAQ Social Anxiety Acceptance and Action Questionnaire LSQ Life Satisfaction Questionnaire
TAQ Tinnitus Acceptance Questionnaire MHC-SF Mental Health Continuum-Short Form
VAAS Voices Acceptance and Action Scale MLM Meaningful Living Measure
WAAQ Work Related Acceptance and Action Questionnaire MLQ-P Meaning in Life Questionnaire-Presence of Meaning
WAAQ-modified Work Related Acceptance and Action Questionnaire-modified MPFI-Committed Action, Values subscales Multifaceted Psychological Flexibility Inventory
PERMA Positive Emotion, Engagement, Positive Relationships, Meaning, Accomplishments/Achievements Profiler
Personal Strivings-Autonomous Personal Strivings Measure
Personal Strivings-Progress Personal Strivings Measure
PROMIS-G-Mental Patient Reported Outcome Measurement Information System-Global Health Scale
PWB-Autonomy, Personal Growth, Purpose in Life, and Positive Relations subscales Psychological Wellbeing Scale
Q-LES-Q-General Activities and Subjective Feelings subscales Quality of Life Enjoyment and Satisfaction Questionnaire
QOLI Quality of Life Inventory
QOLQ Quality of Life Questionnaire
QOL Single Item Quality of Life (single item)
Scales of Social Wellbeing
SF-8-Mental Component Summary Short Form Health Survey (8 items)
SF-12-Mental Component Summary Short Form Health Survey (12 items)
SF-36-Emotional Functioning Short Form Health Survey (36 items)
SF-36-Mental Health Short Form Health Survey (36 items)
SF-36-Role Emotional Short Form Health Survey (36 items)
SF-36-Social Functioning Short Form Health Survey (36 items)
SF-36-Vitality Short Form Health Survey (36 items)
SWEMWBS Short Form-Warwick-Edinburgh Mental Wellbeing Scale
SWLS Satisfaction with Life Scale
VLQ-Consistency Valued Living Questionnaire
VQ-Progress Valuing Questionnaire
VQ-Progress (3-item version) Valuing Questionnaire
VT Values Tracker
WHO-5 World Health Organization-Five Wellbeing Index
WHOQOL-Psychological and Social World Health Organization Quality of Life Questionnaire
WHOQOL-BREF Abbreviated World Health Organization Quality of Life Questionnaire
WHOQOL-BREF-Psychological Abbreviated World Health Organization Quality of Life Questionnaire
WHOQOL-BREF-Social Abbreviated World Health Organization Quality of Life Questionnaire
Open in a new tab

Table 3.

Breakdown of Studies by Country, Language, Psychological Inflexibility Measure, and Well-Being Measure

Count (%)
k = 151 (number of studies) N = 262 (number of correlation analyses)
Country1 Language PI Measure2 Well-Being Measure1
Argentina 1 (0.7%) Chinese 6 (4.0%) AAQ-9 4 (1.5%) CAQ-8 13 (5.0%)
Australia 8 (5.3%) Croatian 1 (0.7%) AAQ-10 8 (3.1%) CAQ-18 7 (2.7%)
Austria 1 (0.7%) Danish 1 (0.7%) AAQ-16 10 (3.8%) CompACT-Valued Action 4 (1.5%)
Belgium 3 (2%) Dutch 5 (3.3%) AAQ-H 2 (0.8%) CPVI-Success 4 (1.5%)
China 5 (3.3%) English 73 (48.3%) AAQ-II 155 (59.2%) ELS 3 (1.1%)
Colombia 3 (2%) Farsi 1 (0.7%) AAQ-US 2 (0.8%) EQ-5D 7 (2.7%)
Croatia 1 (0.7%) Finnish 4 (2.6%) AAQ-W 3 (1.1%) EUROHIS-QOL 3 (1.1%)
Cyprus 2 (1.3%) German 5 (3.3%) CPAQ-8 9 (3.4%) GB-CORE 2 (0.8%)
Denmark 2 (1.3%) Greek 2 (1.3%) CPAQ-20 37 (14.1%) GHQ-12 29 (11.1%)
Ecuador 1 (0.7%) Hungarian 1 (0.7%) CTAQ 2 (0.8%) MHC-SF 3 (1.1%)
England 10 (6.6%) Italian 9 (6%) PIPS 6 (2.3%) MLQ-P 4 (1.5%)
Finland 4 (2.6%) Japanese 1 (0.7%) SA-AAQ 2 (0.8%) Personal Strivings-Autonomous 2 (0.8%)
Germany 4 (2.6%) Korean 2 (1.3%) VAAS 2 (0.8%) Personal Strivings-Progress 2 (0.8%)
Hong Kong 3 (2%) Mandarin 3 (2%) WAAQ 8 (3.1%) PWB 6 (2.3%)
Hungary 1 (0.7%) Norwegian 3 (2%) WAAQ-modified 2 (0.8%) QOLI 4 (1.5%)
India 1 (0.7%) Persian 1 (0.7%) SF-8-Mental Component Summary 2 (0.8%)
Iran 2 (1.3%) Polish 1 (0.7%) SF-12-Mental Component Summary 5 (1.9%)
Italy 9 (6%) Portuguese 4 (2.6%) SF-36-Emotional Functioning 2 (0.8%)
Japan 1 (0.7%) Serbian 1 (0.7%) SF-36-Mental Health 20 (7.6%)
Latin America 1 (0.7%) Spanish 16 (10.6%) SF-36-Role Emotional 4 (1.5%)
Norway 3 (2%) Swedish 9 (6%) SF-36-Social Functioning 11 (4.2%)
Poland 1 (0.7%) Turkish 2 (1.3%) SF-36-Vitality 8 (3.1%)
Portugal 4 (2.6%) SWEMWBS 2 (0.8%)
Serbia 1 (0.7%) SWLS 29 (11.1%)
Singapore 1 (0.7%) VLQ 7 (2.7%)
South Korea 2 (1.3%) VQ-Progress 22 (8.4%)
Spain 9 (6%) WHO-5 5 (1.9%)
Sweden 9 (6%) WHOQOL-Psychological 2 (0.8%)
Taiwan 1 (0.7%) WHOQOL-BREF 5 (1.9%)
The Netherlands 1 (0.7%) WHOQOL-BREF-Psychological 5 (1.9%)
Turkey 2 (1.3%) WHOQOL-BREF-Social 4 (1.5%)
UK 10 (6.6%)
USA 37 (24.5%)
International 6 (4.0%)
Open in a new tab
1

One study did not indicate country in which data were collected.

2

Measures that were only used once are not included in the table. For psychological inflexibility, these were the 6-PAQ, AADQ-6, AAQ-CVD, AAQ-II-6, AAQ-W-R, BI-AAQ, CPAQ-34, IBSAAQ, PIPS-II, TAQ. For well-being, they were the BIEPS-A, Bull’s eye, CompACT-18-VA, CPVI, FACT-G, FS, GHQ-8, GWBS, ICECAP-O, IWB, LET, Life Satisfaction single item, LSQ, MLM, MPFI-Committed Action, MPFI-Values, PERMA, PROMIS-G-Mental, PWB-Autonomy, PWB-Personal Growth, PWB-Purpose in Life, PWB-Positive Relations, Q-LES-Q, Q-LES-Q-General Activities, Q-LES-Q-Subjective Feelings, QOLQ, Scales of Social Well-Being, QOL single item, VLQ-Consistency, VQ-Progress (3-item version), VT, WHOQOL, and WHOQOL-Social.

Note. Please see Table 2 for a complete list of measures included in this review and their corresponding acronyms.

Meta-Correlation

Across 262 correlational analyses, the overall effect size of the correlation between psychological inflexibility and well-being was r = −.467, 95% CI[−.488, −.446], p < .0001, indicating that more psychological inflexibility was associated with worse well-being on average. The amount of heterogeneity in the model was estimated to be τ2 = 0.0269, 95% CI[0.0152, 0.0227], which was significant, Q(261) = 5408.64, p < .0001. In addition, the proportion of variability in effect size estimates across studies due to heterogeneity rather than sampling variability was I2 = 95.11%, 95% CI[91.67, 94.27]. The ratio of total variability in observed outcomes to sampling variability was H2 = 20.47, 95% CI[12.01, 17.44] (Higgins & Thompson, 2002). Both indices suggested considerable variability that support follow-up moderation analyses.

Moderator Analyses

In the first meta-regression model of study characteristics, only sample diagnosis significantly predicted variance in effect size estimates, with subclinical samples showing a higher overall correlation coefficient (or lower absolute correlation coefficient) relative to samples without diagnoses (β = 0.157, 95% CI[0.051, 0.263], p = .0038; see Table 4). In other words, samples without diagnoses generally showed a stronger relationship between psychological inflexibility and well-being relative to samples with elevated symptoms. There were no differences in the magnitude of the correlation between clinical—including psychological, physical/medical, and chronic conditions—and nonclinical samples. Year of publication, mean age, sample size, proportion of women, and study quality did not significantly moderate the relationship between psychological inflexibility and well-being (ps > .10).

Table 4.

Beta Coefficients From Meta-Regression Models Predicting Correlation Between Psychological Inflexibility and Well-Being

Predictor β 95% CI p R 2
Study Characteristics (k = 232) .016
 Sample diagnosis
  Psychological1 0.029 −0.027; 0.084 .312
  Medical1 0.044 −0.025; 0.113 .213
  Chronic condition1 0.045 −0.009; 0.099 .100
  Subclinical1 0.157 0.051; 0.263 .004
 Sample mean age −0.001 −0.003; 0.001 .181
 Year of publication −0.002 −0.007; 0.003 .429
 Percent female −0.001 −0.002; 0.001 .319
 Sample size 0.000 −0.000; 0.000 .848
 Quality rating 0.001 −0.010; 0.012 .867
 
Measure Characteristics (k = 262) .168
 Type of psychological inflexibility measure
  Domain-specific2 0.087 0.049; 0.125 <.0001
  AAQ (old version)2 0.041 −0.018; 0.100 .178
 Type of well-being measure
  Life satisfaction/flourishing3 0.046 0.002; 0.089 .042
  Valued action3 0.113 0.074; 0.152 <.0001
Open in a new tab
1

Reference level is no diagnosis.

2

Reference level is the AAQ-II.

3

Reference level is quality of life measures.

Note. Significant differences based on Bonferroni-corrected p-value are bolded.

In the second meta-regression model of measure characteristics, type of AAQ and type of well-being measure significantly predicted variance in effect size estimates (see Table 4). Specifically, relative to the AAQ-II, domain-specific AAQs were associated with a higher correlation coefficient (or lower absolute correlation coefficient), indicating a stronger relationship between general psychological inflexibility and well-being than between domain-specific psychological inflexibility and well-being (β = 0.087, 95% CI[0.049, 0.125], p < .0001). There was no significant difference between older versions of the AAQ and the AAQ-II. In addition, compared to valued-action measures, overall quality-of-life measures were more strongly associated with psychological inflexibility (β = 0.113, 95% CI[0.074, 0.152], p < .0001). There was no significant difference between quality-of-life and life satisfaction/flourishing measures based on our Bonferroni-corrected p-value; p = .030. Forest plots with studies separated by type of AAQ are presented in Appendices D, E, and F.

Publication Bias

Given that sample diagnosis, type of AAQ, and type of well-being measure were significant moderators in their respective models, they were included in the funnel plot to account for heterogeneity. Thus, the x-axis of the funnel plot shows residuals rather than observed effect sizes. The funnel plot in Figure 2 suggests there may be publication bias, though it is plausible that the observed asymmetry is due to other reasons (Sterne et al., 2011). For example, the asymmetry in Figure 2 indicates bias toward higher positive residuals, which would translate to lower absolute effect size estimates given that all observed effect sizes were negative. Consequently, any correction for the observed bias would be toward a stronger relationship between psychological inflexibility and well-being. Because publication bias is typically in the direction of larger effect sizes, it is less likely that the observed bias is due to reporting bias. Rather, it could be attributed to other factors, such as unobserved moderators, sampling variation, and methodological differences. The regression test for funnel plot asymmetry supported the interpretation of presence of asymmetry (t = 5.441, df = 252, p < .0001).

Figure 2.

Figure 2.

Open in a new tab

Funnel plot of studies included in current meta-analysis, accounting for significant moderators.

Discussion

The present meta-analysis examined the relationship between psychological inflexibility, measured by the AAQ and its variants, and well-being, variously defined as quality of life, life satisfaction/flourishing, and valued action. Based on 262 correlation coefficients reported from 151 studies across over 35 countries and 22 languages, the overall correlation between psychological inflexibility and well-being was −.47, 95% CI[−.49, −.45]. That is, psychological inflexibility and well-being were moderately and negatively associated, with well-being decreasing by .47 units for each 1-unit increase in psychological inflexibility. This result is consistent with our prediction that psychological inflexibility, generally understood as a maladaptive process (Hayes et al., 2006), would be associated with worse well-being. Furthermore, the magnitude of the correlation coefficient suggests that psychological inflexibility explains sizable variance in well-being (approximately 22%), supporting ongoing research efforts to identify effective ways to target inflexibility in psychological interventions (e.g., Fledderus et al., 2012; McCracken & Gutierrez-Martinez, 2011).

At the same time, the 78% of variance in well-being not accounted for by psychological inflexibility points to the importance of considering other factors that influence well-being when developing and delivering treatments, including other putative mediators of intervention outcomes (e.g., mindfulness, dysfunctional thoughts, anxiety sensitivity; Hayes et al., 2022) and environmental or contextual variables (e.g., adverse childhood events, social support, and subjective socioeconomic status; Kalmakis & Chandler, 2015; Präg et al., 2016; Secor et al., 2017). Having coherent intervention strategies or treatment kernels that target multiple processes of change relevant to well-being with precision may give clinicians more flexibility with respect to how they effect healthy change in their clients (Hofmann & Hayes, 2019; see e.g., Ong et al., 2022). For instance, a clinician may switch from targeting process A with kernel X to targeting process A with kernel Y or process B with kernel Z if they notice that the client’s gains in process A have plateaued, understanding that the kernel and/or process of change that works for one client may not work for another. In addition, attending to the influence of extra-individual variables on well-being may naturally lead to organization-level or community-based interventions focused on environmental factors (e.g., increasing accessibility to resources)—beyond equipping people with psychological skills—as part of a more holistic approach to enhancing well-being (Eriksson et al., 2018; Hawley & Williford, 2015).

Moderator analyses indicated that the relationship between psychological inflexibility and well-being was stronger in samples without reported symptoms (e.g., community sample) relative to subclinical samples with elevated symptoms (not formally assessed for diagnoses); there was no difference between nonclinical and clinical samples. In other words, the moderating influence of sample symptom severity may be nonlinear, wherein nonclinical and clinical samples are more similar in their presentation than a subclinical sample. One interpretation is that the AAQ and its variants do not perform as well in subclinical populations, given that many AAQ measures have been developed and validated in either community/student or clinical samples (e.g., Bond et al., 2011; McCracken et al., 2004). Another is that the AAQ and its variants tend to treat psychological inflexibility as a trait, and nonclinical and clinical samples represent more stable, predictable trajectories, making them a more appropriate audience for the AAQs. In contrast, subclinical samples may be more likely to comprise people responding to an acute or temporary stressor. As such, environmental factors—rather than an ingrained pattern of responding—may play a bigger role in determining well-being, undercutting the association between psychological inflexibility and well-being in the subclinical group.

Relative to domain-specific AAQs, general AAQs tended to estimate a stronger relationship between psychological inflexibility and well-being, which is a surprising finding, given that domain-specific AAQs have generally shown incremental validity with respect to predicting symptom severity over the general AAQ-II (see Ong et al., 2019 for a review). However, if we consider symptom severity and well-being as distinct constructs rather than two ends of a continuum of psychological health (Kashdan & Rottenberg, 2010; Keyes, 2002), it follows that they may demonstrate different relationships with psychological inflexibility. In fact, results from this meta-analysis suggest that a general AAQ may be more relevant to understanding overall well-being than domain-specific AAQs. Whether domain-specific AAQs are better at predicting symptom severity bears empirical testing in another meta-analysis.

We also found that quality-of-life measures were more strongly associated with psychological inflexibility relative to valued-action measures, with no difference between quality-of-life and life satisfaction/flourishing measures. That is, psychological inflexibility as measured by AAQs was more closely tied to general indices of quality of life than valued action, suggesting that psychological inflexibility is a broad construct better understood as a determinant of overall wellness rather than a particular facet of functioning like valued action. This finding is surprising given that valued action is embedded in the definition of psychological flexibility (Hayes et al., 2006). However, recent research supports psychological inflexibility and flexibility as independent constructs (Ciarrochi et al., 2014; Rolffs et al., 2016), so inflexibility may not be as closely related to valued action as we might expect. Furthermore, researchers have found that the AAQ-II weights the experience avoidance aspect of psychological inflexibility more heavily than its behavioral side (Francis et al., 2016). Regardless, weaker correlations do not necessarily mean that measures are less useful, because the utility of a measure depends on context and stated goals. For example, clinicians may still be more concerned with increasing valued action (e.g., person struggling with depression who reports that life feels meaningless), just as they may be interested in how domain-specific inflexibility influences a client’s presenting problem. On the other hand, an epidemiological researcher interested in examining coping strategies linked to overall life functioning may prefer the general AAQ-II.

Limitations

Results should be interpreted in light of this study’s limitations. First, we used cross-sectional data and were only able to establish a correlational relationship between psychological inflexibility and well-being. Hence, we cannot conclude that psychological inflexibility causes poorer well-being or vice versa. While theoretical models hypothesize such a causal relationship and longitudinal analyses have yielded findings consistent with this hypothesis (Hayes et al., 2022), the results of this meta-analysis only corroborate a contemporaneous association. Second, when determining our eligibility criteria, we tried to balance exclusivity and inclusivity of measures, with the goal of being sufficiently representative and diverse without being so heterogeneous that our findings would be too diluted and not applicable to any group. The heterogeneity indices of our study suggest that our sample may err on the heterogeneous side, thus, it is important to consider results from our moderation analyses in tandem with the primary meta-correlation. Relatedly, despite including 151 studies in our review, we excluded measures of psychological inflexibility not based on the AAQ, including the MPFI (Rolffs et al., 2016), which treats psychological flexibility and inflexibility as independent constructs, and the PPFI (Kashdan et al., 2020), which has been found to have the strongest conceptual and psychometric properties among psychological flexibility measures (Cherry et al., 2021). Thus, our findings may not generalize to these measures. Third, the research team selected what counts as “well-being” based on an internal consensus, but there are differences in how various cultures conceptualize well-being (Headey et al., 2022). Consequently, our findings suffer from ethnocentric bias in terms of how we defined well-being and may not be readily applicable to cultures that formulate well-being differently. Finally, to manage the scope of this review and focus on our research questions, we chose not to assess the relationship between psychological inflexibility and symptom severity, which precluded us from testing incremental validity of the AAQs over symptom severity or discriminant validity between the two constructs (e.g., Tyndall et al., 2019; Wolgast, 2014).

Conclusion

This study represents the first meta-analytic investigation of the relationship between psychological inflexibility and well-being to our knowledge. Based on present findings, we conclude that psychological inflexibility, as measured by the AAQ and its variants, is on aggregate associated with worse well-being, with well-being defined as quality of life, life satisfaction/flourishing, or valued action. Our results provide meta-analytic evidence to support the hypothesized negative relationship between psychological inflexibility and well-being in the ACT model (Hayes et al., 2006). In addition, moderation results point to differential functions of the AAQ and its domain-specific variants, with the former serving a broader role of predicting general well-being and the latter potentially being more relevant in a clinical setting when working with specific symptom presentations. Thus, we recommend that researchers and clinicians who wish to examine psychological inflexibility and well-being consider the context in which they are working and select a measure that matches their goals and needs.

Highlights.

  • Meta-correlation between psychological inflexibility and well-being was −.47.

  • Moderators included sample diagnosis and type of AAQ (e.g., general, domain-specific).

  • Results are consistent with ACT theory.

  • Limitations include use of cross-sectional data.

Acknowledgements:

The authors thank Emily Brown, Hannah McCarthy, and Jiaxuan Tong for assisting with screening abstracts and Steven C. Hayes, Ph.D., for his conceptual input. Dr. Hofmann receives financial support by the Alexander von Humboldt Foundation (as part of the Alexander von Humboldt Professur) and the Hessische Ministerium für Wissenschaft und Kunst (as part of the LOEWE Spitzenprofessur).

Role of Funding Sources:

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest: SGH receives financial support by the Alexander von Humboldt Foundation (as part of the Alexander von Humboldt Professur), the Hessische Ministerium für Wissenschaft und Kunst (as part of the LOEWE Spitzenprofessur), NIH/NIMH ROIMH128377, NIH/NIMH U0IMH108168, Broderick Foundation/MIT. and the James S. McDonnell Foundation 21st Century Science Initiative in Understanding Human Cognition – Special Initiative. He receives compensation for his work as editor from SpringerNature. He also receives royalties and payments for his work from various publishers. All other authors declare that they have no conflicts of interest.

CRediT Author Statement

CWO: Data curation; Formal analysis; Roles/Writing - original draft; Writing - review & editing. ALB: Conceptualization; Data curation; Roles/Writing - original draft; Writing - review & editing. SGH: Conceptualization; Writing - review & editing; Supervision; Funding acquisition.

References

  1. Altman DG (1991). Practical statistics for medical research Chapman & Hall/CRC. [Google Scholar]
  2. Arch JJ, & Craske MG (2008). Acceptance and commitment therapy and cognitive behavioral therapy for anxiety disorders: Different treatments, similar mechanisms? Clinical Psychology: Science and Practice, 15(4), Article 4. 10.1111/j.1468-2850.2008.00137.x [DOI] [Google Scholar]
  3. Arch JJ, Fishbein JN, Finkelstein LB, & Luoma JB (2022). Acceptance and commitment therapy processes and mediation: Challenges and how to address them. Behavior Therapy 10.1016/j.beth.2022.07.005 [DOI] [PMC free article] [PubMed]
  4. Arch JJ, Wolitzky-Taylor KB, Eifert GH, & Craske MG (2012). Longitudinal treatment mediation of traditional cognitive behavioral therapy and acceptance and commitment therapy for anxiety disorders. Behaviour Research Therapy, 50(7–8), Article 7–8. 10.1016/j.brat.2012.04.007 [DOI] [PubMed] [Google Scholar]
  5. Bond FW, Hayes SC, Baer RA, Carpenter KM, Guenole N, Orcutt HK, Waltz T, & Zettle RD (2011). Preliminary psychometric properties of the Acceptance and Action Questionnaire-II: A revised measure of psychological inflexibility and experiential avoidance. Behavior Therapy, 42(4), 676–688. 10.1016/j.beth.2011.03.007 [DOI] [PubMed] [Google Scholar]
  6. Brown SL, Roush JF, Marshall AJ, Jones C, & Key C (2020). The intervening roles of psychological inflexibility and functional impairment in the relation between cancer-related pain and psychological distress. International Journal of Behavioral Medicine, 27(1), 100–107. 10.1007/s12529-019-09838-8 [DOI] [PubMed] [Google Scholar]
  7. Cherry KM, Hoeven EV, Patterson TS, & Lumley MN (2021). Defining and measuring “psychological flexibility”: A narrative scoping review of diverse flexibility and rigidity constructs and perspectives. Clinical Psychology Review, 84, 101973. 10.1016/j.cpr.2021.101973 [DOI] [PubMed] [Google Scholar]
  8. Ciarrochi J, Sahdra B, Marshall S, Parker P, & Horwath C (2014). Psychological flexibility is not a single dimension: The distinctive flexibility profiles of underweight, overweight, and obese people. Journal of Contextual Behavioral Science, 3(4), 236–247. 10.1016/j.jcbs.2014.07.002 [DOI] [Google Scholar]
  9. Correa-Fernández V, McNeel M, Sandoval JR, Tavakoli N, Kahambwe JK, & Kim H (2020). Acceptance and Action Questionnaire II: Measurement invariance and associations with distress tolerance among an ethnically diverse university sample. Journal of Contextual Behavioral Science, 17, 1–9. 10.1016/j.jcbs.2020.04.002 [DOI] [Google Scholar]
  10. Daks JS, & Rogge RD (2020). Examining the correlates of psychological flexibility in romantic relationship and family dynamics: A meta-analysis. Journal of Contextual Behavioral Science, 18, 214–238. 10.1016/j.jcbs.2020.09.010 [DOI] [Google Scholar]
  11. Downes MJ, Brennan ML, Williams HC, & Dean RS (2016). Development of a critical appraisal tool to assess the quality of cross-sectional studies (AXIS). BMJ Open, 6(12), e011458. 10.1136/bmjopen-2016-011458 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Egger M, Smith GD, Schneider M, & Minder C (1997). Bias in meta-analysis detected by a simple, graphical test. BMJ, 315(7109), 629–634. 10.1136/bmj.315.7109.629 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Eisenbeck N, & Szabó-Bartha A (2018). Validation of the Hungarian version of the Acceptance and Action Questionnaire-II (AAQ-II). Journal of Contextual Behavioral Science, 9, 80–87. 10.1016/j.jcbs.2018.07.007 [DOI] [Google Scholar]
  14. Eriksson M, Ghazinour M, & Hammarström A (2018). Different uses of Bronfenbrenner’s ecological theory in public mental health research: What is their value for guiding public mental health policy and practice? Social Theory & Health, 16(4), 414–433. 10.1057/s41285-018-0065-6 [DOI] [Google Scholar]
  15. Field AP (2001). Meta-analysis of correlation coefficients: A Monte Carlo comparison of fixed- and random-effects methods. Psychological Methods, 6(2), 161–180. 10.1037/1082-989X.6.2.161 [DOI] [PubMed] [Google Scholar]
  16. Field AP (2005). Is the meta-analysis of correlation coefficients accurate when population correlations vary? Psychological Methods, 10(4), 444–467. 10.1037/1082-989X.10.4.444 [DOI] [PubMed] [Google Scholar]
  17. Finger I. da R., de Freitas BI, & Oliveira M. da S. (2020). Psychological inflexibility in overweight and obese people from the perspective of acceptance and commitment therapy (ACT). Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity, 25(1), 169–175. 10.1007/s40519-018-0541-y [DOI] [PubMed] [Google Scholar]
  18. Fledderus M, Bohlmeijer ET, Pieterse ME, & Schreurs KM (2012). Acceptance and commitment therapy as guided self-help for psychological distress and positive mental health: A randomized controlled trial. Psychological Medicine, 42(3), 485–495. 10.1017/s0033291711001206 [DOI] [PubMed] [Google Scholar]
  19. Francis AW, Dawson DL, & Golijani-Moghaddam N (2016). The development and validation of the Comprehensive assessment of Acceptance and Commitment Therapy processes (CompACT). Journal of Contextual Behavioral Science, 5(3), Article 3. 10.1016/j.jcbs.2016.05.003 [DOI] [Google Scholar]
  20. Garner EV, & Golijani-Moghaddam N (2021). Relationship between psychological flexibility and work-related quality of life for healthcare professionals: A systematic review and meta-analysis. Journal of Contextual Behavioral Science, 21, 98–112. 10.1016/j.jcbs.2021.06.007 [DOI] [Google Scholar]
  21. Gentili C, Rickardsson J, Zetterqvist V, Simons LE, Lekander M, & Wicksell RK (2019). Psychological flexibility as a resilience factor in individuals with chronic pain. Frontiers in Psychology, 10. 10.3389/fpsyg.2019.02016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gloster AT, Klotsche J, Chaker S, Hummel KV, & Hoyer J (2011). Assessing psychological flexibility: What does it add above and beyond existing constructs? Psychological Assessment, 23(4), 970–982. 10.1037/a0024135 [DOI] [PubMed] [Google Scholar]
  23. Gloster AT, Klotsche J, Ciarrochi J, Eifert G, Sonntag R, Wittchen HU, & Hoyer J (2017). Increasing valued behaviors precedes reduction in suffering: Findings from a randomized controlled trial using ACT. Behaviour Research and Therapy, 91, 64–71. 10.1016/j.brat.2017.01.013 [DOI] [PubMed] [Google Scholar]
  24. Gloster AT, Meyer AH, & Lieb R (2017). Psychological flexibility as a malleable public health target: Evidence from a representative sample. Journal of Contextual Behavioral Science, 6(2), Article 2. 10.1016/j.jcbs.2017.02.003 [DOI] [Google Scholar]
  25. Gwet KL (2014). Handbook of inter-rater reliability: The definitive guide to measuring the extent of agreement among raters Advanced Analytics, LLC. [Google Scholar]
  26. Hawley PH, & Williford A (2015). Articulating the theory of bullying intervention programs: Views from social psychology, social work, and organizational science. Bullying Prevention and Intervention, 37, 3–15. 10.1016/j.appdev.2014.11.006 [DOI] [Google Scholar]
  27. Hayes SC, Ciarrochi J, Hofmann SG, Chin F, & Sahdra B (2022). Evolving an idionomic approach to processes of change: Towards a unified personalized science of human improvement. Behaviour Research and Therapy, 104155. 10.1016/j.brat.2022.104155 [DOI] [PubMed]
  28. Hayes SC, Hofmann SG, Stanton CE, Carpenter JK, Sanford BT, Curtiss JE, & Ciarrochi J (2019). The role of the individual in the coming era of process-based therapy. Behaviour Research and Therapy, 117, 40–53. 10.1016/j.brat.2018.10.005 [DOI] [PubMed] [Google Scholar]
  29. Hayes SC, Luoma JB, Bond FW, Masuda A, & Lillis J (2006). Acceptance and commitment therapy: Model, processes and outcomes. Behaviour Research and Therapy, 44(1), Article 1. 10.1016/j.brat.2005.06.006 [DOI] [PubMed] [Google Scholar]
  30. Hayes SC, Strosahl KD, & Wilson KG (2011). Acceptance and commitment therapy, second edition: The process and practice of mindful change Guilford Press. [Google Scholar]
  31. Headey B, Trommsdorff G, & Wagner GG (2022). Alternative recipes for life satisfaction: Evidence from five world regions. Applied Research in Quality of Life, 17(2), 763–794. 10.1007/s11482-021-09937-3 [DOI] [Google Scholar]
  32. Higgins JPT, & Thompson SG (2002). Quantifying heterogeneity in a meta-analysis. Statistics in Medicine, 21(11), 1539–1558. 10.1002/sim.1186 [DOI] [PubMed] [Google Scholar]
  33. Hofmann SG, & Hayes SC (2019). The future of intervention science: Process-based therapy. Clinical Psychological Science, 7(1), Article 1. 10.1177/2167702618772296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Jankowski PJ, Sandage SJ, Bell CA, Davis DE, Porter E, Jessen M, Motzny CL, Ross KV, & Owen J (2020). Virtue, flourishing, and positive psychology in psychotherapy: An overview and research prospectus. Psychotherapy, 57, 291–309. 10.1037/pst0000285 [DOI] [PubMed] [Google Scholar]
  35. Kalmakis KA, & Chandler GE (2015). Health consequences of adverse childhood experiences: A systematic review. Journal of the American Association of Nurse Practitioners, 27(8), 457–465. 10.1002/2327-6924.12215 [DOI] [PubMed] [Google Scholar]
  36. Kashdan TB, Disabato DJ, Goodman FR, Doorley JD, & McKnight PE (2020). Understanding psychological flexibility: A multimethod exploration of pursuing valued goals despite the presence of distress. Psychological Assessment, 32(9), 829–850. 10.1037/pas0000834 [DOI] [PubMed] [Google Scholar]
  37. Kashdan TB, & Rottenberg J (2010). Psychological flexibility as a fundamental aspect of health. Clinical Psychology Review, 30(7), 865–878. 10.1016/j.cpr.2010.03.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Keyes CL (2005). Mental illness and/or mental health? Investigating axioms of the complete state model of health. Journal of Consulting and Clinical Psychology, 73(3), Article 3. 10.1037/0022-006X.73.3.539 [DOI] [PubMed] [Google Scholar]
  39. Keyes CLM (2002). The Mental Health Continuum: From Languishing to Flourishing in Life. Journal of Health and Social Behavior, 43(2), 207–222. 10.2307/3090197 [DOI] [PubMed] [Google Scholar]
  40. Kioskli K, Winkley K, & McCracken LM (2019). Might psychological flexibility processes and Acceptance and Commitment Therapy (ACT) apply in adults with painful diabetic neuropathy? A cross-sectional survey. Journal of Contextual Behavioral Science, 13, 66–73. 10.1016/j.jcbs.2019.07.002 [DOI] [Google Scholar]
  41. Kılıç A, Hudson J, Scott W, McCracken LM, & Hughes LD (2022). A 12-month longitudinal study examining the shared and unique contributions of self-compassion and psychological inflexibility to distress and quality of life in people with Type 2 Diabetes. Journal of Psychosomatic Research, 155, 110728. 10.1016/j.jpsychores.2022.110728 [DOI] [PubMed] [Google Scholar]
  42. Knapp G, & Hartung J (2003). Improved tests for a random effects meta-regression with a single covariate. Statistics in Medicine, 22(17), 2693–2710. 10.1002/sim.1482 [DOI] [PubMed] [Google Scholar]
  43. Krafft J, Ferrell J, Levin ME, & Twohig MP (2017). Psychological inflexibility and stigma: A meta-analytic review. Journal of Contextual Behavioral Science 10.1016/j.jcbs.2017.11.002 [DOI]
  44. Levin ME, MacLane C, Daflos S, Seeley JR, Hayes SC, Biglan A, & Pistorello J (2014). Examining psychological inflexibility as a transdiagnostic process across psychological disorders. Journal of Contextual Behavioral Science, 3, 155–163. 10.1016/j.jcbs.2014.06.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. McCracken LM, & Gutierrez-Martinez O (2011). Processes of change in psychological flexibility in an interdisciplinary group-based treatment for chronic pain based on Acceptance and Commitment Therapy. Behaviour Research and Therapy, 49(4), 267–274. 10.1016/j.brat.2011.02.004 [DOI] [PubMed] [Google Scholar]
  46. McCracken LM, & Morley S (2014). The psychological flexibility model: A basis for integration and progress in psychological approaches to chronic pain management. The Journal of Pain, 15(3), 221–234. 10.1016/j.jpain.2013.10.014 [DOI] [PubMed] [Google Scholar]
  47. McCracken LM, Vowles KE, & Eccleston C (2004). Acceptance of chronic pain: Component analysis and a revised assessment method. Pain, 107(1), 159–166. 10.1016/j.pain.2003.10.012 [DOI] [PubMed] [Google Scholar]
  48. Olkin I, & Pratt JW (1958). Unbiased estimation of certain correlation coefficients. The Annals of Mathematical Statistics, 29(1), 201–211. JSTOR. [Google Scholar]
  49. Ong CW, Hayes SC, & Hofmann SG (2022). A process-based approach to cognitive behavioral therapy: A theory-based case illustration. Frontiers in Psychology, 13. 10.3389/fpsyg.2022.1002849 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Ong CW, Lee EB, Levin ME, & Twohig MP (2019). A review of AAQ variants and other context-specific measures of psychological flexibility. Journal of Contextual Behavioral Science, 12, 329–346. 10.1016/j.jcbs.2019.02.007 [DOI] [Google Scholar]
  51. Ong CW, Pierce BG, Woods DW, Twohig MP, & Levin ME (2019). The Acceptance and Action Questionnaire – II: An item response theory analysis. Journal of Psychopathology and Behavioral Assessment, 41, 123–134. 10.1007/s10862-018-9694-2 [DOI] [Google Scholar]
  52. Paladines-Costa B, López-Guerra V, Ruisoto P, Vaca-Gallegos S, & Cacho R (2021). Psychometric properties and factor structure of the Spanish version of the Acceptance and Action Questionnaire-II (AAQ-II) in Ecuador. International Journal of Environmental Research and Public Health, 18(6). 10.3390/ijerph18062944 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Präg P, Mills MC, & Wittek R (2016). Subjective socioeconomic status and health in cross-national comparison. Social Science & Medicine, 149, 84–92. 10.1016/j.socscimed.2015.11.044 [DOI] [PubMed] [Google Scholar]
  54. R Core Team. (2022). R: A language and environment for statistical computing R Foundation for Statistical Computing. [Google Scholar]
  55. Rochefort C, Baldwin AS, & Chmielewski M (2018). Experiential avoidance: An examination of the construct validity of the AAQ-II and MEAQ. Behavior Therapy, 49(3), Article 3. 10.1016/j.beth.2017.08.008 [DOI] [PubMed] [Google Scholar]
  56. Rolffs JL, Rogge RD, & Wilson KG (2016). Disentangling components of flexibility via the hexaflex model: Development and validation of the Multidimensional Psychological Flexibility Inventory (MPFI). Assessment, 25(4), Article 4. 10.1177/1073191116645905 [DOI] [PubMed] [Google Scholar]
  57. Secor SP, Limke-McLean A, & Wright RW (2017). Whose support matters? Support of friends (but not family) may predict affect and wellbeing of adults faced with negative life events. Journal of Relationships Research, 8, e10. Cambridge Core. 10.1017/jrr.2017.10 [DOI] [Google Scholar]
  58. Sterne JAC, & Egger M (2005). Regression methods to detect publication and other bias in meta-analysis. In Publication Bias in Meta‐Analysis (pp. 99–110). Wiley. 10.1002/0470870168.ch6 [DOI] [Google Scholar]
  59. Sterne JAC, Egger M, & Smith GD (2001). Investigating and dealing with publication and other biases in meta-analysis. BMJ, 323(7304), 101–105. 10.1136/bmj.323.7304.101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Sterne JAC, Sutton AJ, Ioannidis JPA, Terrin N, Jones DR, Lau J, Carpenter J, Rücker G, Harbord RM, Schmid CH, Tetzlaff J, Deeks JJ, Peters J, Macaskill P, Schwarzer G, Duval S, Altman DG, Moher D, & Higgins JPT (2011). Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ, 343. 10.1136/bmj.d4002 [DOI] [PubMed] [Google Scholar]
  61. Tyndall I, Waldeck D, Pancani L, Whelan R, Roche B, & Dawson DL (2019). The Acceptance and Action Questionnaire-II (AAQ-II) as a measure of experiential avoidance: Concerns over discriminant validity. Journal of Contextual Behavioral Science, 12, 278–284. 10.1016/j.jcbs.2018.09.005 [DOI] [Google Scholar]
  62. VanderWeele TJ, McNeely E, & Koh HK (2019). Reimagining health—Flourishing. JAMA, 321(17), 1667–1668. 10.1001/jama.2019.3035 [DOI] [PubMed] [Google Scholar]
  63. Viechtbauer W (2010). Conducting meta-analyses in R with the metafor package. Journal of Statistical Software, 36(3), 1–48. 10.18637/jss.v036.i03 [DOI] [Google Scholar]
  64. Wolgast M (2014). What does the Acceptance and Action Questionnaire (AAQ-II) really measure? Behavior Therapy, 45, 831–839. 10.1016/j.beth.2014.07.002 [DOI] [PubMed] [Google Scholar]
  65. Zhao C, Lai L, Zhang L, Cai Z, Ren Z, Shi C, Luo W, & Yan Y (2021). The effects of acceptance and commitment therapy on the psychological and physical outcomes among cancer patients: A meta-analysis with trial sequential analysis. Journal of Psychosomatic Research, 140, 110304. 10.1016/j.jpsychores.2020.110304 [DOI] [PubMed] [Google Scholar]

RESOURCES