The relationship between low levels of mindfulness skills and pathological worry: The mediating role of psychological inflexibility

Full reference: Ruiz, F. J. (2014). The relationship between low levels of mindfulness skills and pathological worry: The mediating role of psychological inflexibility. Anales de Psicología, 30(3), 887–897.

Study type: Cross-sectional correlational study with mediation analyses.

Background and objectives

Pathological worry is a central feature of generalized anxiety disorder (GAD) and represents a pattern of repetitive and distressing thoughts about possible future threats. Although several mindfulness-based psychological interventions have been proposed for treating GAD, the specific relationship between mindfulness skills and pathological worry remains poorly understood. The Acceptance and Commitment Therapy (ACT) model emphasizes that psychological inflexibility—the inability to remain in contact with present experiences when they interfere with valued actions—plays a fundamental role in anxiety disorders.

This study was designed to explore whether psychological inflexibility mediates the relationship between low levels of mindfulness skills and pathological worry. As a preliminary step, it was necessary to adapt and validate the Kentucky Inventory of Mindfulness Skills (KIMS) into Spanish, given that the previous Spanish translation lacked formal validation. The author hypothesized that pathological worry would be related to low levels of mindfulness skills, and that this relationship would be completely mediated by psychological inflexibility.

Method

Participants

The sample consisted of 132 participants aged 18 to 69 years (M = 33.56, SD = 12.88) recruited from a university in southern Spain. Sixty percent of participants were women. Regarding educational level, 65% were college graduates or currently studying for a college degree. The sample was recruited from undergraduate psychology and pedagogy students and their personal contacts. No clinical inclusion or exclusion criteria were specified; the sample was non-clinical.

Design

A cross-sectional correlational design was employed. Participants completed a packet of self-report instruments measuring pathological worry, psychological inflexibility, and mindfulness skills. Data were collected at a single time point.

Intervention / Conditions

There was no experimental intervention. The study was purely observational and correlational.

Outcome measures

Penn State Worry Questionnaire (PSWQ; Meyer, Miller, Metzger, & Borkovec, 1990). The PSWQ is a 16-item, 5-point Likert-type self-report instrument designed to assess the persistent and nonspecific degree of worry that characterizes GAD. Sample items include "I shouldn't worry about things, but I just can't help it" and "I find it hard to dismiss worry about things." The scale shows internal consistency and test-retest reliability ranging from .93 to .95. The authors administered the Spanish translation by Sandín and colleagues.

Acceptance and Action Questionnaire – II (AAQ-II; Bond et al., 2011). The AAQ-II is a 7-item general measure of experiential avoidance and psychological inflexibility on a 7-point Likert scale. Items reflect unwillingness to experience unwanted emotions and thoughts (e.g., "I'm worried about being unable to control my worries and feelings") and inability to act in the present consistent with values when psychological experiences could interfere. The AAQ-II has good psychometric properties and a clearer factor structure than the original AAQ. The Spanish adaptation by Ruiz, Langer, Luciano, Cangas, and Beltrán (2013) was used.

Kentucky Inventory of Mindfulness Skills (KIMS; Baer et al., 2004). The KIMS is a 39-item, 5-point Likert-type scale measuring mindfulness skills. It consists of four subscales: "Observe" (noticing a variety of internal stimuli such as bodily sensations, cognitions, emotions, and external stimuli), "Describe" (labeling or naming observed phenomena using words), "Act with Awareness" (conscious focus on one activity at a time), and "Accept without Judgment" (being nonjudgmental or non-evaluative about present-moment experience). The KIMS has good internal consistency and factor structure. Although no previously validated Spanish translation existed, the author conducted a translation that was evaluated in this study.

Data analysis

Internal consistency analyses (Cronbach's alpha) and exploratory factor analysis were conducted to verify the factor structure of the Spanish KIMS version. Zero-order correlations between variables of interest were examined. Mediation analyses were conducted using the nonparametric bootstrapping procedure to estimate direct and indirect effects, as described in Preacher and Hayes (2004). This method offers greater statistical power than traditional mediator step approaches. Separate mediation analyses were conducted for each KIMS subscale. PSWQ scores served as the dependent variable, total KIMS scores (and subscales) as independent variables, and AAQ-II scores as the proposed mediator. Educational level was included as a covariate. 20,000 bootstrap samples were used to estimate 95% confidence intervals for direct and indirect effects, with indirect effects considered significant if the confidence interval did not include zero. Additionally, standardized effect sizes (ab₀; Preacher & Kelley, 2011) were calculated with 95% bootstrap confidence intervals.

Results

Internal consistency and factor structure of the KIMS-R

Preliminary analyses showed the KIMS was appropriate for factor analysis (Kaiser-Meyer-Olkin index = .82; Bartlett's sphericity test statistically significant, χ² = 1314.44, p < .001). Exploratory factor analyses identified a four-factor structure with eigenvalues greater than 1.0, accounting for 61.82% of variance. Items with factor loadings below .50 on their expected factor were eliminated. Items with the lowest item-total correlations were eliminated to obtain a reduced KIMS version (termed KIMS-R) more suitable for subsequent analyses. The final five-item version of each subscale showed good internal consistency. Specifically, the factor loading table revealed that the four factors (Accept without Judgment, Act with Awareness, Describe, and Observe) showed adequate psychometric properties in the Spanish translation. Cronbach's alphas were: Accept without Judgment (α = .84), Act with Awareness (α = .84), Describe (α = .88), and Observe (α = .77). Item-total correlations ranged from .40 to .81 depending on the factor. The complete scale showed adequate internal consistency (α = .84).

Descriptive data, internal consistencies, and zero-order correlations

Mean scores were: PSWQ (M = 50.5, SD = 11.7, α = .92), AAQ-II (M = 21.9, SD = 7.9, α = .86), and KIMS-R total (M = 68.6, SD = 11.4, α = .84). Results showed that mindfulness skills significantly correlated with pathological worry (r = -.20), with the complete KIMS-R scale correlating negatively with PSWQ. Psychological inflexibility (AAQ-II) significantly correlated with both pathological worry (r = -.38) and mindfulness skills (r = -.20). KIMS-R subscales also showed significant correlations with AAQ-II in the expected direction.

Mediation analyses of the effect of low levels of mindfulness skills on pathological worry through psychological inflexibility

Mediation analyses explored whether psychological inflexibility mediated the relationship between mindfulness skills (overall and each subscale separately) and pathological worry.

Overall mindfulness skills. The analysis examined whether psychological inflexibility mediated the relationship between total KIMS-R score and pathological worry. Results indicated that total KIMS-R score significantly predicted psychological inflexibility (path a: IE = -.265, SE = .058, t = -4.571, p < .001) and psychological inflexibility significantly predicted pathological worry (path b: IE = .767, SE = .123, t = 6.253, p < .001). The total direct effect of KIMS-R on pathological worry was significant (IE = -.195, SE = .089, t = -2.197, p < .05). However, when the mediator (psychological inflexibility) was included in the model, the direct effect of KIMS-R on pathological worry became nonsignificant (IE = .009, SE = .084, t = .105, p > .05). The indirect effect of total KIMS-R on pathological worry through psychological inflexibility was significant (IE = -.203, SE = .054, 95% CI [-0.324, -0.110]). The effect size of the indirect effect was medium to large (ab₀ = -.203, 95% CI [-0.324 to -0.110]), indicating that psychological inflexibility completely mediated the relationship between low levels of overall mindfulness skills and pathological worry.

Accept without Judgment. The "Accept without Judgment" subscale significantly predicted psychological inflexibility (path a: IE = -.736, SE = .136, t = -5.409, p < .001) and psychological inflexibility significantly predicted pathological worry (path b: IE = .664, SE = .125, t = 5.328, p < .001). The total effect was significant (IE = -.847, SE = .204, t = -4.159, p < .001), but when the mediator was included the direct effect became nonsignificant (IE = -.359, SE = .205, t = -1.749, p > .05). The indirect effect was significant (IE = -.489, SE = .120, 95% CI [-0.770, -0.287]), with a medium to large effect size.

Act with Awareness. The "Act with Awareness" subscale significantly predicted psychological inflexibility (path a: IE = -.597, SE = .162, t = -3.683, p < .001) and psychological inflexibility significantly predicted pathological worry (path b: IE = .862, SE = .116, t = 7.443, p < .001). The total effect was significant (IE = .060, SE = .246, t = .246, p > .05), but the direct effect when the mediator was included was also nonsignificant (IE = .575, SE = .215, t = 2.681, p < .05). The indirect effect was significant (IE = -.515, SE = .171, 95% CI [-0.890, -0.221]), with a medium effect size.

Describe. The "Describe" subscale predicted psychological inflexibility (path a: IE = -.295, SE = .150, t = -1.959, p > .05, nonsignificant) and predicted pathological worry (path b: IE = .754, SE = .115, t = 6.575, p < .001). The total effect was nonsignificant (IE = -.291, SE = .218, t = -1.338, p > .05). The indirect effect was nonsignificant (IE = -.222, SE = .127, 95% CI [-0.492, 0.009]).

Observe. Finally, the "Observe" subscale predicted psychological inflexibility (path a: IE = -.071, SE = .184, t = -.386, p > .05, nonsignificant) and predicted pathological worry (path b: IE = .761, SE = .113, t = 6.737, p < .001). The total effect was nonsignificant (IE = -.068, SE = .264, t = -.257, p > .05). The indirect effect was nonsignificant (IE = -.054, SE = .134, 95% CI [-0.267, 0.172]).

In summary, mediation analyses revealed that psychological inflexibility completely mediated the relationship between low levels of overall mindfulness skills and pathological worry. Specifically, regarding individual mindfulness subscales, psychological inflexibility completely mediated the relationship between "Accept without Judgment" and pathological worry, as well as between "Act with Awareness" and pathological worry. However, the "Describe" and "Observe" subscales did not show significant relationships with pathological worry, nor indirect effects mediated by psychological inflexibility.

Discussion and conclusions

This study examined the hypothesis that the relationship between low levels of mindfulness skills and pathological worry would be mediated by psychological inflexibility. As a preliminary step, the KIMS was translated and adapted into Spanish, showing good internal consistency and factor structure. Low levels of mindfulness skills were significantly related to pathological worry, and mediation analyses revealed that psychological inflexibility completely mediated this relationship.

These results have several clinical implications. First, not all mindfulness skills appear to relate to pathological worry in the same way. Specifically, the "Accept without Judgment" and "Act with Awareness" skills were most relevant in the relationship with pathological worry, while "Describe" and "Observe" did not show significant relationships. Second, according to this study's results, the ability to be nonjudgmental about present-moment experience (i.e., the "Accept without Judgment" skill) was the mindfulness factor most closely related to lower pathological worry, and this effect was completely mediated by psychological inflexibility. Finally, the fact that psychological inflexibility functioned as a complete mediator between the relationship of mindfulness skills and pathological worry suggests that developers of mindfulness-based interventions for pathological worry and GAD might consider: (a) reducing the number of avoidant interactions (e.g., those specifically related to using worry as an avoidance strategy); (b) explicitly linking the use of awareness skills with valued behavior according to ACT model goals; and (c) making a major effort to promote acceptance without judgment as a way of allowing people to contact fears and worries that emerge in the present moment without needing to avoid or escape from them.

The present study has important limitations. A cross-sectional design was used, which does not allow determination of causal relationships between variables. Additionally, the sample consisted of non-clinical participants generalizable primarily to the university population, limiting generalization of findings to individuals with GAD or other anxiety disorders. All data were obtained through self-report measures, which could introduce common method bias. Finally, the Spanish translation of the KIMS lacked previous formal validation, although the item reduction process and factor validation were conducted in this study.

Despite these limitations, this study contributes to the literature by specifically demonstrating that psychological inflexibility completely mediates the relationship between low levels of mindfulness skills and pathological worry. These results support the ACT model of psychopathology and the efficacy of treatment for disorders characterized by pathological worry. Findings also indicate that not all mindfulness skills relate equally to pathological worry, and that developers of mindfulness-based interventions for GAD could benefit from specifically emphasizing acceptance without judgment and acting with awareness skills in promoting psychological flexibility.

Significance and contribution

This study contributes to the field of anxiety psychopathology by clarifying the mechanism through which mindfulness skills relate to pathological worry. It demonstrates that psychological inflexibility is a complete mediator of this relationship, suggesting that the benefits of mindfulness may not be due solely to greater present-moment awareness, but rather to changes in how individuals functionally relate to their internal experiences. This finding offers a transdiagnostic perspective on generalized anxiety and has implications for designing interventions that promote acceptance and flexibility in the face of unwanted thoughts and emotions.