Psychological inflexibility and clinical impact: Adaptation of the Acceptance and Action Questionnaire‑II in a sample of patients on haemodialysis treatment

Context and Objectives

Psychological inflexibility (PI) is defined as the difficulty to behave adaptively in response to changes in internal or external context. In patients with chronic kidney disease undergoing hemodialysis (HD), this characteristic may have significant clinical consequences, affecting treatment adherence, health-related quality of life (HRQOL), and emotional well-being. Although there is extensive evidence of the impact of PI across various clinical populations, its specific investigation in hemodialysis patients was limited in Spanish-speaking contexts.

The primary objective was to adapt and validate the Acceptance and Action Questionnaire-II (AAQ-II) in a sample of Spanish patients undergoing hemodialysis treatment, developing a disease-specific version called AAQHD-II. The authors sought not only to develop a psychometrically valid instrument for this population but also to evaluate the relationship between psychological inflexibility and relevant clinical indicators in dialysis (such as phosphorus levels) and its association with psychological symptoms and quality of life.

Method

Participants

The sample consisted of 186 outpatients receiving hemodialysis treatment from 4 Fresenius Medical Care clinics located in Córdoba, Spain. Demographic characteristics were: 113 men (60.8%) and 73 women (39.25%), with a mean age of 70.17 ± 1.01 years. A notable characteristic of the sample is that 97.85% of participants were pensioners or retired, reflecting a predominantly elderly population with typical comorbidities associated with chronic kidney disease. The sample represented a typical population of patients in chronic hemodialysis from a European healthcare system.

Developed Instrument: AAQHD-II

The primary instrument was the Acceptance and Action Questionnaire for patients on Hemodialysis-II (AAQHD-II), an adapted version of the original AAQ-II tailored for the specific needs of this clinical population. The instrument consists of 7 items evaluated on a 5-point Likert scale, where responses range from 1 (never true) to 5 (always true). Total scores theoretically range from 7 to 35, with higher scores indicating greater psychological inflexibility.

The development process followed rigorous methodology: an exhaustive review of items from the Spanish version of the original AAQ-II was conducted and disease-specific versions developed for other conditions were considered. An interdisciplinary team comprising a nephrologist, psychologist, social worker, and nurse evaluated the relevance and clarity of items for the hemodialysis population. Initially, a pool of 10 potential items was compiled and piloted with 6 hemodialysis patients to assess comprehension, relevance, and acceptability. After this piloting process, adjustments were made, reducing the final version to 7 items that best captured the construct of psychological inflexibility in the specific context of chronic hemodialysis.

Other Measurement Instruments

Complementary instruments were used to evaluate construct validity and expected correlations:

1. COOP-WONCA Quality of Life Charts: An instrument assessing multiple dimensions of health-related quality of life. It showed internal consistency of α = 0.77, indicating acceptable reliability.

2. Hospital Anxiety and Depression Scale (HADS): A widely used instrument for evaluating anxiety and depression symptoms in clinical populations. In this study, it achieved internal consistency of α = 0.83, demonstrating good reliability.

Statistical Analysis

Multiple psychometric analyses were conducted to evaluate the validity and reliability of the AAQHD-II:

• Internal consistency analysis: Cronbach's alpha was calculated to assess instrument homogeneity.
• Corrected item-total correlations: The correlation of each item with the total instrument score (excluding the item itself) was examined.
• Exploratory Factor Analysis (EFA): An EFA was performed to examine the factor structure of the instrument. Bartlett's test of sphericity was used to evaluate whether correlations between variables were significantly different from zero. The Kaiser-Meyer-Olkin (KMO) Adequacy Index was calculated to assess sampling adequacy. Parallel analysis was performed to determine the optimal number of factors to retain.
• Convergent validity: Pearson correlations between AAQHD-II scores and measures of anxiety, depression, and quality of life dimensions were examined.
• Group analysis: Patients were classified as "inflexible" (AAQHD-II ≥ 20) versus "flexible" (AAQHD-II < 20) to compare differences in quality of life, anxiety, and depression.
• Multivariate regression analysis: Regression was performed to examine which variables predicted psychological inflexibility and whether PI predicted clinical variables such as phosphorus levels.

Results

Internal Consistency and Item Analysis

The AAQHD-II demonstrated adequate internal consistency with a Cronbach's alpha of 0.72. Corrected item-total correlations ranged from 0.28 (item 4, lowest correlation) to 0.61 (item 3, highest correlation), indicating that all items contribute reasonably to measurement of the construct.

Factor Structure

Exploratory factor analysis revealed solid psychometric characteristics. Bartlett's test of sphericity was significant (χ² = 290.8 [21], P < 0.001), indicating that variables are intercorrelated and appropriate for factor analysis. The KMO index was 0.71, suggesting good sampling adequacy for proceeding with analysis. Parallel analysis recommended retaining a unidimensional structure. This single factor explained 44.31% of total variance, providing a parsimonious and clear solution. Item factor loadings ranged from 0.35 (item 4) to 0.81 (item 3), confirming that all items have substantial loadings on the psychological inflexibility factor.

Convergent Validity

The AAQHD-II showed expected correlation patterns with related measures, confirming its convergent validity. Moderate to moderately-high correlations were found with:

• Anxiety: r = 0.54 (P = 0.000)
• Depression: r = 0.51 (P = 0.000)
• General health-related quality of life: r = 0.45 (P = 0.000)
• Social activity: r = 0.42 (P = 0.000)
• Daily activities: r = 0.40 (P = 0.000)
• Emotional/feelings: r = 0.36 (P = 0.000)
• Perceived general health: r = 0.32 (P = 0.000)
• Overall quality of life: r = 0.31 (P = 0.000)
• Physical fitness: r = 0.23 (P = 0.000)
• Change in health: r = 0.14 (P = 0.000)

All coefficients were statistically significant, with magnitudes coherent with the hypothesis that greater psychological inflexibility would be associated with greater psychological symptoms and worse quality of life.

Group Analysis and Clinical Differences

Patients were classified as "inflexible" (AAQHD-II ≥ 20, n = 39) versus "flexible" (AAQHD-II < 20). Patients classified as inflexible presented:

• Poorer quality of life (24.67 ± 5.91 vs 20.54 ± 4.89, P < 0.001)
• Higher anxiety and depression (P < 0.001)

These differences were clinically significant and validated the proposed cutoff point.

Multivariate Regression Analysis

Multivariate regression to predict psychological inflexibility yielded an explanatory model with R² = 0.469 (P = 0.000), indicating that approximately 47% of the variance in PI was explained by the included predictors. Significant predictors were:

• Anxiety (P = 0.000)
• Depression (P = 0.009)
• Phosphorus levels (P = 0.013)

Additionally, in a multivariate model, psychological inflexibility was a significant predictor of phosphorus levels (P = 0.026), suggesting a bidirectional relationship between the psychological construct and a relevant clinical marker.

Discussion and Conclusions

The study results demonstrate that the AAQHD-II is a valid, reliable, and clinically relevant version of the AAQ-II specifically adapted for hemodialysis patients. The unidimensional structure, adequate internal consistency, and expected convergent validity patterns corroborate the instrument's viability. The development methodology, which included a participatory process with multidisciplinary team involvement and piloting with the target population, ensures that the instrument is culturally sensitive and clinically appropriate.

The significant relationship between psychological inflexibility and psychological symptoms (anxiety and depression) is consistent with Acceptance and Commitment Therapy (ACT) theory, which posits that inflexibility is a transdiagnostic mechanism involved in psychological distress. The innovative finding that PI also predicts serum phosphorus levels suggests that this psychological construct has concrete clinical consequences in hemodialysis, possibly through mechanisms of treatment adherence and dietary restrictions.

The validation of the AAQHD-II provides clinicians and researchers with a concise tool (7 items) to assess psychological inflexibility in this specific population, allowing identification of patients at greater risk of psychological and clinical complications.

Significance and Contribution

This study makes significant contributions in several aspects:

1. Population-Specific Instrument Validation: Provides the first validated version of the AAQ-II specifically adapted for Spanish hemodialysis patients, filling a gap in the psychometric assessment of relevant psychological constructs in this population.

2. Psychology-Clinical Connection: Establishes the relationship between psychological inflexibility and relevant clinical markers (phosphorus), suggesting that ACT-based interventions could have implications not only for psychological well-being but also for clinical outcomes.

3. Multidisciplinary Methodology: Demonstrates the value of a collaborative approach in instrument development, integrating perspectives from multiple clinical disciplines.

4. Foundation for Intervention: Provides a tool to identify patients with greater psychological inflexibility who might benefit from specific psychological interventions such as ACT.

Psychometric Verification Checklist

• Sample clearly described: N = 186, recruited from 4 clinics, with detailed demographic characteristics
• Instrument clearly specified: 7 items, 5-point Likert scale, development with multidisciplinary team
• Adaptation process documented: Review of previous versions, interdisciplinary team, prior piloting
• Internal consistency reported: α = 0.72
• Factor analysis conducted: Unidimensional EFA with 44.31% variance explained
• Factor loadings reported: Range 0.35-0.81
• Convergent validity examined: Multiple correlations with related constructs (r = 0.14-0.54)
• Group comparison: Significant differences between inflexible and flexible groups
• Predictive analysis: Multivariate regression with R² = 0.469
• Clinical significance addressed: Relationship with psychological symptoms and clinical markers
• Specific population: Hemodialysis, Spanish-speaking population
• Bilingual language: Article published in English and Spanish versions