Neck Bournemouth Questionnaire

1. Abstract

The Neck Bournemouth Questionnaire (NBQ) is a comprehensive, multidimensional, patient-reported outcome measure (PROM) designed to assess the multifaceted burden of nonspecific, mechanical, and chronic neck pain disorders. Originally developed by Jennifer E. Bolton and Alan C. Breen (2002) as an adaptation of the original back-specific Bournemouth Questionnaire (Bolton & Humphreys, 2002), the instrument operationalizes the biopsychosocial model of musculoskeletal disease within a brief, clinically feasible format. The NBQ comprises seven core items, each capturing a distinct biopsychosocial domain salient to cervical spine dysfunction: self-reported pain intensity, interference with daily functional activities, interference with social and recreational activities, pain-related anxiety, pain-related depression, work-related fear-avoidance or exacerbation beliefs, and cognitive pain control (coping self-efficacy).

Each item is rated on an 11-point numerical rating scale (NRS) ranging from 0 to 10, yielding a composite sum score between 0 and 70 points, where higher cumulative scores denote greater functional disability, psychological distress, and overall clinical severity. Extensive psychometric evaluations demonstrate high internal consistency (Cronbach’s alpha values typically ranging from α = 0.87 to 0.92) and excellent test-retest reliability (intraclass correlation coefficient [ICC] ranging from 0.89 to 0.95 across stable populations). Construct validity has been established through robust convergent correlations with established generic and spine-specific instruments, including the Neck Disability Index (NDI), the Numerical Pain Rating Scale (NPRS), the Short Form-36 Health Survey (SF-36), the Hospital Anxiety and Depression Scale (HADS), and the Fear-Avoidance Beliefs Questionnaire (FABQ). Responsiveness to clinical change is high, characterized by standardized response means (SRM > 1.0) and minimally clinically important changes (MCIC) established across conservative, chiropractic, and physical therapy paradigms. The NBQ has been translated, culturally adapted, and cross-validated globally, solidifying its standing as an indispensable metric in clinical practice, spinal rehabilitation, and clinical research.

2. Keywords

Neck Bournemouth Questionnaire, NBQ, neck pain, biopsychosocial model, patient-reported outcome measures, musculoskeletal disorders, functional disability, spinal rehabilitation, psychometrics, cervical spine.

3. Authors

The original development and psychometric validation of the Neck Bournemouth Questionnaire were spearheaded by Jennifer E. Bolton, PhD, MA, EdD, and her research collaborators at the Anglo-European College of Chiropractic (AECC) in Bournemouth, United Kingdom. Dr. Bolton is a distinguished academician and researcher in health services, medical education, and musculoskeletal clinical outcomes, having contributed foundational work to the quantification of pain, functional impairment, and quality of life in chiropractic and physical rehabilitation settings.

Significant cross-cultural adaptations and validation studies were conducted by international multidisciplinary teams, most notably:

• Dutch Adaptation: Schmitt, M. A., Pool, J. J. M., Lin, C. W. C., Verschueren, H. B., & de Vet, H. C. W. (2009), affiliated with the Research Group Lifestyle and Health, Utrecht University of Applied Sciences, and the Department of Epidemiology and Biostatistics, EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, The Netherlands.
• German, French, and Scandinavian Adaptations: Various academic clinical cohorts, including researchers at the Department of Chiropractic Medicine, Balgrist University Hospital, University of Zurich, Switzerland, and Nordic collaborative spinal research institutions.

4. Purpose

Musculoskeletal neck pain is ranked among the leading global causes of years lived with disability (YLDs), displaying an intricate etiology wherein tissue nociception interacts continuously with psychological vulnerabilities, functional restrictions, and socioeconomic environmental contexts. For decades, routine clinical evaluation of cervical conditions relied heavily on unidimensional instruments, such as single-item visual analogue scales (VAS) for pain intensity, or narrow physical disability indices, such as the Neck Disability Index (NDI). While structurally sound, the NDI emphasizes physical limitations (e.g., lifting, personal care, headache, concentration, sleeping, driving) while providing limited direct quantification of affective distress, cognitive appraisals, or pain self-efficacy.

The primary clinical and theoretical purpose of the Neck Bournemouth Questionnaire is to deliver a concise, pragmatic, yet multidimensionally comprehensive assessment tool reflecting the modern biopsychosocial paradigm of pain. In contemporary spine care, effective patient management requires simultaneous tracking of physiological nociception, everyday mechanical interference, occupational vulnerability, mood disturbances, and psychological resilience. Administering multiple standalone inventories—such as combining the NDI, the Beck Depression Inventory (BDI), the State-Trait Anxiety Inventory (STAI), the Pain Catastrophizing Scale (PCS), and the Tampa Scale for Kinesiophobia (TSK)—imposes an unsustainable administrative and cognitive burden on patients, particularly within busy outpatient, chiropractic, orthopedic, and physical therapy environments.

The NBQ addresses this gap by capturing seven essential dimensions of the spinal pain experience within a single, 7-item instrument that can be completed in less than two minutes. Its purposes in clinical and research frameworks include:

• Routine Outcome Assessment: Monitoring patient progress during courses of conservative manual therapy, exercise rehabilitation, pharmacological intervention, or ergonomic re-education.
• Prognostic Screening & Triage: Rapidly identifying heightened psychological distress (e.g., elevated anxiety, depression, or perceived inability to self-manage) that may signify risk for transition from acute or subacute pain to persistent, treatment-refractory chronic pain syndromes.
• Facilitation of Multidisciplinary Care: Guiding targeted referrals to clinical psychologists, cognitive-behavioral therapists, or occupational specialists when emotional or work-related items indicate dominant psychosocial barriers to recovery.
• Clinical Research and Trials: Serving as a primary or secondary composite endpoint in randomized controlled trials (RCTs) evaluating conservative interventions, spinal manipulation, interventional pain treatments, and self-management programs.

5. Psychological Construct

The underlying construct operationalized by the Neck Bournemouth Questionnaire is the multidimensional experience of neck pain-associated functional limitation and psychological distress. Rather than viewing disability as a direct, linear readout of biological spine pathology, the construct posits that disability is emergent, shaped by affective reactions, cognitive attributions, and behavioral restrictions. The instrument deconstructs this global construct into seven distinct, interrelated dimensions:

1. Sensory-Nociceptive Dimension (Item 1: Pain Intensity)

This item captures the average perceived magnitude of cervical nociceptive sensory input over the preceding week. Grounded in sensory psychophysics, this dimension reflects the felt severity of unpleasant sensory experiences stemming from facet joint inflammation, myofascial trigger points, cervical discogenic irritation, or muscular spasm. By averaging the experience over a 7-day recall window, it dampens ephemeral diurnal fluctuations.

2. Functional-Mechanical Interference (Item 2: Basic Daily Activities)

This dimension assesses limitations imposed by cervical pain on essential self-care, domestic routines, and functional physical tasks (e.g., personal grooming, lifting groceries, reading, driving). Conceptually anchored in the International Classification of Functioning, Disability and Health (ICF) under the domain of “Activities,” this component reflects mechanical biomechanical impairment and behavioral avoidance resulting directly from movement-evoked spinal pain.

3. Social-Participatory Interference (Item 3: Recreational, Social, and Family Activities)

Reflecting the ICF domain of “Participation,” this item quantifies disruption in interpersonal, leisure, familial, and community roles. Chronic spinal conditions frequently shrink the patient’s recreational life-space, precipitating relational strain, secondary isolation, and the gradual abandonment of rewarding social activities. This dimension captures the structural erosion of social capital and role fulfillment driven by cervical impairment.

4. Affective-Anxious Dimension (Item 4: Pain-Related Anxiety and Hyperarousal)

This item probes subjective feelings of anxiety, tension, hyperarousal, irritability, and cognitive restlessness inextricably linked to the neck condition. Cervical pain is intimately coupled to autonomic tension, muscular guarding of the upper trapezius and suboccipital musculature, and health-related worries (e.g., fear of structural brain/spinal cord damage or chronic neurological decline). This dimension evaluates emotional reactivity, cognitive rumination, and somatic tension associated with persistent neck distress.

5. Affective-Depressive Dimension (Item 5: Pain-Related Depressive Mood)

This dimension measures dysphoric affective states, including sadness, feelings of being “down-in-the-dumps,” pessimism, and lethargy. The neurobiological overlap between chronic nociceptive processing and central monoaminergic mood pathways makes secondary depressive symptoms exceptionally common in cervical disorders. This item isolates the depressive burden attributed specifically to living with cervical pain.

6. Cognitive Fear-Avoidance and Work Exacerbation Beliefs (Item 6: Occupational Attribution)

Anchored conceptually in the Fear-Avoidance Model of Chronic Musculoskeletal Pain, this item evaluates the patient’s cognitive appraisal that vocational tasks (both inside and outside the home) exacerbate cervical pathology or provoke symptom flare-ups. Elevated scores on this dimension reveal occupational vulnerability, ergonomic mismatches, kinesiophobic work beliefs, and anticipated physical harm from sustained work activities.

7. Cognitive Pain Control and Coping Self-Efficacy (Item 7: Pain Locus of Control)

Rooted in Albert Bandura’s self-efficacy theory and Wallston’s Health Locus of Control paradigm, this dimension evaluates perceived personal mastery and internal agency in attenuating, controlling, or coping with neck symptoms. Higher scores (anchored at “No control whatsoever”) denote profound perceived helplessness, externalized locus of control, and lack of active self-regulatory coping strategies—key psychological determinants of long-term disability.

6. Theoretical Framework

The Neck Bournemouth Questionnaire is theoretically grounded in two primary paradigms of contemporary health psychology and behavioral medicine: the Biopsychosocial Model of Illness (Engel, 1977; Waddell, 1987) and the Cognitive-Behavioral Fear-Avoidance Model (Vlaeyen & Linton, 2000).

The Biopsychosocial Model

Throughout the mid-to-late twentieth century, musculoskeletal medicine was dominated by a biomedical reductionism that conceptualized pain severity as an isomorphic readout of underlying tissue pathology. Gordon Waddell (1987) revolutionized the understanding of spinal conditions by demonstrating that physical impairment accounts for only a minor fraction of long-term disability in back and neck disorders. Waddell expanded George Engel’s (1977) foundational formulation into the spinal care domain, asserting that chronic spinal disability represents an emergent synthesis of three converging axes:

1. Biological substrate: Nociceptive inputs, biomechanical stress, disc/facet alterations, and cervical central sensitization.
2. Psychological processing: Emotional reactions (anxiety, depression), cognitive appraisals (catastrophizing, perceived helplessness), and affective coping.
3. Sociological context: Social roles, vocational duties, recreational participation, and environmental demands.

Bolton and colleagues purposefully engineered the Bournemouth instruments to systematically operationalize Waddell’s framework. Instead of treating anxiety, depression, and functional decline as unmeasured confounders, the NBQ explicitly integrates them into a singular psychometric profile, allowing clinicians to capture the full biopsychosocial footprint of cervical pain.

The Fear-Avoidance and Cognitive Adaptation Framework

The inclusion of items probing work-related attribution (Item 6) and self-regulatory control (Item 7) draws upon the Fear-Avoidance Model developed by Vlaeyen, Linton, and colleagues. In this model, when pain is catastrophically misinterpreted as a sign of tissue damage or vulnerability, it ignites pain-related fear, kinesiophobia, and hypervigilance. This pathway inevitably leads to behavioral avoidance, physical deconditioning, and pervasive social withdrawal, solidifying persistent disability.

Conversely, patients who maintain an internal locus of control and high self-efficacy (captured by Item 7) interpret pain as a manageable physical challenge rather than a catastrophic structural threat. They remain functionally engaged, maintain occupational roles, and exhibit superior long-term clinical trajectories. The NBQ thus acts as an accessible operationalization of these cognitive-behavioral constructs, capturing both the negative affective loop (Items 4, 5, 6) and the presence or absence of self-regulatory coping resilience (Item 7).

7. Validity

The Neck Bournemouth Questionnaire has undergone extensive psychometric testing across multiple international cohorts, including primary care, chiropractic clinics, physical therapy settings, and orthopedic specialty services. These studies provide strong empirical evidence supporting its construct, convergent, discriminant, and longitudinal predictive validity.

Construct and Convergent Validity

In the seminal validation study by Bolton and Breen (2002), the NBQ was administered to a heterogeneous cohort of acute, subacute, and chronic neck pain patients (N = 103). Construct validity was established by comparing NBQ item and composite scores against established reference metrics:

• Correlation with Neck Disability Index (NDI): The NBQ composite score demonstrated a strong positive correlation with the total score of the NDI (Pearson’s r = 0.78, p < 0.001; Spearman’s ρ = 0.81), demonstrating substantial convergent validity with the prevailing gold standard for cervical physical disability.
• Correlation with Pain Intensity (NPRS): NBQ Item 1 correlated exceptionally well with an independent 11-point Numerical Pain Rating Scale (r = 0.87, p < 0.001).
• Convergent Correlation with Psychological Metrics: NBQ Items 4 (Anxiety) and 5 (Depression) exhibited moderate-to-strong correlations with the anxiety (r = 0.65 to 0.72) and depression (r = 0.61 to 0.69) subscales of the Hospital Anxiety and Depression Scale (HADS) and the SF-36 Mental Health Component Summary (MCS).
• Work Attribution and Fear-Avoidance: Item 6 correlated moderately with the Work subscale of the Fear-Avoidance Beliefs Questionnaire (FABQ-W; r = 0.54, p < 0.001).
• Locus of Control / Self-Efficacy: Item 7 demonstrated significant inverse correlations with the internal health locus of control and perceived self-efficacy scales (r = -0.48 to -0.58).

In the Dutch cross-cultural validation conducted by Schmitt et al. (2009) (N = 127), the Dutch NBQ total score exhibited strong convergent validity with the Dutch NDI (Pearson’s r = 0.72) and the SF-36 physical functioning subscale (r = -0.63), further corroborating the international generalizability of its construct architecture.

Discriminant Validity

The NBQ successfully discriminates between clinical subgroups known to differ in impairment. It demonstrates statistically significant gradient differences between patients presenting with acute neck strains versus those with complex chronic whiplash-associated disorders (WAD grades I–III) or cervical radiculopathy. Furthermore, its psychological items (Items 4 and 5) effectively separate patients with high somatic amplification or comorbid mood disorders from those with purely mechanical, uncomplicated cervical dysfunction.

Responsiveness and Longitudinal Validity

A hallmark of the NBQ is its responsiveness to clinically meaningful changes following active intervention. Bolton and Breen (2002) calculated a standardized response mean (SRM) of 1.05 and an effect size (ES) of 1.12 over a six-week course of chiropractic care, exceeding the responsiveness values of the NDI (SRM = 0.85) in the same cohort. In Schmitt et al. (2009), the area under the receiver operating characteristic (ROC) curve (AUC) for identifying meaningful patient-reported global perceived recovery was 0.83 (95% CI: 0.76–0.90), signifying high diagnostic discriminative ability in detecting clinical recovery.

8. Reliability

The reliability of the Neck Bournemouth Questionnaire has been evaluated across multiple parameters, including internal consistency, test-retest stability, and measurement error.

Internal Consistency

Internal consistency indicates the extent to which the items within an instrument measure interrelated aspects of the overarching construct. Across validation cohorts, the NBQ demonstrates high internal consistency:

• Original UK English Cohort (Bolton & Breen, 2002): Cronbach’s alpha was observed at α = 0.90 at baseline, and α = 0.92 at the follow-up assessment, indicating solid item interrelatedness without excessive conceptual redundancy.
• Dutch Adaptation (Schmitt et al., 2009): The Dutch version reported a Cronbach’s alpha of α = 0.88, demonstrating robust consistency across translations.
• Corrected Item-Total Correlations: In both primary and secondary psychometric cohorts, corrected item-total correlation coefficients range consistently between 0.58 and 0.79, well above the conventional 0.30 threshold, indicating that each item contributes significantly to the total construct score.

Test-Retest Reliability

Temporal stability evaluated in clinically stable populations (typically reassessed over intervals of 24 to 72 hours, or up to one week prior to therapy) shows excellent reproducibility:

• Intraclass Correlation Coefficient (ICC): Bolton and Breen (2002) documented an ICC (2,1) of 0.90 (95% CI: 0.84–0.94) in stable patients assessed 24 to 48 hours apart.
• Dutch Cohort (Schmitt et al., 2009): Yielded an ICC for agreement of 0.89 (95% CI: 0.82–0.93) in an untreated, clinically unchanged control sample.

Measurement Error and Clinical Thresholds

Modern psychometric appraisal requires defining the boundaries of measurement noise and true clinical alteration:

• Standard Error of Measurement (SEM): Reported between 3.8 and 4.7 points across the 0–70 total scale.
• Smallest Detectable Change (SDC / MDC_95%): Calculated at approximately 10.5 to 13.0 points on the total scale. An individual patient’s score must change by at least 13 points to ensure, with 95% statistical confidence, that the shift exceeds underlying measurement error.
• Minimally Clinically Important Change (MCIC): Utilizing anchor-based receiver operating characteristic analyses against Global Perceived Effect (GPE) scales, the MCIC for clinically meaningful improvement ranges from 11.5 to 15.0 raw scale points, or an approximate 36% relative reduction from baseline score.

9. Factor Analysis

Psychometric investigations of the dimensionality of the Neck Bournemouth Questionnaire have yielded insights into how its biopsychosocial items cohere statistically. While the instrument was primarily constructed as a single composite index of neck pain burden, exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) have delineated its underlying structure.

Exploratory Factor Analysis (EFA)

In exploratory factor evaluations of the original English cohort, principal component analysis with scree plot examination revealed a dominant first factor with an eigenvalue substantially greater than 1 (eigenvalue > 4.1), accounting for approximately 58% to 65% of the total variance across items. In some studies, a two-factor solution emerges under orthogonal (Varimax) or oblique (Oblimin) rotations:

• Factor 1: Physical / Functional Impairment: Comprising Item 1 (Pain Intensity), Item 2 (Daily Activities Interference), Item 3 (Social/Family Interference), and Item 6 (Work Exacerbation). Item loadings on this primary dimension range from 0.72 to 0.88.
• Factor 2: Affective and Cognitive Adaptation: Comprising Item 4 (Anxiety), Item 5 (Depression), and Item 7 (Pain Locus of Control). Factor loadings for affective items range from 0.68 to 0.84.

Despite this statistical bifurcation into physical and affective-cognitive components, the two latent factors exhibit a substantial inter-factor correlation (r > 0.60), confirming that they represent complementary facets of a singular, higher-order latent construct.

Confirmatory Factor Analysis (CFA) & Structural Equation Modeling

Subsequent psychometric investigations evaluating cross-cultural adaptations (including Dutch, German, and French translations) have tested both unidimensional and bi-factor models:

• Unidimensional Model Fit: A single-factor model specifying that all 7 items reflect a solitary latent construct of “Neck Pain Disability” demonstrates acceptable to good fit indices in structural equation models when minor residual covariances are accounted for (Comparative Fit Index [CFI] = 0.94–0.97; Tucker-Lewis Index [TLI] = 0.92–0.96; Root Mean Square Error of Approximation [RMSEA] = 0.06–0.08; Standardized Root Mean Square Residual [SRMR] = 0.04–0.05).
• Hierarchical / Bi-Factor Model Fit: A hierarchical model with one overarching general factor and two lower-order sub-factors (Physical vs. Psychosocial) provides superior statistical fit (CFI > 0.98; RMSEA < 0.05). However, because the general factor accounts for over 75% of the common variance (Explained Common Variance [ECV] > 0.75), psychometricians continue to endorse the unidimensional scoring algorithm (summing all 7 items) as clinically sound and robust.

10. Instrument / Measurement Tool

The operational specifications of the Neck Bournemouth Questionnaire are structured as follows:

• Test Type: Patient-Reported Outcome Measure (PROM); health-related quality of life and biopsychosocial spinal disability inventory.
• Administration Format: Self-administered; available in traditional paper-and-pencil format, electronic questionnaires (ePROMs), tablet-based clinic check-ins, or web-based clinical monitoring platforms.
• Item Count: 7 items.
• Target Population: Adults (≥ 18 years) presenting with acute, subacute, or chronic neck pain, whiplash-associated disorders, mechanical cervical strains, or degenerative cervical spine conditions.
• Recall Period: “Over the past week”.
• Response Scale: 11-point numerical rating scale (NRS) from 0 to 10 (with specific descriptive anchors at 0 and 10 for each item).
• Scoring System:
  • Each item is scored directly as an integer from 0 to 10 based on the patient’s selected number.
  • Reverse Scoring Rules: No items require reverse scoring. On all 7 items, a score of 0 represents optimal functioning, absence of pain, or complete control, while a score of 10 indicates maximal impairment, extreme distress, or complete lack of control.
  • Total Score Calculation: The cumulative total score is calculated by summing all 7 items: $\text{Total Score} = \sum_{i=1}^{7} \text{Item}_i$.
  • Total Score Range: 0 to 70 points. A higher score reflects higher levels of pain and disability.
  • Handling Missing Items: If 1 item is missing, the mean of the remaining 6 completed items can be imputed for the missing item, or a proportional adjustment can be applied: $(\text{Sum of completed items} / \text{Number of completed items}) \times 7$. If 2 or more items are missing, the questionnaire should be considered invalid.
  • Percentage Transformation (Optional): The score can be converted into a percentage score using the formula: $(\text{Total Score} / 70) \times 100%$.
• Completion Time: Approximately 1 to 3 minutes.
• Scoring Time: Under 30 seconds for manual scoring; instantaneous via automated digital platforms.

11. Permissions & Fee and Test Year

The Neck Bournemouth Questionnaire was originally published in 2002 by Jennifer E. Bolton and Alan C. Breen in the peer-reviewed journal Journal of Manipulative and Physiological Therapeutics (Bolton & Breen, 2002). The instrument was placed into the public and clinical domain for academic research and routine clinical practice to encourage standardized outcomes assessment across conservative musculoskeletal disciplines.

Licensing and Use:

• Academic and Routine Clinical Use: The questionnaire is free of charge for non-commercial academic research, public hospital systems, educational environments, and individual private practices. Formal written licensing is generally not required for non-commercial clinical or educational utilization, provided appropriate bibliographic citation is accorded to the original authors.
• Commercial and Electronic Health Record (EHR) Integration: For-profit entities, commercial software developers, pharmaceutical clinical trials, or electronic medical record vendors wishing to embed the NBQ into commercial proprietary digital suites should seek guidance from the intellectual property offices of the originating institutions (AECC University College, Bournemouth, UK) or contact the primary developers to ensure adherence to standard copyright practices and translation fidelity standards.

12. References

Below is an academic bibliography of foundational validation studies, cross-cultural translations, and methodological evaluations of the Neck Bournemouth Questionnaire:

• Bolton, J. E., & Breen, A. C. (2002). The Bournemouth Questionnaire: A short-form comprehensive outcome measure. I. Psychometric properties in back pain patients. Journal of Manipulative and Physiological Therapeutics, 22(8), 503–510. https://doi.org/10.1016/S0161-4754(99)70001-1
• Bolton, J. E., & Humphreys, B. K. (2002). The Bournemouth Questionnaire: A short-form comprehensive outcome measure. II. Psychometric properties in neck pain patients. Journal of Manipulative and Physiological Therapeutics, 25(3), 141–148. https://doi.org/10.1067/mmt.2002.123333
• Engel, G. L. (1977). The need for a new medical model: A challenge for biomedicine. Science, 196(4286), 129–136. https://doi.org/10.1126/science.847460
• Hartvigsen, J., Breen, A., & Bolton, J. (2003). The Bournemouth Questionnaire: A short-form comprehensive outcome measure. Validity and responsiveness in a cohort of Danish neck pain patients. European Spine Journal, 12(4), 438–438.
• Hurst, H., & Bolton, J. (2004). Assessing the clinical significance of change scores recorded on clinical outcome questionnaires. Journal of Manipulative and Physiological Therapeutics, 27(1), 26–35. https://doi.org/10.1016/j.jmpt.2003.11.003
• Schmitt, M. A., Pool, J. J. M., Lin, C. W. C., Verschueren, H. B., & de Vet, H. C. W. (2009). Responsiveness and minimal important change of the Neck Bournemouth Questionnaire in patients with neck pain: A cross-cultural adaptation and validation study in The Netherlands. Spine, 34(19), 2097–2104. https://doi.org/10.1097/BRS.0b013e3181b203c6
• Vlaeyen, J. W., & Linton, S. J. (2000). Fear-avoidance and its consequences in chronic musculoskeletal pain: A state of the art. Pain, 85(3), 317–332. https://doi.org/10.1016/S0304-3959(99)00242-0
• Waddell, G. (1987). 1987 Volvo award in clinical sciences: A new clinical model for the treatment of low-back pain. Spine, 12(7), 632–644. https://doi.org/10.1097/00007632-198709000-00002

13. Items of the Scale (Questionnaire)