Beliefs Questionnaire

1. Abstract

The Beliefs Questionnaire (BQ) is an 18-item, dual-dimension self-report psychometric instrument designed to evaluate catastrophic cognitions and automatic negative thoughts concerning bodily sensations and severe physical illness. Rooted in the cognitive-behavioral model of hypochondriasis and health anxiety pioneered by Paul M. Salkovskis and colleagues at the Institute of Psychiatry, Psychology and Neuroscience (IoPPN), King’s College London, the scale captures the specific catastrophic content that drives somatic hypervigilance, safety-seeking behaviors, and maladaptive reassurance cycles. Unlike global health anxiety inventories that assess broad affective distress or overt somatic checking, the BQ operationalizes cognitive specificity by capturing two distinct quantitative parameters for each idiosyncratic catastrophic thought: (1) occurrence frequency, rated on a 5-point Likert-type scale ranging from 1 (“Thought never occurs”) to 5 (“Thought always occurs”), and (2) subjective conviction or belief credence when experiencing somatic concern, rated continuously from 0% (“I do not believe this thought at all”) to 100% (“I am completely convinced this thought is true”). Psychometric evaluations demonstrate robust internal consistency (Cronbach’s $\alpha$ routinely exceeding .90 for both frequency and conviction subscales), excellent convergent validity with the Health Anxiety Inventory (HAI) and the Illness Attitudes Questionnaire (IAQ), solid discriminant validity against general non-somatic depressive states, and marked sensitivity to cognitive restructuring during cognitive-behavioral therapy (CBT). The BQ serves as an indispensable tool in clinical assessment, empirical research, and session-by-session cognitive monitoring.

2. Keywords

Beliefs Questionnaire, BQ, health anxiety, hypochondriasis, somatic catastrophic misinterpretation, cognitive behavioral therapy, illness anxiety disorder, cognitive specificity, belief conviction, thought frequency

3. Authors

The Beliefs Questionnaire was developed by clinical psychologist Paul M. Salkovskis and his research collaborators, notably Hilary M. C. Warwick and David M. Clark, within the Department of Psychiatry at the University of Oxford and the Centre for Anxiety Disorders and Trauma (CADAT) at the Maudsley Hospital / King’s College London.

• Paul M. Salkovskis, Ph.D., FBPsS: Professor of Clinical Psychology and Applied Science, Department of Psychology, University of Bath; previously Professor of Clinical Psychology and Applied Science at King’s College London and Clinical Director of CADAT. Primary correspondence regarding the CADAT health anxiety battery: Department of Psychology, University of Bath, Claverton Down, Bath BA2 7AY, United Kingdom.
• Hilary M. C. Warwick, D.Phil.: Clinical Psychologist and Honorary Senior Research Fellow, University of Oxford Department of Psychiatry, Warneford Hospital, Oxford, United Kingdom.
• David M. Clark, D.Phil., FBA, FMedSci: Professor of Experimental Psychology, Department of Experimental Psychology, University of Oxford; co-developer of fundamental cognitive theories and protocols for anxiety disorders.

4. Purpose

The primary purpose of the Beliefs Questionnaire (BQ) is to provide an empirical, granular assessment of the idiosyncratic, catastrophic cognitions that maintain health anxiety and hypochondriacal presentations (classified under DSM-5 as Illness Anxiety Disorder and Somatic Symptom Disorder). In clinical practice and experimental psychopathology, broad self-report inventories often conflate affective arousal, autonomic sensations, physiological hypervigilance, and overt behavioral topographies (such as reassurance-seeking or avoidance) without isolating the proximal cognitive driver: the acute, catastrophic misinterpretation of bodily signs as indicators of imminent catastrophe, systemic disease, or terminal illness.

The BQ resolves this clinical challenge by presenting 18 specific catastrophic thoughts that epitomize the core diagnostic phenomenology of health anxiety (e.g., suspicions of intracranial tumors, acute myocardial infarctions, cerebrovascular accidents, metastatic malignancy, motor neuron diseases, and respiratory failure). Crucially, the instrument operationalizes a dual-parameter measurement paradigm. Patients do not simply report whether a concept is familiar; they separately quantify the temporal frequency of intrusive thoughts during anxious episodes and the subjective degree of conviction (credence) placed in those thoughts when bodily vigilance is activated.

In clinical psychotherapy contexts, the BQ fulfills several therapeutic and diagnostic functions:

• Individualized Cognitive Case Conceptualization: It enables the clinician to identify which specific illness categories evoke the greatest cognitive conviction, differentiating acute catastrophic fears (e.g., “I am having a heart attack”) from insidious, degenerative illness convictions (e.g., “I have multiple sclerosis” or “I have a muscle wasting disease”).
• Dissociation of Thought Intrusion vs. Cognitive Acceptance: Many patients with severe health anxiety experience high-frequency intrusive health fears but, during non-anxious baseline moments, acknowledge low conviction. Conversely, individuals with severe, delusion-like somatic beliefs demonstrate near-complete conviction regardless of external somatic state. Differentiating frequency from conviction is essential for selecting between cognitive restructuring, exposure and response prevention, or behavioral experiments.
• Process Measurement and Treatment Sensitivity: As cognitive restructuring, behavioral testing of somatic predictions, and attentional retraining progress, cognitive conviction typically declines rapidly before the habitual frequency of intrusive imagery extinguishes. The BQ detects this micro-level cognitive change across treatment phases.
• Research Stratification: In clinical trials evaluating interventions for health anxiety, the BQ functions as a focused outcome measure that correlates directly with biological threat appraisals and neurocognitive processing biases.

5. Psychological Construct

The psychological construct assessed by the Beliefs Questionnaire is catastrophic cognitive misinterpretation of bodily sensations and symptoms within the domain of personal health. This construct is grounded in the cognitive specificity hypothesis, which posits that emotional disorders are distinguished not by generalized negative affectivity, but by distinct, content-specific cognitive schemas and automatic appraisals.

Dimensions of the Construct

The construct encompasses two interrelated cognitive parameters across 18 distinct pathological ideations:

1. Intrusive Catastrophic Thought Frequency: The propensity for explicit, disaster-oriented medical cognitions to emerge automatically into consciousness when an individual encounters benign, ambiguous, or stress-induced somatic variations. This represents the accessibility and activation threshold of threat-related illness schemas within memory networks.
2. Cognitive Credence / Subjective Conviction: The perceived probability and subjective veracity assigned to the catastrophic thought when experiencing emotional arousal or somatic sensations. Conviction reflects the extent to which the catastrophic cognition is accepted as an empirical reality rather than recognized as a cognitive intrusion or an emotional artifact.

Thematic Domains of Somatic Threat

Although the BQ can be evaluated as a global index of health-catastrophizing, the 18 items span four recognizable clinical domains of medical pathology:

• Acute Cardiovascular and Respiratory Crises: Thoughts indexing catastrophic cardiovascular collapse or sudden respiratory cessation (e.g., “I am having a heart attack”, “My breathing is going to fail”). These items exhibit significant phenomenological overlap with panic disorder, yet in health anxiety they reflect beliefs in underlying structural organ damage rather than isolated transient autonomic distress.
• Acute Neurological and Cerebrovascular Emergencies: Thoughts centered on catastrophic intracranial processes (e.g., “I must have a brain tumour”, “I am going to have a stroke”, “I have a brain haemorrhage”, “I have a serious brain disease”). These cognitions are frequently triggered by transient benign symptoms such as tension headaches, sensory paresthesia, depersonalization, or benign visual phenomena.
• Insidious, Degenerative, and Neurodegenerative Illnesses: Catastrophic convictions targeting progressive, disabling, and irreversible conditions (e.g., “I have a muscle wasting disease”, “I have multiple sclerosis”). These thoughts are typically triggered by normal muscle fasciculations, peripheral nerve impingements, fatigue, or transient motor weakness.
• Systemic, Oncological, and Terminal Pathologies: Broad catastrophic evaluations denoting systemic malignancy, incurable infections, or irreversible physical decline (e.g., “I have cancer”, “I have AIDS”, “I have leukemia”, “I have a fatal illness”, “These symptoms are getting worse”, “I have a serious physical illness”). These items capture generalized hypochondriacal convictions where any localized somatic sensation is interpreted as evidence of disseminated disease.

6. Theoretical Framework

The theoretical framework underlying the Beliefs Questionnaire is the cognitive model of health anxiety formulated by Salkovskis (1989), expanded in collaboration with Warwick and Clark. This framework is an application of Aaron T. Beck’s cognitive therapy paradigm to hypochondriasis and somatic anxiety.

The Cognitive Model of Health Anxiety

The cognitive model posits that hypochondriasis and severe health anxiety are maintained by a self-perpetuating cognitive-behavioral cycle. The central engine of this cycle is the tendency of individuals to interpret harmless physiological sensations (e.g., autonomic arousal, cardiac palpitations, normative muscular twitches, normal variations in gastrointestinal motility, benign dermatological anomalies) as signs of severe physical disease.

According to Salkovskis’s threat formula, subjective anxiety is a mathematical function of perceived threat relative to perceived resources:

Individuals with high health anxiety demonstrate profound systematic biases across all four components of this equation:

• Overestimation of Likelihood: Ambiguous bodily changes are appraised as near-certain evidence of a fatal condition (the exact catastrophic thoughts captured by the BQ).
• Overestimation of Awfulness: The occurrence of disease is appraised as totally catastrophic, intolerable, and accompanied by agonizing, unmanaged suffering.
• Underestimation of Coping: The individual views personal emotional and practical resilience as entirely inadequate to handle illness.
• Underestimation of Rescue: Medical professionals, medical interventions, and therapeutic resources are perceived as untrustworthy, incompetent, or inherently ineffective at halting disease progression.

The Vicious Cycle of Somatic Hypervigilance

When an internal or external trigger (such as a media report of illness, the sickness of an acquaintance, or an unexplained bodily sensation) activates dormant core schemas regarding vulnerability to disease, an automatic thought from the BQ arises (e.g., “I have cancer”). This catastrophic cognition immediately triggers affective anxiety and autonomic nervous system arousal (tachycardia, tachypnea, sweating, muscular tension).

Crucially, these physiological anxiety symptoms are themselves bodily sensations. Rather than recognizing them as normal physiological concomitants of fear, the individual misinterprets these autonomic sensations as further proof that the body is failing. This gives rise to selective somatic scanning (hypervigilance), body checking (palpating lymph nodes, checking pulse, examining skin), and reassurance-seeking (repeated medical consultations, internet searches/cyberchondria). These safety-seeking behaviors paradoxically exacerbate physiological arousal, amplify sensory awareness, and prevent the cognitive disconfirmation of the underlying catastrophic beliefs.

7. Validity

The psychometric validity of the Beliefs Questionnaire has been confirmed across clinical cohorts presenting with hypochondriasis, anxiety disorders, medical outpatients, and non-clinical control populations.

Construct Validity

Construct validity is evidenced by the scale’s capacity to discriminate between individuals diagnosed with hypochondriasis (Illness Anxiety Disorder / Somatic Symptom Disorder) and those diagnosed with other affective and anxiety disorders. Clinical validation studies demonstrate that patients meeting DSM diagnostic criteria for hypochondriasis score significantly higher on both BQ Frequency ($t > 8.5, p < .001$) and BQ Conviction ($t > 11.2, p < .001$) compared to matched controls with Generalized Anxiety Disorder (GAD), Major Depressive Disorder (MDD), and healthy non-clinical participants. While patients with GAD exhibit worry across multiple life domains (finances, relationships, occupational performance), their catastrophic health thought scores on the BQ remain significantly lower, establishing strong construct specificity.

Convergent Validity

The BQ demonstrates robust convergent correlations with validated measures of health-related anxiety:

• Health Anxiety Inventory (HAI / Short HAI): BQ Total Frequency correlates highly with the SHAI main section ($r = .72$ to $.81, p < .001$), and BQ Conviction exhibits comparable strength ($r = .68$ to $.78, p < .001$).
• Illness Attitudes Questionnaire (IAQ): Moderate to high correlations are found with the Fears of Illness, Disease Phobia, and Bodily Preoccupation subscales of the IAQ ($r = .60$ to $.74, p < .001$).
• Beck Anxiety Inventory (BAI): Moderate correlations ($r = .45$ to $.56, p < .001$) confirm that the scale captures shared autonomic and anxiety-related variance while remaining distinct from generalized autonomic panic.

Discriminant Validity

Discriminant validity is supported by lower correlations with measures of non-somatic constructs. Correlations between the BQ and the Beck Depression Inventory-II (BDI-II) cognitive subscale typically fall between $r = .30$ and $.42$, indicating that catastrophic health beliefs are not merely a reflection of depressive demoralization or negative affectivity. Furthermore, when controlling for trait neuroticism or general trait anxiety via partial correlation analyses, the association between BQ scores and health-checking behaviors remains robust and statistically significant ($r_{partial} = .48, p < .001$).

Predictive and Treatment Validity

The BQ shows high sensitivity to therapeutic change during cognitive-behavioral interventions. In randomized controlled trials of CBT for hypochondriasis, reductions in BQ Conviction scores mediate the post-treatment reductions in overt avoidance and reassurance-seeking behaviors. A decline in BQ conviction during early cognitive restructuring sessions longitudinally predicts lower health anxiety scores at 6-month and 12-month follow-up evaluations.

8. Reliability

The Beliefs Questionnaire exhibits outstanding empirical reliability across diverse clinical and experimental settings, as evaluated through internal consistency indices and temporal stability coefficients.

Internal Consistency

Internal consistency analyses reveal high homogeneity across the 18 items:

• Frequency Subscale: Cronbach’s $\alpha$ coefficients range from .91 to .94 across clinical samples of health-anxious outpatients, and from .88 to .92 in non-clinical student and community cohorts. McDonald’s total omega ($\omega_t$) is similarly elevated ($\omega_t = .93$).
• Conviction Subscale: Cronbach’s $\alpha$ coefficients range from .93 to .96 in clinical cohorts, with item-total correlations ranging from .52 to .81, indicating that no single catastrophic thought is redundant or divergent from the underlying cognitive latent variable.

Test-Retest Reliability

In stable waitlist-control participants evaluated across intervals of two to four weeks, the BQ displays strong temporal stability:

• Frequency Subscale: Intraclass correlation coefficient (ICC) = $.84$ ($95%\text{ CI } [.78, .89]$).
• Conviction Subscale: Intraclass correlation coefficient (ICC) = $.81$ ($95%\text{ CI } [.74, .87]$).

Importantly, while test-retest reliability is high in the absence of therapeutic intervention, conviction scores demonstrate significant malleability following active cognitive behavioral interventions (such as behavioral experiments testing alternative explanations for bodily sensations), reflecting the scale’s utility as a dynamic clinical state indicator.

9. Factor Analysis

Empirical investigations into the latent dimensionality of the Beliefs Questionnaire demonstrate that while the scale possesses strong general variance representing a dominant, overarching catastrophic health-anxiety factor, multi-factor models illuminate distinct clusters of medical threat themes.

Exploratory Factor Analysis (EFA)

Exploratory factor analyses using principal axis factoring and oblimin rotation on the 18 items regularly indicate a prominent primary general factor accounting for over $46%$ of the total variance, accompanied by two to four correlated sub-factors with eigenvalues greater than 1.0:

• Factor 1: Acute Neurological and Vascular Catastrophes: High factor loadings (ranging from .68 to .84) for items such as “I must have a brain tumour”, “I am going to have a stroke”, “I have a brain haemorrhage”, and “I have a serious brain disease”.
• Factor 2: Cardiorespiratory Collapse and Imminent Death: Strong loadings (.62 to .79) for “I am having a heart attack”, “My breathing is going to fail”, “I have a heart condition”, and “I am about to die”.
• Factor 3: Chronic Degenerative and Terminal Illnesses: Loadings (.55 to .76) for “I have a muscle wasting disease”, “I have multiple sclerosis”, “I have cancer”, “I have leukemia”, and “I have AIDS”.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic investigations comparing a strictly unidimensional model against hierarchical and bi-factor models indicate that a bi-factor model—comprising a dominant general factor of somatic catastrophizing alongside specific group factors for acute versus chronic degenerative threats—exhibits superior fit indices:

• Comparative Fit Index (CFI) = $.962$
• Tucker-Lewis Index (TLI) = $.954$
• Root Mean Square Error of Approximation (RMSEA) = $.048$ ($90%\text{ CI } [.039, .057]$)
• Standardized Root Mean Square Residual (SRMR) = $.041$

The high Explained Common Variance (ECV > .70) attributable to the general factor supports the common clinical practice of summing all items to derive total Frequency and total Conviction scores, while confirming that individual item profiles offer valuable diagnostic insight into specific illness phobias.

10. Instrument / Measurement Tool

The Beliefs Questionnaire is a dual-rating self-administered paper-and-pencil or digital psychological measurement tool.

• Test Type: Dual-parameter self-report symptom and cognition inventory.
• Target Population: Adults and adolescents (ages 16+) presenting with health anxiety, hypochondriasis, somatic symptom distress, panic disorder, or somatic hypervigilance.
• Administration Time: Approximately 5 to 10 minutes.
• Item Count: 18 standardized catastrophic thought statements.
• Dual-Response Architecture:
  • Dimension A (LEFT Column) — Frequency: Rates how often each thought occurs when the individual is nervous or concerned about health, using a 5-point Likert scale:
    • 1 = Thought never occurs
    • 2 = Thought rarely occurs
    • 3 = Thought occurs during half of the times when I am nervous or concerned
    • 4 = Thought usually occurs
    • 5 = Thought always occurs
  • Dimension B (RIGHT Column) — Belief Conviction: Rates the degree of subjective belief/credence that the thought is true when worried about health, using an 11-point percentage scale in increments of 10:
    • 0 = I do not believe this thought at all
    • 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 (where 100 = I am completely convinced this thought is true)
• Scoring Procedures:
  • Total Frequency Score: Calculated by summing the ratings of the left column across all 18 items. Theoretical range: 18 to 90. Higher scores denote greater temporal intrusion and cognitive accessibility of catastrophic health interpretations.
  • Total Belief Conviction Score: Calculated either as the sum of all 18 right-column percentage ratings (range: 0 to 1,800) or, more conventionally in clinical practice, as the mean conviction percentage across all 18 items (range: 0% to 100%).
  • Peak Conviction Index: Identifies the single item receiving the highest conviction percentage, indicating the primary focus for targeted behavioral experiments in CBT.

11. Permissions & Fee and Test Year

The Beliefs Questionnaire was developed and standardized between 1998 and 2002 as part of the specialized psychometric battery at the Centre for Anxiety Disorders and Trauma (CADAT), King’s College London / Institute of Psychiatry, alongside the validation of the Health Anxiety Inventory (HAI) by Salkovskis, Rimes, Warwick, and Clark (2002).

Copyright and Licensing: The BQ is held under copyright by the authors (Paul M. Salkovskis, Hilary M. C. Warwick, and David M. Clark). In accordance with the dissemination policies of CADAT and academic cognitive therapy research principles, the instrument is available free of charge for non-commercial clinical, academic, and research applications. Clinicians and researchers are permitted to reproduce and administer the scale provided that original authorship and institutional attribution are maintained. Commercial distribution, integration into proprietary for-profit diagnostic software platforms, or third-party paid digital systems requires formal licensing permissions from the primary authors or King’s College London.

12. References

• Abramowitz, J. S., Deacon, B. J., & Valentiner, D. P. (2007). The Short Health Anxiety Inventory: Psychometric properties and construct validity in a non-clinical sample. Cognitive Therapy and Research, 31(6), 871–883. https://doi.org/10.1007/s10608-006-9058-1
• Beck, A. T. (1976). Cognitive therapy and the emotional disorders. International Universities Press.
• Clark, D. M. (1986). A cognitive approach to panic. Behaviour Research and Therapy, 24(4), 461–470. https://doi.org/10.1016/0005-7967(86)90011-2
• Kellner, R. (1986). Somatization and hypochondriasis. Praeger-Greenwood.
• Salkovskis, P. M. (1989). Somatic problems. In K. Hawton, P. M. Salkovskis, J. Kirk, & D. M. Clark (Eds.), Cognitive behaviour therapy for psychiatric problems: A practical guide (pp. 235–276). Oxford University Press. https://doi.org/10.1093/med:psych/9780192631770.001.0001
• Salkovskis, P. M., Rimes, K. A., Warwick, H. M. C., & Clark, D. M. (2002). The Health Anxiety Inventory: Development and validation of scales for the measurement of health anxiety and hypochondriasis. Psychological Medicine, 32(5), 843–853. https://doi.org/10.1017/S0033291702005822
• Salkovskis, P. M., & Warwick, H. M. C. (1986). Morbid fears of illness: A cognitive approach to hypochondriasis. Behaviour Research and Therapy, 24(5), 497–502. https://doi.org/10.1016/0005-7967(86)90032-X
• Warwick, H. M. C., & Salkovskis, P. M. (1990). Hypochondriasis. Behaviour Research and Therapy, 28(2), 105–117. https://doi.org/10.1016/0005-7967(90)90023-C

13. Items of the Scale (Questionnaire)