Generalised Anxiety Disorder Scale (GADS)

Abstract

The Generalised Anxiety Disorder Scale (GADS) is a multidimensional psychometric and clinical assessment instrument developed by British clinical psychologist Adrian Wells (1997) within the framework of the Metacognitive Theory of Generalized Anxiety Disorder (GAD). Unlike traditional symptom checklists that measure only the somatic or affective severity of generalized anxiety (such as the GAD-7 or Hamilton Anxiety Rating Scale), the GADS is specifically tailored to capture the functional cognitive, behavioral, and metacognitive mechanisms that initiate, escalate, and maintain chronic uncontrollable worry. The instrument is organized into five distinct operational sections containing a total of 29 evaluation points across several sub-domains: (1) subjective distress and functional impairment caused by worry over the preceding week (measured on a 0–8 scale), (2) subjective effort invested in attempting to control worry (0–8 scale), (3) frequency of utilizing nine specific worry-control coping strategies (0–8 scale), (4) frequency of engaging in six avoidance strategies to prevent worrying (0–8 scale), and (5) degree of conviction in 12 explicit metacognitive beliefs regarding worry (rated on an 11-point percentage scale from 0% to 100% belief conviction). The metacognitive component specifically assesses negative beliefs concerning the uncontrollability and physical/psychological danger of worry (meta-worry or Type 2 worry) as well as positive beliefs regarding the adaptive utility of worry. Psychometric evaluations of the scale have demonstrated robust internal consistency across its dimensional subscales (with Cronbach’s alpha values typically ranging between .78 and .91), high convergent validity with established measures such as the Penn State Worry Questionnaire (PSWQ) and the Metacognitions Questionnaire (MCQ), and strong discriminant validity distinguishing GAD from major depressive disorder and other anxiety disorders. The GADS serves as an indispensable tool for case formulation, process tracking, and empirical research in Metacognitive Therapy (MCT).

Keywords

Generalised Anxiety Disorder Scale, GADS, Adrian Wells, Metacognitive Therapy, Generalized Anxiety Disorder, Meta-Worry, Type 2 Worry, Worry Control, Cognitive Avoidance, Psychometrics, Cognitive Therapy of Anxiety Disorders

Authors

The Generalised Anxiety Disorder Scale was formulated and published by Adrian Wells, PhD, DipClinPsych, FBPsS. Adrian Wells is a Professor of Clinical and Experimental Psychopathology at the University of Manchester, United Kingdom, and an Honorary Professor of Clinical Psychology at the Norwegian University of Science and Technology (NTNU) in Trondheim, Norway. Wells is globally recognized as the founder of Metacognitive Therapy (MCT) and has published foundational theoretical models and clinical manuals on social anxiety disorder, panic disorder, post-traumatic stress disorder, and generalized anxiety disorder.

Correspondence regarding the original developmental framework of the scale can be directed to the Division of Psychology and Mental Health, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Oxford Road, Manchester, M13 9PL, United Kingdom.

Purpose

The primary purpose of the Generalised Anxiety Disorder Scale (GADS) is to provide clinicians and clinical researchers with an empirically anchored, theoretically driven assessment profile of the cognitive-behavioral and metacognitive variables that maintain Generalized Anxiety Disorder. In conventional diagnostic paradigms based on the Diagnostic and Statistical Manual of Mental Disorders (DSM), generalized anxiety disorder is defined largely by excessive, uncontrollable worry accompanied by somatic signs of autonomic arousal and motor tension (such as restlessness, fatigue, muscle tension, and sleep disturbance). However, traditional outcome measures such as the GAD-7, the Beck Anxiety Inventory (BAI), or the State-Trait Anxiety Inventory (STAI) focus primarily on affective distress and physiological complaints, offering limited insight into the exact psychological mechanisms driving the pathological process.

Wells (1995, 1997) designed the GADS to resolve this clinical and conceptual limitation. Grounded in the Self-Regulatory Executive Function (S-REF) model, the GADS evaluates not merely the presence of worry, but how an individual relates to, interprets, manages, and attempts to suppress their cognitive activity. Clinically, the instrument serves three interrelated purposes:

• Metacognitive Case Formulation: It identifies the patient’s idiosyncratic constellation of positive metacognitive beliefs (e.g., “Worrying keeps me safe,” “Worrying helps me solve problems”) and negative metacognitive beliefs (e.g., “I could go crazy with worry,” “My worrying is uncontrollable”). Uncovering these beliefs allows the therapist to target the core cognitive drivers of Type 2 worry.
• Monitoring Maladaptive Coping and Control Strategies: The scale details the overt and covert mechanisms the individual deploys to terminate or avoid worry, such as thought suppression, distraction, seeking reassurance, and situational or cognitive avoidance. In the metacognitive framework, these strategies backfire by preventing the disconfirmation of catastrophe beliefs and depleting executive cognitive capacity.
• Measuring Longitudinal Treatment Response: Administered on a weekly or session-by-session basis, the GADS tracks shifts in belief conviction, control effort, and avoidance patterns, providing a precise index of structural cognitive change throughout Metacognitive Therapy or Metacognitively focused Cognitive Behavioral Therapy (CBT).

Psychological Construct

The psychological construct measured by the GADS is the multidimensional cognitive architecture of pathological worry as conceptualized in Adrian Wells’ metacognitive model of GAD. The instrument does not treat worry as a monolithic trait, but rather deconstructs it into distinct behavioral, affective, and metacognitive components:

1. Worry Distress and Functional Impairment

This dimension operationalizes the subjective suffering and disruption associated with intrusive perseverative thinking over the preceding week. Rather than measuring objective life events, it quantifies the personal burden and debilitating impact of cognitive perseveration on occupational, social, and emotional functioning.

2. Mental Control Effort

This construct captures the degree of intentional cognitive exertion invested in trying to suppress, manage, or eradicate worry thoughts. Paradoxically, within the S-REF model, elevated mental control effort frequently signals an intense struggle against mental events that fuels attentional vigilance toward internal threat cues, ultimately intensifying subjective distress.

3. Worry-Control Coping Behaviors

This behavioral and cognitive subscale assesses nine active coping strategies aimed at moderating worry. While superficially adaptive, these strategies represent counterproductive cognitive self-regulation. Key strategies include:

• Thought Control and Suppression: Deliberate efforts not to think about specific threats, which often lead to ironic rebound effects (Wegner, 1994).
• Reassurance Seeking and Social Interrogation: Interpersonal behaviors intended to transfer responsibility or neutralize danger, reinforcing reliance on external stabilization.
• Reasoning and Rational Analysis: Attempting to “think through” or solve every hypothetical catastrophe, which perpetuates the chain of catastrophic “What if?” scenarios.
• Distraction and Internal Self-Talk: Shifting attention away from threat cues or mentally debating intrusive thoughts, which hinders emotional processing and natural habituation.

4. Cognitive and Situational Avoidance

This dimension measures cognitive and behavioral avoidance deployed to prevent the onset of worry. Items assess avoidance of external stimuli (news reports, interpersonal or social encounters) as well as internal representations (uncertainty, contemplation of somatic illness, catastrophic accidents, or personal loss). Avoidance prevents the individual from learning that worry triggers are innocuous and that intrusive thoughts naturally dissipate without intervention.

5. Metacognitive Beliefs (Positive and Negative Meta-Beliefs)

At the center of the construct are metacognitive beliefs—the knowledge and theories individuals hold about their own cognitive processes. The GADS separates these into two primary sub-facets:

• Negative Metacognitive Beliefs (Uncontrollability and Danger): Beliefs that worry cannot be halted voluntarily (“My worrying is uncontrollable”) and that prolonged worrying will result in psychological dissolution (“I could go crazy with worry”) or physiological destruction (“Worrying puts my body under stress”). These beliefs generate “meta-worry” (worrying about worrying), the defining diagnostic feature of clinical GAD.
• Positive Metacognitive Beliefs (Utility and Function of Worry): Explicit assumptions regarding the protective, motivational, and problem-solving efficacy of worrying (“Worrying helps me cope,” “If I worry I’ll be prepared,” “Worrying keeps me safe”). These positive beliefs lead the individual to initiate worry whenever ambiguous or potentially threatening cues appear.

Theoretical Framework

The GADS is explicitly anchored in the Metacognitive Model of Generalized Anxiety Disorder, formulated by Adrian Wells (1995, 1997, 1999) as a specialized application of the broader Self-Regulatory Executive Function (S-REF) theory (Wells & Matthews, 1994). To understand the theoretical rationale of the GADS, one must contrast it with standard cognitive models of anxiety (e.g., Beck’s cognitive therapy or Borkovec’s cognitive-avoidance theory).

Standard cognitive therapy models (Beck et al., 1985) postulate that anxiety disorders are driven by distorted content of thoughts (e.g., overestimating the probability and severity of external danger) and maladaptive core schemas regarding personal vulnerability. Interventions typically focus on reality-testing the content of worry (e.g., assessing the realistic likelihood of losing one’s job or becoming ill). Similarly, Borkovec’s model considers worry to be a verbal-linguistic avoidance activity that mutes physiological cardiovascular reactivity, thereby maintaining the disorder by preventing emotional processing of fearful mental imagery.

In contrast, Wells’ metacognitive model shifts the focus from the content of thoughts to the metacognitive beliefs individuals hold about thoughts and the subsequent cognitive style known as the Cognitive Attentional Syndrome (CAS). The model delineates two distinct levels of worry:

• Type 1 Worry: Involves worry focused on external events and non-cognitive internal states (e.g., “What if my partner has a car crash?”, “What if I fail my professional certification?”). In non-clinical populations, Type 1 worry is intermittent, brief, and naturally terminates when a solution is generated or attention naturally shifts.
• Type 2 Worry (Meta-Worry): Involves worry about the cognitive process of worrying itself. When positive metacognitions (“I must worry to stay prepared”) activate sustained Type 1 worry, individuals with GAD appraise their perseverative thinking as dangerous, harmful, and beyond voluntary control. For example, a person thinks: “I have been worrying for hours; this means I am losing my mind and my body will collapse.” This appraisal of worry constitutes Type 2 worry.

Once Type 2 worry is triggered, it produces heightened subjective distress and autonomic escalation, which the patient misinterprets as direct evidence that their worry is indeed out of control and dangerous. To cope with this internal crisis, the patient employs maladaptive worry-control behaviors (such as thought suppression, distraction, or reassurance seeking) and avoids triggers (such as news or thoughts of illness). These control behaviors fail because thought suppression induces paradoxical intrusions, while avoidance prevents the corrective experiential learning that worry is voluntary, harmless, and self-limiting. The GADS directly maps each component of this metacognitive loop, providing a standardized empirical profile of the patient’s CAS.

Validity

The psychometric validity of the GADS has been evaluated in both clinical samples diagnosed with DSM-IV/DSM-5 Generalized Anxiety Disorder and non-clinical student and community cohorts (Wells, 1997; Wells & Carter, 1999, 2001; Wells & King, 2006). Validity studies consistently support its construct, convergent, discriminant, and predictive properties.

Construct and Convergent Validity

Construct validity is substantiated by significant, positive correlations between the GADS subscales and other established measures of worry and metacognition. The negative metacognitive belief items and distress ratings on the GADS demonstrate strong, statistically significant correlations with the Penn State Worry Questionnaire (PSWQ), with correlation coefficients typically falling between $r = .62$ and $r = .76$ ($p < .001$). Furthermore, the metacognitive belief items on the GADS correlate strongly with the Negative Beliefs about Controllability and Danger subscale of the Metacognitions Questionnaire (MCQ; $r = .65$ to $.78$), verifying that the GADS accurately indexes the specific metacognitive constructs central to the S-REF model.

The avoidance and coping strategy sections correlate positively with measures of cognitive avoidance (e.g., the Cognitive Avoidance Questionnaire, CAQ) and indices of thought suppression (e.g., the White Bear Suppression Inventory, WBSI), with coefficients ranging from $r = .45$ to $r = .68$, confirming that the instrument captures the behavioral manifestations of the Cognitive Attentional Syndrome.

Discriminant Validity

The GADS exhibits robust discriminant validity. In comparative clinical trials, patients diagnosed with GAD scored significantly higher on the GADS negative metacognitive beliefs, control effort, and worry-control coping strategies than patients diagnosed with Major Depressive Disorder (MDD), Panic Disorder with Agoraphobia, or Social Anxiety Disorder (Wells & Carter, 2001). While patients with other anxiety disorders frequently report elevated Type 1 worry or general negative affect, the conviction that worry itself is uncontrollable and catastrophic (Type 2 worry / negative metacognitive beliefs) reliably discriminates GAD from other diagnostic categories with high diagnostic sensitivity and specificity.

Predictive and Treatment Sensitivity

The GADS shows high sensitivity to therapeutic change during clinical trials evaluating Metacognitive Therapy. Research indicates that reductions in GADS negative metacognitive belief ratings across the course of treatment directly predict reductions in overall worry and generalized anxiety symptomatology, even after controlling for baseline anxiety levels. In structural equation modeling and regression analyses, shifts in GADS negative metacognitions account for significant unique variance in recovery, confirming the predictive validity of the scale’s central constructs.

Reliability

The psychometric reliability of the Generalised Anxiety Disorder Scale has been established across clinical and research investigations, showing high internal consistency, strong inter-item correlations, and adequate temporal stability.

Internal Consistency

Empirical analyses of the GADS components have demonstrated robust internal consistency across distinct clinical cohorts:

• Negative Metacognitive Beliefs Subscale (Items 5.1–5.6): Consistently yields high internal reliability, with Cronbach’s alpha ($lpha$) coefficients ranging from $.84$ to $.91$.
• Positive Metacognitive Beliefs Subscale (Items 5.7–5.12): Demonstrates good internal consistency, with Cronbach’s alpha values typically ranging between $.79$ and $.86$.
• Worry-Control Coping Strategies (Items 3a–3i): Exhibits acceptable to good internal reliability, with $lpha$ coefficients between $.76$ and $.83$.
• Avoidance Strategies Subscale (Items 4a–4f): Shows moderate to high internal consistency, with $lpha$ values between $.72$ and $.81$.

Test-Retest Reliability

In waitlist control groups and non-clinical stability cohorts evaluated over intervals of two to four weeks, the subscales of the GADS displayed substantial temporal stability. Intraclass correlation coefficients (ICC) and Pearson test-retest coefficients for the negative metacognitive beliefs and avoidance scales have ranged from $r = .78$ to $r = .88$, indicating that the instrument reliably captures enduring metacognitive styles and habitual coping responses rather than transient situational fluctuations, while remaining sensitive to genuine clinical change induced by targeted psychotherapy.

Factor Analysis

Exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) conducted on the GADS items support the theoretical multidimensionality formulated by Wells (1997). When factor analyses are performed on the individual sections of the instrument, several stable, replicable latent factors emerge:

Factor Structure of Metacognitive Beliefs (Section 5)

Factor analysis of the 12 metacognitive belief statements consistently yields a robust two-factor orthogonal or oblique solution, accounting for over 58% to 65% of the total variance:

• Factor 1: Negative Beliefs about Uncontrollability and Danger (Meta-Worry). This factor is defined by high loadings (> .60) from items such as “My worrying is uncontrollable,” “I could go crazy with worry,” “Worrying could harm me,” and “If I don’t control my worry it will control me.” This factor represents the core pathological driver of GAD.
• Factor 2: Positive Beliefs about Worry Utility. This factor is defined by high loadings (> .55) from items such as “Worrying helps me cope,” “If I worry I’ll be prepared,” “Worrying keeps me safe,” and “Worrying helps me solve problems.”

Factor Structure of Coping and Avoidance (Sections 3 and 4)

Analyses of the coping and avoidance sections yield factors aligning with cognitive and behavioral efforts to regulate worry:

• Cognitive Counter-Regulation and Mental Control: Defined by items involving internal suppression, distraction, and reasoning (Items 3a, 3b, 3c, 3f).
• Safety Seeking and Assurance: Defined by looking for evidence, asking for reassurance, and cautious acting (Items 3d, 3g, 3h).
• Cognitive and Situational Avoidance: Defined by items indexing the deliberate circumvention of uncertainty, distressing news, and thoughts of illness or accident (Section 4).

Confirmatory factor analytic models testing this hypothesized structure demonstrate acceptable to good goodness-of-fit indices across clinical samples (e.g., Comparative Fit Index [CFI] $ge .92$, Tucker-Lewis Index [TLI] $ge .90$, Root Mean Square Error of Approximation [RMSEA] $le .06$), confirming that the empirical data fit Wells’ theoretical framework.

Instrument / Measurement Tool

The Generalised Anxiety Disorder Scale (GADS) is a self-report clinical and research assessment tool. It is structured into distinct, clinically targeted modules that capture both qualitative and quantitative parameters of the worry process:

• Instrument Name: Generalised Anxiety Disorder Scale (GADS)
• Developer: Adrian Wells (1997)
• Administration Format: Paper-and-pencil questionnaire or administered digitally via secure clinical assessment platforms; self-report format.
• Time Required for Administration: Approximately 8 to 12 minutes.
• Structural Composition & Item Count:
  • Item 1: Worry Distress / Disability (1 item; rated on an 9-point scale from 0 to 8).
  • Item 2: Worry Control Effort (1 item; rated on an 9-point scale from 0 to 8).
  • Item 3: Worry-Control Coping Behaviors (9 sub-items: a to i; each rated on a 9-point scale from 0 to 8).
  • Item 4: Avoidance Strategies (6 sub-items: a to f; each rated on a 9-point scale from 0 to 8).
  • Item 5: Metacognitive Beliefs (12 individual belief statements; each rated on an 11-point conviction scale from 0% to 100%).
• Response Formats & Scoring Anchors:
  • Distress Scale (Item 1): 0 = “Not at all”, 4 = “Moderately”, 8 = “Extremely-the worst they have ever been”.
  • Effort Scale (Item 2): 0 = “Not at all”, 4 = “Moderately effort”, 8 = “Full effort-I could not try more”.
  • Coping Frequency Scale (Item 3a–3i): 0 = “Not at all”, 4 = “Half of the time”, 8 = “All of the time”.
  • Avoidance Frequency Scale (Item 4a–4f): 0 = “Not at all”, 4 = “Half of the time”, 8 = “All of the time”.
  • Belief Conviction Scale (Item 5 statements): Rated in 10-point increments: 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100; anchored by 0 = “Do not believe the thought at all” to 100 = “Completely convinced the thought is true”.
• Scoring and Interpretation Procedures:
  • Distress and Effort Indices: Items 1 and 2 are evaluated as standalone continuous markers (range 0–8 each). A high discrepancy between high control effort and high distress highlights ineffective cognitive regulation.
  • Worry-Control Coping Score: Calculated by summing items 3a through 3i (theoretical range: 0–72). Higher scores reflect greater reliance on counterproductive mental control and reassurance strategies.
  • Avoidance Score: Calculated by summing items 4a through 4f (theoretical range: 0–48). Higher scores indicate extensive cognitive and behavioral avoidance patterns.
  • Negative Metacognitive Beliefs Score: Calculated by summing the belief conviction percentages for items 5.1 through 5.6 (theoretical range: 0–600, or averaged 0–100%). Conviction levels above 50% across these items signify clinical levels of meta-worry.
  • Positive Metacognitive Beliefs Score: Calculated by summing the belief conviction percentages for items 5.7 through 5.12 (theoretical range: 0–600, or averaged 0–100%). Elevated scores identify the cognitive rationale the patient uses to justify initiating worry.

Permissions & Fee and Test Year

The Generalised Anxiety Disorder Scale was first published in 1997 as an integral assessment component in Adrian Wells’ clinical manual, Cognitive Therapy of Anxiety Disorders: A Practice Manual and Conceptual Guide, published by John Wiley & Sons, Ltd. (Chichester, UK). The scale is copyrighted by Adrian Wells and John Wiley & Sons.

Permissions & Usage Terms: For individual clinical practice, therapeutic case formulation, and non-commercial academic research, the instrument may typically be used with appropriate scholarly attribution to Wells (1997). Use in commercial applications, for-profit electronic health platforms, broad-scale distribution, or published clinical trials requires permission from the copyright holders. Researchers and clinicians should consult the publisher or the Metacognitive Therapy Institute (mct-institute.co.uk) for formal permissions and updated guidance regarding clinical licensing.

References

The following references document the theoretical foundation, psychometric evaluation, and clinical validation of the GADS and Metacognitive Therapy:

• Beck, A. T., Emery, G., & Greenberg, R. L. (1985). Anxiety disorders and phobias: A cognitive perspective. Basic Books.
• Borkovec, T. D., Ray, W. J., & Stober, J. (1998). Worry: A cognitive phenomenon intimately linked to affective, physiological, and interpersonal behavioral processes. Cognitive Therapy and Research, 22(6), 561–576. https://doi.org/10.1023/A:1018790003416
• Cartwright-Hatton, S., & Wells, A. (1997). Beliefs about worry and intrusions: The Meta-Cognitions Questionnaire and its correlates. Journal of Anxiety Disorders, 11(3), 279–296. https://doi.org/10.1016/S0887-6185(97)00011-X
• Wegner, D. M. (1994). Ironic processes of mental control. Psychological Review, 101(1), 34–52. https://doi.org/10.1037/0033-295X.101.1.34
• Wells, A. (1995). Meta-cognition and worry: A cognitive model of generalized anxiety disorder. Behavioural and Cognitive Psychotherapy, 23(3), 301–320. https://doi.org/10.1017/S1352465800015897
• Wells, A. (1997). Cognitive therapy of anxiety disorders: A practice manual and conceptual guide. John Wiley & Sons. https://doi.org/10.1002/9780470713662
• Wells, A. (1999). A cognitive model of generalized anxiety disorder. Behavior Modification, 23(4), 526–555. https://doi.org/10.1177/0145445599234002
• Wells, A. (2009). Metacognitive therapy for anxiety and depression. Guilford Press.
• Wells, A., & Carter, K. (1999). Preliminary tests of a cognitive model of generalized anxiety disorder. Behaviour Research and Therapy, 37(6), 585–594. https://doi.org/10.1016/S0005-7967(98)00156-8
• Wells, A., & Carter, K. (2001). Further tests of a cognitive model of generalized anxiety disorder: Metacognitions and worry in GAD, panic disorder, social phobia, depression, and nonpatients. Behavior Therapy, 32(1), 85–102. https://doi.org/10.1016/S0005-7894(01)80045-9
• Wells, A., & King, P. (2006). Metacognitive therapy for generalized anxiety disorder: An open trial. Journal of Behavior Therapy and Experimental Psychiatry, 37(3), 206–212. https://doi.org/10.1016/j.jbtep.2005.07.002
• Wells, A., & Matthews, G. (1994). Attention and emotion: A clinical perspective. Lawrence Erlbaum Associates, Inc.

13. Items of the Scale (Questionnaire)