Intolerance of uncertainty and negative metacognitive beliefs as transdiagnostic mediators of repetitive negative thinking in a clinical sample with anxiety disorders

https://doi.org/10.1016/j.janxdis.2013.01.006Get rights and content

Abstract

This study aimed to replicate and extend a hierarchical model of vulnerability to worry, with neuroticism and extraversion as higher-order factors and negative metacognitions and intolerance of uncertainty as second-order factors. The model also included a transdiagnostic measure of repetitive negative thinking (RNT) and depression symptoms as outcome variables to determine whether relationships would extend beyond worry, which has traditionally been studied within the context of generalized anxiety disorder (GAD). Participants (N = 99) were referrals to a specialist anxiety disorders clinic with a principal anxiety disorder who completed a battery of self-report questionnaires assessing neuroticism, extraversion, metacognitions, intolerance of uncertainty, worry, RNT, and depression symptoms. Mediational analyses using bootstrapping provided support for transdiagnostic and diagnosis-specific mediation effects. Negative metacognitions fully mediated the relationship between neuroticism and RNT for the whole sample and for subsamples with and without GAD. Intolerance of uncertainty mediated the relationship between neuroticism and worry (for the whole sample and for those with GAD) and between neuroticism and RNT (for those with GAD). Implications for theory, treatment, and nosology are discussed.

Highlights

► A hierarchical model of vulnerability to worry and repetitive negative thinking was tested. ► Negative metacognitions and intolerance of uncertainty were evaluated as mediators. ► The model was tested in subsamples with and without generalized anxiety disorder. ► Diagnosis-specific and transdiagnostic direct and indirect effects were found.

Introduction

Neuroticism, defined as a tendency to experience negative emotional states and sensitivity to stress (Costa & McCrae, 1987), is a relatively stable personality trait associated with a range of emotional disorders (Brown et al., 1998, Watson, 2005, Watson et al., 2005). Neuroticism is a risk factor for developing anxiety and depressive disorders (Krueger, Caspi, Moffitt, Silva, & McGee, 1996), although the pathway from neuroticism into emotional disorder is not well understood. Identifying mechanisms, or second-order factors, that explain why higher-order vulnerabilities such as neuroticism are expressed as emotional disorders for some individuals but not others is important for guiding efforts at prevention and treatment of emotional disorders. This study will investigate a hierarchical model with two potential mediators between neuroticism and worry, namely intolerance of uncertainty and negative metacognitions.

Intolerance of uncertainty (IU) has recently been defined as cognitive, emotional, and behavioral reactions to uncertainty that bias information processing and lead to faulty appraisals of heightened threat and reduced coping (see Carleton, 2012, for a comprehensive review of this and earlier conceptualizations). The Intolerance of Uncertainty Model (IUM) was initially developed with reference to generalized anxiety disorder (GAD; Dugas et al., 1995, Freeston et al., 1994), which is characterized by excessive and uncontrollable worry (American Psychiatric Association, 1994). The IUM suggests that individuals with GAD find uncertainty distressing, which leads to the commencement of worrying when confronted with an uncertain or ambiguous situation (e.g., What if [something bad] happens?). There is considerable evidence that IU is a cognitive vulnerability factor for worry (Koerner and Dugas, 2008, Ladouceur et al., 2000, Sexton et al., 2003, van der Heiden et al., 2010) and an important maintaining factor for GAD (Behar et al., 2009, Dugas et al., 1998). Intervention trials have found that changes in IU are associated with improvements in worry and anxiety symptoms during cognitive behavioral therapy (Dugas and Ladouceur, 2000, Dugas et al., 2003, Dugas et al., 2010).

The Self-Regulatory Executive Function model (S-REF, Wells & Matthews, 1996) posits that positive and negative metacognitive beliefs increase engagement in repetitive negative thinking (RNT). The S-REF model suggests that positive metabeliefs about RNT being helpful motivate engagement in RNT (e.g., “worrying about my problems helps me to cope”), after which negative metabeliefs about RNT being harmful, dangerous, and uncontrollable escalate perceptions of threat from RNT and result in maladaptive attempts to control negative thoughts which, in turn, further increases engagement in RNT. Research has demonstrated associations between various forms of RNT and metacognitive beliefs (McEvoy et al., 2010, McEvoy et al., 2009, McEvoy et al., 2013, Roelofs et al., 2010), and evidence is emerging that metacognitive therapy, which aims to directly challenge positive and negative metacognitive beliefs, is associated with reductions in RNT and symptoms of anxiety and depressive disorders (Rees and van Koesveld, 2008, Wells and Colbear, 2012, Wells et al., 2012).

Building on previous research (Norton and Mehta, 2007, Norton et al., 2005, Sexton et al., 2003), van der Heiden et al. (2010) recently investigated a hierarchical model with neuroticism and extraversion as higher-order factors, IU and metacognitions as second-order factors, and worry as an outcome variable within a clinical sample with GAD. Depression symptoms were also included as an outcome variable to determine whether direct and indirect effects were specific to the core feature of GAD (i.e., worry) or similar for comorbid symptoms. Extraversion was included as a higher-order variable to test if a direct association with depression found previously would be replicated, and to identify differential relationships between the two vulnerability factors (neuroticism and extraversion) and lower-order variables. These researchers found that together negative metacognitions and IU fully mediated the relationship between neuroticism and worry, and partially mediated the relationship between neuroticism and depression symptoms. Neuroticism and extraversion also demonstrated direct relationships to depression symptoms. van der Heiden et al. found that positive metacognitive beliefs did not mediate these relationships for worry or depression symptoms.

Importantly, van der Heiden et al. (2010) noted that most variables in their hierarchical model are transdiagnostic constructs and speculated that their model may extend to other emotional disorders. IU has been found to be associated with, and to mediate, symptoms of social anxiety disorder, panic disorder and agoraphobia, generalized anxiety disorder, obsessive–compulsive disorder, and depression (Boelen and Reijntjes, 2009, Carleton et al., 2010, Carleton et al., 2012a, Dugas et al., 2001, Gentes and Ruscio, 2011, Holaway et al., 2006, Lind and Boschen, 2009, Mahoney and McEvoy, 2012, McEvoy and Mahoney, 2011, McEvoy and Mahoney, 2012). The S-REF model (Wells & Matthews, 1996) is a transdiagnostic model describing the contribution of metacognitions to engagement in the Cognitive Attentional Syndrome (CAS), which consists of RNT along with heightened self-focus, maladaptive coping behavior, and threat monitoring. The S-REF model argues that the CAS causally contributes to emotional disorder, and there is evidence that metacognitions are associated with a range of symptoms and disorders including depression (Papageorgiou and Wells, 2003, Roelofs et al., 2007), anxiety (Yilmaz, Gencoz, & Wells, 2011), GAD (Wells & Carter, 2001), social anxiety disorder (McEvoy et al., 2009), obsessive compulsive disorder and panic disorder (Cucchi et al., 2012).

van der Heiden et al. (2010) used worry as the primary outcome variable given that their focus was on the core feature of GAD. However, recent research has suggested that worry shares many features with other forms of RNT, including depressive rumination and post-event processing (McEvoy et al., 2010, Watkins et al., 2005). Watkins et al. (2005) found that worry, which is typically studied within the context of GAD, and rumination, which is commonly studied within the context of depression, were more similar than different. These forms of RNT have been found to be associated with various emotional disorders cross-sectionally, experimentally, and longitudinally (Borkovec et al., 2004, McEvoy et al., 2009, Nolen-Hoeksema, 2000, Nolen-Hoeksema and Morrow, 1993, Nolen-Hoeksema et al., 2008), and there is evidence that they are associated with symptoms of multiple emotional disorders (McEvoy and Brans, 2013, Segerstrom et al., 2000). McEvoy et al. (2010) found that after diagnosis-specific confounds were removed from the instructions and items in various well-validated measures of RNT (i.e., worry, rumination, and post-event processing) items loaded on a single RNT factor in a clinical sample with anxiety and depressive disorders. This common factor has been found to be associated with a range of emotions within an undergraduate sample (McEvoy et al., 2010) and with symptoms of various emotional disorders within a clinical sample (Mahoney, McEvoy, & Moulds, 2012). Therefore, there are strong theoretical and empirical reasons to expect that van der Heiden et al.’s hierarchical model would replicate with forms of RNT other than worry, and with emotional disorders other than GAD.

The aim of this study was to replicate and extend van der Heiden et al.’s (2010) study. We sought to replicate the relationships found between the higher-order factors of neuroticism and extraversion, the second-order factors of IU and negative metacognitions, and the outcome variables of worry and depression symptoms. van der Heiden et al. found that positive metacognitions were unrelated to the outcomes, so only negative metacognitive beliefs were considered in this study. We extended van der Heiden et al.’s study in three important ways. First, we used transdiagnostic versions of the second-order factors. Specifically, we modified the Metacognitions Questionnaire (MCQ) by substituting the term ‘worry’ in 13 items with more generic terms such as ‘thinking about my problems’, because using this diagnosis-specific term could artificially inflate the relationship between metacognitions and the outcome measure of worry in particular. Similarly, we used a more recently validated short version of the Intolerance of Uncertainty Scale (IUS-12; Carleton, Norton, & Asmundson, 2007), which has been found to have a more stable factor structure than the original version (Carleton, 2012, Carleton et al., 2007, Khawaja and Yu, 2010, McEvoy and Mahoney, 2011). Second, in addition to worry and depression symptoms we included a transdiagnostic measure of RNT as an outcome variable so that we could determine whether the mediation effects were particular to worry or were common to RNT more generally. Third, we used a mixed-diagnosis clinical sample and tested whether the hierarchical relationships would be common across subsamples with and without GAD. These extensions enabled us to identify whether IU and metacognitions were mediators of worry in patients with GAD in particular, or for RNT and emotional disorders in general. Identifying transdiagnostic mediators of RNT is crucial to guide the development of transdiagnostic treatments and to determine how current diagnosis-specific treatments could be used to effectively and efficiently treat a broader array of emotional disorders.

The primary aim of this study was to test an extended hierarchical model with neuroticism and extraversion as higher-order factors, IU and negative metacognitions as second-order factors, and RNT, worry, and depression symptoms as outcomes. Based on van der Heiden et al.’s (2010) findings, we hypothesized that (1) negative metacognitions and IU would fully mediate the relationship between neuroticism and worry and between neuroticism and RNT, (2) IU would partially mediate the relationship between neuroticism and depressive symptoms, and (3) extraversion would only have a direct relationship to depressive symptoms (see Fig. 1). Consistent with the transdiagnostic hypothesis we expected that these relationships would replicate in patients with and without a diagnosis of GAD.

Section snippets

Participants

Participants (N = 99, 53% women) were referred to a specialist anxiety disorders treatment service by General Practitioners or Psychiatrists. At the initial assessment participants completed a standard questionnaire battery and were administered the Anxiety Disorders Interview Schedule for DSM-IV (ADIS-IV; Brown, Di Nardo, & Barlow, 1994). All participants met criteria for a principal anxiety disorder. Principal diagnoses included social phobia (n = 51), panic disorder with or without agoraphobia (n

Data screening

Prior to data analyses, distributions, skewness and kurtosis were examined for scale total scores. Distributions approximated normality with all scales demonstrating acceptable levels of skewness and kurtosis (<|.69|), with the exception of the EPQ-N which just fell out of the range of −1.0 to 1.0 (skewness = −1.08 and kurtosis = 1.25). The removal of two univariate outliers resulted in the skewness (−.66) and kurtosis (−.35) being within the acceptable range for the EPQ-N. All models were run with

Discussion

This study tested a hierarchical model with neuroticism and extraversion as higher-order factors, negative metacognitions and IU as second-order factors, and RNT, worry and depression as outcome variables. The first hypothesis that the relationships between neuroticism and worry, and between neuroticism and RNT, would be fully mediated by negative metacognitions and IU was partially supported. Negative metacognitions fully mediated the relationship between neuroticism and RNT in all samples,

References (83)

  • M.J. Dugas et al.

    A randomized clinical trial of cognitive-behavioral therapy and applied relaxation for adults with generalized anxiety disorder

    Behavior Therapy

    (2010)
  • M.J. Dugas et al.

    Generalized anxiety disorder: a preliminary test of a conceptual model

    Behaviour Research and Therapy

    (1998)
  • T. Ehring et al.

    The Perseverative Thinking Questionnaire (PTQ): validation of a content-independent measure of repetitive negative thinking

    Journal of Behavior Therapy and Experimental Psychiatry

    (2011)
  • M.H. Freeston et al.

    Why do people worry?

    Personality and Individual Differences

    (1994)
  • E.L. Gentes et al.

    A meta-analysis of the relation of intolerance of uncertainty to symptoms of generalized anxiety disorder, major depressive disorder, and obsessive–compulsive disorder

    Clinical Psychology Review

    (2011)
  • R.M. Holaway et al.

    A comparison of intolerance of uncertainty in analogue obsessive–compulsive disorder and generalized anxiety disorder

    Anxiety Disorders

    (2006)
  • R. Ladouceur et al.

    Experimental manipulation of intolerance of uncertainty: a study of a theoretical model of worry

    Behaviour Research and Therapy

    (2000)
  • C. Lind et al.

    Intolerance of uncertainty mediates the relationship between responsibility beliefs and compulsive checking

    Journal of Anxiety Disorders

    (2009)
  • A.E.J. Mahoney et al.

    Psychometric properties of the repetitive thinking questionnaire in a clinical sample

    Journal of Anxiety Disorders

    (2012)
  • P. McEvoy et al.

    The post-event processing questionnaire in a clinical sample with social phobia

    Behaviour Research and Therapy

    (2006)
  • P.M. McEvoy et al.

    Achieving certainty about the structure of intolerance of uncertainty in a treatment-seeking sample with anxiety and depression

    Journal of Anxiety Disorders

    (2011)
  • P.M. McEvoy et al.

    To be sure, to be sure: intolerance of uncertainty mediates symptoms of various anxiety disorders and depression

    Behavior Therapy

    (2012)
  • P.M. McEvoy et al.

    Changes in metacognitions and post-event processing during a course of cognitive behaviour group therapy for social phobia

    Journal of Anxiety Disorders

    (2009)
  • P.M. McEvoy et al.

    Mechanisms driving pre- and post-stressor repetitive negative thinking: metacognitions, cognitive avoidance, and thought control

    Journal of Behavior Therapy and Experimental Psychiatry

    (2013)
  • T.J. Meyer et al.

    Development and validation of the Penn State Worry Questionnaire

    Behaviour Research and Therapy

    (1990)
  • C.S. Rees et al.

    An open trial of group metacognitive therapy for obsessive–compulsive disorder

    Journal of Behavior Therapy and Experimental Psychiatry

    (2008)
  • J. Roelofs et al.

    On the links between self-discrepancies, rumination, metacognitions, and symptoms of depression in undergraduates

    Behaviour Research and Therapy

    (2007)
  • C. van der Heiden et al.

    A hierarchical model for the relationships between general and specific vulnerability factors and symptom levels of generalized anxiety disorder

    Journal of Anxiety Disorders

    (2010)
  • E. Watkins et al.

    Comparisons between rumination and worry in a non-clinical population

    Behaviour Research and Therapy

    (2005)
  • D. Watson et al.

    Basic dimensions of temperament and their relation to anxiety and depression: a symptom-based perspective

    Journal of Research in Personality

    (2005)
  • A. Wells et al.

    Further tests of a cognitive model of generalized anxiety disorder: metacognitions and worry in GAD, panic disorder, social phobia, depression, and nonpatients

    Behavior Therapy

    (2001)
  • A. Wells et al.

    A short form of the metacognitions questionnaire: properties of the MCQ-30

    Behaviour Research and Therapy

    (2004)
  • A. Wells et al.

    Metacognitive therapy in treatment-resistant depression: a platform trial

    Behaviour Research and Therapy

    (2012)
  • American Psychiatric Association

    Diagnostic and statistical manual of mental disorders

    (1994)
  • A.T. Beck et al.

    Beck Depression Inventory-II manual

    (1996)
  • P.A. Boelen et al.

    Intolerance of uncertainty in adolescents: correlations with worry, social anxiety, and depression

    Journal of Nervous and Mental Disorders

    (2010)
  • T. Borkovec et al.

    Avoidance theory of worry and generalized anxiety disorder

  • T.A. Brown et al.

    Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal

    Journal of Abnormal Psychology

    (1998)
  • T.A. Brown et al.

    Anxiety disorders interview schedule for DSM-IV

    (1994)
  • T.A. Brown et al.

    Reliability of DSM-IV anxiety and mood disorders: implications for classification of emotional disorders

    Journal of Abnormal Psychology

    (2001)
  • R.N. Carleton

    The intolerance of uncertainty construct in the context of anxiety disorders: theoretical and practical perspectives

    Expert Review of Neurotherapeutics

    (2012)
  • Cited by (69)

    View all citing articles on Scopus
    View full text