Are worry, rumination, and post-event processing one and the same?: Development of the repetitive thinking questionnaire

https://doi.org/10.1016/j.janxdis.2010.03.008Get rights and content

Abstract

Accumulating evidence suggests that repetitive negative thinking (RNT) is a transdiagnostic phenomenon. However, various forms of RNT such as worry, rumination, and post-event processing have been assessed using separate measures and have almost exclusively been examined within the anxiety, depression, and social phobia literatures, respectively. A single transdiagnostic measure of RNT would facilitate the identification of transdiagnostic maintaining factors of RNT, and would be more efficient than administering separate measures for each disorder. Items from three existing measures of RNT were modified to remove diagnosis-specific content and administered to a sample of undergraduate students (N = 284). Exploratory factor analysis yielded two factors labeled Repetitive Negative Thinking and Absence of Repetitive Thinking (ART). The RNT scale demonstrated high internal reliability and was associated with anxiety, depression, anger, shame, and general distress. Moreover, the RNT scale was associated with constructs that are theoretically related to engagement in RNT, including positive and negative metacognitions, cognitive avoidance, thought suppression, and thought control strategies. The ART scale had little predictive utility. Theoretical and clinical implications are discussed.

Introduction

The seminal work of Fairburn and colleagues in the context of eating disorders prompted what has become known as a transdiagnostic approach to the maintenance of psychological disorders. Fairburn, Cooper, and Shafran (2003) developed a transdiagnostic model of eating disorders by highlighting the mechanisms that maintain eating pathology across the DSM-IV eating disorder categories, and thus departed from the more traditional disorder-focused approach. Harvey, Watkins, Mansell, and Shafran (2004) adopted and broadened this transdiagnostic perspective and sought to identify shared cognitive and behavioral processes that are linked to the persistence of a range of clinical conditions. Their systematic review of the empirical literature highlighted repetitive negative thought as one such transdiagnostic process that is characteristic of the cognitive profile of a range of psychopathologies (Harvey et al., 2004). In a more recent review, Ehring and Watkins (2008) concluded that elevated repetitive negative thought is a feature of most Axis I disorders, and cited depression, social phobia and posttraumatic stress disorder (PTSD) among their examples. In another recent review, Watkins (2008) identified key factors that determine whether repetitive thought has adaptive or maladaptive consequences; namely, (i) thought valence, (ii) the context of the repetitive thought, and (iii) the mode of processing adopted during repetitive thought.

Depressive rumination (Nolen-Hoeksema, 1991, Nolen-Hoeksema, 2004) is arguably the most studied form of repetitive thought in the clinical literature. Depressive rumination refers to a pattern of repetitive thinking in response to sad mood in which an individual analyses the causes, meanings and implications of their mood, problems and events from the past. Rumination is a core cognitive feature of clinically and residually depressed individuals, and has been linked to the duration, severity and maintenance of depressive episodes (see Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008). Longitudinal evidence has demonstrated that individuals who engage in depressive rumination have an elevated likelihood of depression onset (Nolen-Hoeksema, Morrow, & Fredrickson, 1993) and maintenance (Nolen-Hoeksema, 2000). Experimental studies confirm that rumination exacerbates depressed mood (Nolen-Hoeksema & Morrow, 1993), and causally implicate rumination in poor problem solving (Watkins & Moulds, 2005a), preferential encoding of negative self-referent material (Moulds, Kandris, Starr, & Wong, 2007), the retrieval of overgeneral memories (Watkins & Teasdale, 2001), and poor emotional processing of negative events (Watkins, 2004) in depressed and dysphoric samples.

As the hallmark diagnostic feature of generalized anxiety disorder, worry is a form of repetitive thought that has been most investigated in the context of anxiety. Worry and rumination have primarily been distinguished in terms of their temporal focus (i.e., worry is future-oriented, rumination is past-oriented; Papageorgiou & Wells, 1999). Nonetheless, there is evidence that these two types of repetitive thought share common processes (e.g., Watkins, Moulds, & Mackintosh, 2005) and that both are reported by individuals across different clinical groups (e.g., Gruber, Eidelman, & Harvey, 2008). In addition, some researchers have examined the relationship between worry and rumination by comparing participants’ responses on self-report measures that tap both types of repetitive thought. Segerstrom, Tsao, Alden, and Craske (2000) reported that worry and rumination were significantly associated with a measure of repetitive thinking, and that both types of repetitive thought were, in turn, associated with anxiety and depression (see also Fresco, Frankel, Mennin, Turk, & Heimberg, 2002). However, in other studies, worry and rumination have loaded on distinct factors (e.g., Fresco et al., 2002b, Goring and Papageorgiou, 2008). Thus, the degree to which worry and rumination may be associated with a common underlying construct of repetitive thinking requires further investigation.

According to theoretical accounts of social phobia (Clark and Wells, 1995, Rapee and Heimberg, 1997), repetitive thinking about social events is a core maintaining process. Post-event processing (PEP) has been documented in high socially anxious and socially phobic samples, following real-life social events as well as laboratory-based social-evaluative tasks, such as giving a speech (e.g., Abbott and Rapee, 2004, Kocovski et al., 2005, Lundh and Sperling, 2002, Mellings and Alden, 2000, Perini et al., 2006, Rachman et al., 2000). In line with the suggestion that rumination and worry may be captured by a common construct of repetitive thought, it is plausible that post-event processing might similarly reflect this construct. To our knowledge, the interrelationship of post-event processing and other types of repetitive thought (i.e., worry, rumination) has not been examined.

Lundh and Sperling (2002) recruited an undergraduate sample to record their experience of PEP following a distressing social situation. These researchers found that while PEP following negative-evaluative events (i.e., those involving a degree of negative evaluation or the respondent as the focus of attention) was associated with a measure of social anxiety (i.e., Social Phobia Scale), PEP following other socially evaluative situations (i.e., those involving guilt, anger, or no explicit evaluative component) was not. Interestingly, however, PEP was rated as similarly prevalent (How much did you think about the event afterwards?) and distressing (How distressing was it to think about the event?) regardless of whether all socially distressing situations or only negative-evaluative situations were considered. This finding suggests that respondents experienced distressing PEP to the same degree regardless of whether or not there was an evaluative component to the social experience. Thus, PEP appears to be common regardless of whether the core feature of social phobia is present (i.e. fear of negative evaluation), further supporting the notion that repetitive negative thinking is a common experience following distressing events.

Fehm, Schneider, and Hoyer (2007) examined the specificity of PEP for social anxiety by comparing social situations to non-evaluative phobic situations. These researchers found that while respondents reported engaging in PEP following both phobic and social situations, PEP was more intense in social situations. However, in a multiple regression analysis, PEP following a social situation significantly predicted PEP following a phobic situation after controlling for fear of negative evaluation, phobic anxiety, general anxiety, and depression. The reverse was also true. That is, PEP following a phobic situation significantly predicted PEP following social situations. The authors suggested that these findings point to a general predisposition to ruminative thinking. Interestingly, the Fear of Negative Evaluation scale was only associated with PEP following social situations, rather than phobic situations. Although the authors argued that this is evidence of the specificity of PEP for social fears, an alternative possibility is that fear of negative evaluation is simply a moderator of repetitive negative thinking. It may also be that the version of the Post-Event Processing Questionnaire used in this study was more valid for social fears compared to phobic situations. A transdiagnostic measure could shed light on this issue.

Kocovski and Rector (2008) attempted to examine the diagnostic specificity of PEP in relation to depressive rumination. Measures of social anxiety, depression and depressive rumination were administered to a sample of patients with social anxiety disorder attending group CBT. Measures of PEP were then administered during the second treatment session (with reference to the first treatment session) and again in session four (with reference to an exposure task completed in session 3). Interestingly, PEP was associated with social anxiety but not with depressive rumination (rumination in response to depressive symptoms). Moreover, the partial correlation between PEP and social anxiety controlling for depression was greater in magnitude than the partial correlation between PEP and depression when controlling for social anxiety. However, this finding is perhaps not surprising given that (a) the instructions on the RRS ask respondents to complete the items with respect to times when they feel down, depressed or blue rather than socially anxious, (b) some Ruminative Response Scale (RRS, Nolen-Hoeksema & Morrow, 1991) items include depressive symptoms, and (c) depressive rumination was measured some (unspecified) weeks before PEP. The significant partial correlation between depression and PEP when controlling for social anxiety is also consistent with a general tendency to ruminate, irrespective of content. The lack of a measure not tied to diagnosis-specific content has limited researchers’ ability to test this possibility, and then to examine the notion that common maintaining factors are at play across disorders regardless of the content of the rumination. The alternative hypothesis would of course be that the diagnosis-specific content has important implications for our understanding and treatment of various emotional disorders.

Some researchers who have investigated the association between different types of repetitive thought have administered standard self-report instruments that index the constructs of interest, and examined the degree to which responses on these measures are associated. Several researchers have conducted factor analyses of common measures of worry (typically the Penn State Worry Questionnaire, PSWQ, Meyer, Miller, Metzger, & Borkovec, 1990) and rumination (typically the RRS, Nolen-Hoeksema & Morrow, 1991) to determine whether items from these scales load on a common repetitive thinking factor or separate factors. Some studies have found that the PSWQ and RRS scale items exclusively or almost exclusively load on separate factors (e.g., Fresco et al., 2002a, Fresco et al., 2002b, Goring and Papageorgiou, 2008, Muris et al., 2004). However, there is at least one potentially important source of method variance across the scales that could explain this finding. Specifically, all items within the PSWQ include the word “worry”, whereas none of the RRS items do. At present, a validated measure of repetitive thinking that contains items not anchored to disorder-specific content remains absent from the literature. As highlighted by Ehring and Watkins (2008), an ongoing challenge for investigators in the field is the absence of such a generic measure that is independent of disorder-specific concerns and symptoms. The development of such an instrument will be critical for advancements in theory and clinical practice. The administration of a generic measure of repetitive thinking to a variety of clinical groups will permit confirmation of repetitive thought as a transdiagnostic cognitive process, and clarify its role in predicting the persistence of various pathologies. In terms of clinical practice, it will provide an instrument that can be administered to all patients, irrespective of diagnosis or the presence and nature of comorbidity.

With this absence in the extant literature in mind, our primary goal in this study was to draw on existing measures that index types of repetitive thought (namely, depressive rumination, worry, post-event processing) by using items of the RRS, PSWQ and Post-Event Processing Questionnaire-Revised (PEPQ-R, McEvoy & Kingsep, 2006) to develop a general measure of repetitive negative thinking. Critically, the strategy used by Watkins et al. (2005), that involved replacing the term “worry” with “thoughts or images”, was used prior to subjecting items to factor analysis. To our knowledge, this study was the first to attempt to remove this confound. Moreover, this study was the first to include the PEPQ-R as an additional measure of repetitive thinking, which has more recently been developed from the Post-Event Processing Questionnaire (Rachman et al., 2000) to index rumination in social phobia. The original PEPQ and the PEPQ-R have demonstrated good internal reliability and have been found to be associated with measures of social anxiety (PEPQ, Rachman et al.) and anxiety in social situations for a sample with social phobia (PEPQ-R, McEvoy & Kingsep, 2006). Our development of a new, generic measure that merged the items of the RRS, PSWQ and PEPQ-R positioned us to determine: (i) if these measures in fact tap the same underlying construct of repetitive thinking, alternatively (ii) whether they index different constructs as they apply to different disorders, or (iii) whether a combination of these applies (i.e., that one construct is common to all emotions and others are particular to each disorder). Accordingly, we instructed participants to think about an index event (i.e., ‘the last time you felt especially distressed or upset’), and to respond to modified items from the RSS, PSWQ and PEPQ-R with respect to their experience of the event after it occurred.

In addition, we explored the interrelationship of repetitive thinking and a number of theory-driven constructs that are thought to promote engagement in repetitive thinking and have also been previously linked to rumination and worry; namely, positive and negative metacognitive beliefs (Papageorgiou and Wells, 1999, Watkins and Moulds, 2005b), thought control strategies (Watkins & Moulds, 2009) and avoidance strategies (Moulds et al., 2007a, Moulds et al., 2007b). Wells and colleagues’ (Wells, 1995, Wells, 2000, Wells and Carter, 2001, Wells and Matthews, 1996) Self-Regulatory Executive Function (S-REF) model posits that metacognitive beliefs about the utility of recurrent thinking are key variables that drive its persistence. Specifically, this model suggests that positive metacognitions (e.g., worrying helps me to avoid problems in the future) motivate engagement with worry, while negative metacognitions (e.g., when I start worrying I cannotstop) maintain the ruminative cycle and result in maladaptive attempts to control thoughts. Indeed, previous research has found empirical evidence that metacognitions contribute to the development and maintenance of a number of disorders characterized by recurrent and distressing thinking, such as generalized anxiety disorder (GAD, Ruscio and Borkovec, 2004, Wells and Carter, 2001), obsessive compulsive disorder (OCD, Wells & Papageorgiou, 1998), posttraumatic stress disorder (Holeva, Tarrier, & Wells, 2001), depression (Papageorgiou & Wells, 2003), and psychosis (Morrison & Wells, 2003).

Borkovec and colleagues’ (e.g., Borkovec, Alcaine, & Behar, 2004) avoidance theory of worry suggests that worry's function is to avoid the occurrence of future catastrophes, and that it is negatively reinforced by suppression of sympathetic reactions to anxiety-provoking stimuli and the non-occurrence of the catastrophe. Moreover, worry is more superficial and abstract and thus less distressing than thoughts and images that may naturally occur, which also negatively reinforces engagement in this type of repetitive thought. On the grounds that worry is verbal rather than visual/image-based, this theory posits that worry impedes the activation of somatic/physiological arousal (Vrana, Cuthbert, & Lang, 1986). Stöber (1998) and Borkovec, Ray, and Stöber (1998) hypothesized that the abstractness, or reduced concreteness, of worry is the cognitive mechanism that limits activation of imagery. In this way, worry functions as an avoidance strategy by preventing emotional processing of the content of worries. In support of this account Sexton and Dugas (2008) found that cognitive avoidance strategies were significantly correlated with measures of worry, including the PSWQ (r = .57). Moreover, Wells and Davies (1994) reported correlations between the PSWQ and various thought control strategies (r = .39–.50). There is also evidence for the association between thought control strategies and various clinical disorders that involve repetitive thinking; for example acute stress disorder (Warda & Bryant, 1998), PTSD with co-morbid depression (Reynolds & Wells, 1999), and generalized anxiety disorder (Coles & Heimberg, 2005). Therefore, a measure of repetitive negative thinking should be associated with attempts to control or avoid negative thoughts or emotions.

Our hypotheses were as follows. First, we predicted that a factor analysis would yield one main repetitive thinking factor. Alternatively, it was plausible that that separate factors might emerge that were related to distinct emotions (e.g., depression vs anxiety). Second, we hypothesized that repetitive thinking would be associated with symptoms of depression and anxiety, as well as with specific emotions associated with the reference situation (namely, anxiety, anger, shame, and general distress). Third, we predicted that repetitive thinking would be associated with metacognitions, thought control, and cognitive avoidance strategies.

Section snippets

Participants

Participants (N = 284) were first year undergraduate psychology students (70.4% women). Mean age was 19.91 years (SD = 4.36, Range = 17–56).

Repetitive Thinking Questionnaire

The1 RTQ is a measure of repetitive negative thinking (RNT) developed from an initial pool of 44 items drawn exclusively from the PSWQ (Meyer et al., 1990), the PEPQ-R (McEvoy & Kingsep, 2006) and the RRS from the Response Styles Questionnaire (RSQ; Nolen-Hoeksema & Morrow, 1991). Consistent with previous research (e.g.,

Principal Axis Factor Analysis

Common factor analysis (i.e., principal axis factor analysis) was used to analyze the 33 candidate items derived from the PSWQ, RRS, and PEPQ-R. Oblique rotation was used because if multiple factors were derived it would was expected that they would be correlated with one another. Common factor analysis was used for consistency with previous research (e.g., Fresco, Frankel, et al., 2002). Moreover, estimates from common factor analysis tend to replicate better with confirmatory factor analysis

Discussion

Worry and rumination have traditionally been examined almost exclusively within the anxiety and depression literatures, respectively, and post-event processing has been examined within the context of social phobia. More recently, however, researchers have begun to examine the similarities and differences in the content of, and processes maintaining, repetitive thinking across emotional experiences. In order to test our first hypothesis that three measures of repetitive thinking (PSWQ, RRS, and

Acknowledgements

The authors gratefully acknowledge the assistance of Carol Yap and Alishia Williams.

References (79)

  • T.M.B. Mellings et al.

    Cognitive processes in social anxiety: The effects of self-focus, rumination and anticipatory processing

    Behaviour Research and Therapy

    (2000)
  • T.J. Meyer et al.

    Development and validation of the Penn State Worry Questionnaire

    Behaviour Research and Therapy

    (1990)
  • A.P. Morrison et al.

    A comparison of metacognitions in patients with hallucinations, delusions, panic disorder, and non-patient controls

    Behaviour Research and Therapy

    (2003)
  • M.L. Moulds et al.

    The relationship between rumination, avoidance and depression in a non-clinical sample

    Behaviour Research and Therapy

    (2007)
  • P. Muris et al.

    Mediating effects of rumination and worry on the links between neuroticism, anxiety and depression

    Personality and Individual Differences

    (2005)
  • B.O. Olatunji et al.

    The structure of worry: Relations between positive/negative personality characteristics and the Penn State Worry Questionnaire

    Journal of Anxiety Disorders

    (2007)
  • S. Rachman et al.

    Post-event processing in social anxiety

    Behaviour Research and Therapy

    (2000)
  • R.M. Rapee et al.

    A cognitive-behavioral model of anxiety in social phobia

    Behaviour Research and Therapy

    (1997)
  • J. Roelofs et al.

    Effects of neuroticism on depression and anxiety: Rumination as a possible mediator

    Personality and Individual Differences

    (2008)
  • A.M. Ruscio et al.

    Experience and appraisal of worry among high worriers with and without generalized anxiety disorder

    Behaviour Research and Therapy

    (2004)
  • K.A. Sexton et al.

    The Cognitive Avoidance Questionnaire: Validation of the English translation

    Journal of Anxiety Disorders

    (2008)
  • J. Stöber

    Worry, problem elaboration and suppression of imagery: The role of concreteness

    Behaviour Research and Therapy

    (1998)
  • G. Warda et al.

    Thought control strategies in acute stress disorder

    Behaviour Research and Therapy

    (1998)
  • E. Watkins

    Adaptive and maladaptive ruminative self-focus during emotional processing

    Behaviour Research and Therapy

    (2004)
  • E. Watkins et al.

    Positive beliefs about rumination in depression: A replication and extension

    Personality and Individual Differences

    (2005)
  • E. Watkins et al.

    Comparisons between rumination and worry in a non-clinical population

    Behaviour Research and Therapy

    (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.

    The Thought Control Questionnaire: A measure of individual differences in the control of unwanted thoughts

    Behaviour Research and Therapy

    (1994)
  • A. Wells et al.

    Relationships between worry, obsessive-compulsive symptoms and meta-cognitive beliefs

    Behaviour Research and Therapy

    (1998)
  • M.J. Abbott et al.

    Post-event rumination and negative self-appraisal in social phobia before and after treatment

    Journal of Abnormal Psychology

    (2004)
  • A.T. Beck et al.

    An inventory for measuring clinical anxiety: Psychometric properties

    Journal of Consulting and Clinical Psychology

    (1988)
  • A.T. Beck et al.

    Beck depression inventory-II manual

    (1996)
  • Borkovec, T. D., Alcaine, O. M., & Behar, E. (2004). Avoidance theory of worry and generalized anxiety disorder. In R....
  • T.D. Borkovec et al.

    Worry: A cognitive phenomenon intimately linked to affective, physiological, and interpersonal behavioural processes

    Cognitive Therapy and Research

    (1998)
  • C.A. Calmes et al.

    Repetitive thought and emotional distress: Rumination and worry as prospective predictors of depressive and anxious symptomatology

    Cognitive Therapy and Research

    (2007)
  • Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg & M. R. Liebowitz's (Eds.),...
  • M.E. Coles et al.

    Thought control strategies in generalized anxiety disorder

    Cognitive Therapy and Research

    (2005)
  • Coste, J., Fermanian, J., & Venot, A. (1995). Methodological and statistical problems in the construction of composite...
  • Cited by (238)

    View all citing articles on Scopus
    View full text