Don’t panic: Interpretation bias is predictive of new onsets of panic disorder

https://doi.org/10.1016/j.janxdis.2013.11.008Get rights and content

Highlights

  • A panic-related interpretation bias is predictive for new onsets of panic disorder.

  • Prediction is still found after controlling for anxiety sensitivity and fear of bodily sensations.

  • A panic-related interpretation bias does not predict remission of panic disorder.

  • Anxiety sensitivity is also a significant predictor for new onsets of panic disorder.

Abstract

Psychological models of panic disorder postulate that interpretation of ambiguous material as threatening is an important maintaining factor for the disorder. However, demonstrations of whether such a bias predicts onset of panic disorder are missing. In the present study, we used data from the Dresden Prediction Study, in which a epidemiologic sample of young German women was tested at two time points approximately 17 months apart, allowing the study of biased interpretation as a potential risk factor. At time point one, participants completed an Interpretation Questionnaire including two types of ambiguous scenarios: panic-related and general threat-related. Analyses revealed that a panic-related interpretation bias predicted onset of panic disorder, even after controlling for two established risk factors: anxiety sensitivity and fear of bodily sensations. This is the first prospective study demonstrating the incremental validity of interpretation bias as a predictor of panic disorder onset.

Introduction

A pounding heart, hot flashes, lightheadedness – these could merely be signs that one is falling in love. However, individuals suffering from panic disorder often interpret these bodily sensations as signifying an imminent catastrophe such as a heart attack. Psychological models of panic disorder postulate that such biased interpretations serve to maintain the disorder (e.g., Beck et al., 1985, Clark, 1986, McNally, 1994). In particular, these models postulate that patients suffering from a panic disorder automatically interpret bodily sensations as threatening, inciting a vicious circle that can culminate in panic.

Modifying a measure by Butler and Mathews (1983), McNally and Foa (1987) developed an Interpretation Questionnaire containing ambiguous scenarios that were panic-related or panic-unrelated. They found that patients suffering from agoraphobia and panic interpreted panic-related scenarios as threatening more often than did treated agoraphobia/panic patients and healthy control subjects. Other investigators replicated and extended these findings. Harvey, Richards, Dziadosz, and Swindell (1993) found that relative to social phobia patients and healthy control subjects, panic patients exhibited an interpretation bias specific for the ambiguous panic scenarios, whereas both anxiety groups exhibited a threatening interpretation bias for the panic-unrelated scenarios. Results of Clark et al. (1997) further clarified this issue, demonstrating that patients suffering from panic disorder are more likely to believe their (biased) interpretations compared to other anxiety patients and healthy controls. Finally, Rosmarin, Bourque, Antony, and McCabe (2009) showed that panic patients exhibited a self-referential, not a global interpretation bias for threat.

Extending this work, Teachman, Smith-Janik, and Saporito (2007) studied the role of dysfunctional panic-related interpretations by combining a scenario based assessment with a reaction time (RT) based assessment. The scenario based assessment (Brief Body Sensation Questionnaire, BBSQ; Clark et al., 1997) included ambiguous panic-related scenarios as well as ambiguous scenarios describing generally threatening situations. The RT assessment involved the Implicit Association Test (IAT; Greenwald, McGhee, & Schwartz, 1998), a computerized categorization task using RTs as indices for the strength of memory associations. The IAT results showed that panic patients, compared to healthy controls, had stronger associations of concepts related to the self and panic.1 The BBSIQ results showed a panic-related interpretation bias in patients, but not in healthy subjects.

It remains unclear whether this interpretation bias is a consequence of panic disorder or a predictor, and possible causal risk factor, for the disorder (Kraemer et al., 1997). To (partly) investigate this question, Schneider, Unnewehr, Florin, and Margraf (2002) administered the Anxiety Interpretation Questionnaire for Children (AIQ-C), based on McNally and Foa's Interpretation Questionnaire, to children of panic patients, children of parents with animal phobia, and children of healthy subjects. The AIQ-C included three types of ambiguous scenarios, i.e., descriptions of panic-related and panic-unrelated body sensations as well as animal-related situations. Results demonstrated that children of parents who suffered from a panic disorder exhibited a panic-related interpretation bias, but only after they had been primed with panic-relevant but not with panic-irrelevant material.

Another way to investigate this issue is to study people who are at risk of developing panic disorder. For example, cross-sectional and longitudinal research shows that anxiety sensitivity predicts the onset of panic attacks (e.g., Cox et al., 1991, Schmidt et al., 1997) and anxiety disorders (Schmidt, Zvolensky, & Maner, 2006). Hence, examining panic-related interpretation biases in people scoring high on anxiety sensitivity may provide valuable information. Teachman (2005) found that individuals high on anxiety sensitivity exhibited a panic-related interpretation bias (see also Richards, Austin, & Alvarenga, 2001).

Studies on patients undergoing cognitive behavior therapy (CBT) suggest that reduction in interpretation biases over the course of therapy predicts reduction in symptom severity and panic frequency (Teachman, Marker, & Clerkin, 2010). Similarly, reduction in the strength of automatic panic associations predicts symptom reduction during CBT (Teachman, Marker, & Smith-Janik, 2008). Though longitudinal, these studies could not test whether premorbid interpretation biases predict the onset of panic disorder.

Accordingly, in this study we used a prospective design to test whether a version of McNally and Foa's Interpretation Questionnaire predicted new onset of panic disorder in an epidemiologic study of young German women tested at two time points over an approximately 17-month time interval. We hypothesized that women who interpret ambiguous panic-related scenarios in a threatening manner at baseline are more likely to develop panic disorder at follow-up than are women who have benign interpretations of these scenarios at baseline. In addition, we expect that the panic-related interpretation bias retains its predictive significance, even after controlling for levels of anxiety sensitivity and for fear of bodily sensations, which are two established correlates of panic disorder.

Section snippets

Participants

Participants were 1538 German women who took part in the Dresden Predictor Study (DPS; Trumpf et al., 2010). The study involved two assessments. The baseline assessment occurred between July 1996 and September 1997, and the follow-up assessment occurred about 17-months after that (M = 16.9 months, SD = 6, range = 7–30 months). During both assessments, participants completed a diagnostic interview and a battery of self-report questionnaires including the Interpretation Questionnaire (see Trumpf et

Participant characteristics

At baseline (i.e., T1), 45 of the 1538 women met criteria for lifetime panic disorder with or without agoraphobia, and 114 women diagnosed with another lifetime psychological disorder served as a comparison group. Within this latter group, 70 suffered from a mood disorder, 19 from a somatoform disorder, 10 from a substance use disorder, and 31 from an eating disorder.2 However, this group did not include any

Discussion

To the best of our knowledge, this is the first longitudinal study to investigate whether new onsets of panic disorder are predictable by a panic-related interpretation bias. As hypothesized, analyses revealed that women who interpreted ambiguous panic-related scenarios in a threatening manner at baseline were more likely to develop panic disorder at follow-up than were women who had benign interpretations of these scenarios. Strikingly, interpretation bias for panic-related (but not other)

Acknowledgment

We would like to thank all the people who helped with this study. This research was supported by grant DLR 01EG9410 from the German Ministry of Science, Research and Education.

References (37)

  • J.G. Beck et al.

    Attention and memory for threat in panic disorder

    Behaviour Research and Therapy

    (1992)
  • D.L. Chambless et al.

    Assessment of fear in agoraphobics: the Body Sensations Questionnaire and the Agoraphobic Cognitions Questionnaire

    Journal of Consulting and Clinical Psychology

    (1984)
  • D.M. Clark et al.

    Misinterpretation of body sensations in panic disorder

    Journal of Consulting and Clinical Psychology

    (1997)
  • B.J. Cox et al.

    Situations and specific categories associated with clinical and non-clinical panic attacks

    Behaviour Research and Therapy

    (1991)
  • C. Creswell et al.

    Interpretation of ambiguity in children: a prospective study of associations with anxiety and parental interpretation

    Journal of Child and Family Studies

    (2011)
  • P.A. Di Nardo et al.

    Anxiety Disorders Interview Schedule for DSM-IV: lifetime version (ADIS-IV-L)

    (1995)
  • S. Ebert

    Zur Verarbeitung von Gefahrenreizen bei Patienten mit dem Paniksyndrom

    (1993)
  • A. Ehlers

    Angst-Sensitivitäts-Index

    (1986)
  • Cited by (0)

    View full text