Article Text

Download PDFPDF

Guided threat focus and reappraisal was better than safety seeking behaviour for reducing fear in claustrophobia
  1. Phil Richardson, PhD,
  2. Amaryllis Holland, MA, PG Dip Psychol
  1. Tavistock and Portman NHS Trust
    London, UK

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

OpenUrlCrossRefPubMedWeb of Science

QUESTION: In people with claustrophobia, how effective are safety behaviour utilisation (SBU) and guided threat focus and reappraisal (GTR) for reducing fear during exposure?


Randomised (unclear allocation concealment*), unblinded*, controlled trial with 2 weeks of follow up.


A large university in southwestern US.


46 college students who were 18–51 years of age (mean age 20 y, 93% women) and had claustrophobia (were unable to remain in a chamber for 2 min or reported a Subjective Units of Distress level ≥50 on 2 behavioural approach tests [BATs]). Follow up was 89%.


Participants were allocated to 1 of three 30 minute exposure conditions: GTR (n=13); SBU (n=17); and exposure alone (n=12). Exposure consisted of 2 BATs: a claustrophobia chamber (BAT 1) and a tall filing cabinet (BAT 2). All participants were told that claustrophobia is driven by concerns about lack of fresh air or being trapped. The GTR group was told of the efficacy of eliminating fear by focusing on the perceived threat and using evidence to counter the threat. The SBU group was told that several safety strategies were available to them: opening a small window in the chamber, standing near the exit door, checking the latch to make sure it was unlocked, and talking with the experimenter through an intercom. The exposure alone group was told that repeated exposure to the phobic situation would help them overcome their fear.

Main outcome measures

Peak fear during exposure (scale range 0 [no fear] to 100 [very severe]) and clinically significant change (change from baseline was statistically reliable and post-treatment functioning was outside range of claustrophobic population).

Main results

Multivariate analysis of covariance was used. A treatment effect was seen for peak fear for BAT 1 (p<0.001) and BAT 2 (p<0.001) after treatment. The GTR and control groups had greater improvement from baseline than the SBU group on BAT 1 (table) and BAT 2. At 2 weeks, the SBU group had greater peak fear than the GTR group for BAT 2 (p<0.01); the main treatment effect for BAT 1 was not statistically significant (p=0.06). A treatment effect was seen for number of patients with clinical improvement after treatment for BAT 1 (table); at 2 weeks, the pattern was similar but the percentages were smaller (p<0.007). Results were similar for BAT 2 but the percentages were smaller and were not statistically significant at 2 weeks.

Guided threat focus and reappraisal (GTR), safety behaviour utilisation (SBU), or exposure alone (control) for claustrophobia


In people with claustrophobia, guided threat focus and reappraisal was better than safety behaviour utilisation for reducing fear.


The role of safety seeking behaviours in reducing initial anxiety when treating anxiety disorders has been well documented, supporting the usefulness of making safety aids available during exposure treatment. Several studies provide evidence, however, that these safety seeking behaviours can maintain anxiety in social phobia and panic disorder with agoraphobia, thereby limiting the effectiveness of exposure alone.1

The results of the study by Sloan and Telch extend the suggestion that the use of safety seeking behaviours in exposure based interventions may be counter therapeutic to an additional phobic domain: claustrophobia. This study clarifies the specific effects of safety seeking behaviours by separating them from the effects of focusing on perceived threats. The findings show the superiority of GTR over SBU for clinically significant fear reduction.

The reliability of these findings is reinforced by manipulation and attentional focus checks and the use of measures of reliable and clinically significant change.

The authors note the failure of most participants to meet DSM-IV criteria and emphasise that the results should be interpreted with caution until replicated in a clinical sample. Participants in the aforementioned study on social phobia met clinical criteria, however, and it seems probable, therefore, that the results of this study could be generalised to a clinical sample.

The study has features of both efficacy and effectiveness research. It exhibits the high internal validity and experimental control characteristic of efficacy research. Furthermore, the time spent by clients in assessment was more than double that spent in treatment, which suggests an intervention rather atypical of everyday life.

Nevertheless, the use of reliable and clinically significant change measures and the reasonable likelihood of these findings generalising beyond this sample support the clinical implications identified by the authors: that the availability of safety aids may undermine the effects of exposure; that the identification and subsequent discarding of safety strategies during exposure should be encouraged; and that fear reduction may be increased by focusing on perceived threats and reevaluating their significance.



  • Source of funding: not stated.

  • For correspondence: Dr M J Telch, The University of Texas at Austin, Austin, TX, USA. telch{at}

  • * See glossary.