The Nature of the Association between Anxiety Sensitivity and Pain-Related Anxiety: Evidence from Correlational and Intervention Studies

Abstract

High anxiety sensitivity (AS) has been associated with elevated pain-related anxiety in anxiety and pain samples. The present study investigated (a) the associations among the lower order dimensions of AS and pain-related anxiety, using a robust measure of AS, and (b) the pain-related anxiety outcomes of a telephone-delivered cognitive behavioural treatment (CBT) designed to reduce high AS. Participants were 80 anxiety treatment-seeking participants with high AS (M age = 36 years; 79% women). After providing baseline data on AS and pain-related anxiety, participants were randomly assigned to an eight-week telephone CBT or a waiting list control. At baseline, bivariate correlations showed AS physical and cognitive, but not social, concerns were significantly associated with pain-related fear and arousal but not escape/avoidance behaviours. Multiple regression revealed that after accounting for emotional distress symptoms, AS physical, but not cognitive or social, concerns uniquely predicted pain-related anxiety. Multilevel modelling showed that the AS-targeted CBT reduced pain-related anxiety and treatment-related changes in global AS and AS physical concerns mediated changes in pain-related anxiety. Results suggest that an AS-targeted intervention may have implications for reducing pain-related anxiety. Further research is needed in a chronic pain sample.

Keywords::

• anxiety sensitivity
• cognitive behavioural therapy
• pain-related anxiety

Acknowledgements

This work was supported by the Marvin Burke Endowment, Faculty of Medicine, Dalhousie University. Dr Olthuis was supported by a Canadian Institutes of Health Research Vanier Canada Graduate Scholarship.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

^* Email: [email protected].

^** Email: [email protected].

^† Email: [email protected].

^‡ Email: [email protected].

^§ Four or more measurement occasions are required for random slopes in a quadratic growth curve (Mroczek & Griffin, 2007). Given the pattern of means observed in Table 1, quadratic slopes seemed most appropriate to describe the data for ASI-3 and pain anxiety.

^¶ AS treatment outcomes have been previously reported in Olthuis et al. (in press) and are reproduced here only to test the mediating role of AS in pain-related treatment outcomes.