Experiential avoidance in the context of obsessions: Development and validation of the Acceptance and Action Questionnaire for Obsessions and Compulsions

Highlights

• We aimed to develop a version of the AAQ specific to obsessions and compulsions.

• We examined the factor structure, reliability, and validity of the AAC-OC.

• We explored relationships with OC symptom dimensions and related cognitions.

• AAQ-OC demonstrates a two-factor structure and strong internal consistency.

• The AAC-OC demonstrated convergent, discriminant, and incremental validity.

Abstract

The unwillingness to remain in contact with obsessions and anxiety (i.e., experiential avoidance, EA) may explain how normally occurring unwanted intrusive thoughts develop into clinical obsessions as seen in obsessive-compulsive disorder (OCD). Studies examining the relationship between EA and OC symptoms are mixed, potentially because the existing self-report measure of EA (i.e., the Acceptance and Action Questionnaire, AAQ-II) is a general measure that does not adequately capture EA specific to obsessions and compulsions. Thus, we aimed to develop and evaluate an OC-specific version of the AAQ-II. First, we used exploratory factor analysis to empirically reduce an initial pool of 49 items (adapted from original AAQ-II items to reference “intrusive thoughts”) to 13 items. A two-factor solution (Valued Action and Willingness) provided the best fit to the data, accounting for 60.57% of the variance. Second, the reduced AAQ-OC was administered, along with other self-report measures, to an independent sample of adults. The AAQ-OC subscales evidenced good internal consistency as well as convergent, discriminant, and incremental validity. Future work examining the psychometric properties of the AAQ-OC in a clinical sample, as well as the measure's treatment sensitivity are needed.

Introduction

Obsessive–compulsive disorder (OCD), a chronic condition that ranks in the top ten causes of health-related disability worldwide (World Health Organization, 2008), consists of unwanted intrusive thoughts, images, impulses, doubts, or fears that are seemingly uncontrollable and anxiety-provoking (i.e., obsessions) and observable and mental rituals performed to neutralize the anxiety that arises from such thoughts (i.e., compulsions; American Psychiatric Association, 2013). Although OCD as a diagnostic entity only affects 2% of individuals in their lifetime (Ruscio, Stein, Chiu, & Kessler, 2010), between 80% and 99% of the general population experiences unwanted intrusive thoughts similar in content to clinical obsessions (e.g., Belloch, Morillo, Lucero, Cabedo, & Carrió, 2004; Radomsky et al., 2014). Non-clinical and clinical obsessions are associated with the same developmental and maintenance factors (for a review see Abramowitz et al., 2014), suggesting that intrusive thoughts occur along a continuum – differing quantitatively in severity, but not qualitatively in nature. Accordingly, identifying malleable risk and maintenance factors that explain how pre-clinical unwanted intrusive thoughts develop into clinically-significant OCD symptoms can inform the prevention and treatment of this burdensome problem.

Experiential avoidance (EA; i.e., psychological inflexibility) is one such candidate process (Grayson, 2013, Twohig, 2009, Twohig et al., 201), defined as the unwillingness to remain in contact with internal experiences (i.e., thoughts, feelings, or physical sensations) that are perceived as negative (Hayes et al., 2006, Hayes et al., 1996). For instance, a woman with OCD and elevated EA might literally interpret an intrusive image of her husband being harmed in a car accident as a sign of true danger and take steps to remove the unwanted image from her mind to regulate her distress. These attempts to control or “push away” unwanted internal experiences may temporarily decrease anxiety, but are ultimately ineffective at long-term anxiety reduction as attempts to minimize anxiety and obsessions become distressing themselves (Twohig, 2009). Furthermore, these time-consuming efforts can interfere with life functioning, as EA invariably leads to actions that are inconsistent with one's values (e.g., the woman's relationship with her husband may become strained due to excessive reassurance seeking).

Despite the theoretical relevance of EA to OC symptoms, findings from existing empirical studies are equivocal and difficult to interpret (Abramowitz et al., 2009, Blakey et al., 2016, Manos et al., 2010, Reuman et al., 2016, Wetterneck et al., 2014). One explanation for these discrepant findings (as noted by authors of these studies), is that the most widely used self-report measure of EA – the Acceptance and Action Questionnaire (AAQ-II; Bond et al., 2011) – assesses this construct in a general way (e.g., “I’m afraid of my feelings”) and does not capture the construct of EA as it specifically relates to obsessional thoughts. Thus, sole reliance on the AAQ-II is a barrier to effectively understanding the relation between EA and OC symptoms.

This limitation has been recognized by researchers studying other domains of psychopathology who have developed disorder-specific versions of the AAQ-II for body image concerns (Sandoz, Wilson, Merwin, & Kellum, 2013), psychotic symptoms (Shawyer et al., 2007), trichotillomania (Houghton et al., 2014), substance use (Luoma, Drake, Kohlenberg, & Hayes, 2011), and social anxiety (MacKenzie & Kocovski, 2010). These content-specific versions have demonstrated strong psychometric properties including incremental validity over the general AAQ-II in predicting disorder-specific symptom severity (e.g., Houghton et al., 2014; Lillis & Hayes, 2008; Lundgren, Dahl, & Hayes, 2008; MacKenzie & Kocovski, 2010), thus advancing the study of EA in these areas. It would also be desirable to have a specific measure of EA in the context of obsessions and compulsions to monitor patient progress in treatments aimed to enhance psychological flexibility, such as Acceptance and Commitment Therapy (ACT),¹ which demonstrates promising efficacy for the treatment of anxiety and OCD (Twohig et al., 2018; For a review, see Bluett, Homan, Morrison, Levin, & Twohig, 2014).

Accordingly, the aims of the current study were to (a) develop a measure of EA specific to obsessions and compulsions, the AAQ-OC, and (b) evaluate its psychometric properties in an unselected sample with a range of unwanted intrusive thoughts. We hypothesized that the AAQ-OC would be significantly positively associated with purportedly similar measures of OCD symptoms and beliefs (i.e., convergent validity) more so than with other psychological symptoms (e.g., depression, social anxiety symptoms; i.e., discriminant validity). Second, we hypothesized that the AAQ-OC would account for significant variance in OC symptom severity above and beyond the non-specific AAQ-II (i.e., incremental validity).