Development of self-efficacy and expectancy measures for benzodiazepines Jannette M. Parr a , David J. Kavanagh b, ⁎, Ross McD. Young b , Jason P. Connor c a The University of Queensland, School of Medicine, Australia b Institute of Health & Biomedical Innovation and School of Psychology & Counselling, Queensland University of Technology, Australia c The University of Queensland, Centre for Youth Substance Abuse Research, Australia abstract article info Keywords: Benzodiazepines Expectancy Factor analysis Questionnaire Self-efficacy Self-report This study aimed to develop and assess the reliability and validity of a pair of self-report questionnaires to measure self-efficacy and expectancy associated with benzodiazepine use, the Benzodiazepine Refusal Self- Efficacy Questionnaire (BRSEQ) and the Benzodiazepine Expectancy Questionnaire (BEQ). Internal structure of the questionnaires was established by principal component analysis (PCA) in a sample of 155 respondents, and verified by confirmatory factor analyses (CFA) in a second independent sample (n =139) using structural equation modeling. The PCA of the BRSEQ resulted in a 16-item, 4-factor scale, and the BEQ formed an 18-item, 2-factor scale. Both scales were internally reliable. CFA confirmed these internal structures and reduced the questionnaires to a 14-item self-efficacy scale and a 12-item expectancy scale. Lower self-efficacy and higher expectancy were moderately associated with higher scores on the SDS-B. The scales provide reliable measures for assessing benzodiazepine self-efficacy and expectancies. Future research will examine the utility of the scales in prospective prediction of benzodiazepine cessation. © 2009 Published by Elsevier Ltd. 1. Introduction Patients experiencing persistent anxiety or insomnia are typically prescribed benzodiazepines as the first-line treatment (Vicens et al., 2006). However, ongoing prescription of these medications often then occurs, leading to development of dependence (Parr, Kavanagh, Young, & McCafferty, 2006). Consequences of long-term prescription include impaired psycho-motor performance and impaired memory (Griffiths & Weerts, 1997). Elderly users report dizziness, sleep disturbance, morning tiredness and dysphoria (Puustinen et al., 2007), and risks of hip fracture are increased (Cumming & Le Couteur, 2003). Significant attention has therefore been paid to strategies to assist cessation of benzodiazepine use. A meta-analysis of strategies for benzodiazepine cessation estab- lished that psychological approaches provided an effective adjunct to a withdrawal program managed by general practitioners, whereas adjunctive pharmacological approaches provided no additive effect (Parr, Kavanagh, Cahill, Mitchell, & Young, 2009). However, limited research has been undertaken into the role of psychological factors in benzodiazepine cessation (Belanger, Morin, Bastien, & Ladouceur, 2005). A potentially important psychological variable—self-efficacy, or conf- idence in performing tasks (Bandura, 1986)—has largely been ignored in relation to benzodiazepine cessation. Higher self-efficacy influences performance attainments via more ambitious task selection, increased effort, greater persistence in the face of setbacks, and reduced anticipatory arousal (Bandura, 1986). In consequence, it predicts achievements across a wide range of tasks (Bandura, 1982), including control of problematic substance use (Bandura, 1999; Kavanagh, Sitharthan, Spilsbury, & Vignaendra, 1999). It typically retains a sig- nificant predictive effect even after past performance is taken into account (Kavanagh, Pierce, Lo, & Shelley, 1993; Kavanagh et al., 1999). In a clinical context, self-efficacy has utility in identifying when clients require additional assistance in consolidating skills, so they have suff- icient confidence to maintain their performance gains. When the self- efficacy scale samples across contexts, it also identifies the specific situational challenges that require further attention in skills-focused intervention. The only existing measure of self-efficacy for benzodiazepine cessa- tion comprises a single-item measure about confidence in reducing consumption over the following week (Belanger et al., 2005). While that item was associated with concurrent quit status, its ability to make prospective predictions has not been reported. The restriction to a single item necessarily constricts its predictive power (Bowling, 2005), as does the absence of cues to consider the full spectrum of contextual chal- lenges respondents will face. A self-efficacy measure with multiple items and greater situational specificity is required. Other published self-efficacy measures concerning substance use do ask about confidence in resisting consumption in specific situations (Annis, 1984; Annis & Graham, 1988; Bandura, 1982; Kavanagh et al., 1993; Kavanagh et al., 1999; Oei, Hasking, & Young, 2005). These Addictive Behaviors 34 (2009) 751–756 ⁎ Corresponding author. Institute of Health & Biomedical Innovation, Queensland University of Technology, GPO Box 2434, Brisbane Qld 4001, Australia. Tel.: +617 3138 6143; fax: +61 7 3138 6030. E-mail address: david.kavanagh@qut.edu.au (D.J. Kavanagh). 0306-4603/$ – see front matter © 2009 Published by Elsevier Ltd. doi:10.1016/j.addbeh.2009.04.014 Contents lists available at ScienceDirect Addictive Behaviors