Psychometric examination of short forms from the University of Washington pain-related self-efficacy and concerns about pain item banks in patients with low back pain

Participants with low back pain (LBP) in a clinical trial (n = 241) examining nonpharmacologic treatments completed the 8-item UWCAP and 9-item UWPRSE and a numeric pain intensity rating, Oswestry Disability and Fear-Avoidance Beliefs questionnaires at baseline, 1-, 4- and 12-weeks after enrollment. Cronbach’s alpha and intraclass correlation coefficients estimated internal consistency and test–retest reliability, respectively. Floor and ceiling effects for the UWCAP and UWPRSE were examined. Concurrent validity was evaluated with univariate correlation coefficients and predictive validity with multivariate regression models. Participants were divided into categories of treatment responsiveness based on a single-item global rating measure, and UWPRSE and UWCAP change scores and standardized effect sizes were calculated in each category.

Both short forms had good internal consistency (α = 0.89–0.90) and test–retest reliability (ICC = 0.77–0.85), without substantial floor or ceiling effects. As expected, the UWCAP was positively correlated, and UWPRSE negatively correlated, with concurrent measures of pain intensity, disability and fear-avoidance beliefs. The UWCAP added to the prediction model for 4-week disability outcomes (β = 0.25, p = 0.008). Responsiveness was supported by the mean change scores and effect sizes across treatment response categories.