Predictors of Participation in a Nonpharmacological Intervention for Chronic Back Pain

Objective. Cognitive behavioral therapy for chronic pain (CBT-CP) has been identified as an evidence-based adjunct or alternative to opioid pain care. However, little is known about which patients participate in CBT-CP. This study examined predictors of enrollment in a noninferiority trial of in-person vs technology-based CBT-CP for patients with chronic back pain. Setting. A single Veterans Health Affairs (VHA) medical center. Subjects. Veterans with chronic back pain. Design and Methods. For eligible participants (N = 290), individual factors (demographics, distance from a VHA medical center, pain intensity, receipt of opioid prescription, and recruitment method) collected at trial screening were examined to identify predictors of enrollment (i.e., signed consent form). Of those who enrolled, duration of participation in the treatment portion of the study was examined. Results. Among eligible patients, 54% declined enrollment due to lack of interest. Regression analyses revealed that patients not in receipt of an opioid were more likely to enroll. The probability of being in the trial long enough to receive a "dose" of treatment (3 visits or more) was 0.76 (0.04). Conclusions. Overall, enrollment rates were low. However, most patients who enrolled in the study (102 of 134 signed consent) were retained and received a treatment dose. Patients not receiving opioids were more likely to enroll, suggesting that patients who are prescribed opioids, an important group for treatment outreach, are likely underengaged. Identifying predictors of enrollment in CBT-CP may help increase recruitment efficiency and assist in targeting patients who may benefit but are not currently interested in treatment.