Opioid Use as a Predictor of Pain Outcomes in Iraq and Afghanistan Veterans with Chronic Pain: Analysis of a Randomized Controlled Trial

Objective Our objectives were to: 1) assess the relationship between self-reported opioid use and baseline demographics, clinical characteristics and pain outcomes; and 2) examine whether baseline opioid use moderated the intervention effect on outcomes at 9 months. Design We conducted a secondary analysis of data from the Evaluation of Stepped Care for Chronic Pain (ESCAPE) trial, which found stepped-care to be effective for chronic pain in military veterans. Setting A post-deployment clinic and five general medicine clinics at a Veteran Affairs Medical Center. Subjects In total 241 veterans with chronic musculoskeletal pain; 220 with complete data at 9 months. Methods Examination of baseline relationships and multivariable linear regression to examine baseline opioid use as a moderator of pain-related outcomes including Roland Morris Disability Questionnaire (RMDQ), Brief Pain Inventory (BPI) Interference scale, and Graded Chronic Pain Scale (GCPS) at 9 months. Results Veterans reporting baseline opioid use (n = 80) had significantly worse RMDQ (16.0 ± 4.9 vs. 13.4 ± 4.2, P < .0001), GCPS (68.7 ± 12.0 vs. 65.0 ± 14.4, P = .049), BPI Interference (6.2 ± 2.2 vs. 5.0 ± 2.1, P < .0001), and depression (PHQ-9 12.5 ± 6.2 vs. 10.6 ± 5.7, P = .016) compared to veterans not reporting baseline opioid use. Using multivariable modeling we found that baseline opioid use moderated the intervention effect on pain-related disability (RMDQ) at 9 months (interaction Beta = −3.88, P = .0064) but not pain intensity or interference. Conclusions In a stepped-care trial for pain, patients reporting baseline opioid use had greater improvement in pain disability at 9 months compared to patients not reporting opioid use.