Abstract
OBJECTIVES: To identify biopsychosocial factors associated with opioid use for neuropathic pain in adults with SCI. METHODS: We studied 283 adults with SCI who were enrolled in a collaborative project at 1 of 6 participating SCI Model System centers. Using International Spinal Cord Injury Pain Basic and Extended Data Sets, participants were asked about pain, current and prior pain medication use in the past year, and nonpharmacological pain treatment use in the past year, for up to 3 pain conditions. The relationship between active opioid use and neuropathic pain severity, adjusting for relevant clinical and demographic characteristics, was examined using logistic regression using odds ratios (OR) and 95% confidence intervals (CI). RESULTS: There were 104 opioid users and 179 nonusers. In multivariable regression models, employment was associated with a 33% reduction in opioid use compared to unemployment. Opioid users were more likely to use anti-epileptics, a combination of cannabis and anti-epileptics, or a combination of anti-epileptics and nonsteroidal anti-inflammatory drugs than nonusers. Individuals with paraplegia were almost twice as likely to use opioids compared to those with tetraplegia. After adjusting for these factors, the odds of using opioids increased 10% with every 1-point increase in reported worst neuropathic pain severity (0-10 numeric rating scale). This model has good ability to discriminate between opioid users and nonusers based on the area under the curve (AUC) of 0.74. CONCLUSIONS: We have developed a biopsychosocial base model of opioid use in adults with chronic pain after SCI with good predictive performance cross-sectionally and identified key potential risk factors for opioid use.