Prevalence and predictors of persistent post-surgical opioid use: a prospective observational cohort study.

Post-surgical opioid prescribing intended for the short-term management of acute pain may lead to long-term opioid use. This study was undertaken to determine the prevalence of persistent post-surgical opioid use and patient-related factors associated with post-surgical opioid use. One thousand and thirteen opioid-naïve patients awaiting elective surgery in a tertiary private hospital in Sydney were enrolled. Preoperatively, patients completed a questionnaire comprising potential predictors of persistent post-surgical opioid use. Patients underwent surgery with routine perioperative care, and were followed up at 90 to 120 days after surgery to determine opioid use. Factors associated with opioid use were assessed with logistic regression. We had an overall response rate of 95.8% (n=970) of patients, of whom 10.5% (n=102) continued to use opioids at >90 days after surgery. On surgical subtype analysis, the prevalence of persistent opioid use was 23.6% after spinal surgery, and 13.7% after orthopaedic surgery. Four factors were independently associated with persistent post-surgical opioid use in a multivariate model: having orthopaedic (odds ratio [OR] 4.6, 95% confidence interval [CI] 2.0 to 10.8, <0.001) or spinal surgery (OR 4.0, 95% CI 1.7 to 9.2, <0.001), anxiety (OR 2.1, 95% CI 1.1 to 4.1, =0.03), attending pre-admission clinic (OR 3.7, 95% CI 1.6 to 8.6, =0.002), and higher self-reported pain score at >90 days after surgery ( <0.001). More than 10% of opioid-naïve patients undergoing elective surgery experience persistent post-surgical opioid use. Identification of factors associated with persistent post-surgical opioid use may allow development of a risk stratification tool to predict those at highest risk.