ABSTRACT
Guidelines now discourage opioid analgesics for chronic noncancer pain because the benefits frequently do not outweigh the harms. We aimed to determine the proportion of patients with chronic noncancer pain who are prescribed an opioid, the types prescribed and factors associated with prescribing. Database searches were conducted from inception to 29 October 2018 without language restrictions. We included observational studies of adults with chronic noncancer pain measuring opioid prescribing. Opioids were categorized as weak (e.g. codeine) or strong (e.g. oxycodone). Study quality was assessed using a risk of bias tool designed for observational studies measuring prevalence. Individual study results were pooled using a random-effects model. Meta-regression investigated study-level factors associated with prescribing (e.g. sampling year, geographic region as per World Health Organization). The overall evidence quality was assessed using Grading of Recommendations Assessment, Development and Evaluation criteria. Of the 42 studies (5,059,098 participants) identified, the majority (n = 28) were from the United States of America. Eleven studies were at low risk of bias. The pooled estimate of the proportion of patients with chronic noncancer pain prescribed opioids was 30.7% (95% CI 28.7% to 32.7%, n = 42 studies, moderate-quality evidence). Strong opioids were more frequently prescribed than weak (18.4% (95% CI 16.0-21.0%, n = 15 studies, low-quality evidence), versus 8.5% (95% CI 7.2-9.9%, n = 15 studies, low-quality evidence)). Meta-regression determined that opioid prescribing was associated with year of sampling (more prescribing in recent years) (P = 0.014) and not geographic region (P = 0.056). Opioid prescribing for patients with chronic noncancer pain is common and has increased over time.
