OBJECTIVES: We aimed to determine the rate of new persistent opioid use in New Zealand (NZ) after surgery 1^st January 2007 to 31^st December 2019.

METHODS: This was a retrospective population-based study. We included patients of any age who were admitted to any NZ hospital and underwent surgery. Patients were excluded if they had a trauma diagnosis, not dispensed opioids after surgery, had prior history of opioid use, or diagnosed with opioid misuse. Patients who died or had recurrent surgery during follow up were also excluded. All patients were followed for 365 days. The primary outcome was the rate and predictors of persistent opioid use, defined as dispensing of any opioids after surgery within seven days and an additional dispensing of opioids between 91-365 days. Predictors were assessed using multivariable regression analysis.

RESULTS: 260,726 individuals were dispensed opioids on discharge; of these the rate of persistent opioid use was 9.1% (n=23,656). Significant predictors (<0.001) included switching opioid types (adjusted odds ratio (aOR)= 3.21; 95% confidence interval (CI) 3.04-3.38), surgical specialties, higher comorbidity burden, preoperative non-opioid analgesic use (aOR=1.65; 95% CI 1.60-1.71), smoking (aOR=1.44; 95% CI 1.35-1.54), preoperative hypnotic use (aOR= 1.35; 95% CI 1.28-1.42), mental health disorders (aOR=1.29; 95% CI 1.21-1.37), female sex (aOR=1.20; 95% CI 1.16-1.24), older age, and different opioid types. Conversely, increasing surgical severity and postoperative use of non-opioid analgesics were associated with decreased odds of persistent opioid use.

CONCLUSIONS: One in 11 opioid naïve patients who were acutely exposed to opioids after surgery became persistent users. Our findings suggest that there are factors that can be modified to minimize persistent opioid use, irrespective of surgery specialty or severity. We recommend the use of a single opioid with non-opioid analgesic on discharge from hospital after surgery if an opioid is deemed necessary and to avoid unnecessary switching.