LAUSR

OBJECTIVE To quantify opioid use in knee and hip osteoarthritis (OA) patients, and to estimate the proportion of opioids in the population attributable to OA patients. DESIGN Population-based cohort study. METHODS We included 751,579 residents in southern Sweden, aged ≥35 years in 2015. Doctor-diagnosed knee or hip OA between 1998 and 2015 was the exposure. Dispensed weak and strong opioids were identified between November 2013 and October 2015 from the Swedish Prescribed Drug Register (SPDR). We determined age- and sex-standardized 12-month period prevalence of opioid use from November 2014 until October 2015 and calculated prevalence ratios and incidence rate ratios adjusted for age, sex, and other socio-demographic variables. We estimated the population attributable fraction (PAF) of incident opioid use attributable to OA patients. RESULTS The 12-month prevalence of opioid use among OA patients was 23.7% [95% confidence intervals (CI) 23.3-24.2], which was two-fold higher compared to individuals without knee or hip OA: prevalence ratio: 2.1 [95% CI 2.1-2.1]. Similarly, OA patients were more likely to have an incident opioid dispensation, especially for strong opioids (incidence rate ratio: 2.6 [95% CI 2.5-2.7]). Population attributable tractions (PAF) of incident opioid use attributable to OA patients was 12%, 9% for weak and 17% for strong opioids. CONCLUSIONS Every fourth patient with knee or hip OA has opioids dispensed over a 1-year period, and 12% of incident opioid dispensations are attributable to OA and/or its related comorbidities. These results highlight that patients with knee and hip OA constitute a group of patients with an alarmingly high use of opioids.