Co-prescription of gabapentinoids and opioids among adults with and without osteoarthritis in the United Kingdom between 1995 and 2017.

To produce national and regional estimates and trends for gabapentinoid-opioid co-prescribing rates in patients with OA, both in absolute terms and relative to matched controls without OA. Using the UK Clinical Practice Research Datalink database we first constructed age-sex-practice-date 1:1 matched cohorts of patients aged ≥40 years with and without a new diagnosis of OA between 1995-2017 and estimated the relative incidence of a first gabapentinoid prescription. Incident gabapentinoid users in both cohorts were followed to estimate and compare the event rate of gabapentinoid-opioid co-prescription (prescription from both classes within the same 28-day window). The incidence of first gabapentinoid prescription was 3-fold higher in patients with OA than in matched controls [n = 215 357; incidence rate ratio (IRR) 2.93; 95% CI: 2.87, 3.00]. Among incident gabapentinoid users with OA (n = 27 374, median follow-up 3.9 years) the event rate of gabapentinoid-opioid co-prescription was 4.03 (4.02-4.05) per person-year. The rate was higher in OA patients classed as long-term gabapentinoid users (6.24; 6.22-6.26). These rates were significantly higher than in incident gabapentinoid users without OA [adjusted-IRR: 1.29 (1.28-1.30)]. This elevated risk was observed across age, sex, geographic regions, and calendar years, when restricted to strong opioids and to long-term gabapentinoid users, and when co-prescription was defined as within 14 days and same-day prescribing. Patients with OA not only have a higher risk of being prescribed a gabapentinoid but, once prescribed a gabapentinoid, are also at greater risk of opioid co-prescription. Strict restriction of gabapentinoid-opioid co-prescription, and improved access to, and uptake of, effective non-pharmacological and surgical alternatives for OA are required.

Informations de copyright

© The Author(s) 2020. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: [e-mail].