(222) Prevalence and variations of potential hazardous opioid prescribing across the United Kingdom general practices

Background: Opioids frequently contribute to preventable adverse drug reactions, making them a significant focus for intervention. In a previous electronic audit and feedback initiative in Manchester, United Kingdom (UK), implementing prescribing indicators led to a 40% reduction in inappropriate prescribing, but opioid-specific indicators were excluded.

Objectives: This study investigated the various prevalences of potentially hazardous opioid prescribing (PHOP) across general practices and its associated factors that would inform the integration of opioid prescribing indicators into the audit system.

Methods: A cross-sectional study utilised the UK primary care database, Clinical Practice Research Datalink Aurum, focusing on adults (age ≥18 years) at risk of triggering 20 previously developed PHOP indicators. The prevalence was calculated on 1st April 2021 by dividing the numerator (patients triggering an indicator) by the denominator (patients with the potential to trigger an indicator), with 95% confidence intervals (95%CI) from a binomial distribution. Multilevel mixed-effects logistic regression models quantified variation in PHOP prevalence (intraclass correlation coefficients) without covariates and subsequently considered factors associated with PHOP (demographics, etc) in the model.

Results: Overall, 3,224,234 patients with the potential to trigger one of the indicators were included as the study cohort, and 393,119 (12.2%, 95%CI: 12.1, 12.2) of them triggered at least one indicator, with an intraclass correlation coefficient of 0.07. Of the 20 indicators, the prevalence of PHOP ranged from 74.0% to 1.5%, and the most common PHOP is “prescription of opioids for greater than six months without a concurrent laxative”. The prevalence of composite indicators, including indicators associated with prescribing co-medications, such as antidepressants and benzodiazepines, was 32.8% (95%CI: 32.7, 32.9), triggered in 264,787 patients. Female gender, increasing age, and number of medications prescribed were associated with triggering one of the indicators.

Conclusions: The audit revealed the feasibility of applying most of the 20 indicators to UK primary care prescription records. PHOP was common in primary care, particularly concerning co-medications and the lack of laxatives prescribed with opioids. Future studies should explore the feasibility of integrating PHOP indicators into the existing audit system and assess their effectiveness.