Core medication use in general practice prescriptions: A pilot study evaluating the Drug Utilization 90% Index in Irish general practice.

Abstract

Aims: The Drug Utilization 90% Index (DU90%), the number of medicines making up 90% of a doctor's prescribing, is a simple tool that can be used to describe core prescribing patterns. This research aimed to pilot the application of the DU90% in the Irish context, to investigate the relationship between the DU90% and prescriber and practice characteristics and prescribing quality.

Methods: Retrospective observational study using anonymous prescription data from a sample of Irish general practitioners (GPs). Participating GPs provided demographic details and extracted prescription data for 2018-2022 using their existing software systems. The DU90% was calculated annually at both the practice and prescriber level. Prescribing quality indicators included antibiotic, benzodiazepine prescribing rates and high-risk nonsteroidal anti-inflammatory drug prescribing. The association of the DU90% with prescriber and practice characteristics and prescribing quality indicators was explored with multilevel modelling.

Results: Thirty-eight prescribers from 22 different practices were included. The mean DU90% for prescribers was 141.5 (standard deviation 12.9) and for practices was 145.62 (standard deviation 11.87). Practices in receipt of the rural deprivation grant had a significantly lower DU90% (incidence rate ratio 0.94, 95% confidence interval 0.88-0.98). There was no evidence of an association between prescriber-level characteristics and the DU90% (sex, years qualified, number of sessions worked). There was a small positive relationship between the prescriber DU90% and total prescriptions, antibiotic and benzodiazepine prescribing rates, and higher rates of high-risk nonsteroidal anti-inflammatory drug prescriptions.

Conclusion: Applying the DU90% to Irish general practice prescriptions is feasible, revealing that GPs typically use 140 medicines in the bulk of their prescribing.