Text

Background/Aim: Consensus-based prescribing tools are widely used in drug utilisation research and quality measurement to identify drug therapy problems in primary care. Yet, little is known about how consistently these tools define and operationalise medication safety risks.

Our aim is to characterise how contemporary consensus-based prescribing tools define and operationalise drug therapy problems (DTPs), and to assess the implications of tool selection for pharmacoepidemiologic measurement of prescribing quality and drug-related risk at population level.

Methods: We conducted a systematic review of consensus-based prescribing tools applicable to primary care. MEDLINE and Embase were searched for publications from 1 January 2010 to 31 July 2026, with additional reference and citation screening. Eligible studies reported the development of tools comprising explicit prescribing criteria addressing one or more drug therapy problem (DTP) domains (drug-drug, drug-disease, drug-age interactions, overprescribing, underprescribing). Paediatric, single-drug, or highly context-specific instruments were excluded. Prescribing criteria were standardised into consolidated DTP themes defined by risk context, focus drug, and clinical outcome. Themes were assigned to DTP domains and mapped to standardised clinical outcome categories. A binary theme-by-tool matrix was constructed to quantify thematic scope, convergence, and outcome coverage across tools using descriptive distributional analyses.

Results: Data analysis has not been completed. Results will be available by the beginning of the symposium.

Discussion/Conclusion: First results imply that consensus-based prescribing tools might vary markedly in their operationalisation of drug therapy problems and associated clinical outcomes. For applications in drug utilisation studies, quality measurement, and audit – the choice of tool determines which prescribing risks are measurable and which remain unobserved. An outcome-oriented, theme-based framework could enable explicit comparison of measurement scope across tools and support more transparent selection, adaptation, and interpretation in population-level research.