Methodological Development and Assessing Prescribing Determinants Through Cumulative Drug Exposure in Hospitalized...

Area of Science:

• Pharmacology
• Health Informatics
• Clinical Pharmacy

Background:

• Accurate monitoring of drug exposure is crucial for preventing adverse drug reactions.
• Conventional drug exposure metrics at admission/discharge are insufficient for capturing dynamic in-hospital drug burden.
• Clinical data warehouses (CDWs) offer potential for developing more granular drug exposure measures.

Purpose of the Study:

• Introduce and evaluate two novel cumulative drug exposure metrics: CDE and CDED, computed from CDW data.
• Compare the performance of CDE and CDED against conventional drug exposure metrics.
• Develop a methodological framework for characterizing prescribing determinants in hospitalized patients.

Main Methods:

• Retrospective study utilizing the eHOP CDW at Rennes University Hospital.
• Inclusion of adult patients hospitalized for hematological malignancies.
• Analysis of polypharmacy, hyperpolypharmacy, drug-drug interactions, and potentially inappropriate medications using CDE and CDED, with statistical comparisons to conventional metrics.

Main Results:

• CDE and CDED demonstrated stronger correlations with prescribing determinants at discharge compared to admission metrics.
• Factor analysis of mixed data indicated that cumulative metrics capture drug exposure dynamics missed by conventional indicators.
• Potentially inappropriate medications showed strong correlations with both cumulative and conventional metrics.

Conclusions:

• CDE and CDED provide reproducible, scalable, and more accurate alternatives to conventional drug exposure metrics.
• These novel metrics, derived from CDW data, enhance the characterization of drug burden in hospitalized patients.
• The metrics show promise for advancing pharmacoepidemiological research and supporting clinical decision-making.