JMIR Publications

ABSTRACT

Background:

Preventing adverse drug reactions requires accurate monitoring of drug exposure throughout patient care. However, conventional metrics, measured at single timepoints such as admission or discharge, often fail to capture the dynamic and cumulative nature of inpatient drug burden, particularly in complex therapeutic settings. Improving exposure assessment is essential to better support clinical decision-making and medication safety. Clinical data warehouses (CDWs), which store detailed records of drug administration, enable the retrospective reuse of hospital data to develop more granular and dynamic measures of in-hospital drug exposure.

Objective:

This proof-of-concept study aimed to introduce and evaluate two cumulative drug exposure metrics computed from CDW : Cumulative Drug Exposure (CDE) and Cumulative Drug Exposure Density (CDED) and to compare them with conventional metrics for characterizing prescribing determinants in hospitalized patients.

Methods:

We conducted a retrospective analysis using data from the eHOP clinical data warehouse at Rennes University Hospital. The cohort included adults hospitalized for hematological malignancies between November 2020 and November 2021. For each patient, drug administration records were extracted. Four prescribing determinants were analyzed: polypharmacy (PP), hyperpolypharmacy (HPP), drug–drug interactions (DDI), and potentially inappropriate medications (PIM). Two new metrics were computed: CDE, quantifying the number of days each determinant was present, and CDED, normalizing CDE to hospital length of stay. These new metrics were compared to conventional metrics using Spearman correlations, descriptive analysis and factorial analysis of mixed data (FAMD).

Results:

Mean CDE values were 10.5 days for PP, 5.7 for HPP, 64.7 for DDI (accounting for multiple events per day), and 19.0 for PIM (in patients aged ≥65). CDED values ranged from 0.3 to 3.2 across determinants. Conventional metrics at admission were weakly correlated with cumulative exposure measures (e.g., DDI: rs = 0.04, P = .752; PP: rs = –0.04, P = .757; HPP: rs = 0.11, P = .364). Stronger, significant correlations emerged at discharge, particularly for DDI (CDE: rs = 0.44, CDED: rs = 0.46; both P < .001) and HPP. PIM showed strong significant correlations at both timepoints, especially at admission (CDE: rs = 0.73, P < .001). FAMD highlighted that cumulative metrics contributed independently to the principal components, capturing dynamics of drug exposure not reflected by conventional indicators.

Conclusions:

CDE and CDED, derived from real-time CDW data, offer reproducible and scalable alternatives to conventional metrics for characterizing drug exposure in patients hospitalized for severe conditions. By capturing the temporal patterns and intensity of prescribing determinants, these metrics provide a more accurate and nuanced assessment of in-hospital drug exposure. They represent promising tools for future pharmacoepidemiological research and clinical monitoring, particularly in settings where medication burden evolves rapidly over time.