Drug-Event Pairs as Indicators for the Detection of Adverse Drug Reactions during Hospitalization in Routinely...

Area of Science:

• Pharmacovigilance
• Clinical Informatics
• Health Services Research

Background:

• Adverse drug reactions (ADRs) significantly contribute to patient morbidity and mortality in hospitals.
• Effective identification of ADRs is essential for prevention, surveillance, and research.
• Electronic health record (EHR) data offers potential for ADR detection but requires validated indicators.

Purpose of the Study:

• To assess the likelihood of specific drugs causing clinically important inpatient adverse events.
• To develop a validated list of drug-event pairs as indicators of ADRs within EHR data.
• To facilitate ADR identification in clinical practice, surveillance, and research.

Main Methods:

• A consensus process using the RAND/UCLA Appropriateness Method was employed for 14 ADRs.
• Experts rated the causal link between drugs and adverse events on a 4-point Likert scale.
• Drug-event pairs were categorized based on median ratings and expert agreement.

Main Results:

• Of 255 evaluated drug-event pairs, 2 (1%) were validated as indicators of certain ADRs.
• 42 pairs (16%) were validated as indicators of probable ADRs.
• 137 pairs (54%) indicated possible ADRs, and 74 (29%) indicated unlikely ADRs.

Conclusions:

• A content-validated set of indicators for clinically important inpatient ADRs was established.
• These indicators can enhance clinical decision support, quality surveillance, and research outcomes.
• Implementation in German university hospital EHRs aims to determine ADR prevalence and improve resource allocation.