A reference set of clinically relevant adverse drug-drug interactions

Abstract

The accurate and timely detection of adverse drug-drug interactions (DDIs) during the postmarketing phase is an important yet complex task with potentially major clinical implications. The development of data mining methodologies that scan healthcare databases for drug safety signals requires appropriate reference sets for performance evaluation. Methodologies for establishing DDI reference sets are limited in the literature, while there is no publicly available resource simultaneously focusing on clinical relevance of DDIs and individual behaviour of interacting drugs. By automatically extracting and aggregating information from multiple clinical resources, we provide a scalable approach for generating a reference set for DDIs that could support research in postmarketing safety surveillance. CRESCENDDI contains 10,286 positive and 4,544 negative controls, covering 454 drugs and 179 adverse events mapped to RxNorm and MedDRA concepts, respectively. It also includes single drug information for the included drugs (i.e., adverse drug reactions, indications, and negative drug-event associations). We demonstrate usability of the resource by scanning a spontaneous reporting system database for signals of DDIs using traditional signal detection algorithms.

Fig. 1

Data analysis workflow to generate positive and negative controls from DDI online resources with associated evidence for their component drugs. (1) DDI and single-drug online resources data (a) are extracted and stored to separate tables (b,c). (2) Drug names are normalized (d). (3) The intersection of the DDI online resources is extracted to a different table (e) and (4) English language text descriptions (for DDIs and single-drug ADRs) are annotated for AEs (after drug name masking) (f). (A) DDI pairs from (e) are assigned AEs based on the description mappings (g). Positive controls are published in Data Record 1. (B) Negative controls are generated using drugs and AEs from (g), ensuring that drug pairs cannot be found in (b) or in PubMed using a customized query (h). Negative controls are published in Data Record 2. (C) The filtered set of drugs from (d) is linked to AE and indication concepts using available evidence from (c) and negative controls for single drugs are generated following a similar process to the one described above for DDIs (i.e., drug-event is not an ADR mentioned in (c) or in PubMed using a customized query) (i). ADRs, indications and negative controls for single drugs are published in Data Record 3. Data Records 4 and 5 contain mappings of drug names and text descriptions for events, respectively. Column headers appear in grey font next to each Data Record box. Sample records from (b), (c) and (e) can be found in the bottom part of the figure.