Herbal medicine use disclosure, database-flagged potential herb-drug interactions, and inter – interaction database concordance among patients with non-communicable diseases in Vietnam: A multicenter cross-sectional study

Abstract

Background The widespread use of herbal medicines in clinical practice, particularly in Southeast Asia, presents significant challenges for medication safety management, notably regarding interactions between herbs and conventional drugs. Current drug interaction screening tools demonstrate substantial gaps in documenting herb-drug interactions, especially for regionally specific medicinal plants, creating potential risks for patients receiving concurrent therapies. Aim (1) to identify the prevalence of potential herb–drug and herb–herb interactions and the disclosure rate of HM use in patients with NCDs; and (2) to assess the consistency of interaction information across commonly used drug interaction databases. Methods An observational study enrolled 658 patients with non-communicable diseases from two Vietnamese tertiary care facilities. Structured interviews captured herbal medicine utilization patterns, disclosure practices, and specific botanical preparations used. Potential herb-drug and herb-herb interactions were systematically evaluated using four databases: Micromedex®, UpToDate Lexicomp Drug Interactions, Medscape Drug Interaction Checker, and Stockley’s Herbal Medicines Interactions. Inter-database agreement was assessed using Fleiss’ kappa statistics. Results 48.6% of participants reported active herbal medicines use, with 96% failing to disclose this to their healthcare providers. Potential interactions were identified in 31.3% of herbal medicine users, representing 15.2% of the total cohort. Database concordance was remarkably poor, with only 0.7% of interactions consistently documented across all four resources (Fleiss’ κ = –0.0653; p < 0.001). Multiple indigenous medicinal plants commonly used by patients were absent from all evaluated databases. Conclusion The substantial prevalence of undisclosed herbal medicine use, alongside inadequate representation in drug interaction databases, represents a critical gap in medication safety infrastructure. These findings demonstrated the necessity for integrating herbal medicine assessment into routine care, enhanced provider training, and expansion of clinical decision support systems to include region-specific botanical data for populations using concurrent traditional and conventional therapies.