Development and validation of standardized severity rating scale to assess the consistency of drug-drug interaction severity among various drug information resources.

BACKGROUND: The consistency in reporting the severity of drug interactions across the drug information resources is important in guiding the appropriate clinical use of drug-pairs, to minimize the associated adverse events. This necessitates the need of a standardized severity rating scale, that can accommodate the different severity ratings of the same interacting drug-pair into a reasonable severity category, that can ease the consistency assessment among different drug information resources. OBJECTIVE: To develop and validate a standardized severity rating scale that can ease the consistency assessment among the various drug information resources. METHODS: The definitions of various severity rating categories as documented in the eight drug information resources was consolidated to develop a standardized severity rating scale. Thus developed rating scale was validated using twenty commonly used drug-pairs. Fleiss' kappa score was used as an indicator for assessing overall consistency among various drug information resources, whereas, Cohen's kappa was used as an indicator of level of consistency between two drug information resources and between individual drug information resource and newly developed standardized severity rating scale. RESULTS: The newly developed standardized severity rating scale classifies the severity of drug-drug interactions into three categories namely mild, moderate and major. The Fleiss' kappa score was improved from 0.047 to 0.176, indicating improved strength of agreement [Average pairwise agreement: 16% Vs 36.7%] among various drug information resources. The average pairwise Cohen's kappa was 0.082 [Strength of agreement: poor] in original severity ratings whereas it was improved to 0.198 [Strength of agreement: almost equal to fair] in standardized severity rating scale. CONCLUSION: The newly developed standardized severity rating scale can be used as a tool to assess the consistency of severity rating categories among the various drug information resources.

Assessing Consistency of Drug-Drug Interaction-Related Information Across Various Drug Information Resources.

Background Information related to drug-drug interactions (DDIs) varies significantly from one drug information (DI) resource to another. These variations pose challenges for healthcare professionals in making the right decisions regarding using some of the drug combinations in needy patients. The objective of this study was to review eight different DI resources for scope, completeness, and consistency of information related to DDIs. Methodology A total of eight DI resources, namely, Micromedex®, Portable Electronic Physician Information Database©, UpToDate®, Medscape.com drug interaction checker, Drugs.com drug interaction checker, Stockley's Drug Interactions (ninth edition, 2010), Drug Interactions Analysis & Management: Facts and Comparisons 2014 (ninth edition, 2014), and the drug interaction appendix of the British National Formulary-76, were compared. Each DI resource was scored for scope by calculating the percentage of interactions that had an entry in each resource. A completeness score was calculated for each resource describing severity, clinical effects, mechanism, and DDI management. The consistency of the information was assessed using Fleiss Kappa (k) score estimated using ReCal3 0.1 (alpha) web service and Statistical Package for the Social Sciences version 24. Results The scope score was the highest (100%) for UpToDate® and Portable Electronic Physician Information Database©, whereas the completeness score was the highest (100%) for Drug Interaction Analysis & Management: Facts and comparisons 2014. The inter-source reliability scores among the eight different DI sources were poor (k < 0.20, p < 0.05) for documentation of information related to severity, clinical effects, mechanism, and management of DDIs. Conclusions Variations in the information cause uncertainty among healthcare professionals concerning interacting drug pairs in clinical practice. This may also increase the possibility of adverse drug outcomes when interacting drug pairs are used in at-risk patients. We recommend comprehensive preventive and management strategies for DDIs depending on a uniform scale of severity and clinical effects across various DI resources.