Evaluation of Herbal–Drug Interaction Data in Tertiary Resources

Abstract

Purpose

To objectively evaluate various tertiary resources using a set of predetermined criteria to assess which provide the most complete, current, and accurate herbal–drug interaction information.

Methods

A search of the primary literature was conducted to gather evidence related to herbal–drug interactions. Based on the information found, six specific interactions were selected for review: St. John's wort–cyclosporine, St. John's wort–sertraline, coenzyme Q10–warfarin, ginseng–digoxin, danshen–warfarin, and kava kava–alprazolam. Thirteen resources commonly used in our drug information center were selected for evaluation: three drug interaction texts, three computer databases, and seven herbal texts. Criteria were developed to objectively evaluate the accuracy, completeness, and currentness of interaction information provided by each resource. Resources were ranked using an interaction content score, a referencing score, and an overall resource score (a sum of the interaction score and referencing scores).

Results

Interaction content scores and overall resource scores were highest for drug interaction texts and databases, whereas herbal texts generally earned lower scores. Interaction data found within the three Micromedex databases sometimes varied. Only three of the evaluated resources contained all six interactions.

Conclusion

Pharmacists should be familiar with the content and limitations of available resources for herbal–drug interactions. Based on the variability noted, it is recommended that providers use more than one of the more comprehensive resources described in this review. An evaluation of resources for herbal–drug interactions should be performed periodically as they are updated.