Evaluation of personal digital assistant software for drug interactions.

PURPOSE: The accuracy, comprehensiveness, and ease of use of drug interaction software used with personal digital assistants (PDAs) were studied. METHODS: Each program was assessed for accuracy using 40 clinically important and 40 clinically unimportant drug interaction pairs. Accuracy was scored through the summation of software sensitivity, specificity, and positive and negative predictive values. The comprehensiveness of each program was determined by the number of components in the drug interaction monograph. Time needed to identify the management of five important drug interactions defined each program's ease of use. The aggregate scores for accuracy, comprehensiveness, and ease of use were calculated. RESULTS: Scoring 777 and 756 out of a possible 800 points, iFacts and Lexi-lnteract, respectively, provided the most competent, complete, use-friendly compendia for assessment of drug interactions. Mosby's Drug Consult and Mobile Micromedex ranked third and fourth, scoring 688 and 655 points, respectively, while ePocrates Rx v. 6.0 rated seventh, with a score of 559. All drug interaction resources suffer from limitations in the quality or relevance of evidence for the interaction, an absence of identifiable patient and medication risk factors, and a lack of standardization in assigning significance to the interaction. Consequently, clinicians must interpret the importance of the interaction based on all available evidence. Discussion of such evidence was available for only iFacts and Lexi-Interact. CONCLUSION: Both iFacts and Lexi-Interact excelled as PDA pharmacopoeia for assessing drug interactions. However, clinicians should understand the limitations of all current drug interaction resources and exercise vigilance in prevention and recognition of interactions relevant to their patients.