Evaluation of HIV drug interaction Web sites

BACKGROUND: Clinicians frequently consult HIV drug interaction Web sites of unknown quality. OBJECTIVE: To systematically review and identify HIV drug interaction Web sites of high quality and usefulness for healthcare professionals. METHODS: Relevant Web sites were identified through a structured search on commonly used search engines. An assessment tool containing 4 domains (content, reliability, access restrictions, ease of navigation) was developed. English and French Web sites were selected for review if they included HIV drug interaction information directed to healthcare professionals. Web sites were excluded if antiretroviral interaction data were not available or were out of date. Commercial online databases and sites that required payment were not included. Seventeen HIV pharmacists from across Canada participated in the review. The Web sites were ranked with total mean scores. Mean scores for each domain were then analyzed. Interrater agreement and ANOVA using the rater as a covariate were determined. RESULTS: Nine Web sites met the criteria for review. Web sites from Toronto General Hospital (Canada), HIVinSite (beta version) (US), and the University of Liverpool (UK) ranked highest for total mean scores and for content. Other Web sites were found to be reliable, accessible, and easy to navigate; however, they did not consistently include unpublished data or data on herbal preparations, recreational drugs, or multiple interactions. CONCLUSIONS: Three HIV interaction Web sites of high quality were identified that can be valuable tools for HIV and non-HIV healthcare professionals. Regular reviews are necessary in order to keep pace with the growing body of HIV interaction data and the constant evolution of Web sites.