Melanoma information on the Internet: often incomplete--a public health opportunity?

PURPOSE: To assess the accuracy and completeness of information regarding melanoma on the Internet, retrieved by use of search engines.
METHODS: The first 30 uniform/universal resource locators (URLs) from each of eight search engines using the search term "melanoma" were retrieved for evaluation of accuracy and completeness using a 35-point checklist rating system instrument. Four reviewers independently rated each of 35 sites, and one reviewer rated all 74 assessable sites. Kappa statistics were used to evaluate interrater variability.
RESULTS: A total of 74 assessable Web sites were evaluated. The remainder were inaccessible, link pages only, or duplicates. Thirty-five Web sites were each independently rated by four reviewers. The remaining 39 Web sites were each rated by one reviewer. The mean kappa statistic for all variables and all rater pairs for which a kappa could be calculated was 0.824, indicating excellent overall inter-rater reliability. The majority of Web sites failed to include complete information on general information, risk factors, diagnosis, treatment, prevention, and prognosis. Ten Web sites (14%) contained a total of 13 inaccuracies, most relatively minor.
CONCLUSION: Medical information retrieved with the search term melanoma was likely to lack complete basic melanoma information and contained inaccuracies in 14% of sites. Health care providers can help patients by recommending comprehensive and accurate Web sites for patient review, by working to create accurate and thorough Web-based health information material, and by educating patients and the public about the variability in completeness and accuracy.