The "proxy" measures used to determine the quality of a Web site's health information "do little to explain the comprehensiveness and accuracy of the information," according to a study from the Center for Information Therapy. The study found "enormous variation" in the quality of diabetes information on the Internet and concluded that evaluating the content of each individual Web site may be necessary to provide "meaningful guidance to consumers."
Researchers evaluated 90 diabetes Web sites using a list of recommended site characteristics generated from the American Diabetes Association's practice guidelines. Each site was scored based on these guidelines. Quality scores ranged from 14% to 97% among the sites, with an average score of 56%. A quarter of the sites ranked below 41%, and another quarter ranked above 70%. The study found that proxy measures, such as the site's sponsor and disclosure of the sources, "alone are not sufficient to guide consumers to high-quality Web sites." It also found that sponsorship characteristics — whether a site is for-profit, government-sponsored, academic or accepts advertising — were not significantly associated with differences in information quality.
"Structural measures do not correlate highly with performance in the realm of health information quality," the study found. Independent review bodies, such as URAC, should supplement their proxy measures for online health information with evaluations of accuracy and comprehensiveness, the researchers concluded. Objective content reviews can provide a "tangible and useful tool in navigating the online health universe," the study found (Seidman et al., 6/24).