In response to a column in The New York Times last week (http://www.nytimes.com/2007/05/24/fashion/24Cyber.html), I wrote the following letter to the editor (for which they apparently aren’t publishing any letters):

Regarding “Visits to Doctors Who Are Not in, Ever” (May 24), democratization of information has made health content widely available—some would describe this a blessing and others a curse. The question is how do we steer people to information that is accurate, comprehensive, relevant, understandable, and useful?

Empirical research we’ve conducted on diabetes Web sites suggests that the guidance offered by most “experts”—such as a “trustworthy” sponsor, the currency of the content, or the process for updating it—provides little insight into the content’s actual accuracy or comprehensiveness.

Innovative collaborative filtering approaches can be effective, but they also have limitations—in part due to the complexity of medical science and its pace of change. Objective, systematic instruments now exist that could be scaled to provide critical guidance to consumers. These two approaches can complement each other and help people navigate life-and-death matters for themselves and their loved ones.

Since I get more than 150 words here, I’ll expand a bit…. First some quick background: The research I refer to is summarized in a white paper on “The Mysterious Maze of the World Wide Web” on the www.ixcenter.org Web site and the more scientific papers are available from the peer-reviewed Journal of Medical Internet Research (www.jmir.org).

It’s reasonable to approach the evaluation of Web site information quality like many other scientific questions where the answers (to “what is high-quality content?”) are not always straightforward. We can triangulate by using multiple methods to answer the question.

Specifically–and as I proposed in the original research–we can combine three kinds of measurement: structural measures of quality to give a generic content threshold test; performance measures of accuracy and comprehensiveness for specific conditions; and assessment of consumers’ perspectives on content in terms of functionality, understandability and overall utility.

In my original work (before the “Health 2.0″ space existed) , I had envisioned this being done more though random consumer survey methods, but the collective filtering approach is considerably more efficient, even if less scientific.

Bringing these different strategies together will provide everybody with a lot more guidance about health information quality, the first ingredient for good information therapy.

–Josh

This entry was posted on Wednesday, May 30th, 2007 at 9:31 pm and is filed under Health 2.0 Space, Information Quality, Market Rewards/Reimbursement Issues, Standards Development, The New Health Care Consumer. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.