Monday, November 13, 2006

Web-Based Search Engines Help Diagnose Difficult Cases

Web-Based Search Engines Help Diagnose Difficult Cases

Reuters Health Information 2006. © 2006 Reuters Ltd.
Republication or redistribution of Reuters content, including by framing or similar means, is expressly prohibited without the prior written consent of Reuters. Reuters shall not be liable for any errors or delays in the content, or for any actions taken in reliance thereon. Reuters and the Reuters sphere logo are registered trademarks and trademarks of the Reuters group of companies around the world.

NEW YORK (Reuters Health) Nov 10 - Using Google to conduct web-based searches on the internet can assist in the diagnosis of difficult cases, physicians in Australia report in BMJ Online First. In fact, the investigators believe that Google may be more effective than using PubMed.

"Useful information on even the rarest medical syndromes can now be found and digested within a matter of minutes," write Dr. Hangwi Tang and Jennifer Hwee Kwoon Ng.

According to the two clinicians, based at Princess Alexandra Hospital in Brisbane, physicians and patients are more frequently using the world wide web to find answers for health problems.

After witnessing a patient's father correctly diagnose his son's condition as "Paget-von Schrotter syndrome," the two clinicians analyzed in more detail how well Google is at leading doctors to the correct diagnosis.

They obtained 26 cases published in The New England Journal of Medicine during 2005. Without reading the differential diagnosis and conclusion of each case, they selected three to five search terms from each case. The researchers then did a Google search for each case using those terms, and selected the three diagnoses that seemed to fit the symptoms and signs.

It turns out they were correct 58% of the time.

For some of the other diagnoses, Google gave a correct diagnosis that was not specific enough for the physicians to consider it to be correct. For example, the Google diagnosis was "extrinsic allergic alveolitis," instead of the more specific "extrinsic allergic alveolitis caused by Mycobacterium avium."

The authors point out that their chances of success were increased by selecting "statistically improbable phrases" and scanning no more than five pages of the search result.

Drs. Tang and Ng believe that a web search will be more effective for conditions with unique symptoms and signs. Less success will be achieved with non-specific symptoms or common diseases with rare presentations.

Google has access to more than 3 billion articles, they note, and "has far exceeded PubMed as the search engine of choice for retrieving medical articles." It will be even more valuable when it releases a new version, Google Scholar, because it will include only peer-reviewed articles.

BMJ Online First 2006.

buzz this