self-diagnosis by search engine

[Guest guest]

Microsoft Examines Causes of `Cyberchondria' By JOHN MARKOFF

A Microsoft study suggests that self-diagnosis by search engine leads

Web searchers to conclude the worst about what ails them.

If that headache plaguing you this morning led you first to a Web

search and then to the conclusion that you must have a brain tumor,

you may instead be suffering from cyberchondria.

On Monday, Microsoft researchers published the results of a study of

health-related Web searches on popular search engines as well as a

survey of the company's employees.

The study suggests that self-diagnosis by search engine frequently

leads Web searchers to conclude the worst about what ails them.

The researchers said they had undertaken the study as part of an

effort to add features to Microsoft's search service that could make

it more of an adviser and less of a blind information retrieval tool.

Although the term " cyberchondria " emerged in 2000 to refer to the

practice of leaping to dire conclusions while researching health

matters online, the Microsoft study is the first systematic look at

the anxieties of people doing searches related to health care,

Horvitz said.

Mr. Horvitz, an artificial intelligence researcher at Microsoft

Research, said many people treated search engines as if they could

answer questions like a human expert.

" People tend to look at just the first couple results, " Mr. Horvitz

said. " If they find `brain tumor' or `A.L.S.,' that's their launching

point. "

Mr. Horvitz is a computer scientist and has a medical degree, and his

fellow investigator, Ryen W. White, is a specialist in information

retrieval technology.

They found that Web searches for things like headache and chest pain

were just as likely or more likely to lead people to pages describing

serious conditions as benign ones, even though the serious illnesses

are much more rare.

For example, there were just as many results that linked headaches

with brain tumors as with caffeine withdrawal, although the chance of

having a brain tumor is infinitesimally small.

The researchers said they had not intended their work to send the

message that people should ignore symptoms. But their examination of

search records indicated that researching particular symptoms often

led quickly to anxiousness.

They found that roughly 2 percent of all Web queries were

health-related, and about 250,000 users, or about a quarter of the

sample, engaged in a least one medical search during the study.

About a third of the subjects " escalated " their follow-up searches to

explore serious illnesses, the researchers said.

Of the more than 500 Microsoft employees who answered a survey on

their medical search habits, more than half said that online medical

queries related to a serious illness had interrupted their day-to-day

activities at least once.

Mr. Horvitz said that in addition to his interest in creating a Web

search tool that would give more reliable answers, the research was

driven by clear memories from his medical school education of what was

often referred to as " second-year syndrome " or " medical schoolitis. "

He said he remembered " sitting on a cold seat with my legs dangling

off the examination table, " convinced that he was suffering from a

rare and incurable skin disease.

While the doctor was out of the room, Mr. Horvitz said, he took a look

at his medical chart and saw that the doctor's notes read, " is in

medical school, and he has been reading a lot. "

The researchers said that Web searchers' propensity to jump to awful

conclusions was basic human behavior that has been noted by research

scientists for decades.

In 1974, the psychologists Amos Tversky and Kahneman wrote a

seminal paper about decisions that are based on beliefs about the

likelihood of uncertain events, like the outcome of an election or the

future value of the dollar.

They said that people usually employ common sense rules to aid in

decisions. The rules can be quite useful, but they also frequently

lead to systematic errors in judgment.

The Microsoft researchers noted that reliance on the rankings of Web

search results contributes a similar bias to the judgments people make

about illness.

At the same time, Mr. Horvitz said he believed that the Web would

evolve to offer more reliable information.

In the 1990s, Microsoft researchers built a health advisory system for

pregnancy and child care. Mr. Horvitz said that in the future it would

be possible to create search engines that were able to detect medical

queries and offer advice that did not automatically make Web searchers

fear the worst.

http://www.nytimes.com/2008/11/25/technology/internet/25symptoms.html