Microsoft Researchers Study "Cyberchondria"

Slatterz introduces us to the first major study on "cyberchondria" by Microsoft researchers (abstract, paper [PDF]). The news that it can be a bad idea to search the Internet to see if you have a terrible disease should come as no surprise. According to the NYTimes article, the syndrome has been known as "cyberchondria" since at least the year 2000 (we discussed it a few years back). It refers to increased anxiety brought on when people with little or no medical training go searching for answers to common medical complaints on the Web. The article compares cyberchondria with a phenomenon well known among second-year medical students, called "medical schoolitis." The researchers note that Web searchers' propensity to jump to awful conclusions is "basic human behavior that has been noted by research scientists for decades."

Comments

I would disagree

[TubeSteak]

The researchers note that Web searchers' propensity to jump to awful conclusions is "basic human behavior that has been noted by research scientists for decades."

In my anecdotal and limited experience, I've noticed that it is certain personality types which have a "propensity to jump to awful conclusions".

They basically resort to hysterics over things they cannot control, even before they know how [thing] is going to end.

Thats why we built MEDgle

[ashdamle]

We (myself and a couple of doctors) started MEDgle (http://www.medgle.com/ ) specifically to help people understand the possible causes for their symptoms. Hopefully this will help combat cyberchondria by giving people reasonable causes, not the most dire. And as you stated, people should always talk with their health care provider. Any feedback is much appreciated. Cheers

Re:Thats why we built MEDgle

[girlintraining]

I seem to recall sometime in about 2000 there was a report that a computer program had been designed that asked a series of questions and provided a most likely diagnosis. It was apparently better at doing this than the doctors were by a small margin. The project was axed though after a huge outcry that it would put doctors out of business, couldn't be trusted because it couldn't be medically licensed, liability issues, etc. Has anything changed since then?

It seems like a good idea -- people punching in their symptoms and getting an answer about whether to take a trip to the ER (perhaps printout in hand) or not, and with people spending hours in waiting rooms only to find out they have the common cold or just bad menstral cramps, doesn't it make sense to give people the option of entering all their data into a computer first? If we tied such a database into the admissions system, they could show up with all their insurance, contact information, and symptoms list already available for the triage nurse. I'm not advocating taking a person and their clinical experience out of the loop, but certainly there's ways to use medical data to do better targeting. I've felt the same way about pharmaceuticals and wondered why there isn't a database to track adverse reactions to drugs on a per-patient basis... If someone's tried three different anti-depressants and had a poor response to them, maybe that particular reaction projects that Drug X would be 60% more likely to be effective than Drug Y. As it is, it's often drug rhoulette(sp?) until you find one that works.

 

Re:Conclusion...

[westlake]

I'm sure the study concludes that it's a great idea to search the Internet to see if you have a terrible disease using Windows Live Search.

What the study actually found, of course, is that you are far better off using a search engine that links only to known-good sources for public health information.

The AMA. The American Cancer Society. The Center for Disease Control.

The geek shouldn't shrug off studies like these because they expose a serious problem with a general search engine like Google - and a search service that has perhaps become too much a marketing machine.