Friday, November 10, 2006

Googling for a diagnosis

This article will no doubt feature highly in the blogs (click here) and create much controversy! It reminded me of the Isabel decision support tool.

Basically, the authors took a years worth (n=26) of diagnostic cases from the NEJM and, after reading the article, they selected 3-5 pertinent search terms. Added these to google and somehow selected the 3 most likely diagnoses. They then compared these 3 diagnoses with the actual result from the NEJM.

Overall result - 'correct' in 58% of cases (CI - 38-77%).

I'm not sure if I find this disturbing or not, it's certainly too small to make conclusions. But once again Google (not deliberately on their part) have set a standard.....

4 comments:

Dean Giustini said...

Hi Jon,

I do find the hand-wringing today about this article in the BMJ quite amusing.

What physician searchers should be aware of is that Google points to a lot of relevant evidence.

Just as TRIP does. Dean

Anonymous said...

The article did spark quite a reaction from the Isabel staff. You can understand how they might feel threatened. Isabel in part uses copyrighted material for their searches. In contrast Googles full text search come up with numerous different sources, from freePDFs,abstracts,websites,
journals,articles, pubmedcentral,.....to emedicine articles,It would be interesting to see how the respective databases compare in terms of size.

Ward Merkeley, M.D.

jon said...

Hi Mark,

I can imagine how concerned Isabel must be. To me, it reflects the move away from 'subscription' services and centralised resources. I think the rise of Google (as well as new web 2.0 activities) will have a number of people concerned.

Google have a knack of generating publicity - if google is involved it must be 'gold'. So even if they aren't directly involved the google effect gets them the headlines. Isabel would have to pay vast sums to get such PR....

We (well the TRIP Database) moved from a subscription service for a number of reasons, one of which being the 'relentless' rise of Google. Why should people pay for TRIP when they could use Google? Google isn't (in my view) anywhere near as good as TRIP. But it's pretty fine. If I was hard-pressed financially I wouldn't bother subscribing to TRIP. So we adopted the 'new' model and have thrived.

One final thought - now Google have seen this as a potential use I imagine they'll start to focus more attention on this.....

jon

Anonymous said...

Jon,
There continues to be alot of discussion in the BMJ about googling for a diagnosis. When google walks into a room, people look up and have lots of different reactions and thoughts.
Using the internet for medical purposes is rapidly evolving. Many professional are using it in novel ways or just trying it out.
Using the internet as a source for medical information in developing countries with no medical libraries or textbooks is
going to be huge.
Just imagine having a database of standard xrays and MRI for radiology students to look at.
The attitudes about using the internet will always be relative. If you are working in a large emergency room and have a question, you might call one of the attending about tularemia details, but if your in a very very small community in Spain working, you might find a link to the internet priceless?

Ward Merkeley,M.D.
http://internetmedicine.blogspot.com