Stupid pet tricks, aka Google diagnosing

Stupid pet tricks, aka Google diagnosing

From the BBC
A team of Australian doctors googled the symptoms of 26 cases for a study in the New England Journal of Medicine.In 15 cases, the web search came up with the right diagnosis, the paper published on the British Medical Journal website reports.In each of the 26 cases studied, researchers based at the Princess Alexandra Hospital in Brisbane selected three to five search terms from each case and did a Google search without knowing the correct diagnoses.They then recorded the three diagnoses that were ranked most prominently and selected the one which seemed most relevant to the signs.The doctors then compared the results with the correct diagnoses as published in the journal.Google searches found the correct diagnosis in just over half of the cases.

The crucial paragraph from the BMJ paper seems to be:

We then did a Google search for each case while blind to the correct diagnosis (that is, before reading the differential diagnosis and conclusion of each case record). We selected and recorded the three most prominent diagnoses that seemed to fit the symptoms and signs.
They looked through 30-50 results to find three that seemed reasonable. If a real decision support tool like SimulConsult doesn't rank the correct diagnosis within the top 10 it's considered as evidence of failure, not success. When there is such a failure typically there is some important fact missing from the database, a situation that is then corrected. It is not clear if Google has a similar ability to learn from the wisdom of the community of users.
Searches are less likely to be successful in complex diseases with non-specific symptoms ... or common diseases with rare presentations ...
These are the situations where decision support software is most useful.

Google is incredibly powerful. If you are too lazy to think you can type any thoughtless thing in and get something at least somewhat relevant out. That doesn't mean Google should be used for clinical decision support.

Previous
Previous

The diagnostic crisis in child psychiatry

Next
Next

Blown Away