Beware Dr Google! People who use internet to diagnose illness 'can't interpret their own symptoms'
05:47 GMT, 22 July 2012
Many people may believe that the internet has made it easier for us to discover what is wrong when we are sick.
But new research suggests that using Google to diagnose illnesses could in fact be a very bad way of getting appropriate medical treatment.
Of course, a rigorously trained doctor is likely to give a much more accurate diagnosis than the average web user seeking answers from the internet.
But in addition, scientists have warned that individuals do particularly poorly when asked to work out their own chances of having any particular ailment.
Be careful! Scientists warn that self-diagnosis via the internet can be dangerous (picture posed by model)
This misdiagnosis takes two main forms – self-positivity, where we overestimate the risks of falling prey to an illness, and self-negativity, where the opposite is the case.
For example, according to NBC News, people may interpret symptoms which in someone else might seem like indigestion as a sign they are having a heart attack.
Two scientists from the Hong Kong University of Science and Technology used this sort of finding to develop a more systematic study of how people perceive their chances of illness.
They gave college students information on various diseases, telling them both how common they are among the whole population ('base rate') and the details of one specific person's health profile ('case risk').
Sick: But most people are very bad at reading their own symptoms (picture posed by model)
If that person was a stranger, the test subject would tend to rely on the base rate, using a statistical approach.
But if asked to judge their own risk of exposure to disease, the participants primarily used the case risk.
That means that for conditions whose base rate is relatively low, people are more likely to believe that they are affected than that others are.
Conversely, it also means that individuals might underestimate their chances of suffering from relatively common illnesses.
This is why, for example, people are often happy to shake off seasonal flu symptoms, claiming to be 'under the weather' instead, but during an outbreak of swine flu many more people think they are affected than actually are.
The study reminds people that if they try to get medical help from the internet, they are limited by their own biases as well as the haphazard nature of the web.