Doctors don't diagnose purely on the basis of the DSM or similar statistical tools, and haven't for decades, at least here in the UK which is the system I am most familiar with.
This is too big a subject area to go into here in a HN comment but if you want to know more, a good place to start would be by familiarising yourself with the difference between nomothetic and idiographic knowledge, which is completely key to modern diagnostic practices in psychiatry.
VERY VERY simply, while someone might be considered severely abnormal from a nomothetic viewpoint, from an idiographic viewpoint clinicians might decide that they have always been that way and they have a history of being happy and getting on well in their society and with those around them and being successful, and therefore they cannot call that person crazy, however statistically abnormal their actions may be.
But yeah, it's a lot more complex than that.
EDIT: an important point to remember is that not every doctor is some sort of sucker, prescribing whatever big pharma tells them to. When you see a microsoft press release, do you read it and say "well that sounds convincing!" and start building everything on MS platforms? Thought not.
Obviously this largely depends on the economic relationship between healthcare practitioners and big pharma in your country and your healthcare system, whatever that might be.
If you were unlucky enough to live in a country where MS could just pay you per line of ASP.NET checked in, well I guess there would be a lot of misery and shitty code.
This is too big a subject area to go into here in a HN comment but if you want to know more, a good place to start would be by familiarising yourself with the difference between nomothetic and idiographic knowledge, which is completely key to modern diagnostic practices in psychiatry.
http://en.wikipedia.org/wiki/Nomothetic_and_idiographic
VERY VERY simply, while someone might be considered severely abnormal from a nomothetic viewpoint, from an idiographic viewpoint clinicians might decide that they have always been that way and they have a history of being happy and getting on well in their society and with those around them and being successful, and therefore they cannot call that person crazy, however statistically abnormal their actions may be.
But yeah, it's a lot more complex than that.
EDIT: an important point to remember is that not every doctor is some sort of sucker, prescribing whatever big pharma tells them to. When you see a microsoft press release, do you read it and say "well that sounds convincing!" and start building everything on MS platforms? Thought not.
Obviously this largely depends on the economic relationship between healthcare practitioners and big pharma in your country and your healthcare system, whatever that might be.
If you were unlucky enough to live in a country where MS could just pay you per line of ASP.NET checked in, well I guess there would be a lot of misery and shitty code.