There are a lot of issues associated with the idea of "mental illness". A lot. But let's ignore those for the moment.
First, there's way better coverage of what Francis' (and other psychiatrists) are complaining about here:
Second: If we ignore the ridiculous dream of coming up with a definitive dividing line between normal and "ill", and focus instead on what purpose is served by expanding the number of people who can go to a professional, collect a diagnosis and then access resources as a result, it's possible to get some clarity on whether DSM-5 is going to make things better or worse. Every time we collectively decide that more people have a mental illness, the debates surrounding the efficacy of drug and other treatments increases in visibility. The result is generally better treatment that costs less per person, and maybe costs slightly more collectively while being used by a much larger number of people. A diagnosis of mental illness is essentially a measurement of not-fitting-societal-norms. When you get enough people who don't fit in a particular way and stop being willing to lie and say they do, and it's no longer possible to flog the general population into enforcing the norm, the category of illness goes away, and that particular mode of being is treated as being within the range of normal. Those are two mechanisms we've seen in action repeatedly already that mitigate any possible negative effects of diagnosing more and more people.
However loopy #5 is when it comes out, it is actually quite unlikely to be worse than what we have now. The progession, nicely depicted in the table in the wired article, illustrates just how much improvement we've seen. When (over in the NPR coverage) Francis complains about kids who never would have been diagnosed before getting $50K worth of educational services -- as if that's a bad thing -- I can't help but notice that the man is old and his value system reflects it. If we get too many kids who need expensive services for autism disorders, we will figure out a more effective way to provide services that costs less. It's what we do. For that matter, it's what my _town_ does. And we like the results.
Finally, it's always a little sad to see the mental health crowd treating the physical health crowd as somehow "more real". I spend a lot of time making fun of the physical health crowd for defining disease in numerical and statistical ways and then prescribing treatments that will only show beneficial effects over populations, while the risks associated with treatment are born by each individual. The mental health crowd has a long ways to go before they can get nearly as wacky as the physical health crowd.