The basic message is this: everyone in the mental health profession knows full well that the DSM is a work of fiction—that the hundreds of “disorders” described therein are just labels for fuzzy, overlapping clusters of symptoms and that we have never found a definitive biological marker for even one of those disorders. Mental health professionals pretend that the disorders are real, but they're not, period.I have long maintained openly that the categories laid out in the DSM are made up and not very useful for actually helping the people who suffer from the disorders therein. The only part I disagree with above is that "mental health professionals pretend that the disorders are real" - because I am a mental health professional, and clearly, I don't. Nonetheless I am forced to play along with the labeling system because unless you have an "approved" disorder, your insurance probably won't pay for your therapy. Stressed out because of your boss, your significant other, your children? Major depressive disorder (probably unspecified type). Worried about your financial situation? Marital trouble? Generalized anxiety disorder. Does everyone who gets these diagnoses meet all the criteria? Unlikely - but if you want coverage, you have to play by the insurance companies' rules, which thus far still include the DSM's fiats. (I must confess to being ignorant of the rationale for the insurance companies' approach, but I imagine it has to do with practicality over precision - see below.)
I think it is important to note, however, that the DSM and its labels are not entirely useless. Firstly, diagnoses help somewhat with being able to identify clusters of symptoms as representative of a certain disorder which thus requires a certain treatment. You don't do cognitive-behavioral work with psychotic disorders; if someone appears to have borderline personality disorder, dialectical behavior therapy is the only treatment that has demonstrated any effectiveness.
Likewise, categories are helpful for research purposes. If we want to know what has been effective for people who are struggling with exaggerated worry, we need some definitions in order to study what sort of treatment yields the best results. The problem is that the categories often become gospel. Many or even possibly most people with anxiety problems might respond to treatment X, but certainly not all of them will. People, after all, are people - no two alike.
As an on-the-ground practitioner, I use the labels to satisfy the insurance companies; but I am in agreement with the author of this book that by and large, the whole approach is something of a farce.