I've long felt that many disorders are probably the same disorder, manifesting in an individual way within a particular individual.
The trend has been toward more splitting, rather than more lumping, I'm afraid. However, there have been cases of reclassifications where distinct conditions got rejigged with one as a special case of the other.
This is what we should expect, though, as it follows the trend of medicine through the ages. We no longer talk about diseases with such vagueries as "he has the fever" - we now know which of the 3,000 *kinds* of illnesses are causing the fever, and that's what we name it.
This is not meant to be a sweeping statement. And I can't tell you specifically which ones I mean. But ADD and OCD are strongly related, I would think (this is not my area of expertise in the slightest, except as a patient

). What if they weren't just related, but were the same "problem," whatever we call that problem, which manifests differently, depending on the person.
I don't think there's a good case for that. OCD is obviously an anxiety disorder, whereas ADD is a concentration problem. Stimulants make OCD worse, but make ADD better. The underlying organic is almost certainly different.
Watch for the 'axes' and 'clusters'. This is how you find the diagnoses with shared underlying causes, although some axes and clusters are catchalls.
People are different, she said simplistically, so why wouldn't manifestations of a brain disorder naturally differ among them?
I also think folks just get tired of the "31 Flavors" air of mental diagnosis, whether they should or shouldn't. I'm talking, for instance, about that recent article this summer about a "new kind" of anger disorder, a road-rage sort of thing. Can't recall the specifics just now...
I was, oddly enough, taking a basic sociology class at that time, and we discussed it: is this really a "new" disorder, or just a new label for something we already recognize? Did we really need this new label, if so?
Well, the point of diagnostic categorization is to identify appropriate treatment. This disorder is different than others, and its treatment needs to be specialized. That was the thrust of the publication.
Previously, these people have been misdiagnosed as narcissistic personality disorder, but the usual treatment for this condition (cbt) is not effective. In order to return these rage-outburst patients to some sense of control, they need medication. It's a newly-discovered variation on Tourette's tic, and not at all a personality disorder.
How many perfectly "normal" but socially unacceptable behaviors are we medicalizing, maybe when they don't deserve to be such? Are we feeding a health-care/drug manufacturer's monster, or are we helping people?
It should be noted that again, diagnosis is not the same thing as medicalization. Psychology categorizes people all the time: introverts, communicative, fantasy-prone, innovative, &c. The DSMIVR is a tool that focuses on only a small segment of these categories - the ones that cause negative lifestyle impacts. Most diagnoses are treated with non-medical therapy in short timespans.
How many "disorders" are simply products, results, of living in a hostile society? I don't always like to be around people much, but then, people often treat me like crap. Knowing I have to go around people sometimes makes me very anxious. Disorder or learned response? If learned response, can you really give someone a pill that will help?
This is also a myth, though: that non-western, preindustrial, or primitive societies do not see mental problems. The reality is that they just don't have the infrastructure to recognize them as such, and deal with them as they see fit.
For example, my wife was literally half of "psychiatry" in Tobago for about 3 months earlier this year. She said the sad thing about it was that the people down there with mental problems are just as abundant as here, but the difference is that they do not get an iota of compassion. Narcissists are chopped up with machetes by angry neighbours, the depressed are allowed to kill themselves, and the schizophrenic get gunned down by police in a blind panic when they finally do their "I hear voices!" thing in the middle of downtown.
One of the other benefits of classification is predictability. We can predict a person's future behavior, given enough information to understand their diagnostic classification. In the example of the road-rage problem, this is debilitating, and these people now face the legal restriction against having a driver's licence. Just as if they were declared legally blind (an arbitrary legal threshold on a subjective medical diagnosis), the public can assess their risk of causing accidents, and protect ourselves accordingly.
Bipolars often lose their driver's licences, too (not to mention, being institutionalized).