http://www.ncbi.nlm.nih.gov/pubmed/23592047

J Clin Psychol. 2013 May;69(5):516-22. doi: 10.1002/jclp.21992.

Interview: What is PTSD Really? Surprises, Twists of History, and the Politics of Diagnosis and Treatment.

van der Kolk B, Najavits LM.

Source

The Trauma Center at Justice Resource Institute.

"Let's start with how the Diagnostic and Statistical Manual of Mental Disorders (DSM) diagnostic system was born. It came out of a need for medication researchers to agree on a set of criteria so that when they studied a drug for, let's say, depression, people in Los Angeles, New York, and Boston could have a common language for what they meant by “depression.” Its first incarnation was called the “Research Diagnostic Criteria.” At that point everybody seemed to agree that the mind is way too complex, and our knowledge way too limited, to conceptualize distinct diseases of the mind. But drug researchers needed to have a way of talking to each other to get a rough impression of which drugs work for different mental problems. And as crude criteria for drug research, they were serviceable.

However, the limitations of the criteria were well recognized. In the preamble to DSM-III, Robert Spitzer conveyed that the diagnostic system was way too inaccurate and schematic to ever be used for forensic or insurance purposes. That little preamble was left out of subsequent editions and soon psychiatric residents and psychology students came to be taught that the DSM diagnoses actually exist, that they represent concrete disease entities, rather than lists of symptoms. There was little talk that they did not have clearly identifiable, distinct brain or cognitive markers, that there was little research to identity the boundaries between disorders, and that there is an enormous overlap between these supposedly distinct disorders. Thirty years later, this situation has not really improved. After the simplest of field trials for DSM-V diagnoses, based on clinician ratings of written case histories, only five of the 23 DSM-V diagnoses have achieved scientifically acceptable (kappa) levels of agreement."