Some aspects of the DSM-III personality disorders illustrated by a consecutive sample of hospitalized patients.

The main innovations of DSM-III concerning the personality disorders are the assignment of a separate axis and definitions of diagnostic criteria for these disorders. Even with these improvements, diagnosis of the personality disorders shows only moderate interrater reliability. Reasons for this are discussed. Data from a consecutive sample of hospitalized patients are used to illustrate the prevalence of the personality disorders, the coexistence of axis I disorders, and the overlap within the personality disorders. The method of conditional probabilities is presented as a way to get better separation of these disorders.