DSM-5 reviewed from different angles: goal attainment, rationality, use of evidence, consequences—part 2: bipolar disorders, schizophrenia spectrum disorders, anxiety disorders, obsessive–compulsive disorders, trauma- and stressor-related disorders, personality disorders, substance-related and addictive disorders, neurocognitive disorders

Part 1 of this paper discussed several more general aspects of Diagnostic and Statistical Manual of Mental Disorders (DSM-5) and offered a detailed, paradigmatic analysis of changes made to the chapter on depressive disorders. This second part focusses on several other disorders, including bipolar and schizophrenia spectrum disorders. The respective changes and their possible consequences are discussed under consideration of traditional psychiatric classification, particularly from the perspective of European traditions and on the basis of a PubMed search and review papers. The general conclusion is that even seemingly small changes such as the introduction of the mixed feature specifier can have far-reaching consequences. Contrary to the original plans, DSM-5 has not radically changed to become a primarily dimensional diagnostic system but has preserved the categorical system for most disorders. The ambivalence of the respective decision-making becomes apparent from the last minute decision to change the classification of personality disorders from dimensional back to categorical. The advantages and disadvantages of the different approaches are discussed in this context. In DSM-5, only the chapter on addictive disorders has a somewhat dimensional structure. Also in contrast to the original intentions, DSM-5 has not used a more neurobiological approach to disorders by including biological markers to increase the objectivity of psychiatric diagnoses. Even in the most advanced field in terms of biomarkers, the neurocognitive disorders, the primarily symptom-based, descriptive approach has been preserved and the well-known amyloid-related and other biomarkers are not included. This is because, even after so many years of biomarker research, the results are still not considered to be robust enough to use in clinical practice.