The fifth edition of DSM (Diagnostic and Statistical Manual of Mental Disorders) of the American Psychiatric Association, published in May 2013, aimed to a make a breakthrough in the history of the taxonomy of mental disorders.1 The first edition of the manual (DSM-I) by the American Psychiatric Association in 1952 was an attempt to provide a common language for communication between the clinicians. Both DSM-I and its revision, DSM-II (published in 1968), identified only three broad categories of mental disorders: psychoses, neuroses, and character disorders. The terminology used and the description of disorders in these early editions largely reflected the psychoanalytic approach of psychopathology and the identification of specific disorders was based rather on an expert consensus than on research findings. DSM-III, published in 1980, established the most important changes in psychiatric classification so far.2 It defined specific descriptive criteria for the diagnosis of mental disorders while dropping any reference to particular theories on their etiology, relied to a much greater extent on data from clinical and epidemiological studies to define the specific disorders, and introduced the multi-axial diagnosis (apart from the clinical-diagnostic categories, additional separate axes describing personality and intelligence disorders, medical conditions, psychosocial problems, and global functioning). These features of psychiatric classification remained in the next edition of the manual, DSM- IV (published in 1994).3 The general taxonomic approach introduced by DSM-III –using specific descriptive diagnostic criteria, atheoretical with regard to etiology, and based on empirical data– remained in DSM-5, but there are also several important changes in classification methodology as well as in diagnostic categories and criteria.

The shift from the categorical approach of psychopathology (diagnosis based on criteria that either are fulfilled or not) to a dimensional approach (definition of the disorders based on psychopathological dimensions along a continuum of severity or intense) was a central goal in the development of DSM-5, which was reduced, however, to the introduction of some “dimensional” aspects in its current version.4 More precisely, while the categorical diagnostic criteria remain, there are also specifiers of disorder severity and other clinical features, and certain symptoms can be also assessed by clinical scales. Several disorders are grouped into broad categories that are called “spectrums”, e.g. “schizophrenia spectrum”, however only in autism separate disorders of DSM-IV are now unified in a psychopathological continuum, the autism spectrum disorder. Moreover, a multi-dimensional model for the description of personality disorders is proposed as an alternative, whereas in the official model the DSM-IV diagnostic categories and criteria for these disorders are preserved.

The multi-axial system for diagnosis has been dropped in DSM-5, although there was no argument against the clinical importance of the axes in past editions of DSM, i.e. the biopsychosocial approach to assessing and treating the patients. Since the multi-axial diagnosis was not widely used in everyday practice, in the new system diagnoses of any mental disorder (including personality disorders) and any medical condition are made on a single axis. Additionally DSM-5 comprises, although optionally, the identification of certain psychosocial stressors and the assessment of global functioning by a self-administered scale, the WHODAS 2.0.5 As a companion to diagnosis, the use of certain clinical tools is proposed: “cross-cutting symptom measures” for the assessment of symptoms co-occurring across mental disorders (e.g. depression, anger, mania, anxiety),6 and specific clinical scales for the assessment of symptom severity in certain disorders (e.g. dimensions of psychosis symptoms). Another useful clinical instrument, especially in a cross-cultural clinical context, is the Cultural Formulation Interview, a semi-structured interview for the assessment of cultural factors with significant impact on patient’s attitude towards the recognition and treatment of a mental disorder.7

The number of diagnoses included in DSM-5 is much higher than DSM-IV (541 versus 383). This increase is mainly due to splitting or regrouping of disorders already existing in prior editions. However, DSM-5 introduced a significant number of new disorders, such as disruptive mood dysregulation disorder, premenstrual dysphoric disorder, binge eating disorder, hoarding disorder etc. Important changes from DSM-IV have also been made in diagnostic criteria and clinical subtypes of specific disorders.8 Regarding schizophrenia, for example, the special diagnostic significance of bizarre delusions and auditory hallucinations (Schneiderian first-rank symptoms) are eliminated. Moreover, clinical subtypes of schizophrenia are eliminated due to their limited validity, reliability, and longitudinal stability. Instead, the evaluation of clinical heterogeneity and severity of the disorder through the dimensions of symptomatology (positive, negative, disorganized, psychomotor, affective, and cognitive) is recommended. Catatonia is now described as a distinct clinical syndrome and can be specifier to the diagnosis of psychotic, bipolar, or depressive disorders. Some of the possible new disorders that were considered, they are proposed for further research. The most important and most controversial among them, the “attenuated psychosis syndrome”, describes a condition that is considered prodromal of psychosis, although many individuals with this condition do not finally manifest the full blown syndrome.9 Suicidal behavior and nonsuicidal self-injury are also proposed as candidate distinct syndromes for further study, because these behaviors may be associated with specific psychological and biological factors independently of other co-occurring mental disorders. Other conditions (syndromes) proposed for further study are the following: depressive episodes with aort-duration hypomania, persistent complex bereavement disorder, caffeine use disorder, internet gaming disorder, neurobehavioral disorder associated with prenatal alcohol exposure.

The controversies that surrounded the development and publication of DSM-5 rose important issues not only about the changes in this revision but more generally about diagnostic systems and the status of psychiatric diagnosis. The risk of diagnostic overexpansion through the “medicalization” of psychopathology, the currently poor evidence for the validity of psychiatric diagnosis, the “phenomenological poverty” that is resulted from the exclusive use of the DSM in clinical practice and psychiatric training, are issues open for discussion and crucial for the future of psychiatry.10 On the other hand, the revolutionary changes that have been proclaimed but could not be actualized as yet, i.e., the incorporation of neuroscientific findings in the diagnostic system and the dimensional and spectrum approach in psychopathology, also aim to strengthen the scientific status of psychiatric diagnosis.11 The most interest and progressive aspect of the new diagnostic system might be its association with the prospective of faster and continuing revisions (to DSM-5.1, -5.2, etc.). Modern psychiatry should find a shorter route to the direction that combines more reliable and valid diagnosis with individualized care for patients.

Vassilis Kontaxakis
Professor of Clinical & Social Psychiatry, University of Athens, Athens, Greece

George Konstantakopoulos
Research Fellow, First Department of Psychiatry, University of Athens,
Eginition Hospital, Athens, Greece
Section of Cognitive Neuropsychiatry, Department of Psychosis Studies,
Institute of Psychiatry, Psychology and Neuroscience, King’s College London, UK


  1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. American Psychiatric Publishing, Washington, DC, 2013
  2. Decker H. The Making of the DSM-III: A Diagnostic Manual’s Conquest of American Psychiatry. Oxford University Press, New York, 2013
  3. Blashfield RK, Keeley JW, Flanagan EH, Miles SR. The cycle of classification: DSM-I through DSM-5. Ann Rev Clin Psychology 2014, 10:25–51
  4. Narrow W, Kuhl E. Dimensional approaches to psychiatric diagnosis in DSM-5. J Mental Health Policy 2011, 14:197–200
  5. Konecky B, Meyer EC, Marx BP, Kinbrel NA, Morissette SB. Using the WHODAS 2.0 to assess functional disability associated with DSM-5 mental disorders. Am J Psychiatry 2014, 171:818–820
  6. Clarke DE, Kuhl EA. DSM-5 cross-cutting symptom measures: a step towards the future of psychiatric care? World Psychiatry 2014, 13:314–316
  7. Alarcón RD. Cultural inroads in DSM-5. World Psychiatry 2014, 13:310–313
  8. Regier DA, Kuhl EA, Kupfer DJ. The DSM-5: classification and criteria changes. World Psychiatry 2014, 12:92–98
  9. Kontaxakis V, Havaki-Kontaxaki B, Kollias K, Ferentinos P, Papadimitriou G."Attenuated psychosis syndrome" or "subthreshold prodromal state"? Psychiatry Investigation 2013, 10:203–204
  10. Karavatos Th. The reception of DSM-5: "The problem with DSM-5 is the DSM not the 5". Synapsis 2014, 10:4–7 (in Greek)
  11. Adam D. Research suggests that mental illnesses lie along a spectrum - but the field’s latest diagnostic manual still splits them apart. Nature 2013, 496:416–418

Full article in pdf