Psychiatry's most famous rulebook began partly as an army manual
The DSM decides, in practice, what counts as a mental disorder in the United States, and insurers often demand one of its codes before paying. Yet it grew out of census tallies, asylum statistics and a US Army handbook, and its own publisher admits that many diagnoses are applied inconsistently.
The Diagnostic and Statistical Manual of Mental Disorders is published by the American Psychiatric Association to give clinicians a shared vocabulary and standard criteria. The first edition appeared in 1952, and every revision since has added new disorders while dropping conditions no longer regarded as illnesses. The current version, DSM-5-TR, came out in March 2022.
Its reach goes well beyond the consulting room. Researchers use it to sort study participants, drug regulators and pharmaceutical firms rely on it, and courts and policymakers cite it. In the United States, hospitals, clinics and insurers may insist on a DSM diagnosis for anyone treated for a mental disorder. Recent editions have been praised for grounding diagnosis in evidence rather than the theory-driven classification used in DSM-III.
Criticism has been persistent. Skeptics question whether many diagnoses are reliable or valid, argue that the lines between illness and normality are arbitrary, and worry about cultural bias and the medicalisation of ordinary distress. The APA's own figures show low agreement between clinicians for several DSM-5 conditions, among them major depressive disorder and generalised anxiety disorder. Many studies of treatments also track changes on symptom scales rather than whether patients still meet DSM criteria.
Its main rival is the World Health Organization's International Classification of Diseases, which covers all of health and devotes chapter 6 to mental and neurodevelopmental conditions. The ICD dominates in Europe and much of the world, so its reach is far larger. A survey of psychiatrists in 66 countries found ICD-10 favoured for everyday diagnosis and DSM-IV for research, perhaps because the DSM leans on checklists, such as meeting 5 of 9 criteria for at least 6 months, while the ICD trusts clinical judgement more. Since 1980 every DSM code has been an ICD-9 code, and the two manuals have grown closer, though the DSM still centres on specialist care in wealthy countries.
Source: Diagnostic and Statistical Manual of Mental Disorders
