Byline: Nimesh. Desai The traditional practice of understanding and explaining all the symptoms and signs of a person presenting to mental health facilities was synchronous with the hierarchical system of classification. The clinical dictum of 'one person one diagnosis' was in keeping with the early classification systems wherein the diagnostic groups were arranged in hierarchy and clinicians were actively encouraged to arrive at one diagnosis for all the experiences and problem behaviours of each person. When a clinician came across a patient who had dependent pattern of alcohol use and significant depressive features, it was incumbent to decide the primary condition of the two. Research efforts were made to authenticate primary versus secondary alcoholism.[1] In patients with alcohol or other drug dependence and an established pattern suggestive of antisocial personality profile, it was necessary to ascertain if the substance use could be explained as part of or was epiphenomenon of the antisocial personality. The underlying approach and rationale for such practice were to simplify the diagnostic groups, and to adopt one comprehensive management strategy which would be expected to benefit all symptoms. In a person with schizophrenia and depression, it was believed that effective treatment of schizophrenia would relieve the depression and, similarly, successful therapy of the antisocial personality would control the substance use, and in secondary alcoholism, the treatment of the underlying condition would take care of the treatment of alcoholism. In 1970, Robins and Guze put forth a strong argument for making empirical evidence, through precise clinical descriptions and delineation of syndromes, the major criteria for psychiatric diagnoses.[2] About the same time, Feinstein described the term comorbidity for all chronic diseases as 'any additional clinical entity that has existed or that may occur during clinical course of a patient who has the index disease under study for chronic diseases'.[3] A well-described case in point was chronic obstructive pulmonary disease in a person with diabetes mellitus. In the 1970s, the operational diagnostic criteria based on empirical, operational and measurable attributes were popularized in psychiatry initially for research[4],[5] and later for clinical application. The introduction of the multi-axial diagnostic criteria-based classification system with the empirical, atheoretical approach contributed to a paradigm shift. Among other contributions, the one major change was the recognition of the concept of comorbidity in psychiatry by not only permitting but encouraging the clinicians to make more than one diagnoses in a person, if the criteria were adequately met for more than one disorder. At that early stage, the DSM-III stated that one diagnosis cannot be made if it is 'due to' another disorder. This provision, which was gradually made more flexible, in the subsequent classification systems, and later universalized by the ICD-10 in 1992, also following the same approach, modified not only the coding and record keeping, but more significantly broadened the perspectives of clinicians. The assessment of mental health problems of persons with multiple diagnoses, and the related implications in planning treatment and management strategies, has been significantly influenced during the 1980s and 1990s by this one paradigm shift. Soon after DSM-III, in path-breaking research, Boyd et al. in 1984, reported from a large community-based sample in USA that there was 'a general tendency toward co-occurrence, so that the presence of any disorder increased the odds of having almost any other disorder'.[6] They defined comorbidity in epidemiological terms as 'the relative risk of a person with one disorder, to receive the diagnosis of another disorder'. The current systems of DSM-IV and ICD-10 actively encourage multiple diagnoses in the same person, regardless of the possible contribution to aetiology, allowing the maximum amount of diagnostic information. …