Three decades after the introduction of explicit diagnostic criteria and, subsequently, rule-based classifications such as DSM-III [1 ], DSM-III-R , ICD-10  and DSM-IV , it should be possible to examine the impact of these tools on psychiatric nosology. The worldwide propagation of the new classification systems has resulted in profound changes affecting at least four domains of professional practice.
First and foremost, a standard frame of reference has been made available to clinicians, enabling them to achieve better diagnostic agreement and improve communication, including the statistical reporting on psychiatric morbidity, services, treatments and outcomes. Secondly, more rigorous diagnostic standards and instruments have become the norm in psychiatric research.
Although the majority of the research diagnostic criteria are still provisional, they can now be refined or rejected using empirical evidence. Thirdly, the teaching of psychiatry to medical students, trainee psychiatrists and other mental health workers is now based on an international reference system which, while reducing diversity due to local tradition, provides a much needed “common language” to the discipline worldwide.
Fourthly, open access to the criteria used by mental health professionals in making a diagnosis has helped improve communication with the users of services, carers, and the public at large, by demystifying psychiatric diagnosis and making its logic transparent to non-professionals.
While acknowledging such gains, it is important to examine critically the current versions of standardized diagnostic criteria and rule-based classification systems in psychiatry for conceptual and methodological shortcomings. At present, the discipline of psychiatry is in a state of flux.
Advances in neuroscience and genetics are setting new, interdisciplinary agendas for psychiatric research and the results to be expected within the next few decades are likely to affect profoundly the theoretical basis of psychiatry, in particular the understanding of the nature and causation of mental disorders.
New treatments targeting specific functional systems in the brain will require more refined definitions of the clinical populations likely to benefit from them than is possible at present. Even more importantly, the realization that, in all societies, mental disorders contribute a much larger burden of disease than previously assumed will raise critical questions about cost-benefit, equity, right to treatment, and feasibility of prevention.
The conjunction of these powerful factors is likely to have major implications for the future of psychiatric classification as a conceptual scaffold of the discipline. There is little doubt that the classification of mental disorders will undergo changes whose direction and extent are at present difficult to predict.
Although the prevailing view is that an overhaul of the existing classification systems will only be warranted when an accumulated “critical mass” of new knowledge makes change imperative, processes aiming at revisions are already under way and the debates about the future shape of DSM and ICD are gathering momentum. In the light of this, a discussion of the basic principles and “rules of the game” should be timely.
Of course, the complexity of the subject makes it unlikely that any sort of quality assessment checklist will soon emerge and become generally accepted in reviewing new proposals. Nevertheless, a step in that direction is needed if further progress in consolidating the scientific base of the discipline is to be achieved.