Two recent events suggest that the time is ripe for an international and open exchange of views about the development process that will be required for the psychiatric chapter of the eleventh revision of the International Statistical Classification of Diseases and Related Health Problems (ICD–11). In chronological order, the first is the publication of A Research Agenda for DSM–V (Reference Kupfer, First and RegierKupfer et al, 2002), commented upon briefly below. The second is the recent appointment of a new Director General of the World Health Organization (WHO). In the assessments of policies and organisation that are likely to follow this new appointment, it is at least possible that consideration will be given to how the WHO will deal in the near future with its responsibility for the ICD. The general responsibility for all chapters of the ICD is an important issue, and a new problem limited to the successor to Chapter V(F) of ICD–10 will also need to be addressed from the start. The sale of versions of Chapter V(F) of ICD–10 across the world has shown that its successor could have the potential to generate even greater profits, and therefore will need to be handled differently from the rest of the ICD. It is to be hoped that the WHO will take full advantage of this difference.

In the past, the limited time available for each revision process has always cut short what could be achieved. Now, for the first time, there is at least the possibility that the programme of consultation and development can be driven by what is desirable rather than by what can be fitted into an arbitrary and short period of time.

TIME, FOR A CHANGE

Chapter V of ICD–10, in its different versions, has been used much more widely than any other previous revision. This means that the views of its main users, who will inevitably be clinicians rather than researchers, should be a valuable guide for the development and presentation of ICD–11. Some suggestions about the development of ICD–11 are given here, but first, some comments upon the recent publication of A Research Agenda for DSM–V (Reference Kupfer, First and RegierKupfer et al, 2002).

AN AMERICAN AGENDA

This book is both interesting and disappointing. It is interesting because each of its six chapters contains the distilled comments and wisdom of a group of experts who began their deliberations in 1999, with future versions of the DSM in mind. The six chapters deal in turn with nomenclature, neuroscience, developmental science, personality disorders and relational disorders, mental disorders and disability, and culture and psychiatric diagnosis, and each is based upon the proceedings of series of meetings that started in 1999. The chapters are long and detailed and are accompanied by an extensive bibliography; they constitute valuable reviews of recent developments and current practice, plus recommendations for future research. The disappointment is perhaps inevitable because so many recommendations are made, in effect often amounting to a list that implies that all possible research should be carried out on all possible topics.

The title of the book is misleading, in that it implies that the research recommended could form the basis of DSM–V. Inside, there are more realistic comments such as ‘some of the research agendas suggested in these chapters might not bear fruit until the DSM–VI or even DSM–VII revision processes’. The revision process itself is noted as still being several years in the future, with a tentative suggestion that DSM–V might be published in the year 2010. As might be expected, the viewpoints expressed in all the chapters are those of the contemporary research community of the USA. Of the 46 contributors, 42 are American. This is, of course, quite legitimate, since the DSM is produced primarily to serve the interests of the members of the American Psychiatric Association. The most stimulating chapter is the fourth, entitled ‘Personality disorders and relational disorders’. The discussion of the concept of relational disorders, defined as ‘persistent and painful patterns of feeling, behaviour and perceptions involving two or more partners in an important personal relationship’, is a most welcome and public recognition that large parts of psychiatry are necessarily concerned with more than the emotional states and behaviour of individual persons. Problems abound in trying to work out how to cope with these concepts in a classification (which is why every psychiatric classifier has avoided this topic in the past) but the discussion here is a valuable stimulus.

DIFFERENT SORTS OF DIFFERENCES

The chapter on nomenclature is less useful, the weakest section being a discussion of differences between some parts of ICD–10 and DSM–IV found during the recent Australian National Mental Health Survey (Reference Andrews, Henderson and HallAndrews et al, 2001). The discussion is based upon data from the use of the Composite International Diagnostic Interview, a highly standardised interviewing instrument administered by trained lay interviewers (Reference Robins, Wing and WittchenRobins et al, 1988). Such information has its uses in some types of epidemiological studies, but a serious and detailed comparison of psychiatric classifications justifies the use of data of better clinical quality. Before expending time and effort on trying to remove comparatively small differences between the two classifications, it would be better to give urgent priority to an inquiry into how studies in the same country using the same classification can give rise to very different rates for psychiatric disorders, as demonstrated by the current and important debate about surprising differences in survey results in the USA (Reference Regier, Kaelber and RaeRegier et al, 1998; Reference Narrow, Rae and RobinsNarrow et al, 2002). Changes in stem questions and data analysis – originally thought to be unimportant – are probably the main causes, but further studies are required (and they need not be on a large scale). In fact, ICD–10 (World Health Organization, 1992) and DSM–IV (American Psychiatric Association, 1994) are reasonably similar in terms of basic content, for the simple reason that they are both based upon the same body of information, which is published in the international psychiatric literature and is therefore freely available to all. Those differences that exist are of considerable educational interest since they are based upon opinions and clinical traditions, and not upon robust evidence. Some sections of the chapter on nomenclature could be taken as suggesting that so long as there is some ‘international’ input to the various committees that will produce DSM–V, a separate ICD–11 produced by the WHO will not be necessary. If this is what is meant, then surely the suggestion is the wrong way round. From an international point of view, it is far better for the WHO to produce a classification as a result of widespread consultations (including many experts from the USA, as for ICD–10), and then if the psychiatrists of any country feel strongly that something else would be useful locally, some national alternative sub-classifications can be produced with clear explanations about why they are thought to be useful and how they are translatable into the agreed international version. It is not surprising that the authors of these chapters adopt a clearly American approach to this whole subject, but much would be lost if any particular national organisation were to try to supplant the WHO in its function of providing the international psychiatric community with an acceptable ‘common language’. One can only hope that mental health professionals across the world will quickly inform the WHO that a truly international and separate ICD–11 is required, with widespread consultation as for ICD–10.

WHO IS RESPONSIBLE?

It is to be hoped that the WHO will still regard the production of ICD–11 as an important task, not to be delegated to anybody else. It would certainly be helpful to the international psychiatric community if the WHO were to make a policy statement in the near future about the development process, because there is still plenty of time for a wider consultation process for ICD–11 than was possible for ICD–10. The World Psychiatric Association could have a key role in this, as it did for ICD–10. In the first stage of the consultation, the main users (who are clinicians in psychiatry, primary health care and clinical psychology) could be asked specific questions about, for instance, when they think ICD–11 should be published, what general form they think it should take, and whether they are in favour of the ICD–10 policy of ‘different versions for different users’. There were two reasons for publishing the diagnostic criteria for research (DCR–10) separately from the clinical descriptions and diagnostic guidelines (World Health Organization, 1992, 1993). One was to make the task of clinicians as easy as possible by acknowledging that the precise information needed for research work is often not easily available to a busy clinician, and the second was to emphasise to the researcher that selection of patients by comparatively precise criteria is an exercise in restriction as well as in selection. Separate publication was a rational idea, but that does not necessarily mean that it results in a successful policy in practice. The different needs of clinicians and researchers will no doubt continue to be a problem, but feedback on this would be useful.

LIMITED CHANGES ONLY?

Some of the preliminary groundwork relevant to future classifications has already been done. We should all be grateful to Assen Jablensky and the late Robert Kendell for their excellent review of the criteria for assessing a classification in psychiatry (Reference Jablensky, Kendell, Maj, Gaebel and Lopez-IborJablensky & Kendell, 2002). They comment upon all the usual main conundrums of the topic, such as the purposes of classification, the units of classification, diagnostic reliability and validity, clinical utility, categories v. dimensions, and the advantages and disadvantages of detailed lists of criteria. Their conclusions about what might be justified as changes in the next versions of both the ICD and the DSM are reassuringly modest. Like many others over the past few years, they suggest that the section on personality disorders be radically changed, but otherwise they recommend that the temptation to make many small changes in the rest of the classifications should be resisted.

USING THE NETWORK

Finally, there should be mention of how the world at large should be informed of ICD–11, and it would be helpful if the WHO could review its methods of international distribution and sales. Up to now, the WHO has not had any direct responsibility for ensuring that its wide range of excellent health-related publications actually reach their potential buyers. The policy has been to leave this to the governments and professional organisations of the world. This is in contrast to the professional publicity and successful marketing strategies that promote the sale of the DSM products. Large sums of money are now associated with the worldwide sales of psychiatric classifications, and the WHO will continue to miss out on the profits it deserves and needs unless it sets up its own organisation for distribution and sales. It may be that the marketing policy of WHO cannot change; if so, there is one way in which a future ICD–11 could reach a wider audience than was the case for ICD–10. The network of ICD–10 field trial centres which carried out the testing of ICD–10 was based upon the existing network of WHO collaborating centres responsible for many WHO-coordinated studies over the past 30 years or so. This worldwide network has been a priceless international asset, and could be re-energised for consultations about ICD–11. These centres and groups could then be asked to promulgate the sale and use of ICD–11 in their own countries. This was done for ICD–10 by the field trial centres in several European countries, with conspicuous success. Similarly, the same groups could be encouraged to ask their national professional organisations to recommend to psychiatric tutors that, since all governments of WHO member states agree to use the ICD for reporting of health statistics internationally, all professional trainees should be familiar with ICD–11, whatever other national classifications they know about or use.

In addition, WHO could request all editors of major psychiatric journals to accept papers that give diagnostic information in terms of ICD–11, since if properly developed, it would be scientifically equivalent to DSM–V (as ICD–10 Chapter V is to DSM–IV). This was in fact done for ICD–10 Chapter V, but it was not widely advertised and appears to have been forgotten by some researchers and editors. Professional mental health workers, particularly psychiatrists, should not consider themselves properly educated unless they are familiar with whatever major classifications are available. To know what differences exist between them, and the reasons for such differences, should be regarded as part of the knowledge expected of any well-educated professional.

DECLARATION OF INTEREST

None.