Nosology In Psychiatry Research Paper

1. History: From Hippocrates To The 1940s

For at least the last 2,500 years most generations of physicians, and several philosophers, have produced a nosology. Usually this was not because they were preoccupied with classification. It was because a nosology almost inevitably reflects the range and variety of diseases recognized by its authors, the extent of their clinical experience, their assumptions or knowledge about etiology, and their familiarity with or ignorance of the detailed manifestations of illness. Nearly all these nosologies have incorporated several categories of what we now regard as mental disorder, thereby demonstrating that their authors regarded these disorders as diseases to be treated by physicians, and usually with the same range of assumptions about causation and treatment as other diseases.

It is apparent from the corpus of writings attributed to him, that Hippocrates recognized at least six different forms of mental illness:

(a) Phrenitis (acute mental disturbance with fever),

(b) Mania (acute mental disturbance without fever),

(c) Melancholia (chronic mental disturbance),

(d) Epilepsy,

(e) Hysteria (convulsions and other paroxysmal symptoms), and

(f) Scythian disease (cultural transvestism).

His contemporary, Plato, on the other hand, distinguished between Natural Madness originating in physical disease and several kinds of Divine Madness (Prophetic Madness given by Apollo, Religious Mad- ness given by Dionysus, Poetic Madness inspired by the Muses, and Erotic Madness inspired by Aphrodite and Eros).

Many of the nosologies developed over the next 2,000 years are listed by Menninger (1963). Most were fairly simple and incorporated Hippocrates’ categories of mania, melancholia, and epilepsy. Others, like those of Boissier de Sauvages and William Cullen at the end of the eighteenth century, contained several hundred categories of disease arranged in tiers mimicking the orders, genera, and species of plants. Only in the nineteenth century, with the emergence of physicians specializing in the treatment of insanity (known as alienists), did separate classifications of mental illness start to appear. They quickly proliferated, however, because every self-respecting alienist, and certainly every professor, produced his own nosology, some based on behavioral differences, others on presumed differences in etiology, others on psychological differences, some on a mixture of all three. Indeed, as a medical historian subsequently observed, ‘to produce a well-ordered classification almost seemed to have become the unspoken ambition of every psychiatrist of industry and promise.’

It was soon apparent that this multiplicity of classifications, with diverse meanings often attributed to the same terms, rendered communication between one hospital and another almost impossible, and with this came recognition of the need for a single nosology acceptable to everyone. At the Congress of Mental Medicine in Antwerp in 1885 a commission was appointed, chaired by a Belgian alienist, Morel, to produce this. The commission duly delivered a simple typology that was adopted formally by the International Congress of Mental Science in Paris in 1989, but as the English psychiatrist, Hack Tuke, had predicted, its nosology was generally ignored. However, as this was the first formal attempt to secure international agreement, and the forerunner of the present International Classification, its 11 categories of insanity have some historical significance. They were:

(a) Mania,

(b) Melancholia,

(c) Periodical insanity (Baillarger’s folie a double forme),

(d) Progressive systematic insanity,

(e) Dementia,

(f) Organic and senile dementia,

(g) General paralysis,

(h) Insane neuroses (hysteria, epilepsy, hypochondriasis, etc.),

(i) Toxic insanity,

(j) Moral and impulsive insanity, and

(k) Idiocy, etc.

While Morel and his colleagues were engaged on this ultimately fruitless task several national organizations were pursuing similar but more modest aims. In Britain, the Royal Medico-Psychological Association (the forerunner of the Royal College of Psychiatrists) produced a classification in 1882 which it recommended to its members for general use. Further drafts were produced in 1904, 1905, and 1906 before the Association finally accepted that there was no prospect of persuading its members to restrict themselves to the diagnoses listed in any official nomenclature, British or international. A similar sequence of events took place in the USA. In 1886 the Association of Medical Superintendents of American Institutions for the Insane (the forerunner of the American Psychiatric Association) adopted a modified version of the British classification of 1882. This was revised in 1913 to incorporate Kraepelin’s concepts of dementia praecox, involutional melancholia, and manic depressive insanity but, even so, many states and asylums still preferred to use their own nosologies, or declined to be bound by any nomenclature, national or local. Indeed, the New York State Commission in Lunacy continued to use its own private classification until 1968. In 1937 the need for an internationally agreed nosology was discussed once more at the Second International Congress for Mental Hygiene. Although it soon became clear that there was still little hope of any real consensus emerging, the simple eight-category classification proposed by the French psychiatrist Bersot was adopted by two countries, Portugal and Switzerland, and remained in use in both for 20 years.

2. The Origins Of The World Health Organization’s International Classification

Largely at the instigation of two medical statisticians, William Farr in London and Jacques Bertillon in Paris, the International Statistical Congress had, as far back as 1853, recognized the need for ‘une nomenclature uniforme des causes de deces applicable a tout les pays’ (a uniform nomenclature of causes of death applicable to all countries) and had not only produced such a nomenclature but revised it regularly for the next 50 years. Probably because of the importance of mortality statistics to public health and to governments, these successive editions of what came to be known as the Bertillon Classification of Causes of Death were used increasingly widely; and in 1900 the French Government assumed responsibility for the classification and convened a series of international meetings in Paris in 1900, 1920, 1929, and 1938, and thereby produced four successive editions of what was by then called the International List of Causes of Death.

When the World Health Organization (WHO) came into being in 1948 one of its first public actions was to produce a sixth revision of this International List. Renamed the International Statistical Classification of Diseases, Injuries and Causes of Death (ICD), this was for the first time a comprehensive nosology covering the whole range of disease, rather than merely causes of death, and so for the first time included a classification of mental illness. In fact, Section V of this sixth revision (ICD-6), entitled ‘Mental, Psychoneurotic and Personality Disorders,’ contained 10 categories of psychosis, nine of ‘psychoneurosis,’ and seven of ‘disorders of character, behavior and intelligence,’ most of them subdivided further. Unfortunately, although the nomenclature of ICD-6 was adopted unanimously by the 1948 revision conference, and was duly ‘recommended for use’ by all the member states of the WHO, its mental disorders section failed to gain acceptance, and 11 years later it was found to be in official use only in Finland, New Zealand, Peru, Thailand, and the UK. History was repeating itself. Deeply concerned by this state of affairs, the WHO commissioned an English psychiatrist, Erwin Stengel, to investigate the situation and if possible to make recommendations.

The situation Stengel encountered he described as one of ‘almost general dissatisfaction with the state of psychiatric classification, national and international’ and the attitude of many psychiatrists towards the classification of mental disorders seemed to have become ‘one of ambivalence, if not of cynicism’ (Stengel 1959). Many countries had official classifications of their own, some of recent origin, others dating back to the 1930s, but none, new or old, was regarded as satisfactory by its users, or used conscientiously and consistently by them. In the UK, although the General Register Office published mental hospital admission rates and other data in the format of the ICD, the returns from individual hospitals from which these data were derived frequently showed a blatant disregard for, or ignorance of, the requirements of the nomenclature. In the USA, Section V of the ICD was ignored completely, despite the fact that American psychiatrists had taken a prominent part in drafting it. Instead, a nosology drawn up by the American Psychiatric Association and published in 1952 as a Diagnostic and Statistical Manual (DSM-I) was in widespread use. Even so, there were serious local problems. The New York State Department of Mental Hygiene still insisted on retaining its own classification and influential psychoanalysts like Menninger (1963) argued forcefully that classifying patients into categories of illness did more harm than good and should be abandoned.

3. Reasons For Dissatisfaction With Psychiatric Classification

Many factors contributed to the unwillingness of psychiatrists to commit themselves to using a common nosology and to the general dissatisfaction with all classifications. Most fundamentally, there was no rational criterion available to enable the psychiatric community, faced with the need to choose between several alternative nosologies, to decide which was the best, or most valid. In other branches of medicine the old diagnostic concepts and nosologies inherited from the eighteenth century or before had largely been replaced by new concepts and classifications based on at least a partial understanding of etiology. Phthisis had been replaced by pulmonary tuberculosis, myxoedema by hypothyroidism, and Mediterranean anaemia by thalassaemia, and no physician doubted the validity of these new concepts or the value of classification. Psychiatry, on the other hand, remained as ignorant as ever about both the pathology and the underlying causes of most of its common disorders, and so was compelled to continue defining these disorders by their syndromes, just as Sydenham had done in the seventeenth century, and went on using ancient Hippocratic terms like mania and melancholia.

A further problem was that many of these disorders appeared to merge into one another. A few conditions, like Down’s syndrome and Huntington’s chorea, did seem to be clearly demarcated but most were not. Patients with the characteristic symptoms of one disorder not infrequently presented with the typical symptoms of a quite different disorder on a subsequent occasion, and it was commonplace to encounter patients with a mixture of the characteristic symptoms of what were supposed to be unrelated disorders. Such problems bedevilled the distinction between, for example, schizophrenic and manic depressive psychoses, and between depressive illnesses and anxiety states, and resulted in persistent controversies either about how best to distinguish the disorders in question or, even worse, to doubts about the validity or usefulness of the diagnoses themselves.

The combination of these two problems—the lack of understanding of etiology or of any demonstrable pathology, and the absence of obvious boundaries between several of the most common syndromes— meant that, in many instances, there was no rational means of choosing between alternative diagnostic concepts or alternative nosologies. In this situation most psychiatrists, understandably, opted to use whichever diagnostic terms and whichever classification were most familiar, particularly if they were their own.

Other considerations also contributed to the problem. Few of the diagnostic terms in common use held important therapeutic implications. Before the 1960s there were few treatments that were only effective in a single diagnostic category. Electroconvulsive therapy was effective in most acute psychoses—schizophrenic, melancholic, or manic. Chlorpromazine and the other new neuroleptic drugs appeared to have a similarly broad range of efficacy. Most influential of all, particularly in North America, was the fact that a diagnosis had ceased to be a necessary pre-condition to treatment, because psychoanalytically oriented psychotherapy was widely regarded as an effective and appropriate treatment across almost the whole range of psychiatric disorders from mild character disorders to florid schizophrenia. The implications of diagnoses for prognosis were equally tenuous. There were, to be sure, well established differences in outcome between populations of patients with different diagnoses, both in the short and the long term. On average, the course and outcome of schizophrenic and affective psychoses, or of depressions and anxiety states, were significantly different. But the overlap was very substantial, and as a result a diagnosis was of only limited value in predicting outcome in individual patients. By the early 1960s it was also becoming apparent that the reliability of psychiatric diagnoses, even those of experienced psychiatrists, was often alarmingly low, and that even under optimal conditions disagreements were commonplace—either because the patient gave different information to different interviewers, or because that information was interpreted in different ways, or because there were covert differences in the interviewers’ diagnostic concepts.

Finally, many psychiatrists and psychologists were concerned about the harmful effects diagnostic labels often had on their patients. Labels like hysteria or schizophrenia sometimes encouraged hospital staff to treat the patients in their care as ‘hysterics’ or ‘schizophrenics’ rather than as distressed people. Several diagnostic terms also had pejorative overtones, lowering the patient’s self-esteem even further than their illness itself had already done and encouraging relations, neighbors, colleagues, and employers to belittle them.

4. The Response To These Problems

The most important recommendations in Stengel’s report to the WHO were that all diagnoses should be explicitly shorn of their etiological implications and regarded simply as ‘operational definitions’ for specified types of abnormal behavior, and that the next revision of the ICD should have an accompanying glossary ‘available from the beginning in as many languages as possible’ providing a brief description, or thumbnail sketch, of all the disorders in its nomenclature. Other, mainly American, psychiatrists advocated operational definitions as a means of improving reliability. The problem with the brief descriptions Stengel was advocating was that, like traditional textbook descriptions, they described the typical features of a clinical syndrome without making it clear which combinations of those features were adequate to establish the diagnosis in question. A typical textbook description, or the thumbnail sketch of a glossary, lists the typical symptoms of disease X as A, B, C, and D, and often goes on to say that E and F are sometimes present as well. But it does not spell out which combinations of those symptoms are adequate and which are not to establish diagnosis X, and as a result different diagnosticians make different assumptions, and so often end up making different diagnoses. An operational definition is simply a statement, or set of statements, which makes it clear for all possible combinations of A, B, C, D, E, and F which are and are not adequate to establish diagnosis X.

The 1960s also saw the introduction of multi-axial classifications. The Swedish psychiatrist Essen-Moller had long advocated a classification with separate axes for symptomatology (i.e., the syndrome) and etiology, primarily as a means of resolving the difficulties that inevitably occur in a classification in which some disorders are defined by their symptoms (e.g., mania) and others by their presumed etiology (e.g., puerperal psychosis). As he pointed out, having separate axes would make it possible to identify all manic illnesses, whether or not they occurred shortly after childbirth, and all puerperal illnesses regardless of their symptomatology. Shortly afterwards, a WHO working party, set up to produce some much needed recommendations for an improved classification of childhood disorders in the ICD, made a firm recommendation that a tri-axial classification with separate axes for the clinical syndrome, the child’s intellectual level and ‘aetiological and associated factors’ should be tried out and, if these pilot studies proved successful, incorporated into the next revision of the ICD.

4.1 The Eighth And Ninth Revisions Of The International Classification (ICD-8 And ICD-9)

Stengel’s report and the recommendations of its own expert committee were taken seriously by the WHO, and the next (eighth) revision of the International Classification, which was published in 1965 and came into use in 1969, was a considerable improvement on its predecessors. For the first time all psychiatric disorders were brought together in the Mental Disorders section of the classification, thus making it self-sufficient and easier to use. The sections dealing with personality disorders, sexual deviations, mental retardation, alcoholism and drug addiction were all expanded and recast and a new group of reactive psychoses was introduced at the request of Scandinavian psychiatrists. Even more important, the American Psychiatric Association was persuaded, despite considerable domestic opposition, to commit itself to using the nomenclature of ICD-8 instead of that of its own Diagnostic and Statistical Manual. The German and Scandinavian psychiatric societies also agreed to use the new nomenclature so that, for the first time, all the major contributors to the psychiatric literature, with the notable exception of France, were committed officially to using the same classification. Unfortunately, the committee responsible for producing the companion glossary that Stengel had so strongly recommended did not start work until 1967 and the glossary was not published until 1974. The hiatus was partly filled, though, by the publication in 1968 of two national glossaries—American and British—to the nomenclature of ICD-8.

The ninth revision (ICD-9) was published in 1978, and this time the companion glossary was available from the beginning. But despite a series of eight international seminars held between 1965 and 1972, attended by psychiatrists from over 40 countries, there were no major changes. A few new categories, like nondependent abuse of drugs, were added in recognition of the broadening scope of psychiatric practice, a clearer distinction (reminiscent of Plato’s distinction between natural and divine madness) was drawn between organic psychoses and other psychoses, and there was a substantial expansion of the number of categories allocated to disorders of child- hood, but that was all. None of the fundamental problems, like the low reliability of diagnostic assignments, the tensions created by the incorporation of etiological assumptions into diagnostic terms, and the increasingly complex and chaotic classification of depressions, had been tackled effectively. It was also clear why this was so. The need for international consensus—the agreement of nearly 200 sovereign states to use the same nomenclature—was making it almost impossible to introduce radical changes; and an increasing number of sometimes incompatible terms were being added to the nomenclature with each new revision, because individual psychiatric societies always fought for the inclusion of their own favourite terms more persistently than they were prepared to oppose the attempts of others to do likewise. It was becoming apparent, in other words, that if fundamental improvements were ever to be achieved it was unrealistic to expect the WHO to be the driving force.

5. The Third Edition Of The American Psychiatric Association’s Classification (DSM-III)

For the research community the most important shortcomings of all classifications of mental disorders, national and international, were the low reliability and uncertain validity of most of their diagnostic terms. The Washington University Department of Psychiatry in St Louis in the USA reacted to this situation by publishing what came to be known as the Feighner criteria—a set of operational definitions, intended for research use, of the 15 major syndromes they regarded as having been adequately validated (Feighner et al. 1972). Other rival sets of definitions, like the Research Diagnostic Criteria, soon followed, and in this climate the American Psychiatric Association committed itself to producing a new, third, edition of its Diagnostic and Statistical Manual. When it was finally published in 1980 this DSM-III was dramatically different from any previous nosology of mental disorders. It was multi-axial with independent recording systems for five different kinds of information—the clinical syndrome or syndromes; associated personality or developmental disorders; associated physical disorders; the severity of recent psychosocial stresses; and the highest level of social functioning in the past year. The grouping of clinical syndromes was also novel, with all affective disorders brought together and the traditional distinction between psychotic and neurotic disorders abandoned. Even more provocatively, many of psychiatry’s most hallowed terms, like hysteria, manic-depressive illness, and even psychosis and neurosis, were discarded and replaced by unfamiliar and rather unattractive terms like somatoform disorder, substance use disorder, and paraphilia. A brave but only partially successful attempt was also made to define the term ‘mental disorder.’ Finally, and most important of all, nearly all the 200 categories of disorder in the manual were provided with operational definitions, and it was demonstrated in extensive field trials that this made it possible to achieve higher reliability than had ever been attained previously.

None of the innovative features of DSM-III was entirely novel. The advantages of both multiple axes and operational definitions had been pointed out long before and both had already been introduced into clinical practice in some centers. The concept of hysteria had been under attack for over a generation, as had the distinction between psychotic and neurotic disorders. What was new was that a powerful professional body like the American Psychiatric Association was prepared to commit itself and its members to all these innovations simultaneously. For a time the fate of DSM-III hung in the balance, for in the eyes of many older and still influential psychiatrists its authors had gone too far. But before long its advantages were increasingly apparent and eventually it became a best seller, not only in the USA but worldwide.

With hindsight the adoption of DSM-III was a crucial step in the development of a scientific nosology of mental disorders. Certainly, all subsequent nosologies, national and international, have been deeply influenced by it. Of all its innovative features the commitment to operational definitions was the most important. Not only did this result in a major improvement in the reliability of psychiatric diagnoses and eliminate the unsuspected differences in diagnostic criteria that had previously bedevilled international comparisons, they made it explicit that diagnostic terms like schizophrenia were simply convenient designations for more or less arbitrary groupings of symptoms.

6. Contemporary Nosologies

6.1 The Fourth Edition Of The American Psychiatric Association’s Classification (DSM-IV)

The American Psychiatric Association published a revision of DSM-III in 1987 and a new fourth edition (DSM-IV) in 1994. Neither abandoned any of the fundamental features of DSM-III. For the most part they simply corrected the anomalies that had come to light, reworded several operational definitions in order to improve their reliability or to conform with changes in expert opinion, and introduced a few additional categories of disorder. A new ‘clinical significance’ criterion was also added to the definition of many disorders in DSM-IV to counter the criticism that too many people met the DSM-III or DSM-IIIR criteria for common conditions like major depression and phobic disorder and that many of them were not significantly handicapped.

The authors of DSM-IV also endorsed and expanded the rather cumbersome DSM-III definition of ‘mental disorder’ as ‘a clinically significant behavioral or psychological syndrome or pattern that occurs in an individual and that is associated with present distress or disability or with a significantly increased risk of suffering death, pain, disability, or an important loss of freedom.’ In addition, it was stipulated that ‘this syndrome or pattern must not be merely an expectable and culturally sanctioned response to a particular event, for example, the loss of a loved one,’ and that ‘whatever its original cause, it must currently be considered a manifestation of a behavioral, psychological, or biological dysfunction in the individual,’ thereby emphasizing that ‘neither deviant behavior nor conflicts that are primarily between the individual and society are mental disorders unless the deviance or conflict is a symptom of a dysfunction in the individual.’ The introduction to DSM-IV also emphasized, rather more succinctly, that ‘the term mental disorder unfortunately implies a distinction between mental disorders and physical disorders that is a reductionistic anachronism of mind body dualism. A compelling literature documents that there is much physical in mental disorders and much mental in physical disorders. The problem raised by the term mental disorders has been much clearer than its solution, and, unfortunately, the term persists in the title of DSM-IV because we have not found an appropriate substitute.’

6.2 The Tenth Revision Of The International Classification (ICD-10)

The WHO published a tenth revision of the International Classification in 1992, after extensive consultations and field trials in several different parts of the world. Although the titles and definitions of many of the disorders listed in the Mental and Behavioural Disorders section of this ICD-10 are different from those of either DSM-III or DSM-IV its debt to the former is obvious. The general format of the two nosologies is very similar and, crucially, ICD-10 provides ‘diagnostic criteria for research’ for every category (i.e., operational definitions) as well as ‘clinical descriptions and diagnostic guidelines’ for ordinary clinical use (WHO 1992, 1993). This was only possible because, by the time the member states of the WHO were called upon to endorse the draft text of ICD-10, there had been time for the international psychiatric community to come to terms with the radical innovations of DSM-III and to appreciate their advantages. Significantly, though, no attempt was made to define the term ‘mental or behavioral disorder.’ Indeed, there is an apologetic note in the introduction explaining that ‘the term disorder is used … so as to avoid even greater problems inherent in the use of terms such as disease and illness. Disorder is not an exact term, but it is used here to imply the existence of a clinically recognizable set of symptoms or behavior associated in most cases with distress and with interference of personal functions.’

At present, therefore, two nosologies of so-called ‘mental disorders’ are in widespread use—DSM-IV and DC-10. In their essentials they are very similar to one another. In points of detail, particularly the titles and subdivisions of individual syndromes and the detailed definitions of these disorders, they are often different. The American classification is used mainly in North America and by research oriented psychiatrists further afield. The International Classification is used elsewhere, particularly in Europe and developing countries, and by all governments for their formal statistical returns to the WHO. In some ways it is regrettable that there are now, and will be for the foreseeable future, these two rival nosologies. However, the confusion generated by two rivals is much less than that produced by a dozen, the nightmare of the 1950s. There are also advantages in having some choice. The coexistence of alternative titles, subdivisions, and definitions for several of the major syndromes helps to emphasize their arbitrary nature. It also invites formal comparisons, which may help to provide the evidential base for better categorization and definition in the future. It is likely, too, that for financial reasons the American classification will be revised more frequently in the future than the International Classification, and some sections of the nosology may need to be revised quite quickly in the light of etiological research.

7. Shortcomings Of Contemporary Nosologies

7.1 Uncertain Validity

The most obvious weakness of both DSM-IV and ICD-10 is the limited evidence for the validity of most of their constituent categories. The combination of a limited understanding of the etiology of most mental disorders and the lack of evidence for natural boundaries or ‘zones of rarity’ separating their clinical syndromes means that most existing diagnostic concepts and definitions are not securely based. There are exceptions, of course. The etiologies of, for example, Huntington’s chorea, Down’s syndrome, the Fragile X syndrome, and Creutzfeldt Jacob disease are all well established, and enough is known of the neuropathology and etiology of Alzheimer’s disease and vascular dementia, and of the various syndromes associated with the use of psychoactive substances, for them to be reasonably secure. But the same cannot be said for most of the other major syndromes. For many, there is cogent evidence that genetic factors are making a major contribution to their etiology, and that life experiences and other environmental factors are also involved. It seems increasingly likely, though, that many different genes, each on its own of relatively small effect, are responsible for the genetic basis of most syndromes, and that some of these genes are risk factors for more than one syndrome—for affective as well as schizophrenic psychoses, for example, or for anxiety states as well as depressions. Some adverse life experiences appear to predispose to a similarly broad range of apparently unrelated disorders.

7.2 The Dimensional Alternative

An etiological framework of this kind does not easily lend itself to a categorical classification of disorders. For this and other reasons the possibility of representing variation in symptomatology by dimensions rather than categories needs to be considered. The idea is not, of course, new. Wittenborn developed an elaborate multidimensional representation of the phenomena of psychotic illness in the mid-twentieth century and other psychologists and psychiatrists have subsequently advocated and used dimensional models to portray depressive or chronic schizophrenic illnesses, or even mental disorder as a whole. Dimensional classification solves at a stroke all the problems associated with boundaries and comorbidity, and may also be a powerful means of predicting outcome. Even so, there is no realistic prospect at present of any formal national or international classification adopting a dimensional format. Medicine as a whole is too firmly committed to categories of illness, and in any case dimensions are better suited to the portrayal of variation in populations than to day-today decisions about the treatment of individual patients. This does not mean, though, that dimensional systems may not have an important role for research purposes, or as an experimental alternative to a traditional nosology.

8. Future Developments

In time, as understanding of the underlying determinants of most of the major syndromes slowly accumulates, mental disorders will cease being defined by their syndromes and be defined instead by their neuropathology, or perhaps by a characteristic combination of genetic risk factors. This may not happen quickly, and when it does the classification of mental disorders—if they are still so called—may bear little resemblance to our present nosologies. In the immediate future, however, major change seems improbable. The American Psychiatric Association has already committed itself to producing a fifth edition of its Diagnostic and Statistical Manual (DSM-V), probably some time between 2007 and 2010, and at present it seems unlikely that it will differ in any radical way from DSM-IV.

Bibliography:

1. American Psychiatric Association 1980 Diagnostic and Statistical Manual of Mental Disorders (DSM-III), 3rd edn. American Psychiatric Association, Washington, DC
2. American Psychiatric Association 1994 Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), 4th edn. American Psychiatric Association, Washington, DC
3. Feighner J P, Robins E, Guze S B, Woodruff R A, Winokur G, Munoz R 1972 Diagnostic criteria for use in psychiatric research. Archives of General Psychiatry 26: 57–63
4. Kendell R E 1975 The Role of Diagnosis in Psychiatry. Blackwell, Oxford, UK
5. Menninger K A 1963 The Vital Balance: The Life Process in Mental Health and Illness. Viking Press, New York
6. Stengel E 1959 Classification of mental disorders. Bulletin of the World Health Organization 21: 601–63
7. World Health Organization 1974 Glossary of Mental Disorders and Guide to their Classification: For Use in Conjunction With the International Classification of Diseases, 8th revision. World Health Organization, Geneva, Switzerland
8. World Health Organization 1978 Mental Disorders: Glossary and Guide to their Classification in Accordance with the Ninth Revision of the International Classification of Diseases. World Health Organization, Geneva, Switzerland
9. World Health Organization 1992 The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. World Health Organization, Geneva, Switzerland
10. World Health Organization 1993 The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic Criteria for Research. World Health Organization, Geneva, Switzerland