This supplement is devoted to the subject of personality disorder. All the contributions were first presented at the second annual conference of the British and Irish Group for the Study of Personality Disorders (BIGSPD) (which was also supported by the Association of Therapeutic Communities) in Leicester between 31 January and 3 February 2001. The BIGSPD is part of a worldwide network of organisations linked to the International Society for the Study of Personality Disorders, and promotes research and teaching into all aspects of personality disorder. Although to many clinicians personality disorder remains a specialist subject, this supplement demonstrates that its tentacles extend to all parts of clinical practice and that it is almost impossible to be engaged in clinical psychiatry in any discipline without coming across the subject. Quite apart from the front-line treatment of personality disorder, also described in some depth at the BIGSPD meeting (Reference Bateman and FonagyBateman & Fonagy, 1999), there are few clinicians who can disregard the implications of personality disorder in the treatment of common conditions such as depression (Reference MulderMulder, 2002). Clinicians should recognise the part played by personality disorder in cases commonly described as ‘treatment-resistant’ but where the resistance is composed of comorbid personality disorder that requires separate identification and treatment (Reference TyrerTyrer, 2002).

The size of the problem is illustrated by Coid (Reference Coid2003, this issue) in examining personality disorder from the public health perspective. This shows that relatively small gains in prevention of personality disorder would achieve a disproportionate gain in forensic populations, despite the latter being the subject of so much research attention. Borderline personality disorder is often the clinical face of personality disturbance presented to the psychiatrist mainly because of its affective components (Reference CoidCoid, 1993), but in terms of public health it is the antisocial group that has the biggest public health impact.

The potential role of the child psychiatrist in identifying children with the antecedents of adult antisocial personality disorder is emphasised in the next article (Reference HillHill, 2003, this issue). A critical task is to separate this heterogeneous group so that those who are most at risk of developing antisocial personality characteristics become the focus of interventions, and Hill makes an important step towards achieving this task.

Once personality disorder is established it is sometimes forgotten when other mental illness, particularly a severe condition such as schizophrenia, is present. When the personality characteristics are prominent, there is a tendency for hard-pressed clinical staff to want to exclude such patients from in-patient care as they appear to be less deserving than other patients. This may be a mistake unless special attention is paid to treating the personality disorder as well as the mental illness. Tyrer & Simmonds (Reference Tyrer and Simmonds2003, this issue) demonstrate why so many of the general public show antipathy to those with severe mental illness: it is the minority with comorbid personality disorder (mainly antisocial and borderline) who create public disturbance and give all sufferers a bad reputation. There continues to be controversy over whether people with personality disorders should be treated primarily in institutional settings rather than in the community, and gains in symptoms and costs in the community may be offset by the impact of antisocial behaviour.

Although personality disorder is often considered to be very different from the normal variation in personality, Duggan et al (Reference Duggan, Milton and Egan2003, ths issue) show that the psychological constructs that have been well researched by psychologists for the past 50 years are just as applicable to those with mental illness, both mild and severe. Research in personality disorder tends to be empirical, but there is a corpus of theory that needs to be acknowledged also.

Therapeutic communities have the longest ancestry in the treatment of personality disorders, and the longer-term outcome of those admitted to one of these communities, Francis Dixon Lodge, is described by Davies & Campling (Reference Davies and Campling2003, this issue). The good follow-up rate is impressive, but the three patients who died from suicide or undetermined death illustrate what has now become a common finding: those who do not engage in treatment or who drop out early have a much worse outcome (Reference Crawford and WesselyCrawford & Wessely, 1998; Reference Killaspy, Banerjee and KingKillaspy et al, 2000); and significantly more of those who do not comply with treatment satisfactorily have a personality disorder compared with those who do comply (Reference Seivewright, Tyrer and CaseySeivewright et al, 1991).

The final two articles are concerned with aspects of personality disorder in learning disability. This is a relatively neglected subject but highly relevant to the concept of personality disorder. Personality disorder is defined as beginning in adolescence or early in adult life, as a pervasive behavioural condition that impairs relationships and function, and influences cognition, affectivity and impulse control. All of these apply to learning disability equally, and therefore personality disorder and learning disability occupy the same axis in DSM-III and its successors. This difficulty is illustrated by Alexander & Cooray's (Reference Alexander and Cooray2003, this issue) review of the prevalence of personality disorder in learning disability, which shows rates that vary more than 90-fold, with figures that would make any epidemiologist blanch in horror. Finally, Cooray and her colleagues (Reference Oliver, Cooray and TyrerOliver et al, 2003, this volume) show that when global function is affected by personality status, symptoms and learning disability, it is often difficult to get good reliability of function between raters from different backgrounds even when they are trained to some extent beforehand. The assessment of ‘relative incompetence’ is not an easy task.

The subject matter of this supplement ranges across acute psychiatry through to forensic and child psychiatry, learning disability and specialist care for personality disorders. In each of these, personality status has an important role to play. We hope that in illustrating its ramifications in this way, the supplement may make the reader think of the many other areas of clinical practice where personality disorder lies latent, with its importance yet to be discovered.