This article is about the management of eating disorders in which binge eating is a prominent feature. These disorders include bulimia nervosa, the most common eating disorder, and ‘binge eating disorder’, a provisional new diagnosis included in DSM–IV. In addition, binge eating is seen in anorexia nervosa and in many atypical eating disorders.

The category ‘somatoform disorders' was introduced comparatively recently in DSM–III and thereafter in ICD–10: it is the umbrella term currently favoured to cover a heterogeneous group of interrelated and overlapping syndromes, which have been given many names over the years. All these syndromes have in common the process of somatisation: that is, the presence of physical symptoms unexplained by physical disease, with variable degrees of distress and loss of function, about which the patient consults doctors.

In 1981 the condition that was later to be known as the acquired immune deficiency syndrome (AIDS) was described, and in 1983 its causative agent, the human immunodeficiency virus 1 (HIV-1), was isolated. The past 15 years have led to a growing awareness of the global nature of the epidemic and, in parallel with it, to the recognition of its medical, socioeconomic and psychological consequences. In the UK, almost 12 000 AIDS cases have been reported since 1982, and more than 25 000 cases of HIV-1 infection have been identified since 1984. Most cases of AIDS and HIV-1 have been in the Thames Regions, and it is expected that this geographical pattern will continue until the end of the decade.

There are many similarities in the experience and approach of general practitioners (GPs) and psychiatrists. GPs may spend some time in psychiatric posts before settling down as principals in their own practice, while some psychiatrists are members of both their own College and the Royal College of General Practitioners. Although there is great potential benefit for patients in this symbiotic relationship, GPs and psychiatrists work in different settings that require different techniques and time-scales. The professional work of both specialities has been profoundly affected by the National Health Service (NHS) reorganisation of the early 1990s. GPs have developed new relationships not only with psychiatric colleagues, but also with professionals of other disciplines such as psychologists, social workers and counsellors.

Although police surgeons (or forensic medical examiners; FMEs) usually assess fitness to be interviewed by the police (Norfolk, 1996), psychiatrists may also be asked for an opinion (Protheroe & Roney, 1996).

Of all the controversies over the different medications used in psychiatry, there is probably least dispute about the effectiveness of antidepressants in controlling depressive illness. However, the choice of which antidepressant to use is bedevilled by the spectre of potential toxicity from adverse effects, interactions and overdose. The unwanted effects may vary from mild and tolerable to potentially lethal. At one time it was relatively easy to become familiar with the problems surrounding the original two main groups of antidepressants – the tricyclic antidepressants (TCAs) and the monoamine oxidase inhibitors (MAOIs). Recent years have seen the introduction of a number of newer drugs, with a wide range of chemical structures and differing pharmacological activities. They also have a new spectrum of adverse effects and interactions. It is worth reviewing the major differences between old and new antidepressants, and to identify the areas in which clinical caution must be exercised in order to avoid pitfalls and maximise clinical benefit.

The development of the first effective antidepressants in the late 1950s marked a turning point in the treatment of depressive illness. In 1957 the monoamine oxidase inhibitor (MAOI) iproniazid was discovered by chance, while searching for new antituberculous drugs. One year later the tricyclic antidepressant (TCA) imipramine was introduced, having been developed originally as an antipsychotic. A number of other drugs were subsequently added to these two groups of antidepressants, which dominated the field for the next three decades.

In a 17- to 19-year follow-up study it was shown that patients admitted to the Maudsley Hospital, London (whose index episode marked their first psychiatric contact) had a 50% chance of readmission during their lifetime; those with previous admissions had a similar chance of readmission within three years. Less than one-fifth of the patients had remained well, and over one-third suffered severe chronic distress and handicap or had died unnaturally (Lee & Murray, 1988). Similar gloomy pictures were reported in a 15-year follow-up study of patients in London and Sydney (Kiloh et al, 1988) and an 11-year follow-up study of patients in Montreal (Lehmann et al, 1988). As a result of such findings, much emphasis is now put on the importance of continuation and maintenance treatment of depression.

Meta-analysis is the use of statistical techniques to analyse the findings of many individual analyses (Glass, 1977). It covers all aspects of the review process involving formulating relevant research questions, searching the literature, assessing the quality of studies and choosing relevant ones, extracting and combining the data (for review see Henry & Wilson, 1992; Wilson & Henry, 1992). Meta-analysis as part of a systematic review has advantages over a narrative review but there are problems in applying it in practice (Box 1).