Formerly Professor of Psychiatry at the University of Sydney and the Royal Prince Alfred Hospital, Sydney Member of the Order of Australia

Professor Peter Beumont, who died aged 65, held the Chair of Psychiatry at the University of Sydney from 1975. His rich professional life requires that his achievements should be recorded in this Journal: not only had he achieved a truly international reputation for his research in eating disorders, but his early psychiatric career was spent in Britain and he was a Fellow of the Royal College of Psychiatrists.

It was moving that the news of his death should be broken to his close friends and colleagues at a meeting of the Eating Disorders Research Society in Ravello, Italy, which he would normally have attended. This melancholy task was undertaken by Professor Stephen Touyz from Sydney, who told us he had died the previous evening (1 October 2003) after several weeks of illness.

Peter was born Pierre Joseph Victor Beumont in Pretoria, South Africa in 1937. His father came from Belgium, his mother from South Africa. His early education was at Christian Brothers College in Pretoria. He studied at Rhodes University, Grahamstown in the Cape Province and Pretoria University from where he graduated MBChB in 1962. He moved to London with his wife Daphne in 1965.

In the late 1960s, when at the Maudsley, Peter served as registrar on the Metabolic Unit where he first came in contact with patients with anorexia. The result was his MPhil thesis on male anorexia nervosa, as well as an enduring fascination for the subject. Although my role was a minor one, we were colleagues and it pleases me to think I may have helped initiate his conversion experience. After the Maudsley, Peter moved to Oxford where he came under the influence of Professor Michael Gelder and Professor Geoffrey Harris, the pioneer neuro-endocrinologist. There he wrote his MSc thesis on the effects of major tranquillisers on the hypothalamic-pituitary-gonadal axis.

While in England, he achieved the first of many academic distinctions by winning the Gaskell Gold Medal and Prize of the Royal College of Psychiatrists. He was later, in 1994, to win the Max Planck Research Award jointly with Professor K. Pirke from Trier.

In 1971, Peter returned to South Africa and gained rapid promotion from Senior Lecturer to Acting Head of Psychiatry, Groote Schuur Hospital and the University of Cape Town. He finally left South Africa in 1975 to take up the Chair of Psychiatry at the University of Sydney with clinical facilities at the Royal Prince Alfred Hospital, where he established his Eating Disorders Unit.

An appraisal of Peter's research is no easy task in view of his prodigious output of original articles, chapters and books, mostly on anorexia nervosa. He succeeded in giving breadth to a subject that many would formally have considered narrow. It pleased him to point out that the syndrome, first described in the 1870s, has stood the test of time better than most of the psychiatric diagnoses then in vogue. He delighted also in the researcher's opportunity to view and study anorexia nervosa from several different perspectives, none fully satisfactory but each contributing to our understanding: the ascetic and behavioural, the perceptual and psychodynamic and finally, perhaps, the endocrinological. Only a few examples of his contributions can be chosen.

From his days at the Maudsley and Oxford he retained a strong interest in the endocrinology of anorexia nervosa. His own studies on the infusion of luteinizing hormone releasing hormone (LHRH) in anorexic patients showed that their malnutrition caused a reduced gonadotrophin secretion that progressively improved after refeeding. This work helped to confirm that the endocrine disturbances in anorexia nervosa are secondary phenomena and that their pattern is quite different from that of established pituitary disease. This was the final evidence that physicians should abandon the notion of Simmonds’ Cachexia, which attributed anorexia nervosa to pathology of the pituitary gland. In an editorial in the Lancet in 1979, he roundly castigated many physicians for subjecting anorexic patients to unnecessary endocrine investigations, which deflected them from the treatment they really needed.

Peter's observations in 1976 led to his sub-classification of anorexia nervosa. He demonstrated the value of dividing anorexic patients into two types according to their principal method of inducing weight loss. In the better-known type, patients stubbornly avoid food and he named them ‘ dieters’. His new type consisted of patients who resort to vomiting or laxative abuse and he called them ‘vomiters and purgers’. He found that the two sub-types differed as regards their personality, the occurrence of complications and their response to treatment. This approach was largely incorporated in the 1990 DSM-IV classification with a slight difference of names for the two forms of anorexia nervosa; the restricting and the binge-purging types.

Possibly because of a strict adherence to his own sub-division of anorexia nervosa into dieters and purgers, he later resisted the coming of the new syndrome of bulimia nervosa, which was described in 1979. As late as 1988, he wrote an article headed ‘bulimia: is it an illness entity?’ He answered this question in the negative. He had two main reasons for rejecting the concept of bulimia nervosa. First, he disagreed with the view that the excessive concern shown by bulimic patients with their weight was sufficiently pathological to determine their behaviour. After all, their weight often fell within a normal range and thus, he argued that these patients did not really differ much from the majority of young women in modern technologically developed societies. Secondly, he thought there was too great a variation between the kinds of behaviour adopted to counter the overeating: vomiting or purgation was not always present, some exercised to excess, others fasted between bouts of overeating. He also revealed his gift as a polemicist. He disliked the term ‘bulimia’, which he considered disparaging to patients, relying on his considerable knowledge of foreign languages to make his point. He felt it was too reminiscent of words to describe animal-like feeding (the German ‘fressen’ and the Dutch ‘vreten’). In time he came to terms with the new disorder, bulimia nervosa and its name, but nevertheless had a new foray in the field of classification.

One of Peter's most original ideas involved the classification of ‘ eating disorders’. In the early 1990s, he sought to abolish this term and have it replaced by ‘dieting disorder’. He saw this as a single disorder incorporating the salient behaviour of attempted dieting. All patients sought to control their weight, usually with harmful consequences. His proposed revised classification would thus consist of a single category. This, in turn, would be subdivided along three dimensions: the degree of weight loss, the presence of bulimic behaviour and the extent of ‘ purging’ behaviour (in its American sense of vomiting as well as laxative use). He saw it as an advantage that the separate syndrome of bulimia nervosa would be unnecessary. The diagnostic entity of anorexia nervosa would remain as the central disorder for which he preferred the German word Magersucht - seeking after thinness. His dimensional approach would avoid, he thought, the overlap of diagnostic categories in the same patient. The term ‘ dieting disorder’ did not catch on, something he probably foresaw when he ended his 1994 article with a quotation from John Maynard Keynes ‘ the difficulty lies not in the new ideas but in escaping the old’.

Peter was also a sound and imaginative clinical researcher. Again, it is difficult to do justice to his full output but two examples will be chosen - the patients’ propensity for excessive physical activity and the occasional need for compulsory treatment.

From the earliest description of anorexia nervosa, it was recognised that, in spite of their emaciation, the patients were prone to restlessness and overactivity. It is apparent that this behaviour has in recent years become more common, aggravating the malnutrition and interfering with treatment attempts. This change in the illness may be due to the increased commitment to exercise in Western societies during the 1970s and 1980s. Yet little attention was paid to this phenomenon until Peter's article in 1994. Together with members of his team he described the salient behaviour and attitudes displayed by anorexic patients. He identified the addictive aspects of the overactivity and the distress experienced by the patients when prevented from carrying out their routines. He coined the term ‘debting’ to describe their careful titration of activity against their food intake. They will calculate the amount of calories expended in a day and then ensure that their total food intake contains the same or a lesser amount of energy. Previously, treatment programmes simply advocated the prohibition of exercise. The 1994 article advocated the introduction of an exercise component into standard refeeding programmes with the aim of rewarding patients for weight gain, improving their participation in treatment and educating them about healthy levels of exercise. The quality of this combined programme was demonstrated by the patients’ success in continuing to regain weight at an optimum rate.

The second example of Peter's clinical and therapeutic enterprise comes from his method of adapting the legislation in New South Wales, Australia, to the compulsory treatment of the minority of anorexia nervosa patients for whom it is essential. The legal aspects of compulsory treatments in New South Wales differ from those in the UK. In 1990, the Mental Health Act of New South Wales specified that mental illness should be defined by the presence of psychotic features, thus excluding anorexia nervosa, which does not meet the criteria of mental illness under this Act. However, a Guardianship Act was established in 1987 and was ingeniously applied in cases of anorexia nervosa requiring compulsory treatment by appointing a guardian who may be a private person or a public guardian for a period of up to one year. Peter and his team evaluated the compulsory treatment under guardianship in 15 anorexia nervosa patients admitted to eating disorders units between 1991-1994, and compared them with 73 patients admitted voluntarily. The guardianship sample comprised a more severely ill group and a higher proportion of patients had a history of vomiting and purging. The mean weight gain during the compulsory admission was 10 kg. Although this represented a shortfall from target weight in the majority of the patients, the overall response to treatment was considered satisfactory and indeed life-saving in most. This constituted the first systematic study worldwide of anorexia nervosa patients admitted compulsorily and served as a stimulus for at least one other study in the UK.

It is always a pleasure to read Peter's scientific and clinical articles. He had a fine lucid narrative style and his reviews of literature were exemplary. He did not confine himself to a search of recent articles available on the Internet, but always did justice to the old masters, thus displaying his clear grasp of the historical aspects of his subject. But he was at his best when engaged in debate from the floor of a conference hall, when he would often question ideas and conclusions of speakers, or even cross swords with them. Although he could be provocative, he was never boring. Indeed, boredom was an attribute he would not tolerate in colleagues. Hence, he could be a harsh reviewer of a book if the writers were insufficiently stimulating or original with their ideas. He was a gifted linguist, fluent in Afrikaans, French, German, Dutch (and it is said Zulu), as well as English.

Peter would have been the first to concede that, especially in recent years, the research output from Sydney owed a great deal to a fine team of clinical and scientific investigators who shared his passionate interest in anorexia nervosa. He was proud of their accomplishments. In part, this was the stimulus for a conference he organised at Palm Cove near Cairns in Queensland, jointly with Professor Karl Pirke in 1997, in which he displayed the wide range of original research emanating from his department as well as work from German colleagues.

I am indebted to Stephen Touyz for providing details about the manner in which Peter enhanced the provision of services for patients with eating disorders, not only in Sydney, but throughout New South Wales and further afield. In 1991, he began to develop a state-wide service for rural and remote patients, as well as a shared-care programme with the Royal College of General Practitioners. He was also a driving force in developing treatment guidelines for patients with eating disorders and helping carer groups to establish themselves in Sydney. For this work and his academic contributions, he was awarded the Order of Australia in 2001.

Peter was a regular and diligent attender at international conferences on eating disorders, which have proliferated over the years. This was useful for his European and North American colleagues who were thus reminded that research on eating disorders in the Antipodes is vibrant and of the highest quality. Whenever it was possible, Peter was accompanied by his wife Daphne, and on these occasions he found it easier to relax and allow himself visits to art galleries and museums of Egyptology. He is survived by Daphne, their three daughters, Natasha, Clare and Rochelle and his beloved grandchildren.