The British approach to care of elderly people has several virtues but can offer no pretence of being a rational system, having been fathered by expediency and mothered by tradition. However, its successes and failures have enabled us to identify the principles by which a rational system could be established, and in the last few years some of these principles have been incorporated. Rationality in the health and social services implies both a response to the important challenges of disease and disability and also the embodiment of cultural values in the attainment of ideologically determined objectives. A system that serves well the people of one country at one time may not meet the needs and wishes of another nation, or of a later generation.
It is difficult in only a few words to express adequately the cultural values that have underlain the development of the British system of care for the elderly, but some broad generalizations are possible. The elderly people of Britain over the last forty years have espoused what are commonly viewed as traditional British values. Traditionally we are a cautiously sociable but private breed; we like to meet friends and family but do not like to have to live with them. We value individuality and admire eccentricity. We dislike loss of dignity, of privacy or of emotional control. We recognize that there are fates worse than death. We like the basic structure of society to be smoothly but unobtrusively run and detest bureaucracy and regimentation. We like a quiet life and may endure an unpleasantness rather than suffer the disturbance necessary to remove it. Thus we are somewhat torpid by international standards but once roused we can be stubborn of purpose, and uncompromising of principle. Our elderly people have endured two wars and an economic depression. Their aspirations can seem low since however bad a present situation may seem, they have seen much worse. For all these reasons they are politically supine, and the intensely polarized politics of contemporary Britain make them much less of a credible political force than in some other countries notably the United States. With this cultural background the ideological aim of the health and social services for our elderly people is to enable those suffering from physical or mental disability to live where they would wish to live if they were not disabled. For the great majority this means living in a private house with no undue sense either of insecurity or of being a burden to others. Services for older people emphasize care in domiciliary and community settings rather than in institutions. Around 95% of people aged 65 and over live in their own homes.

1. THE SERVICES AS A SYSTEM
The primary characteristic, and virtue, of the British Health and Social Services is that they do comprise a system. Although there is local variation in the quality and intensity of services, the basic components of health care, including specialist geriatric departments and of domiciliary and institutional social services are locally available to every citizen in the country. Day-to-day health care for residents in the community is provided by General Practitioners each with a practice of about 2000 people of all ages. Access to social services is provided through social workers who can be contacted by the general practice team or directly by members of the public. Every citizen is registered with a general practitioner who maintains cumulative medical records on his or her patients, and through whom, as a matter of etiquette and custom, all contact with specialist health care is channeled. This role of the general practitioner as a 'final common pathway' for health care is an important safeguard for elderly people with multiple health and social problems and encountering multiple health and social service agencies.

2. THE SEPARATION OF HEALTH AND SOCIAL SERVICES
A second, and less commendable characteristic of the British system is the division between health and social services. Health services which include hospital and general practitioner care are provided, broadly speaking, from central government funds administered by regional and district health authorities whose members are appointed by central government. Social services are provided through a mixture of central and locally exacted taxation and administered by elected local authorities. Further complication is provided by housing being administered also by the local authority but by a different department from the social services, while social security, including pensions, is administered by a government department separate from health. Each of these separate agencies and departments has an agenda of its own in which the needs of individuals do not directly figure. Thus a social services department may measure its success by how many meals on wheels it has provided rather than on their nutritional value or whether they have been reaching the people who need them. A hospital department may pride itself on its low mean length of hospital stay without knowing whether this is provided by a high death rate or rapid transit into institutional care. Appropriate measures of audit are an important element in any rational system of　care but difficult to add as an afterthought to an empirical system or to sustain without a dedicated budget. There is much enthusiasm for audit in the British Health Services at present but also much uncertainty as to how it should be done. Audit by structure and process is easy enough, but audit of outcome is profoundly difficult in care of the elderly, for whom as we have already implied quality of life, defined on an individual basis, may be more important than mere survival.
The separation of health and social services can lead to unnecessary suffering for elderly people when each of the two agencies attempts to pass responsibility for care to the other's budget. This was the cause of much concern in the 1970s, but negotiations at a local level led to agreed operational criteria for responsibility in most districts. It is widely feared that the problem may resurrect itself in the implementation of the government's proposals for reform of the health and social services.
For many years, health and social services exemplified a conceptual as well as a merely administrative division. Historically, specialized care for elderly people grew in Britain out of the Poor Law, legislation concerned with the management of poverty rather than with the care for health.
In the context of poverty management the disabilities of old age, and the poverty of the afflicted, were seen as normal and inevitable consequences of an inexorable process of biological senescence. It was not until the 1940s and 1950s that the pioneer work of the early geriatricians demonstrated the potential relevance of medical interventions to age-associated disabilities. Thus while the geriatric and health teams increasingly adopted a therapeutic approach to disability, social service teams, whose traditions lay in the management of poverty, continued to follow the traditional prosthetic approach. This division was consolidated by the philosophy of the welfare legislation introduced to Britain after the Second World War. Social workers were significantly recruited from the ranks of the political left-of-centre and saw the services they deployed as instruments of social equity. They felt that poor people should be able to obtain on request anything that rich people could purchase for themselves. Social workers therefore objected to what they saw as an inappropriate 'medical model' of care that required that an old person with a disability should be assessed for remediable causes of disability before being provided with a prosthetic service. In vain did geriatricians urge that the therapeutic approach led to better care and greater independence of older people; as long as the rich were not subject to the same requirement it was seen as a classist policing of the rights of the poor. For a variety of reasons this troublesome mistrust between the medical and social work professions has gradually receded into history. The principle that multidisciplinary assessment of the needs of old people should at least pre cede the use of the ultimate prosthetic service of institutional care has been formally embodied in recent government legislative proposals [1]. There is some fear, however, that the government may see this chiefly as a cost containment procedure rather than a quality audit, in which case its effectiveness may not be as great as hoped by those concerned for the welfare of elderly people. Also the level of competence with which the assessments will be performed remains an open and controversial question.

3. FRAGMENTATION OF CARE
A third characteristic of the British system has been the fragmentation of tasks as well as of administration. This has largely been imposed by professional and other occupational groups, who have pressed for their work to be defined in terms of the tasks to be done rather than the objectives to be attained. In one example of the difficulties this can create, an elderly lady who had suffered a cerebrovascular stroke and for whom care at home was being arranged, had sixteen different people call on her in the first day, but the necessary supervision of her oral medication, one of the services she needed, was not in the job description of any of them. The more progressive districts are overcoming the deficiencies of such a system by various innovative schemes of case management. One such model scheme involves the nominal devolution of a budget to a designated key worker, usually a social worker, who is empowered to use the budget to attain care objectives by purchase of any appropriate mixture of statutory services and contractually agreed care from non-professionals including neighbours and the client's family. This scheme has undergone formal evaluation and been shown to be effective [2], but requires radical changes in the attitudes and modes of working of social services departments and their staff. An associated development has been the widespread creation of 'generic' care workers with permissive rather than the traditional restrictive job descriptions. For example a job description might be framed in terms of readiness to carry out the tasks of care that a concerned relative of an elderly person might undertake. Yet more innovative experiments are exploring the possibilities of pooling resources from the health and social service budgets.

4. GERIATRIC MEDICINE
The fourth characteristic of the British system has been the development of the specialty of geriatric medicine. Geriatric medicine has been a recognized specialty since its first consultant post was established in the 1950s. There are now 600 consultants in England and Wales alone with a planned expansion to 800, a population ratio of one consultant per 10,000 people aged 65 and over, by the end of the decade. As a recognized specialty geriatric medicine receives an allocation of National Health Service training and career posts provided from central funds and has to maintain an approved training programme administered by the Royal Colleges of Physicians. Within the last two years psychogeriatrics has attained recognized status as a subspecialty within psychiatry with analogous rights and obligations.
Although a specialty concerned with elderly people may seem an appropriate accompaniment to the growth in the elderly population its development in Britain has been pragmatic rather than rational. Part of the impetus has been the effort needed to make the fragmented administrative system of care available for elderly people actually work. Furthermore, a powerful influence on the origin of the specialty was the inheritance by the National Health Service in 1948 of the workhouse hospitals left over from the Poor Law institutions containing large numbers of elderly people that doctors were keen not to have to look after. The creation of geriatric medicine was a convenient way of 'hiving off' responsibility for these hospitals and the patients they contained [3]. Not surprisingly, therefore, geriatricians were initially recruited from the ranks of professional failure as well as professional idealism, and the stigma of being professionally second-rate has remained to a diminishing but detectable extent with the specialty ever since. Part of the problem has been that it has never been clear what a geriatrician does that other doctors could not do, and geriatricians do different things in different districts. To some extent geriatricians are characterized by the responsibility they accept rather than the tasks they carry out. The central responsibility is to see that in their health district old people receive the care they can benefit from. How this responsibility is discharged will depend on the range of services available and the activities of other doctors. Unfortunately this can give the impression that the geriatrician is there merely to do the tasks that other doctors are unwilling to undertake. For this reason geriatricians and others have devoted much thought to a more constructive way of defining the role of the specialty.
Since 95% of elderly people live in the community it might be asked whether it would be more rational to have a community-based rather than the hospital-based specialty of geriatric medicine that we have developed. This would however have constituted a breach of the principle that medical care in the community is the sole responsibility of general practitioners. This principle is regarded as fundamental to the system of British health care. Anxieties about the training and expertise of general practitioners in care of their elderly patients have inevitably arisen from time to time, and in the last five years a Diploma in Geriatric Medicine has been introduced as a postgraduate qualification for general practitioners who wish to demonstrate special expertise in this area of medicine. The diploma requires a period of postgraduate training in a department of geriatrics or psychogeriatrics and the passing of a written and clinical examination.
Historically, geriatric medicine began with the care of old people in long-stay nursing institutions. It was rapidly apparent that a proportion of those old people need not have been in such institutions if adequate rehabilitation had been provided. Once geriatricians began to develop systematized rehabilitation services it became clear that some patients had already suffered from being inappropriately treated during the acute phase of their illness. Thus the recent history of British geriatric medicine has been dominated by the provision of an increasing proportion of the acute care of elderly people in hospital, and to have increasing influence on their care in the community. One of the most important lessons of the development of geriatric medicine has been the recognition that hospital acute care needs, if it is to be maximally effective, to be provided in the setting of modern well equipped hospitals with full access to diagnostic facilities. This fact arises from the characteristics of disease in old age which include multiple pathology, and atypical presentation together with rapid deterioration if treatment is delayed and a high frequency of secondary complications (Table 1). Old people therefore require more sophisticated investigation and monitoring than younger people if comparable standards of diagnosis and treatment are to be provided [4]. Three main models of providing acute geriatric care have developed. In the traditional model geriatricians provide parallel acute services to those supervised by general physicians and the referring general practitioners decide which service to send a particular old person to. The rationale for this is based on the assumption that the general practitioner is in a position to judge the needs of the old person appropriately. This is an unreliable assumption since it requires an accurate diagnosis which for the reasons given above is often not available until hospital investigations have been carried out. Operational problems also arise because resources do not necessarily follow the need for them and the partition of the total workload between the separate geriatric and general medical services can be unpredictable. This operational problem is addressed by the age-defined model of geriatric care [5]. In this model, patients requiring hospital medical referral are directed to the geriatric or the general medical department on the basis of age. The defining age varies from 65 to 85 depending on the resources available to the two departments. This model is popular among geriatricians but has a dubious rationale and carries some dangers for old people. Insofar as it has a rationale it is to use age as a screening variable to separate those elderly people who need geriatric care from those who do not. The specificity and sensitivity of different ages for this screening purpose have never been established however, nor have the penalties for the patients of being inappropriately allocated. There is also evidence that geriatric departments are funded and staffed at poorer levels than general medical departments and so elderly people are effectively discriminated against. Under the new funding arrangements proposed for the National Health Service [6] this discrepancy is likely to increase and we are faced with the possible paradox that geriatric medicine which aims to foster the wellbeing of elderly people may be increasingly used as an instrument of discrimination against them. As implied above the present generation of the elderly in Britain, although comprising a large proportion of the electorate, are politically inactive and likely to be seen by administrators and politicians as easy victims of cost cutting.
The integrated model is a rational model based on the properties of disease in old age and on two assumptions. The first is that the needs of an individual should be assessed specifically for that individual not guessed at on the basis of his or her age, and second that old people may need and should have access to the full range of medical expertise not just that of a geriatrician. In this model medical patients regardless of age or perceived problem are referred to a single medical service and geriatricians are members of the multidisciplinary medical team assessing the needs of patients and providing what care is required [7]. Elderly patients are not channeled into geriatric or other specialist services until after they have been fully diagnosed and assessed. In addition to being rational and humane this model is more efficient since it requires the provision of only one pool of emergency hospital beds not two. This model is also particularly valuable in teaching hospitals where medical and nursing students profit from experience in providing care for unsegregated elderly people under the guidance of geriatricians. It seems also to be more successful at recruiting doctors into geriatric medicine as a career [8].
It is important to emphasize that in all these models, at the centre of specialist geriatric care is a multidisciplinary team of which the core members are doctors, nurses, social workers, physiotherapists and occupational therapists. Moreover, acute hospital care is only one aspect of the needs of elderly people, and it is the other aspects that spell out the need for specialist geriatric responsibility as an extension to (not separate from) the acute care (Table 1). The acute care, rehabilitation and long-stay care must function and be managed as a comprehensive and unified service for a defined population.

Table 1 Considerations for a rational geriatric service

5. FORMAL AND INFORMAL CARE
An important series of lessons had to be learned about the relationships between formal statutory care and the so-called informal care provided by family, friends and neighbours. It is commonly asserted that family care has been declining in Britain and that this is due to a disintegration of the family as a social unit. The availability of family care has certainly been declining since the 1940s but this has been due to families being smaller, young people being more mobile geographically, better housing so that more old people live on their own, and a greater proportion of women, who provide most informal family care, returning to work in middle life. All the evidence is that British families that are able to give care to their elderly relatives do so at least as much today as in the past. Certainly the contribution made by informal care is enormous, and surveys suggest that if it were to be withdrawn the statutory provision of residential care and sheltered housing would have to be more than doubled and geriatric beds increased by 60% [9].

6. RATIONING OF CARE
In principle old people have the same right of access to all forms of medical care as do younger patients. The claims that in Britain old people are explicitly excluded from renal dialysis or cardiological interventions, for example, are true, if at all, only for a few places, and such bans are unethical and probably illegal. Conversely a doctor is not legally or ethically obliged to institute expensive therapy for patients if he or she considers it not to be in their best interests. We are not obliged to give anti-influenza prophylaxis or nasogastric feeding to patients with advanced Alzheimer's disease nor necessarily to treat their bronchopneumonia. We have therefore been able to avoid some of the palpable overuse of medical interventions seen in some other countries that have led to demands for specific rationing of medical care away from elderly people. Unless the rising numbers of elderly people in our society is matched by a fall in levels of morbidity, or an increase in national wealth, the issue of rationing will undoubtedly loom in the future. This is a vast topic about which much has been written. Because of our established pragmatic traditions, I suspect that in practice it will be less of a problem in the field of medical care than in the options for social services. It may well prove particularly challenging, for example, where an old person wants expensive care in her own home rather than the cheaper option of institutionalization.

7. CONCLUSION
In this brief overview I have commented on some of the salient features, good and bad, of the British approach to care for elderly people. In general the system works well, but only because it is made to work by the dedication and enthusiasm of health and social services staff. Central to the success of any system will be compassion and respect for older people, and these are things that no government can legislate into existence.

REFERENCES
l) Secretaries of State for Health. Social Security, Wales and Scotland: Working for patients. Cm 555. London. Her Majesty's Stationery Office (1989).
2) Challis D Davies B: Community care schemes: a development in the home care of the frail elderly. In : Grimley Evans J, Caird FI (ed). Advanced Geriatric Medicine. 4. London. Pitman, 35-44 (1984).
3) Editorial. Geriatrics for all? Lancet, l:674-5 (1985).
4) Grimley Evans J: Institutional care. In: Arie T (ed), Health care of the elderly. London. Croom Helm, 176-93 (1981).
5) Horrocks P: The case for geriatric medicine as an age-related speciality. In: Isaacs B (ed), Recent advances in Geriatric Medicine 2. Edinburgh, Churchill Livingstone, 259-77 (1982).
6) Secretaries of State for Health. Social Security. Wales and Scotland: Caring for people. Community care in the next decade and beyond. cm 849. London. Her Majesty's Stationery Office (1989).
7) Grimley Evans J: Integration of geriatric with general medical services in Newcastle. Lancet, l:1430-3 (1983).
8) Parkhouse J, Campbell MG: popularity of geriatrics among Newcastle qualifiers
at preregistration stage. Lancet. 2:221 (1983).
9) Gritmley Evans J: Prevention of age-associated loss of autonomy: epidemiological aspects. Journal of Chronic Diseases. 37:353-63 (1984).