Black Report 1 Concepts of Health and Inequality

1.1 Throughout history different meanings have been given to the idea of “health”. One is freedom from clinically ascertainable disease, which has been central to the development of medicine. Thus, in ancient Greece the followers of Asclepius believed that the chief role of the physician was to “treat disease, to restore health by correcting any imperfections caused by the accidents of birth or life”. (Dubos, 1960, p.109). Beginning with primitive surgical intervention and herbal treatment, a tradition was established which was to prove extraordinarily powerful, accelerating in the 18th century with the rise of science and again in the twentieth century as a consequence of the massive resources provided for research and innovation in medical technologies. The Cartesian philosophy of the body, conceived as a machine and the body controlled as a machine provided an impetus for scientific experiment and a stream of practical outcomes which for an increasing proportion of the population seemed to validate a mechanistic perspective.

1.2 There could be no doubt about the success with which the engineering approach in medicine had been applied. Medical education became concerned with the structure and functions of the body and with disease processes; and medical service became represented predominantly by the acute hospital with its concentration of technological resources. (Abel-Smith,1964). Some have argued that as a consequence this development in medicine distorted our understanding of the problems of human health and that there are alternative or complementary approaches which it is increasingly important to clarify and properly finance. (For example, McKedwn,1976.)

1.3 The relatively restricted and familiar use of the word “health” is therefore associated with the belief systems and the practice of medicine from which its origins can be traced. Health, which derives from a word meaning whole, is the object of the healing process. To heal is literally to make whole or to restore health. The structure of medicine and of the health services helps to sustain this meaning.

l.4 Much wider meanings have been given to “health” which hold major implications for the organisation of society and the pattern upon which personal life may be modelled. Thus to followers of the ideas symbolised in ancient Greece by the goddess Hygeia, rational social organisation and rational individual behaviour were all- important to the promotion of human health. It was an attribute to which men were entitled if they governed their lives wisely. According to them, “the most important function of medicine is to discover and teach the natural laws which will ensure a man a healthy mind in a healthy body.” (Dubos, 1960, p109). Implicit are ideas of the good life: not just freedom from pain, discomfort, stress and boredom, which themselves extend beyond the competence of clinicians to diagnose or treat, but positive expression of vigour, well-being and engagement with one’s environment or community. In some respects this more comprehensive approach reached its apogee in the definition of health adopted at the foundation of the World Health Organisation at the end of the Second World War as a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity”. Adherents of this more comprehensive approach to the meaning of “health” which is usually called ‘social’ have sometimes worked within medicine but sometimes outside it. In most countries there are movements for physical fitness and dietary control. Immunisation is a standard public health practice. And through direct and indirect health education and counselling higher standards of health are encouraged. These practices illustrate the wider meanings that are given to “health”. In the case of children this wider conception of health directs concern not only to the presence or absence of disease, but to development (physical, cognitive, and emotional). There is, moreover, abundant evidence for the interaction of disease and development in infants. Low birth weight babies show a higher mortality and also incidence of neurological and physical disorder (Birch and Gussow, 1970: 52) and, later in life, there is evidence for the aetiological significance of even mild undernutrition in inhibiting growth (Marshall, 1977: 118). It becomes relevant to look at evidence relating to acuity of hearing and vision in children, and at heights, weights and age at onset of puberty, even though none of these things are in any sense ‘disease’.

1.5 We consider that the different meanings of “health” and hence of national objectives in maintaining and promoting health are not given as much attention as they might be, and we discuss the implications of this view further in Chapter 8.

1.6 For the task which we were given, the “social model” of health is clearly more relevant than the “medical model” and we have therefore mainly followed it. However, the two models are not either exclusive or exhaustive, and each has merits. (Discussions based essentially on the “medical model” are given by Black (1979) and by Dollery (1978).) Conceptions of health and illness vary among different groups within a single society and between societies, as well as in any single society over time (Morris, 1975). It is in part for this reason that, as Mechanic (1968) and others have shown, “illness behaviour”, the response to symptoms and the tendency or reluctance to define any symptom as a health problem and to seek medical care – varies between cultural and social groups. Conceptions are in constant process of adaptation or revision. There are considerable cultural differences between the developing, and market, or planned societies. Changes occur by virtue of scientific discovery and innovation, and developments in professional judgements of objective needs and the status of different diseases and treatments. They also occur in response to the pressure of established interests, and the extent of public anxiety about illness, or safety, as well as the current level of demand for health, environmental and social services. Thus one result of research on the elderly and disabled, and the heightening of public interest and concern about their problems, has been that pain, discomfort, debility and different forms of incapacity come to play a more prominent part in social and medical conceptions. If we consider mental illness or mental handicap, or the history of “fringe” medicine to take very diverse examples, we can see how conceptions of health and illness have changed. The priority that is to be given different conditions depends as much on popular interest and concern as upon the judgements reached within medicine. And just as conceptions themselves may gradually change, elements within those conceptions are accorded different weight or priority. We make this point for two reasons. One is that our understanding of “health” will always be evolving, and we must be prepared to absorb new knowledge about changes in health and social conditions. The other is to make better judgements about the strengths and weaknesses of the health care services.

1.7 Within any general approach to the meaning of “health” views are reached about the seriousness of certain states of health.. The construction of the health care services and the priorities which are identified in their development reflect those views. To the extent that a mechanistic model of health holds sway, the health care services will give priority to such matters as surgery, the immunological response to transplanted organs, chemotherapy and the chemical basis of inheritance. Medicine comes to be structured according to a scale of values associated with such a model. The most sought after posts will be those at the heart of the model and medical education and medical careers are similarly influenced. Medicine is not, however, monolithic, as developments in paediatrics, obstetrics, psychiatry and rehabilitation, research in social aspects and in prevention indicate. However, once a conception of disease finds embodiment in a structure of service major changes become difficult to introduce. All professions tend to become over-committed to existing practice and their receptivity to the need for change is liable to become weak. The medical, nursing and other professions are like other professions in this respect. We are pointing out the uncomfortable fact that society cannot look to the professions working within the health services for an account of illness and health which is always as detached or as full as it might be. Indeed, particularised conceptions of illness and health (including their stages and severity) are already institutionalised in medical practice and the organisation, sub-divisions and administration of services.

1.8 Therefore, while the knowledge, experience and views of the health care professions are bound to play a predominant part in the debate, the extension of knowledge about the problems of human health and illness depends also on sources outside the health professions. Under the auspices of the medical and social sciences there needs to be a determined search for evidence of a wide variety of health conditions and their social, environmental and psychological as well as physiological significance.

APPROACHES WITHIN MEDICINE, EPIDEMIOLOGY AND SOCIOLOGY

1.9 In the last hundred and fifty years it could be said that the pursuit of health has increasingly been acknowledged to be a social and not merely a technical enterprise. In part this is due to the success of medical science in reducing mortality from infectious disease and thus directing attention towards chronic diseases of complex aetiology but also to the development of public health services, statistical studies of health, the work of epidemiologists in demonstrating the importance of living standards, protection from hazards and population limitation in improving health, and latterly the work of sociologists on the complex effects of the economy and different forms of social organisation, including the family, upon levels of health. (See for example Susser and Watson,1971; McKeown, 1976; Morris, 1975; Tuckett, 1976.) Biomedical research will continue to be necessary but ‘there is need for a shift in the balance of effort, from laboratory to epidemiology in recognition that improvement in health is likely to come in future, as in the past, from modification to the conditions which led to disease rather than from intervention in the mechanism of disease after it has occurred’. (McKeown, 1976, p179). Thus the sociological contribution is recognised to be, in part, to increase understanding of the social and socioeconomic factors which play a part in the causation of disease and, as in Brown’s work, to take the natural next step and relate these factors themselves to broader sociological concerns with the social structure (see Brown, 1978).

1.10 Some working in medical sociology would emphasise a different perspective. They would argue that their contribution is not only like that of social medicine, to contribute to the understanding of the origins of health and disease in the way people live together in society. ‘Disease’, they would argue, is a medical, not a sociological concept. Sociology is concerned with the social production of understanding, meanings, knowledge; with social structure and process; and with the behaviour of people. Sociologists will try to understand the failure to seek medical attention for what to the physician is a serious disease episode not in terms of simple irrationality but in terms of the individuals own (learned) coping mechanisms, social situation, and the meaning which he attaches to his symptoms. Hence there is a lot of interest in the social production of conceptions of health, in inconsistencies between lay and professional conceptions, and in conditions which are generated by different forms of social organisation.

1.11 While the perspectives adopted in the three – medicine, epidemiology and sociology – tend to be different they are subject to mutual influence and some of the most creative practitioners acknowledge the need to absorb or combine their strengths. Thus, nearly 40 years ago Sigerist (following Virchow long before) argued.

“The task of medicine is to promote health, to prevent disease, to treat the sick when prevention is broken down and to rehabilitate the people after they have been cured. These are highly social functions and we must look at medicine as basically a social science.” (Sigerist, 1943, p241).

THE CHOICE OF INDICATORS OF HEALTH AND ILL-HEALTH

1.12 We have argued that conceptions of health tend to vary in time and according to place, and that depending on their experiences and situation, societies, and groups within those societies, will tend to emphasise some things more than others. Science, however, demands precision, and different aspects of the meaning of health and ill-health have to be translated into operational terms, and applied systematically.

1.13 Measures of the “health” of populations can take many different forms. Among the most familiar are mortality rates, prevalence or incidence morbidity rates, sickness-absence rates, and restricted-activity rates. Each of these indicators poses problems of measurement and has its limitations. For example, overdue dependence on mortality rates can induce comparative indifference towards problems of chronic illness. Undue dependence on morbidity-rates can discourage interest in congenital and other permanently incapacitating conditions, as well as conditions affecting human well-being which fall outside the conventional classification of “morbidity”. Examples of the use of indicators are given below.

1.14 Partly because of the problems of measurement, but also (in the kind of analysis we have been invited to make) because of the need for time-series statistics, we have given precedence to mortality rates. However, we wish to call attention to the need for different measures of health in combination, and therefore to the importance of experimenting with indicators which allow relevant experiences among the population to be captured and examined in relation to the allocation of resources for health and the organisation of the health care system. Thus a combined indicator of pain and restricted activity is being developed (Culyer et al, 1972) and in current Canadian work indicators reflecting social, emotional and physical functioning. (Sackett et al, 1977). Again, the need to relate rather complex indicators of depression to the measurement of life events was felt to apply to the community generally and not just those selected for psychiatric treatment (Brown and Harris, 1978).

1.15 A distinction is often made between acute and chronic sickness and attempts are made to relate the utilisation of health services to these conditions. Thus during the early 1970s acute sickness was defined in the General Household Survey as “restriction at any time during a two week reference period of the level of activity normal to that person, caused by illness or injury”, and chronic sickness as “a long-standing illness, disability or infirmity which limits a person’s overall activity level” (GHS Introductory Report, 1973, p264). In the first case one question was asked, “During the two weeks ending last Sunday, did you have to cut down at all on the things you usually do because of illness or injury?” In the second case two questions were developed: “Do you suffer from any long-standing illness, disability or infirmity? (If yes) Does it limit your activities compared with most people of your own age?” In the 1977 Report a broader approach was adopted. Those designing the survey attempted to distinguish between short-term and chronic health problems. A number of questions were designed to identify persons who “had anything wrong with them” in the preceding 14 days which “was not connected with any permanent or recurring chronic health problems” and, secondly, “people who considered that they had a health problem all the time, or one that kept recurring.” (GHS, 1979, pp 80-84). It is clear from GHS reports for these years that choice of terminology in “indicator” questions can considerably affect the percentage of the population identifying themselves as having “Short tem” and “chronic” health problems. We are not sure that “omnibus” concepts of the kind developed in 1977-78 can be helpful in analysis unless they are broken down into the different kinds of meaning placed upon them by informants. Thus, the approach used in the GHS in the early 1970s stressed “limitation of activity” as a key criterion of ill-health – and this helped to produce a coherent and consistent set of data. After using a different approach in 1977 and 1978 the OPCS reverted to the previous approach in 1979. But whatever is the method adopted of distinguishing acute from chronic sickness we would also want to call attention to the dangers of treating any two sets of people so identified as distinct. Acute and chronic conditions in some patients are difficult to distinguish. Since the evidence suggests that the poorer groups are at greater risk of chronic sickness and disablement there are dangers of distorting conclusions reached about the characteristics of the “acute” sick if all those with chronic sickness are first excluded from any analysis.

1.16 Disablement is also an important related concept. In his work for the World Health Organisation Wood showed that disability “can be conceived of on at least two planes, functional limitation and activity restriction”. (Wood, 1975, P13). Like other authorities he is concerned to distinguish between impairment and disability. He defines impairment as “a generic term that embraces any disturbance of or interference with the normal structure and functioning of the body, including the systems of mental function” and disability as “the loss or reduction of functional ability and activity that is consequent upon impairment”. (Ibid. p2). The distinction between impairment and disability in this WHO classification also follows the practice adopted in a special Government survey (Harris 1971). While interpretation of the concept of disability varies it has been identified increasingly in recent years (as in the GHS surveys) with restriction of activity – which would include self-care, household management, occupational and social activities. Parallel with this trend has been a greater emphasis on treating severity of disablement irrespective of cause or sex or age. Most local authorities have made returns since 1975 which have distinguished the numbers of the physically handicapped according to severity of handicap. Non-statutory bodies are pressing for even wider application of this principle. “All disabled people: the old, young adults and children; the mentally and physic handicapped, those disabled at home as well as at work or in war; and those disabled from birth, after an accident, or a long illness must be treated alike. It is not the origins or type of disablement or age which should count”. This statement is made on behalf of a large number of organisations of and for disabled people (Disability Rights Handbook for 1980, Dec 1979, p48). Surveys of public opinion seem to endorse such statements (Louis Harris International 1974).

CONCEPTS OF INEQUALITY AND SOCIAL CLASS

1.17 The distribution of health or ill-health among and between populations has for many years been expressed most forcefully in terms of ideas on ‘inequality’. These ideas are not just ‘differences’. There may be differences between species, races, the sexes and people of different age but the focus of interest is not so much natural physiological constitution or process as outcomes which have been socially or economically determined. This may seem to be straight forward but the lengthy literature, and widespread public interest in the subject of  inequality, shows that factors which are recognisably or discernibly man-made are not so easy to disentangle from the complex physical and social structure in which man finds himself. Differences between people are accepted all too readily as eternal and unalterable. The institutions of society are very complex and exert their influence indirectly and subtly as well as directly and self-evidently. For some the concept of inequality also carries a moral reinforcement – as a fact which is undesirable or avoidable. For others the moral issue is non-existent or is relatively inconsequential. For them differences in riches or work conditions are an inevitable and hence natural outcome of the history of attempts by man to build society; and they conclude that the scope for modification is small and, besides other matters, of small importance.

1.18 Central to the development of work on inequality has been the development of concepts of ‘social class’. Populations are not simply differentiated uniformly according to income, wealth, housing, and access to education. The differentiation on one distribution tends to correspond with the differentiation on another and the population can be divided into ranks or strata, one above the other. Of course, societies may differ according to the number of strata or the distance between them and the ease with which the boundaries can be identified. And although the boundaries or dividing lines between the classes may be difficult to discern this does not make class divisions any less real. Social classes, then, may be said to be segments of the population sharing broadly similar types and levels of resources, with broadly similar styles of living and (for some Sociologists) some shared perception of their collective condition. There has been, and remains considerable controversy within sociology about the origins and relative importance of class in relation to social inequalities and social change.

THE PROBLEMS OF CHOOSING INDICATORS OF INEQUALITY

1.19 Traditionally inequalities in health have been portrayed through a characterisation of class obtained by ranking occupations according to their social status or prestige. In addition to occupation, a variety of factors may be said to play a part in determining class – income, wealth, type of tenure of housing, education, style of consumption, mode of behaviour, social origins and family and local connections. They are inter-related, but none of them should be regarded as a sufficient indicator of class. Historically occupation was selected as the principal indicator, partly because it came to be regarded as more potent than some other alternatives, but partly because it was most convenient for statistical analysis. Thus, occupation designates not simply type of work but tends also to show broadly how strenuous or difficult for health it is, what are the likely working conditions (for example, whether it is indoors or outdoors and whether there is exposure to noise, dust, or vibration) and working amenities and facilities, as well as level of remuneration and likely access to various fringe benefits. Pay will also determine family living standards, but while members of the family will not be exposed to some features of the working conditions experienced there are others which may affect them indirectly (like the risk of intermittent unemployment, or the stresses of disablement and of shift work).

1.20 As an indicator of social position a named occupation is convenient. Matters like income and wealth, style of living and family origins are difficult to explore and are less acceptable to the public as questions during a formal interview. On the other hand, the economic and industrial changes of recent years make it less likely that for any individual current occupation indicates his life-long occupational position. And in a society with increasing participation of women in the economy the occupation of an individual man is less likely to be a sufficient indicator of his family’s social position.

1.21 It was in the middle and latter part of the l9th century that the use of occupation to analyse the health of the population took hold. For example, in the Registrar General’s Office Farr and Ogle showed striking differences in mortality between workers in particular occupations as well as between occupations. (Annual Reports of the Registrar General in England, 1875 and 1885). By the turn of the century Seebohm Rowntree was able not only to trace in detail the sanitary defects of areas of York but he was able to compare the general mortality rates, infant mortality rates and heights and weights, of children of different ages in three areas of York, distinguished according to the proportions living below his poverty line, and compared with the ‘servant keeping classes’. (Rowntree, 1901, pp 182-221).

1.22 It was a natural next step for someone to attempt to construct a comprehensive classification for the purpose of analysis health experience. This was undertaken by Stevenson in 1911 mainly to analyse infant mortality. He divided the population into 8 social groups. Only the first 5 were ranked, and the other 3 groups, textile workers, miners and agricultural workers, were thought to be sufficiently important to deserve separate identification. With various changes the fivefold scale has survived until the present day. Since 1970 the classification used has been as follows:

• I. Professional (eg accountant, doctor, lawyer) (5%)
• II. Intermediate (eg manager, schoolteacher, nurse) (18%)
• IIIn. Skilled non-manual (eg clerical worker, secretary, shop assistant) (12%)
• IIlm. Skilled manual (eg bus driver, butcher, coal face worker, carpenter) (38%)
• IT. Partly skilled (eg agricultural worker, bus conductor, postman) (18%)
• V. Unskilled (eg laborer, cleaner, dock worker) (9%)

Percentages give distribution of economically active and retired males in each social class (GB 1971).

1.23 In some authorities’ view, it was easier to obtain consensus about the scaling of social class in 1910 than in the 1970s (Susser and Watson, 1971, pI06). The system of classification proved to be a powerful epidemiological tool and has been widely used out this century (Leete and Fox, 1977). At each census, however, there have been changes in the detail of the classification. The most important of these was in 1960. Members of the armed forces were excluded hence increasing the unclassified groups. Aircraft pilots, navigators and engineers for example, were changed from class III to II draughtsmen from II to III: postmen and telephone operators from III to IV and lorry drivers’ mates from IV to V. (See Reid, 1977, for an outline of changes and of other classifications.)

1.24 The successive classifications are published in detail (most recently OPCS 1970) and have been successful in displaying gradients in mortality experience and other conditions. However this occupational classification has met with strong reservations from sociologists, for whom class is a concept having important explanatory significance. The latest classification (OPCS, Classification of Occupations 1970, p x) aggregates ‘Unit groups’ of occupations according to the ‘general-standing of the Occupations concerned’. But, as Goldthorpe (1978, P1) and others have pointed out, the precise method of determining ‘general standing’ has never been disclosed and for that reason the adequacy of the classification cannot be assessed. After the Second World War, sociologists developed a classification more firmly based on social perception of occupational prestige. The Hall-Jones scale (consisting of 7 ranked categories) was adopted in a pioneering study of social mobility (Glass, 1954) and modified subsequently (identifying 8 ranked categories) (Townsend, 1979 Appendix 6). One of the chief merits of this scale was to distinguish different ranks of non-manual occupations.

1.25 It is however difficult to present really convincing alternatives. For example, even if a cross-section of the population is invited to rank occupations as the basis of a normative scale, they can hardly be asked to do so for the 20,000 or more distinguishable occupations of the employed population. Sociologists usually confine themselves to asking individuals about a small number of occupations, say 30, which are believed to be representative or at least common. Inferences are then made about the ranking of the remaining occupations. Quite how one moves to the remaining occupations, and identifies the numbers of ranks remains a problem. The procedure is more a mixture of presupposition with the partial representation of social perceptions. In the Oxford studies of social mobility Goldthorpe and Hope have shown how the ranking of 20 occupations can be related to the ranking of 860 by asking sub samples of informants to rank 2 groups of 20 occupations, one of them being the basic 20 and the other being a variable set of the same number (Goldthorpe and Hope, 1974, p 48-50). They asked people to rate occupations on 4 dimensions: (i) standard of living, (ii) prestige: (iii) power and influence over other people: (iv) value to society (ibid, pp 27-33). A scale with 124 categories, reduced for some users to 36, was produced.

1.26 At a later stage these 36 categories were divided into 7 classes, but were not to be thought of as necessarily having an entirely hierarchical ordering”. (Goldthorpe 1978 p4). (See also Goldthorpe et al, 1977,and 1978).

1.27 A comparison of the Hope-Goldthorpe and Registrar General’s classifications brought out the fact that the latter ”is far from meeting OPCS’s claim of homogeneity”. On the basis of the former’s research “it turns out that we need to change the definition of social class for about 33 per cent of the male population” (Bland, 1979, p286). However, many would be ranked only one class higher or lower in the 7 class scale and we are not aware of studies showing that in practice the results of applying the 2 (abbreviated) measures are markedly different.

1.28 Sociologists have tended to distinguish between a “distributional” and “relational” approach to the measurement of stratification. On the one hand the population may be ranked according to one attribute or a combination of attributes, such as prestige, income or educational qualifications. On the other hand, it may be divided according to the relationship of individuals to the means of production or the market. The problem is that work or market situation varies quite widely and distributional criteria apply more than is generally appreciated. When developed in detail the 2 approaches are not so distinct as they are often believed to be.

1.29 The possibilities of developing operational concepts of “class” rather than “status” have not been fully explored by sociologists, in part because of the practical difficulties. These difficulties are greater than in the United States, where the national census collects, for example, both income and educational as well as occupational data (Duncan, 1961; Blau and Duncan, 1967). We consider there are practical possibilities of measuring certain occupational as well as social aspects of class. Some attempt to measure career expectations in terms of remuneration, security, opportunities for advancement, working conditions and amenities and fringe benefits for different occupations, or at least the main occupations, would be involved.

1.30 One of the problems about any classification based on current occupation is the allocation of married women. As Leete and Fox say,

“The early social class analyses of women were based on the husband’s occupation principally because women generally gave up employment on marriage, and because it was assumed that husbands and wives had similar life styles and were, by implication, of the same social class. The assumption was that if men and their wives exhibited similar mortality differentials the whole socioeconomic context of life was an important factor affecting mortality: but if not occupation exerted a direct influence on mortality. Since the 1930s, with the growth of the numbers of married women entering the labour force, married women have been increasingly classified in census tabulations by their own occupations if they are or have been employed; single, widowed and divorced women have always been classified by their own occupation.” (op cit p4).

1.31 In practice the occupational category in which most women are to be found is class III non-manual whereas men are to be found typically in class III manual. It may be wrong to build analyses on data where many of the married women are classified according to an occupation which may have been held only for a very short period or involve relatively few working hours, and the problem becomes more complicated if decisions are to be taken about which parent’s occupation should be the right indicator for analysis of, say, infant mortality.

1.32 This helps to illustrate the qualifications which must be attached to any analysis of social class, using current occupation. The qualifications apply not just to the occupations held by married women but also to many of those held by men, who may have changed occupations during working life.

1.33 We consider it to be worthwhile for analyses of health to continue to be made on the basis of classifications using current occupations. But first, as suggested above, we believe that an effort should be made to make this classification as objective as possible, by taking account in the ranking of occupations of current and lifetime earnings, and also fringe benefits, security, working conditions and amenities. Our intention is to shift attention from the more elusive subjective rating of “prestige” or “general standing” of occupations to their material or environmental (and more measurable) properties. Second, it would be desirable for the term ‘occupational class’ to be used rather than “social class” when the current occupation of the individual is used as the basis of the classification. Third, it will become increasingly important to use the married man’s occupation in combination with the married woman’s occupation in analysing various health conditions and experiences, for example infant and child mortality. Fourth, the need for a ‘social class’ measure for analysis of the health of the family unit as a whole or of individual members of the family unit will become increasingly important. One possibility is using the current occupations of both parents together with information, where it can be obtained, about the main occupations of the husband’s father and the wife’s father. We return to these possibilities in Chapter 7.

1.34 For many significant social minorities current occupational class of the male is an inadequate indicator of resources or life-style and we are obliged to consider more aggregated indicators. One attempt at dealing with the problem has been made in the context of the Bristol follow-up of the 1970 Birth Cohort Study (Child Health and Education in the 70s). Their social index compounds social class with the level of education, housing (tenure, crowding, availability of bathroom) and an assessment of the neighbourhood of residence. Each of these was scored and weighted to produce a one dimensional index. Preliminary indications, based on a sub-sample of the 16,000 families, are that social index score groupings are better able to discriminate on such variables as child developmental indicators and use of child health services than is the Registrar General’s social class alone (Osborne and Morris, 1979). There are of course theoretical difficulties in using such methods (eg in the choice and weighting of components), and examination of different alternatives will be necessary before conclusions can be reached about the replacement or augmentation of the indices used nationally.

1.35 Finally use of occupation as an indicator of social class has become so widespread in Britain in recent decades that the pre-occupations of some pioneer health statisticians have been forgotten. Farr, Stevenson and others were particularly concerned to relate health experience to riches or poverty (for example, Stevenson 1928). Efforts should be made to restore this tradition, and not only because of the difficulties (discussed above) in taking occupation as a reliable indicator of a family’s social class. The growth of absolute levels of resources, the spread of employer welfare benefits in kind and of social service benefits, and the increase of owner-occupation among the working classes makes a measure of ‘resources’ all the more important. The term ‘resources’ seem to be more appropriate than ‘income’ because of the present day impact of wealth and both employer welfare and social service benefits-in-kind upon living standards. Considerable sums are spent each year on official annual surveys – including the Family Expenditure Survey (FES), the General Household Survey (GHS) and the National Food Survey. The FES provides the best measure of incomes and although some information is collected about employer welfare and social service benefits it is incomplete and rather rough. Valuable data about the distribution of health are collected in the GHS, and although the information collected about income has, since 1979, been the same as in the FES, it is not supplemented by information on other resources. The development of a more adequate measure will not be easy, and the Royal Commission on the Distribution of Income and Wealth took a very cautious view in some of its reports about the possibilities of linking income and wealth in surveys. (See especially Reports Nos 1, 4 and 5 (1975-77). However, its 7th Report took a more positive view about the need to develop joint distributions of income and wealth as a priority (p 160) and about the desirability of sample surveys of personal wealth holdings (Report No 7, Cmnd 7595, 1979, P3). We recommend that in the General Household Survey steps should be taken (not necessarily in every year) to develop a more comprehensive measure of income, or command over resources, through either (a) estimates of total wealth or at least some of the most prevalent forms of wealth, such as housing and savings, or (b) the integration of income and wealth, employing a method of, for example, annuitization.

1.36 These questions of improving knowledge of health in relation to inequalities are further developed in Chapter 7.

CONCLUSION

1.37 In examining the state of health of a population it is necessary to remember there are different meanings of “health” which have different implications for action to improve health. On the one hand “health” can be conceived as the outcome of freeing man from disease or disorder, as identified throughout history by medicine. One the other hands “health” can be conceived as man’s vigorous, creative and even joyous involvement in environment and community of which presence or absence of disease is only a part. While there are many indicators of health and ill-health, including mortality rates, morbidity rates, sickness absence rates and restricted activity rates, we concentrate most attention in this report, partly for practical reasons, on mortality rates.

1.38 Different meanings are also given to the term “inequality”. Interest tends to be concentrated on those (substantial) differences in condition or experience among populations which have been brought about by social or industrial organisation and which tend to be regarded as undesirable or of doubtful validity by groups in society. Inequality is difficult to measure and trends in inequalities in the distribution of income and wealth, for example, cannot be related to indicators of health except indirectly. Partly for reasons of convenience, therefore, occupational status or class (which is correlated closely with various other measures of inequality), is used as the principal indicator of social inequality in this report.

Reid, I. Social Class Differences in Britain, London, Open Books, 1977

Reports Nos 1.4.5.and 7 of the Royal Commission on the Distribution of Income and Wealth, Cmnds 6171,6626,6999 and 7595, LondonHMSO, 1975-79.

Rowntree, B.S., Poverty. A Study of Town Life, London, MacMillan, 19O1

Sackett, D.L. , L.W. Chambers, A.S. MacPherson, C.H. Goldsmith and R.G. Macauley,”The Development and Application of Indices of Health: general methods and a summary of results”, American Journal of Public Health, Vol 67, No 5, 1977-

Sigerist, H.E. , Civilisation and Disease, University of Chicago Press, Phoenix Edition, 1962 (First published Cornell University Press,1943)

Stevenson, T.H.C. – “The Vital Statistics of Wealth and Poverty” – Journal of the Royal Statistical Society, Vol 91, 1928.

Supplement to the 35th Annual Report of the Registrar General in England, London, 1875.

Supplement to the 45th Annual Report of the Registrar General in England,London, 1885.

Susser, M.W. and Watson, W., Sociology in Medicine, London, Oxford University Press, 1971.

Townsend, P., Poverty in the United Kingdom, London, Penguin Books, 1979

Tuckett, D.A. (ed.) – An Introduction to Medical Sociology, London, Tavistock, 1976.

Wood, P.H.N., “Classification of Impairment and Handicaps” WHO International Conference for the 9th Revision of the International Classification of Diseases, 1975-