Health Inequality

An Introduction to Concepts, Theories and Methods

Mel Bartley

• Publisher: Polity
• Year: 2016
• ISBN:
  9780745691138
• Edition: 2
• Topic: Social Sciences
• Subtopic: Sociology
• Index: Social Sciences

At a time when social inequalities are increasing at an alarming rate, this new edition of Mel Bartleys popular book is a vital resource for understanding the extent of health inequalities and why they are proving to be persistent despite decades of growing knowledge and policies on the issue. As in the first edition, by examining influences of social class, income, culture and wealth as well as gender, ethnicity and other factors in identity, this accessible book provides a key to understanding the major theories and explanations of what lies behind inequality in health. Bartley re-situates the classic behavioural, psycho-social, and material approaches within a life-course perspective. Evaluating the evidence of health outcomes over time and at local and national levels, Bartley argues that individual social integration demands closer attention if health inequality is to be tackled effectively, revealing the important part that identity plays in relation to the chances of a long and healthy life. Health Inequality will be essential reading for students taking courses in the sociology of health and illness, social policy and welfare, health sciences, public health and epidemiology and all those interested in understanding the consequences of social inequality for health.

1
What is Social Inequality?

Before being able to go any further into the description of health inequality, we need to think carefully about what is meant by social inequality. From 1931 to 2001, official statisticians in England and Wales used a consistent measure of social class, the Registrar General’s Social Classes (RGSC) schema to compile a series of reports on health inequality, the contents of which will be described in the next chapter. This pioneering work has stimulated a large literature using many different measures of social inequality. The measure used in British official statistics itself changed in 2001. So although we now have many studies from different nations and at different times, it is actually very difficult to make comparisons between nations or to look at trends over time.

It has become a lot clearer in the past 10 years that the conflict and confusion around how we should understand social inequality has major implications for the study of health inequality. If you read the research that has been published over this period, it is striking how awkward many papers are when describing their measures of social position. The terms ‘high’ and ‘low’ are used in a vague manner, attached to income, status or social class without further explanation. What is the difference between class and status in any case? Do they always have the same relationship to each other, or to income? There seems to be little consideration of the possibility that different people might have very different ideas about what might count as ‘high’ and ‘low’ status or income. In fact, the measure of social position presently used in British official statistics does not have a simple ranking of high to low at all. There is a hidden theme throughout most research on health inequality which explains a lot of this confusion. This chapter aims to clarify the confusion and describe the danger it poses to understanding health inequality.

A seminal paper by Krieger and colleagues in 1997 (Krieger, Williams and Moss 1997) used a general term, ‘social position’ to encompass class and status and ‘socio-economic position (SEP)’, to include income or wealth differences. These terms are proposed only as a tool for taking things forward and are by no means a perfect expression of the underlying ideas. For example, income and assets are not really types of ‘position’ in society, although one’s place on the income ladder (your income relative to that of other people) might be regarded in this way. Two people with the same amount of monthly income, whether or not they own their home and have a car, central heating, a computer and so on, may be in different socio-economic positions in terms of their social class, status or income relative to that of their friends, neighbours or workmates.

Importantly, Krieger and colleagues also suggested that in actual research these dimensions of inequality need to be kept separate. The most suitable measure, they argue, will depend on the ways in which the researcher thinks social inequality is producing inequalities in a specific health outcome. For example, there are social inequalities in accidental death and in death from heart disease, but obviously different processes must be at work. Chapters 4, 5 and 6 of this book set out three different ‘aetiological pathways’, and chapters 9 and 10 give some examples of how different pathways may be at work in different cases. But first it is necessary to spell out the differences between the dimensions of class, status and income/wealth, and how these have been measured. In order to try and incorporate the newer research, it is also now necessary to discuss how education might fit into the measurement of social inequality.

Measures of social position

There are two ways in which social position (leaving out income and wealth) is usually represented: social class and social status or prestige. Although in many studies these two concepts have been used interchangeably, these dimensions of inequality may not have the same relationships to different types of disease outcomes. When we understand what social class and status are supposed to measure,we can see more clearly what kinds of effects they might be expected to have on health and how these effects may come about.

Social class

Measures of social class are based on theories of social structure: people choose their measure according to the theory they prefer. The two most prominent theories of social structure used in studies that work with a concept of class are based on the thinking of Marx and Weber. They divide occupations into groups according to typical employment conditions and employment relationships. These groups are the social classes (and there is no concept of some being ‘higher’ or ‘lower’ than others). Both schools of thought agree on the importance of two things. The first is the ownership of assets, such as property, factories or firms. That is what determines whether a person needs to work at all or whether she or he is the owner of a business, land or other assets sufficient to make working for a wage or salary unnecessary. The second feature of social class which is of generally agreed significance is the relationship of all those who do have to work for a living with those who own and manage the establishments in which they work, with those who supervise their work and also with any others whose work they in turn manage or supervise.

The definition of social class most widely used in British sociological and political research is based on the work of Weber, developed by John Goldthorpe and his co-workers. The earliest of these measures was first used in work on social mobility in the United Kingdom (Goldthorpe, Llewellyn and Payne 1980). Social classes are described as combining occupations whose members would tend to have similar sources and levels of income job security and chances of economic advancement, and who would have a similar location within systems of authority and control within businesses, and hence similar degrees of autonomy (Marshall et al. 1988). Erikson, Goldthorpe and Portocarrero further developed the original Goldthorpe schema in order to conduct a large international comparative study of social mobility (Erikson and Goldthorpe 1992). The most basic classificatory division in this schema is between those who are owners, either of a company or of property such as real estate or farmland, and those who are not. Within the group of ‘owners’, there are those who employ large numbers of others, those with a few employees and those with none. Those who own no property or company may either be employees or self-employed workers. The much larger group of people who are employees has more sub-divisions. They are divided according to the skill needed for their work, whether it is manual or non-manual, and the nature of their employment contract.

Erikson and Goldthorpe distinguish two basic forms of employment contract: the ‘service contract’ and the ‘labour contract’. The service contract is what you find in managerial and professional work. Employees with this kind of contract of employment have to be trusted: their work cannot be supervised in any simple way by monitoring their time-keeping or counting how many nuts or bolts they have produced by the end of the day. In order to motivate performance, employees with this kind of job are offered more job security, salary increments and a progressive career as incentives to good and loyal service. In addition, this type of work entails a degree of command, either over the work of other people or at least autonomy over one’s own work. Workers with a service contract are usually paid monthly, may have share options or a similar stake in the profitability of the company and seldom have to do things like clocking in and out. In contrast, employees with a labour contract perform work that is more easily monitored. They have little autonomy and tend to be more closely supervised and restricted in their patterns of work. Payment is more closely tied to hours of work and, in some cases, to how much is produced in that time (‘piece rates’). There is less likelihood of career progression and no annual salary increment, and job security is lower. The Erikson−Goldthorpe−Portocarrero (EGP) classification acknowledges that many occupations have a mix of these conditions, so that allocating them into classes is a matter of deciding which occupations more closely resemble each other in these respects (Evans 1992).

The principles behind the EGP schema, of employment relations and conditions as the basis for defining social classes have been further developed into the new class schema, the National Statistics Socio-economic Classification (NS-SEC). The NS-SEC is now used in all official government reports in the United Kingdom, such as the 2001 and 2011 censuses of England and Wales and of Scotland and the annual Health Surveys. The most important difference between the EGP schema and the NS-SEC is that both the notion of ‘skill’ and the manual/non-manual divide have disappeared from the classificatory principles. The criteria for allocating occupations to the different classes have been made totally explicit. They are: the timing of payment for work (monthly versus weekly, daily or hourly); the presence of regular increments; job security (over or under one month); how much autonomy the worker has in deciding when to start and leave work; promotion opportunities; degree of influence over planning of work; and level of influence over designing their own work tasks (Coxon and Fisher 2002). Because it will be used for a wide range of official statistics as well as for research, the NS-SEC in its ‘full’ form has a large number of categories which can be combined in different ways according to the purpose at hand. A seven-category version is likely to be that most frequently used in reports and studies (see Box 1.1).

---

Box 1.1 National Statistics Socio-Economic Classification (NS-SEC)

1. Higher managerial and professional occupations, including employers in large firms, higher managers and professionals, whether they are employees or self-employed.
2. Lower managerial and professional occupations and higher technical occupations.
3. Intermediate occupations (clerical, administrative, sales workers with no involvement in general planning or supervision but high levels of job security, some career prospects and some autonomy over their own work schedule).
4. Small employers and self-employed workers.
5. Lower technical occupations (with little responsibility for planning own work), lower supervisory occupations (with supervisory responsibility but no overall planning role and less autonomy over own work schedule).
6. Semi-routine occupations (moderate levels of job security; little career prospects; no pay increments; some degree of autonomy over their own work).
7. Routine occupations (low job security; no career prospects; closely supervised routine work).

---

Extensive empirical work went into deciding which occupations to put in each social class. Questions covering each of the seven criteria were asked of some 60,000 citizens in the United Kingdom Labour Force Survey of 1997. Occupations could then be allocated to social classes according to the typical answers of members of each occupation to these questions. For example, among biological scientists, 78.6 per cent had incremental pay and 76 per cent planned their own work; among kitchen porters, the comparable percentages were 27 per cent and 3.8 per cent. As may be imagined, the amount of work involved in the construction of this measure was enormous, and it will be necessary to carry out regular updating of the classification as job conditions change over time, and new occupations appear.

Social status

The sociological definitions of social class described above are very different to what is often meant by ‘class’ in lay terms. In everyday talk, people often use ‘social class’ to refer to what a sociologist would call ‘status’. Unlike social class, the concept of status centrally involves the idea of a hierarchy or ranking ‘from top to bottom’ of society. The term ‘status’ is used often in everyday talk, and everyone thinks they know th...