20,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Geisteswissenschaft
- Sprache: Englisch
The persistence of stark health inequalities in today’s world is painfully clear to see, not least in the effects of the Covid-19 pandemic and falling life expectancy in many parts of the world. How can we advance our understanding of the full extent of health inequality, what drives it, and ways to address it?
The third edition of this popular book closely examines the influence of social class, gender, and race/ethnicity (among other issues) on health in the light of broad macro-political contexts. The classic behavioural, psychosocial, and material approaches to health and their embodiment within a life-course perspective are introduced but, importantly, are also re-situated within the growing understanding of the commercial and political determinants of health. Bartley and Kelly-Irving draw on extensive new evidence that shows how the chances for everyone to lead a long and healthy life depend on where power lies to control health-damaging policies and introduce health-promoting ones.
Health Inequality will continue to 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.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 512
Veröffentlichungsjahr: 2024
Ähnliche
CONTENTS
Cover
Table of Contents
Title Page
Copyright
Acknowledgements
Introduction to the Third Edition
1 Measuring Social Inequality
Measuring social inequality scientifically
Social class
Social status/prestige
Education
Income and wealth
Area measures
2 What is Health Inequality?
How unequal is health?
Why health inequality?
3 Figuring Out Health Inequality
Preliminary concepts
Absolute and relative
Standardization: what is it and why is it needed?
Observational studies and ‘causality’
Models of health inequality
Mediation: the pathways from inequality to illness
4 Explanatory Models I: Behavioural Explanations
Social distribution of health-risk behaviours
How important is health behaviour for health inequality?
Why are ‘risky’ behaviours unequal?
Problems with the behavioural model
5 Explanatory Models II: Material Explanations
What is the materialist model?
Income and poverty
Health risks at home and work
Material factors, power and the ‘fundamental cause’
The cost of a healthy life
Infectious disease and hygiene
6 Explanatory Models III: Psychosocial Factors
The biology of stress: fight, flight and defeat
Types of psychosocial factors
How important are psychosocial factors?
7 The Life Course and Embodiment
Background and emergence of the life-course framework
The life-course framework and its principles
Simultaneous life-course principles
Embodiment: understanding social-to-biological processes
Embodiment mechanisms over the life course
Health inequalities over the life course: linking the explanatory pathways to embodiment
8 Gender and Inequality in Health
What do we mean by gender?
Gender inequalities in health
Macrolevel factors and explanatory pathways
Sex, gender and embodiment
9 Racialization and Health Inequality
What is meant by ‘race’ or ‘ethnicity’?
Ethnicity, biology and health
How great are ethnic or racial differences in health?
Racialization, socioeconomic conditions and intersecting inequalities
Social ecology of ethnicity and health
10 The Macrosocial Environment: Political, Economic and Commercial Contexts
Distribution of income
Welfare regime types
Policies and expenditures
Corporate and commercial determinants of health
A case study of the macrolevel determinants of health and health inequalities
The life course, explanatory models and the macro context
Conclusion: The Way Forward for Research and Policy Debate
Conventional explanations for effectiveness of therapies
International comparative studies
Implications of recent evidence for policy debate
New models for research and policy
Combining macrolevel processes with embodiment to design policy
References
Index
End User License Agreement
List of Figures
Chapter 1
Figure 1.1
Galton’s concept of social structure
Figure 1.2
The Hindu caste system
Figure 1.3
Distribution of disposable income in the UK and household income in the USA
Figure 1.4
Distribution of wealth in Great Britain and the USA
Chapter 2
Figure 2.1
Trends in mortality rates (directly age standardized) in England and Wales, 1970–2010 …
Figure 2.2
Social inequality in life expectancy at birth in England and Wales, 1992–2016
Figure 2.3
Life expectancy by highest qualification at age 25 in the USA, 1996 and 2006
Figure 2.4
Changes in life expectancy between 2010 and 2017 by race and education in the USA
Figure 2.5
Age-standardized mortality by occupational social class in different European nations (men …
Figure 2.6
Changes in life expectancy at birth by area deprivation in England, 2015–2020
Chapter 3
Figure 3.1
Relationship between weight and calories consumed
Chapter 4
Figure 4.1
Trends in inequality in smoking in Great Britain, 1974–1998 (men)
Figure 4.2
Trends in inequality in smoking in Great Britain, 1974–1998 (women)
Figure 4.3
Smoking in the UK, 2014–2019 (men and women aged 18–64)
Figure 4.4
Obesity by social class in England, 2013–2017
Figure 4.5
Smoking represented as a celebration of success and status
Figure 4.6
Percentage of men and women meeting US recommended amounts of exercise by …
Chapter 5
Figure 5.1
Trends in health inequality in England and Wales, 1931–1991 (men aged …)
Figure 5.2
The life expectancy at gradient by social class in England and Wales, …
Figure 5.3
Covid-19 mortality by area deprivation in the USA and England
Chapter 7
Figure 7.1
Social position at birth and mortality in West Scotland, 1970–1973 …
Figure 7.2
Mortality by occupation of father and own occupation at two time points in …
Figure 7.3
A life-course framework for health inequality incorporating social inequalities, …
Chapter 8
Figure 8.1
Female advantage: gap in the average life expectancy at birth in years in twenty-nine …
Figure 8.2
Female disadvantage: difference in proportion of life expectancy at birth spent in good …
Figure 8.3
Trends in age-standardized mortality from cancer, cardiovascular disease and injuries for …
Figure 8.4
Relative contribution of causes of death to sex gap in life expectancy in various countries
Chapter 9
Figure 9.1
Trends in mortality by race/ethnicity in the USA, 2011–2021
Figure 9.2
Trend in mortality rates of younger adults (aged 25–44) by race/ethnic group in the …
Figure 9.3
Life expectancy by ethnic group in England and Wales, 2014
Figure 9.4
Age-standardized mortality rate from Covid-19 for men and women in England and Wales, …
Chapter 10
Figure 10.1
Life expectancy vs. GDP per capita, 2021
Figure 10.2
Gini coefficient and life expectancy in OECD countries, 2017–2020
List of Tables
Chapter 1
Table 1.1
Social class schema of the Registrar General of England and Wales
Chapter 2
Table 2.1
Standardized mortality ratios by Registrar General’s Social Classes (RGSC) in …
Table 2.2
Trend in mortality rates (directly age standardized) using old and new social class measure…
Table 2.3
Mortality (age standardized rate per 100,000) by years of education in men and women aged …
Table 2.4
Inequality in mortality in men aged 30–59 in European nations, 2000–2005
Table 2.5
Annual percentage change in ratio of mortality rates comparing those with lowest and highest …
Chapter 3
Table 3.1
Example of direct standardization
Table 3.2
Odds ratio: depression by gender
Table 3.3
Exposure and disease: odds ratio
Table 3.4
Social class differences in disease before adjustment for smoking
Table 3.5
Social class differences in disease after adjustment for smoking
Table 3.6
Odds of poor health before and after adjustment for smoking
Chapter 4
Table 4.1
Hypothetical relationship between social position, locus of control and health risk …
Chapter 5
Table 5.1
Covid-19 mortality rate by occupation in England and Wales, 2020
Table 5.2
Covid-19 mortality rates by health care occupations in England and Wales, 2020
Chapter 7
Table 7.1
Accumulation of health risk over the life course
Chapter 9
Table 9.1
Mortality rates from leading causes by race/ethnic group in the USA, 2017
Table 9.2
Mortality from all causes (SMR) by sex, cause and country of birth for people aged …
Chapter 10
Table 10.1
Coefficient of variation measuring degrees of inequality in income for a more and …
List of Boxes
Chapter 7
Box 7.1
Criticisms of the research emerging on the foetal origins of adult disease hypothesis
Box 7.2
Principles of life-course theory developed by Elder and colleagues
Guide
Cover
Table of Contents
Title Page
Copyright
Acknowledgements
Introduction to the Third Edition
Begin Reading
Conclusion: The Way Forward for Research and Policy Debate
References
Index
End User License Agreement
Pages
iii
iv
x
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
Health Inequality
An Introduction to Concepts, Theories and Methods
Third Edition
Mel Bartley and Michelle Kelly-Irving
polity
Copyright © Mel Bartley & Michelle Kelly-Irving 2025
The right of Mel Bartley and Michelle Kelly-Irving to be identified as Author of this Work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.
First edition first published in 2003 by Polity PressSecond edition first published in 2017 by Polity PressThis third edition first published in 2025 by Polity Press
Polity Press65 Bridge StreetCambridge CB2 1UR, UK
Polity Press111 River Street Hoboken, NJ 07030, USA
All rights reserved. Except for the quotation of short passages for the purpose of criticism and review, no part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher.
ISBN-13: 978-1-5095-5717-2
A catalogue record for this book is available from the British Library.
Library of Congress Control Number: 2024936495
The publisher has used its best endeavours to ensure that the URLs for external websites referred to in this book are correct and active at the time of going to press. However, the publisher has no responsibility for the websites and can make no guarantee that a site will remain live or that the content is or will remain appropriate.
Every effort has been made to trace all copyright holders, but if any have been overlooked the publisher will be pleased to include any necessary credits in any subsequent reprint or edition.
For further information on Polity, visit our website:politybooks.com
Acknowledgements
Mel Bartley thanks the Centre for Longitudinal Studies, Institute of Education, for the use of the 1958 birth cohort data and the United Kingdom Data Archive and Economic and Social Data Service for making that information available. None of the organizations that provided access to these data had responsibility for their analysis or interpretation as presented in this book.
I owe a great deal to past and present colleagues, in particular Amanda Sacker, Meena Kumari, Pekka Martikainen, Archana Sing-Manoux, Paul Clarke, Mai Stafford and Eric Brunner. Updating the British official data on health inequality would not have been possible without the help of Peter Goldblatt.
Michelle Kelly-Irving thanks the Gender and Health Inequalities Project (Horizon 2020 European Research Council, Gendhi-Synergy Grant Agreement SGY2019-856478) and the Inserm Centre for Epidemiology and Research in Population Health at the University of Toulouse for their support.
A special thanks to my wonderful colleagues for their advice, knowledge and discussion, especially Cyrille Delpierre, Meena Kumari, Raphaële Castagné, Marc Chadeau-Hyam, Muriel Darmon, Nathalie Bajos, Pierre-Yves Geoffard, Thierry Lang, Paolo Vineis and David Blane. A final thanks to my family, for putting up with my rambling on about health inequalities.
Introduction to the Third Edition
The study of health inequality is often traced back to around fifty years ago. Many of those who studied health inequality in the 1970s and ’80s saw it as a chance to improve our understanding of disease and our ability to prevent ill-health in the whole population regardless of social background. Social differences in premature mortality from almost all causes were so large, it seemed that, if we could understand the ways in which socioeconomic adversity ‘gets under the skin’ to produce disease, this might lead to major improvements in population health and ways of making medical care more effective. What is the point, we asked, of treating someone’s bronchitis in hospital and then sending them back to a damp, cold house to become ill again?
The existence of health inequalities was seen as proof that diseases were indeed preventable by changes to the environment. Although the British National Health Service had offered medical care free at the point of use to all citizens since 1947, by the 1970s it was clear that this provision was not reducing the level of health inequalities in the UK; in fact, they increased (see Chapter 2). This made it all the more important to regard prevention as being better than cure. After all, even free operations are not a pleasant experience; most people would prefer not to become ill in the first place. But more surprising was that health inequalities continued to increase even after forty years of a welfare state that, in theory, prevented the worst extremes of poverty, and had actually succeeded in reducing income inequality.
The 1980 Black Report, published by the UK Department of Health and Social Security, was the first attempt of its kind anywhere in the world to drill down into the statistics and identify what it was about social class (the measure of social inequality used in British official statistics, further discussed in Chapter 1) that produced these large differences in risk. The Black Report put forward four possible models of explanation: selection, artefact, material, and behavioural-cultural. In the early 2000s, the artefact explanation (which suggests that there is no causal link between class status and health outcomes, but that they are just coincidental measurements) had been completely discredited and the selection explanation (which makes health the key causal factor, suggesting that social class is the result of a process of ‘selection’ based on health) was little considered outside economics. Accordingly, the first edition of this book concentrated on the material and behavioural-cultural models and tried to assess how well each of them fitted the existing evidence. It also described and assessed three additional factors that had been increasingly investigated in research between 1980 and 2000: psychosocial stress at work, social isolation and life-course effects. This third edition has the exciting task of showing the ways in which the focus on the individual is receding from health inequality research, to be replaced by a much more sophisticated understanding of the ways in which the environment – material, economic, commercial and political – is implicated in health inequality (Bambra et al., 2019).
Between 1997 and 2001, when the ideas behind the first edition of this book were being gestated and written down, the UK had a ‘New Labour’ (moderate social democratic) government that was in many ways determined to reduce health inequality, and was prepared to do so by addressing at least some of the social and economic factors (the so-called ‘upstream factors’) believed to be involved, such as the absence of a minimum wage, low pensions, widespread educational failure and child poverty (Douglas, 2016). The successor to the Black Report, the Acheson Report (1998), set out a large number of rather precise recommendations, and a government plan published in 2002 described the ways in which many of these would be met (Department of Health, 2002). As pointed out by Mackenbach (2010a), the British programme was ‘by far the best resourced of all the Western European strategies to reduce health inequalities which started during the decade’. It should have been a golden age during which the research paid off in terms of real world reductions in health disparities between social groups, and overall improvements in population health.
So it was surprising and dismaying to see the verdict on the years of policy initiatives that followed the Acheson Report. According to most commentators, they did not succeed in reducing health inequality in the UK (Department of Health, 2009; Mackenbach, 2010b; Law et al., 2012). But this is not a universal opinion. Official statistics on health inequality gradually abandoned measuring social inequality in terms of social class and adopted ‘area deprivation’ as the most commonly used measure (Case and Kraftman, 2022). From this perspective, there did seem to be the beginning of some improvement in health inequalities between richer and poorer geographical areas between 1999 and 2010, only for these to widen again after 2010 (Robinson et al., 2019; Vodden et al., 2023). This observation points to one of the new features of the data we now have for several nations: it covers a period of great change in policy, a period of government austerity, and of course a worldwide pandemic.
One major problem for those of us who are attached to testing ideas with data is that the availability of data that can be used to evaluate the success or failure of new policies has changed. When we decided it was time for a third edition of Health Inequality, we could soon see that the problems besetting the second edition had become even more acute. The evaluations of the strategy for reducing health inequality from 2002 had only required a comparison of mortality by geographical areas, not by social class, as had been done every ten years since 1921 in England and Wales, for example (see Chapter 2). Social class gaps were only assessed in infant mortality. It appears that the degree of class inequality in infant mortality did begin to fall quite a lot (Bambra, 2012). But that is not the same as being able to extend the seventy-year-old analysis of social class differences in the premature death of adults that gave rise to the whole issue of health inequality in the first place. There are many problems with the use of area measures, although, as we will see, it also has some advantages. It has long been known that not all poor people live in deprived areas. One study has reported that fewer than half of those who were either unemployed or receiving welfare benefits in Scotland, for example, were living in the most deprived 20 per cent of areas (McCartney et al., 2023).
Moreover, if research is to inform policy, we need to unravel the potential pathways that link factors such as exposure to hazards, housing, health care, labor markets – and many other socially structured facets – to health. Deprivation indices, which are usually composite measures at an area-level including factors like occupation, education, material assets, income, crime, etc., may obfuscate these underlying mechanisms of relevance. These issues will be developed in Chapter 1.
In this edition, the major reorientation of our explanatory efforts is to place the processes (pathways) within the context of the political and commercial determinants of health. In 2005, Bambra and her colleagues had criticized the ways in which health has been ‘depoliticized’, adding that this has not happened ‘by chance: both the masking of the political nature of health, and the forms of the social structures and processes that create, maintain and undermine health, are determined by the individuals and groups that wield the greatest political power’ (Bambra et al., 2005, p. 192). The growing body of research in global health was giving rise to a different perspective on health inequality within nations, focused on the power of big corporations both to persuade us to consume unhealthy products and to motivate governments not to control unhealthy consumption (Schram and Goldman, 2020; Lacy-Nichols et al., 2023; Lancet, 2023). Earlier research focused on the activities of large companies that produce and promote tobacco goods and highly processed foods (Kickbusch et al., 2016; Marteau et al., 2019). This work has now developed even further into an appreciation of how these companies actually formulate government policies (Buse et al., 2017; McKee and Stuckler, 2018), and even the research that is supposed to inform policymaking. McKee and Krentel write of ‘the use of false experts who seek to lend credibility to the corporation’s arguments, and “moving the goalposts”, where corporations dismiss evidence presented in response to a specific claim by continually demanding some other, often unfulfillable, piece of evidence’ (2022, p. 91–2).
A sobering conclusion from some of these new studies has been that the researcher herself needs to be very aware of the power of corporations to influence the content of science itself. For example, as van Schalkwyk and colleagues point out: ‘There is ample evidence on the ways in which industries conduct research – for example by conducting large numbers of studies in secrecy until the desired results are obtained, and then commissioning independent researchers to conduct the study that will produce those results’ (2022, p. 97). In 20014, when the second edition of this book was being written, such claims would have seemed far-fetched. But since then, the awareness of how easy it is to distort evidence has grown rapidly. Papers are frequently retracted from journals when such distortions are discovered. Increasingly, the student of health inequality needs to ask ‘Says who?’ ‘Who has paid for this?’, ‘In whose interests are these findings?’
Does this mean that the classification of approaches into material, behavioural and psychosocial models placed within a life-course context is no longer useful when we try to understand more recent trends? Certainly, the content of the different aetiological (‘diseasecausing’) models continues to be revised in light of what has happened in life-course research, and reinterpreted from the perspective of ‘embodiment’ (as we shall see in Chapter 8). When we observe the health of people in a certain social category of class, gender or ethnicity, we now know that we are observing the accumulated consequences of their material, emotional and cultural histories. Our objective in this edition is to place this sociogenesis of human bodies and human health into its political context.
In what follows we shall see whether existing research on health inequality can be understood with the help of a model of accumulated biological, psychological and social advantages and disadvantages, within the contexts of different national and local economies and cultures. There is a vast quantity of potential combinations of circumstances that individuals may pass through in their life course, all of which may contribute to healthy life expectancy many years in the future. The challenge for research is to arrive at adequate measures of these, and to find adequate methods for putting them together into causal models that can be useful for policy discussion and help to explain what causes health inequality – how it persists and how to end it.
Further reading
On the relationship of research to policy in health inequality:
Smith, K. E. and K. Garthwaite, ‘Contrasting views on ways forward for health inequalities’, in K. E. Smith, S. Hill and C. Bambra, eds., Health Inequalities: Critical Perspectives. Oxford: Oxford University Press, 2016, pp.81–94.
On the new approaches to health inequality:
Maani, N., M. Petticrew and S. Galea, The Commercial Determinants of Health. Oxford: Oxford University Press, 2023.
McKee, M. and A. Krentel, Issues in Public Health: Challenges for the 21st Century. Maidenhead: McGraw-Hill Education, 2022, Chs. 4 and 5.
Lacy-Nichols, J., S. Nandi, M. Mialon, J. McCambridge, K. Lee, et al., ‘Conceptualising commercial entities in public health: beyond unhealthy commodities and transnational corporations’, The Lancet 401 (2023): 1214–1228.
McKee, M. and D. Stuckler, ‘Revisiting the corporate and commercial determinants of health’, American Journal of Public Health 108 (2018): 1167–1170.
Schram, A. and S. Goldman, ‘Paradigm shift: New ideas for a structural approach to NCD Prevention’, International Journal of Health Policy and Management 9 (2020): 124–127.
1Measuring Social Inequality
Measures of health inequality are not separate from the ideas people have about the causes of health inequality – indeed, they are intimately linked to these ideas. This is so important because most measures currently in use are implicitly underpinned by a theory of inequality that pushes research towards a certain kind of explanation. So it is not really possible to separate ‘How should we measure social inequality?’ from ‘What are the causes of health inequality?’.
The way in which social inequality has been measured in studies of health inequality has a long history and is still a matter of considerable dispute. This has been a major problem for the progress of understanding. In order to understand health inequality well enough to begin to design policies to reduce it, we need what many have called ‘causal narratives’ (Marshall, 1997; Rose and O’Reilly, 1997; Rose et al., 2010). A causal narrative is the kind of story that scientists try to develop in order to explain and predict natural phenomena. For example, it was at one time believed that the sun went around the Earth. But as this narrative did not make sense in the light of many observations of the stars and planets, astronomers came to believe its opposite. Newton’s concept of gravity made it possible to explain the way in which the sun’s gravity kept the Earth and other planets in their captive orbits. Causal narratives are not fixed, but constantly respond to new observations. But first of all, there need to be clear definitions of what is being observed.
At the present time the measurement of inequality is so confused that leading scholars in respected journals still make such comments as: ‘Socioeconomic status (SES) is a central measure in social epidemiology, but its use is complicated by the fact that scholars have not agreed on a definition, and probably never will’ (Oakes and Andrade, 2014). In addition to this recognition that socioeconomic status lacks conceptual clarity, Muntaner pointed out that ‘most studies use socioeconomic status indicators such as educational level, occupational class, or even income without any serious consideration of the social mechanisms that link exposure and outcome. This makes for weak ability to examine causality and thereby to propose effective interventions’ (2013, p. 853). While Glymour et al. add: ‘Good measurement of SES at individual, family, and community levels remains a major challenge.’ And: ‘To translate this evidence into effective publichealth interventions, we need more conclusive evidence on the causal components of highly correlated socioeconomic measures and on the major mediators of inequalities’ (2014, p. 90).
The most frequently used measure of social inequality in the current literature is ‘socioeconomic status’. This confused term originated from research in the USA and is based on a theory called ‘functionalism’. Davis and Moore describe the functionalist theory for why social inequality exists by stating that there is ‘a universal necessity which calls forth stratification in any social system’, which is that:
As a functioning mechanism a society must somehow distribute its members in social positions and induce them to perform the duties of these positions. It must thus concern itself with motivation at two different levels: to instil in the proper individuals the desire to fill certain positions, and, once in these positions, the desire to perform the duties attached to them …
Social inequality is thus an unconsciously evolved device by which societies insure that the most important positions are conscientiously filled by the most qualified persons. Hence every society, no matter how simple or complex, must differentiate persons in terms of both prestige and esteem, and must therefore possess a certain amount of institutionalized inequality. (1945, pp. 242, 243)
According to functionalism, the way in which this mechanism is assured is through the education system. Education sorts people out into the more and the less able. Inequality in working conditions, pay and status ensures that those who have done best in education are attracted to the most functionally important jobs in business, government, the military, science and technology and so on.
American sociologists, accordingly, allocated a ‘socioeconomic status’ to each occupation by a combination of the years of education required and the average income. It was at least implicitly believed that such a measure would be an indicator of ‘ability’. In effect, people’s ‘abilities’ determine their social position, so that the privileges of those with more social advantage are not a cause of their life conditions, but the deserved result of their personal (possibly genetic) ‘worth’. It is an easy step from there to the argument that lack of ‘ability’ is to blame for the kinds of behaviour that causes worse health and shorter lives among those in less privileged social positions. This was the ‘causal narrative’ involved in functionalism: people with lower levels of ability both end up in poorer paid jobs with worse working conditions and are more likely to engage in health-damaging behaviour.
Functionalism became controversial when it was noticed that its central argument was somewhat circular. How do we know that a certain degree of inequality in conditions, pay and status between occupational position is good for society? Because our present society (particularly US society in the 1950s) survives and prospers. Why does our society survive? Because it has the correct amount of inequality. However, the picture of American society as a smoothly functioning organism could not survive the economic, political and cultural crises of the 1960s and 1970s (Wiley, 1985). The idea that inequality assured prosperity looked more and more improbable. Studies also demonstrated that the education system, far from allocating the most able to the most advantaged occupational positions, in fact tended to perpetuate the privilege of children from advantaged families (Dearden et al., 1997). As a result, functionalism was pretty much discredited in sociology. However, it has remained lurking in the background of social epidemiology with serious consequences for the advancement of knowledge.
Measuring social inequality scientifically
In a scientific approach to understanding health inequality, therefore, we need to resist the temptation to use measures that push us towards certain explanations regardless of the evidence. Sociology, anthropology and economics have long traditions of measuring inequality in its different forms: roughly speaking, social class, social status (prestige), income and wealth. A good beginning is to aim for better measurement of these dimensions in our studies of health inequality.
A seminal paper by Krieger et al in 1997 used a general term ‘social position’ to encompass class and status, and ‘socioeconomic 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 or have a car and the latest technological devices, and so on, may be in different socioeconomic positions in terms of their occupational social class or their status.
Importantly, Krieger et al. 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 (the causal narrative). For example there are social inequalities in accidental death and in death from heart disease, but obviously different processes must be at work in these two health outcomes. 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.
Social class
Registrar General of England and Wales class schema
Chapter 2, which shows some of the facts and figures that illustrate health inequality, begins by using the Registrar General’s Social Classes (RGSC), which was the measure used in the classical series that allowed us to trace the increase in health inequality over ninety years. The UK (more strictly, England and Wales, as Scotland and Northern Ireland have separate national statistical organizations) was the birthplace of health inequality research because it has been possible since 1911 to divide the population according to a measure of social class that was used in the ten-yearly Census. Every ten years (the inter-Census interval) 10 per cent of the population of England and Wales was coded into a social class according to the Registrar General (chief statistical official) schema. These social classes formed the denominator for the calculation of class-specific mortality, while the numbers of deaths in the three years around each Census year formed the numerator. Social class was, until 2001, defined as a measure of ‘general standing in the community’ and ‘occupational skill’. Table 1.1 shows the categories used.
Table 1.1 Social class schema of the Registrar General of England and Wales
Class number
Description
I
Professional
II
Managerial
Until 1981
III
Routine nonmanual and skilled manua
After 1981
IIIN
Routine nonmanual
IIIM
Skilled manual
IV
Semi-skilled manual
V
Unskilled manual
The Registrar General’s measure was seldom used in British sociological research. Marshall and colleagues (1988, p. 19) described it in the following way: ‘The scheme embodies the now obsolete and discredited conceptual model of the nineteenth-century eugenicists: namely, that of society as a hierarchy of inherited natural abilities, these being reflected in the skill level of different occupations.’ We can see that this describes functionalism very succinctly.
Mackenzie (1981) has described the way in which the Registrar General’s schema was dependent on earlier work by the eugenicist Galton, who proposed that there was a ‘normal distribution of genetic worth’ in any population, reflected in the division of labour, as shown in Figure 1.1. In this diagram, the horizontal bottom axis represents ‘genetic worth’, running from the lowest at the left to the highest worth at the right. The vertical axis gives us very roughly the numbers of people in each group, showing a ‘normal distribution’ whereby the largest number of people have around average amounts of ‘worth’. The curve is divided into five social classes, from a small class of criminals and paupers (very poor and/or low status people), through to the largest group, the ‘respectable’ workers, to the group with the highest genetic endowments made up of large employers and independent professional people (like Galton himself). There were no actual data to support the ideas represented. But T. H. C. Stevenson, a leading English public health official who originated the Registrar General’s Social Class schema, adopted it for his studies of the urban health crises that beset the new industrial cities of the time (Leete and Fox, 1977; Szreter, 1984).
Figure 1.1 Galton’s concept of social structure
Source: MacKenzie, 1981: 17
It has been easy to translate the reasoning behind functionalism into a relatively pessimistic view on the potential for policy to ameliorate health inequality. In terms of one of Mackenbach’s explanations for rising health inequality in an increasingly affluent Europe,
the lower social strata have become more exclusively composed of individuals with personal characteristics that increase the risks of ill-health. This is the result of decades of upward intergenerational social mobility, which may have increased opportunities for social selection and may have made the lower social groups more homogeneous with regard to personal characteristics like low cognitive ability and less favorable personality profiles. (2012, p. 766)
In other words, as education becomes increasingly meritocratic as nations modernize, it is easier for the more intelligent born into less advantaged social positions to move through the school system into occupations better suited to their ‘worth’, taking with them their other health-promoting characteristics, such as the ability to understand health education messages and the self-control to observe them. The problem with this idea is that social mobility is no higher now in advanced industrial economies than it was in the 1960s (Bukodi et al., 2015; Hecht et al., 2020; Tahir, 2022), so there has been no increase in this proposed ‘sorting’ effect. And yet health inequality has grown. Therefore, using a measure of social inequality based on either functionalism or a nineteenth-century eugenic theory is not the best way to advance our understanding of health inequality.
National Statistics Socioeconomic Classification (NS-SEC)
The Office for National Statistics (ONS) of England and Wales came to the view that a better measure of social class was needed. In preparation for the 2001 Census, it began to develop such a measure that was not a nineteenth-century leftover, but firmly based theoretically and empirically. It turned to the work of Goldthorpe and colleagues to extend Weber’s (1982) sociological theories about social class to the situation of the late twentieth century.
Weber’s theory divides occupations into groups according to typical employment conditions and employment relationships. These groups are the social classes. The first criterion for deciding who is in what social class is the ownership of assets, such as land, 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, property or other assets sufficient to make working for a wage or salary unnecessary. This is useful because it introduces an element of the relationships of wealth and power that prevail between different groups. The second feature of social class that is of generally agreed significance is the relationship between all those who do have to work for a living and those who own and manage the establishments in which they work, those who supervise their work, and any others whose work they in turn may manage or supervise.
The seven criteria that the ONS devised for allocating occupations to the different classes were made explicit (Coxon and Fisher, 2002). All occupations in the economy can be organized into classes on the basis of the seven criteria (Rose and O’Reilly, 1997):
The timing of payment for work (monthly vs weekly, daily or hourly).
The presence of regular pay increments.
Job security (e.g. more than or less than one month’s notice).
How much autonomy the worker has in deciding when to start and leave work.
Promotion opportunities.
Degree of influence over planning of work.
Level of influence over designing their own work tasks.
Applying these criteria to each occupation gives you seven social classes with contrasting conditions and relations of employment:
Higher managerial and professional occupations, including employers in large firms, higher managers, professionals whether they are employees or self-employed.
Lower managerial and professional occupations and higher technical occupations.
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 security).
Small employers and self-employed workers.
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).
Semi-routine occupations (moderate levels of job security; little career prospects; no pay increments; some degree of autonomy over their own work).
Routine occupations (low job security; no career prospects; closely supervised routine work).
This work has advanced the attempts of health inequality researchers by making available a class schema with a firm conceptual grounding in measurable characteristics of employment relations and conditions. In this way, we gain a measure of social position that is completely separate from any measures of ‘genetic worth’, intelligence or ability. It is possible that these kinds of characteristics (intelligence, etc.) may be associated with social class membership, but this has to be demonstrated empirically and cannot be used as an automatic explanation for health inequality. Social class membership is a result of the way in which the economy is organized and has nothing to do with personal characteristics: if everyone in the population had the same level of education, IQ or personality there would still be social classes.
Social status/prestige
A lot of the time when people talk about ‘social class’, what they are really thinking of is status or prestige. Everyone works with an idea of status in their everyday lives, which is one reason why ‘socioeconomic status’ proves to be such a seductive term. However, it is a slippery concept and status orders vary greatly between societies and cultures. The clearest example of a status order is found in Hindu societies, most prominently in India. In Hindu cultures, gradations of prestige are represented by the caste system, based on the traditional occupations of extended kin groups (Beteille, 1992). Figure 1.2 is a very simplified version of the Hindu caste system found in India. As you can see, the caste groups are associated with different types of occupation (priest, warrior, merchant) but this does not mean every member exercises that occupation. Rather, a member of a caste is regarded as descended from ancestors who exercised a certain occupation.
The organizing idea of this caste hierarchy is of being closer to the Divine and purer. The highest status is accorded to those descended from priestly ancestors, who devote their lives to prayer and study rather than earthly power or wealth. The persistence of the caste system in Hindu societies is regarded by many as a great injustice, and is a matter of great dispute and conflict. ‘Untouchables’ have been relabelled ‘Children of God’ and are members of a Scheduled Caste, to which the government accords various forms of positive discrimination in an attempt to improve their life chances.
Anthropologists regard the symbols of caste membership to be the willingness of people to live close together, worship in the same religious spaces, eat together and marry each other. Members of higher castes maintain ‘social distance’ from members of lower castes (those with less prestige) by avoiding these forms of contact. Having close contact with lower castes risks rendering the individual ‘impure’ so that, for example, the marriage prospects of everyone in a family are damaged if one member marries a member of a lower caste.
Figure 1.2 The Hindu caste system
It might seem bizarre that the study of health inequality in modern industrialized nations would use measures of status, whose use seems outdated in such societies. However, you do not have to look far to see that status differences are still a powerful influence on social life. Here is a description from a widely read London newspaper of the ways in which a British audience might decide on the status of an individual woman who had recently married the heir to the British throne.
Nothing in the past four years has so exposed Britons’ obsessions with policing class boundaries as the coverage of Catherine Middleton’s family, wealth, upbringing and ancestors. The undertone of much of it has not been celebratory, but incredulous and indignant.
Newspapers have dwelt endlessly on the fact that her family tree includes coal miners, domestic servants, road sweepers and butchers. Her origins are described as humble … There is far less criticism of her father, or his family. He comes from a background described as ‘solid’; lines of provincial solicitors and landed gentry. The worst that is said of him is that he is self-made. (‘As a society, we still distrust the upwardly mobile’, Jenni Russel, London Evening Standard, 18 April 2011)
So in the terms of the caste diagram above, Catherine’s father Mr Middleton’s caste lies partly in the fact that his ancestors included gentlemen of leisure and scholarship (perhaps equivalent to ‘Brahmin’), although he himself has had to make his own money (akin to ‘Vaishya’). But, horrors! Mrs Middleton’s ancestors include servants (‘Shudra’) and even people with ‘untouchable’ occupations such as butchers and road sweepers.
This story will seem quite alien to many readers outside the UK (although coverage of Catherine Middleton and another royal bride, Meghan Markle, has been widespread across the world), because, unlike class, the sources of status are specific to historical time and place (Chan and Goldthorpe, 2007). Social groups may display caste-like social distancing behaviour for many different reasons. A famous modern example was the insistence in many American states that African American people go to different schools, sit in a different section of buses, and eat in different restaurants to white Americans. This segregation was regardless of the occupations or income of the discriminated group.
In the sociology of industrial societies, the measurement of status is similarly based on social proximity and distancing, although status is far more closely linked to occupation than to ancestry and this is reflected in the measures used. The most recent measure, devised by Chan and Goldthorpe (2004, 2007; Chan, 2010) is based on asking questions about the occupations of best friends and marriage partners of people in each occupation. Occupations can be ordered in terms of who mixes with whom, making these measures somewhat similar to the Hindu caste system. The difference here is that status is attached to each occupation rather than to that of individuals’ ancestors. For example, if university lecturers tend to choose friends among doctors, lawyers and architects, this defines a cluster regarded as having the same status. The reason why members of certain occupations favour members of certain others as friends are left rather vague, although Chan and Goldthorpe observe that the degree to which occupations share a preponderance of manual work (‘manuality’) is a major factor, manual work being associated with low status.
In fact, the status measures developed by British sociologists have not been widely adopted in health studies. Instead, both the study of status and the functionalist eugenic tradition have left a persistent thread for epidemiology in the use of education as a measure of social position.
Education
Functionalist ideas are reflected in the current common use of education as a measure of social position in many studies of health inequality. Higher educational attainment is regarded as the sign that an individual is well suited to one of the more privileged occupations. In this way, the use of education to measure social position imports eugenic assumptions into modern-day social epidemiology, and we can see that the ways in which social inequality is defined and measured in studies has profound consequences for the ways in which health inequality might be explained (the ‘causal narratives’).
There have been some practical reasons for the increasing use of education in health inequality research (Elo, 2009). A lot more international comparative research has become possible and many nations use measures of social class, if at all, of widely varying types. As more longitudinal research is done, we begin to see how much people move between social classes, income, or status groups over time. In contrast, education is thought of as being fixed by the time of entry into work. Education is often found to be a powerful predictor of health outcomes (Mirowsky and Ross, 2003). So what is the problem with using education as a measure of social inequality in studies of health?
There is no doubt that education is very strongly related to health (Mirowsky and Ross, 2003). But in fact, using it as a simple measure of social position does not do justice to the complexity of health inequality (Khalatbari-Soltani et al., 2022). What happens over the whole of childhood is increasingly understood to have a cumulative effect on both personal development and social destination (Liuet al., 2010) – something we will explore in Chapter 7. In general, the greater the number of favourable influences over this period, whether they be family living standards, family culture or the quality of relationships, the better the child and young person will do in education. It will not help in reducing the size of the health gap between more and less socioeconomically advantaged groups of adults to reduce these powerful processes in early life to a one-dimensional measure.
The second problem with using education as a simple measure of social position is that, over time and between nations, the numbers of people who reach a given level of education vary enormously. In 1965 around 6 per cent of the 18–19-year-old age cohort, and only 2 per cent of women, went to university in the UK (Egerton and Halsey, 1993). In 2013 it was around 43 per cent (Department for Business, Innovation and Skills, 2014). Is it possible, then, to regard ‘degreelevel education’ as the same thing for people born in these two cohorts? Or indeed, for men and women? In the USA, those with a high school diploma in 1959 had much higher pay on average than those without one; but by 2012 this ‘income premium’ had reduced greatly, and only having a college degree was sufficient to ensure a higher income than no qualifications at all (Chen et al., 2013). The differences between different nations, and how these have varied over time, are also large.
The third closely related problem relates to the differences in access to education for girls and boys, and for people of different ethnic or status groups. In some nations, ethnicity, caste and gender either exclude people from university (or indeed from school at all) or raise significant barriers. For the age cohort of one of us (MB), at the time of attending university there were quotas limiting the numbers of Jewish people who could be admitted to some US universities, and in the UK quotas limiting how many girls could attend certain types of technical training, including medical school.
The fourth problem with regarding education as a measure of social position is that it is far too easy to fall into the ‘functionalist’ assumptions discussed at the beginning of this chapter, and let these influence the ways in which we reason about health inequality. It is clear from a lot of what gets written in the literature on health inequality that educational attainment, like tastes in music, art or literature, is actually being used as a measure of individual ‘virtues’ that have resulted in the ‘socioeconomic status’ of the person. So we don’t really have to explain health inequality; it just follows from the natural superiority of the people who find themselves in the best jobs with the highest income and status.
Income and wealth
These measures are far less often used in studies of health inequality. This is partly because they are often absent from large-scale datasets such as a national Census. But income and wealth have other problems as well. Income is unequal in the majority of industrial nations, with most people concentrated in a group that ranges from very low to medium, and then a long tail of richer and much richer households (Figure 1.3). In the graph for the UK, the mean (the total amount of income divided equally amongst each individual or household in the population) is well to the right of the median (the midpoint amount, which half of all households earn more than and half of all households earn less than). It is very unlikely that adding, say, £1,000 to income has a similar effect on health if it takes the household from £80,000 a year to £81,000 than if it took a household from £6,000 to £7,000.
Figure 1.3 Distribution of disposable income in the UK and household income in the USA
Note: UK data for 2020; USA data for 2022
Source: ONS 2021a; data from US Census Bureau
A similar picture is visible for household income distribution in the USA (bottom graph of Figure 1.3), where the median is less than the mean, due in part to the large number of households found in the highest income groups to the right of the distribution. Adding income to the left of this distribution is much more likely to have a positive impact upon health, especially in a political system lacking universal health care where households are reliant on using their income to pay for inefficient private health insurance to access basic health care needs.
Wealth has been shown to have a stronger relationship to health than income (Martikainen et al., 2003; Demakakos et al., 2016; Finegood et al., 2021). But wealth is even more unevenly distributed than income, with the great majority of people having none, or very little, apart from their property, which in any case is usually still being paid for by a mortgage, which is a burden on their income rather than a benefit.
The graphs in Figure 1.4 show how uneven the distribution of wealth was in Great Britain and especially in the USA in recent years. The poorest half of the British population (up to the 5th wealth decile) held only just above 3 per cent of the total wealth of the nation (including the value of houses, pension plans and so on), a proportion that reduces to 1 per cent in the US context. The wealthiest 10 per cent held almost half of all wealth in Britain, and well over half the wealth in the USA.
The position we take in this book is that if we want to measure status, we need to use the results of studies that do in fact show how the prestige of different occupations is ranked in the culture that people inhabit, determining which groups mix together socially and intermarry. If we want to measure social class, we need to use a measure such as the NS-SEC and the growing number of adaptations for other nations, based on employment relations and conditions. If we want to measure income or wealth, then we can use other pretty obvious methods (such as asking people what their income is and how much savings they have, although this is not as simple as it sounds). We may want to look at these in combination to ask whether, for example, income is more or less important than occupational class or status (Geyer et al., 2006; Fliesser et al., 2018; Qi et al., 2019). To do this, then we should use the separate measures in the same analysis. As Hoffman et al. (2020, p. 551) have pointed out, ‘only analyses of multiple dimensions allow for comprehensive measurements of SEP [socioeconomic position], take into account the fact that some SEP dimensions are mediated by others, and provide insights into the social mechanisms underlying the stable structure of inequalities in mortality’.
Figure 1.4 Distribution of wealth in Great Britain and the USA
Note: for Great Britain, data is April 2018 to March 2020; for US, data is for 2022
Source: ONS 2022a; data from Bruenig 2023
As subsequent chapters will show, there are often good reasons for being interested in these kinds of question. As to education, if it is the only measure available, then it may be impossible to avoid using it. But this needs to be done with great caution and keeping in mind that it is only a ‘proxy’ or a probabilistic indicator (Khalatbari-Soltani et al., 2022). Education does not measure class, status or income at all, but gives some indication of what these may be, which will vary by time, place, gender and ethnicity amongst other things.
Area measures
Due to the difficulties and expense of doing surveys of individual social position and health, area measures have come to be widely adopted. Geographical areas are classified according to measures of social and economic conditions. In the USA it is possible to use forms of average income in different areas, taken from censuses. In the UK, where income is not included in the census, various ‘deprivation indices’ have been devised. These combine more readily available items, such as the proportion in each area of those who are low paid, unemployed, unskilled and low educated, and who live in substandard housing or experience crime (Department for Communities and Local Government, 2015). Items are taken from a range of sources, not just censuses. There is also an item on the proportion of the local population with poor health or disability, which creates obvious problems for using this kind of measure for health inequality studies, although individual studies often correct for it (Knies and Kumari, 2022) and it does not seem to produce too large a bias (McCartney et al., 2023).
A more serious problem is the ‘ecological fallacy’. We cannot know for sure whether, for example, high mortality in a deprived area is explained by the most deprived individuals being the ones more likely to die (Case and Kraftman, 2022). A deprivation index is based on averages across the local population, where not everyone will be poor or unemployed, etc. Especially with a rare event such as death, it is quite possible that it is the better-off people living in a deprived area who have a higher mortality rate. The UK Institute for Fiscal Studies health inequality review cautions us that
For policy purposes … we need to identify causal mechanisms that take us back to labour markets, pollution, social support, healthcare and other aspects of the way that society is structured. For these purposes, the deprivation index – a blend of components that includes income, employment, education, crime, environment and even health – is over-aggregated. (Case and Kraftman, 2022, p.11)
However, it is a lot easier to implement area-level policy interventions than to intervene in the structure of occupations or status within a whole society. The well-known Marmot reviews of health inequality in the UK use area deprivation measures more often than other measures of socioeconomic position (Marmot et al., 2020). One of the best recent studies of health inequality in the UK using area deprivation found that around 30 per cent of premature deaths (deaths before age 75) could be attributed to deprivation (Lewer et al., 2020); but the index of deprivation used to define socioeconomic position combined ‘information about seven domains: income, employment, education levels, crime, health, availability of services, and local environment’ (p. 34). It is not straightforward to pin down the causal narratives that might link these various items to the causes of premature mortality. Deprivation and mortality risk could have been linked by more people being murdered, having damp insanitary housing, being unable to access health services, or any combination of these plus the effects of other items in the index.
