Health inequalities

Authors


Address for correspondence: UCL Medical School, Research Department of Infection and Population Health, University College London, London W1A 8AA
e-mail: g.scambler@ucl.ac.uk

Abstract

The evidence bearing on the nature and extent of health inequalities documented globally and in the UK is addressed, twin foci within the UK being (a) associations between socioeconomic classification and health and longevity, and (b) the notion of a ‘social gradient’. A consideration of the various ‘models’ that have been developed by sociologists and their allies – most conspicuously social epidemiologists – to account for (a) and (b) is offered, drawing on government-sponsored commissions and reviews as well as the peer-reviewed literature. This is followed by a portrayal of specifically sociological theories of health inequalities, featuring those that hold social structures as well as cultural shifts in convention and behaviour to be causally efficacious for health inequalities. The summary and conclusions of the review incorporate an outline of pertinent questions the sociological community has so far been reluctant to address and an agenda for future research.

Introduction

Commentators on equality often distinguish between equality of opportunity and equality of outcome. In Butler’s Education Act in England and Wales in 1944, the IQ-oriented ‘11+’ examination was introduced. This was to be taken by all children aged 11 to sort the wheat (20%) from the chaff (80%). It was apparently intended to institutionally underpin equality of opportunity; but now we know better. Such are the ‘structured’, if not structurally-determined, advantages accruing to those brought up in high-rather-than-low income households that sitting the same examination in the same place at the same time falls far short of equality of opportunity. But what if equality of outcome is substituted? Would this seemingly more realistic option not bring other rival principles into play? It seems likely that (state-sponsored) social engineering towards equality of outcome would have to be assessed against the exercise of freedom. And what relation does justice bear to equality and/or freedom, or democracy? Who, on behalf of whom, might engineer equality of outcome; and if such persons were democratically accountable, to whom and by what means?

Class, socioeconomic classifications (SECs), health and longevity

The notion that material poverty predisposes, however circuitously, to poverty of health and life expectancy has a long pedigree; and doubtless as much was apparent in all types and levels of society long before the accumulation of what would now be accepted as evidence. In the early 1840s, Engels (1987/1845) and Virchow (1985/1848) alike ventured beyond merely noting the social patterning of disease to assert that this patterning was a consequence of the mode of social organisation: disease was socially produced. Engels used statistics and case studies to argue powerfully and poignantly that class exploitation was the prime cause of premature mortality in England. de Maio (2010: 15), who tracks this literature, selects this quotation well:

When one individual inflicts bodily injury upon another, such injury that death results, we call the deed manslaughter; when the assailant knew in advance that the injury would be fatal, we call his deed murder. But when society places hundreds of proletarians in such a position that they inevitably meet a too early and an unnatural death, one which is quite as much a death by violence as that by the sword or bullet; when it deprives thousands of the necessaries of life, places them under conditions in which they cannot live – forces them … to remain in such conditions until that death ensues which is the inevitable consequence – knows that these thousands of victims must perish, and yet permits these conditions to remain, its deed is murder (Engels 1987/1845: 127).

Engels refers in this context to inequity, which is now typically understood to refer to inequality that is avoidable, unnecessary or unfair. It is an emphasis re-captured in some current discussions, and we shall return to it later.

Virchow was a Prussian physician with far-reaching interests and, like Engels, revolutionary aspirations. His report on a typhus outbreak in Upper Silesia in the fateful year of 1848 bears comparison with Engels’The Condition of the Working Class in England. The typhus epidemic, he insisted, bore an intimate relation to famine. Those most imperilled:

were poorer and more ignorant, more servile and submissive than almost any other people in the world; they had lost all their energy and self-confidence. [Thus:] it is no longer a question of the medical treatment and care of this or that person taken ill with typhoid, but of the well-being of one and a half million fellow citizens who find themselves at the lowest level of moral and physical decline. With one and a half million people you cannot begin with palliatives, if you want to achieve anything you have to be radical … That is why I insist that free and unlimited democracy is the single most important principle. If we get free and well-educated people then we shall undoubtedly have healthy ones as well (Virchow cited in Taylor and Rieger 1984: 205–8).

In a celebrated pronouncement, Virchow surmised that ‘disease is not something personal and special, but only a manifestation of life under (pathological) conditions … Medicine is a social science and politics is nothing else but medicine on a large scale’ (Virchow 1985/1848, cited in de Maio 2010: 20). Implicit in his analysis too is a relation between inequity and democracy.

Coincidentally with the proto-sociological stances of Engels and Virchow, there emerged in the UK especially, a form of social accounting consonant with their ‘theories’ but focused on social patterning (that is, more epidemiological than sociological). From the outset positive associations between absolute and relative poverty and a diminution in health and longevity were apparent. Until 2000, and dating back to the beginning of the twentieth century, two socioeconomic classifications (SECs), sometimes taken as proxies for the more theoretical notions of class developed by the likes of Marx and Engels, featured in many government-sponsored reports as well as in much of the peer-reviewed literature: ‘social class based on occupation’ (SC, formerly Registrar General’s Social Class) and Socioeconomic Group (SEG). Neither classification was compelling, either as a proxy for class or as a representation of employment relations, recognition of which resulted in their eventual displacement by the National Statistics Socioeconomic Classification (NS-SEC), derived from the neo-Weberian ‘Goldthorpe Scheme’ (Rose and Pevalin 2003). The conceptual base for the NS-SEC is the structure of employment relations operating in modern developed economies like the UK. Occupations are differentiated in terms of reward mechanisms, promotion prospects, autonomy and job security. The most advantaged NS-SEC classes (higher managerial and professional) typically exhibit personalised reward structures, have good opportunities for advancement, relatively high levels of autonomy within the job, and are relatively secure (these attributes tending to be reversed for the most disadvantaged, or routine, class) (Langford and Johnson 2010).

In their recent review of the literature, on class, SECs and health, Muntaner et al. (2010) make the point that the use of NS-SEC is spreading from the UK to Europe and further afield, and, as a measure of employment relations, is of interest and significance for health inequalities in its own right (see Rose and Harrison 2009). The fact that classificatory schema like NS-SEC are considerably removed from Marxian-oriented theorisations of social class circumscribes without undermining their relevance for sociology of health inequalities (Wright 2009).

As far as England and Wales are concerned, the changes captured through studies of a succession of SECs on the one hand, and, say, life expectancy on the other, are unambiguous. They reveal an upward trend in life expectancy for men and women across the SEC spectrum. However, the improvement in life expectancy has been ‘more rapid among those at the top than the bottom of the socioeconomic hierarchy’ (Graham 2009: 12). The same pattern is evident in other developed or high-income countries (Mackenbach 2005). The latest data from the Office of National Statistics for England and Wales, using NS-SEC, are consistent with this broad-brush summary. A study of male mortality between 2001 and 2008, for example, found that in 2001 the mortality rate of those in routine and manual occupation was 2.0 times that of those in managerial and professional occupations; in 2008 that ratio had risen to 2.3. The authors note that ‘this pattern of declining absolute but rising relative inequalities is a well-known phenomenon in the context of declining overall mortality rates’ (Langford and Johnson 2010: 1). The recent Strategic Review of Health Inequalities (The Marmot Review 2010) affords comprehensive coverage of SECs and health and longevity.

It has been suggested that a ‘social gradient’ exists: in other words, relationships between SECs and measures of health and longevity are finely tuned: not only are there dramatic differences between best-off and worst-off in England and Wales, but the higher one’s social position (or for that matter, level of education, occupational status or housing conditions) the better one’s health is likely to be.

Muntaner et al. (2010), however, have shown that SECs like NS-SEC that are particularly sensitive to employment relations do not always reveal a graded relationship: small employers can exhibit worse health than highly skilled workers for example, and supervisors can display worse health than frontline workers.

There is a relative paucity of material for developing societies, certainly as far as social gradients are concerned. There is ample evidence nonetheless for ubiquitous inter- and intra-national health inequalities (WHO 2008).

Relations other than class and SECs

Passing reference has been made to class and SEC-related phenomena like poverty, income, educational qualifications, occupational status and housing tenure, to which might be added neighbourhood amenities, car ownership and so on. None of these phenomena are perfectly correlated with class/SECs, each being answerable to other forms of privilege and social division. Gender, ethnic and spatial relations have received most attention with regard to health inequalities in the UK (Graham 2007, 2009, Annendale 2010, Bradby and Nazroo 2010). This is not the occasion to review these distinctive and important fields, none of which reduces to class or SEC; but it is apposite to note both the heterogeneity and what Pease (2010) describes as the ‘normativity’ of privilege. The latter involves processes of ‘othering’, interpreted here as a method of portraying difference as if it were somehow alien from the socially orthodox or that which passes as normal. Table 1 lists a number of statuses that are valued, alongside their more negatively valued counterparts.

Table 1. Valued versus devalued social statuses
Social statusValuedDevalued
  1. Source: From Pease (2010).

GenderMaleFemale
RaceWhiteNon-white
LanguageEnglishOther
ReligionChristianMuslim
SexualityHeterosexualHomosexual
Life cycleAdultChild/Senior citizen
EducationEducatedIlliterate
Physical healthHealthyUnhealthy
AbilityAble-bodiedDisabled
Mode of incomeWorkState benefits
HousingHome ownerTenant
Marital statusMarriedUnmarried
Family sizeTwo childrenChildless/4+ children

While investigations of the salience for health inequalities of a number of items on Pease’s list have been fairly comprehensive, others remain neglected. A distinction needs to be drawn too between objective and subjective approaches to status. Marmot (2006), for example, maintains that people’s subjective sense of their social positioning is salient for their health. This issue is revisited below.

Alternative models and approaches

In a frequently cited review of the research literature, Link and Phelan (1995) conclude that social position is a ‘fundamental cause’ of health. Their case rests on the persistence of this association across time and place, independent of changes in the major causes of death (i.e. from infectious to chronic diseases). As Graham (2009: 13) observes, however, ‘noting that the association persists is not, of course, the same thing as explaining how it persists’. There has been a plenitude of models in a steady gathering of momentum over the last generation, The Black Report (1980) in the UK being an important catalyst (see Bartley 2003, Solar and Irwin 2005). Most of these are socio-epidemiological in orientation. Core data on SEC differences – which for some indicate a social gradient – are succinctly represented in this paragraph by Wilkinson and Marmot (2003: 10):

Both material and psychosocial factors contribute to these differences and their effects extend to most diseases and causes of death. Disadvantage has many forms and may be absolute or relative. It can include having few family assets, having a poorer education during adolescence, having insecure employment, becoming stuck in a hazardous or dead-end job, living in poor housing, trying to bring up a family in difficult circumstances and living on an inadequate retirement pension. These disadvantages tend to concentrate among the same people, and their effects on health accumulate during life. The longer people live in stressful economic and social circumstances, the greater the physiological wear and tear they suffer, and the less likely they are to enjoy a healthy old age.

There is no fully comprehensive and non-contentious way of categorising explanatory orientations to the enduring associations between SECs and health, each of which informs the quotation from Wilkinson and Marmot. The literature is voluminous, international and heterogeneous. There is a case, however, for distinguishing between: behavioural, material and psychosocial orientations.

Behavioural

The behavioural approach emphasises SEC differences in behaviours that are either damaging to or protective of health. In The Black Report (1980) these were referred to as ‘cultural/behavioural’ factors in acknowledgement of the cultural origins of many, apparently individual, behaviours. Most frequently noted and researched is the SEC-patterned consumption of food and drink (especially alcohol) and of cigarettes; but other salient behaviours are the degree of use of preventive medical services and of immunisation, contraception and antenatal care. Bartley (2003) maintains that longitudinal studies of behaviours pertinent for health show that such behaviours only account for about a third of SEC differences in morbidity and mortality. Moreover in the 30 years since the publication of The Black Report it has become apparent that people’s behaviours are often anchored not just in their culture but in their social and economic circumstances: eating healthily is not cheap and smoking can afford temporary relief in the face of the monotony of everyday lives devoid of tangible hope.

Material

The Black Report (1980) allocated clear causal priority to ‘material/structural’ factors for explaining SEC-related health inequalities. This approach highlights levels of wealth and income, but extends also to their social concomitants. Protracted unemployment, for example, has been associated with elevated rates of morbidity and mortality both for unemployed men and for their spouses. Work exposure to toxic or unsafe environments has been similarly exposed as injurious to health and longevity, as has damp housing. And as noted in the preceding paragraph, ‘risk behaviours’ for health are in part the issue of material disadvantage or deprivation. The authors of The Marmot Report (2010: 16) report that in England:

People living in the poorest neighbourhoods, will, on average, die seven years earlier than people living in the richest neighbourhoods … Even more disturbing, the average difference in disability-free life expectancy is 17 years … So people in poorer areas not only die sooner, but they will also spend more of their shorter lives with a disability. To illustrate the importance of the social gradient: even excluding the poorest five per cent and the richest five per cent the gap in life expectancy between low and high income is six years, and in disability-free life expectancy 13 years.

Bartley (2003) gives weight to such findings but maintains that the full significance of material living standards for health can only be appreciated over the life-course as a whole.

Psychosocial

Advocates of a psychosocial approach insist that a credible explanation of health inequalities must incorporate ‘psychosocial risk factors’ like social support, work autonomy, and what Seigrist (2009) refers to as the balance between efforts and rewards (within and without the workplace). It is an orientation that accents the ramifications of social inequality for how people see, define and evaluate themselves and their projects. Wilkinson (1996) complements Marmot’s (2006, 2010) studies by contending that income inequality leads to social fragmentation and dislocation, and thence to a breakdown in social networks, mutuality and trust; these, in their turn, jeopardise health and well-being. Social capital is critically diminished. Some commentators hold that prolonged stress due to psychosocial factors results in an increase in ‘allostatic load’: if too many negative changes occur too rapidly, bodily adjustment is compromised, resulting in overload and exhaustion. An inconsiderate manager would be harder to accommodate in a workplace that was overly hot, cold or noisy, or when the individual had been on an inadequate diet. There is prima facie evidence for an association between SECs and allostatic overload.

Of course these three subsections do scant justice to the research programmes they purport to represent, but then these are literatures that are exhaustively, one might almost say tediously, reviewed. Nor do they by any means exhaust the range of perspectives on health inequalities either emanating from or addressed by sociologists.

If the behavioural, material and psychosocial orientations are mainstream, each with its evidence base, it would be remiss not to mention other, divergent perspectives. One example focuses on what in the Black Report (1980) was called ‘social selection’. Canning and Bowser (2010: 1223) see the socioeconomic gradient in health ‘to be in large part the result of differentials in health, reversing the direction of causality put forward in the Marmot reports’. They advocate direct health interventions, especially in child health, as mechanisms for improving both health and socioeconomic outcomes. Ironically, as we shall see, this emphasis on health interventions in childhood mirrors Marmot’s own recommendations. Another argument is that it is intelligence, or IQ, that strikes as the most impressive independent variable for health inequalities (see for example Batty and Deary 2004).

Although more vehicle than theory, life-course approaches to explaining health inequalities call on the temporal dimension and warrant a special mention. The underlying premise is that factors disadvantageous for health and longevity tend both to cluster and to accumulate over time. The phenomenon of clustering comes as no surprise: people residing in low-income households and sub-standard housing are most likely to find themselves in neighbourhoods lacking in social amenities and to be impoverished socially as well as economically (although there are of course exceptions). Childhood emerges in the research literature and in rival paradigms alike as a pivotal time for health, and this is reflected in the later discussion; but childhood is not of course the only life-stage of significance for health.

In a contribution on life-course influences on health at older ages, Bartley and Blane (2009) develop the themes of clustering and accumulation to address the issue of how the social becomes biological. They cite their own work on the ‘inverse housing law’ to show that residence in poor-quality housing in a region with severe climate is associated with both reduced lung function (Blane et al. 2000) and elevated diastolic and systolic blood pressure (Mitchell et al. 2002). The ‘inverse housing law’, they maintain, particularly affects older people since they (a) often lack the means to renovate their homes against the local climate, and (b) are more vulnerable to air pollution because of age-related respiratory and cardiovascular decline. They close their contribution with an elegant proposition: ‘the past is written into the body while the present shapes behaviour and reactions to life’ (Bartley and Blane 2009: 61).

There is merit in each of the mainstream behavioural, material and psychosocial approaches while the concept of the life-course provides an important temporal frame for future research. Others have ventured a more comprehensive or ‘inclusive’ statement. For example, Solar and Irwin (2005) provide a synthesis of studies conducted within this research programme in the form of a model. Their model implies that social and political contexts generate sets of unequal socioeconomic positions, leading to a stratified difference by ‘income levels, education, professional status, gender, race/ethnicity and other factors’. Socioeconomic positioning underwrites the structural determinants of health inequalities. Socioeconomic positions translate, via differential exposure and differential vulnerability to factors known to compromise health, into specific or intermediary determinants. Of relevance here are material conditions, like work and housing circumstances; psychosocial conditions, like stressful life events; and behavioural factors, like smoking. A distinctive feature of Solar and Irwin’s (2005) model is its incorporation of the health care system. They draw attention to the active role of the Department of Health in Britain in trying to counter health inequalities through equitable access and public health and cross-department policy initiatives.

Towards a sociological theory of health inequalities

Approaches or orientations rarely add up to a theory, even if they step tentatively in that direction. Nor is a model the same as a theory: it is an heuristic device, its merits extrinsic rather than intrinsic (Scambler 2007). Coburn’s (2000) critique of Wilkinson’s (1996)‘relative income hypothesis’ offers a much closer approximation to a sociological theory of health inequalities. Coburn’s objection to Wilkinson’s position stems from: (1) its use of level of income as a starting point; and (2) the priority it then accords to psychosocial pathways (with loss of social cohesion and trust providing the mechanisms linking high rates of income inequality with diminished health and longevity (and many other ‘social evils’; see Wilkinson and Pickett 2009). Wilkinson argues that the common assumption that lack of material resources led directly to the social gradient has proven false; ‘it now looks as if a major part of the association between low social status and poorer health springs from the experience of low social status or subordination itself’ (Wilkinson 2000: 998–9). This conclusion is in accord, he maintains, with ‘the increasing success of psychosocial factors in explaining health inequalities’. Marmot’s (2006) emphasis on ‘status syndrome’ is in similar vein.

Coburn rejects the socio-epidemiological fixation on associations between SECs and health via the social gradient. This is too far removed from the empirical explication of Link and Phelan’s (1995)‘fundamental cause’ of health inequalities: ‘there has been an overwhelming tendency to focus on the possible social/psycho-biological mechanisms through which social factors might be tied to health rather than on examination of the basic social causes of inequality and health’ (Coburn 2000: 137). For Coburn, it is the growing dominance of neo-liberalism from the mid-1970s that is the causal progenitor of income inequality and its health-compromising sequelae. This new neo-liberal order, he suggests, rests on three premises: (1) that markets are the optimally effective allocators of resources in production and distribution; (2) that societies are composed of individual producers and consumers motivated in the last resort by material considerations; and (3) that competition is the prepotent vehicle for innovation. Economic globalisation and neo-liberalism are natural bedfellows. With the possible exception of Japan, the more neo-liberal or market-oriented the nation-state, the greater the income inequality within its borders, and the higher the level of social fragmentation and the lower the social cohesion and trust (see also Wilkinson and Pickett 2009).

Coburn developed his position. In a 2004 paper he posited a more explicitly ‘class-based’ or ‘class/welfare regime’ theory, contending that global and national social and political trends have increased the power of business classes and lowered that of working classes, although international pressures towards neo-liberal doctrines and policies have been differentially resisted by nation-states, largely because of their differing historical and institutional structures. The essence of his revised thesis is that socio-epidemiological research indicates that neo-liberalism is associated with greater income inequalities, inter- and intra-nationally (see WHO 2008). Countries with less neo-liberal – that is, social democratic – welfare regimes (e.g. Finland, Sweden, Norway) have better health than those with more clearly neo-liberal regimes (e.g. USA, Britain). Previous analyses have for the most part, Coburn suggests, linked factors like income inequality and social cohesion with health or well-being, while his version ‘deepens’ the causal explanation by including economic globalisation, neo-liberalism and the power of capital, as well as welfare regimes and markets.

Scambler and Higgs (2001), sympathetic critics of Coburn’s early paper, anticipated his subsequent shift in their proclamation of the causal power of class. The former has argued that the period since the early 1970s has witnessed a change in the class/command dynamic; namely, an intensification of class power relative to that of the (increasingly privatised and regulatory) state. Polemical illustration is afforded by the greedy bastards hypothesis (or GBH) (Scambler 2007, 2009). This asserts that Britain’s widening health inequalities can be seen as a largely unintended consequence of the voracious, ‘strategic’ appetites of a hard core or cabal in its strongly globalised capitalist-executive, backed by its more weakly globalised power elite. If men of money have always bought men of power, to paraphrase the historian Landes (1998), they have got more for their money since the early 1970s than they did in the immediate postwar years of consensual welfare-statism.

Underlying the GBH is the claim that new ‘flexible’ or de-standardised work patterns, the growth of income inequality, welfare cuts, and ‘derivative’ processes like the postmodernisation of culture and novel and divisive forms of individualism, have their origins in the strategic behaviour of the GBs (Harvey 2010). When, for example, the CEOs and directors of large transnational companies, along with financiers and rentiers, pocket huge pay packages, pension pots and ‘honours’ for (a) downsizing workforces, (b) substituting part-time for full-time workers, (c) reducing work autonomy in favour of micro-managerial control, (d) outsourcing, and (e) ending final salary pension schemes, they deleteriously affect the health and longevity of (ex-)employees, and by doing so contribute positively to health inequalities. Strategic decision-making on the part of CEOs, plus the financiers, rentiers and so on that help comprise the capitalist-executive (Scambler and Scambler forthcoming), is in and of itself a risk behaviour for health. It is an approach that has more resonance with Engels and Virchow than with much recent social epidemiology. Analogously, researchers have emphasised the corporate role in influencing social and health policies and in peddling commodities like tobacco (Baba et al. 2005, Farnsworth and Holden 2006).

Scambler (2007) argues that there are empirically discernible media through which class and other structural relations realise their influence on health and life expectancy. He refers to these as capital or asset flows. These are summarised in Table 2. Asset flows vary temporally and so are rarely either possessed or not, it almost invariably being a matter of degree or strength of flow. Moreover there is frequently interaction or compensation between flows. A reduced biological asset flow, for example, might be compensated for by a strong psychological asset flow, or an arrest in the flow of material assets by strong flows of social or cultural assets. As epidemiological research on the clustering of risk factors for health suggests, there is a tendency for flows to be weak or strong across assets. Scambler accords prepotent significance to the flow of material assets.

Table 2. Types of asset flow salient for health and longevity
  1. Source: Adapted from Scambler and Scambler (forthcoming).

Biological (or ‘body’) assets can be affected by class relations even prior to birth. Low-income families, for example, are more likely to produce babies of low birthweight; and low birthweight babies carry an increased risk of chronic disease in childhood, possibly through biological programming.
Psychological assets yield a generalised capacity to cope, extending to what is increasingly conceptualised as ‘resilience’. In many ways the ‘vulnerability factors’ that Brown and Harris (1978) found reduced working-class women’s capacity to cope with ‘life events’ of causal importance for clinical depression are class-induced interruptions to the flow of psychological assets.
Social assets have come to assume pride of place in many accounts of health inequalities, and feature strongly in the contributions of Marmot and Wilkinson. Social assets, or ‘social capital’, refer to aspects of social integration, networks and support. The political manipulation of social capital should not lead to its neglect.
Cultural assets, or ‘cultural capital’, are generated initially through processes of primary socialisation, subsequently encompassing formal educational opportunities and attainment. Class-related early arrests to the flow of cultural assets can have long-term ramifications for job prospects, income levels, and therefore health.
Spatial assets have been revealed as salient for health via area-based studies. These studies have shown that areas of high mortality tend to be areas with high net out-migration; and it tends to be the better qualified and affluent who exercise the option to move.
Symbolic assets represent the variable distribution of social status or ‘honour’. Status/honour is known to impact on people’s health. It has been shown for example to exercise an impact via people’s sense of ‘where they stand’ relative to others comprising their reference groups.
Material assets translate into ‘standard of living’. Relative poverty or deprivation due to impoverishment and meagre standard of living has long been associated with diminished health and reduced longevity, although the mechanisms linking material disadvantage with health remain hotly debated.

There are subjective as well as objective dimensions to asset flows. In Marmot’s (2006)Status Syndrome, for example, bringing to sociologists’ minds the insight that if people define their situations as real they are real in their consequences, it is suggested that people’s own assessments of their status have pertinence for their health and longevity. So people’s definitions of their situations vis-à-vis each of their current asset flows carry a potential to impact on their health.

Asthana and Halliday (2006) point out the difficulties of measuring asset flows, and, more generally, of verifying or falsifying their role as media of enactment of a new class/command dynamic. Scambler’s (2010) defence is that difficulties of measurement, raised almost exclusively by epidemiologically-oriented quantitative sociologists prejudicially committed to ‘variable analysis’, must not be allowed to get in the way of ‘making a good case’, which is all sociologists can aspire to. Moreover, the making of a good empirical case amounts to much the same as making a good theoretical case.

Tackling health inequalities

The phrase ‘tackling health inequalities’ means different things to different people. Articulated as Weberian ideal types, for those coming from a ‘policy sociology’ perspective engagement for change involves working with people of influence; while for those coming from a ‘critical sociology’ perspective it can involve working against them (Burawoy 2005). Goraya and Scambler (1998) write of a location paradox: ‘insiders’ are heard (but typically co-opted), while ‘outsiders’ remain largely unheard.

Health inequalities in North America have long been studied by social epidemiologists, sociologists and others, most persistently in sociological vein by the critical Marxist sociologist Navarro (2002); but the term ‘health disparities’ has often been the term of preference, and references to SECs, let alone class, have lagged behind those to race and ethnicity (for a while ‘health variations’ displaced health inequalities in Britain, heralding risks of insider co-optation). Race, Navarro has always insisted, masks class in the US. One persisting difficulty in the US, however, has been the relative lack of good data sets (Lahelma 2010).

In the last decade there has been some reconfiguring of the goals of public health policy, principally but not exclusively in developed countries. The emphasis has increasingly been on the ‘twin goals’ of (1) improving population health, and (2) reducing health inequalities. In Britain, Saving Lives: Our Healthier Nation spoke of ‘improving health for all and tackling health inequality’ (SSH 1999: 2). A year later it was announced in the US that the prevailing strategy ‘is designed to achieve two over-arching goals: increase quality and years of healthy life (and) eliminate health disparities’ (USDHHS 2000: 2). These changes of emphasis were far from exceptional. Moreover the WHO 2008 Commission on Social Determinants of Health committed later to the same aspirations at a global level. This timeframe has seen a greater acknowledgement too of Link and Phelan’s (1995)‘fundamental causes of disease’, sometimes explicitly anchoring risky health behaviours in social and material circumstances (Graham 2009). The multi-dimensional and comprehensive recommendations of the WHO’s Commission are outlined in Table 3.

Table 3. Recommendations of the WHO Commission on Social Determinants of Health
  1. Source: WHO 2008.

(1) Daily living conditions
(a) a comprehensive approach to early child development, building on existing child-survival programmes and extending interventions in early life to include social/emotional and language/cognitive development;
(b) sustained investment in rural development, addressing policies of exclusion that lead to rural poverty, landlessness and displacement of people from their homes; urban governance and planning;
(c) economic and social policy responses to climate change and other environmental degradation take into account health equity;
(d) full and fair employment and decent work as a central aim of national and international social and economic policy-making; safe, secure and fairly-paid work, year-round work opportunities, and a healthy work-life balance for all; and improved working conditions for all workers in order to reduce exposure to material hazards, work-related stress, and health-damaging behaviours;
(e) comprehensive social-protection policies that support an income level conducive to healthy living for all; and
(f) specifically with regard to the health sector, the Commission calls for the building of universal health-care systems oriented around primary health care.
(2) Inequitable distribution of power, money and resources
(a) place responsibility for action on health and health equity at the highest level of government, and ensure its coherent consideration in all policies;
(b) adjust the health sector as appropriate – include social determinants in the policy and programmatic functions of health ministries’ stewardship role in supporting a social determinants approach throughout government;
(c) strengthen public financing for action on social determinants; increase international financing for health equity, and coordinate increased finance by means of a framework for acting on social determinants;
(d) reinforce the primary role of the State in providing basic services essential to health (such as water and sanitation) and regulating goods and services with a major impact on health (such as tobacco, alcohol and food);
(e) address gender bias in the structures of society – in laws and their enforcement, in the way organisations are run and interventions designed, and in how a country’s economic performance is measured;
(f) reaffirm commitment to addressing sexual and reproductive health and rights universally;
(g) empower all groups in society through fair representation in decision-making about how society operates, particularly in relation to its effect on health equity, and create and maintain a socially inclusive framework for policy-making; and
(h) enable civil society to organise and act in a manner that promotes and realises the political and social rights affecting health equity.
(3) Monitoring the problem and interventions
(a) ensure that routine monitoring systems for health equity and social determinants are in place locally, nationally and internationally;
(b) invest in generating and sharing new evidence on how social determinants influence population health and health equity, and on the effectiveness or otherwise of measures to reduce health inequities through action on social determinants;
(c) provide information about social determinants to policy actors, stakeholders and practitioners, and invest in raising public awareness.

Unsurprisingly, the recommendations of the UK’s Marmot Report (2010) are in line with those of the WHO Commission. Economic growth is not all. Fairness and social justice are emphasised, as is the linkage between social and health inequalities. ‘Proportionate universalism’ is commended, namely, universal action, but with a scale and intensity that is proportionate to level of disadvantage. In this context six policy objectives are set:

  • 1 give every child the best start in life;
  • 2 enable all children, young people and adults to maximise their capabilities and have control over their lives;
  • 3 create fair employment and good work for all;
  • 4 ensure healthy standard of living for all;
  • 5 create and develop healthy and sustainable places and communities; and
  • 6 strengthen the role and impact of ill health prevention.

One condition of realising these objectives is action by central and local government, the NHS, the third and private sectors and community groups; another is effective participatory local decision-making (which in turn depends on empowering individuals and communities).

The omens are not auspicious. Echoing the analyses of Engels (1987/1845) and Vichow (1985/1848), as well as Navarro (2009), Scambler and Scambler (forthcoming) highlight enduring contradictions within capitalism – social and private appropriation, the strategic rationality of the corporation and the economic ‘anarchy’ of the wider society, the polarisation of wealth and poverty, the privileging of profit over need (Beuchler 2008) – and the ‘management’ of crisis tendencies. If not managed, crisis tendencies can become full-blown legitimation crises (Habermas 1975) as was briefly threatened with the ‘global financial crisis’ in 2008–2009 (seen by a number of commentators as a ‘window of opportunity’ for reformers (Graham 2009).

While in the liberal capitalism of the nineteenth century the market provided ‘system integration’ by coordinating the production and distribution of material goods, and ‘social integration’ by providing norms, values and identities that reinforced people’s economic motivation, this is no longer the case. Through twentieth and twenty-first century capitalism transnational corporations have come to monopolise production, set prices and manipulate demand, thus nullifying the putative benefits of competition, price reduction and so on. They are also characterised by much more state intervention (in part in response to liberal capitalism’s failure to provide system and social integration). The state now underwrites needed but unprofitable goods and services, maintaining the infrastructure, subsidising education and training for workers, providing social insurance for the unemployed, people with disabilities and the retired, and mitigating and repairing the environmental by-products of capitalism. The potential for crisis is thereby enhanced.

It has been contended that the new class/command dynamic alluded to above has precipitated an assault on forms of state intervention that have hitherto insured capitalism against potential trauma. Harvey (2005: 115–16) queries the lack of analysis of this same class/command dynamic:

The possibility, for example, that the ruling ideas might be those of some ruling class is not even considered, even though there is overwhelming evidence for massive interventions on the part of business elites and financial interests in the production of ideas and ideologies: through investment in think-tanks, in the training of technocrats, and in the command of the media. The possibility that financial crises might be caused by capital strikes, capital flight, or financial speculation, or that financial crises are deliberately engineered to facilitate accumulation by dispossession, is ruled out as far too conspiratorial even in the face of innumerable suspicious signs of co-ordinated speculative attacks on this or that currency.

Interestingly this passage was written prior to the financial crisis of 2008–2009.

The WHO Report of the Commission on Social Determinants of Health (2008) and The Marmot Report in the UK (2010) appear to take strong lines on the need for urgent and radical action. The WHO Report, for example, includes among its targets ‘the inequitable distribution of power, money and resources’ (see Table 3). On the face of it, this is subtle, focused and apt. Moreover it is strategic, calibrated not only to inform but to facilitate relevant shifts in governance. On other levels, however, it not only glosses over the contradictions of capitalism, but denies them: it presents social reality as other than it is, as if social structures like class do not exist. In doing so it effectively underwrites an ideology reflective of the new class/command dynamic. Social stratification and social structures are fundamental for health inequalities and for attempts to tackle them (Lahelma 2010).

After the manner of Scambler (2007), Coburn (2009: 44) insists that the sociologists must see beyond SEC qua variable to re-focus on class qua social structure:

People with high SES (socioeconomic status – GS) do indeed live longer than those with less. SES, however, is a mere ranking of people according to income, educational attainment or occupational position. It reflects standard of living generally, and because these standards are related to many different types of disease, it is a good correlate of health status. But SES is itself a result of class forces. The nature of the capitalist class structure, and the outcome of class struggles, determine the extent and type of socioeconomic inequalities in a given society, and the socioeconomic inequalities in turn shape the pattern of health – and health care. But while many theorists of the social determinants of health proclaim an interest in the basic determinants of health and health inequalities, much of their literature omits any consideration whatsoever of the political and class causes of SES and the SES-health relationship. When they speak of analyzing the ‘causes of disease’, they seldom go far enough up the causal chain to confront the class forces and class struggles that are ultimately determinant.

To reiterate a point made earlier, this does not render research on the associations between an array of SECs and health inequalities redundant. To paraphrase Wright (2009), the fact that sociologists interested in health inequalities have tended to neglect Marxian approaches to ‘exploitation and domination’ does not mean there is no sociological return on accessing class via Weberian emphases on ‘opportunity hoarding’ or even via ‘attributes and conditions’.

Summary, conclusion and agendas

The literature on health inequalities is immense. Quantitative sociology seems often to be subsumed by impressive social-epidemiological research programmes (perhaps in the UK due to a deficit in skills). The contributions of qualitative sociologists have been fewer and have coalesced into a more or less discrete discourse. In what has necessarily been a selective review, more references have been made to quantitative than to qualitative research; to SECs and class than to other social divisions; and the overall thrust has been theoretical, that is, sociological. A genuine sociology of health inequalities, it has been implied, must be rooted in the discipline, regarding data, whether acquired through quantitative, qualitative or ‘mixed methods’, as mere clues as to what is happening and why. It is a field in which Wright Mills (1963) plea for ‘sociological imagination’ is more often cited than heeded. It is a field, too, in which those Hegel would have defined as ‘unserious’ (i.e. those who advocate positions they neither believe in nor live by) are currently over-represented, possibly another by-product of the prevailing class/command dynamic. ‘No reviews or policies’, Pickett and Dorling (2010: 1233) rightly insist, ‘ “boldly go” where all public health researchers know they need to go’ (see also Dorling 2010). Many sociologists operate in wilful denial of those ‘enduring social structures’, like class, but not only class, that underwrite health inequalities and undermine reforms to reduce them.

Research conducted within the categories of ‘behavioural’, ‘material’ and ‘psychosocial’ have much to offer, as do the assorted models offered as syntheses. They are all grist to the sociological mill. Arguably, however, they do not add up to a sociology of health inequalities. Coburn’s (2000) critique of Wilkinson’s (1996) bold work in epidemiology has paved the way for a more distinctively sociological contribution. However, there remain issues either far from resolution or as yet substantially unaddressed, including the following:

  • 1 There is no question that fields of enquiry ranging from genetics to anthropology – embracing, for short, biological, psychological and social mechanisms – (1) each contribute causally to health and health inequalities, and (2) are simultaneously active. Given also the intrusions of contingency and agency into everyday events, just what causal weight can be attached to social mechanisms in the genesis and durability of health inequalities?
  • 2 There have been few attempts to integrate macro-, meso- and micro-oriented sociological research on health inequalities (Williams 2003). There is in particular a dearth of Mertonian theories of the middle-range (after the manner of Siegrist’s 2009 postulation of effort-reward imbalance for example). How, and following what timetables and routes, do social structures like relations of class, command, gender, ethnicity and so on come to impact on individuals and their bodies?
  • 3 As Doyal (1979) prophetically indicated a generation ago, and as is perhaps best captured now in the contributions of Szreter (2002, Szreter and Woolcock 2004), there is a need for a political economy of health that transcends the nation-state: the global financial crisis of 2008–2009 bears eloquent testimony to this. National health inequalities can no longer be explained without reference to transnational social mechanisms; and no more can they be effectively ‘tackled’ nationally.
  • 4 The obduracy of the latest reincarnation of global financial capitalism and its contradictions cannot be glossed over. As Marmot (2006, 2010) documents and Harvey (2005, 2010) helps explain, health inequalities from the USA to China, Brazil, Russia and Nigeria betray the global or transanational nature and salience of social structures. To what extent are inter- and intra-national health inequalities (1) the product of transnational social structures or mechanisms, which (2) account no less compellingly for the repeated failure of reforms to reduce them?
  • 5 Are the contradictions and structures of contemporary capitalism such that addressing the health of populations and health inequalities in the developed nations will likely have adverse effects on the health of populations and health inequalities in developing nations? Indeed, is there a prior contradiction between addressing the health of populations and health inequalities?
  • 6 The dearth of sociological interest in capitalism’s contradictions and their bearing on health inequalities can itself be said to involve contradiction, involving a denial of sociology’s own classical heritage. Much policy sociology is at one and the same time (1) tailored to appeal to insiders, and (2) a distortion of social reality.

It is appropriate to return finally, if obliquely, to some of the philosophical questions posed at the beginning of this paper. Chandra and Vogi (2010) note the gap between knowledge and implementable policy in relation to health inequalities in the UK and elsewhere. Half a century ago Miliband (1961) showed why the ‘formal’ democracy represented by the British parliamentary system could not deliver on policies of the kind required to reduce health inequalities. Research since suggests he was right, and further that today’s ‘career politicians’ may lack even the political will of their predecessors (Oborne 2007). Marmot (2010) frequently quotes Neruda’s injunction to ‘rise up with me against the organisation of misery’; and he intends this globally rather than nationally. Is this revolutionary talk or not? If he and sociologists of health inequalities are serious, then there will have to be a sociological reckoning with the contradictions of capitalism and the likes of transnational and national relations of class and command, a step well beyond a fascination with SECs and research designed to refine our understanding of the social gradient.

There is a case for rescuing principles of equality, freedom, democracy, justice and so on from the largely asocial and self-referential discourse of Anglo-Saxon philosophers, a process Sen in particular has pioneered. Sociological engagement with the elimination or reduction of health inequalities calls for an interrogation of what this might mean in global citizens’actually existing social and natural worlds. It is evident that under the sway of Western neo-liberal ideology many a social formation is becoming more not less equal. The urgent questions for sociologists would seem to be: why this is; whether, and under what conditions, trends to greater inequality might be arrested and reversed; and what the prospects are for realising these conditions (for Marmot’s ‘rising up’). It is a challenge more often posed than taken up (Bambra et al. 2010, Mackenbach 2010). Addressing these questions will require a grounding of independent but not absolute (philosophical) principles, and further reflection on trade-offs between them in actually existing social and natural worlds. In the absence of attempts to answer these questions, and of such reflection, it is difficult to argue that the sociological project is altogether coherent or serious.

Ancillary