Research highlights

• • England has been proactive in coordinating health inequalities (HI) policies. Yet, HI continue to widen.

• • Much of the policy focus has centred around the role of the National Health Service (NHS).

• • This is in part an outcome (albeit unintended) of flows of investment.

• • NHS resource allocation has been used to deflect attention away from macroprocesses of social inequality.

• • Although well-meaning, the targeting of NHS resources to this end does not promote health equity.

The widening gap: whose problem?

The UK Labour Government of 1997–2010 invested considerable policy effort and resources into the problem of HI. During its first term in office, Britain was arguably already at a more advanced stage than its European counterparts with regard to coordinating HI policies (Mackenbach and Bakker, Reference Mackenbach and Bakker2002). Thereafter, there was the launch of a cross-government HI strategy (A Programme for Action) and the introduction of the first ever national Public Service Agreement target for HI, measures which raised awareness, recognised the need to tackle the wider determinants of HI and the fact that many levers lay outside the NHS (Bull and Hamer, Reference Bull and Hamer2001). This was supported by a series of toolkits, best practice guidance and other resources (such as a National Support Team) designed to improve the capacity of health and other agencies to counter HI, as well as the publication of a number of landmark reports on the causes of and potential strategies to address HI (e.g. Acheson, Reference Acheson1998; Wanless, Reference Wanless2004; Marmot et al., Reference Marmot, Allen, Goldblatt and Boyce2010).

Despite this high policy profile, HI, as measured by life expectancy at birth and infant mortality, have continued to widen. Between 1992–1994 and 2001–2003, the absolute difference in life expectancy between local authority areas with the highest and lowest life expectancies increased from 8.9 to 9.4 years (Shaw et al., Reference Shaw, Davey Smith and Dorling2005). By 2007–2009, a boy born in Kensington & Chelsea would be expected to live for 84.4 years, compared with 73.7 years in Blackpool [and, illustrating the continued divergence in life chances across the United Kingdom, 71.2 years in Glasgow City (Office for National Statistics (ONS), 2010a), a difference of 13.1 years]. Similarly, although infant mortality rates (IMRs) have narrowed marginally after a period of widening, in 2008, IMRs for routine occupations (NS-SEC group 7) were twice that for higher managerial occupations (NS-SEC group 1), at 5.5 and 2.8 deaths per 1,000 live births, respectively (ONS, 2009). As a result, the Government's target of a 10% reduction in the relative gap between 1997–1999 and 2009–2011 is unlikely to be reached. Similarly, although health improvements have been made [witness the narrowing of inequalities in premature mortality from circulatory diseases and cancers (DH, 2009)], the inequalities target for all-age all-cause mortality will not be met if current trends continue (National Audit Office (NAO), 2010).

There are a number of plausible explanations for these persistent variations. Over the past 20 years, HI research, previously dominated by a concentration on individual-level ‘lifestyle factors’, saw a profusion of pioneering work exploring, for example, the role of risk factors at different periods of the life course (Barker, Reference Barker1994; Hertzman et al., Reference Hertzman, Power, Matthews and Manor2001; Davey Smith, Reference Davey Smith2003; Kuh et al., Reference Kuh, Ben-Shlomo, Lynch, Hallqvist and Power2003) and the psychological effects of living in an unequal society (Wilkinson, Reference Wilkinson1996; Brunner and Marmot, Reference Brunner and Marmot1999; Kawachi et al., Reference Kawachi, Colditz, Ascherio, Rimm, Giovanucci, Stampfer and Willett1999; Brunner, Reference Brunner2001; Marmot, Reference Marmot2004; Wilkinson and Pickett, Reference Wilkinson and Pickett2009). Today, it is broadly agreed that the unequal distribution of health outcomes reflects the unequal distribution of the social factors that influence health (Graham, Reference Graham2004) and that early life circumstances, the life trajectories of children and young people, and adult circumstances can all have a bearing on adult health, although the relative strengths of these effects may vary across different cohorts and health outcomes (Graham, Reference Graham2002).

Evidence suggests that with respect to several key indicators, notably income, wealth, education and child welfare, there has been growing socio-economic polarisation in the United Kingdom. During the past decades income inequality for all households has increased substantially, from a Gini coefficient of 28 in 1983 to 34 in 2008/09 (ONS, 2010b). The share of income received by the richest 1% in Britain has become particularly skewed, returning to 1935 levels from an all-time low in 1974 (Dorling, Reference Dorling2010). Because of rising house prices (a major repository for wealth), there has also been an increase in wealth inequality between those who own their properties and those (32% of English households) who do not (Levin and Price, Reference Levin and Price2010). Despite some signs that social differences in examination results may have started to reduce (Equality and Human Rights Commission (EHRC), 2010), educational inequality in the United Kingdom remains higher than the average for 24 Organisation for Economic Co-operation and Development (OECD) countries (The United Nations Children's Fund (UNICEF), 2010). Indeed, internationally Britain is one of the poorest performing OECD countries with respect to the lives and well-being of its children, scoring particularly badly in terms of children's material well-being, family and peer relationships, behaviours and risks and subjective well-being (UNICEF, 2007).

This is not to argue that the government neglected the wider determinants of health. Initiatives such as the National Minimum Wage, tax credits, pension credits, the Child Poverty Strategy, the Sure Start Programme and the Decent Homes Programme (see e.g. Hills and Stewart, Reference Hills and Stewart2005) variously made important contributions. However, the emphasis was on disadvantaged areas and groups rather than on addressing either the social gradient or the continued polarisation of income and wealth at the upper end of the spectrum. Inequalities thus continued to widen, fuelled inter alia by an over-reliance on financial services, the forces of globalisation and the way the taxation system continued to disadvantage those on lower incomes.

Given what we know about the multiple pathways linking the social determinants of health to a range of health outcomes across the life course (Asthana and Halliday, Reference Asthana and Halliday2006), the most obvious explanation for the continuing health gap thus lies in the unequal distribution of biological, psychological, social, cultural, spatial and material capital and in the generative mechanisms that give rise to the unequal distribution of these factors (Scambler, Reference Scambler2001). Yet, in spite of this, HI policy – and responses to that policy – has come to focus increasingly on the role of the NHS. This culminated in 2006 with HI being listed at the top of the NHS's six service priorities (DH, 2006). At the national level, lead responsibility for delivering the Government's HI targets lay with the DH, whereas locally it was Primary Care Trusts (PCTs) that were expected to take a lead in agreeing priorities, strategies and plans for Local Strategic Partnerships (LSPs).

These lines of accountability and leadership may go some way towards explaining why, despite a policy rhetoric that acknowledged that improvements in health and well-being require cross-governmental action, HI have often failed to become an area of corporate responsibility. At the national level, for example, although Health Impact Assessment is now a mandatory component of the regulatory Impact Assessment (IA) of government policy, a recent review of 310 IA reports (spanning 15 government departments) found that very few policy teams demonstrated a deep understanding of the ways in which their policies impinged upon health. When health impacts were considered, there was a tendency to focus on negative impacts (as opposed to the role non-NHS policies could play in promoting health), on a small number of health determinants, and to ignore HI (DH, 2010a).

Moving to the local level, for a large part of this period, the drive to tackle HI in a local context was centred on LSPs and the allied Local Area Agreements (LAAs; Perkins et al., Reference Perkins, Smith, Hunter, Bambra and Joyce2010), which provided a formal framework for partnership working between public, private and third sector organisations. The LAAs, developed by local authorities (LAs) with their partners in the LSP, were three-year action plans focusing on a mix of national and local priorities and targets relevant to the local area. Yet, progress through this medium has been difficult. The Audit Commission found health and HI were major issues when setting LAAs but they were also ‘one of the issues that local partners found most taxing’ and inspectorates had significant concerns relating to progress in tackling health issues in twenty-one areas (Audit Commission, 2010: 6). The Improvement and Development Agency for Local Government (Idea.gov) highlights how, because of time pressures and direction given by central government, LAAs also tended to focus on national priorities and established delivery mechanisms. As a result, ‘the role of wider determinants of health has not been explored locally in any depth’ (Idea.gov, see also Elliot et al., Reference Elliot, Bjogal and Williamsnd). Thus, although LAs hold some levers that shape the social determinants of health and, since 2006, have been supported by the National Support Team to promote wider, social and environmental improvements on HI, their efforts have tended to focus more narrowly on service access and/or lifestyle issues.

Locally, the tendency has been for non-health sectors to take insufficient account of the HI agenda. Again, this reflects a perception that responsibility for delivery lies with the local NHS, as well as financial pressures, a fragmentation of funding streams and a lack of relevant public health evidence for upstream interventions (Marmot et al., Reference Marmot, Allen, Goldblatt and Boyce2010: 88). A systematic review of the literature (Hunter et al., Reference Hunter2008), suggested, for example, that area-based partnerships (such as the New Deal for Communities (NDC) and Health Action Zones) had not achieved better improvements to population health than comparator areas. Indeed, within the NDC, residents with lower education not only experienced the least favourable health profiles at baseline but also the smallest improvements (Stafford et al., Reference Stafford, Nazroo, Popay and Whitehead2008). Like the NDC, the National Strategy for Neighbourhood Renewal represented an attempt to include health within the wider regeneration agenda. Introduced in 2001 in the most deprived areas, health was one of the six key outcomes. Outcomes have improved and inequalities decreased in some areas, most notably housing and education (Blackman, Reference Blackman2006), suggesting HI might reduce in the longer term. However, one of the key obstacles to embedding effective multi-sectoral action has been the ongoing tension between short-term targets and policy cycles and the time required to address the structural determinants of health, which, in consequence, tend to attract limited attention (see e.g. Lawless, Reference Lawless2004).

With the NHS at the heart of HI policies, the political response to enduring health disparities has, naturally enough, been to focus on the failure of the health service to tackle HI effectively. For example, echoing an observation made in Lord Darzi's NHS Next Stage Review (DH, 2007a), a recent report by the House of Commons Committee of Public Accounts (2010) examining why the DH has failed to meet its HI target hones in on the failure of the Department to address GP shortages in areas of ‘highest need’. The report also highlights what it describes as widespread ‘funding shortfalls in the most deprived areas’, as well as the fact that funds were not allocated specifically to tackle HI but instead had to be drawn from general allocations.

The idea that the failure to tackle HI reflects inadequate investment by the NHS in preventative measures is also pervasive (House of Commons Health Committee, 2009; Royal College of Physicians, 2010). Yet, there has been significant growth in prevention expenditure in the United Kingdom since 2000, from an estimated 1.8% of total health expenditure (0.6 percentage points below the then OECD average of 2.4%) to around 4.0%. This now places the United Kingdom above the OECD average, which, in 2006/07, stood at 2.8%. Indeed, if expenditure on preventative pharmaceuticals is included, 5.4% of total UK health expenditure is directed towards prevention (Butterfield et al., Reference Butterfield, Henderson and Scott2009).

Given evidence that, compared with other sources of variation such as education, income and job status, the targeting of NHS efforts (through, e.g. assigning Spearhead status to the worst health areas and investing in preventive health) appears to have had a negligible impact on HI (Costa-Font et al., Reference Costa-Font, Hernández-Quevedo and McGuire2010), it is important to explore how a meta-narrative has emerged in which the health problems facing the most deprived areas – and the solution to those problems – have increasingly come to be linked to levels of NHS funding. In other words, why have we come to conceptualise public health in such narrow health terms?

One explanation, we propose, lies in the way in which commentators and stakeholders have linked patterns of funding with HI. For much of the past decade, areas with the worst health status have tended to receive less than their formula-based ‘targeted’ allocations. These large ‘Distance from Targets’ (DFTs) gave rise to a widespread perception that such areas were underfunded, attracting far more attention (and robust lobbying on the part of urban PCTs) than the fact that they were nevertheless receiving by far the largest per capita NHS allocations. The implication, apparent in several parliamentary sessions (House of Commons Health Committee, 2009, 2010; House of Commons Committee of Public Accounts, 2010; Hansard, 2011) is that HI are amenable to (indeed demand) intervention by the health sector.

In fact, as we outline below, the correlation between DFTs, deprivation and HI owed more to an NHS formula that set inappropriately high targets for deprived areas than any causal link between these factors. Nevertheless, the power of the meta-narrative was such that, when a review commissioned by the previous government confirmed suspicions that, relative to underlying health care needs, younger deprived populations had been very generously funded, while demographically older, affluent areas appeared to have been historically underfunded, a fundamental shift in target allocations was inconceivable. Instead, the new HI adjustment was used in the words of the then Minister of Health Services, to keep “the distribution of funding between the most and least deprived areas in line with the previous formula” (Hansard, 2009). The fact that, in some of England's most deprived areas, over 20% of total PCT-level allocations derived from the new HI adjustment reinforced the link between HI and the NHS. At the same time, the role played by wider policy developments (in particular growing socio-economic polarisation) in perpetuating inequalities in health has continued to receive little policy comment.

NHS resource allocation and HI

Since 1976, a core aim of NHS resource allocation has been to promote health care equity, that is, equal opportunity of access to health care for people at equal risk. In 1999, however, the Advisory Committee on Resource Allocation (ACRA), a body established to advise the Secretary of State for Health on the distribution of NHS resources across primary and secondary care, was tasked with meeting a second key objective. This was that resource allocation should also contribute to the reduction of avoidable inequalities in health. This second interpretation of equity, that all should have an ‘equal opportunity to be healthy’, is known as health equity. Notwithstanding the recent publication of a research report commissioned by ACRA on the treatment of HI in successive NHS weighted capitation formulae (Morris et al., Reference Morris, Sutton and Dixon2010), there has been little explicit debate about the tensions that arise between the two principles of health care equity and health equity or about how, in practice, these goals have proved difficult to reconcile within a single NHS resource allocation formula (Asthana and Gibson, Reference Asthana and Gibson2008).

In order to promote health care equity, the distribution of funding should reflect the cost of addressing the existing burden of disease. For most chronic conditions that place significant demands on health care resources, age is a far more significant determinant of morbidity than deprivation. Older people are also far more likely to die than younger people and, because progressive and fatal illness often requires high intensity care, this has important implications for the provision and funding of health care services. The areas that tend to have the highest crude rates of both mortality and chronic disease, and thus the highest needs for health care, are the areas with the oldest demographic profiles.

Conversely, in order to promote health equity, funding needs to be targeted so as to reduce the health gap between the most advantaged and least advantaged groups. This implies that resources should be directed at populations with the worst health in terms of age-standardised measures, that is at the most socio-economically deprived areas. The difficulty that arises for resource allocation purposes is that the geographical pattern of social deprivation in England is negatively correlated with that of age. [For instance, comparing average IMD 2007 scores for PCTs (Yorkshire and Humber Pubic Health Observatory (YHPHO), 2008) with 2007 mid-year quinary age-group population estimates of the proportion of people aged 75 years and above (ONS, 2010c) yields a correlation coefficient of −0.430]. In other words, the most deprived areas (with the worst health in standardised terms) tend to have younger demographic profiles and thus lower rates of mortality and disease in crude terms. The geographical distribution of population ‘needs’ with respect to health care equity is thus very different to the geographical distribution of needs with respect to health equity.

Until the decision, implemented in 2009, to introduce separate formulae to meet ACRA's health care equity and health equity objectives, little progress had been made in determining the proportion of the NHS budget that should be dedicated to reducing HI (Bevan, Reference Bevan2009). A small HI adjustment was made between 2001 and 2003, but this amounted to less than £150 million per annum and was concentrated on Health Authorities with the highest Years of Life Lost index values (Morris et al., Reference Morris, Sutton and Dixon2010). The idea of explicitly defining a separate HI component of the formula was then dropped between 2003–2004 and 2007–2008 as it was felt that the new Allocation of Resources to English Areas (AREA) formula [used to distribute funds for Hospital and Community Health Services (HCHS) and prescribing] had responded adequately to observed levels of unmet need in areas with high ethnic minority and/or deprived populations.

Without a designated, let alone ring-fenced, HI formula component, spending on HI during this period had to be drawn from general funding allocations. As PCTs failed to direct additional funding towards HI (House of Commons Health Committee, 2009), concerns surrounding inadequate investment in public and preventive health efforts may well be legitimate. However, the perception that areas suffering the worst HI have been generally underfunded is more problematic. The introduction of the AREA formula had actually initiated a significant shift in resources towards PCTs serving younger, more deprived populations and further widened variations in per capita funding between PCTs. However, to avoid financial destabilisation, allocations for deprived areas were only slowly brought up to their AREA targets (NAO, 2010) resulting, as noted above, in the observed correlation between deprivation, HI and DFTs. This reinforced the understanding that areas suffering the worst HI needed more NHS money if their problems were to be rectified.

Yet, analysis of the most recent NHS weighted capitation formula casts doubt on the perception that England's most deprived areas have been systematically underfunded. In 2009–2010, the CARAN (Combining Age Related and Additional Needs) report introduced a new methodology for dealing with the HCHS and prescribing components of the formula. Among other innovations, this calculates HCHS age and additional needs in a one-stage model, stratified by age (Morris et al., Reference Morris, Carr-Hill, Dixon, Law, Rice, Sutton and Vallejo-Torres2007). In this way the formula responded to concerns that, in the two-stage AREA approach, age and additional needs indices opposed each other, so that the positive effect on funding of a relatively old population was effectively cancelled out by the negative effect of affluence (Stone, Reference Stone2007; Asthana and Gibson, Reference Asthana and Gibson2008).

The new CARAN methodology, if implemented without adjustment in 2009–2010, would have resulted in a very significant redistribution of revenue income away from the most deprived PCTs. The scale of this redistribution was such that for PCTs in the four least deprived deciles there would have been a 4.6% increase relative to the previous year's AREA-based allocation (£58.58 per capita), while PCTs in the four most deprived deciles would have seen an overall fall of 8.7% in their allocation (£151.92 per capita; Asthana and Gibson, Reference Asthana and Gibson2010). This is not, of course, an entirely like-for-like comparison in that CARAN, unlike AREA, makes no allowance for unmet need. As a result, in order to meet ACRA's second objective, the NHS weighted capitation formula has since 2009–2010 needed to incorporate an explicit HI element.

Following ACRA's conclusion that it was not possible to determine on a technical basis the weight that should be applied to this element of formula (DH, 2008), it was set by Ministers at 15% of the overall allocation to PCTs. As noted above, the explicit rationale behind this particular figure was to keep the distribution of funding between the most and least deprived areas in line with the previous formula (Hansard, 2009). This desire to minimise funding volatility is understandable and not uncommon (Gibson and Asthana, 2011), but one must recognise that the choice of 15% was in order to obtain a particular outcome (sectoral funding stability) rather than to achieve a demonstrably equitable distribution of NHS resources.

By distinguishing between the resources required to meet health care needs and those targeting HI, the new formula makes it possible to identify the funding – over and above need for health care – that has been available to deprived communities and that, since 2009–2010, has been explicitly earmarked for HI. The weighted capitation mechanism is complex and disentangling the contribution made by the two formula shares is not entirely straightforward. However, the exposition spreadsheets which detail each year's settlement (DH, 2010b, 2011) define the mechanisms used and the sums distributed according to the two formula shares. On this basis, it is difficult to argue that the most deprived areas have been underfunded.

Although the HI share was set by Ministers at 15% of the overall PCT allocation, there was naturally huge variation in the degree to which individual PCTs benefited from its application. In 2009–2010, for instance, the 10 most deprived PCTs (defined, as above, using 2007 Index of Multiple Deprivation scores) received a target allocation of £1417 per capita through the application of the utilisation formula, and £365 per capita through the HI formula. In other words, 20.5% of NHS funding targeted at these areas was allocated with a view to addressing HI. This compares with £1152 and £77, respectively, across the 10 least deprived PCTs, where just 6.3% was thus directed at meeting the HI objective. The scale of this difference is illustrated by the fact that, in per capita terms, the HI allocation in the 10 most deprived PCTs amounts to 31.7% of the allocation deemed sufficient to meet the entire health care needs of people in the 10 least deprived PCTs.

The HI adjustment [renamed the Disability Free Life Expectancy (DFLE) formula] was reduced from 15% to 10% in the 2011–2012 settlement. This resulted in per capita target allocations across the 10 most deprived PCTs of £1575 and £281 through the application of the utilisation and HI/DFLE formulae, respectively – as opposed to £1237 and £65 across the 10 least deprived PCTs. The net effect was a 4.2% increase in target allocations for the most deprived PCTs, compared with a 10.6% increase in target allocations for the 10 least deprived PCTs. This reinforced a process redistributing resources away from younger, more deprived, populations and towards older, less deprived populations, which had been initiated by the introduction of the new CARAN-based formula – a shift charted by how DFTs have changed over time. DFTs capture how relatively volatile formulae-based targets relate to actual PCT allocations, which, being constrained by the DH's Pace of Change policy, evolve far more slowly. Thus, although allocations to the 10 most deprived PCTs were, overall, 2.7% below target in 2007–2008 (the last settlement under the direct aegis of the AREA-based formula), the 2009–2010 introduction of the CARAN-based formula (along with the 15% HI adjustment) meant that the 10 most deprived PCTs now received allocations that were 2.0% above target. Changing the HI adjustment from 15% to 10% continued the re-evaluation of the relative needs of these PCTs, which, in 2011–2012, were deemed to be have allocations that were now, overall, 5.2% above target. Although this serves to undermine the perception that such areas are underfunded, it does nothing to counter the broader meta-narrative, which explicitly links addressing HI with NHS activity and spending. Very large sums (£8.2 billion in 2011–2012) continue to be distributed in order to meet the HI objective.

The medicalisation of HI

In this paper, we have sought to explore how a meta-narrative has emerged that, while critical of persistent variations in health, rests on a very narrow understanding of the causes of and solutions to HI. The precise size of the HI/DFLE share of the weighted capitation formula may have originated with a desire to minimise sectoral funding volatility, but it reflects and reinforces a belief that the NHS can – and should – play a central role in promoting health equity. The outcome has been a conceptualisation of HI (and of interventions to address HI), which remains overwhelmingly focused on the individual and, in particular, on individual behaviour.

This shift from a public to a private conception of public health policy (Hunter, Reference Hunter2005) was apparent in the previous Government's White Paper, Choosing Health: making healthy choices easier (DH, 2004). With its strong emphasis on lifestyle issues and individual choice, this placed responsibility for health at the personal level and, by reducing the Government's role to that of ‘supporting people in making better choices’, showed a clear preference for consumerism over collectivism (Hunter, Reference Hunter2005). The White Paper's choice of national health inequality targets is also symptomatic with, for example, rates of adult smoking, under-18 conceptions and childhood obesity being used to assess the extent to which the underlying determinants of ill health and HI were being addressed. The use of these targets hardly suggests an understanding of the structural causes of HI. Similarly, over the past few years, social marketing has become the buzzword in public health, the assumption being that, properly (commercially) packaged, healthy lifestyles will be de rigueur for the discerning British consumer. Good health, it would seem, is no longer an outcome of access to economic, social, cultural and psychological capital. It is an individual achievement.

HI targets have been abolished by the Conservative-led Coalition Government. Unsurprisingly, their 2010 Public Health White Paper, Healthy Lives, Healthy People: Our strategy for public health in England (DH, 2010c) emphasises the role of personal responsibility and social marketing techniques in helping people to make healthier choices. More surprisingly it also accepts the limitations of simply offering people information and acknowledges the potential case for central action (e.g., by eliminating people's choice through legislation). It also acknowledges the need for upstream initiatives (i.e. those that address the context of individual behaviour), albeit primarily by proposing new ‘responsibility deals’ with industry in order to, for example, provide better information about food and promote more socially responsible retailing and consumption of alcohol.

In contrast to the 2004 White Paper, which received general acclaim from the public health community (Hunter, Reference Hunter2005), concerns have been raised about the latest White Paper proposals in leading UK medical journals (Lancet, Reference Lancet2010; Wise, Reference Wise2010). Yet, there is little sense that public health policy – and the core political values driving that policy – differ fundamentally from the approach taken by the previous Government (which also emphasised the roles of individual and corporate social responsibility). Indeed, as there are precedents from social democratic regimes such as Finland for actively involving the private sector in public health policy, coalition plans – which have been described as nothing less than a corporate takeover of public health policy (Lawrence, Reference Lawrence2010) – hardly represent a radical shift towards the political right. This raises the question of whether what lies at the heart of objections to the new White Paper is the proposal to transfer responsibility (and budgets) for public health from the NHS to LAs.

Insofar as the move of public health to LAs arguably promotes a closer relationship between public health teams and those involved in the distal determinants of health – for example, environmental health, housing and transport (Kessel and Haines, Reference Kessel and Haines2010), this is one of the more potentially radical proposals of the new White Paper. Yet, the UK Faculty of Public Health has made strong representations for the public health budget to be ring-fenced in a way that protects the existing public health specialist workforce with its remits across health protection, health improvement and health care commissioning. Resistance to transferring public health responsibilities to LAs perhaps reflects the perception that central rather than local government action is required to address the inequalities in the underlying determinants of health as well as concerns that public health will not enjoy the same level of authority in LAs as in the NHS. However, the danger of maintaining the roles of existing public health staff is that, far from becoming core business for LAs in which social, economic, environmental and health priorities are aligned, public health will continue to comprise the individualised behavioural interventions delivered by the NHS.

Although not denying the cost-effectiveness of preventive interventions such as statins, anti-hypertensive drugs and smoking cessation, it seems unlikely that a further targeting of NHS funding towards the health equity objective will have much of an impact. It is very difficult to reconcile the significant levels of NHS funding that have been targeted at HI with what the NHS can realistically do to address socio-economic gaps in health outcomes. The unequal distribution of health reflects the unequal distribution of the social and economic factors that influence health (Graham, Reference Graham2004) and, as noted above, the distribution of key determinants of health has become more unequal. Against this background, it is hardly surprising that HI have been persistent and in some cases widening. What is surprising is that this has not been attributed to the actions of a government that, in many respects, embraced a neo-liberal agenda.

This lack of social critique owes much to the extent to which HI have been medicalised. Because of a meta-narrative that links addressing HI to levels of NHS funding and to the generous settlements that have been made to the health service (particularly in the most deprived areas), economic policies that have resulted in growing socio-economic (and health) polarisation have been subject to remarkably little political opposition. At the same time, public health policy has been used to legitimise the kind of society that neo-liberal ideology has produced in the United Kingdom – one in which health (like other assets) has become a matter of individual and not collective responsibility.

This does not serve the interests of promoting greater health equity. Focusing attention (and resources) on the role that the NHS can play in reducing HI depoliticises the socio-structural determinants of health and reinforces dominant ideologies of individualism, choice and the virtues of the market. This becomes ever more urgent as the regressive effects of public spending cuts become apparent. What is instead required is a critical and socially transforming agenda that acknowledges the role of the environment and corporate capitalism in structuring health (Porter, Reference Porter2006; Renault, Reference Renault2006) and that explicitly addresses the fact that human health and disease are ‘the embodiment of the successes and failures of society as a whole’ (Mackenbach, Reference Mackenbach2009: 181).