Social inequality in health: Dichotomy or gradient?

Available online at www.sciencedirect.com

Health Policy 85 (2008) 71–82

Social inequality in health: Dichotomy or gradient? A comparative study of problematizations in national public health programmes夽 Signild Vallg˚arda ∗ Department of Health Services Research, Institute of Public Health, University of Copenhagen, Øster Farimagsgade 5, P.O. Box 2099, 1014 Copenhagen K, Denmark

Abstract Recent public health programmes from four countries: Denmark, England, Norway, and Sweden, are studied to analyse how social inequality in health is described, explained and suggested to be tackled, i.e., the problematization or the discursive process whereby the issue is framed and made accessible to political action. Social inequality in health is defined in these programmes both as a disadvantaged minority with major health problems, in contrast to the rest of the population, i.e., as a dichotomy; and as a gradient in which health problems are seen as increasing with lower social class or educational level. The causes of health inequality are identified as behaviour, social relations and underlying social structures. Policies aimed at reducing health inequality can be characterized as either in accordance with a residual welfare state model, targeting the disadvantaged, or a universal model, addressing the whole population. All countries have policies that are mixtures of these problematizations, but with some systematic differences between the countries. In this field England resembles the Scandinavian countries, as much as they resemble each other dispelling the idea of a Nordic or Scandinavian welfare state model. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Social inequalities; Public health; Policy; England; Scandinavia

1. Introduction Denmark: “Social inequality in health should be minimized” [1, p. 8]. 夽 The Danish, a shorter version of the Swedish, and the Norwegian programmes on health inequalities are published in English. I have translated other quotations from the Swedish and the general Norwegian programme. ∗ Tel.: +45 35 32 79 68; fax: +45 35 32 76 29. E-mail address: [email protected].

England: “Improve the health for everyone and the worst off in particular” [2, p. viii]. Norway: “A broad, long-term strategy to reduce social inequalities in health”[3, p. 5]. Sweden: “Reduce disparities in health between different population groups” [5, p. 18]. As the above quotations indicate, reduction of social inequality in health was one of the main targets of

0168-8510/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.healthpol.2007.07.004

72

S. Vallg˚arda / Health Policy 85 (2008) 71–82

public health programmes launched by the Danish, English, Norwegian and Swedish governments in the years around the millennium. Social inequality in health has entered the political agenda in several other countries in recent decades, notably the Netherlands, New Zealand [6], Finland [7] and Ireland [8]. The purpose of this article is to present a study of how social inequality in health was problematized in the four countries’ policy declarations. The process of problematization influences how policies are created and differ. A second question is if the problematization in Scandinavian countries is similar to or different from that in England, i.e., whether a Scandinavian model exists in this policy field. Several studies have been undertaken to analyze how social inequality in health has been dealt with politically. Many of them have two characteristics in common: they assume that a political consensus exists as to what social inequality is; and the development of policies in the area is assumed to follow a chain of events from documentation of a problem to political awareness and onward to concern, willingness to take action, initiatives and comprehensive coordinated policies. The policies may be in different phases, and there may be backlashes along the way, but they are not seen as following different paths [9–12]. This paper has a different approach; namely, to study the development of policies as a discursive process, in which the various ways of problematising an issue are studied, and where it is assumed that different discourses and problematizations are possible both between and within countries. Furthermore, it is presumed that different developments or paths are possible in the political process when dealing with the issue. It is an approach utilized by several social scientists [13–16]. The process of problematization is a necessary step in any political process. It is a discursive process whereby issues are framed and thereby made accessible to political action [17–19]. In the words of Michel Foucault, the “transformation of a group of obstacles and difficulties into problems to which diverse solutions will attempt to produce a response, this is what constitutes the point of problematization” [17, p. 118]. The process of problematization is a way of exercising power by setting the political agenda. G¨osta Esping-Andersen has described three welfare state typologies that have had great impact on welfare state research during the last decades. He differentiates

between the ‘liberal’ welfare state, in which beneficiaries are mainly people of low-income, the ‘corporatist’ welfare states where rights are “attached to class and status”, and, finally, there is the ‘social democratic’ regime type, which is characterized by the principle of universalism. [20] Welfare state measures may thus be characterized as either universal or targeted/residual. The Scandinavian states have been characterized as universal welfare states [21] while the characteristics of the British state are slightly more blurred. A questions addressed in the paper is whether these characteristics of the Scandinavian and British welfares states, respectively, apply to public health policies toward health inequalities. In this paper England, rather than the other UK countries, is studied because it has the most comprehensive programmes on inequalities.

2. Materials and methods The programmes studied are the most recent general public health programmes launched by the governments [1,2,5,22] and specific programmes on social inequality in health published by the English [23] and Norwegian [4,24] governments. In England, Norway and Sweden one or several green papers [25–29] preceded the general final programmes. In England, a committee was set up in 1997 to investigate social inequality in health and submit recommendations [30]. Since the publication of the general English programme in 1999, more papers about tackling inequalities have been published and in 2003 a special action programme was launched [23,31,32]. In Norway an action plan on social inequalities was published in 2005 [24] and in 2007 the government launched a national strategy to reduce health inequalities [3]. In Sweden the programme was preceded by the work of a committee which published three white papers with data and policy recommendations and it was followed by a white paper in 2006 evaluating the situation [33]. In Denmark and Sweden only general public health programmes have been published. In the Danish programme little is written about social inequalities, while the issue is widely addressed in the Swedish programme [1]. Policy programmes display how governments wish to present their concerns and intentions. Since public health programmes contain policy statements rather

S. Vallg˚arda / Health Policy 85 (2008) 71–82

than specific introductions of bills, they have an agenda-setting and symbolic or normative objective. Target groups involve members of parliament, politicians at other levels, civil servants and public health professionals, all of whom should be motivated to act accordingly, as well as the general public—via the mass media. Four countries imply four languages. I have chosen to translate: ‘ulighed’, ‘ulikhet’ and ‘oj¨amlikhet’ with the word inequality, assuming that they are roughly understood in the same way. In Norway the words “helseforskjeller” (health disparities) and in Sweden “sociala skillnader i h¨alsa” (social disparities in health) are also used as synonyms for social inequalities in health. The Scandinavian languages have no distinction between inequality and inequity, but since the English programmes do not use the term inequity, it is not a problem in this comparison. Inequality will be used throughout the paper.

3. Social inequality—a political problem? In 1978 the Alma Ata Declaration stated that health was a human right and that: “The existing gross inequality in the health status of the people particularly between developed and developing countries as well as within countries is politically, socially and economically unacceptable and is, therefore, of common concern to all countries” [34]. In 1984 the countries of the European region of the World Heath Organization (WHO) agreed on the goals of “Health for all by the year 2000” [35]. The second goal concerned the reduction of health inequalities in and between countries. Formally the four countries of this study had been committed to the issue since the late 1970s. The British government set up a committee in 1977 to address health inequalities. By the time the so-called Black report appeared (in 1980) government had changed and no immediate political action was taken [36,37]. Nevertheless, the report had a considerable impact on researchers and politicians [38–40], inside as well as outside Britain. Health inequalities gradually entered the British political agenda in the 1990s [9,41,42]. In Sweden, social inequality in health was addressed for the first time in a green paper in 1984, and it has since been mentioned in all public health statements [43,44]. In Denmark, it was not until the late 1990s that the issue

73

was given noticeable political attention [44] and in Norway it was as late as the general programme from 2003, although it was briefly mentioned in programmes from 1987 and 1993 [45,46]. Although the timing was very different, by the beginning of this century the reduction of social inequality in health had arisen as an issue in health policy. Politicians had also been aware of health inequalities before this time, e.g., high incidence and mortality of tuberculosis or infant mortality among the poor, but they had not been framed as a general health inequality problem. One might ask why the question of health inequality was addressed so late in countries where social democratic parties had substantial influence throughout the twentieth century. Equal access to healthcare had been a political issue as far back as the Second World War. Margaret Whitehead suggests that increased knowledge about inequalities was the reason why the issue entered the political scene [9]. This may have been one reason, but substantial knowledge about social inequalities in health had been available since at least the 1970s in Denmark and Sweden [47] in the form of research, reports and routinely collected data [44]. In the UK, the registrar general had published mortality statistics for different social classes since 1921 [30]. Thus, knowledge does not seem to have been the sole – or even a crucial – reason for the political attention. The heightened interest among researchers might as well have been fostered by the political attention as the reverse situation.

4. What is the problem: dichotomy or gradient? Social inequality is defined either as a disadvantaged minority with major health problems in contrast to the rest of the population, i.e., as a dichotomy or a gap, where one can say that the problem belongs to a minority; or as a gradient in which health problems increase with lower social class or educational level, and inequality therefore concerns the whole population. Some programmes use the terms gradient and gaps; the use of the term dichotomy is mine. In the English programmes and papers a change has occurred from seeing health inequalities as a gradient to seeing them as a dichotomy. In the Independent inquiry from 1998 it was stated: “The penalties of inequalities in health affect the whole social hierarchy and

74

S. Vallg˚arda / Health Policy 85 (2008) 71–82

usually increase from the top to the bottom” [30]. Similar statements can be found in the programme from 1999, Saving Lives [2]. The Cross-Cutting Review from 2002 explicitly mentions the gradient: “Health indicators show a stepwise relation to social position in a gradient” [32]. However, the action programme from 2003 in most statements defines the problem as a gap or a dichotomy. It compares the health problems of routine and manual groups or of the fifth of areas with the lowest life expectancy with “the population as a whole”. The national indicators used to monitor the success of the programme all measure dichotomies: “deprived areas or areas with poorer outcomes as compared to England as a whole”, or “manual groups with the whole population” [23, p. 65]. Hilary Graham in her analysis of the English policies states that “there is evidence that the narrow focus on the poor health of the poor groups is giving way to a broader orientation to health gradients” [8, p. 126]. As far as I read the English programmes this, however, is not the case. On the contrary, the idea of a gradient seems to be giving way to a focus on the poor and disadvantaged. The Danish programme has many similarities to the most recent English one and defines the problem as a dichotomy. The problem of social inequalities is seen as higher morbidity and mortality among the disadvantaged or the “vulnerable and distressed adults,” who “are both socially marginalized and at great risk in terms of health” [1, p. 77]. The Danish definition is special inasmuch as the group of disadvantaged is defined by both social and health characteristics. Unlike its English counterpart, the Danish programme does not mention deprived areas and communities. The social context is not included in the definition. Contrary to the focus on the disadvantaged in the Danish programme, the indicators listed at the end of the programme set out to follow development in the distribution of health and risk factors in social, occupational and educational groups; thus, assuming a gradient [1]. The Norwegian programme on social inequality in health from 2005 [24] unequivocally approaches the problem as a gradient, whereas the latest from 2007 [3,4] also mentions the health of excluded minorities. In 2005 social inequality in health was described by stating that: “the link between social position and health forms a gradient and affects all levels in society” [4, p. 8] as measured by income or educational level. The action plan from 2005 explicitly rejected the dichoto-

mous view and stated: “it does not appear to be the case that only people under the threshold of absolute poverty are less healthy due to their low social status. On the contrary, studies indicate that there is a continuous increase in health afflictions with declining socioeconomic status throughout the population” [24, p. 8]. The last strategy emphasises the gradient interpretation but also mentions groups with specific health problems such as: “prisoners, long-term recipients of social assistance, heavy drug addicts and alcoholics and some immigrant groups” [3, p. 25], included in these groups are also the Sami people and the large and increasing group of people living alone. In addition the plan mentions that other differences such as those between people of different “gender, ethnic background and place of residence often play a part in social inequalities” [3, p. 7]. While in the English programmes the problematization of social inequality changed from a gradient to primarily a dichotomy-based approach, the movement in Norway seems to have been almost the reverse [10]. In 1987 the national board of health wrote about “groups that in terms of health were exposed” [45]. The general public health programme from 2003 mentioned both the disadvantaged and the gradient. In 2005 only the gradient was considered a relevant description, while with the programme from 2007 both definitions are included with the emphasis on the gradient. The Swedish programme is the most comprehensive in its definition of health inequalities, as it includes “social class, ethnic or cultural background, sexual preferences, disability and age” and inequality between the genders. Some of these categories, such as gender, are by definition dichotomous. The description of social inequality is, apart from these exceptions, dominated by the idea of the gradient: “Differences may be seen on all educational levels” and “with increasing salary the health is gradually improved” [48, p. 26]. The interpretation of social inequality as a gradient has by and large been consistent in Swedish programmes since 1984 when it was first addressed. Beside defining social inequality as a gradient or as the problem of the disadvantaged, Hilary Graham mentions a description of it as a gap between the highest and the lowest social classes [8]. The term “health gap” is frequently used in English programmes but seems to refer to the difference between the disadvantaged and

S. Vallg˚arda / Health Policy 85 (2008) 71–82

the rest. I do not think that this third category adds to the understanding of the problematization. Either the problem is seen as primarily belonging to a minor group, the disadvantaged or the lowest social class, whether compared to the highest class or the population as a whole, or as a gradient pervading the entire population, at least in the programmes studied of the four countries. However, as Paula Braveman writes, there is the difference that the gap between the most privileged and the least will reveal greater differences than the difference between the excluded and the rest [49]. While the problematization of social inequality in Norway went from regarding social inequality in health as the problem of disadvantaged groups to a phenomenon involving the whole population, the English problematization changed the other way round: from seeing it mainly as a gradient to seeing at as dichotomy. The political path of the problematization of social inequalities in health may thus go in opposite directions. The Danes and Swedes were more consistent in adhering to dichotomy and gradient interpretation, respectively.

5. Health inequalities in the four countries The actual development and level of health inequalities in the four countries studied could influence the political problematization [50] and the extent to which these inequalities are seen as important to deal with. The size of inequalities and whether one can observe changes depends on how they are measured or problematized. Both aspects of the concept ‘social’ and ‘health’ can be defined and measured in a number of

75

different ways. ‘Social’ could be measured for example as educational level, occupational class or income. The categories are furthermore often defined differently in different countries. ‘Health’ could be measured as self-assessed health, mortality, causes of death or morbidity, etc. Different age groups could also be considered; additionally, finally, inequalities can be measured as absolute or relative differences. All these choices influence the picture given. Using relative differences Mackenbach et al. found greater differences in mortality and morbidity in Norway and Sweden than in Denmark and Great Britain [51]. V˚ager¨o and Erikson pointed at absolute differences being smaller in Norway and Sweden than in the other countries [52]. The picture of the existing social inequalities in health in a given country and when countries are compared is thus ambiguous. The same is the case when the question of changes is addressed [53,54]. What is of major importance for the way politicians perceive the issue is, of course which pictures they are presented to. In all countries there was a perception that health inequalities were persistent and even increasing, they do not, however, present comparison with other countries. To illustrate the issue I have chosen to give one example, see Table 1. According to Table 1 the countries are fairly similar, but which country could be said to be most unequal depends on the measure. Sweden has the highest relative difference and Denmark the highest absolute difference. Furthermore the Danish death rate is the highest, implying that Swedish blue-collar workers have almost as low a death rate as Danish white-collar workers. Since the countries do not seem to differ greatly concerning inequalities measured in this way, these data do not contribute to an explanation of why the

Table 1 Death per 100,000 by socioeconomic group 1990–1994 Country

Socioeconomic group

Deaths per 100,000 person years

Relative difference

Absolute difference

Denmark

Blue-collar White-collar

570 390

1.5

180

England and Wales

Blue-collar White-collar

460 300

1.5

160

Norway

Blue-collar White-collar

430 280

1.5

150

Sweden

Blue-collar White-collar

410 250

1.6

160

Men 30–59 years [59].

76

S. Vallg˚arda / Health Policy 85 (2008) 71–82

problematizations in the four countries differ as they do.

6. Causes of social inequality: structural or behavioural? Since social inequality in health is defined in different ways, that which is to be explained also differs: the situation of the disadvantaged or the gradient. When the problem of health inequalities is seen as the health problems of the disadvantaged it is relevant to see how the programmes explain the situation of the excluded, especially as regards health. Ruth Levitas [18] proposed a model to describe different discourses on exclusion in the English political debate: RED (redistributionist discourse) in which exclusion is explained by poverty; and MUD (moral underclass discourse) where the behaviour of the excluded is seen as the cause of exclusion and the behaviour is explained by moral deficits. Using Levitas’ model Sandra Carlisle suggests considering culture instead of morality: “lower socioeconomic groups share and reproduce unhealthy, underclass ‘culture”’ [16, p. 270]. Finally, Levitas suggests SID (social integrationist discourse) where lack of social integration is the cause of the problems, and unemployment is seen as decisive in this respect. All three explanations are put forward in the English action programme. Health inequalities are “a consequence of differences in opportunity, in access to services, in material resources, as well as differences in lifestyle choices of the individuals. Unfortunately, the effects can be passed on from generation to generation” [23, p. 6]. The idea that behaviour could be inherited must refer to a cultural or social inheritance. Among the explanations the behaviour of the disadvantaged is given precedence. The problems could be said to belong to the disadvantaged—not to society. In the Danish programme a variant of the moral underclass discourse [MUD] is presented. However, without a moral judgment it is rather a cultural underclass discourse. The poor health of the disadvantaged is caused by their norms and thereby their behaviours, and it is stated that “the concepts of quality of life and a good life among vulnerable and distressed groups of people may challenge the usual norms and values of society”

[1, p. 78]. One reason for this behavioural and cultural interpretation may be the predominant idea of Denmark as an egalitarian society, and therefore the interpretation is that those who do not reach this equality in health must have some personal deficits. The most recent Norwegian programme mentions as causes material conditions above all income, childhood conditions, working and housing environment, behaviour and health care [3, p. 6]. The causal path from income to health goes via the possibility of financing health promoting food, housing and leisure-time activities [p. 8], i.e., a direct influence (as in RED). However, when the situation of ethnic minority groups is explained social exclusion rather than poverty is seen as the cause. “The main social explanation seems to be unemployment on arrival in Norway, and to a lesser extent financial problems” [3, p. 27]. This interpretation is more in accordance with the social integrationist discourse, SID. Programmes in Norway and Sweden offer behaviours as an explanation of health inequalities but emphasize that they are not mainly the result of individual choices or deficits. With almost identical phrasing the two programmes state, regarding differences in health-related behaviours: “Since they follow very clear social patterns, the principle cause of the disparities is not an individual’s choice of lifestyle” [5,22, p. 35]. In the Norwegian and Swedish programmes, the poorer health of the lower social classes is also seen as a result of a combination of living conditions, occupational hazards (both physical and psychological), and poor social relations, social networks and social capital. In the Swedish programme it is stated “the lower paid have the poorest health. It should be stressed that it is not low pay that determines health, but the strong correlation between low pay and poor working conditions, restricted freedom at work, economic stress, poor housing and limited access to holiday trips, recreation, culture, etc.” [48, p. 26]. The 2005 Norwegian programme has similar explanations. The most important material factor is not considered to be deprivation but unhealthy working conditions [24]. The 2007 strategy states, as mentioned, that income and the level of consumption it enables as direct causes of health or illhealth. Sandra Carlisle [16] maintains that materialistic and psychosocial explanations of disease express two different discourses. This might be true for research

S. Vallg˚arda / Health Policy 85 (2008) 71–82

[14], but in the programmes the two are difficult to separate, particularly in the Swedish one. Finally, explanations considering underlying social structures are briefly mentioned in the Swedish and Norwegian programmes: “The structure of society can itself be the negative factor that triggers bad health” [5, p. 33]. “Some of the causes of social inequalities in health are therefore to be found in social conditions” [24, p. 15]. The Danish programme presents the least comprehensive causes of health inequalities, its focus being almost exclusively on the behaviour of the disadvantaged. The Norwegian and the Swedish explanations, meanwhile, encompass different levels of society, behaviour, social relations and social structure. This applies also to the English explanations, though to a lesser extent. Definitions and explanations are linked to responsibility. If social inequality in health is seen as a gradient caused by living conditions, it cannot be the responsibility of the individual citizen. If it is seen as pertaining to the disadvantaged or the lowest classes and caused by their behaviour, then it can be seen both as their responsibility and as that of politicians. The Norwegian and Swedish programmes more often than the others state that responsibility for reducing inequalities is political, not individual. The Norwegian government is very explicit on this point: “As long as systematic inequalities in health are due to inequalities in the way society distributes resources, then it is the community’s responsibility to take steps to make distribution fairer” [3, p. 5].

7. What is to be done: universal or residual welfare state policies? Overall, one may identify two strategies for reducing health inequalities that are correlated to the ways in which inequalities are defined. The idea that social inequality in health is the problem of a disadvantaged segment of the population is related to a strategy in which improving the lives of this group is in focus. It is a residual or liberal welfare state approach targeting the poor or excluded, which could also be named a propoor policy [55]. If the problem is seen as a gradient, the task is to level it out with initiatives that address society as a whole, an approach that could be labelled universal,

77

social-democratic or pro-equality. Another way of categorizing the initiatives presented in the programmes is to distinguish between (1) those that address behaviour, (2) those that address immediate living conditions (be they material or social), and (3) those that address the underlying social structure causing these conditions. The Danish strategy for reducing social inequalities in health is exclusively focused on the behaviour of the excluded. Social and healthcare sectors are given a central role, not least because “vulnerable and distressed adults have considerable contact with the social and healthcare services” [1, p. 71]. In general it is stated regarding public health policy that it “is important to avoid paternalism,” and that “the public sector should not govern our lives” [1, p. 58]. As regards the disadvantaged, the attitude is somewhat different, and the health and social sectors should perform tracing and outreach activities. The disadvantaged should be induced to change their behaviour, even though they may not wish to do so. “Being able to tackle this challenge is important in working with these target groups” [1, p. 77]. The Danish policy may be characterized as a residual or pro-poor policy targeting the behaviour of the excluded or marginalized. Initiatives suggested in English programmes on the basis of the Acheson report from 1998 and onwards focused on improving the situation and behaviour of disadvantaged groups and communities, i.e., a residual or pro-poor policy [14,30,56]. The 1999 programme stressed the importance of living conditions: “Tackling underlying social, economic and environmental conditions is vital. Those factors operate independently, as well as through specific lifestyle factors. So health inequality can be reduced only by giving more people better education; creating employment so that people can achieve greater prosperity; building social capital by increasing social cohesion and reducing social stress by regenerating neighbourhoods and communities; and tackling those aspects of the workplace that are damaging to health” [2, p. 81]. Thus, it is a very broad strategy addressing social conditions and the social structure. This changed in 2003 when the behaviour of the disadvantaged and the role of social and health services in helping individuals through antenatal care and smoking cessation were in focus. Studying other papers launched by the British government, Raphael and Bryant reach a similar conclusion about a change in focus from broader determinants to behavioural fac-

78

S. Vallg˚arda / Health Policy 85 (2008) 71–82

tors, although the focus is still broad compared to the USA and Canada [57]. The programme does, however, also mention the importance of “tackling the wider determinants of health inequalities, such as poverty, poor educational outcomes, worklessness, poor housing, homelessness and the problem of disadvantaged neighbourhoods” [23, p. 7]. Like the Danish solution, this policy is residual or pro-poor, albeit one that addresses many more determinants. The Norwegian action plan on social inequalities from 2005 opts for a universal policy: “research indicates that we will not address the relation between socioeconomic position and health if we base our activities on strategies that focus on ‘the poor’ as an isolated target group” [24, p. 9]. The 2003 general public health programme recommended initiatives focused on the disadvantaged groups and on the whole population and the social structures [22]. As behaviours were seen as related to social conditions, the initiatives aimed at changing them had to take that into account: “It may be difficult to change people’s behaviour if we do not succeed in improving their entire life situation” [22, p. 48]. Initiatives from the health sector were not seen as sufficient. There has thus been a shift from a partially residual policy in 2003 to a universal policy in 2005. In the 2007 programme the two approaches are combined but the universal approach dominates and it explicitly refers to the Nordic welfare state tradition: “In many cases, targeting, for example on the basis of means testing, can have a stigmatising effect and actually undermine the purpose. General welfare schemes are less stigmatising and serve to prevent people ending up in high-risk situations. In addition, social inequalities in health affect all social classes, not only the most disadvantaged. We must therefore continue to build on the Nordic tradition of general welfare schemes and at the same time implement special measures to help the people with the most problems” [3, pp. 6–7]. The universal approach is also seen as more farsighted, preventing inequalities appearing rather than remedying them when they have done so. The interventions shall deal both with social differences, i.e., by changing taxation, behaviours, health care, and prevent social exclusion. The initiatives suggested by the Swedish programme cover a number of aspects. Strong commitment to the welfare state as a means of ensuring equality is stressed:

“It is essential to adopt a broad welfare perspective on public health policy in order to reduce the disparities in health among different social groups” [5, p. 28]. Thus, it is the responsibility of politicians to achieve the overall goal and to “create social conditions that will ensure good health on equal terms for the entire population” [5]. Despite the general statements on social and material conditions, many of the Swedish initiatives concern behavioural factors and are often targeted at the disadvantaged. The healthcare sector shall undertake focused health-promoting activities towards smoking, nutrition, physical exercise, rest and sleep, because these behaviours “to a high degree contribute to inequalities in health between different groups of society” [48, p. 68]. It is stated that “we also need to give special support to certain individuals and social groups” [5, p. 33]. Thus, when the Swedish policy is specified, it mentions several targeted interventions, even though a universal policy would have been more at terms with the dominant definitions and explanations given in the programme. The picture is somewhat mixed. While Denmark and England adopt a residual or targeted policy towards social inequalities in health, Norway and Sweden combine universalism with targeting.

8. Discussion The analysis shows that social inequality in health is not simply a problem to be recognized. Rather, it is created as a problem, and the problematization process may follow different paths. Two ideal types can be distinguished. One that understands social inequality in health as a problem belonging to an excluded minority, i.e., as a dichotomy between this minority and the rest of the population. Their poor health is caused by the behaviour or specific conditions of the groups and solved by targeting this group. The other understands social inequality as a gradient affecting the whole population. It is caused by differences in living conditions as determined by underlying social structures that distribute wealth, power and opportunities unequally and where the solution is a universal approach aiming at eliminating differences in living conditions or change of social structures. What is the implication of the problematization?

S. Vallg˚arda / Health Policy 85 (2008) 71–82

79

Table 2 Political problematization of social inequalities in health

The problem Change in problem construct The causes

The solutions Addressing

Denmark

England

Norway

Sweden

Health of the disadvantaged None

Health of deprived areas and manual groups From gradient to dichotomy

Health gradient and the health of the disadvantaged None

Behaviour

Behaviour, deprivation, exclusion

Targeted Behaviour

Targeted Behaviour and living conditions

Health gradient and the health of the excluded From dichotomy to gradient and dichotomy Income, working conditions, socially determined behaviour, exclusion Universal and targeted Living conditions and behaviour

If followed consistently, the problematizations lead to fundamentally different policies and, consequently, to different effects. They influence initiatives taken and people’s perceptions of causes of inequalities and thus responsibilities for health problems. If the problem is seen as belonging to a defined minority and caused by their behaviour, the group or the individuals within that group may be held at least partly responsible. Conversely, if the problem is seen as pervading the whole society and caused by living conditions and social structures, it is a political responsibility. As has been shown in the analysis and can be seen from Table 2 none of the countries fits totally into these ideal types. However, the Danish and the 2005 Norwegian programme on social inequalities are fairly consistent with them. Whereas the Danish plan advocates a dichotomous view of inequality and suggests targeting the behaviour of the disadvantaged, the Norwegian plan views inequality as a gradient, and considers social relations and structures as causes and potential targets of the policy. The latest English programme is consequent in addressing the problem as a dichotomy or gap, and suggests interventions towards the behaviour of the excluded and improvement of their living conditions; a broader focus than the Danish programme but still within the exclusion discourse. The Swedish programme presents an understanding of the problem as a gradient caused by living conditions and social structures, but the interventions suggested also include interventions targeted at the behaviour of the disadvantaged, as does the latest Norwegian programme.

Working conditions, social structure, socially determined behaviour Universal and targeted Living conditions and behaviour

The Danish programme focuses only on changing the behaviour of the disadvantaged or the excluded while the English and the Norwegian programmes also aim at reducing the causes of exclusion and thereby preventing citizens from becoming excluded in the first place. The Danish policy may, theoretically at least, become a never-ending story. The programmes were drawn up by governments of different political colours: the English by New Labour; the Danish by a liberal-conservative government—the content was, however, not greatly different from that of the programme launched by the preceding Social Democratic government in 1999. The Swedish programme was launched by a social democratic government and the Norwegian programme by a liberal-conservative governments, except for the last one which was launched by a social democratic government. It is not possible to detect a direct relation between the content of the programmes and the political colours of the government launching the them; however, the implications of being liberal or social democratic could well differ from one country to another. The liberal Swedish government which came into office in 2006 has not yet (June 2007) launched a public health policy, but it has withdrawn [58] the last white paper from the former social democratic government [33]; a change in policy is therefore to be expected. As stated initially the countries studied have many similarities, but had the comparison included countries with greater contrasts, e.g., the USA the similarities between the four countries would probably have been more salient [57].

80

S. Vallg˚arda / Health Policy 85 (2008) 71–82

The texts analyzed in this article are all public health programmes and, as mentioned, they differ in style. Nevertheless, these programmes probably serve similar functions in their respective national political contexts and can be characterized as examples of symbolic policies that may eventually lead to political initiatives. The fact that England and Norway have published separate programmes on inequalities in health can be seen as a sign of high priority given to the issue. No separate programme has been issued in Sweden ostensibly because the issue is integrated in all aspects of the public health policy and health inequalities are mentioned throughout the programme. In Denmark the issue seems to have low priority and is mentioned only sporadically in the programme. The reason for this may be a general understanding of Denmark as an egalitarian society, making the question of health inequality seeming less relevant. Likewise, compared to Norway and Sweden, the question about gender equality has an inauspicious placing on the Danish political agenda. Another reason for the relatively limited Danish interest in the topic could be that more focus is on general life expectancy, where Denmark is faring comparatively badly in a European context, and this may overshadow the inequality issue. In contrast to the idea of a Nordic or Scandinavian model, the similarities between the Scandinavian countries in this field are small, except for the fact that social inequality has entered the political agenda. There is more variation among the Scandinavian countries than between Scandinavia and England. Universalism which is often claimed to be the hallmark of the model is not on the Danish agenda in this policy field. The reason why so many countries and political parties in each country can agree on the goal of reducing inequality in health might be that, as Hilary Graham notes in reference to Humpage, “the appeal of unifying goals like tackling health inequalities lies precisely in their capacity for alternative readings” [8, p. 116]. As has been shown, there are many alternative readings of social inequalities in health as regards what they are, what causes them, and what must be done to tackle them. Endeavours to define the issue more precisely represent a closure that could jeopardize the unanimous political support for putting social inequalities in health on the political agenda. In this article I have contributed to a problematization of problematizations as a means of clarifying

the political content of public health programmes. My classifications may, of course, be challenged, and I hope they will be. It has been done from a belief that what we say, write and label has an impact on the way we perceive society and, consequently, also on our actions, and that reflections on these problematizations are therefore important. If social inequalities are seen as a dichotomy, as the problem of a disadvantaged minority with major health problems, it might be difficult to perceive differences in health between other groups of the population. If health inequalities are seen as representing a gradient, the particular health problems of the poorest may not be as manifest. Further research is needed on the implementation of the policies to determine if these differences in the policies presented actually make a difference and if so, what difference. References [1] Government of Denmark. Healthy throughout Life—the targets and strategies for public health policy of the Government of Denmark, 2002–2010. Copenhagen: Government of Denmark; 2002. [2] Department of Health. Saving lives. Our healthier nation. London: The Stationary Office Limited; 1999. [3] Norwegian Ministry of Health and Care Services. National strategy to reduce social inequalities in health. Oslo: Norwegian Ministry of Health and Care Services; 2007. [4] Det kongelige helse-og omsorgsdepartment. Nasjonal strategi for a˚ utjevne sosiale helseforskjeller. Oslo: Det kongelige helseog omsorgsdepartment; 2007. [5] Govt.Bill 2002/03:35. The Swedish public health policy and the National Institute of Public Health. Journal Scandinavian of Public Health 2004;32(Supplement 64):3–64. [6] Mackenbach JP, Stronks K. A strategy for tackling health inequalities in the Netherlands. BMJ 2002;325:1029–32. [7] Social-och H¨alsov˚ardsministeriet. Statsr˚adets principbeslut om folkh¨alsoprogrammet H¨alsa 2015. Social-och h¨alsov˚ardsministeriet 2001;2001(5):4–35. [8] Graham H. Tackling inequalities in health in England: remedying health disadvantages, narrowing health gaps or reducing health gradients? Journal of Social Policy 2004;33(1):115–31. [9] Whitehead M. Diffusion of Ideas on Social Inequalities in Health: a European perspective. The Milbank Quarterly 1998;76(3):469–92. [10] Dahl E. Health inequalities and health policy: the Norwegian case. Norsk Epidemiologi 2002;12(1):69–75. [11] Mackenbach JP, Bakker MJ. Tackling socioeconomic inequalities in health: analysis of European experiences. The Lancet 2003;362:1409–14. [12] Mackenbach J, Bakker MJ, editors. Reducing inequalities in health. A European perspective. London: Routledge; 2002.

S. Vallg˚arda / Health Policy 85 (2008) 71–82 [13] V˚ager¨o D. Health inequalities as policy issues—reflections on ethics, policy and public health. Sociology of Health & Illness 1995;17(1):19. [14] Birch S. The 39 steps: the mystery of health inequalities in the UK. Health Economics 1999;8:301–98. [15] Graham H, Kelly MP. Health inequalities: concepts, framework and policy. London: Health Development Agency; 2004. [16] Carlisle S. Inequalities in health: contested explanations, shifting discourses and ambiguous policies. Critical Public Health 2001;11(3):267–81. [17] Foucault M. Polemics, politics, and problematizations: an interview with Michel Foucault. In: Rabinow P, editor. Ethics. Subjectivity and truth. New York: The New Press; 1994. p. 111–9. [18] Levitas R. The inclusive society? Social exclusion and new labour. Hampshire: Palgrave; 1998. [19] Rose N, Miller P. Political power beyond the State: problematics of government. British Journal of Sociology 1992;43(2):173–205. [20] Esping-Andersen G. The three worlds of welfare capitalism. Cambridge: Polity Press; 1990. [21] Christiansen NF, Petersen K, Edling N, Haave P. The Nordic model of welfare. A historical reappraisal. Copenhagen: Muesum Tusculanum Press; 2006. [22] Det kongelige helsedepartement. Resept for et sunnere Norge. Folkehelsepolitikken. Stortingsmelding nr. 16 (2002–2003). Oslo: Det kongelige helsedepartement; 2003. [23] Department of Health. Tackling health inequalities. A programme for action. London: Department of Health; 2003. [24] Directorate for Health and Social Affairs. The challenge of the gradient. Oslo: Directorate for Health and Social Affairs; 2005. [25] Nationella Folkh¨alsokommitt´en. SOU 1998:43. Hur skall Sverige m˚a batter, -f¨orsta steget mot nationella folkh¨alsom˚al. Delbet¨ankande av Nationella folkh¨alsokommitt´en. Stockholm: Fritzes; 1998. [26] Nationella Folkh¨alsokommitt´en. SOU 1999:137. H¨alsa p˚a lika villkor, andra steget mot nationella folkh¨alsom˚al. Delbet¨ankande av Nationella folkh¨alsokommitt´en. Stockholm: Faktainfo; 1999. [27] Nationella Folkh¨alsokommitt´en. SOU 2000:91. H¨alsa p˚a lika villkor-nationella m˚al f¨or folkh¨alsan. Slutbet¨ankande av nationella folkh¨alsokommitt´en. Stockholm: Fritzes; 2000. [28] Sosial-og helsedirektoratet. Der er bruk for alle. Norges Offentlige Utredninger 1998:18. Oslo: Statens forvaltningstjeneste, Staten Trykningstjeneste; 1998. [29] Secretary of State for Health. Our Healthier Nation A Contract for Health. Green paper by Secretary of State for Health. London: The Stationary Office; 1998. [30] Acheson D. et al. Independent inquiry into inequalities in health. The Acheson report. London: 1998. [31] Department of Health. Reducing health inequalities: an action report. Our healthier nation. London: Department of Health; 1999. [32] Department of Health. Tackling health inequalities. 2002 Crosscutting review. London: Department of Health; 2002.

81

[33] Regeringen. Folkh¨alsopolitik f¨or j¨amlikhet i h¨alsa och h˚allbar tillv¨axt. Regeringens skrivelse 2005/06:205. Stockholm: Regeringen; 2006. [34] WHO. Declaration of Alma-Ata International Conference on Primary Health Care, Alma-Ata, USSR, 6–12 September 1978. http://www.who.int/hpr/NPH/docs/declaration almaata.pdf. [35] WHO. Health for all by the year 2000; 1984. [36] Townsend P, Davidson N. Inequalities in health. Middlesex: Penguin Books; 1982. [37] Berridge V, Blume S. Poor Health. Social inequality before and after the black report. London: Frank Cass Publishers; 2003. [38] Macintyre S. Before and after the Black Report: Four fallacies. In: Berridge V, Blume S, editors. Poor health. London: Frank Cass; 2003. p. 198–219. [39] Macintyre S. The Black Report and Beyond what are the Issues. Soc Sci Med 1997;44(6):723–45. [40] Marmot M. From Black til Acheson: two decades of concern with inequalities in health. A celebration of the 90th birthday of Professor John Morris. International Journal of Epidemiology 2001;30:1165–71. [41] Berridge V, Blume S. Poor health. Social Inequality before and after the Black Report. London: Frank Cass; 2003. [42] Oliver A, Nutbeam D. Addressing health inequalities in the United Kingdom: a case study. Journal of Public Health Medicine 2003;25(4):281–7. [43] Socialstyrelsen. SOU 1984:39. H¨also-och sjukv˚ard inf¨or 90talet. HS90. Huvudrapport. Stockholm: 1984. [44] Vallg˚arda S. Folkesundhed som politik. Danmark og Sverige ˚ 1930 til i dag. Arhus: Aarhus Universitetsforlag; 2003. [45] Helsedirektoratet. Helse for alle i Norge? Oslo: Helsedirektoratet; 1987. [46] Sosialdepartementet. Utfordringer i helsefremmende og forebyggende arbeid. Stortingsmelding nr. 37, 1992–93. Oslo: Sosialdepartementet; 1993. [47] Vallg˚arda S. Health inequalities—political problematisations in Denmark and Sweden. Critical Public Health 2007;17(1):45–56. [48] Regeringen. M˚al f¨or folkh¨alsan. Regeringens proposition 2002/03:35. Stockholm: Socialdepartementet; 2002. [49] Braveman P. Health disparities and health equity: concepts and measurements. Annual Review Public Health 2006;27:167–94. [50] Vallg˚arda S. Hvad er social ulighed i sundhed? Om kategoriseringers betydning. In: Vallg˚arda S, Koch L, editors. Forskel og lighed i sundhed og sygdom. København: Munksgaard; 2006. p. 15–28. [51] Mackenbach JP, Kunst AE, Cavelaars AEJ, Groenhof F, Geurts JJM. Socioeconomic inequalities in morbidity and mortality in western Europe. The Lancet 1997;349:1655–9. [52] V˚ager¨o D, Erikson R. Socioeconomic inequalities in morbidity and mortality in western Europe. The Lancet 1997;350:516. [53] Kunst AE, Bos V, Lahelma E, Bartley M, Lissau I, Regidor E, et al. Trends in socioeconomic inequalities in self-assessed health in 10 European countries. International Journal of Epidemiology 2005;34:295–305. [54] Mackenbach JP, Bos V, Andersen O, Cardano M, Costa G, Hardin S, et al. Widening socioeconomic inequalities in mor-

82

S. Vallg˚arda / Health Policy 85 (2008) 71–82

tality in six Western European countries. International Journal of Epidemiology 2003;32:830–7. [55] Vega J, Irwin A. Tackling health inequalities: new approaches in public policy. Bulletin of the World Health Organisation 2004;82(7):482. [56] Davey Smith G, Ebrahim S, Frankel S. How policy informs the evidence—‘evidence-based’ thinking can lead to debased policy making. In: Davey Smith G, editor. Health inequalities: lifecourse approaches. Bristol: The Policy Press; 2003. p. 507–11.

[57] Raphael D, Bryant T. The state’s role in promoting population health: Public health concerns in Canada, USA, UK, and Sweden. Health Policy 2006;78:39–55. ˚ [58] Regeringen. Aterkallelse av skrivelse 2005/06: 205 Folkh¨alsopolitik f¨or j¨amlikhet i h¨alsa och h˚allbar tillv¨axt. Stockholm: Regeringen; 2006. [59] Bostr¨om G, Ros´en M. Measuring social inequalities in health—politics or science? Scandinavian Journal of Public Health 2003;31:211–5.