Relationship between social class and health

Prior to 1980, it was generally assumed that Britain was turning into an egalitarian society (Wilkinson, 1986:1). With the introduction of the NHS in 1948 and the increase in protective and regulatory legislations (Wilkinson, 1986), the predominant impression was that class divisions and socioeconomic inequalities were no longer of significance (Bartley, 2004). However, it became clear in 1980 with the publishing of the Black Report, which dealt with class [1] differences in health, that this was not the case. The report revealed very large differences in death rates between social classes and although general health was improving, these differences were not declining (Wilkinson, 1986; Wadsworth, 1997) (Appendix 1). The majority of scientific evidence supports this socioeconomic explanation of the health inequalities (Acheson, 1998; Marmot, 2005), and even though such inequalities of the British population have been recorded since the mid-nineteenth century (Chadwick, 1842), disagreement continues to prevail over the causes of this phenomenon (Blane, 1985). This essay argues that Black’s materialist theory offers the best explanation, and will critically analyse the alternative theories of key thinkers on this topic to further strengthen this argument. After discussing Black’s materialist theory, his other suggestions of artefact, theories of natural selection and behavioural/cultural theory (Black, 1980) will be analysed. The remainder of the paper will critically discuss Wilkinson’s psychosocial theory followed by Barker’s biological programming and Wadsworth and Kuh’s arguments for the life-course approach.

In the Black report, materialist, or structural, reasoning was seen as the most important factor, and many authors have since reinforced this argument (Marmot, 2005; Shaw, 1999; Blane et.al, 1997). The Black report described materialist as;

those explanations emphasizing hazards inherent in society, to which some people have no choice but to be exposed given the present distribution of income and opportunity (Black, 1980:_).

Most studies of geographical areas illustrate poor health and high mortality characterised by poverty (Gorey and Vena, 1995 quoted in Bartley, 2004), unemployment (Sloggett and Joshi quoted in Bartley, 2004) or pollution (Mackenbach, Looman and Kunst quoted in Bartley, 2004). As an example, people who work in occupations exposed to accident hazards and to dangerous substances, as well as extremes of temperature, also tend to be poorly paid (Bartley, 2004:96).

Vagero and Illsely (1995) claim that Black’s favoured explanation is ambiguous, arguing that there is no apparent idea about how poverty leads to disease in modern society, however the results of low or no income are rather self explanatory. Individuals with higher incomes can afford better housing, reside in safer environments and have better access to healthcare (Grundy and Holt, 2001) and “no-one would seriously argue with the contention that sufficient, safe, nutritious food is an essential ingredient for good health”(Tansey and Worsley quoted in Baggott, 2000). Roberts et.al (1993) argue that the twelve-fold class difference in child mortality down to accidental falls can logically be blamed on the inadequate interior and unsafe exterior space in which the children of lower social class families are forced to play. Additionally, many studies, such as that by Carstairs and Morris (1989) (Appendix 2) show the strong correlations between material deprivation and both mortality and morbidity (Davey Smith et.al., 1990:376). Adding to the materialist argument is Tudor Hart’s inverse care law (1971). Tudor Hart (1971:_) claims that “no market will ever shift corporate investment from where it is most profitable to where it is most needed”. In every society where market forces determine who gets what in health care, inequalities are a big feature of the system. However this argument is no longer as relevant in Britain today due to the NHS (_, 2002) however Watt (2002) claims that the inverse care law continues to remain true even with the NHS, as deprived areas lose out in NHS resource distribution.

In contrast to the materialist theory, some question whether socioeconomic health inequalities exist at all (Mackenbach and Kunst, 1997, Bloor et.al, 1987). Black (1980) suggested that;

Both health and class are artefacts of the measurement process and it is

implied that their observed relationship may itself be an artefact of little

causal significance (Black, 1980:154).

This means that mortality discrepancies are down to numerator-denominator bias occurring because class may be allocated differently on the death certificate, numerator, than at the census, denominator (Davey Smith et.al 1990:375). However Black (1980) concluded that should this possibility have truth, it would not play a significant factor. On the other hand Bloor et.al (1987) argue that the role of artefact explanation is bigger, more pervasive, and more complex than Black implied. However this debate has been overcome in a study by the Office of Population Censuses and Surveys. Using social class as assigned at the 1971 census to categorise individuals at death, it was found that eliminating numerator-denominator bias in this way had no effect on the mortality differentials (Vagero and Illsely, 1995:220). Mackenbach and Kunst (1997:767) suggest that the increase in poor health amongst lower social classes could have instead come from an increasing tendency to complain or the criteria against which one’s own health is evaluated could have changed. They except that this suggestion is merely speculative, but attempt to emphasize the need to look at objective health indicators. However it is “almost universally agreed in the academic literature that social class differences in health are real, a property of social relations in all societies, and not the by-product of measurement errors or errors of definition” (Vagero and Illsely, 1995:220). As Bloor et.al. (1987) point out, the measurement process may be hiding as well as creating health inequalities, which does not reject, but rather emphasises, the importance of measurement. “Measurement problems may affect the size and pattern of differences, but do not cast doubt on their existence” (Vagero and Illsely, 1995:220).

The Black report also presented Health selection as a potential explanation for socioeconomic health differences. According to this suggestion, those with initially poorer health may be restricted in improving their social position and are likely to have lower paid and unskilled jobs (health-related social mobility). In this way, socioeconomic inequalities in health may to some extent occur due to the selection of comparatively unhealthy people into lower social classes (the health selection hypothesis) opposed to the result of social position on health (the social causation hypothesis)(Acheson, 1998). According to West (1991:373) this hypothesis is generally regarded as having a very insignificant role in the creation of inequalities, and a “theoretical debt to social Darwinism”. Illsley (1983), Stern (1983) and West (2002) argue that the Black Report gives insufficient attention to the health selection explanation and that inter- and intra-generational mobility and assortative mating appear to be imperative factors in the persistence of socioeconomic health inequalities (Bloor et.al, 1987). However, the data on health selection is conflicting, somewhat due to a need for a distinction between health-related social mobility and how it explains social gradients in health (the health selection hypothesis). Although there is evidence for health-related social mobility (Rodgers & Mann, 1993), this may not necessarily effect or create socioeconomic health differences (Davey Smith et.al. 1990:373). Additionally, the examination of social class data from the 1971 and 1981 census has shown that downward social mobility does not account for the mortality differentials (Goldblatt, 1988, Goldblatt, 1989). As a result, Blane et.al (2008:11) consider the idea of health selection to be the least plausible explanation for social class health inequalities.

The last of the theories presented by Black is behavioural or cultural reasoning. This explanation is preferred by the medical profession and health experts, and often suggests that such behaviours are basically under individual control. From this point of view class can affect health by conditioning behaviours which harm or encourage health (Blane et.al 1997). The assumption is that lower social classes are more likely to consume harmful commodities (refined foods, tobacco, alcohol), perform less leisure-time exercise and poorly use preventive health care (vaccination, ante-natal surveillance, contraception) (Davey Smith et.al. 1990, Grundy and Holt, 2001). There is a formidable bulk of evidence which connects these behaviours to causes of death such as coronary heart disease, lung cancer and chronic bronchitis (___). Appendix 3 shows an example of the extreme class differences in such behaviours (Simpson _).

According to Blane (1985:436), though material factors are accepted as part of a ‘multi-factorial aetiology’, behavioural factors are thought to make the larger contribution. Air pollution and occupational dust exposure, as illustration, can be part of the aetiology of chronic bronchitis, but this illness is considered largely to be due to cigarette smoking. Material factors are therefore considered of secondary importance. This assumption however may be misleading as official mortality statistics understate the health effects of occupational hazards, since material factors of likely aetiological significance have been ignored, and because their combined effect is barely researched (Blane 1985:439).

Blane et.al (1997:385) suggest that “research into health inequalities has been biased towards behavioural explanations and that materialist explanations have been relatively neglected”. If this is the case it could lead to ineffective policy interventions and to holding lower classes responsible for proceedings which are outside their control. Additionally, the ‘behaviours’ being referred to are stereotypical of what one would assume from lower classes. The problem is that many of these do not always hold true. For example, it is assumed that there are higher rates of smoking amongst lower social classes however this has only been the case since mid 20th century (Simpson, 1987). Additionally, the social class differences in fat consumption are very small, and although the manual class may perform less exercise in leisure time they exert more energy at work (Davey Smith et.al. 1990:375). However Tudor Hart’s inverse care law, as mentioned earlier, explains how lower class behaviours affect the treatment of disease, rather than the creation. Cartwright and O’Brien (1978) found that middleclass people ask more questions to their doctors with an average consultation of 6.2 minutes opposed to 4.2 minutes for the working class, meaning the working class benefit less. He also claims that doctors struggle to relate to the working class, leaving their appointments being awkward and the problem not fully discussed (Baggott, _).

Another criticism of the behaviour argument accepts the aetiological significance but discards the autonomous role it assigns to them (__). They see behaviour as conditioned by the material context in which it occurs, which means that disease producing behaviours are “relegated from the status of independent variables to that of intervening variables between social structure and disease” (Blane et.al., 1997:_). The Black Report demonstrates ways in which behaviour may be embedded in material conditions. Whilst discussing the less frequent use of contraception in lower social classes, he asks:

Is it lack of knowledge, outmoded ideas, or lack of access to the means of

contraception – or is it due to an underdeveloped sense of personal control

or self-mastery in the material world? It can certainly be argued that what

is often taken for cultural variation in cognition and behaviour is merely a

superficial overlay for differing group capacities of self-control or mastery which

are themselves a reflection of material security and advantage(Black, 1980:169).

Vagero and Illsely (1995) call this distinction between materialist and behavioural ‘obscure’. Strong (1990) likened this to “the Marxist distinction between the ‘economic base’ and the ‘ideological superstructure’”. In that hypothesis, the first ultimately determines the second. This also appears to be the case here, as much behaviour does in fact depend on materialist matters. Another example is diet; “the effect of low income renders it meaningless to consider diet a matter solely of choice” (Davey-Smith et.al, 1990). However, the distinction between behavioural and materialist is significant for scientific and policy reasons. Merging the two explanations discounts the influence of the material environment which is not mediated through behaviour (Blane et.al, 1997).

This interrelationship suggests that “part of the differential distribution of behaviour is actually caused by the skewed distribution of living conditions” (_). As a result, the independent contribution of behaviour can only be measured after controlling structural conditions. Such a study by Stronks et.al (1996) found that contribution of materialist conditions was much greater (Appendix 4-9). Additionally, they proved that if the overlap between behaviour and material conditions had been ignored, as it generally had been before their study, the role of behaviour would have been overestimated (Stronks et.al, 1996:667). Therefore given this suggestion that behaviour is to an extent embedded in the environment, material conditions can either effect health directly or indirectly through behaviour (Stronks et.al., 1996). This suggestion can be schematised as shown in appendix 10.

The psychosocial environment interpretation suggests that income inequality has a considerable effect on ones psychological health (Lynch et.al, 2000). When referring to deprivation as relative, not absolute, Hasan viewed;

the aetiological pathway as psychological, related to dissatisfaction about unmet social needs, perhaps inspired by the Townsend definition of poverty as the lack of resources to obtain the living conditions and amenities which are customary or widely encouraged (1989:384).

Such needs, Hasan suggested, were historically first formed and satisfied in the way-of-life of the highest social class (Hasan, 1989:384). Lower classes would endure the frustration of not meeting new needs immediately. As a result the class gradient in health will be eternally reproduced, no matter the level of wealth. This idea contradicts the traditional and powerfully held sociological theory that the level of expectation adjusts to fit ones social circumstances (Siegel in _, 1957). Hasan (1989) assumes that this possibly progressively changed with the beginning of mass television and instant world communication. Wilkinson (1992:_) builds upon this and suggests a “psychological mechanism, based on stress, social support, and self-esteem, and their relationship to the income distribution”. Wilkinson (_) speculates that income inequalities affect health through perceptions of position in the social hierarchy which are based on relative placing according to income. Such perceptions create negative emotions such as humiliation and doubt which are translated ‘inside’ oneself to poorer health via psycho-neuro-endocrine mechanisms and anxiety induced behaviours such as drinking alcohol. Simultaneously, these negative emotions are translated ‘outside’ oneself into antisocial behaviour, reduced civic involvement, and less unity within a community. Therefore perceptions of relative income link individual and social pathology. Lynch et.al (2000) accept that income inequality may produce negative psychosocial consequences, but also argue that interpretation of links connecting income inequality and health have to start with the structural reasons for inequalities, and cannot not simply focus on perceptions of the inequality. The psychosocial interpretation conflates the structural sources with the subjective consequences of inequality and reinforces the “impression that the impact of psychosocial factors on health can be understood without reference to the material conditions that structure day to day experience” (__). The structural, political-economic processes which create inequalities exist prior to their effects being experienced at the individual level (__).

According to Davey-Smith (_) the life-course perspective presents a way of moving beyond generalisations about health inequalities. The life-course approach provides evidence that the biological and social beginnings of life are important in determining the child’s potential for adult health. “Biological programming may set the operational parameters for certain organs and processes. Social factors in childhood influence the processes of biological development, and are the beginnings of socially determined pathways to health in adult life”. Life history studies show the vital factors related with the development of these pathways, and the life phases at which intervention to lessen adult health inequalities could be most effective (Wadsworth, 1997). Mackenbach and Howden-Chapman (2003) state that health inequalities begin in the womb, whilst ___ claims that the ages of _ to _ is the most important time to condition positive behaviours. Sweeting and West (1995) suggest that family life in adolescence could have more direct effects on wellbeing than structural factors and may be circuitously linked to health inequalities in adulthood, through social mobility. However, according to Hardy et.al (2004:64) there does not appear to be a phase in life which has particular priority of health status. Each stage appears capable of adding its own defence or disadvantage (Blane quoted in Marmot and Wilkinson, 2003). However it appears that to completely ignore the life-course approach would be unwise. Without a good education, one may not be able to secure sufficient material conditions (_). Without positive behaviours being reinforced from a young age, one may develop negative health behaviours (_). By looking at the observed relationship between life-course with behaviours and material factors, and the relationship already identified between behavioural and material factors, the ‘healthy mix’ model has been created (appendix 11). The direction of the arrow illustrates the offering which can be made. It shows that materialist matters still have the most importance, as they can provide optimal behaviours, a good life-course and, above all, choice, where a positive upbringing and good behaviours can have its limitations. Taking materialist matters out of the mix could be detrimental to health, as the good life-course reinforcing the positive behaviours to want to buy a safe house and fruit and vegetables is not enough. Material factors offer that choice. Taking away the positive behaviours can lead to poor health as having the money to buy fruit and vegetables and the best health care does not mean one will. Taking away a good life-course may lead to psychological health issues. This illustrates, as this essay does, that life-course and behaviours can have some effect on health, but material factors have the most importance when determining ones health.

Since the emergence of the 1980 Black report, an abundance of studies have attempted to contribute to a broader understanding of socioeconomic health inequalities. After analysing the alternative options, the materialist explanation continues to offer the greatest contribution when explaining socioeconomic health inequalities. Artefact and health-selection are widely agreed to offer little to no significance. The behavioural theory offered the greatest rivalry for this explanation, however it was discovered that the distribution of income sets the parameters within which choices are made. However both behaviour and life-course theories can also play a significant role at the same time. From this observation a model has been derived. The model aims to explain the relationship between life-course, behaviour and materialist matters in a clear manner. The model further emphasises the significance of material factors, and therefore tackling these inequalities should be the aim of health policies.

Appendix 1: (Blane et.al, 1997)

Appendix 2: (Carstairs,_)

Appendix 3: (Simpson,_)

Appendix 4: (Stronks,_)

Appendix 5: (Stronks, )

Appendix 6: (Stronks,_)

Appendix 7: (Stronks,_)

Appendix 8: (Stronks,_)

Appendix 9: (Stronks,_)