Commentary: General or cause-specific factors in explanations of class inequalities in health

Jake M Najman

Department of Sociology, Anthropology and Archaeology, The University of Queensland, St Lucia, Queensland, Australia. E-mail: j.najman{at}mailbox.uq. edu.au

The series of papers published in this issue of the journal provide the opportunity to revisit one of the more persistent debates associated with class inequalities in health. The consistency of class-based health inequalities, over time, between countries, for most age groups and for most causes of death led some researchers to label the economically/occupationally disadvantaged as ‘generally susceptible’ to a wide variety of diseases.1 Recent conceptualizations advocate structurally similar general explanations which suggest a lack of ‘social capital’,2 ‘social cohesion’3 and the psychosocial work environment4 are causes of these health inequalities. By contrast, the papers in this issue emphasize the importance of risk behaviours but substantially extend the way we think about these risk behaviours.

Middelkoop et al.5 and Turrell and Mathers6 in their papers note that ischaemic heart disease and a variety of other circulatory diseases, lung cancer and ill-defined conditions make the major contribution to class-related health inequalities. The main (proximate) risk behaviours for these causes of death are small in number and involve cigarette consumption, poor diet, low physical activity and/or exercise and excessive alcohol consumption.7 These behaviours are unambiguously more prevalent in the lowest class/socioeconomic groups. Certainly there is a need to recognise, at a more general level of abstraction, the extent to which these behaviours co-occur in the same social class group. Why these co-occur and what might be done to change this pattern remain uncertain, as does the value of general interpretations attributing this co-occurrence to a lack of social capital/social cohesion.

Two other findings in these papers are particularly relevant to our understanding of the causes of class inequalities in health. The first of these concerns the extent to which controlling for the known higher risk behaviours in the most economically disadvantaged reduces the association between class and mortality. Martikainen et al.8 here examine the association between one indicator of class, cigarette consumption and rates of lung cancer. They find that controlling for current and past cigarette consumption reduces but does not eliminate the association between class and lung cancer mortality, suggesting other factors (e.g. diet) account for the residual association. Hart et al.9 in their paper using 20-year prospective data reinforce this finding and suggest that the residual association might be attributed to poor socioeconomic conditions through the life course, perhaps leading to generally poor lung health.

The second finding extends the search for an account of the residual associations (between class and mortality after control for known risk factors) in another direction. This involves examining the extent to which the health of a current generation may be influenced by factors in early childhood or the health of the previous generation. In this view, health is biologically embedded (say via poor diet, housing, etc.) in the fetal/ childhood period (a life-course view) or is biologically transmitted by a previous generation. Wamala, Lynch and Kaplan10 not only point to earlier and later life exposures associated with socioeconomic disadvantage, each independently contributing to heart disease morbidity, but they also note the continuing effect of biological transmission of risk via short stature, an indicator of inherited disadvantage.

Kaplan et al.,11 in their paper using data from a longitudinal Finish study, use the term ‘early life imprinting’ to describe the observed association between parental socioeconomic position and the subsequent cognitive abilities of men aged 58–64 years. Cognitive ability is associated with a number of risk behaviours. The finding that parents’ socioeconomic circumstances remain associated with these abilities after controlling the respondent's own socioeconomic circumstances suggests either that these are biologically transmitted or that they reflect embedding which occurred early in childhood.

In sum, the major risk factors for poor health outcomes are well known and are all found to be substantially more common in the most economically disadvantaged groups. Once there is statistical adjustment and control for these risk factors, the magnitude of the association between indicators of class and health diminishes, but some association generally remains. The remaining effects are generally attributed to other lifestyle/risk behaviours or to the biological transmission and/or embedding associated with the economic circumstances of the previous generation or of the early childhood experiences of the current generation. Are these findings, which extend our understanding of the importance of risk behaviours, consistent with a more general explanation of the association between class and health?

One way of reconciling the more general explanations of class inequalities in health with the observed importance of risk factors, is to acknowledge the conceptualization of proximate and distant causes of health inequalities. More general explanations (emphasizing more distant causes) will be of value if they encompass the known risk factors and point to a common response. Is a lack of social capital/cohesion causally associated with adverse risk behaviours? Until we are able to understand whether and how these more general (distant) accounts are associated with risk, general approaches to explaining class inequalities remain no more than interesting but unsubstantiated theories.

References

1 Syme SL, Berkman LF. Social class, susceptibility and sickness. Am J Epidemiol 2000;104:1–8.[ISI][Medline]

2 Kawachi I. Social capital and community effects on population and individual health. Ann NY Acad Sci 1999;896:120–30.[Abstract/Free Full Text]

3 Kawachi I, Kennedy BP. Health and social cohesion: why care about income inequality? BMJ 1997;314:1037–40.[Abstract/Free Full Text]

4 Marmot MG et al. Contributions of job control and other risk factors to social variations in coronary heart disease incidence. Lancet 1997; 350:235–39.[ISI][Medline]

5 Middelkoop BJC, Struben HWA, Burger I, Vroom-Jongerden JM. Urban cause-specific socioeconomic mortality differences. Which causes of death contribute most? Int J Epidemiol 2001;30:240–47.[Abstract/Free Full Text]

6 Turrell G, Mathers C. Socioeconomic inequalities in all-cause and specific-cause mortality in Australia: 1985–1987 and 1995–1997. Int J Epidemiol 2001;30:231–39.[Abstract/Free Full Text]

7 Suadicani P, Hein HO, Gyntelberg F. Socioeconomic status and ischaemic heart disease mortality in middle-aged men: importance of the duration of follow-up. The Copenhagen Male Study. Int J Epidemiol 2001;30:248–55.[Abstract/Free Full Text]

8 Martikainen P, Lahelma E, Ripatti S, Albanes D, Virtamo J. Educational differences in lung cancer mortality in male smokers. Int J Epidemiol 2001;30:264–67.[Abstract/Free Full Text]

9 Hart CL, Hole DJ, Gillis CR, Davey Smith G, Watt CM, Hawthorne VM. Social class differences in lung cancer mortality: risk factor explanations using two Scottish cohort studies. Int J Epidemiol 2001;30:268–74.[Abstract/Free Full Text]

10 Wamala SP, Lynch J, Kaplan GA Women's exposure to early and later life socioeconomic disadvantage and coronary heart disease. The Stockholm Female Coronary Risk Study. Int J Epidemiol 2001;30: 275–84.[Abstract/Free Full Text]

11 Kaplan GA, Turrell G, Lynch JW, Everson SA, Helkala E-A, Salonen JT. Childhood socioeconomic position and cognitive function in adulthood. Int J Epidemiol 2001;30:256–63.[Abstract/Free Full Text]