Editorials

Social mobility and health: cause or effect?

Since the publication of the Black report in 1980,¹ health researchers throughout the industrialised world have given extensive attention to the issue of inequalities in health. Health inequalities are large, widespread, and remarkably persistent. Different socioeconomic indicators give roughly the same picture, showing inequalities with a variety of health measures² in both sexes and at all ages, with adolescence as a possible exception.³ The consistent and robust links between socioeconomic status and health suggest that scientists of different disciplines have a lot to explain.

Much discussion has been devoted to the relative explanatory power of two hypotheses: social causation and health selection. The social causation hypothesis maintains that health is related to socially determined structural factors such as working environment or behavioural factors such as diet. The health selection hypothesis maintains that social mobility is affected by health, and that the healthy move up the class hierarchy while the less healthy move down.

In this week's issue of the BMJ, two studies shed new light on the health selection-social causation controversy. The two studies have different methodological designs, age span, and health measures. What they share is an interest in health related social mobility--that is, selection. It may seem contradictory that one paper finds selection to be important⁴ while the other does not.⁵ However, the term "selection" is ambiguous and multidimensional. It can refer to mobility between and within generations, between social classes, and into and out of the labour market. Power et al (p 449) focus on health related class mobility between and within generations.⁵ Bartley and Owen (p 445) examine intragenerational health related mobility into and out of the labour market in different social classes,⁴ a process that in occupational medicine has been labelled the "healthy worker effect."⁶

Power et al use longitudinal follow up data to analyse the effects of health related mobility and cumulation of social circumstances among men and women aged 33 years.⁵ The authors find that health related mobility does occur but does not explain health inequalities at age 33, a paradox that is due to the small numbers affected. This finding is supported by earlier evidence.^{7 8} What does explain the inequalities is lifetime social circumstance as indexed by earlier social class recorded at several points of time.

Even the most persistent opponents of the health selection hypothesis admit that health selection occurs from time to time.⁹ However, the proponents and opponents divide when it comes to the explanatory power of this hypothesis: the proponents argue that its effect could be substantial,¹⁰ the opponents maintain that its contribution is only marginal.⁹ In light of this disagreement it is noteworthy that Power et al show that health related mobility between generations seems to have little role in producing health inequalities.⁵ One of the most distinguished defenders of the selection hypothesis, West,¹⁰ argues that health related selection is most likely to occur between childhood and early adulthood--that is, as people move from their parents' class to their own achieved class. This is exactly the life stage covered by Power et al. Another advantage of Power et al's analysis is that it accounts for both direct and indirect health selection--that is, mobility related to manifest illness as well as to latent health potential such as height. The study is limited in that it looks at only self assessed health, so that we cannot be sure that the conclusion applies to more specific health measures such as serious psychiatric disorders like schizophrenia.¹¹

Bartley and Owen analyse time series data on men aged 16-59 years derived from the general household survey for the period 1973-93.⁴ They show that men with limiting long standing illness are much less likely to be employed if they are manual workers, a tendency that increases as unemployment rises but does not diminish as unemployment falls. This suggests that manual occupations have become more "health selective" over the past 20 years. It is noteworthy that this healthy worker effect has increased over the past 20 years. Similar patterns have been found elsewhere.^{12 13} As Bartley and Owen point out, this contradicts the theory that people in poor health are more likely to enter manual occupations with low status. In addition, two important methodological implications should be mentioned. The first concerns bias in the measurement of trends in class inequalities in health. Unless people who have been employed are systematically included in the analysis according to their previous occupations, the increasing healthy worker effect will give a false picture of shrinking health inequalities over time. Secondly, comparisons of health inequalities between countries might be severely distorted if the healthy worker effect is not properly accounted for.

The social and health policy implications of the two papers are that measures of intervention and prevention should be implemented early in life in order to resist the cumulation of adverse social circumstances, if at all possible. However, Power et al's analysis remains abstract.⁵ We do not know what the specific environmental or behavioural factors are. There would be great advantage if these health factors could be identified.

Even in the highly developed and egalitarian Scandinavian welfare states, which provide relatively generous benefits to groups outside the labour market, work is by far the most important source of welfare.¹⁴ Being excluded from the labour market because of poor health will, in the long run, result in a reduction of income and standard of living, which in turn might result in even poorer health. Thus, measures aimed at improving the employment prospects of people with chronic health problems would maintain their living standards, which is an important task in itself, and in addition contribute directly to better health and wellbeing.

ESPEN DAHL Research director Institute for Applied Social Science, Pb 2947 T(empty set)yen, 0608 Oslo, Norway

Espen Dahl 

1. Townsend P, Davidson N, eds. Inequalities in health. The Black report. Harmondsworth: Penguin, 1982.
2. Dutton DB, Levine S. Socieconomic status and health: Overview, methodological critique, and reformulation. In Bunker JP, Gomby DS, Kehrer BH. Pathways to health. The role of social factors. Menlo Park, CA: Henry J Kaiser Family Foundation, 1989.
3. Macintyre S, West P. Lack of class variation in health in adolescence: an artefact of an occupational measure of social class? Soc Sci Med 1991;32:395-402.
4. Bartley M, Owen C. Relation between socioeconomic status, employment, and health during economic change, 1973-93. BMJ 1996;313:445-9. [Abstract/Free Full Text]
5. Power C, Matthews S, Manor O. Inequalities in self rated health in the 1958 birth cohort: lifetime social circumstances or social mobility? BMJ 1996;313:449-53.
6. Carpenter LM. Some observations on the healthy worker effect. Br J Ind Med 1987;44:289-91. [Medline]
7. Lundberg O. Childhood living conditions, health status, and social mobility: A contribution to the health selection debate. European Sociological Review 1991;7:147-61.
8. Power C, Manor O, Fox AJ. Health and class: the early years. London: Chapman and Hall, 1991.
9. Blane D, Smith GD, Bartley M. Social selection: what does it contribute to social class differences in health? Sociology of Health and Illness 1993;15:1-15.
10. West P. Rethinking the health selection explanation for health inequalities. Soc Sci Med 1991;32:373-84.
11. Dohrenwend BP, Itzhak L, Shrout PE, Schwartz S, Naveh G, Link BG, et al. Socioeconomic status and psychiatric disorders: the causation-selection issue. Science 1992;255:946-52. [Abstract/Free Full Text]
12. Dahl E. Social inequality in health--the role of the healthy worker effect. Soc Sci Med 1993;36:1077-86.
13. Elstad JI. Employment status and women's health--exploring the dynamics. Acta Sociologica 1995;389:231-49.
14. Esping-Andersen G. The three worlds of welfare capitalism. Cambridge: Polity Press, 1990.