Samuel Preston and Paul Taubman

Socioeconomic Differences in Adult Mortality and Health Status

1. General Feature of the Article

An introductory review of theoretical and empirical works on the topic nicely summed up by the title - how positions in different socioeconomic status levels affect the mortality as well as health status of people. Among principal indicators of socioeconomic status, which are education, income and occupational prestige, the authors note that traditional studies paid greatest attention to education as the principal indicator in studying relationship of socioeconomic status and mortality and health levels. This is because education is least susceptible to the problem of reverse causation (health status is likely to affect income and occupational status to significant extent, less likely to affect education to the same extent), as well as being stable and measurable on an interval scale.

2. Empirical Data on Socioeconomic Differences in Mortality and Health Status

First comes the review of empirical data that indicates the recent trend of socioeconomic differences in mortality and health status. Briefly, some of the major points that should be noted are as follows. First, quite obviously, except for minor aberrations there is a consistent tendency for those with higher education to be better off in terms of mortality. Outside of that, then: (1) There is a tendency for the narrowing of the range of educational differences in mortality as age advances. (2) During the period from 1975 to 1982-84, educational differences in mortality for males widened, while those of females remained equiproportionate. (3) Data from Western nations reveal that quite similar patterns of socioeconomic differences in mortality hold for most of those nations including the U.S. (4) Heart disease is the principal form of disease that accounts for the difference in mortality depending on the educational levels. (5) The educational differences in disability and ill health are huge compared to the educational difference in mortality. However, like the mortality difference, disability and ill health differences contract as age advances.

3. Review of Two Theories on Socioeconomic Difference in Mortality and Health Status

Reviews two basic theoretical models attempting to account for this socioeconomic difference in mortality and health. They are economic models and social-psychological models. Economic models, which have better conceptual clarity, takes the form that is familiar to all other social scientific models taking after economics: that is, it views individuals as attempting to maximize an utility function, in which concerned items to be included in the function may or may not be related to health. But, anyway, the calculation of this utility function determines how "healthy" a person is going to be. The economic model clearly distinguishes between endogenous and exogenous variables that are to be input into this utility function. In fact, though the assumption that some variables can be perceived of as completely exogenous (e.g., choice to live or not to live in environmentally polluted areas, which may be said to be "exogenous" but also appears to be partially under control of individual agency) may be questionable, one of the beauty of the economic models is this clear cut distinction of endogenous and exogenous factors. On the whole it is clearer to analyze the effects of the latter variables on socioeconomic differences, and the authors give consideration to several of these variables that can be considered under the economic model. Among the variables considered are income/wages/education (straightforward, sometimes known as the "deprivation model"), prices of health-related goods and services, levels of medical knowledge and technique (patterns of differential access to different classes, as well as individual people's skills in allocating scarce resources), personal endowments, environmental factors and personal tastes (it is one of the weakness of the economic model that these factors are treated as solely exogenous).

On the other hand, social-psychological approaches pay little attention to the issue of causation and individuals are seen to be possessors of certain traits, but not pursuers of definitive goals. On the whole this approach is not as systematic, yet it does provide for the advantage of paying more attention to such factors as tastes and personality, as well as better attention to interpersonal relationships. Sorting out causal pathways in the connections among social relations, socioeconomic status and health is rather difficult, and on the whole this approach has not been so successful in solving this issue of causal direction.

4. Suggestion for Future Course of Research

Further, neither of these approaches have paid sufficient attention to the longitudinal data. The authors suggest that future studies must pay more systematic attention to how individuals' personal histories cumulate to create the socioeconomic differences in mortality and health at any given moment in time - yet the latter is all the demographers are studying currently.

5. Proximate Determinants and the Effects of Race

The authors note of the number of empirical studies that tried to identify "proximate determinants" of health - biomedical variables such as smoking that act as intervening variables between class and health status to account for the effects of the former on the latter. These have not been completely successful, and they tended to show that a large portion of the difference in health status was left unexplained by these proximate determinants. On the other hand, past researches tended to show that effects of race on the difference in health status are actually the spurious effects of class and status distinctions, such as education and income. As such, the authors suggest that formulating policies based on the idea that race may have independent effects on the difference in health status and mortality is largely useless.

6. Critique, Relevance

The three articles of this one, The Risk of Being Sick by James Riley, and The Changing Pattern of American Mortality Decline by Eileen Crimmins, make up the mortality and morbidity section of our demography reading list. As far as this article is concerned, while there is nothing surprising that there is a socioeconomic difference in human mortality and morbidity, one of the most notable insight is probably that the effects of socioeconomic difference are cumulative so that future researches need to make more effort in collecting longitudinal data. The problem with most of today's empirical research is that they are mostly based on the cross-section data. Outside of this basic insight, the empirical trends pointed out in the section 2. should probably remembered.