Published by Oxford University Press on behalf of the International Epidemiological Association
Book Review |
The Political and Social Contexts of Health. Navarro V (ed). New York: Baywood, 2004, pp. 246, $40.00 (PB), ISBN: 0-89503-299-6
Vicente Navarro is one the world's leading authorities on how political economy shapes population health. As the editor of this ambitious volume, he joins the list of distinguished scholars to engage in a series of international and country-based studies of how different political contexts and social policies have shaped public health in Europe. The studies described in this book arise out of a collaborative research group—the International Network on Social Inequalities in Health (INIH), whose goal was to produce reports that were ‘... easily read and understood, that can contribute to the design of specific public interventions to improve the health of populations’ (p. 2).
The book contains 241 pages comprised of an introduction, followed by an international comparative analysis of OECD countries from 1950 to 1998, and then five chapters of country-specific case studies from the UK, Italy, Germany, Spain, and Sweden, and a final chapter of summary and conclusions. The introductory chapter lays out the conceptual model developed by the INIH group that guides the empirical analyses. Essentially, the model suggests that power relations, in the form of electoral behaviour (e.g. voter participation, time in government of different types of political parties) and trade union activity (e.g. density of unions) are the prime movers behind how different societies structure the labour market and social welfare systems for their populations. Labour market structure and welfare provisions in turn determine the extent of social inequality, indexed in this case by income inequality. The extent of social inequality then determines the health of the population. In the empirical analyses presented, population health is measured by the infant mortality rate (IMR) and life expectancy (LE), both considered overall indicators of population health. The great strength of the approach applied in this volume is that the same conceptual model and the attendant variables that correspond to the model inform every empirical analysis.
There is a large amount of data presented in this book that will be of interest to those concerned with associations between social variables and IMR and LE. For instance, in the cross-country comparisons of 17 OECD countries from 1950 to 1998 the authors present both bivariate descriptive associations among a large number of variables, and also model them in a pooled cross-sectional analysis intended to show which factors are the most important in determining levels of population health. The conclusion of this analysis is that lower income inequality, larger proportion voting for left-wing political parties, greater public health expenditures, and larger labour market participation overall and of women in particular, the lower the IMR. However, there was little effect of these factors on LE—a result that was not discussed by the authors.
On one hand it is important to have evidence that aspects of progressive social policy are linked to lower infant mortality. On the other hand, results like these also raise questions for those interested in understanding how broadly defined political and social contexts affect population health.1 Why are upstream social factors like voting behaviour or income inequality associated with IMR but not LE? What are the mechanisms through which broad social policy may affect some health outcomes and not others? And are cross-sectional analyses even appropriate—i.e. would we expect an instantaneous effect of social policy on population health indicators or should time lags be considered? For instance, even though there is an apparently strong cross-sectional correlation between income inequality and health in the 1990s among US states, long-term trends in income inequality do not match trends in most indicators of population health.2 Nor is there even a consistent cross-sectional association among US states for each decade from 1949 to 1999.3 At the US national level, over the last 30 years, all-cause mortality, life expectancy and infant mortality have declined for both blacks and whites, at the same time that trends in voter participation were declining and income inequality increasing.1 A naïve interpretation of this would be that higher income inequality and lower voter participation are both associated with improvements in population health.
Results of longer-term analyses like the ones mentioned above, present challenges to overly simplistic notions of how broad social factors drive levels of population health.4 Evidence gained from individual and aggregate cross-sectional studies needs to be examined over time at the population level. The most important determinants of different components of population health should move in some temporally consistent fashion with the changing levels of health in the population, as for instance, do lagged trends in smoking and lung cancer. Of course, it is easier to track the concordance between changes in proximal risk factors and changes in outcomes, when a small number of risk factors account for a large number of cases.5 The challenge to better understand how social factors affect population health can be informed by documenting trends in the outcome of interest, decomposing those trends into age and relatively homogeneous component causes and documenting what the main proximal risk factors are for those components. Then it may be possible to think more clearly about how so-called ‘upstream’ social factors could be linked and de-linked over time to the main proximal risk factors which determine the bulk of the cases and hence the levels and trends in the population.2 This approach to ‘triangulating’ individual level knowledge of risk factor–disease associations with population level data is important to better understanding the social determinants of population health.6,7
The points made above are certainly not a criticism of The Political and Social Contexts of Health, rather they are an extension that may point towards a way in which to build upon the foundation laid down by Navarro and his colleagues. This book is important because it raises the very basic issue of how political context and social policy affect population health and the words offered by Professor Richard Wilkinson on the back cover of the book sum it up nicely. ‘At last we have a book that asks all the right questions. For anyone with a serious interest in health policy, this is where to start.’
References
1 Lynch J, Davey Smith G. Rates and states: reflections on the health of nations. Int J Epidemiol 2003; 32:663–70.
2 Lynch J, Davey Smith G, Harper S, Hillemeier M. Is income inequality a determinant of population health? Part 2. U.S. national and regional trends in income inequality and age- and cause-specific mortality. Milbank Q 2004; 82:355–400.[CrossRef][Web of Science][Medline]
3 Lynch JW, Harper S, Kaplan GA, Davey Smith G. Income inequality and mortality across US States, and the importance of time period and source of income data: Cross-sectional study. Am J Public Health (in press).
4 Kunitz S, Engerman S. The ranks of death: secular trends in income and mortality. Health Transit Rev 1992;2(Suppl.):29–46.
5 Lynch JW, Davey Smith G. A life course approach to chronic disease epidemiology. Annu Rev Public Health 2005; 26: (in press).
6 Leon DA, Chenet L, Shkolnikov VM et al. Huge variation in Russian mortality rates 1984–94: artefact, alcohol, or what? Lancet 1997; 350:383–88.[CrossRef][Web of Science][Medline]
7 Leon DA. Common threads: underlying components of inequalities in mortality between and within countries. In: Leon DA, Walt G (eds). Poverty, Inequality and Health. Oxford: Oxford University Press, 2001, pp. 58–87.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||