Last updated on April 20, 2014 at 8:28 EDT

Macrosocial Determinants, Epidemiology, and Health Policy: Should Politics and Economics Be Banned From Social Determinants of Health Research?

September 24, 2008

By Muntaner, Carles Chung, Haejoo

In “Epidemiology and the Macrosocial Determinants of Health,” Putnam and Galea highlight important new developments in health policy and social epidemiology. We aim, in this commentary, to enhance the issues raised by the authors while complementing their arguments with our analyses. We agree that social epidemiologists interested in health policy might be drawn more often to macrosocial analysis because political forces and health policies affecting the public’s health tend to begin at the national level (i). Nevertheless, between holism (the focus on macrosocial factors and outcomes) and individualism or the microsocial analysis of common social epidemiology (e.g., social capital, education, personal income, occupation), there is a third option. This third option is systemism, which allows us to integrate macro- and micro-levels when policies demand it (z). Following Bunge’s postulates (3):

1. everything is a system or a component of a system;

2. systems have system level characteristics that their components lack (e.g., income inequality is a feature of society that a person does not have);

3. all problems need to be addressed in a systematic way rather than in a fragmented way (this is why social epidemiology rarely produces explanations as it follows the fragmented risk factor approach);

4. ideas should be put into systems (theories).

Thus, individualists study only the components of social systems (low-income persons, women), ignoring their structure (e.g., class relations, patriarchy). Macrosocial epidemiology will correct this lacuna as far as it embraces macro-micro analyses. A BoudonColeman diagram (3) illustrates the advantages of systemic thinking.

The macrosociologist might find it difficult to explain how economic growth leads to stagnation in, say, a contemporary wealthy country, and the individualist does not understand from where old age pensions originate (from a government’s political economy). Only the systemic view has a complete understanding of the social mechanisms linking macro- and micro-social levels.

Putnam and Galea are correct in acknowledging that the Social Determinants of Health (SDOH) approach broke with the microsocial dominance in epidemiology – social support, anger, isolation, John Henryism, Type A, job strain, life events, sense of coherence, educational attainment, occupational prestige, “SES,” among others. We hope that the WHO Commission on SDOH (4) will establish social epidemiology as an important field to be reckoned with in the future. We note, however, that models such as the “gradient” or the “fundamental causes” propose a grand generality that is often contradicted by historical context. In specific historical periods, for example, the wealthiest have been exposed to more risk factors than the poor – smoking, animal fat, and cocaine (5).

SDOH models still bear the imprimatur of physics, while their object of study might be closer to history or early evolutionary theory (6). Sociology may involve drawing strong conclusions from observations rather than experiments, summarizing vast amounts of data into relatively simple theories, favoring explanation over prediction (6). Thus, lack of macrosocial explanations also affects SDOH research, even in its most established findings: although, for example, there are hundreds of studies on income and health, very few of them consider income as endogenous to politics, governments, class conflict, unions, or other social institutions (7). Putnam and Galea are also right on the mark when they point to the avoidance of macrosocial studies in epidemiology (8). Already in the 19705, the studies of Harvey Brenner on macroeconomics and population health were received with controversy (9). They failed to capture the interest of epidemiologists. It took two decades for a rekindling of interest for Brenner’s macroeconomic models in epidemiology (10).

To clarify our models, we can divide macrosocial determinants of health into three categories: economic, political, and cultural (including science, technology, ideology, religion, etc.) (z). Thus a tax reduction on capital gains is political, income inequality is economic, and the belief in the collective benefits of private property is cultural. All are exogenous to microsocial factors (e.g., work organization, interpersonal violence, social support). Putnam and Galea astutely signal the lack of social mechanisms that follows from the neglect of macrosocial determinants. Yet some mechanisms are ignored even at the microlevel due to their controversial (political) nature. Class exploitation can be measured at the macro- (e.g., national), meso- (organizational), and micro- social levels (n). On the other hand, some macrosocial determinants that carefully bypass any political mechanism, such as social capital, have been enthusiastically embraced (12.).

Interestingly, Putnam and Galea trace adoption of multilevel models in social epidemiology to the theoretical and ethical critique of individualism in epidemiology and public health (13). Theory and values preceded the adoption of statistical techniques. As a coauthor (C.M.) of some of the first papers using this multilevel analysis in social epidemiology (14-16), I can bear witness that the emphasis on context by Nancy Krieger (with precedents in Engels and Sydenstryker among others) provided a powerful incentive to find the adequate statistical method to test contextual and multilevel hypotheses. As Putnam and Galea illustrate, the field of “macrosocial determinants of health” needs foundational conceptual work. Concepts are often confusing as sociopolitical and socioeconomic are redundant. Every economic or political fact is surely social. Eco-social would also be redundant, unless “eco” means the physical and biological environment of a society. Concepts are inadequate structural should refer to the relations among the components of a system and not to its level. Similarly, global is too restrictive; too idiosyncratic – societal; and metaphors abound – upstream, ultimate, macro.

The authors identify four barriers to explain why macrosocial determinants have been given so little attention in epidemiology. Yet, the definition of epidemiology as an applied science that studies the distribution and determinants of disease in populations implicitly includes macrosocial determinants. Populations are systems of persons defined not only by biological relations but also by social relations – political, economic, and cultural (17). Why then do epidemiologists want to restrict the scope of their discipline? Internal factors could be the biological training of the leadership of the discipline – dentistry, medicine, pharmacy, biology – or the controversy surrounding social science, policy, and politics in general, which can threaten the scientific status and funding of a discipline. External factors would involve, for example, the institutional linkages between epidemiology and medicine and the economics of biomedical research.

The supposed lack of objectivity of macrosocial science constitutes a second objection. Such concerns apparently do not stop macroeconomists from being among the most influential applied scientists of our time. For decades social scientists have studied objectively subjective social issues such as political ideology. They have transformed these soft sciences into hard technologies on which we all depend. In fact, the contemporary accuracy of predictions of electoral behavior, for example, is impressive.

The Weberian ethos of value free social science, fully endorsed by epidemiology, has hurt epidemiology by placing too many topics off limits for fear of not being objective. In applied science (epidemiology) and technology (public health), values are explicit since the goal of these disciplines is to change the world. Just as using firearms and tobacco are surely bad for health, a political system that generates poverty can also be considered bad for health. In fact, failing to declare policy preferences implies not fulfilling the epidemiologist’s role.

Another popular barrier involves the “ecological fallacy” or Robinson bias. Geoffrey Rose has put forward the case for macrosocial studies in his classical article on the health of individuals vs. the health of populations (18). Macrosocial analysis provides insight into major determinants of population health that go beyond individual risk factors. It answers questions such as “why is the homicide rate among young men much higher in Chicago than in Wales?” Most relevant to public health, macrosocial analyses lead to interventions that also occur at the macrosocial level – health and social policies – such as restricting access to firearms or eliminating racially segregated poverty.

Epidemiology cannot afford to ignore the health and social policy consequences of the macrosocial determinants of the health approach. What is more, the inadequacy of epidemiologic methods to deal with macro-determinants can be easily overcome. Cross-fertilization among methods is common in transdisciplinary domains such as genetic epidemiology. Social epidemiology has already shown openness to methods from other disciplines: multilevel modeling was developed in sociology of education; SEM with latent variables in political science. Important methodological issues in macrosociology (19) such as the use of case studies, the small n problem, or the analysis of time series will have to be addressed in social epidemiology. CAUSE FOR OPTIMISM?

Despite resistance to macrosocial determinants in epidemiology – the notion that only policies matter, not politics (an argument that would make democracy irrelevant as well) – signs for moderate optimism persist. Recently, global health, that analyzes macrosocial factors such as financial capital flows, wars, trade agreements, or welfare regimes and their impacts on population health, has blossomed (20). Nevertheless, global health is dominated by the new charity model of large foundations (Bill and Melinda Gates foundation) and donors (Warren Buffet, Bill Gates). Their emphasis is notably technological and biomedical. Owing to the huge influence that funding from the two richest persons can yield over this discipline, this giving could deter research on macrosocial determinants, in particular those that envision a central role for government redistribution.

There has been much interest in conditional programs for the poor, such as in the “Oportunidades” program in Mexico. Owing to the means-tested nature of these programs and the involvement of the private sector in the delivery of health services, it is important to compare them to government-financed and delivered universal programs such as Venezuela’s Barrio Adentro. Again, due in part to the support of global health donors (who, unlike the WHO, only represent themselves), the former type of program is capturing most attention in global health.

There is, however, a vibrant area of macrosocial research on the relation between politics, welfare state, and population health in comparative perspective (1, 21-25). Here the challenge is that many epidemiologists are more comfortable dealing with isolated policies than inquiring about the political and economic roots (social class alliances, political parties, unions, social movements) of sets of policies (the welfare state).

We conclude that Putnam and Galea have raised important issues that should be widely discussed among epidemiologists and public health scholars. Epidemiologists, health policy and public health scholars, health economists, and sociologists interested in macrodeterminants of health have reason to be optimistic about the chances to pursue these interests in the years to come. Resistance will continue within and outside epidemiology, but the reality of global integration, economic inequality, and an unstable political economy that does not meet the needs of the majority of the persons in the planet might be too overwhelming to ignore. Social epidemiologists should welcome transdisciplinary areas of inquiry and the growth of knowledge that emerges. Ultimately the fortunes of social epidemiology will also be influenced, not surprisingly, by macrosocial events outside the academic world.


1. Navarro V, Muntaner C, Borrell C, Benach J, Quiroga A, RodriguezSanz M, et al. Politics and health outcomes. Lancet. 2.006; 368(9540):1033-7.

2. Muntaner C, Lynch JW. Income inequality, social cohesion, and class relations: a critique of Wilkinson’s neo-Durkheimian research program. Int J Health Serv. 1999;29(1):59-81.

3. Bunge M. Systemism: the alternative to individualism and holism. In: Boyd R, Florez-Malagon A, editors. Social Sciences and Transdisciplinarity. Montreal: CDAS; 1999.

4. http://www.who.int/social_determinants/, accessed 15 May 2008.

5. Benach J, Muntaner C. Aprender a Mirar la Salud. Barcelona: El Viejo Topo; 2005.

6. Lieberson S, Lynn FB. Barking up the wrong tree: scientific alternatives to the current model of sociological science. Ann Rev Social. 2002; 28:1-19.

7. Muntaner C, Oates GL, Lynch JW. Social class and social cohesion: a content validity analysis using a nonrecursive structural equation model. Ann N Y Acad Sci. 19995896:409-13.

8. Muntaner C, Chung H. Psychosocial epidemiology, social structure, and ideology. J Epidemiol Community Health. 2005;59(7):540-1.

9. Wagstaff A. Time series analysis of the relationship between unemployment and mortality: a survey of econometric critiques and replications of Brenner’s studies. Soc Sci Med. 1985;21(9):985-96.

10. Brenner MH. Unemployment, economic growth, and mortality. Lancet. 1979;1(8117):672.

11. Muntaner C, Lynch JW, Oates GL. The social class determinants of income inequality and social cohesion. Int J Health Serv. 1999;29(4): 699-732.

12. Muntaner C, Lynch JW, Davey Smith G. Social capital and the third way in public health. In: Labonte R, editor. Critical Perspectives in Public Health. Oxford: Oxford University Press; 2008.

13. Tesh S. Hidden Arguments. New Haven: Yale University Press; 1989.

14. O’Campo P, Gielen AC, Faden RR, Xue X, Kass N, Wang MC. Violence by male partners against women during the childbearing year: a contextual analysis. Am J Public Health. 1996;85(8, Part 1)11092-7.

15. Diez-Roux AV, Nieto FJ, Muntaner C, Tyroler HA, Comstock GW, Shahar E, et al. Neighborhood environments and coronary heart disease: a multilevel analysis. Am J Epidemiol. 1997;146(1):48-63.

16. Soderfeldt B, Soderfeldt M, Jones K, O’Campo P, Muntaner C, Ohlson CG, et al. Does organization matter? A multilevel analysis of the demand-control model applied to human services. Soc Sci Med. 1997;44(4)=527-34.

17. Muntaner C. Social epidemiology: no way back. A response to Zielhuis and Kiemeney. Int J Epidemiol. 2001;30:625-6.

18. Rose G. Sick individuals and sick populations. Int J Epidemiol. 2001;30(3):427-32.

19. Peters BG. Comparative Politics: Theory and Methods. New York, NY: New York University Press; 1998.

20. Chung H, Muntaner C. Welfare state typologies and global health: an emerging challenge. J Epidemiol Community Health. 2008;62:282-3.

21. Navarro V, Shi L. The political context of social inequalities and health. Int J Health Serv. 2001;31(1):1-21.

22. Navarro V, Borrell C, Benach J, Muntaner C, Quiroga A, Rodriquez-Sanz M, et al. The importance of the political and the social in explaining mortality differentials among the countries of the OCED, 1950-1998. Int J Health Serv. 2003;33(3):419-94.

23. Chung H, Muntaner C. Political and welfare state determinants of population health. Soc Sci Med. 2006;63(3):829-42.

24. Chung H, Muntaner C. Welfare state matters: a typological multilevel analysis of wealthy countries. Health Policy. 2007;80(2):328-39.

25. Espelt A, Borrell C, Rodriguez-Sanz M, Muntaner C, Pasarin MI, Benach J, et al. Inequalities in health by social class dimensions in European countries of different political traditions. Int J Epidemiol. 2008, (March 13 [Epub ahead of print]).


1 Departments of Psychiatry, Public Health Sciences, and Nursing,

University of Toronto, Toronto, ON, Canada

2 Department of Political Science, Faculty of Arts and Science, University of Toronto,

Toronto, ON, Canada

Correspondence: Carles Muntaner, Health Sciences Building, University of Toronto, 386-155,

College St, Toronto, ON, MsT iP8, Canada, E-mail: carles.muntaner@utoronto.ca

Journal of Public Health Policy (zoo8) 2.9, 199-306.



Carles Muntaner is the Chair, Psychiatry and Addictions Nursing Research in the Social Policy and Prevention Research Department at the Centre for Addictions and Mental Health (CAMH). He is also a professor at the faculty of nursing with a cross-appointment in the Department of Public Health Sciences, Faculty of Medicine, University of Toronto.

Haejoo Chung is currently CIHR IHSPR post-doctoral fellow at the University of Toronto and research scientist for the employment relations HUB (EMCONET) of the WHO commission on Social Determinants of Health.

Copyright Palgrave Macmillan Limited 2008

(c) 2008 Journal of Public Health Policy. Provided by ProQuest LLC. All rights Reserved.