Glossary a Glossary for Social Epidemiology

J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from J Epidemiol Community Health 2001;55:693–700 693

Glossary

A glossary for social epidemiology

N Krieger

Why “social epidemiology”? Is not all epidemi- Whether these biological expressions of social ology, after all, “social” epidemiology? In so far inequality are interpreted as expressions of as people are simultaneously social and biologi- innate versus imposed, or individual versus cal organisms, is any biological process ever societal, characteristics in part is shaped by the expressed devoid of social context?—or any very social inequalities patterning population social process ever unmediated by the corporal health.16 The construct of “biological expres- reality of our profoundly generative and mortal sions of social inequality” thus stands in bodies?12 Yet, despite the seeming truism that contrast with biologically deterministic formu- social as well as biological processes inherently lations that cast biological processes and traits shape population health—a truism recognised tautologically invoked to define membership in even in the founding days of epidemiology as a subordinate versus dominant groups (for scientific discipline in the early 19th century— example, skin colour or biological sex) as not all epidemiology is “social epidemiol- explanations for social inequalities in health. ogy”.34 Instead, “social epidemiology” (which first attained its name as such in English in 195035) is distinguished by its insistence on explicitly investigating social determinants of Discrimination population distributions of health, disease, and Discrimination refers to “the process by which a wellbeing, rather than treating such determi- member, or members, of a socially defined nants as mere background to biomedical group is, or are, treated diVerently (especially phenomena. Tackling this task requires atten- unfairly) because of his/her/their membership tion to theories, concepts, and methods of that group”(page 169).9 This unfair treat- conducive to illuminating intimate links be- ment arises from “socially derived beliefs each tween our bodies and the body politic; toward [group] holds about the other” and “patterns this end, the glossary below provides a selection of critical terms for the field. of dominance and oppression, viewed as expressions of a struggle for power and One brief note of explanation. Some entries 10 contain only one term; others include several privilege” (pages 125–6). related terms whose meanings are interde- People and institutions who discriminate http://jech.bmj.com/ pendent or refer to specific aspects of a broader adversely accordingly restrict, by judgement and action, the lives of those against whom they construct. Additionally, each entry is cast in 6 relation to its significance to social epidemiol- discriminate. At issue are practices of domi- ogy; explication of salience to other disciplines nant groups—both institutionally and is beyond the scope of this particular glossary. interpersonally—to maintain privileges they accrue through subordinating the groups they Biological expressions of social inequality oppress (intentionally and also by maintaining Biological expressions of social inequality refers to the status quo) and the ideologies they use to on September 28, 2021 by guest. Protected copyright. how people literally embody and biologically justify these practices, with these ideologies express experiences of economic and social revolving around notions of innate superiority 6 inequality, from in utero to death, thereby pro- and inferiority, diVerence, or deviance. Pre- ducing social inequalities in health across a dominant types of adverse discrimination are wide spectrum of outcomes.126Core to social based on race/ethnicity, gender, sexuality, epidemiology, this construct of “biological disability, age, nationality, and religion, and, expressions of social inequality” has been although not always recognised as such, social class. By contrast, positive discrimination (for Department of Health evident in epidemiological thought—albeit not and Social Behavior, always explicitly named as such—since the dis- example, aYrmative action) seeks to rectify Harvard School of cipline’s emergence in the early 19th century, inequities created by adverse discrimination. Public Health, 677 as exemplified by early pathbreaking research Social epidemiological analyses of health Huntington Avenue, (for example, conducted by Louis René consequences of discrimination require con- Boston, MA 02115, Villermé (1782–1863)) on socioeconomic gra- ceptualising and operationalising diverse ex- USA dients in—and eVects of poverty on— pressions of exposure, susceptibility, and resist- 378 Correspondence to: mortality, morbidity, and height. ance to discrimination, as listed below, Professor Krieger Examples include biological expressions of recognising that individuals and social groups ([email protected]) poverty and of diverse types of discrimination, may be subjected simultaneously to multiple— Accepted to publication for example, based on race/ethnicity, gender, and interacting—types of discrimination: (page 16 March 2001 sexuality, social class, disability, or age. 42)6

www.jech.com J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from 694 Krieger

Aspects of discrimination: (2) pathways of embodiment, structured si- Type: defined in reference to constituent domi- multaneously by: (a) societal arrangements of nant and subordinate groups, and justifying power and property and contingent patterns of ideology production, consumption, and reproduction, Form: structural, institutional, interpersonal; and (b) constraints and possibilities of our legal or illegal; direct or indirect; overt or cov- biology, as shaped by our species’ evolutionary ert history, our ecological context, and individual Agency: perpetrated by state or by non-state histories, that is, trajectories of biological and actors (institutional or individuals) social development. Expression: from verbal to violent; mental, (3) cumulative interplay between exposure, physical, or sexual susceptibility, and resistance, expressed in path- Domain: for example, at home; within family; ways of embodiment, with each factor and its at school; getting a job; at work; getting distribution conceptualised at multiple levels housing; getting credit or loans; getting medical (individual, neighbourhood, regional or politi- care; purchasing other goods and services; by cal jurisdiction, national, inter-national or the media; from the police or in the courts; by supra-national) and in multiple domains (for other public agencies or social services; on the example, home, work, school, other public set- street or in a public setting tings), in relation to relevant ecological niches, Level: individual, institutional, residential and manifested in processes at multiple scales neighbourhood, community, political jurisdic- of time and space. tion, national, regional, global (4) accountability and agency, expressed in pathways of and knowledge about embodi- Cumulative exposure to discrimination: ment, in relation to institutions (government, Timing: intrauterine period; infancy; child- business, and public sector), communities, hood; adolescence; adulthood households, and individuals, and also to Intensity: severe to mild accountability and agency of epidemiologists Frequency: chronic; acute; sporadic and other scientists for theories used and Duration: timespan over which discrimina- ignored to explain social inequalities in health; tion is experienced a corollary is that, given likely complementary Responses to discrimination can similarly be causal explanations at diVerent scales and lev- analysed.6 els, epidemiological studies should explicitly name and consider the benefits and limitations Ecosocial theory of disease distribution of their particular scale and level of analysis. Ecosocial12and other emerging multi-level epi- More than simply adding “biology” to demiological frameworks11 12 seek to integrate “social” analyses, or “social factors” to “bio- social and biological reasoning and a dynamic, logical” analyses, the ecosocial framework historical and ecological perspective to develop begins to envision a more systematic integrated new insights into determinants of population approach capable of generating new hypoth- distributions of disease and social inequalities eses, rather than simply reinterpreting factors in health. The central question for ecosocial identified by one approach (for example, theory is: “who and what is responsible for popu- biological) in terms of another (for example, lation patterns of health, disease, and wellbeing, as social).1 manifested in present, past, and changing social http://jech.bmj.com/ inequalities in health?” Adequate epidemiologi- Embodiment cal explanations accordingly must account for A core concept for understanding relationships both persisting and changing distributions of between the state of our bodies and the body disease, including social inequalities in health, politic; see definition in entry on “ecosocial across time and space. To aid conceptualisa- theory” tion, ecosocial theory uses a visual fractal metaphor of an evolving bush of life inter- on September 28, 2021 by guest. Protected copyright. twined with the scaVolding of society that Gender, sexism, and sex diVerent core social groups daily reinforce or Gender refers to a social construct regarding seek to alter.12 A fractal metaphor is chosen culture-bound conventions, roles, and behav- because fractals are recursive structures, re- iours for, as well as relationships between and peating and self similar at every scale, from among, women and men and boys and girls.13–15 micro to macro.2 Thus, ecosocial theory invites Gender roles vary across a continuum and both consideration of how population health is gen- gender relationships and biological expressions of erated by social conditions necessarily engag- gender vary within and across societies, typically ing with biological processes at every spatio- in relation to social divisions premised on temporal scale, whether from subcellular to power and authority (for example, class, global, or nanoseconds to millenniums.1 race/ethnicity, nationality, religion).615 Sexism, Core concepts for ecosocial theory accord- in turn, involves inequitable gender relation- ingly include1: ships and refers to institutional and interper- (1) embodiment, a concept referring to how sonal practices whereby members of dominant we literally incorporate, biologically, the mate- gender groups (typically men) accrue privileges rial and social world in which we live, from in by subordinating other gender groups (typi- utero to death; a corollary is that no aspect of cally women) and justify these practices via our biology can be understood absent knowl- ideologies of innate superiority, diVerence, or edge of history and individual and societal ways deviance.6 13–15 Lastly, sex is a biological con- of living. struct premised upon biological characteristics

www.jech.com J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from A glossary for social epidemiology 695

enabling sexual reproduction.14 16 Among peo- social) over time, as shaped by the historical ple, biological sex is variously assigned in rela- period in which they live, in reference to their tion to secondary sex characteristics, gonads, society’s social, economic, political, technologi- or sex chromosomes; sexual categories include: cal, and ecological context. One component male, female, intersexual (persons born with may involve what has been termed “biological both male and female sexual characteristics), programming”, referring to “the process and transsexual (persons who undergo surgical whereby a stimulus or insult, at a sensitive or and/or hormonal interventions to reassign their “critical” period of development, has lasting or sex).14 Sex linked biological characteristics (for lifelong significance” (page 13)27; which of example, presence or absence of ovaries, testes, these processes, under what circumstances, are vagina, penis; various hormone levels; preg- reversible is an important empirical and public nancy, etc) can, in some cases, contribute to health question. gender diVerentials in health but can also be construed as gendered expressions of biology and Multi-level analysis erroneously invoked to explain biological expres- Multi-level analysis refers to statistical method- sions of gender.116 For example, associations ologies, first developed in the social sciences, between parity and incidence of melanoma which analyse outcomes simultaneously in among women are typically attributed to preg- relation to determinants measured at diVerent nancy related hormonal changes; new research levels (for example, individual, workplace, indicating comparable associations between neighborhood, nation, or geographical region parity and incidence of melanoma among men, existing within or across geopolitical however, suggests that social conditions linked boundaries).28–31 If guided by well developed to parity, and not necessarily—or solely—the conceptual models clearly specifying which biology of pregnancy, may be aetiologically rel- variables are to be studied at which level,28 evant.17 these analyses can potentially assess whether individuals’ health is shaped by not only “indi- Human rights and social justice vidual” or “household” characteristics (for Human rights, as a concept, presumes that all example, individual or household income) but people “are born free and equal in dignity and also “population” or “area” characteristics; the rights”18 and provides a universal frame of ref- latter may be “compositional” (for example, erence for deciding questions of equity and proportion of people living in poverty) or social justice.18–21 Operationally, translated to “contextual” (irreducible to the individual the realm of political and legal accountability, level, for example, income distribution, popula- “international human rights law is about defin- tion density, or absence of facilities, such as ing what governments can do to us, cannot do to supermarkets, libraries, or health centres).30 31 us, and should do for us”19 [italics in the original], so as to respect, protect, and fulfill Poverty, deprivation (material and their human rights obligations.19 20 Human social), and social exclusion rights norms are premised, in the first instance, To b e impoverished is to lack or be denied upon the 1948 Universal Declaration of adequate resources to participate meaningfully Human Rights18 and its recognition of the indi- in society. A complex construct, poverty is visibility and interdependence of civil, political, inherently a normative concept that can be economic, social, and cultural rights.18–21 A defined—in both absolute and relative http://jech.bmj.com/ “health and human rights” framework thus not terms—in relation to: “need”, “standard of liv- only spurs recognition of how realisation of ing”, “limited resources”, “lack of basic human rights promotes health but also helps security”, “lack of entitlement”, “multiple translate concerns about how violation of deprivation”, “exclusion”, “inequality”, human rights potentially harms health into “class”, “dependency”, and “unacceptable concrete and actionable grievances that gov- hardship”32; see “socioeconomic position” ernments and the international community are (below). Also relevant is whether the experi- legally and politically required to address. ence of poverty is transient or chronic. on September 28, 2021 by guest. Protected copyright. Understanding of what prompts violation of According to the United Nations, as elabo- human rights and sustains their respect, rated in the Human Development Report 2000, protection and fulfillment is, in turn, aided by two forms of poverty can be distinguished: social justice frameworks, which explicitly ana- “human poverty” and “income poverty”(page lyse who benefits from—and who is harmed 17).21 Human poverty is “defined by impover- by—economic exploitation, oppression, dis- ishment in multiple dimensions—deprivations crimination, inequality, and degradation of in a long and healthy life, in knowledge, in a “natural resources”.21–24 Together, both frame- decent standard of living, in participation”; works provide concepts relevant for analysing income poverty, by contrast, “is defined by dep- social determinants of health and for guiding rivation in a single dimension—income” (page action to create just and sustainable societies. 17.21 From this perspective, income poverty constitutes a critical (but not exclusive) deter- Lifecourse perspective minant of human poverty, including the latter’s Lifecourse perspective refers to how health status expression in compromised health status. at any given age, for a given birth cohort, Deprivation (pages 10–11, 36–37)33 can be reflects not only contemporary conditions but conceptualised and measured, at both the indi- embodiment of prior living circumstances, in vidual and area level, in relation to: material utero onwards.25–27 At issue are people’s devel- deprivation, referring to “dietary, clothing, opmental trajectories (both biological and housing, home facilities, environment, location

www.jech.com J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from 696 Krieger

and work (paid and unpaid),” and social depri- possession of selective and arbitrary physical vation, referring to rights in relation to characteristics (for example, skin colour).613 “employment, family activities, integration into Racism refers to institutional and individual the community, formal participation in social practices that create and reinforce oppressive institutions, recreation and education”(page systems of race relations (see “discrimination”, 93).34 above).61541 Ethnicity, a construct originally Poverty thresholds accordingly can be set at: intended to discriminate between “innately” (a) an income level (for example, poverty line) diVerent groups allegedly belonging to the determined inadequate for meeting subsist- same overall “race”,42 43 is now held by some to ence needs, or (b) “the point at which refer to groups allegedly distinguishable on the resources are so seriously below those com- basis of “culture”44; in practice, however, “eth- manded by the average individual or family nicity” cannot meaningfully be disentangled that the poor are, in eVect, excluded from ordi- from “race” in societies with inequitable race nary living patterns, customs, and activities”, relations, hence the construct “race/ such that the poverty line equals “the point at ethnicity”.642 which withdrawal escalates disproportionately Two diametrically opposed constructs are to the falling resources” (pages 116–17).33 thus relevant to understanding research on and Social exclusion, another term encompassing explaining racial/ethnic disparities in health.645 aspects of poverty, in turn focuses attention on The first is: racialised expressions of biology, not only the impact but also the process of whereby measured average biological diVer- marginalisation (pages 54–6).33 35 Avenues by ences between members of diverse racial/ethnic which social groups and individuals can groups are assumed to reflect innate, geneti- become excluded from full participation in cally determined diVerences (premised, in the social and community life include: (a) legal first instance, on the arbitrary phenotypic exclusion (for example, de jure discrimina- characteristics seized upon to define, tautologi- tion), (b) economic exclusion (due to eco- cally, racial categories). The second is: biological nomic deprivation), (c) exclusion due to lack of expressions of racism (see “biological expressions provision of social goods (for example, no of social inequality”, above). For example, fol- translation services or lack of facilities for disa- lowing dominant ideas construing “race” as an bled persons), and (d) exclusion due to innate biological characteristic, epidemiologi- stigmatisation (for example, of persons with cal research has been rife with studies attempt- HIV/AIDS) and de facto discrimination. ing to explain racial/ethnic disparities in health in relation to presumed genetic diVerences, Psychosocial epidemiology absent consideration of eVects of racism on A psychosocial framework directs attention to health.6 45–46 47 Alternatively, considering lived both behavioural and endogenous biological experiences of racism as real but the construct responses to human interactions.1 At issue is of biological “race” as spurious, social epide- the “health-damaging potential of psychologi- miological research investigates health conse- cal stress”, as “generated by despairing circum- quences of economic and non-economic ex- stances, insurmountable tasks, or lack of social pressions of racial discrimination.6 13 45–48 support” (page 41)36; see also “stress” (below). Typically conceptualised in relation to indi- viduals, its central hypothesis is that chronic Sexualities and heterosexism http://jech.bmj.com/ and acute social stressors: (a) alter host suscep- Sexuality refers to culture bound conventions, tibility or become directly pathogenic by roles, and behaviours involving expressions of aVecting neuroendocrine function, and/or (b) sexual desire, power, and diverse emotions, induce health damaging behaviours (especially mediated by gender and other aspects of social in relation to use of psychoactive substances, position (for example, class, race/ethnicity, diet, and sexual behaviours).1436“Social capi- etc).49 Distinct components of sexuality in- tal” and “social cohesion”, in turn, are clude: sexual identity, sexual behaviour, and proposed (and contested) as population level sexual desire. Contemporary “Western” cat- on September 28, 2021 by guest. Protected copyright. psychosocial assets that potentially can im- egories by which people self identify or can be prove population health by influencing norms labelled include: heterosexual, homosexual, and strengthening bonds of “civil society”, with lesbian, gay, bisexual, “queer”, transgendered, the caveat that membership in certain social transsexual, and asexual. Heterosexism, the type formations can potentially harm either mem- of discrimination related to sexuality, consti- bers of the group (for example, group norms tutes one form of abrogation of sexual rights50 encourage high risk behaviours) or non-group and refers to institutional and interpersonal members (for example, harm caused to groups practices whereby heterosexuals accrue privi- subjected to discrimination by groups support- leges (for example, legal right to marry and to ing discrimination).1 37–40 have sexual partners of the “other” sex) and discriminate against people who have or desire Race/ethnicity and racism same sex sexual partners, and justify these Race/ethnicity is a social, not biological, cat- practices via ideologies of innate superiority, egory, referring to social groups, often sharing diVerence, or deviance. Lived experiences of cultural heritage and ancestry, that are forged sexuality accordingly can aVect health by path- by oppressive systems of race relations, justified ways involving not only sexual contact (for by ideology, in which one group benefits from example, spread of sexually transmitted dis- dominating other groups, and defines itself and ease) but also discrimination and material con- others through this domination and the ditions of family and household life.49 50

www.jech.com J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from A glossary for social epidemiology 697

Society, social, societal, and culture managerial class—exist in relationship to and Society, originally meaning “companionship or co-define each other. One cannot, for example, fellowship”, now stands as “our most general be an employee if one does not have an term for the body of institutions and relation- employer and this distinction—between em- ships within which a relatively large group of ployee and employer—is not about whether people live and as our most abstract term for one has more or less of a particular attribute, the condition in which such institutions and but concerns one’s relationship to work and to relationships are formed”(page 291).51 Social, others through a society’s economic structure. as an adjective, likewise has complex meanings: Class, as such, is not an a priori property of “as a descriptive term for society in its now pre- individual human beings, but is a social dominant sense of the system of common life”, relationship created by societies. As such, social and also as “an emphatic and distinguishing class is logically and materially prior to its term, explicitly contrasted with individual and expression in distributions of occupations, especially individualist theories of society”(page income, wealth, education, and social status. 286) [italics in the original].51 Societal, in turn, One additional and central component of class serves as a “more neutral reference to general relations entails an asymmetry of economic social formations and institutions” (page 294).51 exploitation, whereby owners of resources (for By this logic, social epidemiology and its social example, capital) gain economically from the theories of disease distribution stand in con- labour or eVort of non-owners who work for trast to individualistic epidemiology, which relies them. on individualistic theories of disease causation Socioeconomic position, in turn, is an aggre- (see “theories of disease distribution”, below). gate concept that includes both resource-based Culture, originally a “noun of process” refer- and prestige-based measures, as linked to both ring to “the tending of something, basically childhood and adult social class position.54-56 crops or animals,” (page 87)51 presently has Resource-based measures refer to material and three distinct meanings: “(i) the independent social resources and assets, including income, and abstract noun which describes a general wealth, and educational credentials; terms used process of intellectual, spiritual, and aesthetic to describe inadequate resources include “pov- development . . .; (ii) the independent noun, erty” and “deprivation” (see “poverty”, above). whether used generally or specifically, which Prestige-based measures refer to individuals’ indicates a particular way of life, whether of a rank or status in a social hierarchy, typically people, a period, a group, or humanity in gen- evaluated with reference to people’s access to eral; and . . . (iii) the independent and abstract and consumption of goods, services, and noun which describes the work and practices of knowledge, as linked to their occupational intellectual and especially artistic activity” prestige, income, and educational level. Given (page 90).51 In social epidemiology, meaning distinctions between resource-based and (ii) predominates, with “culture” typically con- prestige-based aspects of socioeconomic posi- ceptualised and operationalised in relation to tion and the diverse pathways by which they health related beliefs and practices, especially aVect health, epidemiological studies should dietary practices. By this logic, “acculturation” state clearly how measures of socioeconomic (or, perhaps more accurately “decultura- position are conceptualised. The term “socio- tion”45) refers to members of one “culture” economic status” should be eschewed because adopting beliefs and practices of another (and it arbitrarily (if not intentionally) privileges http://jech.bmj.com/ typically dominant) “culture”.52 53 Related, “status”—over material resources—as the key examples abound44 53 in epidemiological litera- determinant of socioeconomic position.54 ture whereby the construct of “culture” is con- flated with “ethnicity” (and “race”) and together are inappropriately invoked to explain Social determinants of health socioeconomic and health characteristics of Social determinants of health refer to both diverse population groups on the basis of specific features of and pathways by which “innate” qualities, rather than as a conse- societal conditions aVect health and that on September 28, 2021 by guest. Protected copyright. quence of inequitable social relationships potentially can be altered by informed ac- between groups.52 tion.42457 As determinants, these social proc- esses and conditions are conceptualised as Social class and socioeconomic position “essential factors” that “set certain limits or Social class refers to social groups arising from exert pressures”, albeit without necessarily interdependent economic relationships among being “deterministic” in the sense of “fatalistic people (pages 60–69).51 54–56 These relation- determinism” (pages 98–102).51 ships are determined by a society’s forms of Historically contingent, social determinants property, ownership, and labour, and their of health, broadly writ, include: connections through production, distribution, (a) a society’s past and present economic, and consumption of goods, services, and infor- political, and legal systems, its material and mation. Social class is thus premised upon technological resources, and its adherence to people’s structural location within the norms and practices consistent with inter- economy—as employers, employees, self em- national human rights norms and standards; ployed, and unemployed (in both the formal and and informal sector), and as owners, or not, of (b) its external political and economic capital, land, or other forms of economic relationships to other countries, as imple- investments. Stated simply, classes—like the mented through interactions among govern- working class, business owners, and their ments, international political and economic

www.jech.com J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from 698 Krieger

organisations (for example, United Nations, Social production of scientific knowledge World Bank, International Monetary Fund), Social production of scientific knowledge refers to and non-governmental organisations. ways in which social institutions and beliefs One term appearing in social epidemiologi- aVect recruitment, training, practice, and cal literature to summarise social determinants funding of scientists, thereby shaping what of health is “social environment”.4757 This questions we, as scientists, do and do not ask, metaphor invokes notions of “environment”, a the studies we do and do not conduct, and the term literally referring to “surroundings” and ways in which we analyse and interpret data, initially used to denote the physical, including consider their likely flaws, and disseminate both “natural” and “built”, environment. Both results.68–71 “social environment” and the related metaphor That scientists’ ideas are shaped, in part, by “social ecology” are problematic in that they dominant social beliefs of their times is well can conceal the role of human agency in creat- documented.3 72–74 Relevant to social epidemiol- ing social conditions that constitute social ogy, a substantial body of literature demon- determinants of health.1 strates how scientific knowledge and, more importantly, real people, have been harmed by scientific racism, sexism and other related Social inequality or inequity in health ideologies, including eugenics, which justify dis- and social equity in health crimination and discount the importance of Social inequalities (or inequities) in health refer to understanding and ameliorating social inequali- health disparities, within and between coun- ties in health.6 Tellingly, as of the year 2000, only tries, that are judged to be unfair, unjust, 0.05% of the approximately 34 000 articles avoidable, and unnecessary (meaning: are nei- indexed in Medline by the keyword “race” had ther inevitable nor unremediable) and that sys- explicitly investigated racial discrimination as a tematically burden populations rendered vul- determinant of population health.6 nerable by underlying social structures and political, economic, and legal institutions.21 58 59 Stress As such, social inequalities (or inequities) in Stress, a term widely used in the biological, health are not synonymous with “health physical, and social sciences, is a construct inequalities”, as this latter term can be whose meaning in health research is variously interpreted to refer to any diVerence and not defined in relationship to “stressful events, specifically to unjust disparities.58 59 For exam- responses, and individual appraisals of situa- ple, recently proposed measures of “health tions” (page 3).75 Common to these definitions inequalities” deliberately quantify distributions is “an interest in the process in which environ- of health in populations without reference to mental demands tax or exceed the adaptive capac- either social groups and or social inequalities in ity of an organism, resulting in psychological or health.59–62 biological changes that may place persons at risk for Social equity in health, in turn, refers to an disease” [italics in original] (page 3).75 An absence of unjust health disparities between “environmental stress perspective” focuses on social groups, within and between countries.58 “environmental demands, stressors, or events” Promoting equity and diminishing inequity (page 4)75; a “psychological stress perspective”, requires not only a “process of continual on “an organism’s perception and evaluation of equalization” but also a “process of abolishing http://jech.bmj.com/ the potential harm posed by objective environ- or diminishing privileges” (pages 117–19).51 mental exposures” (page 6)75; a “biological Thus, pursuing social equity in health entails stress perspective”, on “activation of the physi- reducing excess burden of ill health among ological systems that are particularly respon- groups most harmed by social inequities in sive to physical and psychological demands” health, thereby minimising social inequalities (page 8).75 Whether social epidemiological in health and improving average levels of health research conceptualises stress in relation to overall.21 structural, interpersonal, cognitive, or biologi- cal parameters, and whether it uses “environ- on September 28, 2021 by guest. Protected copyright. Social production of disease/political ment” as a term or metaphor that reveals or economy of health conceals the role of human agency and Social production of disease/political economy of accountability in determining distributions of health refers to related (if not identical) “stress”, depends on the underlying theories of theoretical frameworks that explicitly address disease distribution guiding the research (see economic and political determinants of health “theories of disease distribution”, below, and and distributions of disease within and across “social determinants”, above). societies, including structural barriers to peo- ple living healthy lives.1 63–66 These theories Theories of disease distribution accordingly focus on economic and political Theories of disease distribution seek to explain cur- institutions and decisions that create, enforce, rent and changing population patterns of disease and perpetuate economic and social privilege across time and space and, in the case of social and inequality, which they conceptualise as epidemiology, across social groups (within and root—or “fundamental”67—causes of social across countries, over time).1 Using—like any inequalities in health. Although compatible theory (pages 316–18)51 71—interrelated sets of with the ecosocial theory of disease distribu- ideas whose plausibility can be tested by human tion, they diVer in that they do not systemati- action and thought, theories of disease distribu- cally seek to integrate biological constructs into tion presume but cannot be reduced to mech- explanations of social patternings of health.12 anism oriented theories of disease causation.1

www.jech.com J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from A glossary for social epidemiology 699

Explicit attention to aetiological theory is essen- 32 Spicker P. Definitions of poverty: eleven clusters of meaning. In: Gordon D, Spicker P, eds. The international tial, because shared observations of social glossary on poverty. London: Zed Books, 1999:150–62. disparities in health do not necessarily translate 33 Gordon D, Spicker P, eds. The international glossary on pov- 1 erty. London: Zed Books, 1999. to common understandings of causes. Excess 34 Townsend P. The international analysis of poverty. New York: risk of HIV/AIDS among poor women of colour, Harvester/Wheatsheaf, 1993. 35 Shaw M, Dorling D, Davey Smith G. Poverty, social exclu- for example, is attributed to social inequity by sion, and minorities. In: Marmot M, Wilkinson RG, eds. ecosocial and social production of disease theo- Social determinants of health. Oxford: Oxford University Press, 1999:211–39. ries of disease distribution, but is attributed to 36 Elstad JI. The psycho-social perspective on social inequali- “bad behaviours” by biomedical lifestyle theo- ties in health. In: Bartley M, Blane D, Davey Smith G, eds. ries of disease causation.176 The sociology of health inequalities. Oxford: Blackwell, 1998: 39–58. 37 Kawachi I, Berkman L. Social cohesion, social capital, and Thanks to Sofia Gruskin, Mary Northridge, and George Davey health. In: Berkman L, Kawachi I, eds. Social epidemiology. Smith for helpful comments. Oxford: Oxford University Press, 2000:174–90. 38 Wilkinson RG. Unhealthy societies: the aZictions of inequality. 1 Krieger N. Emerging theories for social epidemiology in the London: Routledge, 1996. 21st century: an ecosocial perspective. Int J Epidemiol (in 39 Lynch JW, Davey Smith G, Kaplan GA, House JS. Income press). inequality and mortality: importance to health of individual 2 Krieger N. Epidemiology and the web of causation: has any- incomes, psychological environment, or material condi- one seen the spider? Soc Sci Med 1994;39:887–903. tions. BM J 2000;320:1200–4. 3 Krieger N. Epidemiology and social sciences: towards a 40 Kunitz SJ. Accounts of social capital: the mixed health eVects critical reengagement in the 21st century. Epidemiol Rev of personal communities and voluntary groups. In: Leon D, 2000;11:155–63. Walt G, eds. Poverty, inequality, and health: an international 4 Berkman L, Kawachi I, eds. Social epidemiology. Oxford: perspective. Oxford: Oxford University Press, 2001:159–74. Oxford University Press, 2000. 41 Essed P. Understanding everyday racism: an interdisciplinary 5 Yankauer A. The relationship of fetal and infant mortality to theory. London: Sage, 1992. residential segregation: an inquiry into social epidemiology. 42 Statistics Canada and US Bureau of the Census. Challenges Am Sociol Review 1950;15:644–8. of measuring in an ethnic world: Science, politics, and reality. 6 Krieger N. Discrimination and health. In: Berkman L, Washington, DC: US Government Printing OYce, 1993. Kawachi I, eds. Social epidemiology. Oxford: Oxford Univer- 43 Hobsbawm EJ. Nations and nationalism since 1780: pro- sity Press, 2000:36–75. gramme, myth, reality. 2nd ed. Cambridge: Cambridge Uni- 7 Sydenstricker E. Health and environment. New York: versity Press, 1992. McGraw-Hill, 1933. 44 Haynes MA, Smedley BD, eds. The unequal burden of cancer: 8 Morris JN. Uses of epidemiology. Edinburgh: Livingston, an assessment of NIH research and programs for ethnic minori- 1957. ties and the medically underserved. Washington, DC: 9 Jary D, Jary J, eds. Collins dictionary of sociology. 2nd ed. National Academy Press, 1999. Glasgow, UK: HarperCollins Publishers, 1995. 45 Krieger N. Refiguring “race”: epidemiology, racialized biol- 10 Marshall G, ed. The concise Oxford dictionary of sociology. ogy, and biological expressions of race relations. Int J Oxford: Oxford University Press, 1994. Health Services 2000;30:211–16. 11 Susser M, Susser E. Choosing a future for epidemiology: II. 46 Williams DR. Race, socioeconomic status, and health. The from black boxes to Chinese boxes and eco-epidemiology. added eVects of racism and discrimination. Ann NY Acad Am J Public Health 1996;86:674–7. Sci 1999;896:173–88. 12 McMichael AJ. Prisoners of the proximate: loosening the 47 Lillie-Blanton M, LaVeist T. Race/ethnicity, the social envi- constraints on epidemiology in an age of change. Am J Epi- ronment, and health. Soc Sci Med 1996;43:83–92. demiol 1999;149:887–97. 48 Davey Smith G. Learning to live with complexity: ethnicity, 13 Krieger N, Rowley DL, Herman AA, et al. Racism, sexism, socioeconomic position, and health in Britain and the and social class: implications for studies of health, disease, United States. Am J Public Health 2000;90:1694–8. and well-being. Am J Prev Med 1993;9 (suppl):82–122. 49 Parker RG, Gagnon JH, eds. Conceiving sexuality: approaches 14 Fausto-Sterling A. Sexing the body: gender politics and the con- to sex research in a post-modern world. New York: Routledge, struction of sexuality. New York: Basic Books, 2000. 1995. 15 Essed P. Diversity: gender, color, and culture. Amherst, MA: 50 Miller AM. Sexual but not reproductive: exploring the junc- University of Massachusetts, 1996. tion and disjunction of sexual and reproductive rights. 16 Ruiz MT, Verbrugge LM. A two way view of gender bias in Health and Human Rights 2000;4:68–109. medicine. J Epidemiol Community Health 1997;51:106–9. 51 Williams R. Keywords: a vocabulary of culture and society. 17 Kravdal O. Is the relationship between childbearing and Revised ed. New York: Oxford University Press, 1983. cancer incidence due to biology or lifestyle? Examples of the importance of using data on men. Int J Epidemiol 1995; 52 Kunitz SJ. Disease and social diversity: the European impact on the health of non-Europeans. New York: Oxford University 4:477–84. 18 United Nations. Universal declaration of human rights.GA Press, 1994. http://jech.bmj.com/ Res 217A(III), UN GAOR, Res 71, UN Doc A/810, 1948. 53 Lin SS, Kelsey JL. Use of race and ethnicity in epidemio- 19 Gruskin S, Tarantola D. Health and human rights. In: logic research: concepts, methodologic issues, and sugges- Detels R, McEwen J, Beaglehole R, et al,eds.The Oxford tions for research. Epidemiol Rev 2000;22:187–202. textbook of public health. 4th ed. New York: Oxford Univer- 54 Krieger N, Williams D, Moss N. Measuring social class in sity Press (in press). US public health research: concepts, methodologies and 20 Mann JM, Gruskin S, Grodin MA, et al,eds.Health and guidelines. Annu Rev Public Health 1997;18:341–78. human rights. New York: Routledge, 1999. 55 Wright EO. Class counts: comparative studies in class analysis. 21 UNDP 2000: United Nations Development Programme New York: Cambridge University Press, 1997. (UNDP). Human development report 2000:Human rights and 56 Lynch J, Kaplan G. Socioeconomic position. In: Berkman human development. New York: Oxford University Press, L, Kawachi I, eds. Social epidemiology. Oxford: Oxford Uni- 2000. versity Press, 2000:13–35. 22 Boucher D, Kelly P, ed. Social justice: from Hume to Walzer. 57 Marmot M, Wilkinson RG, eds. Social determinants of health. on September 28, 2021 by guest. Protected copyright. London: Routledge, 1998. Oxford: Oxford University Press, 1999. 23 Krieger N, Birn A-E. A vision of social justice as the founda- 58 Whitehead M. The concepts and principles of equity and tion of public health: commemorating 150 years of the Spirit health. Int J Health Services 1992;22:429–45. of 1848. Am J Public Health 1998;88:1603–6. 59 Leon DA, Walt G, Gilson L. International perspectives on 24 People’s Health Assembly 2000. People’s charter for health. health inequalities and policy. BMJ 2001;332:591–4. Gonoshasthaya Kendra, Savar, Bangladesh December 4–8, 60 Murray C, Gakidou EE, Frenk J. Health inequalities and 2000. At: http://www.pha2000.org [last accessed: 11 Feb social group diVerences: what should we measure? Bull 2001]. WHO 1999;77:537–43. 25 Kuh D, Ben Shlomo Y, eds. A lifecourse approach to chronic 61 Murray CJL, Frenk J, Gadikou EE. Measuring health disease epidemiology. Oxford: Oxford University Press, 1997. inequality: challenges and new directions. In: Leon D, Walt 26 Davey Smith G, Gunnell D, Ben-Shlomo Y. Life-course G, eds. Poverty, inequality, and health: an international perspec- approaches to socio-economic diVerentials in cause- tive. Oxford: Oxford University Press, 2001:194–216. specific adult mortality. In: Leon D, Walt G, eds. Poverty, 62 Braveman P, Krieger N, Lynch J. Health inequalities and inequality, and health: an international perspective. Oxford: social inequalities in health. Bull WHO 2000;78:232–4. Oxford University Press, 2001:88–124. 63 Doyal L. The political economy of health. London: Pluto Press, 27 Barker DJP. Mothers, babies, and health in later life. 2nd ed. 1979 (1935). Edinburgh: Churchill Livingston, 1998. 64 Breilh J. Epidemiologia economia medicina y politica. 4th ed. 28 Blalock HM Jr. Contextual-eVects models: theoretic and Mexico City, Mexico: Distribuciones Fontamara, 1988. methodologic issues. Annu Review Sociol 1984;10:353–72. 65 Conrad P, Kern R, eds. The sociology of health and illness: 29 Bryk AS, Raudenbush SW. Hierarchical linear models:applica- critical perspectives. New York: St Martin’s Press, 1981. tions and data analysis methods. Newbury Park, CA: Sage, 66 Navarro V. Crisis, health, and medicine: a social critique.New 1992. York: Tavistock, 1986. 30 Diez-Roux AV. Bringing context back into epidemiology: 67 Link BG, Phelan JC. Editorial: understanding socio- variables and fallacies in multilevel analysis. Am J Public demographic diVerences in health—the role of fundamen- Health 1998;88:216–22. tal social causes. Am J Public Health 1996;86:471–3. 31 Macintyre S, Ellaway A. Ecological approaches: rediscover- 68 Fleck L. Genesis and development of a scientific fact. Chicago: ing the role of the physical and social environment. In: University of Chicago University Press, 1979. Berkman L, Kawachi I, eds. Social epidemiology. Oxford: 69 Rose H, Rose S, eds. Ideology of/in the natural sciences. Cam- Oxford University Press, 2000:332–48. bridge, MA: Schenkman, 1979.

www.jech.com J Epidemiol Community Health: first published as 10.1136/jech.55.10.693 on 1 October 2001. Downloaded from 700 Krieger

70 Haraway D. Primate visions: gender, race, and nature in the 74 Porter D. Health, civilization, and the state: a history of public world of modern science. New York: Routledge, 1989. health from ancient to modern times. London: Routledge, 71 Ziman JM. Real science: what it is, and what it means. 1999. Cambridge: Cambridge University Press, 2000. 75 Cohen S, Kessler RC, Underwood L. Measuring stress: a 72 Fee E. Disease and discovery: a history of the Johns Hopkins guide for health and social scientists. New York: Oxford Uni- School of Hygiene and Public Health, 1916–1936. Baltimore: versity Press, 1995. Johns Hopkins University Press, 1987. 76 Fee E, Krieger N. Understanding AIDS: historical interpre- 73 Rosenberg CD, Golden J, eds. Framing disease:studies in cultural tations and the limits of biomedical individualism. Am J history. New Brunswick, NJ: Rutgers University Press, 1992. Public Health 1993;83:1477–86.

7th European Forum on Quality Improvement in Health Care 21–23 March 2002 Edinburgh, Scotland

We are delighted to announce this forthcoming conference in Edinburgh. Authors are invited to submit papers (call for papers closes on Friday 5 October 2001) and delegate enquiries are welcome. The themes of the Forum are: x Leadership, culture change, and change management x Achieving radical improvement by redesigning care x Health policy for lasting improvement in health care systems x Patient safety x Measurement for improvement, learning, and accountability x Partnership with patients x Professional quality: the foundation for improvement x Continuous improvement in education and training x People and improvement. Presented to you by the BMJ Publishing Group (London, UK) and Institute for Healthcare Improvement (Boston, USA). For more information contact: [email protected] or look at http://jech.bmj.com/ the website www.quality.bmjpg.com. Tel: +44 (0)20 7383 6409; fax: +44 (0)20 7373 6869. on September 28, 2021 by guest. Protected copyright.

www.jech.com