Year: 2004
Type: article
Abstract: Social epidemiology is defined as the study of the distribution of health outcomes and their social determinants (1). It builds on the classic epidemiologic triangle of host, agent, and environment to focus explicitly on the role of social determinants in infectious disease transmission and progression. These determinants are the “features of and pathways by which societal conditions affect health” (2, p. 697). Early studies of human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) focused on individual characteristics and behaviors in determining HIV risk, an approach that Fee and Krieger (3) refer to as “biomedical individualism.” Biomedical individualism is the basis of risk factor epidemiology; by contrast, the social epidemiology perspective emphasizes social conditions as fundamental causes of disease (4) (table 1). Social epidemiologists examine how persons become exposed to risk or protective factors and under what social conditions individual risk factors are related to disease. Social factors are thus the focus of analysis and are not simply adjusted for as potentially confounding factors or used as proxies for unavailable individual-level data. Social factors are indeed critical to understanding nonuniform infectious disease patterns that emerge as a result of the dependent nature of disease transmission or the idea that an outcome in one person is dependent upon outcomes and exposures in others (5, 6). Contact patterns that enhance HIV/AIDS vulnerability may be conceptualized at multiple levels. Figure 1 distinguishes determinants of HIV/AIDS at three levels: individual, social, and structural. Individual factors include biologic, demographic, and behavioral risk factors that may influence the risk of HIV acquisition and disease progression. Social-level factors include critical pathways by which community and network structures link persons to society. These structures are central to understanding the diffusion and differential distribution of HIV/AIDS in population subgroups. Structural-level factors include social and economic factors, as well as laws and policies. These factors, in turn, affect HIV transmission dynamics and the differential distribution of HIV/AIDS. Infectious disease epidemiology provides models of the mechanisms through which social determinants affect HIV transmission (7). For example, the basic reproductive number of an infectious disease, R0 (8), describes secondary infections that arise from a primary infection. In the equation R0 = βCD, β is the probability of infection per contact, C is the number of contacts, and D is the duration of infectivity. The goal of intervention efforts is to reduce the empirical value of these terms by modifying the social conditions under which individual risk factors lead to disease. Examples of factors that affect the component terms of R0 in HIV epidemiology are presented in table 2. In this review, we present existing evidence linking social and structural determinants to HIV/AIDS. In addition, we discuss the implications of these findings for future social epidemiology research on HIV/AIDS as well as the design of more effective HIV/AIDS interventions. We searched the published literature to identify conceptual and empirical research reports on the social epidemiology of HIV/AIDS. Five databases were searched: PsycINFO (American Psychological Association, Washington, DC), PubMed (MEDLINE; National Institutes of Health, Bethesda, Maryland), Social Science Citation Index (Web of Science; Thomson ISI, Stamford, Connecticut), Sociological Abstracts (CSA, Bethesda, Maryland), and Digital Dissertations (ProQuest; UMI, Ann Arbor, Michigan). Searches were designed to include the factors we specified in our framework as social or structural factors (figure 1). Searches were limited to published articles in the English language for the period 1981–2003. The following keywords were included in each search: AIDS/acquired immunodeficiency syndrome, HIV, and epidemiol*. Additional searches were conducted by using combinations of keywords listed in Appendix table 1 corresponding with our framework. We identified four categories of social-level factors of importance to HIV/AIDS epidemiology: cultural context, social networks, neighborhood effects, and social capital. Each uses different conceptual and methodological approaches to examine the effects of social forces on population HIV/AIDS vulnerability. Anthropologist Edward Tylor defined culture as “that complex whole which includes knowledge, belief, art, law, morals, custom, and any other capabilities and habits acquired by man as a member of society” (9, p. 1). Anthropologic and epidemiologic approaches may be integrated in a variety of ways to identify features of the social environment that affect HIV/AIDS risk. One way to explore how the social environment affects HIV/AIDS epidemiology is through the use of mixed research methods. Mixed-methods study designs integrate qualitative and quantitative research methods either sequentially or concurrently (10). In sequential study designs, qualitative methods may be used to explore a topic under study or to explain quantitative epidemiologic findings. Concurrent study designs are meant to confirm, cross-validate, or corroborate findings within a single study. A common type of concurrent mixed methods study is “triangulation,” and this approach has been used extensively in rapid assessments of illicit drug use and HIV/AIDS (11–13). Mixed methods approaches are particularly well suited to the investigation of the often hidden and stigmatizing behavioral and social factors underlying HIV epidemics. One exemplary study combining qualitative methods with quantitative methods was conducted by Beyrer et al. (14) to examine the role of overland heroin trafficking routes in shaping explosive HIV/AIDS epidemics among injection drug users in Southeast Asia. Piecing together data from a variety of sources, including existing epidemiologic data, key informant interviews, and laboratory data, this study revealed that distinct HIV subtypes emerged and recombined along drug trafficking routes originating in Myanmar, one of the world’s largest heroin producers. Along these trafficking routes, communities of injection drug users formed, facilitating the spread of HIV into local communities in Laos, Thailand, Vietnam, India, and China (refer, for example, to Panda et al. (15)). This illustration highlights the broader understanding of HIV/AIDS epidemiology that can be achieved by examining the interplay between contextual factors and social and behavioral factors. Investigation of social networks in HIV/AIDS began with the mapping of relationships between one of the first identified AIDS cases, an airline steward, and a large number of his male sex partners in the early 1980s (16). Social network analysis generates measures of the quality, density, position, and structure of relationships between persons, including dyads (partnerships), personal networks (“egocentric” networks), and larger communities (“sociometric” networks) (17, 18). Social networks can influence health outcomes in direct and indirect ways, including 1) social influence, 2) social engagement and participation, 3) prevalence of infectious disease and network member mixing, 4) access to material goods and informational resources, and 5) social support (19). Researchers have demonstrated that patterns in the structure of relationships—rather than differences in individual risk behaviors alone—explain observed HIV patterns (20, 21). The theoretical foundation for examining social networks in HIV research is closely tied to advances in sexually transmitted disease (STD) epidemiology. A key concept from STD epidemiology is the notion of the “core group,” a small group of disease transmitters responsible for a large proportion of cases (22). Friedman et al. (23) found that individuals’ locations within sociometric risk networks were associated with HIV risk among a group of injection drug users in New York City. Other concepts from STD epidemiology, such as partner concurrency, bridging, and mixing patterns, are also important in understanding HIV risk (24–29). Specific network characteristics that have been associated with HIV/AIDS include the size of subgroups and their distribution in a network (23), the centrality of HIV-positive persons within networks (30), partner selection patterns (24, 31–33), and concurrent sexual partnerships (28). Inclusion of these variables has been shown to improve transmission estimates in mathematical modeling (34, 35). Social and normative influences have also been associated with individual HIV risks (36, 37). Network-related social and normative influences are predictive of illicit drug use (38) and condom use behavior (37, 39), highlighting the importance of network-based interventions for HIV prevention (18). Kelly et al. (40, 41) developed a popular opinion leader model that has been effective in reducing HIV risk in several populations, including men who have sex with men and women in low-income housing (42). The success of this model has led to its adaption for international use by the National Institute of Mental Health Collaborative HIV/STD Prevention Trial in China, India, Peru, Russia, and Zimbabwe. Neighborhoods represent the intersection of social networks and physical spatial locations, a confluence Wallace (43) has called the “sociogeographic networks” through which infectious diseases spread. Early interest in the role of neighborhood social environment in disease transmission was sparked by a study in Colorado Springs, Colorado, in which researchers found that gonorrhea was highly focused geographically in core residential neighborhoods (44). Both direct and indirect mechanisms may determine how neighborhood-level factors shape population HIV/AIDS patterns. Direct mechanisms are those that increase the likelihood of a person coming into contact with someone who is HIV positive, for example, through residential segregation and the social isolation of marginalized populations. Indirect mechanisms include those that increase population vulnerability to HIV/AIDS, such as exposure to poor socioeconomic conditions, high unemployment, or the proliferation of illicit drug markets. A range of neighborhood-level factors have been examined in relation to infectious disease, including poverty and income (45, 46), residential segregation (47), and neighborhood physical environment (48, 49). Current research in neighborhood and area effects on health emphasizes the importance of moving beyond documentation of associations to analyze the social and epidemiologic mechanisms through which neighborhood effects might operate (50–54). Increasing concentrations of affluence and poverty are contributing to what demographer Douglas Massey has called “a radical change in the geographic basis of human society” (55, p. 395). Powerful social and economic forces in US cities are increasing neighborhood segregation by class and race/ethnicity (56, 57). Resulting social disorganization and loss of resources and services in poor neighborhoods are in turn shaping HIV/AIDS patterns at the neighborhood level. In a number of studies in New York City, for example, Wallace (58–63) has examined the complex interplay of public policies such as “planned shrinkage” with HIV epidemic dynamics in the Bronx, documenting the “synergy of plagues” that has accompanied rapid social change and the destruction of essential protective networks in poor communities. Using AIDS surveillance data, ecologic studies conducted in various US cities have also consistently found significant associations between income and poverty measures and neighborhood-level AIDS incidence and prevalence rates, and these findings have been consistent across census block groups (46), census tracts (64), and zip codes (65, 66). Length of survival an AIDS has also been with neighborhood measures of income and the of highly a of health at the population may also a role in shaping HIV/AIDS patterns, associations between HIV/AIDS and income at the neighborhood have not been well segregation by race/ethnicity is neighborhood-level that may an important role in HIV/AIDS may affect infectious disease patterns through the and isolation of persons in one increasing the probability of transmission within that For example, found that measures of residential isolation were protective for at risk of disease. Indirect effects of segregation are associated with of neighborhood and with for those in poor neighborhoods segregation may to understanding disease we of studies examining neighborhood segregation in relation to HIV/AIDS that have been The physical environment of neighborhoods has also been examined in relation to infectious disease. et al. examined gonorrhea and neighborhood physical environment in New by using an of physical to explore and to this the of physical such as and a in social in a of community this concept to public et al. found a significant between neighborhood physical and gonorrhea rates, a by a ecologic study of US physical environment may HIV risk by illicit drug use such as injection behaviors and of the mechanisms through which the observed associations may be and associations between the physical environment and HIV/AIDS is research is to support the design of neighborhood-level HIV/AIDS interventions. has differences are not result from social and economic by policies. are and to p. The of evidence ecologic associations between neighborhood-level factors and infectious disease that to be to identify of and defined social as of social structures that that is the of that in its not be p. Social may affect health through 1) the of 2) access to services and 3) of in a and 4) participation, to policies that are more to published studies have explicitly examined social in the of HIV/AIDS. In the and examined income and social as of STD and AIDS found social to be the of STD and AIDS In et al. examined one of social participation, as a for understanding community influences on HIV infection. found that in of and was in other social groups groups with high of social HIV risk. findings from these studies are and We identified categories of factors to HIV/AIDS epidemiology: structural and the and and Each is in the that highlights a of by often and forces that through as is more the the of p. is in patterns of on sexual and HIV A of how structural might influence HIV/AIDS risk is presented in 2. The and uses of race/ethnicity in epidemiologic research have been the of analysis and Social epidemiologists race/ethnicity as an of social forces than physical has that race/ethnicity is a for exposure to which may be defined as the and individual that and of p. The study of disease is of central importance in the study of HIV/AIDS In the for example, the of HIV HIV/AIDS and of also high HIV/AIDS with of behavioral risk factors at the individual have not observed HIV/AIDS or STD by race/ethnicity individual pathways by which HIV/AIDS in a group complex of economic and social patterns, of illicit and health Social epidemiology is and evidence as to what factors and these HIV/AIDS and found that sexual networks for in sexually transmitted infection in the National Health and Social et al. found that risk network structure in a of injection drug users in differences in HIV prevalence between and and structural factors key in shaping the patterns that to HIV/AIDS For example, the effects of sex from the of men may be relationships and sexual segregation by race/ethnicity also to shape social and risk networks in ways that to disease patterns. residential segregation was and associated with gonorrhea at the in the patterns might be observed for HIV/AIDS. is in the proportion of women among HIV/AIDS cases the for of HIV-positive in through and for of HIV-positive in HIV infections in women are at an and women are and more to HIV infection. theoretical for understanding in HIV/AIDS have been including and human beyond as a risk these approaches structural for in HIV/AIDS. focus has been on women in the of sex or transmission women HIV from their partner and reducing acquisition of HIV among men is key to reducing the spread of HIV to women the of and and than their male partners upon of HIV-positive in of of and HIV risks by women The effects of include individual to HIV and a of to HIV prevention The for and Prevention estimates that one of those with HIV not their HIV and have central in shaping to HIV/AIDS the effects of these social forces on the differential distribution of HIV/AIDS have not been well has been examined at the individual in studies of and and to of and in a that the of to p. et al. have defined as and that are at to have AIDS or HIV and at the and communities with which these are p. and have that a conceptual framework for understanding is our of and to as social that can be in relation to broader of and p. in et al. that HIV transmission and with AIDS the of or structural interventions change the in which and communities HIV infection The effective have been those in which communities have to and by increasing community of HIV Social interventions to and to affect community The for this is found in diffusion which on social networks, opinion and change these are by cultural through the in social networks within communities are essential for and behavior change to to HIV/AIDS structures refer to as well as to the and with their and et al. that laws can affect health in 1) may be a through which social determinants affect health direct and 2) may to social conditions associated with health outcomes indirect of direct effects of on HIV risk are on access to injection which have been associated with HIV incidence of an indirect of structures is the of laws on income which may social conditions that increase HIV vulnerability. key social determinants of HIV/AIDS, including poverty and income and community social change may affect HIV/AIDS patterns through population and and the and structures of Each of these factors may be as modifying between and persons in populations. the HIV infection than populations, of HIV prevalence in the or and may affect HIV transmission studies of have into 1) studies of the spread of HIV along and 2) studies of the that vulnerability to HIV/AIDS can the spread of HIV subtypes and to patterns of and evidence linking patterns and Beyrer et al. (14) found that distinct HIV subtypes were associated with different illicit drug trafficking routes in Southeast Asia. has to the spread of HIV in India, and In addition, studies have identified the importance of in the of for HIV/AIDS is a classic of an health have examined the role of in population HIV/AIDS patterns that might for the effects of on HIV/AIDS patterns include sexual and drug use patterns to in patterns, and of and persons, and on the health sex that men have also been associated with high HIV/AIDS prevalence at the This ecologic is to be by the effects of cultural at the local of the effects might have on and network patterns. the of resources in the public and and the environment a central role in the and of HIV/AIDS epidemics. of importance to HIV/AIDS include health social and illicit drug HIV/AIDS is a high on the health of and policies are to be contributing to increasing HIV/AIDS The complex and between policies and HIV/AIDS are to be policies affect health and by poverty in the distribution of and health have that structural designed to and in have a on the HIV/AIDS by increasing and and reducing in the health and social services as to how policies can be designed to to in HIV/AIDS Structural-level health policies and can to in HIV/AIDS HIV prevention have on individual behavior the of the HIV prevention is with of the for that the social and economic of HIV/AIDS The condom is an exemplary of an effective structural HIV prevention to the environment in which HIV risk behaviors the of may also affect in HIV/AIDS transmission by reducing among HIV-positive Social policies a critical role in the of those to HIV/AIDS, such as or Social policies such as and public affect access to resources and can also affect HIV transmission and access to quantitative research has social change to population health outcomes qualitative research has the importance of social in shaping risk In for example, et al. the effects of the and of which Social and Social on the basis of drug and in this study into the drug of income of and in behaviors in an to HIV to studies of the associations between social policies and HIV/AIDS are to the human of policies of with the of those to HIV/AIDS and to identify drug also has a significant on HIV/AIDS. drug particularly of is HIV epidemics in the The on has focused on to the of which of prevention and measures policies for those trafficking or illicit have in of persons for drug The direct of patterns to be For example, et al. found that on policies such as the of and of those with use from Social were associated with in is a risk factor for HIV risk behaviors have been shown to associated with of and HIV risk of in have access to can increase HIV/AIDS risk and by social and risk networks, or and increasing poverty and social in in risk behaviors in of have been For example, et al. found that the in was associated with among injection drug users in of the of heroin trafficking from In the of the of has also been associated with HIV/AIDS forces are often and of men from In a study for the found that a in the size of the from to as a proportion of population reduce HIV among by 1 to the of in are to reduce HIV in with HIV/AIDS prevention and for The of social epidemiology to the HIV/AIDS have in This is in to a that from a that risk factors for disease to one that the that patterns of disease in p. and methodological in the have this our understanding of multiple causes of risk in modeling geographic and databases linking public health data with on social factors enhance our to and in ways that more closely the of HIV/AIDS epidemics. social epidemiology research in HIV/AIDS determine how we can design more effective of interventions at multiple of social A number of key which of the larger social environment in HIV/AIDS transmission and disease are model complex and be has these include data variables and of analysis at multiple contextual effects, and complex with and p. approaches are for the effective of social-level interventions. persons are with HIV/AIDS, and an HIV infections in effective are high drug and in are to of an HIV more These the to underlying social and structural determinants of HIV/AIDS through policies and This was by from the National Institute on The of the New York of for for and A framework for the social epidemiology of human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome The the the nature of the between of In are between factors at that to observed epidemic patterns. sexually transmitted infection. 2. through which various of structural might influence the risk of human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome of how epidemiology is examined by using different research HIV, human immunodeficiency acquired immunodeficiency In this a is between three approaches to social epidemiology: a a social of disease and an This table is on by Krieger of how epidemiology is examined by using different research HIV, human immunodeficiency acquired immunodeficiency In this a is between three approaches to social epidemiology: a a social of disease and an This table is on by Krieger terms in the equation for the basic reproductive and factors that affect the empirical estimates of the terms in the of HIV, human immunodeficiency acquired immunodeficiency terms in the equation for the basic reproductive and factors that affect the empirical estimates of the terms in the of HIV, human immunodeficiency acquired immunodeficiency keywords used to databases for published literature on the social epidemiology of of Bethesda, HIV, human immunodeficiency acquired immunodeficiency keywords used to databases for published literature on the social epidemiology of of Bethesda, HIV, human immunodeficiency acquired immunodeficiency
Source: Epidemiologic Reviews
Author: K. E. Poundstone
Institution: Johns Hopkins University
Language: English
Subfield: Infectious Diseases
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2160999442