Collective Properties and Healthy Communities
The previous chapter demonstrated the strong connection between personal ties and a variety of health outcomes . A large body of research also documents the association of health-related outcomes with the socioeconomic characteristics of community contexts . For example , the ecological concentration of poverty correlates to infant mortality , low birth weight , child maltreatment , and adolescent violence as well as depression , homicide , victimization , cardiovascular diseases , and all-cause mortality . At the other end of the spectrum , communities with high socioeconomic status appear to promote health in both children and adults . Research indicates that risk and protective factors in the community environment are not solely attributable to individual-level attributes or behavioral lifestyles . Moreover , predisease pathways in individuals may interact with community characteristics in ways that alter the risk for disease .
In this chapter we therefore highlight the collective properties of social environments , alongside rather than in opposition to individual factors and personal ties . Such a focus treats community contexts as important units of analysis in their own right , calling for new measurement strategies as well as methodological frameworks that do not simply treat the social context as a “trait” of the individual . Understanding the causes of variation in collective processes associated with healthy communities may provide opportunities for preventive intervention at lower cost than traditional strategies .
COMMUNITY CONTEXTS AND MULTILEVEL RESEARCH
Social characteristics vary systematically across communities along dimensions of socioeconomic status (e.g., poverty, wealth, occupational attainment), cultural context (e.g., normative guidelines), family structure and life cycle (e.g., female-headed households, child density), residential stability (e.g., home ownership and tenure), and racial/ethnic composition (e.g., racial segregation). A long history of research shows that health-related problems also vary systematically by community, often in conjunction with socioeconomic characteristics (Yen and Syme, 1999). As far back as the 1920s, urban neighborhoods characterized by poverty, residential instability, and dilapidated housing were found to suffer disproportionately higher rates of infant mortality, crime, mental illness, low birth weight, tuberculosis, physical abuse, and other factors detrimental to health (see e.g., Shaw and McKay, 1942).
This general empirical finding continues to the present day, as illustrated by the ecological “comorbidity” or spatial clustering of homicide, infant mortality, low birth weight, accidental injury, and suicide. In the period 1995-1996, for example, data from the city of Chicago reveal that census tracts with high homicide rates tend to be spatially contiguous to other tracts high in homicide. Perhaps more interesting, more than 75 percent of such tracts also contain a high level of clustering for low birth weight and infant mortality and more than 50 percent for accidental injuries (Sampson, forthcoming). Suicide is more distinct, although even here the spatial clustering is significant. The ecological concentration of homicide, low birth weight, infant mortality, and injury indicates that there may be geographic “hot spots” for unhealthy outcomes.
Not only do social characteristics vary systematically with health across communities, a growing body of contextually oriented research has linked community social characteristics with variations in individual-level health. Simply put, even when individual attributes and behaviors are taken into account, there is evidence of direct risk factors linked to environmental context (Robert, 1999a). Recent analyses of the longitudinal Alameda County Health study in northern California, for example, found that self-reported fair/poor health was 70 percent higher for residents of concentrated poverty areas than for residents of nonpoverty areas, independent of age, sex, income, education, smoking status, body mass index, and alcohol consumption (Yen and Kaplan, 1999a). In a related study, the age and sex-adjusted odds for mortality were more than 50 percent higher (odds ratio = 1.58) for residents in areas characterized by poverty and deteriorated housing, after adjusting for income, race/ethnicity, smoking, body mass index, alcohol consumption, and perceived health status (Yen and Kaplan, 1999b). Such patterns are not limited to the United States. A multilevel study in
Sweden found a similar elevated risk of poor health for residents of lower socioeconomic status communities, controlling for age, sex, education, body mass index, smoking, and physical activity (Malmstrom et al., 1999).
Correlational and observational studies suffer well-known weaknesses with respect to making causal inferences. It may be, for example, that individuals with poor health selectively migrate to or are left behind in poor neighborhoods. In the case of individual selection, the correlation of health with community characteristics may be spurious. Experimental and quasi-experimental studies have thus begun to explore community-level effects on health outcomes. One such example is found in the Moving to Opportunity (MTO) program, a series of housing experiments in five cities that randomly assigned housing project residents to one of three groups: an experimental group receiving housing subsidies to move into low-poverty neighborhoods, a group receiving conventional (Section 8) housing assistance, and a control group receiving no special assistance. A study from the Boston MTO site showed that children of mothers in the experimental group had significantly lower prevalence of injuries, asthma attacks, and personal victimization during follow-up. The move to low-poverty neighborhoods also resulted in significant improvements in the general health status and mental health of household heads (Katz et al., 1999). Because the experimental design was used to control individual-level risk factors, a reasonable inference from these studies is that an improvement in community socioeconomic environment leads to better health and behavioral outcomes.
In short, research in social and behavioral science has established a reasonably consistent set of findings relevant to the community context of health:
There is considerable inequality between neighborhoods and local communities along multiple dimensions of socioeconomic status.
A number of health problems tend to cluster together at the neighborhood and larger community levels, including but not limited to violence, low birth weight, infant mortality, child maltreatment, and the risk of premature adult death.
These two phenomena are themselves related; community-level predictors common to many health-related outcomes include concentrated poverty and/or affluence, racial segregation, family disruption, residential instability, and poor-quality housing.
The ecological differentiation of U.S. society by factors such as social class, cultural background, race, and health (see also Chapter 7) is a
robust and apparently increasing occurrence that emerges at multiple levels of geography, whether neighborhoods, local community areas, or even states.
The relationship between concentrated poverty and many health outcomes, especially all-cause mortality, depression, and violence, maintains when controls are introduced for individual-level risk factors. Thus, there appears to be a direct association between the social environment and health, an emerging pattern in experimental studies as well.
Taken together, these findings yield a potentially important clue in thinking about why it is that communities and larger collectivities might matter for health. If multiple and seemingly disparate health outcomes are linked together empirically across communities and are predicted by similar characteristics, there may be common underlying causes or mediating mechanisms. In particular, if “neighborhood effects ” of concentrated poverty on health exist, presumably they stem from social processes that involve collective aspects of neighborhood life, such as social cohesion, spatial diffusion, local support networks, informal social control, and subcultures of violence. Yet we know little about these and other social mechanisms, especially how to measure them at the community level (Mayer and Jencks, 1989; Sampson et al., 1999). Questions about collective properties and mediating social processes pertain equally to observational and experimental studies. For example, what accounts for the apparent improved health among public-housing residents in the Boston MTO experimental group: level of safety, housing quality, social support? Establishing an effect of the environment on health is not tantamount to explaining its biological pathways or its collective-level sources.
An emerging body of research has therefore begun to explore how social processes such as mutual trust among community residents, shared expectations, density of acquaintanceship, reciprocated exchange of information, social control of public spaces, institutional resources, local support networks, and participation in voluntary associations bear on public health outcomes. For example, an index combining informal social control and social cohesion (labeled “collective efficacy”) has been shown to predict rates of violence across more than 300 Chicago neighborhoods. Collective efficacy showed a strong negative relationship with the rate of neighborhood violence, after controlling for poverty, residential stability, immigrant concentration, and a set of individual-level characteristics such as age, sex, socioeconomic status, race/ethnicity and home ownership (Sampson et al., 1997). This finding held up after adjusting for prior levels of neighborhood violence that may have depressed later collective efficacy
because of fear; in this model, a two-standard-deviation elevation in collective efficacy was associated with a 26 percent reduction in the expected homicide rate (Sampson et al., 1997:922). Measures of social cohesion and trust have also been found to predict mortality rates at the state level. The level of distrust (the proportion of residents in each state agreeing that most people cannot be trusted) was strongly correlated in one study (Kawachi et al., 1997) with the age-adjusted mortality rate (r = .79, p < .001). Lower levels of trust were associated with higher rates of most major causes of death, including coronary heart disease, unintentional injury, and cerebrovascular disease. A one-standard-deviation increase in trust was associated with a 9 percent lower level of mortality.
METHODOLOGICAL CHALLENGES AND RESEARCH PRIORITIES
Despite promising leads from existing research, numerous limitations must be addressed if scientific knowledge is to progress. Indeed, methodological issues such as the differential selection of individuals into communities (compositional and selection effects), indirect environmental effects that work through family and peer mechanisms, measurement error, spatial interdependence (e.g., diffusion processes), and simultaneity bias (e.g., poor health causing poverty) represent serious challenges to our ability to draw definitive conclusions on the role of neighborhood and community social contexts. Equally important, there is a need to further develop multilevel methodologies for contextually based research. Health data collected at nested levels of aggregation (e.g., neighborhood, city, state) pose important challenges to the standard analytic procedures used by health researchers. A methodological program of research is thus needed to develop tools for the proper evaluation and analysis of community-level data.
A central challenge in this regard is to build strategies for direct measurement of the social mechanisms and collective properties hypothesized to predict health. As interest in the social sciences turns increasingly to an integrated scientific approach that emphasizes individual factors in social context, a mismatch has arisen in the quality of measures (Raudenbush and Sampson, 1999). Standing behind individual measurements are decades of psychometric and biological research, producing measures that often have excellent statistical properties. In contrast, much less is known about measures of ecological settings. Neighborhood-level research in particular is dominated by the study of poverty and other demographic characteristics drawn from census data or other government statistics that do not provide information on the collective properties of administrative units. We thus recommend a concerted methodological effort to enhance the science of ecological assessment (“ecometrics ”) of social environments relevant to
health. A major component of ecometrics is the development of systematic procedures for directly measuring community social processes, such as in population-based health surveys and systematic social observation of community environments (Sampson et al., 1999; Raudenbush and Sampson, 1999). The latter approach has used videotaping techniques to capture aspects of microcommunity environments (such as street blocks) that bear on health risks (e.g., garbage in the streets, public intoxication, unsafe housing).
A potentially fruitful direction for the National Institutes of Health (NIH) would be to support the systematic collection of benchmark data on social environments that can be compared across communities. An exemplar that might be used as a framework on which to build is the Sustainable Seattle project (Sustainable Seattle, 1999), where some 40 indicators have been collected for use as a benchmark to gauge the progress of Seattle in meeting various goals of public and civic health. An innovative combination of archival records, census data, and surveys characterizes sustainability trends across five basic areas: environment (e.g., air quality), population and resources (e.g., fuel consumption), economy (e.g., housing affordability, poverty), youth and education (e.g., high school graduation, literacy), and community health (e.g., low birth weight, neighborliness). The Leaders Roundtable in Portland, Oregon, has undertaken a similarly ambitious initiative (The Caring Community) that collects data on community health using a combination of focus groups, surveys, key stakeholder interviews, and document reviews (Green, 1999). If agreed measurement standards at the national level could be developed under NIH leadership, communities could use benchmark data to develop early warning signs with respect to changes in the quality of health environments. Ultimately, understanding the dynamics of change in communities themselves and not just the aggregated characteristics of individuals is important for establishing the sources and effects of collective properties that bear on health, in addition to effective policy responses.
In promoting efforts to reach these goals, the institutes are well positioned to take advantage of recent developments that augur well for analytic advances in the behavioral and social sciences. One is the “quasi-experimental” changes now underway across many American cities in public housing, such as the Moving to Opportunity experiments (Katz et al., 1999), voucher programs, and dispersion of concentrated poverty. By integrating econometric strategies for collecting theoretically relevant data on the collective properties of social environments with the random assignment of individuals to new social contexts, researchers are in a better position to sort out selection mechanisms and social causation mechanisms in health outcomes. We thus recommend that NIH support creative efforts to analytically exploit the planned changes that are unfolding in the public
housing arena. A second window of opportunity can be found in new technologies for the mapping and identification of environmental “hot spots.” Geographic information systems (GISs) exploit technological advances in ways that are transforming how research is being conducted in the social sciences. For instance, data on health outcomes can now be linked virtually in real time to address-level data bases on employment, density of liquor stores, mixed land use, and building code violations. A principal advantage of GISs is community profiling and the ability to overlay multiple health-related phenomena (e.g., deaths, cancer clusters, and accident hot spots) in time and space.
INTERACTIONS OF INDIVIDUAL AND COLLECTIVE PROPERTIES
Previous sections of this report discussed the importance of predisease pathways, but there is little research on how such pathways interact with community, environmental, and cultural contexts. Research on the collective properties of healthy communities thus needs to be integrated with the vigorous body of research on individual pathways. For example, do allostatic loads build up faster and in more destructive ways for racial/ethnic or cultural minorities living in some contexts than others? What are these environments? What mechanisms explain person-environment interactions? Are there threshold effects or leveling effects of the environment on disease risk such that individual factors become overwhelmed? What are these thresholds and for what aspects of the environment? As one example, there is evidence that race and income are not significant predictors of disease in areas of concentrated disadvantage (Yen and Syme, 1999). Multilevel studies share a unifying theme in stressing the interaction of individuals and context. Unfortunately, research has yet to systematically link validated measures of community context with the developmental course of predisease pathways and individual-level health outcomes. Although in its infancy, we believe that the multilevel study of developmental and community processes related to health is a crucial research frontier that deserves priority.
OTHER SOCIAL CONTEXTS
To be sure, health environments are not limited to geographical communities. As described elsewhere in this report, families, workplaces, religious institutions, and peer groups generate their own collective properties that bear on health. Many of these factors are in turn influenced by cultural context and background. Nonetheless, strong friendship ties and family social support networks have been found to promote individual health (Berkman and Syme, 1979). Nor are the relevant health environments
limited to urban settings and areas of disadvantage. Most of the U.S. population lives in suburban areas, and the relationship of socioeconomic status and health holds at the upper end of the socioeconomic distribution as well as the lower end (Robert, 1999b). Yet much of the extant research literature is limited to the study of poverty in inner-city communities, underscoring the need to assess suburban and rural contexts. Moreover, there is a need for research on how public policies (e.g., on housing, transportation, and economic development) influence the collective properties of environments. Understanding community social processes requires a simultaneous focus on multiple social contexts and institutional (including governmental) domains.
We recommend a coordinated effort in the institutes to investigate the collective properties of social environments that extant research suggests are promising for a deeper understanding of the etiology of health outcomes and for the development of community-based prevention strategies. Because community contexts are important units of analysis in their own right, they call for concrete strategies that have heretofore been neglected in the institutes. We also underscore the importance of attending to cultural diversity in how healthy communities are defined and realized. Regarding the future, NIH-supported research on healthy communities should include the following kinds of work:
development of a “benchmark” assessment (standardized approach) of the collective health of communities;
selection of and support for longitudinal studies that target data augmentation and multilevel analysis, with a particular focus on personenvironment interactions;
investigation of contextual factors (e.g., cohesion, informal social control, physical disorder, local support networks) as mediators of health or disease outcomes;
design of prevention strategies to promote aggregate-level health by changing social and community environments (e.g., regulation of smoking in public places, taxation policies).
Berkman LF , Syme SL. 1979 “Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents” American Journal of Epidemiology 109/2 : 186-204 .
Green B. 1999 An Evaluation of the Caring Community Initiative of the Leaders Roundtable ( Portland OR : Northwest Professional Consortium, Inc. )
Katz L , Kling J , Liebman J. 1999 “Moving to opportunity in Boston: Early impacts of a housing mobility program” Paper presented at the conference “Neighborhood Effects on Low Income Families,” Joint Center for Poverty Research, Northwestern University/University of Chicago, Chicago IL, September.
Kawachi I , Kennedy BP , Lochner K , Prothrow-Stith D. 1997 “Social capital, income inequality, and mortality” American Journal of Public Health 87/9 : 1491-1498 .
Malmstrom M , Sundquist J , Johansson SE. 1999 “Neighborhood environment and self-reported health status: A multilevel analysis” American Journal of Public Health 89/8 : 1181-1186 .
Mayer SE , Jencks C. 1989 “Growing up in poor neighborhoods: How much does it matter?” Science 243/4897 : 1441-1445 .
Raudenbush SW , Sampson RJ. 1999 “‘Ecometrics': Toward a science of assessing ecological settings, with application to the systematic social observation of neighborhoods” Sociological Methodology 29 : 1-41 .
Robert SA. 1999a “Neighborhood socioeconomic context and adult health. The mediating role of individual health behaviors and psychosocial factors” Annals of the New York Academy of Sciences 896 : 465-468 .
Robert SA. 1999b “Socioeconomic position and health: The independent contribution of community socioeconomic context” Annual Review of Sociology 25 : 489-516 .
Sampson RJ. Forthcoming “How do communities undergird or undermine human development? Relevant contexts and social mechanisms” in Booth A and Crouter AC (eds) , Does It Take a Village? Community Effects on Children, Adolescents, and Families ( Mahwah NJ : Lawrence Erlbaum ).
Sampson RJ , Raudenbush SW , Earls F. 1997 “Neighborhoods and violent crime: A multilevel study of collective efficacy” Science 277/5328 : 918-924 .
Sampson RJ , Morenoff JD , Earls F. 1999 “Beyond social capital: Spatial dynamics of collective efficacy for children.” American Sociological Review64/5 : 633-660 .
Shaw CR , McKay HD. 1942 Juvenile Delinquency and Urban Areas, A Study of Rates of Delinquents in Relation to Differential Characteristics of Local Communities in American Cities ( Chicago : University of Chicago Press ).
Sustainable Seattle . ( Seattle : Sustainable Seattle ).
Yen IH , Kaplan GA. 1999a “Poverty area residence and changes in depression and perceived health status: Evidence from the Alameda County Study” International Journal of Epidemiology 28/1 : 90-94 .
Yen IH , Kaplan GA. 1999b “Neighborhood social environment and risk of death: Multilevel evidence from the Alameda County study” American Journal of Epidemiology149/10 : 898-907 .
Yen IH , Syme SL. 1999 “The social environment and health: A discussion of the epidemiologic literature” Annual Review of Public Health 20 : 287-308 .