The previous chapters of this report focused on health systems and individual and household-level risks that might explain the U.S. health disadvantage, but it has been increasingly recognized that these health determinants cannot be fully understood (or influenced) in isolation from the environmental contexts that shape and sustain them. In contrast with traditional environmental health approaches that focus primarily on toxic substances in air, water, and soil, this more recent approach conceptualizes the environment more broadly to encompass a range of human-made physical and social features that are affected by public policy (Frumkin, 2005). These economic, social, urban or rural, transportation, and other policies that affect the environment were not traditionally thought of as relevant to health policy but are now attracting greater attention because decision makers are beginning to recognize their health implications (Cole and Fielding, 2007).
By definition, environmental factors affect large groups that share common living or working spaces. Thus, they are key candidates as explanatory factors for health differences across geographic areas, such as countries. Indeed, a major motivation for the research on environmental determinants of health has been the repeated observation that many health outcomes are spatially patterned. These patterns are present across countries and across regions within countries, as well as at smaller scales, such as across urban neighborhoods (Center on Human Needs, 2012b; Kawachi and Subramanian, 2007). Strong spatial variation is present for a large range of
health outcomes, including many of the outcomes for which there are cross-national health differences, such as noncommunicable diseases, associated risk factors, injuries, and violence.
Understanding the reasons for the spatial patterns of health within countries may shed light on environmental factors that may contribute to differences across countries. Several factors may explain the strong spatial patterns that are observed within countries. A key contender is the spatial sorting of people based on their socioeconomic position, race, or ethnicity. However, evidence suggests that regional and neighborhood differences in health persist even after adjusting for these socioeconomic and demographic factors (Diez Roux and Mair, 2010; Mair et al., 2008; Paczkowski and Galea, 2010; Pickett and Pearl, 2001). This evidence suggests that broad environmental factors may play an important role in health. Moreover, environmental factors linked to space and place may in turn contribute to and reinforce socioeconomic and racial or ethnic health disparities (Bleich et al., 2012; Laveist et al., 2011). Thus, individual and environmental factors may be part of a reinforcing cycle that creates and perpetuates health differences. These reinforcing processes by which environmental factors and individual-, family-, and community-level factors reinforce each other over time may also play an important role in generating cross-national differences in health.
This chapter focuses on both the physical and social environment in the United States as potential contributors to its health disadvantage relative to other high-income countries. This chapter, like others before it, focuses on three questions:
• Do environmental factors matter to health?
• Are environmental factors worse in the United States than in other high-income countries?
• Do environmental factors explain the U.S. health disadvantage?
Many aspects of the physical and social environment can affect people’s health.1 Spatial contexts linked to regions or neighborhoods are among
1Although analytically distinct, physical and social environments may also influence and reinforce each other: for example, physical features related to walkability may contribute to social norms regarding walking, which may in turn promote more walkable urban designs and community planning.
Physical Environmental Factors
The factors in the physical environment that are important to health include harmful substances, such as air pollution or proximity to toxic sites (the focus of classic environmental epidemiology); access to various health-related resources (e.g., healthy or unhealthy foods, recreational resources, medical care); and community design and the “built environment” (e.g., land use mix, street connectivity, transportation systems).
The environment can affect health through physical exposures, such as air pollution (OECD, 2012b). A large body of work has documented the effects of exposure to particulate matter (solid particles and liquid droplets found in the air) on cardiovascular and respiratory mortality and morbidity (Brook et al., 2010; Laumbach and Kipen, 2012; Mustafić et al., 2012; Tzivian, 2011). Research has identified specific physiologic mechanisms by which these exposures affect inflammatory, autonomic, and vascular processes (Brook et al., 2010; Tzivian, 2011).
The effects of particulate matter on mortality appear to be consistent across countries. For example, a recent review of studies from the late 1990s to mid-2000s found a consistent inverse relationship between airborne particulate matter and birth weight in Australia, Brazil, Canada, France, Italy, the Netherlands, South Korea, the United Kingdom, and the United States (Parker et al., 2011a). Another notable example is the evidence linking lead exposures to cognitive development in children (Bellinger, 2008; Levin et al., 2008). The evidence of environmental effects of air pollution and lead has been reflected in legislation in many countries directed at reducing levels of these pollutants in the environment.
Increasing attention has focused on the implications for health behaviors and social interactions that are created by the built environment. The
2Much early work on the spatial patterns of health used variables such as aggregate summaries of area socioeconomic or race/ethnic composition or measures of residential segregation by various attributes as proxies for a range of broadly defined environmental factors that may be relevant to health (see, e.g., Diez Roux and Mair, 2010). The identification of causal effects using these aggregate summaries raises a number of methodological challenges and does not allow one to identify the specific environmental attributes that may be relevant. More recent work has attempted to identify the specific environmental factors that may be important to specific health outcomes, as well as the pathways through which these factors may operate.
3The environment can also be considered on a larger geographic scale, especially in seeking explanations for cross-national health differences. For example, the health of some nations is affected by their geography or climate.
built environment refers to the presence of (and proximity to) health-relevant resources as well as to aspects of the ways in which neighborhoods are designed and built (including land use patterns, transportation systems, and urban planning and design features). An important example is evidence that links proximity to healthy or unhealthy food stores with dietary behaviors and related chronic disease outcomes (Babey et al., 2008; Larson et al., 2009; Moore et al., 2008; Morland et al., 2006).4 Food availability and food advertising influence energy intake and the nutritional value of foods consumed (Grier and Kumanyika, 2008; Harris et al., 2009; Institute of Medicine, 2006a).
Another large body of work has documented how walking and physical activity levels are affected by access to recreational facilities, land use mix, transportation systems, and urban planning and design (Auchinloss et al., 2008; Diez Roux et al., 2001; Ding et al., 2011; Durand et al., 2011; Gordon-Larsen et al., 2006; Heath, 2009; Kaczynski and Henderson, 2008; McCormack and Shiell, 2011; Transportation Research Board, 2005). Studies conducted in the United States and other high-income countries have found that “walkability” (which is measured by such proxies as building density, land use mix, and street connectivity) predicts walking patterns (Durand et al., 2011; Inoue et al., 2009; Sundquist et al., 2011; Van Dyck et al., 2010). Across countries, studies have also shown that physical activity by children is associated with features of the built environment, including walking-related features, and physical activity resources (Bringolf-Isler et al., 2010; Davison and Lawson, 2006; Galvez et al., 2010; Sallis and Glanz, 2006).5
Although more definitive evidence is needed (see Feng et al., 2010), it has been hypothesized that these environmental features may contribute to the obesity epidemic (Galvez et al., 2010; Papas et al., 2007; Sallis and Glanz, 2009). The importance of residential environments to obesity and related conditions, such as diabetes, was recently highlighted by a randomized housing intervention: low-income participants who were randomly assigned to move into low-poverty areas experienced significant improvements
4Although in the U.S. context a number of studies have reported associations of local access to healthy foods with diet, some studies have not detected such associations (Cummins et al., 2005; Pearce et al., 2008). An important difficulty in comparing results across countries is that the proxy measure for the local food environment is often the type of food stores or restaurants available (such as supermarkets or fast food outlets), but the extent to which these typologies reflect relevant differences in the foods actually available to consumers may differ significantly across countries.
5Studies that compare the effects of built environment features across countries are limited and inconclusive. One recent review found that access to open space (parks and other green spaces) in neighborhoods was associated with physical activity levels in both the United States and Australia (Pearce and Maddison, 2011).
in weight and diabetes indicators (Ludwig et al., 2011). Unfortunately, the study was not designed to identify the specific environmental features responsible for the observed effect.
A range of other physical environmental features have been linked to other health outcomes. For example, the density of alcohol retail outlets has been linked to alcohol-related health complications (Campbell et al., 2009; Popova et al., 2009), including injury and violence (Cunradi et al., 2012; Toomey et al., 2012). Transportation systems and other aspects of physical environments that influence driving behaviors are also related to injury morbidity and mortality (Douglas et al., 2011). Living in socioeconomically disadvantaged neighborhoods (as a proxy for a range of environmental exposures) has been linked to higher rates of injury in both adults and children (Cubbin et al., 2000; Durkin et al., 1994).
Social Environmental Factors
Factors in the social environment that are important to health include those related to safety, violence, and social disorder in general, and more specific factors related to the type, quality, and stability of social connections, including social participation, social cohesion, social capital, and the collective efficacy of the neighborhood (or work) environment (Ahern and Galea, 2011).6 Social participation and integration in the immediate social environment (e.g., school, work, neighborhood) appear to be important to both mental and physical health (DeSilva et al., 2005). What also seems important is the stability of social connections, such as the composition and stability of households7 and the existence of stable and supportive local social environments or neighborhoods in which to live and work.
A network of social relationships is an important source of support and appears to be an important influence on health behaviors. Work on the “transmission” of obesity through social networks has highlighted the possible importance of social norms in shaping many health-related behaviors (Christakis and Fowler, 2007; Hruschka et al., 2011; Kawachi and Berkman, 2000).8 A long tradition of sociological research links these social features not only to illness, but also to risks of violence (Morenoff et al., 2001; Sampson et al., 1997). Social environments may also operate
6Other factors that are also frequently discussed, such as social norms, have been more difficult to study because of a variety of methodological and data challenges.
8Analytical complexities make the isolation of these effects difficult in observational studies.
through effects on drug use, which also has consequences for violence and mental-health-related outcomes.9
Neighborhood conditions can create stress (Cutrona et al., 2006; Do et al., 2011; Merkin et al., 2009), which have biological consequences (see Chapter 6). Features of social environments that may operate as stressors (including perceptions of safety and social disorder) have been linked to mental health, as have factors that could buffer the adverse effects of stress (e.g., social cohesion, social capital) (DeSilva et al., 2005; Mair et al., 2008).
One mechanism through which the social environment can enhance health is through social support. Social support has appeared in many (but not all) studies to buffer the effects of stress (Cohen and Wills, 1985; Matthews and Gallo, 2011; Ozbay et al., 2007, 2008). Resilience to the adverse health effects of stress has also been tied to factors that could influence how one perceives a situation (threat versus challenge) and how one responds to stressors (Harrell et al., 2011; Hennessy et al., 2009; Matthews and Gallo, 2011; Ziersch et al., 2011). One theory for the tendency of some immigrant groups to have better health outcomes than might be expected on the basis of their incomes and education (see Chapter 6) is the social support immigrants often provide one another (Matthews et al., 2010).
Social capital refers to “features of social organization, such as trust, norms, and networks, that can improve the efficiency of society by facilitating coordinated actions” (Putnam, 1993, p. 167). Studies have shown consistent relationships between social capital and self-reported health status, as well as to some measures of mortality (Barefoot et al., 1998; Blakeley et al., 2001; Kawachi, 1999; Kawachi et al., 1997; OECD, 2010c; Schultz et al., 2008; Subramanian et al., 2002). Social capital depends on the ability of people to form and maintain relationships and networks with their neighbors. Characteristics of communities that foster distrust among neighbors, such as neglected properties and criminal activity, can affect both the cohesiveness of neighbors as well as the frequency of poor health outcomes (Center on Human Needs, 2012b).
Spatial Distribution of Environmental Factors
In addition to considering differences between the United States and other countries in the absolute levels of environmental factors, it is also important to consider how these factors are distributed within countries. Levels of residential segregation shape environmental differences across neighborhoods (Reardon and Bischoff, 2011; Subramanian et al., 2005).
9Although findings have not always been consistent, levels of safety, violence, and other social environmental features have also been found to be associated with walking and physical activity (Foster and Giles-Corti, 2008).
Neighborhoods with residents who are mostly low-income or minorities may be less able to advocate for resources and services. Perceptions and stereotypes about area reputation, local demand for products and services, and the purchasing power of residents may also influence the location of health-relevant resources. Physical environmental threats (such as proximity to hazardous sites) may be more prevalent in low-income or minority neighborhoods, a concern of the environmental justice movement (Brulle and Pellow, 2006; Evans and Kantrowitz, 2002; Mohai et al., 2009; Morello-Frosch et al., 2011). These neighborhoods may also lack the social connections and political power that can help remedy adverse conditions.
Other Environmental Considerations
The panel focused its attention on the role of local physical and social environments as potential contributors to the U.S. health disadvantage and did not systematically examine whether other contexts, such as school or work environments, differ substantially across high-income countries. Nor did the panel examine whether neighborhood conditions exert a greater influence on access to health care in the United States than in peer countries. However, these conditions are important to health. For example, the school environments of children, adolescents, and college students can affect diet, physical activity, and the use of alcohol, tobacco, and other drugs (Katz, 2009; Wechsler and Nelson, 2008). Dietary options on cafeteria menus and in vending machines, opportunities for physical activity, and health education curricula are all important to children’s health.
Workplaces have also long been recognized as important determinants of health and health inequalities, occupational safety, and access to preventive services (Anderson et al., 2009; Schulte et al., 2011). Physical working conditions (e.g., exposure to dangerous substances, such as lead, asbestos, mercury), as well as physical demands (e.g., carrying heavy loads), human factors, and ergonomic problems can affect the health and safety of employees. Stressful psychosocial work environments and “job strain”—which refers to high external demands on a worker with low levels of control or rewards—have become recognized as prominent determinants of health and have been linked to self-reported ill health (Stansfeld et al., 1998), adverse mental health outcomes (Clougherty et al., 2010; Low et al., 2010; Stansfeld and Candy, 2006), and markers of chronic disease (Fujishiro et al., 2011). Exposure to job strain exhibits a strong social gradient, which influences inequalities in the health of workers (Bambra, 2011).10
10Findings on job strain have not been consistent, raising the question of whether these are primarily markers of socioeconomic position, which can influence health through other plausible material or psychosocial pathways (Eaker et al., 2004; Greenlund et al., 2010).
Although the panel did not undertake a systematic comparison of workplace conditions in the United States and other countries, it did note that U.S. employees work substantially longer hours than their counterparts in many other high-income countries. In 2005, annual hours worked in the United States were 15 percent higher than the European Union average (OECD, 2008a). Other working conditions and work-related policies for U.S. employees often differ from those of workers in peer countries. For example, U.S. workers have a larger gender gap in earnings, which could potentially affect the health of women, and U.S. workers spend more time commuting to work (OECD, 2012g), which decreases cardiorespiratory fitness (Hoehner et al., 2012). Other important differences in work-related policies include employment protection and unemployment benefits, as well as family and sickness leave (see Chapter 8). However, cross-national comparisons of workplace safety, other occupational health characteristics, labor market patterns, and work-related policies were beyond the scope of the panel’s review.
There is scant literature comparing social and physical environmental features across countries. Here we provide selected examples of the ways in which levels or distributions of physical and social environments relevant to health might differ between the United States and other high-income countries.
Few data are available to make cross-national comparisons of exposure to harmful physical or chemical environmental hazards. There is, for example, little evidence that air pollution is a more severe problem in the United States than in other high-income countries (Baldasano et al., 2003; OECD, 2012a; Parker et al., 2011a).11 Although cross-national comparisons of the volume of emissions and carbon production per gross domestic product show that the United States is a major emitter, this finding does not provide a basis for comparing the cleanliness or healthfulness of air, water, or other resources. The heavy reliance on automobile transportation in the United States is linked to traffic levels, which contribute to air pollution and its health consequences (Brook et al., 2010; Laumbach and Kipen,
11Averages could mask important spatial heterogeneity in air pollution, and this heterogeneity could have important implications for differences in aggregate health if some populations are systematically exposed to high levels of pollution.
2012). Data on population exposures to air pollution across countries are relatively scarce (OECD, 2008b). One available measure is the concentration of particulate matter less than 10 micrometers in diameter (PM-10):12 in the United States, the concentration of PM-10 levels is 19.4 micrograms per cubic meter, lower than the OECD average of 22 micrograms per cubic meter (OECD, 2012a).
An important factor that influences a range of environmental features relates to patterns of land use and transportation. In general, U.S. residential environments are highly dominated by Americans’ reliance on private automobile transportation. This characteristic has promoted dispersed automobile-dependent development patterns (Transportation Research Board, 2009) with consequences for population density, land use mix, and walkability (Richardson, 2004), all of which may have health implications. In 2008, the United States had 800 motor vehicles per 1,000 people compared with 526 in the United Kingdom, 521 in Sweden, 598 in France, and 554 in Germany (World Bank, 2012b). Cities in the United States tend to be less compact and have fewer public transportation and nonmotorized travel options and longer commuting distances than cities in other high-income countries (Richardson and Bae, 2004). Many European countries have strong antisprawl and pro-urban centralization policies that may contribute to environments that encourage walking and physical activity as part of daily life (Richardson and Bae, 2004).13
International comparisons of the social environment are complicated by difficulties in obtaining comparable measures of social environments. For example, aside from their direct links to injury mortality (see Chapter 1), violence and drug use may be indirect markers of social environmental features that affect other health outcomes. As noted in Chapters 1 and 2, homicide rates in the United States are markedly higher than in other rich nations. There are fewer data to compare rates of other crimes across countries. As noted in Chapter 5, certain forms of drug use (which is often linked to other social environmental features) also appear to be more prevalent in the United States than in other high-income countries.
Although Chapter 6 documented a long-standing trend of greater poverty and other social problems in the United States than in peer countries,
12Particulate matter less than 10 micrometers in diameter (PM-10) poses a health concern because it can accumulate in the respiratory system. In particular, particles that are less than 2.5 micrometers in diameter (“fine” particles) are thought to pose the largest health risks (U.S. Environmental Protection Agency, 2007).
13Even in these countries, however, automobile use is rising quickly.
evidence is more limited to compare these countries in terms of social cohesion, social capital, or social participation. For example, OECD data indicate that the United States has the highest prevalence of “pro-social behavior,” defined as volunteering time, donating to charities, and helping strangers (OECD, 2011e). At least one study of cross-national differences in social capital found that the United States ranked at an intermediate level compared with other high-income countries in measures of interpersonal trust; the study also found that the United States ranked higher than many other countries on indicators of membership in organizations (Schyns and Koop, 2010). A previous National Research Council (2011) report and a paper prepared for that study (Banks et al., 2010) did not find much evidence that the United States had unique social networks, social support, or social integration. However, the focus of that paper was on the social isolation of individuals rather than on social cohesion or social capital measured as a group-level construct. Other data indicate that nearly 3 percent of people in the United States report “rarely” or “never” spending time with friends, colleagues, or others in social settings. This figure is one of the lowest in the OECD (2012a).
On another measure, OECD data suggest that levels of trust14 are lower in the United States than the OECD average and than in all peer countries but Portugal, with Nordic countries showing the highest levels (OECD, 2011e). According to the World Gallup Poll, people in the United States are less likely than people in other high-income countries to express confidence in social institutions, and Americans also have the lowest voting participation rates of OECD countries.
In an interesting link between physical and social environments, Putnam (2000) has argued that increasing sprawl could contribute to declining social capital in the United States because suburban commutes leave less time for social interactions. However, it remains unclear whether sprawl helps explain differences in levels of social capital, or health, across countries.
Spatial Distribution of Environmental Factors
Research in the 1990s demonstrated that people of low socioeconomic status were more likely to experience residential segregation in the United States than in some European countries (Sellers, 1999). More
14Trust data are based on the question: “Generally speaking would you say that most people can be trusted or that you need to be very careful in dealing with people?” Data come from two different surveys: the European Social Survey (2008 wave 4) for OECD European countries and the International Social Survey Programme (2007 wave) for non-OECD Europe (OECD, 2011e).
recent evidence also suggests that residential segregation by income and neighborhood disadvantage has been increasing over time in the United States (Reardon and Bischoff, 2011). Given the established correlation between neighborhood, race, and socioeconomic composition and various health-related neighborhood resources in the United States, this greater segregation could also result in greater exposure of some population sectors to harmful environments (Lovasi et al., 2009). Although studies of residential segregation do not directly assess environmental factors, to the extent that segregation is related to differences in exposure to environmental factors, countries with greater segregation may also experience greater spatial inequities in the distribution of environmental factors, resulting in greater health inequalities and possible consequences for overall health status. Studies that use measures of area socioeconomic characteristics as proxies for environmental features have generally reported similar associations of area features with health in both the United States and other countries (van Lenthe et al., 2005), but there is some evidence that area effects may be greater in countries, like the United States, which have relatively greater residential segregation (Moore et al., 2008; Stafford et al., 2004).
At least two studies have suggested that spatial variation in health-related resources may have very different distributions in the United States than in other countries. A review of spatial variability in access to healthy foods found that food deserts—areas with limited proximity to stores that sell healthy foods—were more prevalent in the United States than in other high-income countries (Beaulac et al., 2009). A New Zealand study found that area deprivation was not always consistently associated with lack of community resources (including recreational amenities, shopping, educational and health facilities) (Pearce et al., 2007). This finding is in sharp contrast to studies of the United States, which have found associations between neighborhood socioeconomic disadvantage and the absence of resources that are important to public health (Diez Roux and Mair, 2010).
Large geographic disparities in toxic exposures to environmental hazards and in healthy food access have been repeatedly noted in U.S. communities (Diez Roux and Mair, 2010; Mohai et al., 2009; Pastor et al., 2005). Similar geographic disparities may exist for other environmental features. For example, the distribution of walkable environments may be more variable in the United States than in other countries, creating “unwalkable” islands, where walking is not a viable transportation alternative to driving. These barriers may inhibit physical activity for parts of the population, resulting in worse overall health. Levels of safety and violence may also be more strongly spatially segregated in the United States than in other countries, resulting in areas with greater exposure to violence and its harmful health consequences.
Although no studies have collected the necessary data to determine directly the contribution of the environment to the U.S. health disadvantage, existing evidence on the health effects of environmental factors and on differences in levels and distributions of environmental factors between the United States and other high-income countries suggest that environmental factors could be important contributors to the U.S. health disadvantage. Below we review the possible contributions of the environment to major conditions for which U.S. health disadvantages have been documented.
Obesity, Diabetes, and Cardiovascular Disease
Environmental factors that affect physical activity (primarily through their effect on active life-styles, including walking) and access to healthy foods (rather than calorie-dense foods) may help explain differences in obesity and related conditions between the United States and other high-income countries. As noted above, land use patterns and transportation systems differ starkly between the United States and other high-income countries (Richardson and Bae, 2004; Transportation Research Board, 2009). Transportation behavior also differs between the United States and other high-income countries, with U.S. residents walking and cycling substantially less than Europeans (Bassett et al., 2008; Buehler et al., 2011; Hallal et al., 2012). For example, analyses of comparable travel surveys show that between 2001-2002 and 2008-2009, the proportion of “any walking” was stable in the United States, at 18.5 percent, while it increased in Germany from 36.5 to 42.3 percent. The proportion of “any cycling” was extremely low and stable in the United States, at 1.8 percent, while it increased in Germany from 12.1 to 14.1 percent. There was also less variation in active travel among socioeconomic groups in Germany than in the United States (Buehler et al., 2011). Although the precise effects of these transportation differences on people’s energy expenditure is difficult to quantify, it seems reasonable to expect that different transportation patterns would have important implications for U.S. levels of obesity (Pucher et al., 2010a).
The food intake of the U.S. population is influenced by both supply and demand, particularly food availability, advertising, and other aspects of the way in which meals are socially produced, distributed, and consumed (including mass production and marketing of cheap calorie-dense foods and large portion sizes) (Institute of Medicine, 2006a; Nestle, 2002; Story
et al., 2008).15 In addition, there is evidence that food access is more inequitably distributed in the United States than in other high-income countries (Beaulac et al., 2009; Franco et al., 2008; Moore and Diez Roux, 2006), which may create problems of food access for vulnerable populations.
Importantly, these various features of the physical environment may act synergistically, reinforcing their effects and creating an “obesogenic” environment that affects all U.S. residents, at least to some extent. In addition, these environmental effects may contribute to the development of social norms regarding behaviors and weight (Christakis and Fowler, 2007), which then reinforce certain features of the physical environment, making them increasingly difficult to modify. This reinforcement creates a vicious cycle in which the environment contributes to the development of social norms (such as reliance of automobile transportation) and the behavior resulting from the norm reinforces the environmental features (such as absence of bicycle lanes or public transportation) that sustain it.
The dominant land use and development pattern espoused in the United States for decades (Richardson and Bae, 2004) has created dependence on private automobile transportation, with important implications for traffic volume and associated traffic injuries and fatalities (Transportation Research Board, 2009). Once established, the land use patterns and transportation systems are self-reinforcing and may in turn hinder the development of efficient and inexpensive public transportation alternatives. A physical environment that promotes and incentivizes automobile transportation also reinforces social norms regarding travel, which complicates efforts to modify the patterns. The existing land use patterns and reliance on private automobile transportation not only contribute to traffic volume and injury fatalities, but probably also contribute to physical inactivity, air pollution, and carbon emissions. In this way, a common physical environmental feature may explain the coexistence of the U.S. health disadvantage on apparently unrelated health domains (obesity and injuries).
Homicides, Violence, Drug-Related Deaths, and HIV Risk
Environmental factors, broadly defined, may also contribute to at least part of the U.S. health disadvantage in homicide, violence, and drug-related deaths. As noted above, residential segregation by income in the United States is associated with violence and related outcomes (Sampson et al.,
15Advertising also plays an important role in promoting alcohol and tobacco use (Chuang et al., 2005; Kwate and Meyer, 2009; Mosher, 2011; Primack et al., 2007).
1997; U.S. Department of Justice, 2007). Residential segregation by income and race have also been linked to drug use (Cooper et al., 2007) and HIV/AIDS risk (Poundstone et al., 2004), other contributors to the U.S. health disadvantage. Neighborhood violent crime has in turn been linked to low birth weight (Morenoff, 2003) and childhood asthma (Wright, 2006), two other health conditions that appear to be more common in the United States than in other high-income countries. Residential segregation (and its many social and physical correlates) may be another environmental factor that affects multiple, seemingly unrelated health domains in which the United States has a health disadvantage.
Another important environmental influence on homicide and suicide rates is the ease of access to firearms, which has a strong association with homicide rates (Hepburn and Hemenway, 2004). Legislative policies in other countries limit circulation and ownership of firearms by civilians. As stated in a thorough review by Hepburn and Hemenway (2004, p. 429):
High-income countries outside the United States have much lower rates of handgun ownership than the United States, and the licensing, registration, and safe storage regulations they have make it much harder for known criminals to obtain firearms. Thus, relatively few of the homicides in these countries are firearm homicides.
There is some evidence that environmental factors that could affect the U.S. health disadvantage are worse or are more inequitably distributed in the United States than in other high-income countries. It is plausible to hypothesize that factors in the built environment related to low-density land development and high reliance on automobile transportation; environmental factors related to the wide availability, distribution, and marketing of unhealthy foods; and residential segregation by income and race (with its social and economic correlates) may be important contributors to the U.S. health disadvantage in many domains.
It is noteworthy that these environmental factors may interact with other factors at both “higher” levels of broad social policy and “lower” levels that operate at the individual level. For example, high levels of residential segregation may create large social inequalities across neighborhoods that, in the presence of easy access to guns, may result in high gun violence and homicide rates. Easy access to unhealthy foods may interact with personal sources of stress (e.g., from work) in promoting the consumption of calorie-dense foods. Environments that discourage physical activity may also limit social interactions, with potential implications for violence and drug use.
Environments also help to create and reinforce social norms (Hruschka
et al., 2011) that influence health outcomes. In this way, environmental factors are undoubtedly part of a self-perpetuating cycle that operates across multiple domains, but delineating exactly how this occurs—and how this may differ across place and time—will require further research.
Many of the environmental factors relevant to health are directly amenable to policy. Therefore, identifying which of these factors are important contributors to the U.S. health disadvantage could point to policy interventions that might reduce the disadvantage. For example, cross-national comparisons show that levels of active transportation, such as walking or cycling, can be effectively modified by specific land use and transportation policies (Pucher and Dijkstra, 2003; Pucher et al., 2010b). Although many of the data reviewed in this chapter are highly suggestive of an important role for environmental factors, more empirical evidence is needed to draw definitive conclusions. Important areas for future cross-national research on environmental factors and health include (1) characterizing levels and distributions of environmental risk factors using comparable measures across countries; (2) documenting inequalities in the distribution of these environmental factors; (3) identifying the extent to which these environmental factors affect health and the extent to which their effects are modulated by individual-, community-, or country-level factors; (4) examining directly the contribution of environmental factors to health differences between the United States and other high-income countries; and (5) studying national, regional, and local country policies that may curb levels of adverse environmental exposures, reduce the extent to which they are inequitably distributed, or buffer their effects.
The contribution of environmental factors to the U.S. health disadvantage is likely to result from dynamic and reinforcing relationships between environmental and individual-level factors. Environmental factors also operate over a person’s life course, so that the environments one experiences early in life may influence health trajectories over time. Environmental factors are in turn linked to upstream social and policy determinants. In many ways, the environment can be thought of as the mid- or “meso-” level of influence linking macrolevel factors (e.g., economic and social policy) and microlevel processes (e.g., individual behavior). A comprehensive understanding of the causes of the U.S. health disadvantage will require recognizing how the environment interacts with these other factors and helps perpetuate or mitigate the disadvantage across a broad set of health domains.