Social Environment Notebook
- Economic Status
- Occupational Status
- Educational Status
- Physical Work
- Workplace Social Environment
- Income Inequality
- Residential Segregation
- Physical Environment
- Social Capital
- Measuring Sources of
Stress in the Environment
- Measuring Aspects of the Environment Related to Physical Activity
- Measuring Aspects of the Environment Related to Availability and Accessibility
of Healthy Foods
- Childhood Chaos and Socioeconomic Status
Socioeconomic Status and Health: The Potential Role of Suboptimal
This summary was prepared by Gary W. Evans, Elyse Kantrowitz, and Michelle Schamberg of Cornell University. It was most recently revised in July, 2008.
- Socioeconomic Status and Environmental Quality
- Environmental Quality and Health
Satisfactory explanation for the ubiquitous socioeconomic status-health gradient remains elusive suggesting, in part, that an adequate model of this relation is probably complex and multifaceted. In this chapter we provide an overview of data indicating that income is inversely correlated with exposure to suboptimal environmental conditions. By environmental conditions we mean the physical properties of the ambient and immediate surroundings of children, youth, and families including pollutants, toxins, noise, crowding as well as exposure to settings such as neighborhoods, housing, schools and work environments. We will also briefly cite evidence that each of these environmental factors, in turn, is linked to health. Health is defined in its broadest terms to encompass physical and psychological well being. We also note where there are relevant psychosocial processes known to be directly linked to health. For example some physical environmental conditions affect helplessness and individual beliefs about self-efficacy. These psychological processes are well identified precursors to psychological ill health.
It is important to state clearly at the outset that the implicit conceptual model under discussion is as follows:
Figure 1. Basic underlying conceptual model.
As can be seen above, what we will discuss is evidence for two necessary prerequisites for this model to be valid—namely that SES is associated with environmental quality and, in turn, that environmental quality affects health. This is not equivalent, however, to the conclusion that SES effects on health are caused by differential exposure to environmental quality. There is little if any data directly testing this proposition. What is necessary to verify the model shown in Figure 1 is that the SES ==> health link is mediated by environmental quality.
In addition to this fundamental shortcoming in the extant data base, results on SES and environmental exposure tend to be restricted to income and, in several cases, are not continuous instead comparing individuals below and above the poverty line. Furthermore for certain salient environments, especially work and school settings, scant data are available on income-related differential exposures to hazardous, polluted, or inadequate building conditions. The reader should also bear in mind that for several of the income-related environmental exposure results, the data are confounded with race. Given that there is also evidence that nonwhite individuals, at least in the United States, are more likely to be exposed to health threatening environmental conditions than are white individuals, it can be difficult to disentangle associations between income and environmental quality from environmental racism.
Another more subtle limitation in the data base on SES, environmental quality, and health is the unit of analysis. With the increasing popularity of Geographic Information Systems (GIS), data on both subcomponents of the model depicted in Figure 1, i.e., SES ==> ENVIRONMENTAL QUALITY and ENVIRONMENTAL QUALITY ==> HEALTH are being analyzed at the level of census tracts or higher levels of aggregation because of ready access to clustered information that can then be geocoded. Thus one can, for example, look at some index of pollution at the level of the census tract along with median or average tract income. Similarly one can look at environmental pollutants at the tract level and rates of morbidity and mortality. It is important to remind the reader of the problem of the ecological fallacy.
The ecological fallacy refers to the collection and analysis of data at one level of aggregation and the drawing of conclusions at another level of aggregation. An illustration might prove instructive at this point. Durkheim (1897; 1997) predicted that Catholic countries would have lower rates of suicide than Protestant countries because of greater social integration (cohesion). When he analyzed rates of suicide by the dominant religion of country, he uncovered support for his model. That is, the rates of suicide for predominantly Catholic countries were indeed lower than those for predominantly Protestant countries. However upon closer inspection of the data, one finds that the individuals actually committing suicide were predominantly Catholics living in Protestant countries. The picture is actually, however, more complex. It is possible, in fact even likely, that the variable of religious affiliation is indeed a potent factor in understanding normative behavior when measured at both levels of aggregation: country (or culture) and individual. Neighborhood physical or social characteristics may exert their own, independent effects on individual health and behavior just as individual level characteristics can have potent effects on health or well being. Moreover neighborhood level effects can interact with individual level effects, providing a context wherein the relations between individual environmental exposure and health can be modified (Bronfenbrenner & Morris, 1998). For example the association between housing quality and women's mental health is potentiated by neighborhood level poverty (Kasl et al., 1982).
An overarching problem that plagues much of epidemiology is the difficulty of disentangling environmental from individual based explanations of morbidity or mortality. How does one know that differential exposure to certain environmental conditions as a function of income which, in turn, is associated with some index of morbidity is due to the environment rather than to the person? Many people choose where they live. Individual health, mental, and cognitive status can affect trajectories of exposure to suboptimal environmental conditions. Short of random assignment to environments of varying quality, the best way to get at this issue is with longitudinal rather than cross sectional comparisons. What happens when the same person's exposure to environmental conditions changes over time? Such data are extremely rare in the literature reviewed below with nearly all comparisons occurring cross-sectionally. On the other hand, in the few instances where longitudinal data are available or when other types of selection controls have been incorporated, parallel trends emerge, closely matching cross sectional results.
There is also a conceptual issue we wish to briefly discuss before overviewing some of the evidence for linkages among SES, environmental quality, and health. Nearly all of the empirical work, and for that matter theoretical discussion about this issue, has examined individual environmental risk factors. Research and discussion tends to be focused on specific pollutants, toxins, or particular ambient conditions such as housing quality and each respective factor's link to income or health. We suspect that the potential of environmental exposure to account for the link between SES and health derives from multiple exposures to a plethora of suboptimal environmental conditions. That is, we would argue that a particularly important and salient aspect of reduced income is exposure to the confluence of multiple, suboptimal environmental conditions. The poor are most likely to be exposed to not only the worst air quality, the most noise, the lowest quality housing and schools, etc. but, of particular consequence, the poor are significantly more likely to be exposed to lower quality environments on a wide array of multiple dimensions. We hypothesize that it is the accumulation of exposure to multiple, suboptimal physical conditions rather than any singular environmental exposure, that will provide a fruitful explanation for the SES health gradient. To express it metaphorically, when it rains, it pours. The poorer the individual, the greater the deluge of environmental degradation.
GIS techniques provide a potentially powerful analytic tool to examine this hypothesis in the aggregate, depicting, for example, the overlay of poverty, disease specific morbidity/mortality, and multiple environmental exposure conditions. For a striking and prescient example of the power of this approach see the classic book, Design with nature by the landscape architect Ian Mc Harg (1969). Mc Harg envisioned early on the importance of evaluating proposed changes in land use (e.g., highway construction) on multiple environmental conditions that could, in turn, converge on human health and disproportionately burden the poor and ethnic minority citizens within the United States.
Should this multiple exposure, health, and income hypothesis prove accurate, then current estimates of the importance of suboptimal environmental exposure to explain the SES ==> health link are likely conservative. Nearly all of the available data on environmental health risk, SES, and health emanate from economically developed countries; whereas the greatest convergence of multiple suboptimal environmental characteristics with the severest health threatening consequences occurs in economically underdeveloped countries.
Socioeconomic Status and Environmental Quality
In this section we overview data on the relations between income or SES and exposure to environmental risks. We examine both individual environmental conditions such as toxic wastes, air pollution, crowding, and noise as well as the physical quality of specific settings such as the home, school, work, and neighborhood.
The environmental justice movement, launched in the '80s, called attention to the fact that low income, and especially low income minority individuals, were much more likely to be exposed to toxic wastes and other forms of health threatening environmental conditions relative to their more affluent and white fellow citizens (Institute of Medicine, 1999). An influential book, Dumping in dixie (Bullard, 1990) documented the geographic association of toxic waste dumps in the Southeastern region of the United States with low income, minority neighborhoods. The percentage of families below the Federal Poverty line in census tracts inclusive of EPA Region IV Hazardous Waste Landfills ranged from 26% (South Carolina) to 42% in Alabama. Twenty nine percent of families living within one mile of a commercial hazardous waste facility in Detroit are below the poverty line and 49% of them are nonwhite. More than 1.5 miles away, 10% are poor and 18 percent are people of color (Mohai & Bryant, 1992). One hundred percent of U.S. Government uranium mining and 4 out of the largest 10 coal strip mines are located on Native American reservations (Goldtooth, 1995). Nearly half of Native Americans live below the Federal poverty line. More recent analyses of income and race differentials in hazardous waste exposure reveal similar trends (White, 1998). Children's body lead burden is strongly associated with both income and race. For example in an EPA Task Force report, "Environmental Equity: Reducing Risk for All Communities" (1992) 68% of urban Black children in families with incomes below $6000 had blood lead levels that exceeded safe limits in comparison to 15% of the same population with incomes above $15,000. For white children, the comparable data were 36% and 12%. A more recent report from the EPA illustrates differences in blood lead concentration for young children according to their ethnicity and family income. Black children between the ages of 1 to 5 and living below the federal poverty line had a blood lead concentration of 3.6 µg/dl; in comparison, the blood lead concentration for Black children living 200% above the federal poverty line was substantially lower at 2.2 µg/dl. For Hispanic children, the comparable data were 2.4 µg/dl and 1.6, respectively (Environmental Protection Agency, 2003).Similarly, poor Native American and White children are 2.8 times more likely to exceed current safe levels of blood levels compared to non-poor children from the same geographic areas of rural Oklahoma (Malcoe et al., 2002). The National Health and Nutrition Survey conducted in 1980 and 1990 documents elevated blood lead levels in low income individuals, particularly among inner city residents (Pirkle et al., 1994). As shown in Figure 2, low-income black and white children are more than three times likely as non-poor children to exceed current US safety standards for body lead burden (Dilsworth-Bart & Moore, 2006). These data also illustrate the pernicious and all too common role of environmental racism in America as well. Note that non-poor black children have approximately the same body lead burden as poor white children in this national sample.
Figure 2. Income, race, and body lead burdens from a US national sample. Adapted from Figure 1 in Dilworth-Bart & Moore (2006).
Ambient pollutant exposure reveals similar race and income related trends. Figure 3, for example, depicts exposure to levels of ambient sulfur oxides in the St. Louis metropolitan region in relation to income levels (Freeman, 1972).
Figure 3. Air quality and income levels in St. Louis. Adapted from Table 7.2, Freeman (1972).
These data are particularly interesting to consider in light of the linear SES-Health gradient. Analogous data were found for several other, common ambient air pollutants with known pathogenic effects. Exposure to ozone, a principal toxic component of photochemical smog, as well as fine particulate matter, in the South Coast Air Basin of California, is inversely related to income levels (Brajer & Hall, 1992). See O'Neill et al. (2003) for an interesting summary of linkages between several ambient air pollutants and SES. These authors also review an emerging perspective that not only do exposures to many ambient pollutants increase as one moves down the SES gradient, susceptibility to pollutants may also be graded by SES with heightened vulnerability among those with fewer resources. The World Bank has become interested as well in environmental justice issues, publishing sobering statistical summaries about worldwide environmental health threats. For example in low income cities from the 1970s to the late 1980s the average levels of suspended particulate matter in all cities increased from approximately 300 micrograms per cubic meter of air to 325. Furthermore the total range of measured particulates for all of these cities at both time periods exceeded even marginal let alone acceptable limits from a respiratory health standpoint. Cities in middle income countries over the same time period witnessed improved air quality (from approximately 180 down to 150 micrograms/cubic meter of air) and wealthy countries improved from about 100 to 75 micrograms per cubic meter of air. Analogous data are provided by the World Bank for water quality (World Development Report, 1992).
Death and injury from excessive heat exposure is also not randomly distributed with greater risks typically occurring among the low-income and ethnic minority elderly (Klinenberg, 2002). In an in depth analysis of eight Phoenix neighborhoods during a record heat wave in the summer of 2003, Harlan and colleagues (2006) found higher temperatures and more people at thermal comfort indices indicative of danger in low income neighborhoods. These same neighborhoods were also on average more bereft of shade trees and ground vegetation plus individuals were less likely to have air conditioning.
Although most attention to environmental pollutants and income has been focused on hazardous wastes and air pollution, there are several case studies suggesting higher levels of contaminated water among low income populations (Calderon et al., 1993; World Health Organization, 2006). For example 44% of water supplies for migrant farm workers in North Carolina tested positive for coliform and 26% for fecal coliform. For comparable farm areas in the same region both levels were at 0% (Ciesielski et al., 1991). Low income Chicano populations living along the US/Mexico border (Colonias) are plagued by contaminated drinking water. Estimates indicate, for example, that in Texas nearly 50% of the Colonias population lacks safe drinking water which is largely believed to be the source of the three fold increase in this population's risk for waterborne diseases relative to the overall morbidity rate in Texas (Texas Governor's Border Working Group, 1992). In 1984 EPA surveyed rural drinking water supplies in the US and found significantly higher levels of coliform in low income households. Finally, low SES families are much more likely to swim in polluted beaches (Cabelli & Dufour, 1983) as well as consume fish from contaminated waters (West et al., 1989).
Today, increasing interest has focused on exposure to indoor air quality which may play an even greater role in the respiratory health and well being of individuals, particularly young children. Levels of several common air borne toxin exposures are often higher indoors and, particularly for young children, the duration of exposure to many of these contaminants is greater inside relative to the outdoor environment. Although there are some suggestive data, with the exception of secondary cigarette smoke, very little is known about the association between income levels and exposure to indoor air contaminants.
Parental smoking which is inversely related to income levels increases children's exposures to a wide variety of indoor toxins. For example, in the United States 40.2% of children living in poverty have been exposed to cigarette smoke at home in comparison to 30.6% of those not in poverty (Machlin, Hill, & Liang, 2004). Mothers who are poorer are also less likely to quit and smoke more than their higher income counterparts (Graham, 1995; Groner et al., 1998; Jun, Subramanian, Gortmaker, & Kawachi, 2004). Length of tenure on welfare also predicts maternal smoking prevalence and consumption levels (Graham & Blackburn, 1998). In an interesting study, Jarvis et al. (1992) showed that third grade children's levels of salivary cotinine, a marker of exposure to tobacco, was linearly related to occupational status. Moreover cumulative risk factors associated with poverty increase smoking prevalence in mothers of newborns. Rental occupied housing, lack of higher education, and single parenthood status are associated with a nine fold increase in smoking among mothers of newborns (Spencer & Coe, 2001). This association is independent of mother's age, parity, and ethnicity. Smoking during pregnancy is also highly correlated with maternal education. For example 38% of American women who dropped out of high school smoke during pregnancy compared to 9% going beyond high school and 2% who are college graduates (National Center for Health Statistics, 2001).
In rental units in the United States, 10% percent of households with incomes below the poverty line rely primarily upon hot air units without ducts and 4% use unvented gas heaters as their primary heat source. For rental households with incomes exceeding $30,000 comparable figures are 7% and 1% for ductless hot air heat and unvented gas heaters, respectively (Statistical Universe, 2000). Toxic indoor air pollutants, NO2 and CO, related to combustion processes (stoves, heating, smoking), are substantially higher in low income, inner city residences relative to U.S. averages (Goldstein et al, 1988; Schwab, 1990). Exposure to radon, a known carcinogen, is related to income levels in rural counties in New York state (see Table 1). Chi and Laquatra (1990) suggest that income-related differences in radon exposure are probably related to structural deficiencies that provide more permeable radon vectors for radon to enter into the residence.
Table 1. Radon exposure as a function of household income characteristics. Adapted from Table 5, Chi and Laquatra (1990).
|Percentage of Households Exceeding EPA Safe Limits (4 pCi/L) for Radon|
|Rental||Owner Occupied < $40,000||Owner Occupied > $40,000|
Acute respiratory obstructive diseases such as asthma are associated with serum IgE antibodies to dust mite feces, cats, cockroaches, and certain pollens. Exposure to cockroach allergens as well as antibody sensitivity is associated with socioeconomic status (Sarpong et al., 1996, Rauh et al., 2002). Rosenstreich et al. (1997) found high levels of allergeric reactions to cockroaches in a general population sample of inner city children and more than half of low income asthma patients in several urban, inner city samples evidenced specific IgE antibodies and positive skin test results to cockroaches (Bernton et al., 1972; Kang, 1976). Furthermore dampness in houses which is inversely associated with household income is conducive to dust mites as well as molds and fungi, all related to respiratory obstructive disorders (Gold, 1992). In developing countries, household income can literally mean the difference between life and death. In Uganda for instance the odds of having mosquito bed netting amongst the highest versus the lowest income quartiles is 3.5 to 1 and for more effective nets impregnated with insecticide, greater than 20:1 (Kemble et al., 2006).
Exposure to ambient noise levels are associated with income as well. A nationwide survey of ambient noise in 24 metropolitan sites, exclusive of airports or major highways revealed a strong, consistent relationship between 24 hour ambient, community noise levels and household income (r=-.61) (Environmental Protection Agency, 1977). Families with household incomes below $10,000 were exposed to levels more than 10 decibels higher than families with incomes above $20,000. Note that decibels is a log scale. A 10 decibel increase is perceived as about twice as loud. Data from the American Housing Survey reveals that low income residents are nearly twice as likely (9.1%) to report that neighborhood noise is bothersome in comparison to families not in poverty (5.9%) (Sherman, 1994). A recent analysis of airport noise and children's health and cognitive performance around Heathrow airport documents linkages between income and actual, objective indices of noise exposure. As shown in Table 2, elementary schools with higher levels of aircraft noise exposure have greater percentages of children eligible for free lunches (Haines et al., 2000). Leq is an index of average intensity of sound exposure, measured in decibels. Leq or other weighted indices are typically used to index ambient noise exposure.
Table 2. Aircraft noise exposure and elementary school poverty index. Adapted from Table 1 in Haines et al. (2000).
|Low Noise||Moderate Noise||High Noise|
|<57 Leq||57-63 Leq||64-72 Leq|
|% Eligible free lunch||14||23||28|
We have also measured indoor average noise levels in rural families living in upstate New York. The average Leq for families living at or below the poverty line is 65 and for middle class families (those living two to four times the poverty line), the average Leq is 61.
Residential crowding which is typically indexed by the ratio of people to number of rooms, is also linked to income. Figure 4 depicts national data from the 1990 census, showing a clear income-related gradient (Myers et al., 1996). The official U.S. Census definition of a crowded household is greater than one person per room.
Figure 4. Residential crowding (greater than one person per room) and household income in the United States. Adapted from Table 1, Myers et al. (1996).
We have also examined residential crowding in our work on rural poverty in upstate New York. The mean level of people per room in families living at or below the poverty line in our rural upstate New York sample is .68. The mean density level for middle income families is .50.
The quantity and quality of space proximate to residences may also bear upon health and quality of life. Low income neighborhoods in New York city have 17 square yards of park space per child whereas all other New York city neighborhoods average 40 square yards of park space per child (Sherman, 1994). In the United Kingdom, 86% of professionals and supervisors have access to a private garden at home in comparison to 69% of manual laborers (Townsend, 1979). Manual laborers are four times more likely (14%) to have a garden or yard at home too small to sit outside in the sun relative to professionals, managers, or supervisors.
In addition to examining linkages between constituents of environmental quality and SES, one can also look at bundles of environmental quality as embodied in the overall quality of settings such as housing, schools, work, or neighborhoods. Housing quality is strongly tied to income levels in the United States. Income levels are positively associated with home ownership and negatively correlated with residential mobility (Federman et al., 1996). For example approximately three quarters of those above the U.S. Federal poverty line own their own home compared with 40% of those who are poor. Low income families are five times more likely to be evicted than their non-poor counterparts. Statistics from the American Housing Survey, conducted by the U.S. census, reveal that the poor are more than three times as likely to have substandard quality housing than the not poor (22% vs 7%) (Sherman, 1994). Thirty six percent of all American households with a child under the age of 18 report at least one problem with housing compared to 77% of those households at or below 50% of the median income for the surrounding geographic area (Trends in the Well Being of America's Children and Youth, 2000). Poor children residing in older housing are often at an increased risk for injury. For example, for every 10% increase in the proportion of residential housing built prior to 1950, risk for falling increases by 17% and risk for being burned increases by 34% (Shenassa et al., 2004). In low-income countries, children are disproportionately exposed to poor living conditions and are at least five times more likely to suffer injury than children living in wealthier countries (Bartlett, 2002). As is evident in Table 3, income is inversely related to various indicators of housing adequacy.
Table 3. Percentage of children living in houses with selected problems from the 1985-1989 American Housing Survey. Adapted from Table 4.6, (Mayer, 1997).
|Income Decile||Income Decile|
|No sewer/septic system||1.7||.9||.1||.0|
|No central heat||32.3||34.7||21.4||9.6|
|Holes in floor||7.0||5.8||1.4||.6|
|Open cracks (walls, ceiling)||19.9||15.9||6.3||3.2|
|≥ 1 person/room||19.2||23.4||10.9||5.3|
Analogous trends have been uncovered in a representative national sample of households in the United Kingdom (Townsend, 1979). We have also found that housing quality is significantly correlated with the income to needs ratios (r=-.39) of rural families in upstate New York. The income to needs ratio is a per capita poverty index formed by taking the ratio of family income to the Federally defined poverty index. Thus an income to needs ratio of one equals the poverty line. The federal formula is adjusted annually to the cost of living index and is derived with a focus on financial needs for food and shelter. We used a housing composite scale that relied on raters' assessments of cleanliness/clutter, indoor climate quality, privacy, exposure to safety hazards, and structural quality (Evans et al., 2000). Social class differentials in childhood injuries from accidents in the home (e.g. falls) are correlated with hazardous characteristics of residential structures (Blane et al., 1997).
Poor families are also much less likely to have basic amenities such as clothes washers (72%), clothes dryers (50%), air conditioning (50%) or telephone (77%) than the not poor (clothes washer 93%, clothes dryer 87%, air conditioning, 72%, telephone, 97%) (Federman et al., 1996; Mayer; 1998). In the Netherlands, the percentage of persons with one or more housing deficiencies (no refrigerator, no washing machine, no clothes dryer, ≥ one person/room) is linearly related to income, ranging from 16% for families in the lowest sixth of income to 1% of those in the highest sixth of income (Stronks et al., 1998).
Developmental psychologists have developed rating instruments to assess psychosocial dimensions of the home environment of families. These instruments encompass both measurements of physical qualities as well as evaluations of parenting and other aspects of the social environment. Figure 5 depicts a linear relation between income to needs ratio and scores on a common residential environment rating scale, the HOME (Garrett et al., 1994).
Figure 5. HOME scale values and income, adapted from Figure 2, Garrett et al. (1994). Permission to reprint portions of this Figure from the University of Chicago.
Bradley and Caldwell (1984), the principal authors of the HOME scale reported the lower the SES, the poorer the HOME scores for infants and two year olds. Specific subscales of the HOME reveal that SES was most strongly linked to the provision of appropriate places for play and the degree of structure and organization in the residence. In another analysis using the HOME scale, 6-9 year olds in families below the poverty line suffered a 34% deficit in overall HOME scores relative to those in families with an income to needs ratio above four (Miller & Davis, 1997). Moreover, the longer the duration of childhood poverty, the stronger the negative association. Dubow and Ippolito (1994) found a correlation of -.54 between HOME scores and the number of years elementary school aged children lived below the poverty line.
Sherman (1994) provides a sobering statistic that may be indicative of the quality of the home environment available to children. Fifty nine percent of children ages 3-5 who are poor have 10 or more books at home; 81% of children who are not poor have 10 or more books at home. Sadly only 38% of low income parents in America read on a daily basis to their preschoolers. Although substantially higher, the figure for their more affluent counterparts, 58% is also dismal (Trends in the Well Being of America's Children and Youth, 2000). According to the National Survey of America's Families (NSAF), 45% of parents with earnings below $32,000 read to their children twice per week or less. Comparatively, only 10% of parents earning $69,000 report the same level of decreased involvement with their children (Turner & Kaye, 2006). Not surprisingly, the higher the socioeconomic status of the family, the more time youth spend reading on a daily basis (Larson & Velma, 1999). An interesting companion statistic that may inter-relate to reading activity is television watching. Numerous studies have documented an inverse relationship between household SES and youth TV viewing (Larson & Verma, 1999). For example, the percentage of 13-year-olds in the United States who watch more than five hours of television is 18 and 10 with household heads who did not graduate from high school or are college graduates, respectively (Trends in the Well Being of America's Children and Youth, 2000).
In 1998 ninety four percent of American urban children in predominantly low income (≥ 40% below poverty line) versus 57% of urban children living in neighborhoods with little poverty (< 10% below poverty line) had no internet access (Kids Count Data Book, 2000). Eighty four percent of the former households and 35% of the latter had no access to a computer. Across the entire United States, 52 and 15% of elementary and secondary school children, respectively who are in the bottom income quintile have computer access at home. This contrasts markedly with the 74 and 79% of elementary and secondary children, respectively in the highest income quintile who have home computer access.
An important setting for children are schools and daycare environments. The quality of the school environment is tied to income. Per capita school expenditures vary greatly according to community resources given the reliance of many school districts on local property taxes. In 1999 the Federal government surveyed a representative sample of 903 public elementary and secondary schools in the United States about their physical facilities (National Center for Education Statistics, 2000). One in five schools had a building in less than adequate repair, 43% had at least one infrastructure deficiency (e.g., heating, indoor air quality) and about 10% were seriously overcrowded (greater than 125% capacity). Not surprisingly as shown in Figure 6, predominantly low income schools suffered a disproportionate burden of inadequate school facilities.
Figure 6. The percentage of inadequate original buildings and permanent additions in relation to the percentage of children in the school eligible for free or subsidized lunches. Adapted from Table 2, page 12 National Center for Education Statistics (2000).
Table 4 provides summary data from the National Center for Education Statistics report on the Condition of America's Public School Facilities: 1999. As is apparent on every dimension, low income schools fare worst. Moreover on several indices of facility quality, there appear to be linear gradients in relation to income levels for the school.
Table 4. Percentage of building components inadequate in relation to percentage of children in the school eligible for free or subsidized lunch. Adapted from Tables 4 and 8 of the National Center for Education Statistics (2000).
|% Eligible Children||Roof||Plumbing||Heating||Electric Power||Lighting||Ventilation||Indoor Air Quality||Acoustics||Physical Security|
The percentage of schools seriously overcrowded in the intermediate ranges of income, 20-39% school lunch eligible and 40-69% school lunch eligible, are eight and seven percent, respectively.
Children in schools with a larger proportion of poor children are also more likely to be crowded. Twelve percent of American public schools with more than 70% of their children eligible for subsidized or free lunch programs are above 125% of building capacity in comparison to 6% of schools with less than 20% eligible for lunch programs (National Center for Education Statistics, 2000). In terms of health outcomes it is important to recall that low income children are also more likely to live in seriously overcrowded households, defined as more than one person per room (see Figure 3 above). The adverse impacts of residential crowding are exacerbated among children in more crowded daycare facilities (Maxwell, 1996).
It is, of course, difficult to disentangle the quality of the physical plant from the social environment of schools. Perhaps the most fundamental resource in a school is the quality of its teachers. Secondary teachers in low income schools are significantly less likely to have undergraduate majors or minors in the subjects they teach relative to those in more affluent schools. For example 27% of secondary math teachers in poor school districts majored or minored in mathematics in college compared to 43% in school districts that are not predominantly low income (Ingersoll, 1999). Comparable differences occur in the sciences whereas the differential in English is smaller.
School safety is associated with income as well. Blue collar adolescents are twice as likely to report the presence of weapons at school (12%) or fighting in school (32%) than their white collar counterparts (Gallup, 1993).
Recently several authors have examined the quality of daycare in relation to income levels. The ratio of daycare staff to children as well as expenditure is related to income levels (NICHD Early Child Care Research Network, 1997; Phillips et al., 1994). The educational level and pay scales of child care workers are related to income as well (Phillips et al., 1994). Both of these studies suggest that for the very poor subsidies appear to offset daycare quality relative to the lower middle and working class for institutional daycare center care. For home care, the more typical income-quality gradient is seen, with poorer quality home daycare associated with reduced family income. Phillips and colleagues have also documented that the quality of childcare provider-child interaction (e.g., sensitive, harshness, detachment) is also correlated with income levels.
Outside of home and school, poorer people may be subject to greater health risks on the job. In a large Swedish sample of workers, Lundberg (1991) assessed different environmental and behavioral factors believed to account for SES gradients in health. Of particular interest, the strongest predictor of the gradient was poor working conditions, defined as a heavy lifting or tasks with repetitive strain plus daily contact with toxins, fumes, dust, explosives, vibration and the like. Furthermore, in multiple regression models, poor working conditions was the only independent (i.e., entered last after all other factors) predictor of the SES health gradient. An American study of 14 blue collar worksites throughout Boston uncovered evidence of income differences even within occupational class in high level exposures to ergonomic hazards and some ambient exposures (chemical but not dust or noise) (Quinn et al., 2007). Comparing multiple exposures (> 1), 48% of workers making less than $10.54 hourly compared to 45% of those with wages above that were exposed to 3+ high levels of risk exposure. Note that the Quinn et al findings are within a narrow band of income since all of the individuals were blue collar workers within four different industries. Emerging evidence documents pervasive race differentials in occupational exposure to toxins and physically hazardous, risky working conditions (Frumkin & Walker, 1998; Lucas, 1974; Wright, 1992). For example steel workers located in the most hazardous component of the production process (topside of the coke ovens) are nearly three times more likely to be black than white. Among the most notoriously unhealthy labor sectors are seasonal agricultural work and sweatshop garment production—settings predominated by low income workers. Quinn and colleagues (2007) showed however that racial disparities in occupational risk exposures were more muted when considered within the same spectrum of job types. Moses et al. (1993) review several studies suggested a greater body burden of persistent chlorinated hydrocarbons among low income, Chicano/Latino and black agricultural workers. Although these substances are now banned in the U.S. because they are lipophilic, they remain sequestered in fatty tissue for many years. DDT serum levels are related to SES among blacks and whites in Dade County, Florida (Davies et al., 1972). Table 5 depicts DDE levels in blood among adult blacks and white from Davies et al., 1972). DDE is a major metabolite of DDT and more indicative of life long exposure. Two aspects of these data are noteworthy. First the data reveal a nearly perfect linear SES gradient, and second African-Americans suffer much higher body pesticide burdens. DDT concentrations in human breast milk among indigent black women in rural counties in Mississippi and Arkansas averaged 447 ppb. Average levels for middle class women in Nashville averaged 14 ppb (Woodard et al., 1976). In the National Health and Nutritional Examination Survey II, conducted from 1976-1980, living below the federal poverty line had a significant, independent association with serum DDT (odds ratio=1.48) and dieldrin (odds ratio=1.43) levels (Stehr-Green, 1989).
Table 5. DDE (DDT metabolite) serum ppb in relation to SES in Dade County, Florida. Adapted from Table 5 (Davies et al., 1972).
|Social Class (Holllingshead 2 factor index)|
Given the robust association of race and income among American workers, it is reasonable to suspect that differential income-work setting quality relations exist as has been documented with respect to race. We do know with some certainty that work-related injuries are inversely related to wages. Moreover injury caused sick days and duration of sick days per injury are both inversely associated with wages (Hamermesh, 1999). As shown in Table 6, occupational status in a large, representative sample of workers in the United Kingdom (Townsend, 1979) is inversely related to more difficult working conditions.
Table 6. Percentage of UK men with difficult working conditions as a function of occupational status. Adapted from Table 12.2 in Townsend (1979).
|On feet all of the time||2||16||28||79|
|Work before 8 or at night||15||19||18||50|
Berney and colleagues (2000) asked elderly individuals (M=67.9 years) to retrospectively report the number of years they had been exposed to various environmental hazards, including at work. Exposure to combined occupational hazards (i.e., fumes and dust, physically arduous tasks, lack of job autonomy) was inversely related to class. For example male manual laborers had more than double the number of years working in hazardous conditions (M=51.1 years) than non-manual laborers (M=20.9 years). Combined occupational hazards are expressed in years, cumulatively across hazards. Thus for example an individual exposed to 10 years of dust, 5 years in arduous labor and 20 years in a job with low autonomy would be assigned a score of 35 hazardous years. Table 7 illustrates some additional analyses of these data, focusing on dust/fumes and arduous task demands as a function of occupational class.
Table 7. Age-adjusted mean number of hazard years (males). This table is based on previously unpublished data collected by L. Berney, D. Blane, G. Davey Smith and P. Holland. This research was funded by the U.K. Economic and Social Research Council (Grant # L128251003).
|British Registrar General Social Class|
|I||II||II Non-manual||III Manual||IV||V|
Townsend (1979) in his report of occupational class and working conditions in the United Kingdom developed a composite index of working amenities that included sufficient heat in the winter for those outside, availability of tea/coffee, indoor toilet, facilities for washing/changing clothes, place to buy or eat lunch, secure place to keep coat/spare clothing, lockable personal storage, first aid kit/facilities, possibility to make at least one call daily, and control over task lighting. He then constructed summary Working Conditions based on the number of amenities available: Very poor working conditions—Less than four amenities, Poor working conditions—between four and six amenities, Adequate working conditions -six amenities, and Good working conditions consisted of more than six amenities. Table 8 depicts data from this study on men in the UK working under different levels of overall work quality as a function of occupational status.
Table 8. Occupational status and percentage of men in the UK working at different levels of overall work setting quality. Adapted from Table A.41 in Townsend (1979).
|Overall Work Setting Quality|
Stressful, psychosocial conditions of working settings also appear related to occupational status. Marmot and colleagues (1999) have shown among British Civil Servants that grade level is inversely related to autonomy (decision latitude) on the job, monotonous working conditions, and work pace. The trends are linear in relation to Civil Service Grade (1-6) and, in turn, are related to sickness absence and incidence of coronary heart disease.
In addition to school, work, and home, local surroundings may contribute to health and well being. Low income urban neighborhoods suffer poorer basic municipal services (e.g., police, fire sanitation (Wallace & Wallace, 1998); are less able to get problems in their neighborhoods addressed (Altschuler et al., 2004), and experience greater residential mobility (Leventhal & Brooks-Gunn, 2000) relative to more affluent, urban neighborhoods. Nine to eleven year old children in Sydney Australia rated their overall neighborhood quality as higher in relation to an objective composite index of neighborhood risk, based upon census data (Homel & Burns, 1987). A primary component of this neighborhood risk index was SES. The higher the neighborhood risk index, the more likely it was that children rated their setting as having too much traffic, being dirty and polluted, too much noise, no safe places to play, and having fewer parks and outdoor play spaces. Even within predominantly low income areas, family income is positively related to the overall quality of neighborhood housing and other amenities (Spencer et al., 1997). Not surprisingly, rates of pediatric injuries are higher in lower income families on the order of about 30% (Haynes et al., 2003). Pedestrian accidents occur more often among low income children (Macpherson et al., 1998; Roberts & Power, 1996) which is likely due in part to higher levels of street traffic in low income neighborhoods (Macpherson et al., 1998).
Macintyre and colleagues (1993) found that working class areas of Glasgow, Scotland in comparison to upper middle class sections had fewer shops, paid more for food, had dramatically fewer recreational opportunities, were further from mass transit stops in combination with lower rates of car ownership, and had poorer street cleaning and maintenance. Steptoe and Feldman (2001) found similar data in London across a wider SES spectrum. The lower neighborhood SES the more physical (e.g., litter, noise, air pollution, traffic) and social (e.g., fear of crime, vandalism, insufficient recreation opportunities, insufficient shopping options) problems present. As noted earlier (Sherman, 1994), low income children have less access to parks and suitable nearby nature (e.g. gardens). Furthermore as shown by the Sydney study, children seem well aware of this. Playgrounds in low income areas are more hazardous (as assessed by independent, trained raters) relative to those in higher income neighborhoods (Suecoff et al., 1999). Moreover young children of low income families are much more likely to have no safe play areas nearby their home (Townsend, 1979). Furthermore, basic housing stock is of significantly lower quality (percentage dilapidated housing) in low income neighborhoods than in middle or upper income areas (Joint Center for Housing Studies at Harvard University, 1999). Abandoned lots and boarded up houses and other structures also occur more frequently in low income areas (Taylor & Harrell, 1999; Wandersman & Nation, 1998).
Rates of exposure to crime are strongly tied to family income levels as well as neighborhood income composition (Sampson et al., 1997). Children from low SES neighborhoods are more likely to be exposed to aggressive peers than children from higher income areas (Sinclair et al., 1994). Low income adolescents perceive their neighborhood as more dangerous, violent, and of poorer overall quality (graffiti, cleanliness, housing quality) than their middle class counterparts (Aneshensel & Sucoff, 1996). Homel and Burns (1994) in their Sydney neighborhood study also found that neighborhood risk was linearly related to young children's judgments about the presence of unfriendly people Thus both the immediate residential environment as well as the neighborhood infrastructure of low income individuals are likely to be of lower overall quality than the home or surroundings of people with more financial resources.
Environmental Quality and Health
The section above documents pervasive income related differences in exposure to environmental risks. The present section provides a much briefer summarization of evidence that the disproportionate burden of suboptimal environmental exposure shared by those who are poorer could have health consequences. The amount and quality of research on environmental effects on health and well being are substantially greater than evidence of income differentials in exposure to poor environmental quality.
A voluminous literature relying on epidemiological studies as well as human and animal experiments demonstrates that ambient air pollutants cause various respiratory problems including bronchitis, emphysema and asthma. Less well documented links exist between certain ambient pollutants and lung cancer. Exposure to carbon monoxide may also be a risk factor for coronary heart disease. In addition, ambient air pollution may increase risk for respiratory infection (Holgate et al., 1999; Lippman, 1992; National Research Council, 1991). Exposure to ambient pollutants, principally ozone, a toxic component of photochemical smog, has been linked to psychological distress, negative emotional affect, and behaviors including interpersonal attraction and aggression. The latter function appears to be curvilinear with moderate levels of irritable pollutants causing increased aggression (Evans, 1994; Rotton, 1983). Although a relatively new area of inquiry, there is already an impressive body of literature linking indoor air quality, including environmental tobacco smoke, with various respiratory illnesses (Bardana & Montanaro, 1997; Institute of Medicine, 2000; Samet & Sprngler, 1991).
Environmental toxins, principally heavy metals (e.g., lead), solvents (e.g., cleaning fluids), and pesticides, occur in hazardous waste disposal facilities and various manufacturing, mining, and agricultural activities. Toxicological impacts include cancer, respiratory morbidity, brain damage, and various neurotoxicological difficulties (National Research Council, 1991; Johnson, 1990; Scott, 1990; Morello-Frosch et al., 2002). In utero exposure to several toxins also produce teratological effects. Many of these same toxins in much lower doses produce cognitive and behavioral abnormalities including attentional and memory disorders, lower IQ, and poorer academic achievement. It is estimated that for each incremental increase (from 1-10 µg/dl) of lead in the blood, there is a 1.4 point drop in a child's IQ (Canfield et al., 2003). Behavioral problems including impulse control, frustration intolerance, and aggression have been associated with several toxins as well (Araki, 1994; Riley & Vorhees, 1991). The low dose behavioral toxicological effects appear to be especially dangerous during the critical period of fetal development.
Another aspect of environmental quality, ambient noise levels, also appears to threaten health. Links between chronic noise exposure and hearing damage are well documented (Kryter, 1994). Both intensity and duration of exposure are important parameters of noise exposure and health. Suggestive data link noise exposure to coronary heart disease and hypertension, but the evidence is not solid (Berglund & Lindvall, 1995; Thompson, 1993). Several community studies have shown that children's blood pressure and possibly neuroendocrine stress hormones are elevated when living or attending schools in the flight paths of major airports (Evans, 2001). There are contradictory findings on ambient noise exposure and prematurity and birth defects, as well as a small number of studies suggesting immunosuppression from noise in animal models (Evans, 2001).
Noise clearly interferes with complex task performance (e.g., dual tasks) but has inconsistent impacts on simple tasks (e.g., vigilance) (Evans & Cohen, 1987). Several studies have uncovered evidence that both acute as well as chronic noise exposure can lead to motivational deficits linked to learned helplessness (Cohen, 1980; Evans, 2001). Glass and Singer (1972) in a famous set of studies found, for example, that immediately following exposure to 20 minutes of noxious noise in the laboratory, subjects were less likely to persist at challenging puzzles. Their data also indicate that it is the uncontrollability of noise, in particular, that is problematic for motivation. It is perhaps noteworthy that the initial learned helplessness experiment with humans used uncontrollable noise to induce helplessness (Hiroto, 1974). A large number of studies have shown that chronic noise exposure is linked to reading deficits in young children. The effects on reading are not due to hearing loss. Moreover some of this effect is due to problems with speech perception in noise-exposed children (Evans & Lepore, 1993). Noise also has adverse consequences for interpersonal processes including altruism and aggression (Cohen & Spacapan, 1984). Conclusions about an association between ambient noise exposure and mental illness are not well substantiated (Stansfeld, 1993).
Crowding, like noise, functions as a stressor, elevating blood pressure and neuroendocrine parameters (Evans, 2001). Several studies have indicated that infectious diseases are more likely in relation to crowding among vulnerable subgroups (e.g., prisoners, refugee camps) and that residential crowding (i.e., people per room) is associated with psychological distress in the general population (Evans, 2001). There is no evidence to substantiate the widespread perception of cultural differences in tolerance for crowding (Evans et al., 2000). Areal indices of density (e.g. people/acre) appear less important than interior density measures such a people per room for understanding health outcomes associated with crowding. Several studies indicate that a principal pathway linking residential crowding to psychological distress is problems with unwanted social interaction (Baum & Paulus, 1987; Evans, 2001). Residents of more crowded homes are more socially withdrawn and perceive lower levels of social support in comparison to individuals living in less crowded settings. Parents in crowded homes are also less responsive to their children and tend to employ harsher, more punitive parenting styles (Evans, 2001). Crowding may also interfere with complex task performance and has been linked to learned helplessness (Evans, 2001; Evans & Cohen, 1987). Relations between crowding and aggression are unclear but several studies have indicated reduced altruism and more negative interpersonal interactions in more crowded settings (Baum & Paulus, 1987).
Concerns about housing quality and physical health are a longstanding interest within the field of public health. Because of the design of research projects investigating housing and health it is extremely difficult to draw definitive conclusions; nonetheless the preponderance of evidence suggests that substandard and more hazardous construction is associated with more unintentional injuries, especially among young children and the elderly. Inadequate heating systems, and the presence of dampness, molds and other allergens are also associated with poor respiratory health (Burridge & Ormandy, 1993; Ineichen, 1993; Matte & Jacobs, 2000). Epidemic increases in asthma in inner city settings may be partially attributable to elevated ambient pollutants along with exposure to allergens in the home. It is worth noting that the evidence linking housing and health includes several longitudinal analyses of housing improvements and at least one study with random assignment.
Work investigating a possible link between housing quality and mental health is more controversial. The findings are less numerous and consistent than the physical health research. It would be fair to conclude that evidence suggests that high rise housing may be linked to elevated psychological distress among low income women with young children as well as with restricted, outdoor play activities in young children (Evans, Wells & Moch, 2000; Gifford, in press).
There is also a good deal of evidence showing relations between the design of public housing and both fear of crime and actual incidence of crime (Taylor & Harrell, 1999). One of the problems with research on mental health and housing is reliance on housing measurements developed originally for the purpose of assessment of physical health. Recent work indicates that scales indexing behaviorally relevant aspects of housing may prove more fruitful in research on housing and psychological well being (Evans et al., 2000).
The quality of the home environment has also been linked to children's cognitive development. The provision of adequate learning materials and the absence of chaotic conditions have been shown to predict better achievement, both cross sectionally and longitudinally (Bradley, 2000; Matheny et al., 1995; Wachs & Gruen, 1982). The role of structure and predictability in family routines has also been implicated in children's socioemotional development (Fiese & Kline, 1993). Research on the quality of the physical environment of daycare settings and early school environments and children's development is not sufficiently developed to draw definitive conclusions, but trends indicate that the physical environment may play a role directly impacting children's cognitive and social development and indirectly by way of changes in teachers' behaviors (Moore & Lackney, 1993; Transik & Evans, 1995). Some of the physical characteristics of schools, in addition to noise and crowding, believed to be important to cognitive development include structure and predictability, arrangement and quality of activity areas, degree of openness, privacy, access to nature, availability and variety of age appropriate toys and learning aids, and play materials for fine and gross motor development that provide graduated challenge and natural light (Weinstein & David, 1987).
There has been a recent upsurge of interest in neighborhood effects on well being, focusing on cardiovascular health, crime and violence, and children's development. The more recent work reflects sensitivity to the problem of the ecological fallacy described earlier (drawing conclusions at one level of analysis with data at another level). Some of these studies look at neighborhood effects, after statistically controlling for individual variation in, for example, SES or income levels. Other studies employ multiple level modeling techniques that account for both individual and areal level variation in SES or income. Low SES neighborhood characteristics, independent of household SES, are associated with higher all cause mortality (Davey Smith et al., 1998; Haan, Kaplan & Camacho, 1987), greater cardiovascular risk in men (Davey Smith et al., 1998; Harburg et al., 1973), as well as women (Davey Smith et al. 1998); cardiovascular disease in men and women (Davey-Smith et al., 1998; LeClere et al., 1998) and with injury mortality (Cubbin et al., 2000). As noted earlier, exposure to urban crime is positively associated with both individual income levels and neighborhood income characteristics (Sampson et al., 1997). Interestingly from a psychological health perspective, a key underlying mechanism to explain the linkage between neighborhood poverty and crime is diminished collective efficacy. Residents of low income, high crime neighborhoods perceive less social cohesion and diminished social control in their neighborhoods relative to persons living in lower crime areas (Sampson et al., 1997). Fear of crime in adults, particularly the elderly population, has reached epidemic proportions in low income, inner city neighborhoods (Perkins & Taylor, 1996; Wandersman & Nation, 1998). Finally, exposure to violence has well documented, adverse consequences on children's socioemotional development (Garbarino, 1995; Garbarino, Dubrow, Kostelny & Pardo, 1992; Osofsky, 1995; Richters & Martinez, 1993).
Adult mortality, physical morbidity, and health related-behaviors (e.g., diet, exercise, smoking) are all associated with neighborhood SES, independent of individual SES (Pickett & Pearl, 200; Robert, 1998; Waitzman & Smith, 1998). Children growing up in high SES neighborhoods have a clear advantage in school readiness and perform better academically, independently of familial income or education (Leventhal & Brooks-Gunn, 2000). Children and youth's mental health, particularly externalizing behaviors (acting out, aggression), are associated with residence in low income neighborhoods. Studies controlling for individual SES as well as multiple level analyses converge on these findings. Adolescents in low income neighborhoods also appear to become sexually active earlier and are more likely to become teenage parents compared to their peers living in more affluent neighborhoods (Leventhal & Brooks-Gunn, 2000).
We have reviewed data showing that income is associated with exposure to a wide variety of environmental quality indicators in the ambient environment, at home, in school, on the job, and in one's neighborhood. Differential income and racial exposure to environmental health risks constitute an important and emerging field of scholarship and public policy, frequently termed environmental justice. It would be fair to summarize this body of work as showing that the poor and especially the non-white poor bear a disproportionate burden of exposure to suboptimal, unhealthy environmental conditions in the United States. Moreover the more researchers scrutinize environmental exposure and health data for racial and income inequalities, the stronger the evidence becomes that grave and widespread environmental injustices have occurred throughout the United States. Such findings moved former President Clinton in 1994 to issue an executive order requiring all federal agencies to identify and address disproportionately high and adverse human health or environmental effects of federal programs and policies on minority and low income populations.
There are several gaping holes in the current data base necessary to critically examine whether the SES health gradient could be partly attributed to environmental exposures. First, data on income or SES, and environmental exposure are quite thin for several important environments, especially work, schools, and neighborhood settings. In several instances a dose response function is not available, instead measures of environmental risk for low income individuals are compared to persons above the poverty line. It would be preferable to have data across the continuum of income or SES and environmental risk exposure. Moreover in many instances the poverty: not poverty comparison is entangled with ethnicity. In the cases of exposure to hazardous waste sites and to occupational risk exposure, respectively the data on race differentials in exposure are better developed than they are for income.
Second, we hypothesize that the likelihood of singular environmental exposure accounting for the SES health gradient is small—instead we believe that it is the confluence of suboptimal conditions that is most likely to function as a potent mechanism helping to account for SES related differences in health. Research on cumulative risk exposure among children offers a useful analogue. This work shows that children exposed to one or perhaps two serious risk factors suffer at most modest decrements in psychological or cognitive functioning. However the accumulation of multiple risk factors dramatically elevates the probability of adverse socioemotional and cognitive developmental outcomes (Rutter, 1981). The gap in our analysis of income, environmental risk and health is that hardly any data exist showing the relation between income and multiple sources of environmental risk. We do know with some clarity that income is inversely related to exposure to a higher frequency of social stressors and to more adverse social stressors (Attar et al., 1994; Brown et al., 1986) but parallel data for physical stressors do not exist.
The third serious deficiency in the current data base for claiming that adverse environmental exposure might account for the SES health gradient is the absence of any data testing for the mediational model depicted in Figure 1. To our knowledge no data indicate that the effects of poverty or income on health are mediated by exposure to multiple environmental risk factors. Therefore what we have shown herein can be summarized as follows:
- income is often directly related to environmental quality, especially when low income samples are contrasted with samples that are not poor;
- environmental quality is inversely related to multiple physical and psychological health outcomes.
Greater progress in addressing the model shown in Figure 1 will require the collection of environmental risk and health data broken down by income or SES levels. Currently such data bases remain the exception. The absence of longitudinal studies also leaves the door open for the relations among income, environmental risk, and health to be due to selection factors rather than environmental effects. Such a person-based explanation seems unlikely to account for the wide array of differential, environmental exposure shown herein but changes in environmental conditions intra-person would provide even stronger evidence of an environmentally based mechanism for the SES health gradient than the current preponderance of cross sectional data.
There is clearly consistent evidence that people who are poorer in the United States are more likely to be exposed to multiple, environmental risks that portend adverse health consequences. Exposure to multiple, suboptimal environmental risk factors is one viable mechanism among several that could account for some of the covariance between SES and health.
Altschuler, A., Somkin, C.P., & Adler, N.E. (2004). Local services and amenities, neighborhood social capital, and health. Social Science and Medicine, 59, 1219-1229.
Aneshensel, C. & Sucoff, C. (1996). The neighborhood context of adolescent mental health. Journal of Health and Social Behavior, 37, 293-310.
Araki, S. (Ed.). (1994). Neurobehavioral methods and effects in occupational and environmental health. NY:Academic.
Attar, B., Guerra, N. & Tolan, P. (1994). Neighborhood disadvantage, stressful life events, and adjustment in urban elementary school children. Journal of Clinical Child Psychology, 23, 391-400.
Bardana, E. & Montanaro, B. (Eds.). (1997). Indoor air pollution and health. NY:Marcel Dekker.
Bartlett, S.N. (2002). The problem of children's injuries in low-income countries: A review. Health Policy and Planning, 17, 1-13.
Baum, A. & Paulus, P.D. (1987). Crowding. In D. Stokols & I. Altman (Eds.). Handbook of environmental psychology (pp. 533-570). NY:Wiley.
Berglund, B. & Lindvall, T. (1995). Community noise. Archives of the Center for Sensory Research, 2, 1-195.
Berney, L, Blane, D., Davey Smith, G., Gunnell, D., Holland, P. & Montgomery, S. (2000). Socioeconomic measures in early old age as indicators of previous lifetime exposure to environmental health hazards. Sociology of Health and llness, 22, 415-430.
Bernton, H., Mc Mahon, T. & Brown, H. (1972). Cockroach asthma. British Journal of Diseases in Children, 66, 61.
Blane, D., Barley, M. & Davey Smith, G. (1997). Disease aetiology and materialist explanations of socioeconomic mortality differentials. European Journal of Public Health, 7, 385-391.
Bradley, R. H. (1999). The home environment. In S.L. Friedman & T.D. Wachs (Eds.). Measuring environment across the lifespan (pp. 31-58). Washington, DC: American Psychological Association.
Bradley, R.H. & Caldwell, B. (1984). The HOME inventory and family demographics. Developmental Psychology, 20, 315-320.
Brajer, V. & Hall, J. (1992). Recent evidence on the distribution of air pollution effects. Contemporary Policy Issues, 10, 63-71.
Bronfennbrenner, U. & Morris, P. (1998). The ecology of developmental processes. In W. Damon & R. Lerner (Eds.). Handbook of child psychology, 5th ed. (pp. 992-1028). NY:Wiley.
Brown, L., Cowen, E., Hightower, A. & Lotyczewski, B. (1986). Demographic differences among children in judging and experiencing specific stressful life events. Journal of Special Education, 20, 339-346.
Bullard, R.D. (1990). Dumping in dixie. Boulder,CO: Westview Press.
Burridge, R. & Ormandy, D. (Eds.). (1993). Unhealthy housing. London:E.F. Spon.
Cabelli V. & Dufour, A. (1983). Health effects criteria for marine recreational waters. EPA-600/1-80-031. U.S. EPA, Office of Research and Development. Research Triangle Park, NC.
Cakmak, S., Dales, R.E., & Judek, S. (2006). Respiratory health effects of air pollution gases: modification by education and income. Archives of Environmental & Occupational Health, 61, 5-10.
Calderon, R., Johnson, C., Craun, G., Dufour, A., Karlin, R., Sinks, T. & Valentine, J. (1993). Health risks from contaminated water: Do class and race matter? Toxicology and Industrial Health, 9, 879-900.
Canfield, R.L., Henderson, C.R.J., Cory-Slechta, D.A., Cox, C., Jusko, T.A., & Lanphear, B. (2003). Intellectual impairment in children with blood lead concentrations below 10 micrograms per deciliter. New England Journal of Medicine, 348, 1517-1526.
Chi, P. & Laquatra, J. (1990). Energy efficiency and radon risks in residential housing. Energy, 15, 81-89.
Cieselski, S., Handzel. T. & Sobsey, M.(1991). The microbiologic quality of drinking water in North Carolina migrant farmer camps. American Journal of Public Health, 81, 762-764.
Clinton, W.J. (1994). Federal actions to address environmental justice in minority populations and low income populations. Federal Register, 59, 7629-7633.
Cohen, S. (1980). Aftereffects of stress on human performance and social behavior: A review of research and theory. Psychological Bulletin, 88, 82-108.
Cohen, S. & Spacapan, S. (1984). The social psychology of noise. In D.M. Jones & A.J. Chapman (Eds.). Noise and society (pp. 221-245). NY:Wiley.
Cook, D., Whincup, P., Jarvis, M., Strachan, D., Papacosta, O. & Bryant, A. (1994). Passive exposure to cigarette smoke in children aged 5-7 years: Individual, family, and community factors. British Medical Journal, 308, 384-389.
Cubbin, C., LeClere, F. & Smith, G. (2000). Socioeconomic status and injury mortality: Individual and neighbourhood determinants. Journal of Epidemiology and Community Health, 54, 517-524.
Davey Smith, G., Hart, C., Watt, G., Hole, D. & Hawthorne, V. (1998). Individual social class, area-based deprivation, cardiovascular disease risk factors, and mortality in Renfrew and Paisley study. Journal of Epidemiology and Community Health, 52, 399-405.
Davies, J., Edmundson, W., Raffonelli, A., Cassady, J. & Morgade, C. (1972). The role of social class in human pesticide pollution. American Journal of Epidemiology, 96, 334-341.
Dubow, E. & Ippolito, M. (1994). Effects of poverty and quality of the home environment on changes in the academic and behavioral adjustment of elementary school-age children. Journal of Clinical Child Psychology, 23, 401-412.
Durkheim, E. (1897; 1997). Suicide. NY: Free Press.
Environmental Protection Agency (1977). The urban noise survey. EPA 550/9-77. Washington, DC:EPA.
Environmental Protection Agency (1992). Environmental equity: Reducing risk for all communities. Office of Solid Waste and Emergency Response. Washington, DC. EPA230-R-92-008.
Environmental Protection Agency. (2003). America's children and the environment: Measure of contaminants, body burdens, and illnesses. Office of Children's Health Protection. Washington, DC. EPA240-R-03-001).
Evans, G.W. (1994). The psychological costs of chronic exposure to ambient air pollution. In R.L. Isaacson & K.F. Jensen (Eds.). The vulnerable brain and environmental risks, Vol. 3: Toxins in air and water (pp. 167-182). NY:Plenum.
Evans, G.W. (2001). Environmental stress and health. In A. Baum, T. Revenson & J.E. Singer (Eds.). Handbook of health psychology (pp. 365-385). Mahwah,NJ:Erlbaum.
Evans, G.W. & Cohen, S. (1987). Environmental stress. In D. Stokols & I. Altman (Eds.). Handbook of environmental psychology (pp. 571-610). NY:Wiley.
Evans, G.W. & Lepore, S.J. (1993). Nonauditory effects of noise on children: A critical review. Children's Environments, 10, 31-51.
Evans, G.W., Lepore, S.J. & Allen, K. (2000). Cross cultural differences in tolerance for crowding: Fact or fiction? Journal of Personality and Social Psychology, 79, 204-210.
Evans, G.W., Wells, N.M., Chan, E. & Saltzman, H. (2000). Housing and mental health. Journal of Consulting and Clinical Psychology, 68, 526-530.
Evans, G.W., Wells, N.M. & Moch, A. (2000). Housing and mental health: A review of the evidence and a methodological and conceptual critique. Manuscript under review.
Federman, M., Garner, T., Short, K., Cutter, W., Levine, D., Mc Gough, D. & Mc Millen, M. (1996). What does it mean to be poor in America? Monthly Labor Review, May, 3-17.
Fiese, B. & Kline, C. (1993). Development and validation of the family ritual questionnaire: Initial reliability and validation studies. Journal of Family Psychology, 6, 290-299.
Freeman, A.M. (1972). The distribution of environmental quality. In A.V. Kneess & B. Bower (Eds.). Environmental quality analysis (pp.243-280). Baltimore:Johns Hopkins Press.
Frumkin, H. & Walker, D. (1998). Minority workers and communities. In Wallace, R. (Ed.), Maxcy-Rosenau Last public health and preventative medicine, 14th ed. (pp.682-688). Stamford, Conn:Appleton & Lange.
Gallup, G. (1993). America's youth in the 1990s. Princeton: Gallup Institute.
Garbarino, J. (1995). Raising children in a socially toxic environment. San Francisco:Jossey-Bass.
Garbarino, J., Dubrow, N., Kostelny, K. & Pardo, C. (1992). Children in danger: Coping with the consequences of community violence. San Francisco:Jossey-Bass.
Garrett, P. Ng'andu, N. & Ferron, J. (1994). Poverty experiences of young children and the quality of their home environments. Child Development, 65, 331-345.
Gifford, R.(in press). Satisfaction, health, security and social relations in high rise buildings. In A. Seidel & T. Heath (Eds.). Social effects of the building environment. London:E. & F.N. Spon.
Glass, D.C. & Singer, J.E. (1972). Urban stress. NY:Academic.
Goldstein, I., Andrews, L. & Hartel, D. (1988). Assessment of human exposure to nitrogen dioxide, carbon monoxide and respirable particulates in New York inner city residents. Atmospheric Environment, 22, 2127-2139.
Goldtooth, T.B.K. (1995). Indigenous nations: Summary of sovereignty and its implications for environmental protection. In B. Bryant (Ed.). Environmental justice (pp. 138-148). Washington, DC:Island Press.
Graham, H. (1995). Cigarette smoking: A light on gender and class inequality in Britain? International Journal of Social Policy, 24, 509-527.
Graham, H. & Blackburn, C. (1998). The socioeconomic patterning of health and smoking behaviour among mothers with young children on income support. Sociology of Health and Illness, 20, 215-240.
Groner, J., Ahijevych, K., Grossman, L. & Rich, L. (1998). Smoking behaviors of women whose children attend an urban pediatric primary care clinic. Women and Health, 28, 19-32.
Haan, M., Kaplan, G. & Camacho, T. (1987). Poverty and health. American Journal of Epidemiology, 125, 898-998.
Haines, M., Stansfeld, S., Head, J., & Job, R.F.S. (2000). Multi-level modelling of aircraft noise on performance tests in schools around Heathrow Airport London. International Congress of Psychology. Stockholm, Sweden. August.
Hamermesh, D. (1999). Changing inequality in work injuries and work timing. Monthly Labor Review, October, 22-30.
Harburg, E., Erfurt, J., Hausentstein, L., Chape, C., Schull, W. & Schork, M. (1973). Socioecological stress, suppressed hostility, skin color, and black-white male blood pressure: Detroit. Psychosomatic Medicine, 35, 276-296.
Harlan, S.L., Brazel, A.J., Prashad, L., Stefanov, W.L., & Larsen, L. (2006). Neighborhood microclimates and vlunerability to heat stress. Social Science and Medicine, 63, 2847-2863.
Haynes, R., Reading, R., & Gale, S. (2003). Household and neighborhood risks for injury to 5-14 year old children. Social Science and Medicine, 57, 625-636.
Hiroto, D.S. (1974). Locus of control and learned helplessness. Journal of Experimental Psychology, 102, 187-193.
Holgate, S., Samet, J., Koren, H. & Maynard, R. (1999). Air pollution and health. NY:Academic.
Homel, R. & Burns, A. (1987). Is this a good place to grow up in? Neighborhood quality and children's evaluations. Landscape and Urban Planning, 14, 101-116.
Hyndman, S. (1998). Making connections between housing and health. In R. Kearns & W. Gesler (Eds.). Putting health into place. Syracuse, NY:Syracuse University Press.
Ineichen, B. (1993). Housing and health. London: E & FN Spon.
Ingersoll, R.M. (1999). The problem of underqualified teachers in American secondary schools. Educational Researcher, 28, 26-37.
Institute of Medicine (1999). Environmental justice. Washington, DC: National Academy Press.
Institute of Medicine (2000). Clearing the air: Asthma and indoor air exposure. Washington, DC: National Academy Press.
Jarvis, M., Strachan, D. & Feyerbrand, C. (1992). Determinants of passive smoking in children in Edinburgh, Scotland. American Journal of Public Health, 82, 1225-1229.
Johnson, B.L. (1999). Impact of hazardous waste on human health. NY:Lewis.
Joint Center for Housing Studies at Harvard University. (1999). The State of the Nation's Housing. Cambridge, MA: Harvard University.
Jun, H.J, Subramanian, S.V., Gortmaker, S., & Kawachi, I. (2004). Socioeconomic disadvantage, parenting responsibility, and women's smoking in the United States. American Journal of Public Health, 94, 2170-2176.
Kang, B. (1976). Study on cockroach antigen as a probably causitive agent in bronchial asthma. Journal of Allergy and Clinical Immunology, 58, 357-365.
Kasl, S.W., Will, J., White, M. & Marcuse, P. (1982). Quality of the residential environment and mental health. In A. Baum & J.E. Singer (Eds.). Advances in environmental psychology (pp. 1-30). Hillsdale, NJ: Erlbaum.
Kemble, S.K., Davis, J.C., Nalugwa, T. Njama-Meya, D., Hopkins, H., Dorsey, G., & Staedke, S.G. (2006). Prevention and treatment strategies used for the community management of childhood fever in Kampala, Uganda. American Journal of Tropical Medicine and Hygiene, 74, 999-1007.
Kids Count Data Book, 2000. (2000). Seattle: Annie Casey Foundation.
Klinenberg, E. (2002). Heat wave: A social autopsy of disaster in Chicago. Chicago: University of Chicago Press.
Kryter, K. (1994). The handbook of hearing and the effects of noise. NY:Academic.
Larson, R.W. & Verma, S. (1999). How children and adolescents spend time around the world: Work, play and developmental opportunities. Psychological Bulletin, 125, 701-736.
Le Clere, F.B., Rogers, R. & Peters, K. (1998). Neighborhood context and racial differences in women's heart disease. Journal of Health and Social Behavior, 39, 91-107.
Leventhal, T. & Brooks-Gunn, J. (2000). The neighborhoods they live in: The effects of neighborhood residence on child and adolescent outcomes. Psychological Bulletin, 126, 309-337.
Lippman, N. (1992). Environmental toxicology. NY:Van Nostrand.
Lucas, R.E.B. (1974). The distribution of job characteristics. The Review of Economic Statistics, 56, 530-540.
Lundberg, O. (1991). Causal explanations for class inequality in health-an empirical analysis. Social Science and Medicine, 32, 385-393.
Machlin, S. R., Hill, S. C., and Liang, L. (2006). Children Living with Adult Smokers, United States, 2004. Statistical Brief 147. Rockville,MD:Agency for Healthcare Research and Quality.
Macpherson, A.,, Roberts, I., & Pless, I. (1998). Children's exposure to traffic and pedestrian injuries. American Journal of Publich Health, 88, 1840-1845.
Macintyre, S., Maciver, S. & Sooman, A. (1993). Area, class and health: Should we be focusing on places or people? International Social Policy, 22, 213-234.
Malcoe, L.H., Lynch, R.A., Kegler, M.C., Skaggs, V.J. (2002). Lead sources, behaviors, and socioeconomic factors in relation to blood lead of Native American and white children: a community based assessment of a former mining area. Environmental Health Perspective, 110, 221–232.
Marmot, M., Siegrist, J., Theorell, T. & Feeney, A. (1999). Health and the psychosocial environment at work. In M. Marmot & R.G. Wilkinson (Eds.). Social determinants of health (pp. 105-131). NY:Oxford University Press.
Matheny, A., Wachs, T.D., Ludwig, J. & Phillips, E. (1995). Bringing order out of chaos: Psychometric characteristics of the confusion, hubub, and order scale. Journal of Applied Developmental Psychology, 16, 429-444.
Matte, T. & Jacobs, D. (2000). Housing and health: Current issues and implications for research and progress. Journal of Urban Health Bulletin of the New York Academy of Medicine, 77, 7-25.
Maxwell, L.M. (1996). Multiple effects of home and day care crowding. Environment and Behavior, 28, 494-511.
Mayer, S.E. (1997). Trends in the economic well being and life chances of America's children. In G.J. Duncan & J. Brooks-Gunn (Eds.). Consequences of growing up poor. NY:Russell Sage.
Mc Harg, I. (1969). Design with nature. Garden City, NY:Natural History Press.
Miller, J. & Davis, D. (1997). Poverty history, marital history, and quality of children's home environments. Journal of Marriage and the Family, 59, 996-1007.
Moore, G.T. & Lackney, J. (1993). School design. Children's Environments, 10, 99-112.
Morello-Frosch, R., Pastor, M., Porras, C., & Sadd, C. (2002). Environmental justice and regional inequality in southern California: Implications for future research. Environmental Justice, 110, 149-154.
Moses, M., Johnson, E., Anger, W., Burse, V., Horstman, S., Jackson, R., Lewis, R., Maddy, K., Mc Connell, R., Meggs, W. & Zahm, S. (1993). Environmental equity and pesticide exposure. Toxicology and Industrial Health, 9, 913-959.
Myers, D., Baer, W. & Choi, S. (1996). The changing problem of overcrowded housing. Journal of the American Planning Association, 62, 66-84.
National Center for Education Statistics. (2000). Condition of America's public school facilities: 1999. NCES 2000-032. Washington, DC:U.S. Department of Education.
National Center for Health Statistics (1991). Children's exposure to environmental cigarette smoke. Advance data from vital and health statistics: Number 202. Hyattsville, MD.
National Center for Health Statistics. (2001). Smoking during pregnancy in the 1990s. National vital statistics reports: Number 7. Hyattsville, MD.
National Institute for Child Health and Human Development. (1997). Poverty and patterns of child care. In G.J. Duncan & J. Brooks-Gunn (Eds.). Consequences of growing up poor (pp. 100-131). NY:Russell Sage.
National Research Council (1991). Environmental epidemiology, Vol 1. Washington, DC:National Academy Press.
O'Neill, M.S., Jerrett, M., Kawachi, I., Levy, J.I., Cohen, A.J., Gouveia, N., Wilkinson, P., Fletcher, T., Cifuentes, L., & Schwartz, J. (2003). Health, wealth, and air pollution: Advancing theory and methods. Environmental Health Perspectives, 111, 1861-1870.
Osofsky, J. (1995). The effects of exposure to violence on young children. American Psychologist, 50, 782-788.
Perkins, D. & Taylor, R.B. (1996). Ecological assessments of community disorder: Their relationship to fear of crime and theoretical implications. American Journal of Community Psychology, 24, 63-107.
Phillips, D.A., Voran, M. Kisker, E., Howes, C. & Whitebook, M (1994). Childcare for children in poverty: Opportunity or inequity? Child Development, 65, 472-492.
Pickett, K. & Pearl, M. (2001). Multilevel analyses of neighborhood socioeconomic context and health outcomes: A critical review. Journal of Epidemiology and Community Health, 55, 111-122.
Pirkle, J., Brody, D., Gunter, E., Kramer, R., Paschal, D., Flegal, K. & Matte, T. (1994).The decline in blood lead levels in the United States. Journal of the American Medical Association, 272, 284-291.
Quinn, M.M., Sembajwe, G., Stoddard, A.M., Kriebel, D., Krieger, N., Sorensen, G. Hartman, C., Naishadham, D., & Barbeau, E.M. (2007). Social disparities in the burden of occupational exposures: Results of a cross sectional study. American Journal of Industrial Medicine, 50, 861-875.
Rauh, V.A., Chew, G.L., & Garfinkel, R.S. (2002). Deteriorated housing contributes to high cockroach allergen levels in inner-city households. Environmental Justice, 110, 323- 327.
Richters, J.E. & Martinez, P. (1993). The NIMH community violence project. Psychiatry, 56, 7-21.
Riley, E. & Vorhees, C. (Eds.). (1991). Handbook of behavioral teratology. NY:Plenum.
Robert, S. (1998). Socioeconomic position and health: The independent contribution of community socioeconomic status. Annual Review of Sociology, 25 489-516/
Roberts, I. & Power, C. (1996). Does the decline in injury mortality vary by social class? British Medical Journal, 313, 784-786.
Rosenstreich, D., Eggleson, P., Kattan, M., Baker, D., Slavin, R., Gergen, P., Mitchell, H., McNiff-Mortimer, K., Lynne, H., Ownby, D. & Malveaux, F. (1997). The role of cockroach allergy and exposure to cockroach allergens in causing morbidity among inner-city children with asthma. New England Journal of Medicine, 336, 1356-1363.
Ross, C. (2000). Walking, exercising, and smoking: Does neighborhood matter? Social Science and Medicine, 51, 265-274.
Rotton, J. (1983). Affective and cognitive consequences of malodorous pollution. Basic and Applied Social Psychology, 4, 171-191.
Rutter, M. (1981). Protective factors in children's responses to stress and disadvantage. In M. Kent & J. Rold (Eds.). Prevention of psychopathology, vol. 1 (pp. 49-74).Hanover, NH:University Press.
Samet, J. & Spengler, J. (Eds.). (1991). Indoor air pollution: A health perspective. Baltimore:Johns Hopkins Press.
Sampson, R., Raudenbush, S. & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918-924.
Sarpong, S., Hamilton, R., Eggleston, P. & Adkinson, N. (1996). Socioeconomic status and race as risk factors for cockroach allergen exposure and sensitization in children with asthma. Journal of Allergy and Clinical Immunology, 97, 1393-1401.
Schwab, M. (1990). An examination of intra-SMSA distribution of carbon monoxide exposure. Journal of the Air and Waste Management Association, 40, 331-336.
Scott, R. (1990). Chemical hazards in the workplace. Chelsea, MN:Lewis.
Shenassa, E.D., Stubbendick, A., & Brown, M.J. (2004). Social disparities in housing and related pediatric injury: A multilevel study. American Journal of Public Health, 94, 633-639.
Sherman, A. (1994). Wasting America's future. Boston:Beacon Press.
Sinclair, J., Pettit, G., Harrist, A., Dodge, K. & Bates, J. (1994). Encounters with aggressive peers in early childhood: Frequency, age differences, and correlates of risk for behavior problems. International Journal of Behavioral Development, 17, 675-696.
Spencer, M.B., Mc Dermott, P., Burton, L. & Kochman, T. (1997). An alternative approach to assessing neighborhood effects on early adolescent achievement and problem behavior. In J. Brooks-Gunn, G.J. Duncan & J.L. Aber (Eds.). Neighborhood poverty, Vol. II: Policy implications in studying neighborhoods (pp. 145-163). NY:Russell Sage Foundation.
Spencer, N.J. & Coe, C. (2001). The additive effects of social factors on risk of smoking in households with newborn infants. Unpublished manuscript. University of Warwick. United Kingdom.
Statistical Universe (2000). Income of families and primary individuals by selected characteristics: Renter occupied units, 1999. http://web.lexis-nexis.com/statuniv/.
Stehr-Green, P. (1989). Demographic and seasonal influences on human serum pesticide residue levels. Journal of Toxicology and Environmental Health, 27, 405-421.
Steptoe, A. & Feldman, P. (2001). Neighborhood problems as sources of chronic stress: Development of a measure of neighborhood problems, and associations with socioeconomic status and health. Annals of Behavioral Medicine, 23, 177-185.
Stronks, K., Dike van de Mheen, H. & Mackenbach, J. (1998). A higher prevalence of health problems in low income groups: Does it reflect relative deprivation. Journal of Epidemiology and Community Health, 52, 548-557.
Suecoff, S., Avner, J., Chou, K. & Drain, E. (1999). A comparison of New York City playground hazards in high and low income areas. Archives of Pediatrics and Adolescent Medicine, 153, 363-366.
Taylor, R.B. & Harrell, A. (1999). Physical environment and crime. National Institute of Justice. Washington, DC.
Texas Governor's Border Working Group (1992). The Texas-Mexico border: Challenges and opportunities for improving public health and access to health care. Subgroup on infectious disease and environment and the Subgroup on access to health care. Cited in Calderon et al. (1993). Health risks from contaminated water: Do class and race matter? Toxicology and Industrial health, 9, 879-900.
Thompson, S.J. (1993). Review: Extraaural health effects of chronic noise exposure in humans. In H. Ising & B. Kruppa (Eds.). Larm und krankheit (Noise and disease)(pp. 107-117). NY:Verlag.
Townsend, P. (1979). Poverty in the United Kingdom. Berkeley:University of California Press.
Trancik, A. & Evans, G.W. (1995). Spaces fit for children: Competency in the design of daycare center environments. Children's Environments, 12, 311-319.
Trends in the Well Being of America's Children and Youth 2000. (2000). U.S. Department of Health and Human Services. Washington, DC:U.S. Government Printing Office.
Turner, M.A. & Kaye, D.R. (2006). How Does Family Well-Being Vary across Different Types of Neighborhoods? Low-Income Working Families, Paper 6. Washington, DC: The Urban Institute.
Wachs, T.D. & Gruen, G. (1982). Early experience and development. NY:Plenum.
Waitzman, N. & Smith, K. (1998). Phantom of the area: Poverty-area residence and mortality in the United States. Journal of Public Health, 88, 973-976.
Wallace, D. & Wallace, R. (1998). A plague on your houses. London: Verso.
Wandersman, A. & Nation, M. (1998). Urban neighborhoods and mental health. American Psychologist, 50, 647-656.
Weinstein, C. & David, T. (Eds.). (1987). Spaces for children. NY:Plenum.
West, P., Fly, J., Marans, R. & Larkin, F. (1989). Michigan sports anglers fish consumption survey. University of Michigan School of Natural Resources.Natural Resource Sociology Research Lab. Technical Report 1. Ann Arbor, MI.
White, H.L. (1998). Race, class, and environmental hazards. In D.E. Camacho (Ed.). Environmental injustices, political struggles (pp. 61-81). Durham, NC:Duke University Press.
Woodard, B., Ferguson, B. & Wilson, D. (1976). DDT levels in milk of rural indigent blacks. American Journal of Diseases in Children, 130, 400-403.
World Bank. (1992). World development report. NY:Oxford University Press.
World Health Organization. (2006). Guidelines for drinking water quality, 3rd Edition. Geneva: World Health Organization.
Wright, B.H. (1992). The effects of occupational injury, health, and disease on the health status of Black Americans. In B. Bunyan & P. Mohai (Eds.). Race and the incidence of environmental hazards (pp. 114-125). Boulder, CO:Westview Press.