Skip to main content

Table 2 Characteristics of studies of residential environment and breast cancer risk or incidence

From: Residential environment and breast cancer incidence and mortality: a systematic review and meta-analysis

Author, year (location)

Individual-level data source; area level data source

Study design and sample characteristics

Geographic location and unit

Main area based measures (measurement)

Outcome

[26]Blot, 1977 (United States)

NIH publication on US cancer mortality by county; 1960 US Census

Cross-sectional; ≥ 20 years old

Contiguous US; county

Income (Median family income, categorized into 2 groups: <50%, >50% by region and population-size)

Mortality

[17]Devesa, 1980 (United States)

Third national cancer survey 1969–1971; US Census 1970

Cross-sectional Females ≥ 15 years; n=20,914 cases; 92.5% white, 7.5% black

18 US Standard Metropolitan Statistical Areas; Census Tracts

Education (Median years of education categorized into 5 groups for Whites: <10, 10–10.9, 11–11.9, 12–12.9, and ≥13 years; categorized into 3 groups for blacks: <10, 10–10.9, and ≥11 years) Income (Median family income categorized into 5 groups for Whites: <$9,000, $9,000-10,999, $11,000-12,999, $13,000-14,999, ≥$15,000; and categorized into 3 groups for blacks: <$5,000, $5000-6,999, ≥$7,000)

Incidence

[20]Krieger, 1990 (United States)

SEER 1979–1981; US Census 1980

Cross-sectional; Females; n=4,454 cases; 86% white, 9% black, 5% Hispanic

San Francisco Bay Area; Census block group

Occupational class composition (% employed in “working class” occupations, categorized into 2 groups ≤ 66% and >66% in working class occupations); Poverty (% living below poverty, categorized into two groups (≥20% and < 20%)

Incidence

[16]Baquet, 1991 (United States)

SEER 1978–1982; US Census 1980

Cross-sectional; Females; ≥ 25 years old

San Francisco/Oakland, Atlanta, Detroit; Census tract

Education (Median years of education, categorized into 4 groups: <high school, high school graduates, some college, at least 4 years of college) Income (Median family income, categorized into 4 groups: <$15,000, $15,000-24,999, $25,000-29,999, ≥$30,000)

Incidence

[23]Williams, 1991 (Australia)

Victorian Cancer Registry 1982–1983 Melbourne statistical division 1979–1983; Australian Census 1981

Cross-sectional; Females; 40–74 years old

Melbourne; Local government area

Composite SES (Based on occupational status, income, educational attainment, family instability, persons living in low standard housing likely to have difficulty with English, categorized into deciles)

Incidence mortality

[21]Nasca, 1992 (United States)

New York State Cancer Registry 1978–1982; US Census 1980

Cross-sectional

New York state exclusive of New York City; Minor civil divisions

Urbanization (Population density: [persons/square miles], categorized into quinitles

Incidence

[22]Pollock, 1997 (United Kingdom)

The Thames Cancer Registry 1987–1992; UK Census 1991

Cross-sectional; Females; 40–99 years old; n=22,399 cases

South Thames; Enumeration district

Composite SES “Townsend Index of Social Deprivation” (based on % unemployed, % private household lacking a car, % private household not owner occupied, % private household subject to overcrowding; categorized into deciles)

Incidence mortality

[30]Gorey, 1998 (Canada)

Ontario cancer registry 1986–1993; Canadian Census 1991

Cross-sectional; Females; ≥ 25 years old; n=1,3227 cases

Metropolitan Toronto; Census tract

Poverty (annual household income adjusted for household size, categorized into low (≥23% households below low-income cutoff) and high (<7% of households below criterion))

Incidence

[34]Liu, 1998 (United States)

The Los Angeles County Surveillance Program 1979–1992; US Census 1970, 1980, 1990

Cross-sectional; Females; ≥ 15 years old; n=82,453 cases; 77.9% white, 9.1% black, 9.0% Hispanic, 3.2% Asian, <1% other

Los Angeles County; Census tract

Composite SES (Based on weighted average educational attainment and median household income; categorized into quintiles)

Incidence

[39]Prehn, 1998 (United States)

Northern California cancer center’s greater Bay Area cancer registry 1988–1992; US Census 1990

Cross-sectional; Females; n=22,757 cases; 100% white

San Francisco Bay Area and 20 counties from adjoining regions; Census block group

Education (% with college education, categorized into 2 groups: ≥45% vs. <45%); Income (Median household income, categorized into 2 groups ≥ $50000 and <$50000); Occupational Class (% employed in working class occupations, categorized into 2 groups ≤50% and >50% working class); Poverty (% below poverty level, categorized into 2 groups ≤5% vs.>5%)

Incidence

[33]Krieger, 1999 (United States)

Population-based cancer registry 1988–1992; US Census 1990

Cross-sectional; Females; n=16,120 cases; 78% white, 7% black, 7% Hispanic, 8% Asian

San Francisco Bay Area; Census block group

Composite SES (Combination of occupational class (% employed in “working class” and “professional” occupations) and poverty (% below poverty level); categorized into 3 groups: 1) professional (non-poor and poor), 2) working class, non-poor, 3) working class, poor)

Incidence

[35]Mackillop, 2000 (Canada and United States)

Ontario Cancer Registry (1989–1993); SEER 1988–1992; Canadian Census 1991 and US Census 1990

Cross-sectional

Ontario, Canada; 9 SEER regions in US; Enumeration area in Canada and census tract in the U.S.

Income (Median household income, categorized into deciles. Race-specific deciles in the US for secondary analysis) Natural log of relative income for regression

Incidence

[40]Reynolds, 2004 (United States)

The California Teachers Study cohort with annual linkage to the California Cancer Registry, baseline in 1995 with follow up through Dec 1999; US Census 1990

Prospective cohort; Females; 21–108 years at baseline; n=114,927

California; region

Urbanization (a priori specification of urban counties, categorized into San Francisco Bay area, Southern Coastal area, rest of California)

Incidence

[42]Robert, 2004 (United States)

Population-based case control study 1988–1995; US Census 1990

Case–control; Females; 20–79 years old; n=7,179 cases, 7,488 controls

Wisconsin; Census tract and Zip code

Composite SES (Based on median income, % adults below poverty, % unemployed, % college graduate, categorized into quintiles) Urbanization (Residence in census-defined “urban areas”, categorized into 3 groups: 100% rural, mixed rural/urban, 100% urban)

Risk

[18]Hall, 2005 (United States)

North Carolina State Registry 1995–1999; US Census 1990

Cross-sectional; Females; 27,989 cases, 82% white, 18% non-white

North Carolina; County

Metropolitan areas (Urban Influence Code based on by adjacency or non-adjacency to a Metropolitan Area, and size of the largest communities, categorized into 3 groups- metropolitan; non-metropolitan adjacent to metropolitan; non-metropolitan, nonadjacent to metropolitan areas)

Incidence

[41]Reynolds, 2005 (United States)

The California Cancer Registry 1988–1997; US Census 1990

Cross-sectional; Females; ≥20 years old; 176,302 cases

California; Block group

Composite SES (Based on % with college degree, median family income and % employed in managerial/professional occupations, categorized into quartiles) Urbanization (Population size and density, categorized into 4 groups -urban suburban, city, small town/rural)

Incidence

[32]Krieger, 2006 (United States)

Northern California Cancer Center’s San Franciso; Oakland SEER Registry, Los Angeles Surveillance program, Massachusetts Cancer registry 1978–1982, 1988–1992, 1998-2002/ US Census 1980, 1990, 2000

Cross-sectional; Females; 154,083 cases

San Francisco/ Oakland, Los Angeles county, Massachusetts; Census tract

Composite SES (% below poverty level and % high income ho'useholds (defined as ≥4 times the US median household income), categorized into 5 groups: 1) <5% poverty-<10% high income, 2) <5% poverty-≥10% high income, 3) 5-9% poverty, 4) 10-19% poverty, 5) ≥20% poverty)

Incidence

[44]Shack, 2008 (United Kingdom)

English cancer registries 1998–2003; UK Census 2001 and government databases

Cross-sectional; Females; 210,020 cases

8 UK cancer registries; Postal code of residence

Income deprivation (Based on the income domain of the Index of Multiple Deprivation, categorized into quintiles)

Incidence

[46]Webster, 2008 (United States)

The Massachusetts Cancer Registry 1987–1993; US Census 1980, 1990

Case–control; Females; 548 cases, 490 controls

Cape Cod, Massachusetts; Census block group

Composite SES (Based on median income, % adults below poverty, % unemployed, % college graduate, categorized into quintiles) Poverty (% of adults below poverty level, categorized into 3 groups based on the 20th and 80th percentiles of control women)

Risk

[28]Clegg, 2009 (United States)

National Longitudinal Mortality Study and SEER, 1973–2001; US Census 1970, 1980, 1990

Cross-sectional; Females; ≥ 25 years old; 1739 cases; 78% white, 7% black, 4% Mexican, 1% other Hispanic, 4% Asian/Pacific Islander, 2% other

11 SEER regions

Urbanization (Census definition of urban/rural)

Incidence

[4] Harper, 2009 (United States)

SEER 1987–2004; US Census 1990

Cross-sectional

SEER regions; County

Poverty (% below poverty level, categorized into 4 groups: <10%, 10-14%, 15-19%, ≥20%)

Incidence mortality

[19]Hausaer, 2009 (United States)

NAACCR Registries 1997–2004; USDA 2003

Cross-sectional; Females; 50–74 years old; 587,408 cases; 100% white

29 population-based cancer registries in the North American Association of Central Cancer Registries (NAACCR); County

Poverty (% below poverty level, categorized into 3 groups: <10%, 10-19%, ≥20%) Urbanization (US Dept. of Agriculture codes and population size, categorized into urban, suburban and rural areas)

Incidence

[29]Dobson, 2010 (Australia)

Australian Longitudinal Study on Women’s Health, baseline survey in 1996 with follow up through 2006; The Australian Standard Geographic Classification

Longitudinal; Females; 70–75 years at baseline; 12,400 with 2,803 breast cancer deaths

Australia

Area of residence (Road distance to the closest service center, a measure of population size)

Mortality

[31]Keegan, 2010 (United States)

The California Cancer Registry 1988–2004; US Census 2000

Cross-sectional; Females; 12,563 cases; 100% Hispanic

California/ Cross-sectional; Block groups averaged over census tracts (for SES)

Composite Hispanic Enclave (Based on % linguistically isolated overall and who speak Spanish, speak limited English, speak limited English and speak Spanish, % recent immigrants, % Hispanic, % foreign-born) Composite SES (Based on income, occupation, and housing costs, categorized into quintiles) Combined SES and Hispanic Enclave (Combination of SES and Hispanic enclave, categorized into 4 groups: low SES-high enclave, high SES-low enclave, low SES-low enclave, high SES-high enclave)

Incidence

[45]Spadea, 2010 (Italy)

The Turin Longitudinal Study and the Piedmont Cancer Registry, 1985–1999; Italian Census 1971

Cross-sectional; Females; 30–84 years old; 9,203 cases

Turin, Italy; Census tract

Composite SES (Based on % manual workers, % with low education, % tenants, % living % in houses without bath, % families with a single parent with children, and a crowding index, categorized into quintiles) Relative Index of Inequality (Ratio of regression-based rates for extreme points of the social hierarchy)

Incidence/Risk

[27]Borugian, 2011 (Canada)

The Canadian Cancer Registry 1992–2004; Canadian Census 1991, 1996, 2001, 2006

Cross-sectional; Females; ≥19 years old; 226,169 cases

Canada; postal code

Income (Average income per single person equivalent in the enumeration area or dissemination area, categorized into quintiles)

Incidence

[47]Yost, 2001 (United States)

The California Cancer Registry 1988–1992 US Census 1990

Cross-sectional; Females; ≥15 years old; 97,227 cases; 80% white, 6% black, 9% Hispanic, 5% Asian

California; Census block group

Composite SES (Based on education index, proportion with a blue-collar job, % in workforce without a job, median household income, % below 200% poverty level, median rent, median house value, categorized into quintiles)

Incidence

[37]Palmer, 2012 (United States)

The Black Women’s Health Study, baseline in 1995 with follow-up through 2009; US Census 2000

Longitudinal; Females; 21–69 years at baseline; total n=55,896, analysis on n=1,343 cases with geocoded data; 100% black

17 US states; Census block group

Composite SES (Based on median household income, median housing value, % household receiving interest, dividends or net rental income, % with college degree, % employed in managerial, executive or professional specialty, % families with children headed by a single female; categorized into quintiles)

Incidence

[38]Panczak, 2012 (Switzerland)

The Swiss National Cohort 2001–2008;Swiss Census 2000

Longitudinal; Females; ≥ 30 years old; n=4,300,000 (including males), breast cancer deaths unknown

Switzerland; Neighborhood boundaries

Swiss-SEP Index (SES composite measure based on occupational status, income, educational attainment, family instability, persons living in low standard housing likely to have difficulty with English, categorized into deciles)

Mortality

[43]Schlichting, 2012 (United States)

SEER 2000–2007; US Census 2000

Cross-sectional; Females n=34,3627 cases; 75% white, 9% black, 9% Hispanic, 7% Asian/Pacific Islander, <1% American Indian/ Native Alaskan

17 SEER regions; County

Education (% without high school degree, categorized into quartiles) Poverty (% below federal poverty level, categorized into three groups (<10%, 20-19%, ≥20%) Urbanization (Rural–urban continuum definition per US Dept. of Agriculture, categorized into metro counties and non-metro counties)

Incidence

[36]Markossian, 2012 (United States)

SEER 1992-2007

Cross-sectional; Females; ≥15 years old; n= 23,500 cases; 69% white, 31% black

Georgia (15 counties; County

Urban/Rural residence (County-level urban/rural residence)

Mortality