Income diversity and neighborhood variation in low birth weight rates, Chicago, 1990-2006: Results using longitudinal and cross-sectional measures

doi:10.4236/ojpm.2013.37061

Paper Menu >>

Journal Menu >>

Vol.3, No.7, 454-459 (2013) Open Journal of Preventiv e Me dic ine

http://dx.doi.org/10.4236/ojpm.2013.37061

Income diversity and neighborhood variation in low

birth weight rates, Chicago, 1990-2006: Results using

longitudinal and cross-sectional measures

Jessica Kubo1, Diana S. Grigsby-Toussaint2*

1Department of Statistics, University of Illinois, Urbana Champaign, USA

2Department of Kinesiology and Community Health and Division of Nutritional Sciences, University of Illinois, Urbana Champaign,

USA; *Corresponding Author: [email protected]

Received 27 July 2013; revised 2 September 2013; accepted 19 September 2013

Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is

properly cited.

ABSTRACT

Although increased risk for adverse birth out-

comes has been associated with neighborhood

socioeconomic disadvantage, most studies have

used cross-sectional measures to account for

neighborhood context. Consequently, dynamic

neighborhood processes that may influence

adverse birth outcomes are not fully understood.

In this study, a longitudinal measure of socio-

economic change was used to explore variation

in low birth weight (LBW) rates between 1990

and 2006 in Chicago neighborhoods. A cross-

sectional measure of neighborhood socioeco-

nomic characteristics was then used to compare

the LBW rates across Chicago neighborhoods

during the same time frame to determine whether

the cross-sectional measure would capture the

same nuances in LBW variation as the longitu-

dinal measure. Consistent with previous studies,

both measures identified higher low birth weight

rates in neighborhoods entrenched in poverty

during the study period. However, the longitu-

dinal measure showed that mothers residing in

low income neighborhoods with high concen-

trations of immigrants had LBW rates that were

lower than mothers residing in high income

neighborhoods. Our results suggest that while

cross-sectional measures of neighborhood so-

cioeconomic context may capture global varia-

tions in low birth weight rates, longitudinal

measures may illuminate subtleties between

neighborhoods that might provide an opportu-

nity for targeted policies to reduce adverse ma-

ternal and child health outcomes.

Keywords: Low Birth Weight; Neighborhood;

Socioecon omic Status

1. INTRODUCTION

In the United States (US), low birth weight (LBW),

defined as the percentage of babies weighing <2500

grams (5.5 lbs) at birth, persists as a major public health

problem [1,2]. Between 1970 and 2008, LBW prevalence

in the US increased from 7.93% to 8.18% [1,2]. During

this time frame, racial disparities in LBW also persisted,

with African-American LBW rates consistently almost

twice that of Whites [1]. Several studies have demon-

strated that socioeconomic characteristics of neighbor-

hoods may mediate these observed disparities [3-9]. In

addition, higher levels of neighborhood deprivation, usu-

ally measured using constructs of education, income,

poverty, or unemployment, have been found to be pre-

dictive of LBW and other adverse birth outcomes [3-7,

9-11].

Notwithstanding the fact that these studies do capture

variations in birth outcomes, a major limitation is the use

of single decennial census estimates to measure neighbor-

hood context. Although several studies demonstrate that

longitudinal measures of neighborhood context may bet-

ter explain variations in health [12,13], most studies ex-

amining neighborhood influences on LBW and other

adverse birth outcomes use cross-sectional measures [5,

6,14-17]. In a study by Do [12], results suggest that lon-

gitudinal measures of neighborhood context are particu-

larly salient for disentangling health disparities between

Blacks and Whites. As such, using cross-sectional meas-

ures to investigate differences in birth outcomes between

Blacks and Whites in neighborhoods in the US may un-

derestimate or misestimate the gap in LBW rates, par-

J. Kubo, D. S. Grigsby-Toussaint / Open Journal of Preventive Medicine 3 (2013) 454-459 455

ticularly given dynamic neighborhood processes that

might change over time. Neighborhood disinvestment,

for example, may impact social networks or access to

health service agencies that conceivably influence ma-

ternal stress, and as a consequence, increase poor birth

outcomes. Conversely, neighborhood revitalization may

improve access to better quality foods, and a higher tax

base for resources that could result in a more supportive

environment for health-promoting behaviors for expec-

tant mothers, thus, reducing LBW risk. In the context of

public health planning and surveillance, these neighbor-

hood processes are critical for targeting and tracking

interventions, particularly in major metropolitan areas of

the US that have experienced substantial changes in so-

cioeconomic composition over the last three decades

[18,19]. Thus, incorporation of a longitudinal assessment

metric of neighborhood processes could provide a more

complete picture in understanding LBW risk.

In this study, we use a longitudinal measure of neigh-

borhood socioeconomic change in the city of Chicago to

examine associations with trends in LBW rates. Time

trends in LBW are examined as well as mean LBW over

the study period 1990-2006. We compare the longitudi-

nal measure of neighborhood characteristics to a cross-

sectional measure of neighborhood characteristics based

on quintiles of income categories from the US 2000 cen-

sus.

2. METHODS

2.1. Low Birth Weight

Publically available de-identified vital statistics data

from the Illinois Project for Local Assessment of Needs

(IPLAN) website was used for this study. The IPLAN

website serves as a repository for102 health indicators

used to assist health departments with community as-

sessments and five-year planning goals [20]. The Center

for Health Statistics at the Illinois Department of Public

Health maintains the IPLAN website, updating vital sta-

tistics data based on birth certificate records reported by

local health departments across the state. Total counts of

live births and LBW infants born to Black and White

women in each of Chicago’s 77 community areas (i.e.,

neighborhoods) between 1990 and 2005 were abstracted

from the IPLAN database. IPLAN defines low birth

weight using the conventional definition of weight at

birth of less than 2500 grams [21]. The final sample size

consisted of 804,289 births; 350,681 to Black mothers,

and 453,428 to White mothers.

2.2. Neighborhood Socioeconomic Context

2.2.1. Definition of Neighborhoods

The administrative unit used for public health planning

in the city of Chicago is the community area, which is

used as a proxy for neighborhoods in this study [22].

Designated in the 1920’s by the Social Science Research

Council at the University of Chicago, these 77 areas

were intended to reflect the cultural and social history of

Chicago [23]. On average, community areas in Chicago

consist of 3 to 4 census tracts and the population count

may range from approximately 3000 to 120,000. On av-

erage, each community area has 37,000 residents [22].

2.2.2. Measure of Socioeconomic Change

An income diversity index of neighborhood socioeco-

nomic context developed by the Metro Chicago Informa-

tion Center (MCIC), an official census center, was used

to characterize neighborhoods. To develop the income

diversity index, the MCIC used data available from the

Neighborhood Change Database (NCDB), a commer-

cially available database of social, demographic, eco-

nomic and housing data on census tracts in the US for

1970, 1980, 1990, and 2000 [24]. The purpose for de-

veloping the index was to examine trends in the socio-

economic composition of neighborhoods over time in

order to develop sound community development policies

[25].

Neighborhoods were categorized as stable diversity,

emerging low income, emerging high income, desertifi-

cation, and emerging bipolarity based on patterns of

economic change in Chicago community areas over 30

years [25]. Briefly, stable diversity neighborhoods (n =

19) consist of community areas that have maintained a

socioeconomically diverse population between 1970 and

2000. Emerging low income neighborhoods (n = 11) have

experienced a loss of high income families, while the

reverse occurred with emerging high income neighbor-

hoods (n = 21), where the number of low income fami-

lies is decreasing. Desertification neighborhoods (n = 11)

show patterns of entrenched levels of poverty with a

predominantly African-American population. Finally,

emerging bipolarity neighborhoods (n = 15) show an

increase in both high and low income residents. Table 1

summarizes the characteristics of each type of neighbor-

hood based on the index using US Census 2000 data.

(For detailed methodology on the MCIC income diver-

sity index, see http://www.mcic.org).

2.2.3. Quintiles of Low Income Families

The percent of families considered to be low income

in 2000 was obtained from the MCIC database for each

of the 77 Chicago neighborhoods. These percentages

were used to divide the 77 neighborhoods into quintiles.

The first quintile, which contains the lowest percentage

of low income families, has 17 neighborhoods. The sec-

ond has 16, the third 15, the fourth 14, and the fifth 15.

J. Kubo, D. S. Grigsby-Toussaint / Open Journal of Preventive Medicine 3 (2013) 454-459

456

2.2.4. Neighborhood Demographics

Census 2000 data was obtained for the 77 Chicago

neighborhoods from the Chicago Metropolitan Agency

for Planning (http://www.cmap.illinois.go v/). Race/eth-

nicity, percent college educated, percent foreign born,

median income, percent in poverty, median property

value, percent female-headed households, and percent

unemployment were calculated using neighborhood-level

data and aggregated by each set of indices used—the IDI

and low income quintiles.

2.3. Statistical Analyses

Race-stratified mean LBW rates were calculated using

counts of LBW and total live births for each neighbor-

hood type based on the income diversity index and for

each quintile of low income families in 2000. Simple

linear regression was performed for each income diver-

sity index category and each quintile of low income. The

year (1990-2006) was used as the predictor, with percent

of LBW as the response. R2 values from simple linear

regression, as well as estimates and p-values of the in-

tercept and predictors are reported. All analyses were

performed using SAS 9.1 (SAS Institute Inc., Cary, NC).

3. RESULTS

Demographics from the 2000 census were compared

for neighborhoods aggregated using both index methods

(IDI and low-income quintiles) in Table 1. It is observed

that the range of median incomes in thousands of dollars

is lower (28.7 versus 34.4) when using the IDI than when

using income quintiles. Desertification neighborhoods

are 96.2% Black (quintiles range from 12.2 in 0 - 20th

quintile to 80.7 in 80 - 100th quintile). Also, while the 60

- 80th quintile is 61.7% Black, emerging low income

neighborhoods are 37.9% Black.

Between 1990 and 2006, mean LBW rates for Whites

ranged from 6.38% (95% CI, 6.05, 6.71) in emerging

low income neighborhoods to 7.82% (95% CI, 7.07, 8.57)

in desertification neighborhoods (Table 2). Linear re-

gression (Table 3) indicates significant increases in

LBW rates in emerging low income, emerging bipolarity,

stable diversity, and emerging high income neighbor-

hoods during the study period. LBW ranged from 6.61%

(95% CI, 6.18, 7.04) in the 0 - 20 quintile to 6.83% (95%

CI, 6.47, 7.20) in the 80 - 100 quintile during the study

period. Linear regression results show a significant in-

crease in LBW for all quintiles at the 95% level of sig-

nificance.

The mean LBW rates for Black mothers ranged from

13.9% (95% CI, 13.5, 14.3) in emerging low income

neighborhoods to 16.3% (95% CI, 15.9, 16.7) in deserti-

fication neighborhoods (Table 2). Overall, LBW rates

for this demographic showed a downward trend between

Table 1. Selected demographic characteristics of neighborhood socioeconomic context using the US Decennial Census 2000 esti-

mates.

Income

Diversity

Index

% foreign

born

% of adults

with a college

education

% White

non-Hispanic

% Black

non-Hispanic % Hispanic

Median

property value

in $1000s

Median

household income

in $1000s

% of families

below poverty

rate

Desertification

N = 11 1.9 13% 1.6 96.2 1.6 105.6 18.9 39.1

Emerging Low

Income

N = 11

28.2 16% 31.9 37.9 44.8 104.9 34.0 18.4

Emerging

Bipolarity

N = 15

22.5 26% 50.0 30.0 23.0 160.0 39.2 13.6

Stable Diversity

N = 19 21.8 27% 33.9 45.0 20.9 147.8 40.2 13.6

Emerging High

Income

N = 21

19.1 39% 63.6 18.9 20.2 222.6 47.6 10.7

US Census 2000

Income Quintile

% foreign

born

% of adults with

a college

education

% White

non-Hispanic

% Black

non-Hispanic % Hispanic

Median

property

value $

Median

household

income

% of

families below

poverty rate

80 - 100

N = 17 8.8 13% 8.4 80.7 15.5 102.0 21.4 35.0

60 - 80

N = 16 17.0 24% 19.1 61.7 18.3 129.1 31.5 21.5

40 - 60

N = 15 27.8 23% 39.3 28.9 41.4 159.9 37.3 16.4

20 - 40

N = 14 25.8 27% 51.9 28.8 23.0 150.8 41.3 10.5

0 - 20

N = 15 16.7 43% 75.6 12.2 11.7 238.3 55.8 4.9

J. Kubo, D. S. Grigsby-Toussaint / Open Journal of Preventive Medicine 3 (2013) 454-459 457

Table 2. Mean low birth weight rate by income diversity index and US Census 2000 quintile of income families for White, Black and

combined mothers.

White Black Combined

All Neighborhoods 6.78 (6.48, 7.08) 15.27 (14.93, 15.61) 10.44 (10.12, 10.76)

Income Diversity Index Mean (95% CI) Mean (95% CI) Mean (95% CI)

Desertification 7.82 (7.07, 8.57) 16.30 (15.87, 16.73) 15.75 (15.41, 16.10)

Emerging Low Income 6.38 (6.05, 6.71) 13.89 (13.51, 14.27) 8.11 (7.84, 8.39)

Emerging Bipolarity 7.15 (6.76, 7.55) 15.27 (14.93, 15.61) 10.86 (10.56, 11.16)

Stable Diversity 6.66 (6.34, 6.99) 14.55 (14.17, 14.93) 10.41 (10.04, 10.77)

Emerging High Income 6.85 (6.53, 7.18) 15.27 (14.83, 15.72) 8.99 (8.71, 9.26)

White Black Combined

US Census 2000 Income Quintile Mean (95% CI) Mean (95% CI) Mean (95% CI)

80 - 100 6.83 (6.47, 7.20) 15.97 (15.56, 16.38) 13.26 (12.90, 13.61)

60 - 80 6.83 (6.50, 7.17) 14.92 (14.55, 15.29) 11.43 (11.10, 11.75)

40 - 60 6.87 (6.57, 7.17) 15.08 (14.64, 15.52) 9.24 (8.93, 9.55)

20 - 40 6.76 (6.43, 7.09) 14.60 (14.07, 15.13) 8.61 (8.25, 8.96)

0 - 20 6.61 (6.18, 7.04) 14.37 (13.89, 14.84) 7.86 (7.53, 8.20)

Table 3. Results of linear regression of low birth weight rates on years for White, Black, and combined mothers.

White Black Combined

All Neighborhoods 0.0007 <0.0001 −0.0006 0.0127 −0.0007 0.0004

Income Diversity Index Coefficient P-value Coefficient P-value Coefficient P-value

Desertification 0.0012 0.3967 −0.0012 0.0009 −0.0005 0.0362

Emerging Low Income 0.0007 0.0029 −0.0001 0.7692 0.0003 0.0636

Emerging Bipolarity 0.0011 0.0002 −0.0002 0.5674 −0.0002 0.4448

Stable Diversity 0.0006 0.0032 −0.0002 0.5168 −0.0007 0.0203

Emerging High Income 0.0007 0.0009 −0.0004 0.3986 −0.0004 0.0302

White Black Combined

US Census 2000 Income Quintile Coefficient P-value Coefficient P-value Coefficient P-value

80 - 100 0.0006 0.0474 −0.0008 0.0146 −0.0007 0.0075

60 - 80 0.0006 0.0102 −0.0004 0.2330 −0.0003 0.1360

40 - 60 0.0004 0.0254 0.0001 0.8527 −0.0002 0.2588

20 - 40 0.0006 0.0035 −0.0006 0.3283 −0.0004 0.1227

0 - 20 0.0014 <0.0001 −0.0007 0.1502 0.0007 0.0014

1990 and 2006 (P < 0.05), with significant decreases in

LBW rates for desertification neighborhoods. Black

LBW rates ranged from 14.4% (95% CI, 13.9, 14.8) in

the 0 - 20 quintile to 16.0 (95% CI, 15.6, 16.4). A

non-significant increase in LBW rates was noted for the

40 - 60 quintile; non-significant decreases occurred in

the 0 - 20, 20 - 40, and 60 - 80 quintile. Neighborhoods

in the 80 - 100 quintile experienced a significant de-

crease in LBW rates.

Regression results for White mothers were significant

for all quintiles, however, for desertification neighbor-

hoods no significant time trend was observed. For Black

J. Kubo, D. S. Grigsby-Toussaint / Open Journal of Preventive Medicine 3 (2013) 454-459

458

mothers only the 80 - 100 quintile with the highest per-

centage of low income families had significant regres-

sion results; similarly, using the IDI only desertification

neighborhoods had significant time trends. Combined

results also showed a difference in the two index meth-

ods—for Black and White mothers, desertification, sta-

ble diversity, and emerging high income had significant

time trends; only the 0 - 20 and 80 - 100 quintiles had

significant time trends. Further, the 0 - 20 quintile with

the least percentage of low income families had a posi-

tive slope; emerging high income neighborhoods had a

negative slope.

4. DISCUSSION

Consistent with other studies exploring adverse birth

outcomes and neighborhood context using cross-sectional

measures, we found that Black and White women resid-

ing in neighborhoods entrenched in poverty, with high

concentrations of Blacks, were more likely to have higher

LBW risk [3,4]. Additionally, consonant with national

estimates, LBW rates for Blacks were consistently dou-

ble those of Whites for the study period. Notwithstanding,

it is interesting to note that emerging low income neigh-

borhoods have the lowest LBW rates of all neighbor-

hoods and seem to confer some protection against LBW

risk for both Black and White mothers. This finding may

be due to high concentrations of Latino immigrants in

these neighborhoods (Table 1), where some studies have

suggested that better social support may attenuate the

impact of limited material resources on health [26].

Higher LBW risk for Black and White mothers residing

in emerging high income neighborhoods may be due to

more highly educated women residing in these neighbor-

hoods who tend to be older at first birth [27]. As a con-

sequence of being older at first birth, these women may

also be more likely to use assisted reproductive technol-

ogy that has also been shown to be associated with high-

er LBW risk [28].

Comparing the two classification schemes, aggregated

demographics over 77 neighborhoods using the US cen-

sus 2000 estimates shows an increasing gradient of per-

cent Black, percent in poverty and proportion of female

headed households and a decreasing gradient of median

income from the lowest quintile of low income house-

holds to the highest. The IDI, on the other hand, shows

desertification neighborhoods in stark contrast with other

indices for these measures. The IDI as a longitudinal

measure appears to better capture the entrenched poverty

of the desertification neighborhoods, while the method of

income quintiles misses this in the quintile with the high-

est low income families.

To our knowledge, few studies have undertaken inves-

tigations of neighborhood influences on birth outcomes

using a longitudinal measure of socioeconomic context.

The income diversity index is particularly unique as it

allows us to examine patterns in LBW risk in Chicago

neighborhoods while accounting for the influence of

neighborhood revitalization and immigration patterns,

thus providing a more nuanced view of neighborhood

influences on LBW. In addition, the income diversity

index was specifically designed to address issues related

to community development for the city of Chicago, thus

it may serve as a more practical tool for public health

planning to address LBW risk.

This study was not without its limitations. First, the

use of vital statistics data at the neighborhood level

without individual level covariates limited our ability to

account for potential confounders. For example, infor-

mation on mothers’ length of residence would have im-

proved our ability to better quantify the influence of

neighborhoods on LBW risk. Second, while we had data

on racial categories of the mothers in our study, we did

not have data on ethnicity. As such, some of the women

in our population may be misclassified as Black or White,

when their ascribed status was Hispanic. This could ac-

count for the lower LBW rates for both Blacks and

Whites in the emerging low income neighborhoods in

our study. Third, while our measure of neighborhood

context accounts for socioeconomic change over a thirty-

year period, it is primarily based on a measure of family

income, which is only one aspect of neighborhoods that

may influence health.

The goal of this study was to use a longitudinal meas-

ure of neighborhood context based on population-based

data for LBW in the city of Chicago. While our results

corroborate previous findings [8,10] that neighborhoods

with high concentrations of African-Americans and high

levels of poverty have higher rates of LBW, it also pro-

vides a nuanced view of the impact of neighborhood

immigrant and revitalization patterns on the birth out-

comes of Black and White residents. Thus, measures that

seek to capture socioeconomic characteristics of neighbor-

hoods over time may provide better insight for targeted

policies to reduce adverse health outcomes as part of

city-wide public health planning efforts.

REFERENCES

[1] Centers for Disease Control and Prevention (2008) Health,

United States, 2008 with chart book on trends on the

health of Americans. National Center for Health Statistics,

Hyattsville.

[2] Martin, J.A., Hamilton, B.E., Ventura, S.J., Osterman,

M.J.K, Wilson, E.C. and Mathews, T.J. (2012) Births:

Final data for 2010. National Vital Statistics Reports, 61,

National Center for Health Statistics, Hyattsville.

[3] Roberts, E.M. (1997) Neighborhood social environments

and the distribution of low birthweight in Chicago. Ame-

rican Journal of Public Health, 87, 597-603.

J. Kubo, D. S. Grigsby-Toussaint / Open Journal of Preventive Medicine 3 (2013) 454-459

459

http://dx.doi.org/10.2105/AJPH.87.4.597

[4] O’Campo P., Burke, J.G., Culhane, J., Elo, I.T., Eyster, J.,

Holzman, C., Messer, L.C., Kaufman, J.S. and Laraia,

B.A. (2008) Neighborhood deprivation and preterm birth

among non-Hispanic Black and White women in eight

geographic areas in the United States. American Journal

of Epidemiology, 167, 155-163.

http://dx.doi.org/10.1093/aje/kwm277

[5] Buka, S.L., Brennan, R.T., Rich-Edwards, J.W., Rauden-

bush, S.W. and Earls, F. (2003) Neighborhood support

and the birth weight of urban infants. American Journal

of Epidemiology, 157, 1-8.

http://dx.doi.org/10.1093/aje/kwf170

[6] Messer, L.C., Kaufman, J.S., Dole, N., Savitz, D.A. and

Laraia, B.A. (2006) Neighborhood crime, deprivation,

and preterm birth. Annals of Epidemiology, 16, 455-462.

http://dx.doi.org/10.1016/j.annepidem.2005.08.006

[7] Subramanian, S.V., Chen, J.T., Rehkopf, D.H., Waterman,

P.D. and Krieger N. (2006) Comparing individual- and

area-based socioeconomic measures for the surveillance

of health disparities: A multilevel analysis of Massachu-

setts births, 1989-1991. American Journal of Epidemiol-

ogy, 164, 823-834.

http://dx.doi.org/10.1093/aje/kwj313

[8] Collins, J.W. and David. R.J. (2009) Racial disparity in

low birth weight and infant mortality. Clinics in Perina-

tology, 36, 63-73.

http://dx.doi.org/10.1016/j.clp.2008.09.004

[9] Blumenshine, P., Egerter, S., Barclay, C.J., Cubbin, C. and

Braveman, P.A. (2010) Socioeconomic disparities in ad-

verse birth outcomes. American Journal of Preventive

Medicine, 39, 263-272.

http://dx.doi.org/10.1016/j.amepre.2010.05.012

[10] Cubbin, C., Marchi, K., Lin, M., Bell, T., Marshall, H.,

Miller, C. and Braveman, P. (2008) Is neighborhood dep-

rivation independently associated with maternal and in-

fant health? Evidence from Florida and Washington. Ma-

ternal and Child Health Journal, 12, 61-74.

http://dx.doi.org/10.1007/s10995-007-0225-0

[11] Farley, T.A., Mason, K., Rice, J., Habel, J.D., Scribner, R.

and Cohen, D.A. (2006) The relationship between the

neighbourhood environment and adverse birth outcomes.

Paediatric and Perinatal Epidemiology, 20, 188-200.

http://dx.doi.org/10.1111/j.1365-3016.2006.00719.x

[12] Do, D.P. (2009) The dynamics of income and neighbor-

hood context for population health: do long-term meas-

ures of socioeconomic status explain more of the black/

white health disparity than single-point-in-time measures?

Social Science & Medicine, 68, 1368-1375.

http://dx.doi.org/10.1016/j.socscimed.2009.01.028

[13] Benzeval, M. and Judge, K. (2001) Income and health,

the time dimension. Social Science & Medicine, 52, 1371-

1390. http://dx.doi.org/10.1016/S0277-9536(00)00244-6

[14] Reagan, P.B. and Salsberry, P.J. (2004) Race and ethnic

differences in determinants of preterm birth in the USA:

Broadening the social context. Social Science & Medicine,

60, 2217-2228.

http://dx.doi.org/10.1016/j.socscimed.2004.10.010

[15] Kaufman, J.S., Dole, N., Savitz, D.A. and Herring, A.H.

(2003) Modeling community-level effects on preterm

birth. Annals of Epidemiology, 13, 377-384.

http://dx.doi.org/10.1016/S1047-2797(02)00480-5

[16] Pickett, K.E, Collins Jr., J.W., Masi, C.M. and Wilkinson,

R.G. (2005) The effects of racial density and income in-

congruity on pregnancy outcomes. Social Science & Me-

dicine, 60, 2229-2238.

http://dx.doi.org/10.1016/j.socscimed.2004.10.023

[17] Masi, C.M., Hawkley, L.C., Piotrowski, Z.H. and Pickett,

K.E. (2007) Neighborhood economic disadvantage, vio-

lent crime, group density, and pregnancy outcomes in a

diverse, urban population. Social Science & Medicine, 65,

2440-2457.

http://dx.doi.org/10.1016/j.socscimed.2007.07.014

[18] Zielenbach, S. (2005) Understanding Community Change:

A look as low-income Chicago neighborhoods in the

1990s. Urban Institute, No. 4.

[19] Kingsley, G.T. and Pettit, K.L.S. (2007) Concentrated

poverty: Dynamics of change. Urban Institute, No. 5.

[20] Illinois Department of Public Health (2006) Vital Statis-

tics Illinois. Illinois Center for Health Statistics, Spring-

field.

[21] World Health Organization (1993) International statistics

classification of diseases and related health problems. 10th

Edition, Vol. 2. WHO, Geneva.

[22] Bocksay, K.A., Harper-Jemison, D.M., Gibbs, K.P., Wea-

ver, K. and Thomas, S.D. (2007) Community area health

inventory part one: Demographic and health profiles.

Health Status Index Series Vol. XVI No. CV, Chicago

Department of Public Health Office of Epidemiology,

Chicago.

[23] Hunter, A. and Janowitz, M. (1984) Symbolic communi-

ties: The persistence and change of Chicago’s local com-

munities. University of Chicago Press, Chicago.

[24] University of Michigan Population Studies Center (2002)

Neighborhood Change Database (NCDB) Tract Data

from 1970-2000.

http://www.psc.isr.umich.edu/dis/data/resource/detail/144

[25] Taylor, D.G. (2007) Income diversity and the context of

community development. Metro Chicago Information Cen-

ter, Chicago.

[26] Acevedo-Garcia, D., Soobader, M.J. and Berkman, L.F.

(2007) Low birthweight among US Hispanic/Latino sub-

groups: The effect of maternal foreign-born status and

education. Social Science & Medicine, 65, 2503.

[27] Edwards, M.E. (2002) Education and occupations: Re-

examining the conventional wisdom about later first

births among American mothers. Sociological Forum, 17,

423-443.

[28] Centers for Disease Control and Prevention (2009) As-

sisted reproductive technology and trends in low birth-

weight—Massachusetts, 1997-2004. Morbidity and Mor-

tality We ekly Report, 58, 49-52.