Embed Size (px)
Citation preview
Racial/ethnic inequality amongchildren: Implications for healthdisparities across the lifecourse
Dolores Acevedo-Garcia, PhD, MPA-URPAssociate Professor
With funding from the W.K. Kellogg Foundation
Outline
• Influences on child health and development– Neighborhood effects
• Racial/ethnic inequality in neighborhood/school environments
• How do we integrate them?
A definition of children’s health
Children’s health is the extent to whichindividual children or groups of children areable or enabled to (a) develop and realizetheir potential, (b) satisfy their needs, and (c)develop the capacities that allow them tointeract successfully with their biological,physical, and social environments.
(IOM, 2004, Children’s Health: The Nation’s Wealth)
Racial/ethnic composition ofUS child population, 1980-2009
74 6961 56
9 1217
22
0
10
20
30
40
50
60
70
80
90
100
1980 1990 2000 2009CPS
White Black Hispanic/Latino Asian/PI Native American Other
Acevedo-Garcia, Soobader & Berkman, 2005, The differential effect of foreign-born status on low-birthweight byrace/ethnicity and education. Pediatrics.
Acevedo-Garcia, Soobader & Berkman, 2007, Low birthweight among US Hispanic/Latino subgroups: The effectof maternal foreign-born status and education. Social Science & Medicine.
Low birthweight rates by race/ethnicity, nativity and maternal education, US 1998
0.02.0
4.06.08.0
10.012.0
14.016.0
US-born Foreign-born US-born Foreign-born US-born Foreign-born
Whites Blacks Latinos
0-11 years 12 years 13-15 years 16+ years
Low birthweight rates by race/ethnicity,nativity, and maternal education, US 2002
0-11yrs 12yrs 13-15 yrs 16+ yrs
Influences on Child Healthand Development
ChildSchoolFamily
NeighborhoodNeighborhoodAdultHealth,
education, productivity,
earnings
[M]ost of the prevalent conceptual frameworks [ofchild development] do not emphasize the socialstratification system, or the social positions thatcomprise the scaffolding or structure of the system(i.e., social class, ethnicity, and race) and theprocesses and consequences that these relativepositions engender for a child's development. Thisshortcoming is found even in most of the contextuallybased theoretical frameworks identified in thedevelopmental literature as organizational,transactional, and ecological.
(Garcia-Coll et al 1996)
Neighborhood, school and familyenvironments matter for child health
and development
• Science of brain development in early childhood hasfocused on family environment (e.g. parenting).
• Relative little knowledge about influence of largerenvironments such as neighborhoods, andinteraction effects (e.g. between neighborhoods andparenting).
Neighborhood effects on childhealth and development
• Increasing evidence but largely fromcross-sectional studies (selectionproblems).
• Sampling frames (e.g. central citysamples) limited for understandingdistributional issues– May mis-estimate neighborhood effects
and disparities
Children in unsafe neighborhoods more likely to be in fair/poor health and have emotional problems
NSCH 2003 (adjusted odds ratios) Larson et al., 2008, Pediatrics
1.00 1.00
1.39 1.40
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
Fair/poor health Emotional problems
Safe neighborhood Unsafe neighborhood
Percent of children who are always/usually safe in their neighborhood by race/ethnicity:
NSCH 2005
91.7
68.969.8
0102030405060708090100
White Black Hispanic
MTO – strongest evidence onneighborhood effects on child health
(Acevedo-Garcia et al. 2004; Turner & Acevedo-Garcia, 2005; Acevedo-Garcia & Osypuk 2008)
• Evidence from Moving to Opportunity, ahousing desegregation program, suggeststhat moving from low- to high-povertyneighborhoods can improve mental health.– Only experimental evidence of neighborhood
effects on health;– Remarkable effects given that health was not
addressed explicitly.
Housing mobility benefits on girls: NCSR-AS lifetime generalized anxiety disorder scale
(Moving To Opportunity cross-site evaluation, ITT 4-7 year effects)
9.1
4.9
0
2
4
6
8
10
High-poverty neighborhood (control) Low-poverty neighborhood (treatment)
Housing mobility benefits: Percent feeling safe at night (Moving To Opportunity cross-
site evaluation, ITT 4-7 year effects)
55.0
69.0
01020304050607080
High-poverty neighborhood (control) Low-poverty neighborhood (treatment)
Distributional aspects andneighborhood environment
• Large racial/ethnic disparities in cross-sectional distributions of neighborhood/schoolenvironment.
(Acevedo-Garcia et al. 2008, Osypuk et al. In Press)
• Large racial/ethnic disparities in longitudinaltrajectories of neighborhood/schoolenvironment.
(Quillian 2003, Sampson and Sharkey, 2008, Phuong Do 2009)
• Disparities in neighborhood environmentlinked to child outcomes.
(Garcia-Coll 1996, Caughy and O’Campo 2006, Acevedo-Garcia et al. 2008)
There are large racial/ethnicinequalities children’s access to
“opportunity neighborhoods/schools”
Black and Latino children experiencedouble and triple jeopardy
• Black and Hispanic children live in families thatdisproportionately experienced disadvantage.
• Disparities among individuals and families areexacerbated by vast inequalities inneighborhood and school environments.
• Inequalities go far beyond what can beexplained by income differences:– Poor black and Hispanic children encounter
environments considerably worse than poor whiteand Asian children.
Pyramid Graph:Hypothetical Equal Neighborhood Environment
for 2 Groups: A Mirror Image
100 75 50 25 25 50 75 100
0-10%
10.1-20%
20.1-30%
30.1-40%
Over 40% WhiteBlack
NeighborhoodPoverty Rate
Share of Children in Neighborhoods with Specified Poverty Rates
0
Metro ChicagoPoverty Composition of Neighborhoods of
Black v. White Children
100 75 50 25 25 50 75 100
0-10%
10.1-20%
20.1-30%
30.1-40%
Over 40% WhiteBlack
NeighborhoodPoverty Rate
Share of Children in Neighborhoods with Specified Poverty Rates
0
Metro ChicagoPoverty Composition of Neighborhoods of
Poor Black v. Poor White Children
100 75 50 25 25 50 75 100
0-10%
10.1-20%
20.1-30%
30.1-40%
Over 40% Poor WhitePoor Black
Share of Children in Neighborhoods with Specified Poverty Rates
NeighborhoodPoverty Rate
0
Metro ChicagoPoverty Composition of Neighborhoods of
All Black v. Poor White Children
100 75 50 25 25 50 75 100
0-10%
10.1-20%
20.1-30%
30.1-40%
Over 40% Poor WhiteAll Black
Share of Children in Neighborhoods with Specified Poverty Rates
NeighborhoodPoverty Rate
0
Not only are average values of indicatorsof neighborhood environment worse forblack and Hispanic children, but theentire distribution is shifted in a worsedirection for blacks and Hispanics.
Quantifying Separate and Unequal: Racial/EthnicDistributions of Neighborhood Poverty…
Osypuk, Galea, McArdle, Acevedo-Garcia, In Press. Urban Affairs Review.
Neighborhood % poverty
Whi
te 3rd
qua
rtile
Min
ority
1st q
uarti
le
Is there IQR overlap?
Whites Minorities
0% 40% 30% 10% 20%
Pop
ulat
ion
Siz
e
Minority IQR White
IQR
White and Black Children Overlap for Neighborhood Poverty: 2000
Metro Area
Tra
ct %
Pov
erty
Non-Hisp. White Non-Hisp. Black
White and Hispanic Children Overlap for Neighborhood Poverty: 2000
Metro Area
Non-Hisp. White Hispanic
Tra
ct %
Pov
erty
1.4
16.8
20.5
White Black Hispanic
Share of Children Who Experience Double Jeopardy:Live in BOTH Poor Families and Poor Neighborhoods (Percent)
Source: Acevedo-Garcia, Osypuk, McArdle & Williams, 2008
Note: Poor neighborhoods are those with poverty rates over 20%.Source: 2000 Census
Black children much more likely than whitechildren to experience “double jeopardy”Source: Acevedo-Garcia, Osypuk, McArdle & Williams, 2008
Proportion of white and black children experiencing "double jeopardy" by segregation level, Metropolitan areas, 2000
1.0% 0.8% 1.4%
26.4%
14.6%
10.0%
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
High segregation Medium segregation Low segregation
White children Black children
Latino children much more likely than whitechildren to experience “double jeopardy”Source: Acevedo-Garcia, Osypuk, McArdle & Williams, 2008
Proportion of white and Hispanic/Latino children experiencing "double jeopardy" by segregation level, Metropolitan areas, 2000
3.5%1.5% 1.2%
25.1%
10.8%
5.0%
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
High segregation Medium segregation Low segregation
White children Latino children
Share of Students Eligible for Free/Reduced Lunch
Source: National Center for Education Statistics, Common Core of Data, 2006-07.
Share of Students Eligible for Free/Reduced Lunch
Source: National Center for Education Statistics, Common Core of Data, 2006-07.
Segregation is associated withdetrimental health outcomes.
Hypersegregation and racialdisparities in preterm birth
Osypuk & Acevedo, 2008, American Journal of EpidemiologyFigure 2. Racial Disparity in Predicted Probability Preterm Birth by Metropolitan Area Hypersegregation and Maternal Age from Final
Model (in percentage points) (random slope model)
0%
2%
4%
6%
8%
10%
12%
14%
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
age
Racia
l disp
arity
in p
redi
cted
pro
babi
lity
. pr
eter
m b
irth
.(blac
k m
inus
whi
te)
Racial disparity innonsegregated MAs
Racial disparityhypersegregatedMAs
5/8/07
What can we do?
• Do we pause?
• Link child development and child equity work.– Effects at individual level plus population wide
inequality patterns• Isolate effect of “race/ethnicity”• Double/triple jeopardy: “black and Latino
children don’t get a break”• Focus on “opportunity”
– Child specific definition of opportunity
What can we do?
• Focus on solutions– Evidence and principles
• Mobility policies, e.g. link housing subsidiesand low-income housing development toopportunity neighborhoods/schools, schoolintegration/school choice programs.
• Place-based policies, e.g. neighborhood foodenvironment, school-based interventions,safeafter-school spaces.
Extra slides
Terminology• Metropolitan (Metro) Areas:
– Comprised of core cities and their surroundingsuburban areas
– Definitions based on strong economic and social ties(commuting patterns)
– Larger than cities or counties.– Often referred to as “regions”
• Racial/ethnic groups:– Our focus: 4 largest racial/ethnic groups in U.S.– Non-Hispanic (NH) white; NH black; Hispanic or Latino;
NH Asian/Pacific Islander• Children:
– Under age 18
What we know• Neighborhood, school and family environments matter for
child health and development.
• Children are highly segregated across neighborhoods andschools.– Segregation is associated with detrimental health
outcomes.• There are large racial/ethnic inequalities in children’s access
to “opportunity neighborhoods” and “opportunity schools”—aresult of segregation.– Not accounted for by income differences
• Moving from high-poverty to low-poverty neighborhoods mayimprove child health.
Black/Hispanic Students Attend Schools with Dramatically Different RacialCompositions Than Those of White Students
(Percent of Students Attending Schools by Black/Hispanic Share of Enrollment: 2006-07)
Source: National Center for Education Statistics, Common Core of Data, 2006-07.
Percent
Racial residential segregation is at the rootof black-white health disparities
Acevedo-Garcia 2000, 2001; Acevedo-Garcia & Lochner 2003; Acevedo-Garcia et al. 2003, 2004, 2008;Acevedo-Garcia & Osypuk, 2008; Osypuk & Acevedo-Garcia, 2008; Subramanian, Acevedo-Garcia & Osypuk,
2005; Williams & Collins 2001
• Segregation constrains socioeconomicadvancement of minorities,
• Increases exposure of minorities to unfavorableneighborhood environments,
• Leads to segregation in health care settings, whichis associated with disparities in quality oftreatment.
Pathways between neighborhoodenvironment and health
(Turner & Acevedo-Garcia, 2005)
• Lack of safety/violence• Stress• Health behaviors:
• Targeting (fast food, tobacco)• Neighborhood physical and social environment
(playground quality, walkability, safety)• Peer networks
• Environmental pollutants• Long term effects of limited access to high
quality education
