|
|
Citation:
Black JL, Macinko J. Neighborhoods and obesity. Nutr Rev. 2008 Jan;66(1):2-20.
PubMed ID: 18254880
Study Design:
Systemic Review
Class:
M - Click here for explanation of classification scheme.
Research Design and Implementation Rating:
 POSITIVE: See Research Design and Implementation Criteria Checklist below.
Research Purpose:
This review summarizes the literature on neighborhood determinants of obesity.
Inclusion Criteria:
- Outcome variables including a measure of body weight, physical activity or diet
- Independent variables including a neighborhood-level measure or assessment of a social, behavioral, or demographic predictor of obesity
- The study was conducted in a human population in an industrialized country.
- Only English-language articles were reviewed.
Exclusion Criteria:
all others not meeting inclusion criteria.
Description of Study Protocol:
Search procedure
The literature review was conducted from August 2005 through March 2007 by systemically searching the PubMed and PsychInfo databases.
Both databases were searched with the following keywords in their title or abstract: "neighborhood AND obesity."
The following combinations of keywords were searched in abstracts and titles: "obesity" AND "multilevel"; "SES"; "income"; "income inequality"; "race"; "supermarket"; "grocery store"; "fast food"; "farmers market"; "food policy"; "food price"; "restaurant"; "built environment"; "physical activity"; "crime"; and "transportation". The keyword "neighborhood" was also combined with "physical activity", "diet", "race", and "socioeconomic status".
Type of intervention and outcomes investigated
- measure of body weight
- physical activity
- or diet
Data Collection Summary:
Type of information abstracted from articles
Results were grouped according to the major neighborhood characteristics analyzed in the literature.
How was data combined:
- Macro-level social, historical, and economic factors that shape overall neighborhood context
- neighborhood or meso-level living conditions, such as infrastructure and services
- local availability and quality of food
- neighborhood characteristics that promote or inhibit physical activity
Description of Actual Data Sample:
# of articles included: 36 included a specific measure of body weight status or obesity
# of articles identified: 2000 potential articles were identified; 90 of which assessed at least one neighborhood determinant of obesity
Studies of neighborhood- and area-level socioeconomic resources and obesity
| Reference |
Country, location (population sampled) |
Sample size |
Study design |
Neighborhood metric |
Height and weight data |
Body weight outcome(s) |
| Chang (2006) |
USA (MSAs with >10% black) |
46,881 (130 MSAs) |
M |
MSAs |
Self-reported |
Overweight=BMI≥25; obese=BMI≥30 |
| Chen & Paterson (2006) |
USA, St Louis, MO (adolescents) |
315 |
I |
Census block group |
Measured |
BMI |
| Inagami et al (2006) |
USA, Los Angeles County, CA |
2620 (65 NHs) |
M |
Census tract |
Self-reported |
BMI |
| Janssen et al (2006) |
Canada (students in grades 6-10) |
6684 (169 schools) |
M |
5 km Radius around school |
Self-reported |
Obese=BMI≥30 |
| King et al (2006) |
Australia, Melbourne |
4913 (50 NHs) |
M |
Census collector district |
Self-reported |
BMI |
| Mobley et al (2006) |
US States: CT, MA, NE, NC, SD (low-income women) |
2692 (222 NHs) |
M |
Zip code |
Measured |
BMI |
| Monden et al (2006) |
Netherlands, Eindhoven |
8802 (86 NHs) |
M |
Administrative unit |
Self-reported |
Overweight=BMI≥25 |
| Nelson et al (2006) |
USA (students in grades 7-12) |
20,745 |
I |
Constructed via cluster analysis |
Self-reported |
Overweight=BMI≥95th percentile |
| Spillsbury et al (2006) |
USA, Cleveland (African American children) |
843 |
I |
Census tract |
Measured |
BMI percentile for age |
| Boardman et al (2005) |
USA |
402,154 |
M |
"Very small areas" from NHIS |
Self-reported |
Obese=BMI≥30 |
| Vandergrift & Yoked (2004) |
USA |
47 |
E |
State |
Self-reported |
Obesity=% per state with BMI≥30 |
| Robert and Reither (2004) |
USA |
3617 |
M |
Census tract |
Self-reported |
BMI |
| Van Lenthe et al (2002) |
Netherlands, Eindhoven |
8897 (86 NHs) |
M |
Census tract |
Self-reported |
Overweight=BMI≥25 |
| Sundquist et al (1999) |
Sweden |
9240 |
I |
Small area market |
Self-reported |
Overweight and obesity |
| Davey Smith et al (1998) |
Scotland, Renfew and Paisley |
6961 men (7991 women) |
I |
Postcode sector and enumeration district |
Measured |
BMI |
| Ellaway et al (1997) |
Scotland, Glasgow |
691 (4NHs) |
I |
Socially contrasting neighborhoods |
Measured |
Obese=BMI≥30 |
Abbreviations: E, ecologic; I, individual; M, multilevel; MSAs metropolitan statistical area; NHs neighborhoods; NHIS, 1990-1994 National Health Interview Survey
Studies of income equality and obesity
| Reference |
Country,location (population sampled) |
Sample size |
Study design |
Main measure(s) |
Association with BMI/weight status |
Metric of income inequality measure |
Height and weight data |
Body weight outcome(s) |
| Mobley et al (2006) |
USA: CT, MA, NE, NC, SD (low income women) |
2692; 88 NHs |
M |
Income sipersion |
Ø |
County |
Self-reported |
BMI |
| Picket et al (2005) |
Large, high income countries |
21 |
E |
Gini coeeficients, UNDPHP indicators |
+ |
Country |
Pooled data from the International Obesity Taskforce |
Proportion obese (BMI≥30) per country |
| Robert & Reither (2004) |
USA |
3617 |
M |
Gini coefficients |
+ |
Census tract |
Self-reported |
BMI |
| Diez-Roux et al (2000) |
USA |
81,557
44 states
|
M |
Robin Hood Index |
+ for women only |
State |
Self-reported |
BMI |
| Kahn et al (1998) |
USA |
34,158 male; 42,741 female
21 states
|
I |
Household Inequality Index |
+ for men only |
State |
Self-reported |
Self-reported weight gain in waist |
Abbreviations: E, ecologic; I, Individual; M, multilevel; MSAs, metropolitan statistical areas; NHs, neighborhoods
Studies of neighborhood and racial composition and obesity
| Reference |
Country, location (population sampled) |
Sample size |
Study type |
Measure(s) of racial composition |
Association with BMI/weight status |
Neighborhood metric of SES measure |
Height and weight data |
Body weight outcome(s) |
| Chang (2006) |
USA (MSAs with >10% black) |
46,881; 130 MSAs |
M |
Index of racial isolation |
+for blacks; Ø for whites |
MSA |
Self-reported |
BMI; overweight=BMI≥25 |
| Mobley et al (2006) |
USA States: CT, MA, NE, NC, SD (low-income women) |
2692; 88 NHs |
M |
Index or racial segregation |
Ø |
Zip code |
Measured |
BMI |
| Boardman et al (2005) |
USA |
402,154 |
M |
Proportion black |
+ |
"Very small areas" from NHIS |
Self-reported |
Obese=BMI≥30 |
| Robert & Reither (2004) |
USA |
3617 |
M |
Percent black |
Ø |
Census tract |
Self-reported |
BMI |
Abbreviations: E, ecologic; I, Individual; M, multilevel; MSAs, metropolitan statistical areas; NHs, neighborhoods
Studies of neighborhood food availability and obesity
| Reference |
Country, location (population sampled) |
Sample size |
Study type |
Main measure |
Method of measuring food access |
Association with BMI/weight status |
Height and weight data |
Body weight outcomes |
| Inagami et al (2006) |
USA, Los Angeles County, CA |
2620; 65 NHs |
M |
Access to primary grocery store |
Distance between residence and census tract centroid |
+
For father distances
|
Self-reported |
BMI |
| Morland et al (2006) |
USA, states: MS, NC, MD,MN |
10,763; 207 NHs |
M |
Availability of food stores |
Number of food stores per census tract |
-
For supermarkets;
+
for convenience stores
|
Measured |
Overweight=BMI≥25; obese=BMI≥30 |
| Jeffery et al (2006) |
USA, state: MN |
1033 |
I |
Access to restaurants |
Restaurant outlet density within 2 mile radius of work and home |
Ø
for fats food;
-for men with more restaurants near work
|
Self-reported |
BMI |
| Mobley et al (2006) |
USA, states: CT, MA, NE, NC, SD (low-income women) |
2692; 222 NHs |
M |
Availability of food stores |
Density of grocery stores, fast food, restaurants and mini-marts per zip code |
Ø |
Measured |
BMI |
| Sturm and Data (2005) |
USA, (children >4 years old followed until 3rd grade) |
6918; 724 schools; 59 MSAs; 37 states |
M |
Access to food stores |
Distance from home and school zip codes to grocery stores, convenience stores and restaurants and food prices |
Ø
+
for fruit and vegetable price index
|
Measured |
BMI |
| Maddock (2004) |
USA |
50 states |
E |
Availability of fast food |
State-level availability (square miles and populationper outlet) of McDonalds and Burger King |
+ |
State-level aggregates based on self-reported data |
Percent obese (BMI≥30) per state |
| Burdette and Whitaker (2003) |
USA, Cincinnati, OH (3-4-year old children in WIC) |
7020 |
I |
Availability of fast food |
Distance from home to fast-food outlet |
Ø |
Measured |
Overweight=BMI≥95th percentile |
Ø, no significant association; +, positive association; -, negative association
Abbreviations: E, ecologic; I, Individual; M, multilevel; MSAs, metropolitan statistical areas; NHs, neighborhoods
Studies of neighborhood physical activity environment and obesity
| Reference |
Country, location (subpopulation studied) |
Sample size |
Study type |
Type of measure |
Main neighborhood variable(s) |
Metric of neighborhood measure |
Association with BMI/weight status |
Height and weight data |
Body weight outcome |
| Boehmer et al (2007) |
USA, Savannah, GA and St. Louis, MO |
1032 |
I |
Perceived and objective |
Recreation facilities, land use, transportation, aesthetics |
Perceived objective 400 m buffers from residence |
+For perceived lack of destinations, sidewalks and objective poor sidewalk quality, physical disorder, garbage;
Øfor recreation facilities, traffic safety
|
Self-reported |
Obese=BMI≥30 |
| Berke et al (2007) |
USA, King County, WA (older adults 65-97 years) |
936 |
I |
Objective |
Walkability score |
1-3 km buffers from residence |
Ø for walkability |
Measured |
BMI |
| Poortinga (2006) |
England |
14,836; 720 postcodes |
M |
Perceived |
Self-rated local environment features (e.g. access to amenities, physical features, reputation, aesthetics, social support and capital) |
Perceived neighborhood |
+for social nuisances;
- for perceptions of the social environment
|
Measured |
Obese=BMI≥30 |
| Mobley et al (2006) |
USA; CT, MA, NE, NC, SD (low-income women) |
2692; 222 NHs |
M |
Objective |
Land use, fitness facilities per 1000 residents, robbery arrest per 100,000 |
Zip code |
- for mised land use, fitness facilities;
+ for crime
|
Measured |
BMI |
| Gordon-Larsen et al (2006) |
USA (adolescents) |
20,745 |
I |
Objective |
Access to physical activity facilities
|
Block group |
- for increased facilities |
Self-reported |
Overweight=BMI≥95th percentile |
| Nelson et al (2006) |
USA (students grade 7-12) |
20,745 |
I |
Objective |
Access to physical activity facilities, walkability, crime used to define neighborhood clusters |
3-km distance from residence |
+ for rural working class and exurban and mixed-race urban areas |
Self-reported |
Overweight=BMI≥95th percentile |
| Lumeng et al (2006) |
USA children (7019 years) |
768; 10 NHs |
I |
Perceived |
Parental perceptions of neighborhood safety |
Perceived neighborhood |
- for perceived safety |
Self-reported |
Overweight=BMI≥95th percentile |
| Glass et al (2006) |
USA, Baltimore, MD (age50-70 years) |
1140; 65 NHs |
M |
Perceived |
Neighborhood psychosocial hazard scale |
Baltimore "city neighborhoods" |
+ for perceived psychosocial hazards |
Self-reported |
Obese=BMI≥30 |
| Timperio et al (2005) |
Australia, Melbourne (families with children ages 5-6 and 10-12 years) |
291 families of 5-6 and 919 families of 10-12 year olds |
I |
Perceived (by parents and children) |
Neighborhood access to physical activity facilities, traffic and safety |
Perceived neighborhood |
+ for parental perception of traffic, concern for road safety with children aged 10-12 years |
Measured (for children) |
Obese=BMI≥30 |
| Ellaway et al (2005) |
Europe |
6919; 8 countries |
I |
Perceived (by surveyors) |
Graffiti, litter, dog mess, and greenery |
Immediate residential environment |
- for green space; + for graffiti, garbage |
Self-reported |
Overweight/obese =BMI≥25 |
| Rutt and Coleman (2005) |
USA, El Paso, TX (mainly Hispanic) |
996 |
I |
Perceived |
Physical environment characteristics, barriers to exercise |
2.5 mile radius |
+ for land use mix |
Self-reported |
BMI |
| Lopez-Zetina (2005) |
USA, CA |
33 counties |
E |
Objective |
Aggregate VNT per county |
County (with >100,000 residents) |
+ for county VMT |
Self-reported |
County-level % obese (BMI≥30) |
| Vanderfrift and Yoked (2004) |
USA |
50 states |
E |
Objective |
Urban sprawl |
State level |
+ for amount of developed land |
Self-reported from secondary data |
State-level percent obese (BMI≥30) |
| Frank et al (2004) |
USA, Atlanta, GA |
10,878 |
I |
Objective |
Land use mix |
1-kb distance from residence |
- for mixed land use |
Self-reported |
Obese=BMI≥30 |
| Saelens et al (2003) |
USA, San Francisco, CA |
107 |
I |
Perceived |
Neighborhood environment walkability scale |
Perceived neighborhood |
- for walkability |
Self-reported |
Overweight=BMI≥25 |
Abbreviations: E, ecologic; I, Individual; M, multilevel; MSAs, metropolitan statistical areas; NHs, neighborhoods; VMT, vehicle miles traveled
 Summary of Results:
Key Findings
- From 37 studies, the influence of neighborhood factors on obesity are mixed.
- Neighborhood-level measures of economic resources were associated with obesity in 15 studies
- The associations between neighborhood income inequality and racial composition with obesity were mixed.
- The availability of healthy versus unhealthy food was inconsistently related to obesity.
- Neighborhood features that discourage physical activity were consistently asociated with increased body mass index.
Other Findings
- This review suggests that, at minimum, individual-level approaches such as diet and exercise guidelines need to recognize potential barriers to good health imparted by the neighborhood context.
Author Conclusion:
Characterisitcs of the built environment and neighborhood opportunities for physical activity are consistently associated with reduced body weight status, while the influence of food avialability on obesity is mixed. The efficacy of targeted neighborhood interventions to reduce obesity remains unknown.
Reviewer Comments:
|
Research Design and Implementation Criteria Checklist: Review Articles
|
|
Relevance Questions
|
| |
1.
|
Will the answer if true, have a direct bearing on the health of patients? |
Yes
|
| |
2.
|
Is the outcome or topic something that patients/clients/population groups would care about? |
Yes
|
| |
3.
|
Is the problem addressed in the review one that is relevant to nutrition or dietetics practice? |
Yes
|
| |
4.
|
Will the information, if true, require a change in practice? |
Yes
|
| |
|
Validity Questions
|
| |
1.
|
Was the question for the review clearly focused and appropriate? |
Yes
|
| |
2.
|
Was the search strategy used to locate relevant studies comprehensive? Were the databases searched and the search termsused described? |
Yes
|
| |
3.
|
Were explicit methods used to select studies to include in the review? Were inclusion/exclusion criteria specified and appropriate? Were selection methods unbiased? |
Yes
|
| |
4.
|
Was there an appraisal of the quality and validity of studies included in the review? Were appraisal methods specified, appropriate, and reproducible? |
Yes
|
| |
5.
|
Were specific treatments/interventions/exposures described? Were treatments similar enough to be combined? |
Yes
|
| |
6.
|
Was the outcome of interest clearly indicated? Were other potential harms and benefits considered? |
Yes
|
| |
7.
|
Were processes for data abstraction, synthesis, and analysis described? Were they applied consistently across studies and groups? Was there appropriate use of qualitative and/or quantitative synthesis? Was variation in findings among studies analyzed? Were heterogeneity issued considered? If data from studies were aggregated for meta-analysis, was the procedure described? |
Yes
|
| |
8.
|
Are the results clearly presented in narrative and/or quantitative terms? If summary statistics are used, are levels of significance and/or confidence intervals included? |
Yes
|
| |
9.
|
Are conclusions supported by results with biases and limitations taken into consideration? Are limitations of the review identified and discussed? |
Yes
|
| |
10.
|
Was bias due to the review’s funding or sponsorship unlikely? |
Yes
|
| |
| |
Copyright American Dietetic Association (ADA).
|
