Role of neighborhood characteristics in explaining child and adolescent obesity

Background: Childhood and adolescent obesity is recognized as a major public health problem in the US.  It has been postulated that neighborhood factors may influence obesity rates.

Methods:   Data are from a hospital-based pediatric primary care clinic in Rhode Island. Home addresses of 8,127 patients, 2-18 years old with valid height and weight data from 2010-2013 were included. Body Mass Index (BMI) was calculated, and patient data was geocoded, and analyzed at the block group (BG) level.  We determined the percent of obese patients in each BG, and compared neighborhood characteristics across BGs. BGs with fewer than 20 patients were excluded (n=683), leaving 129 BGs and 6,603 patients in our sample. 

Results: Rates of obesity varied substantially across BGs, ranging from 0%-44.1% (mean=22.0%). Approximately 17% of BGs had obesity rates of 15% or less, while 16.3% had rates of 30% or greater. Variation was not explained by neighborhood demographics.  Median income of BGs ranged from $11,406 to $79,545 (mean=$33,209; 2 BGs with median income greater than $60,000).  No significant associations were found between obesity rates and median income, poverty rate, housing, or race/ethnicity characteristics of BGs.

Conclusions:  There was substantial variation in obesity rates across BGs in our sample, but no evidence to support an association between obesity rates and neighborhood demographics.  Further research is needed to identify neighborhood characteristics that can explain the variation in obesity at the block group level.