Grant Number:	5R03CA103484-02 Interpret this number
Primary Investigator:	Scribner, Richard
Organization:	Lsu Health Sciences Center
Project Title:	Multilevel Exploration of Obesity Risk of Cancer Preven*
Fiscal Year:	2004

DESCRIPTION (provided by applicant): The overall goal of the project is to develop an approach for identifying modifiable factors in the residential environment responsible for the epidemic of overweight and obese individuals in the United States. The successful identification of such factors would facilitate interventions to primarily prevent obesity related cancer. We believe that the Louisiana Department of Motor Vehicles (DMV) Driver's License File, which contains both geographic identifiers -suitable for nesting individuals by geographic scale - and anthropometric measures - suitable for determining BMI, is ideally suited to this approach. If the effect for an environmental risk factor exists multilevel analyses should be able to observe significant differences in BMI across units at the geographic scale where the effect exists. Measures of various environmental risk factors can then be entered in the model to identify the specific factor accounting for this hierarchical structure in the data. In a final analysis to validate the implicit model we will assess the ecologic relation between obesity and obesity related cancer at the census tract level. The specific alms are: 1.To determine whether individual differences in BMI are grouped at any of a number of levels of geographic scale (i.e., block group, census tract, zip code, city, parish), 2.To identify environmental factors linked to either the social environment (e.g., median income, percent unemployed), the physical environment (e.g., street networks, access to recreational facilities), or the availability of certain foods (e.g., fresh fruits and vegetables, fast food) that explain variance in BMI partitioned to a geographic level of analysis, and 3. To determine if individual level factors that may moderate or confound the observed effect of any environmental risk factor in explaining between area differences in BMI. 4. To assess the ecologic relation between high prevalence of overweight and obesity and the high incidence of overweight and obesity related cancers at the census tract level.





None