The primary objective of this dissertation was to examine the association between neighborhood walkability and self-reported body mass index (BMI) among New York City residents. Neighborhood walkability was defined as mixed land use, population density, pedestrian injuries density, subway stops density, bus stops density, tree density, and street connectivity. In addition walkability index was created based on all available measures of walkability in this study. Although the literature review shows that the relationships between walkability and self-reported BMI are qualitatively consistent across studies, the use of self-reported BMI is not fully justified given the well-known error in BMI estimates that are based on the self-report of height and weight. Preliminary analysis suggests that the impact of such errors is not negligible and in fact may produce modeling results that could be misleading. Consequently, the secondary objective of this dissertation was to investigate the nature of these errors and develop prediction equations which were then used to correct for the reporting error.
In order to address the primary objective, data from the Community Health Survey 2002 were linked with zip code level data. Cross-sectional analysis using multilevel modeling was conducted. Population density, subway stops density, bus stops density, street connectivity, land use mix and walkability index were all inversely associated with BMI. These associations were statistically significant for all walkability indicators considered, with the exception of land use mix and subway stops density. In contrast to my hypothesis, however, pedestrian injuries density was not positively associated with BMI. Instead, an inverse association that was statistically significant was observed.
Prediction equations were developed using data from New York City Health and Nutrition Examination Survey. Consequently, Beta coefficients from these equations were used to adjust for reporting error by calculating “measured” or corrected BMI in the CHS 2002 sample. When corrected BMI was used as an outcome in women, Beta coefficients for each walkability exposure changed by more than 10% as compared to Beta coefficients with self-reported BMI as an outcome, which indicated that results for self-reported BMI were biased towards the null. In men, only two exposures changed by at least 10% with corrected BMI as an outcome. Although the results based on the self-reported BMI and corrected BMI were quantitatively different, mathematical difference did not suggest conceptual differences. Since it is uncommon for epidemiologic studies to obtain clinically measured height and weight, these findings provide important evidence that the association between neighborhood walkability and BMI will be captured even if BMI is calculated from self-reported values of height and weight.
|School Location:||United States -- New York|
|Source:||DAI-B 71/03, Dissertation Abstracts International|
|Keywords:||Body mass index, Built environment, Obesity, Walkability|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be