sources should be devoted to relevant obesity prevention efforts. Funding and implementation priorities are often based on available data; as a result, areas with limited or no data are often overlooked because it is difficult to justify a need or to make a case for investment in intervention efforts without baseline data. Furthermore, the collection of data requires a level of accountability and follow-up, as there is the expectation that data collection will lead to changes and improvements in a community for the public good. Nevertheless, data are often lacking at the community level. Surveys can be expensive to conduct, research and validation of community assessment tools are relatively new, and the intersection of public health and the built environment is only beginning to be explored (Northridge et al., 2003).
Only a limited number of national surveys of childhood obesity prevention efforts have provided data that are aggregated at the regional or city level (Chapter 4 and Appendix D). For example, the CDC’s 2004 School Health Profiles survey had weighted data from 11 school districts that enabled analysis of comparison data at the district level (CDC, 2006a) (Appendix C). The Behavioral Risk Factor Surveillance System (BRFSS) conducted by CDC has recently expanded its capabilities to provide local data for several U.S. cities and communities (CDC, 2006b) (Appendix C). The Selected Metropolitan/Micropolitan Area Risk Trends (SMART) BRFSS project provides data for counties, cities, and geographic areas with 500 or more respondents. The SMART BRFSS is a potential model for other systems such as the Youth Risk Behavior Surveillance (YRBS) system, to provide more local level data. Currently, the YRBS system provides data at the national and state levels and has a few specialized data sets, such as data for Bureau of Indian Affairs schools.
Efforts to provide greater specificity at the local level involve increased sample sizes and are therefore more costly to administer and analyze. However, given the need for local-level data for local-level decision making, research efforts focused on accurate methodologies for the extrapolation of state or regional data into meaningful community-level data should be explored.
State, regional, and city surveys are also conducted. The funding sources for these surveys, frequency with which they are conducted, their consistency, and the extent of data that they obtain on topics relevant to childhood obesity prevention are highly individualized. The California Health Interview Survey is one of the more extensive health surveys and provides data for the state and county levels, including representative information for specific racial/ethnic sub-populations. The Indian Health Service can provide tribal and community leaders with local-level data through its electronic health information system (IHS, 2004) (Chapter 3), which in-