Abstract P123: the Physical Activity Built Environment: Evaluation of the Concordance of Perceived Versus Measured Access to Parks and Bike Paths

Background: Approximately 80% of pregnant women in the US do not meet the CDC recommendation for moderate-intensity physical activity. The built environment is a key contributor to physical activity (PA) behavior. However, there may be a disconnect between physical attributes of the built environment and perceived access to an environment that supports PA. Use of a geographic information system (GIS) to objectively measure aspects of the built environment can yield important information about proximity of resources for supporting PA. We evaluated concordance of perceived access to multi-use bike paths (MUBP) and parks in a population of pregnant women. We also examined demographic characteristics associated with concordance between perceived and actual supports for PA. Methods: A population of 155 pregnant women in the greater Columbus, OH area reported perceptions of access to environmental supports within their neighborhood (defined as 0.5 mi. radius/ 10 min. walk) and community (10 mi. radius/ 20 min. drive), along with self-reported PA information. Addresses were abstracted from medical records and geocoded. Within ArcMap 10.2.2 (ESRI, Redlands, CA), buffers at distances of 0.5, 1, and 10 miles were created around MUBP and parks, respectively. Respondents were classified as “near” a resource if their address fell within those buffers. Logistic regression was used to examine the association between concordance of self-report with objectively measured proximity and demographic factors associated with higher concordance. Results: Within the study population of 155, 12% and 28% of women lived within 0.5 and 1 mile of a dedicated MUBP, respectively. More women lived within close proximity to a park (46% within a 0.5 mile radius, and 69% were within a 1 mile radius). The discordancy for parks ranged from 9.68%-49.03%, and from 16.77%-76.13% for MUBP. Concordance of report of parks within the community was higher among non-Hispanic white women (OR=4.2 [1.1, 15.3]) and that concordance or over-reporting of access to parks within the community was higher among married women (OR=5.3 [1.3, 21.5]). There were no significant differences in self-reported PA or changes in PA, however women living within 1 mile of a park reported more minutes of moderate-to-vigorous PA. A greater percentage of those with access reported increasing or maintaining levels of moderate PA from before pregnancy through the third trimester. Conclusions: The discordance between perception of access and objectively measured proximity to MUBP and parks suggests a substantial portion of women are not aware of their availability, or do not perceive them as relevant factors in influencing their PA behaviors during pregnancy. It may be that physical proximity is not a major driver in perceived access. Characteristics of the built environment, such as safety or social environment, may play a larger role in influencing PA behaviors.