Stepping towards causation: do built environments or neighborhood and travel preferences explain physical activity, driving, and obesity?

Evidence documents associations between neighborhood design and active and sedentary forms of travel. Most studies compare travel patterns for people located in different types of neighborhoods at one point in time adjusting for demographics. Most fail to account for either underlying neighborhood selection factors (reasons for choosing a neighborhood) or preferences (neighborhoods that are preferred) that impact neighborhood selection and behavior. Known as self-selection, this issue makes it difficult to evaluate causation among built form, behavior, and associated outcomes and to know how much more walking and less driving could occur through creating environments conducive to active transport. The current study controls for neighborhood selection and preference and isolates the effect of the built environment on walking, car use, and obesity. Separate analyses were conducted among 2056 persons in the Atlanta, USA based Strategies for Metropolitan Atlanta's Regional Transportation and Air Quality (SMARTRAQ) travel survey on selection factors and 1466 persons in the SMARTRAQ community preference sub-survey. A significant proportion of the population are "mismatched" and do not live in their preferred neighborhood type. Factors influencing neighborhood selection and individual preferences, and current neighborhood walkability explained vehicle travel distance after controlling for demographic variables. Individuals who preferred and lived in a walkable neighborhood walked most (33.9% walked) and drove 25.8 miles per day on average. Individuals that preferred and lived in car dependent neighborhoods drove the most (43 miles per day) and walked the least (3.3%). Individuals that do not prefer a walkable environment walked little and show no change in obesity prevalence regardless of where they live. About half as many participants were obese (11.7%) who prefer and live in walkable environments than participants who prefer car dependent environments (21.6%). Findings suggest that creating walkable environments may result in higher levels of physical activity and less driving and in slightly lower obesity prevalence for those preferring walkability.