Walking Behaviors Reported in the American Time Use Survey 2003-2005

Background: We report walking for shopping, exercise, transportation, and walking the dog, among other sources captured in the 2003 to 2005 American Time Use Survey (ATUS). Methods: We extracted and analyzed 8 walking behaviors (by sex, age, education level, and race/ethnicity) from 24 hours of activities recalled by telephone interview for 15,175 males and 19,518 females age >= 15 years. Results: On any given day in 2003 to 2005, 45.8% of Americans participated in a median of 45 minutes of any walking activities; 31.6% walked for shopping purposes, 12.5% walked for transportation, 4.8% walked for exercise, and 2.5% walked the dog. College-educated respondents more commonly reported walking while shopping, walking for exercise, and dog walking. Those with less than a high school education more commonly reported walking for transportation. Conclusions: Despite limitations identified in imputing explicit and implicit performance of walking behaviors in the ATUS, Americans engage in a wide variety of walking behaviors that are not well represented by surveys focused only on leisure-time behaviors. Public health implications include increased availability of multiple and varied opportunities for walking, especially through environmental shifts toward more walkable places and destinations and policy shifts that support walking behaviors over competing transportation modes.