Initiating and maintaining recreational walking: a longitudinal study on the influence of neighborhood green space

Objective: This study examined prospective relationships of green space attributes with adults initiating or maintaining recreational walking. Methods: Postal surveys were completed by 1036 adults living in Adelaide, Australia, at baseline (two time points in 2003-04) and follow-up (2007-08). Initiating or maintaining recreational walking was determined using self-reported walking frequency. Green space attributes examined were perceived presence, quality, proximity, and the objectively measured area (total and largest) and number of green spaces within a 1.6. km buffer drawn from the center of each study neighborhood. Multilevel regression analyses examined the odds of initiating or maintaining walking separately for each green space attribute. Results: At baseline, participants were categorized into non-regular (n. = 395), regular (n. = 286), and irregular walkers (n. = 313). Among non-regular walkers, 30% had initiated walking, while 70% of regular walkers had maintained walking at follow-up. No green space attributes were associated with initiating walking. However, positive perceptions of the presence of and proximity to green spaces and the total and largest areas of green space were significantly associated with a higher likelihood of walking maintenance over four years. Conclusion: Neighborhood green spaces may not assist adults to initiate walking, but their presence and proximity may facilitate them to maintain recreational walking over time.