BackgroundPhysical activity (PA) has numerous health benefits, but older adults live mostly sedentary lifestyles. The physical and social neighborhood environment may encourage/dissuade PA. In particular, neighborhood crime may lead to feeling unsafe and affect older adults’ willingness to be physically active. Yet, research on this topic is still inconclusive. Older population, probably the age group most influenced by the neighborhood environment, has been understudied, especially in Southern Europe. In this study, we aimed to analyze the association between leisure-time physical activity (LTPA) in older adults and objective crime, alongside other neighborhood characteristics.MethodsWe obtained data from a population-based cohort from Porto (2005–2008) to assess LTPA. Only adults aged 65 years or more were included (n = 532). A Geographic Information System was used to measure neighborhood characteristics. Neighborhood crime was expressed as crime rates by category (incivilities, criminal offenses with and without violence and traffic crime). Neighborhood characteristics such as socioeconomic deprivation, land gradient, street density, transportation network, distance to parks, non-residential destinations and sport spaces were also included. Generalized Additive Models were fitted to estimate the association between neighborhood characteristics and the participation (being active vs. inactive) and frequency (min/day) of LTPA.ResultsForty-six percent of the men and 61 % of the women did not engage in any kind of LTPA. Among the active participants, men spent on average 50.5 (35.2 Standard Deviation, SD) min/day in LTPA, whereas the average among women was 36.9 (35.1 SD) min/day (p < 0.001).Neighborhood crime was unrelated to the participation in, or frequency of, LTPA. On the other hand, two neighborhood characteristics – distance to the nearest park (β = −0.0262, p = 0.029) and to the nearest non-residential destination (β = −0.0735, p = 0.019) – were associated with time spent on LTPA, but only among active older women. No neighborhood characteristic was related to participation in LTPA.ConclusionsFrom a public health point of view, the provision of parks and non-residential destinations (shops, schools, cultural and worship places) might contribute to elevate PA levels of already active older women. On the other hand, in this setting, crime was not a big issue.Electronic supplementary materialThe online version of this article (doi:10.1186/s12889-015-1879-y) contains supplementary material, which is available to authorized users.