Does the environment moderate the impact of a mass media campaign to promote walking?

Purpose. To examine if a mass media campaign influenced walking differently in people in different physical environments. Design. Quasi-experimental study. Setting. Wheeling, West Virginia. Participants. Random sample of adults age 50 to 65 years, response rate: 72.1% (n 5 719 in intervention community, n 5 753 in comparison community). Intervention. Mass media campaign. Measures. Self-reported measures were used in before and after telephone surveys for walking and the physical environment. Measures included 11 environmental walkability items, from which two subscales (i.e., usable sidewalks/aesthetics and facilities) were extracted. Analysis. Multiple linear regression. Results. Overall, walking increased by 2.7 minutes per week (standard deviation [SD] 5 231.1, not significant [NS]). When confined to those insufficiently active at baseline (i.e., ,30 minutes per day) the minutes walked increased by 92.1 minutes (SD 5 152.9, p, .001). For the insufficiently active at baseline in the top half of the environmental factor of usable sidewalks, walking increased by 19 minutes more than in the bottom half (NS). For the factor of aesthetics and facilities, people in the more walkable environment increased walking by 87 minutes more than those in the bottom half (p , .001). Conclusion. In this community-wide physical activity, intervention changes in walking after the campaign were significantly moderated by some environmental attributes. This contributes to the limited evidence on the impact of the environment in enhancing community physical activity interventions. This finding needs to be replicated in other community interventions with greater environmental variation.