Sustainability of violence prevention after a randomized vacant lot greening intervention in Philadelphia: A 10-y follow-up

Vacant lot greening has been shown to reduce violence and crime in the short-term [C. C. Branas et al. , Proc. Natl. Acad. Sci. U.S.A. 115 , 2946–2951 (2018)]. We assessed the 10-y impact of a 2013 clustered randomized controlled trial (RCT) of lot greening in Philadelphia on violence and crime. We measured lot greening by treatment assignment from the original 2013 cluster RCT and a longitudinal greening score to assess the sustainability of the treatment assignment from 2012 through 2023. Our
Vacant lot greening has been shown to reduce violence and crime in the short-term [C. C. Branas et al. , Proc. Natl. Acad. Sci. U.S.A. 115 , 2946–2951 (2018)]. We assessed the 10-y impact of a 2013 clustered randomized controlled trial (RCT) of lot greening in Philadelphia on violence and crime. We measured lot greening by treatment assignment from the original 2013 cluster RCT and a longitudinal greening score to assess the sustainability of the treatment assignment from 2012 through 2023. Our outcomes were the annual reported total crime, violent crime, and firearm violent crime per square mile at each lot. We conducted intent-to-treat and contamination-adjusted intent-to-treat analyses to estimate the local average treatment effect on relative change in crime. Lots randomized to the full greening intervention remained greener over time compared to other lots. In intent-to-treat analyses, compared to lots that received no intervention, lots that received any intervention had a 9% decrease in density of any crime (estimate = 0.91, 95%CI = 0.86, 0.96) and a 9% decrease in violent crime (estimate = 0.91, 95%CI = 0.86, 0.97) per square mile per year. In contamination-adjusted analyses, compared to lots without sustained greening treatment, lots with sustained greening treatment had a 12% decrease in density of any crime (estimate = 0.88, 95%CI = 0.83, 0.93) and an 11% decrease in density of violent crime (estimate = 0.89, 95%CI = 0.83, 0.95) per square mile per year. Even after a decade, vacant lot greening continues to prevent violence.




