Neighborhood – Saunders et al. (2017) [Roanoke, VA]

Study Reference:

Saunders, J., Robbins, M., & Ober, A. J. (2017). Moving from efficacy to effectiveness: Implementing the drug market intervention across multiple sites. Criminology & Public Policy16(3), 787-814.


Location in the Matrix; Methodological Rigor; Outcome:

Neighborhoods; Focused; Highly Proactive; Rigorous; Effective


What police practice or strategy was examined?

This study examined the impact of a Drug Market Initiative (DMI) in Roanoke, Virginia beginning in 2010. The DMI is a focused deterrence program that sought to identify and disrupt overt drug markets. The DMI was intended to operate in five stages: 1) The planning stage, where police officers would identify drug markets and sellers in those markets 2) The targeting phase, where police would make undercover buys and build cases for prosecution against dealers who were deemed to be violent and dangerous (“A-listers”). 3) The community buy-in phase, where officers encouraged community members to take back their neighborhoods by showing disdain for drug dealing and notifying the police of suspicious behavior. 4) The “call-in” phase, where less dangerous offenders were invited to a public community meeting and threatened with prosecution if they continued engaging in drug offenses. 5) Finally, officers conducted post-call-in enforcement and community follow-up. The DMI in Roanoke targeted two neighborhoods, the Hurt Park and Melrose-Rugby neighborhoods, which comprised of 66 and 144 census blocks respectively, and were characterized by crack cocaine and marijuana distribution.


How was the intervention evaluated?

Researchers used a synthetic control design to evaluate the impact of the intervention. This involved the identification and weighting of areas that did not receive the intervention to mimic as closely as possible the crime trends that would have occurred had the intervention not taken place. The authors measured the effect of the intervention relative to the synthetic controls on all crime incidents, as well as property, violent, and drug crime incidents. In the Hurt Park neighborhood, each outcome was evaluated at three, six, nine, and twelve months after the intervention, while in the Melrose-Rugby neighborhood the outcomes were evaluated at three and six months after the intervention. The authors also measured the fidelity of the intervention across each of the planned implementation stages.


What were the key findings?

In the Hurt Park neighborhood, there were significant reductions of 23% in total crime incidents and 50% in Part I property crime incidents. However, there were no significant effects on violent or drug crime. In the Melrose-Rugby neighborhood, there were significant reductions in violent crime only. The intervention was also rated as having high treatment fidelity across all phases.


What were the implications for law enforcement?

DMI efforts appear to be effective when implemented with high fidelity. In this regard, the authors note that – compared to other DMI efforts that have failed to reduce crime – the Roanoke interventions most closely resemble the original High Point (NC) DMI blueprint. It is not clear whether the Roanoke intervention succeeded because of its high program fidelity or because it more closely matched the type of drug market and problems that the DMI was originally designed for. As such, the authors note that the implementation of DMI is challenging, and the effectiveness of DMI efforts may be tied to the specifics of the implementation and the type of markets that are targeted.


Where can I find more information about this intervention, similar types of intervention, or related studies?