Vacant properties: A modifiable target of intervention to reduce aggravated assault in urban cities?

Background/Purpose: Violent assaults in urban cities challenge policymakers and community leaders who seek interventions to reduce violence. Prior literature has established demographic features of neighborhoods in which violence rates are highest, but has been limited in identifying modifiable targets for intervention. This study aimed to determine if vacant properties within blockgroups were associated with increased aggravated assault over time in an urban city.

Methods: Aggravated assaults (2002-2006) were compiled through police data and linked to blockgroup characteristics from Census Bureau data. Vacant properties within blockgroups were calculated yearly, while other sociodemographic characteristics (e.g., population, income, race) were encoded as time invariant. A mixed effects negative binomial regression model examined longitudinally the association of vacant properties and assaults between and within blockgroups, adjusting for sociodemographic characteristics.

Results/Outcomes: Among the 1816 blockgroups, the assault rate rose from 642 per 100,000 to 711 per 100,000 over the study period. Within blockgroups, the median number of vacant properties was 6 and did not change. Between blockgroups over time, total assaults increased by 18.5% (95%CI: 12.2%-25.1%) for every increase in vacancy category, controlling for sociodemographic characteristics. The impact was stronger for gun assaults, which increased by 22.4% (95%CI: 13.6%-32.0%), than for non-gun assaults, which increased by 14.8% (95%CI: 8.4%-21.5%), for each increase in vacancy category.

Conclusions: Vacant properties were associated with increased assaults independent of other indicators of disadvantage within blockgroups. The stepwise association of vacant properties with assault may suggest a modifiable factor for intervention for the reduction of violence in urban cities.