Faculty Bibliography

The New York State Medicaid program is pursuing reforms that would shift reimbursement from fee-for-service to value-based models. Behavioral health populations and services are key to these reforms, but important gaps exist in the breadth and relevance of available quality measures for the behavioral health field. This column describes how the state addressed these gaps by using both inter- and intra-agency collaborations. As part of this effort, the state convened a panel of consultants, including academics, providers, and consumers, to develop a behavioral health quality measurement agenda. Panel recommendations and ongoing partnerships are described as a model for states considering similar value-based reform initiatives.

Racial and ethnic segregation has been linked to a number of deleterious health outcomes, including violence. Previous studies of segregation and violence have focused on segregation between African Americans and Whites, used homicide as a measure of violence, and employed segregation measures that fail to take into account neighborhood level processes. We examined the relationship between neighborhood diversity and violent injury in Oakland, California. Violent injuries from the Alameda County Medical Center Trauma Registry that occurred between 1998 and 2002 were geocoded. A local measure of diversity among African American, White, Hispanic, and Asian populations that captured interactions across census block group boundaries was calculated from 2000 U.S. Census data and a Geographic Information System. The relationship between violent injuries and neighborhood level of diversity, adjusted for covariates, was analyzed with zero-inflated negative binomial regression. There was a significant and inverse association between level of racial and ethnic diversity and rate of violent injury (IRR 0.30; 95% CI: 0.13-0.69). There was a similar relationship between diversity and violent injury for predominantly African American block groups (IRR 0.23; 95% CI: 0.08-0.62) and predominantly Hispanic block groups (IRR 0.08; 95% CI: 0.01-0.76). Diversity was not significantly associated with violent injury in predominantly White or Asian block groups. Block group racial and ethnic diversity is associated with lower rates of violent injury, particularly for predominantly African American and Hispanic block groups.