Longitudinal Trends in Substance Use and Intimate Partner Violence-Related Death: Findings from the National Violent Death Reporting System

Background: Intimate partner violence (IPV) is a common phenomenon that is a significant cause of mortality in the United States. Existing research has identified risk factors associated with IPV to advance prevention efforts, including the robust association between IPV and substance use. Less research has examined how IPV trends vary based on specific types of substance use, though several studies have identified a strong association between IPV victimization and alcohol use. To advance knowledge within these areas of research, we examine the co-incidence of substance use and IPV-related deaths from the years 2003-2015.    

Method: Data is from the National Violent Death Reporting System (NVDRS), which includes all violent deaths from 27 states in the U.S., collected each year from 2003 to 2015 (N = 135,310). NVDRS data includes reports from death certificates, as well as coroner, law enforcement and toxicology reports. The IPV-related binary dependent measure in this study included whether the death was related to immediate or ongoing conflict or violence between current or former intimate partners. To establish substance use related violent deaths, 5 binary independent variables were created from toxicology reports to identify persons who had ingested alcohol, opioids, amphetamine, marijuana and cocaine prior to death. Additional measures in the NVDRS include the year of the death, race/ethnicity, age, gender, education and marital status. To examine whether there are substance use differences in IPV related deaths, fixed-effects multivariate logistic regression models examined whether a substance use by year interaction terms were associated with IPV-related death, controlling for additional demographic factors.

Results: Between 2003 and 2015, there was a significant increase in odds of IPV-related death for all individuals (OR = 1.01, p < .01). Women, racial/ethnic minorities, persons younger in age, with lower education levels, who were married, and who engaged in alcohol, amphetamine, marijuana and cocaine use all showed an increased odds of IPV-related death. Opioid users showed decreased odds of IPV-related death. Additionally, model results showed a significant year by opioid use interaction, which showed opioid users experienced significantly lower odds of IPV-related death over time compared to non-opioid users (OR = 1.05, p < .01). A significant year by amphetamine use interaction also emerged, which showed amphetamine users experienced significantly higher odds of IPV-related death over time compared to non-amphetamine users (OR = 1.06, p < .01). There were no significant changes over time among alcohol, marijuana, and cocaine users in odds of IPV-related death.

Conclusion: This study clarifies existing research that has identified a relationship between substance use and IPV. Our results suggest that given the increases in substance use and IPV co-occurrence over time, additional consideration should be given to how other substances such as amphetamines may be associated with IPV and why opioid use may be related to a lower risk for IPV-related death. These findings suggest that social work practitioners may consider the specific type of substances used among clients and associated risks to improve concrete intervention and prevention efforts associated with IPV.