Examining Relationships between the Social Vulnerability Index and North Carolina Residents' Calls to a National Domestic Violence Hotline

Background/Purpose:

The Centers for Disease Control and Prevention (CDC) and the Agency of Toxic Substances and Diseases Registry (ATSDR) define social vulnerability as the degree to which a community can prevent human suffering and financial loss due to a disaster or disease outbreak; the Social Vulnerability Index (SVI) was created to measure this construct. Could this index have predicted increased human suffering associated with domestic violence (DV) during COVID-19? Studies have demonstrated the SVI is associated with other forms of violence including violent crime in the US as well as DV and animal abuse in Brazil; however, DV in the US has not been explored. The purpose of this study is to determine whether the SVI can predict DV hotline calls two years after stay-at-home orders ended. This remains a prominent issue because the SVI can be used to assess post-disaster needs and prepare for future outbreaks.

Methods:

Analyses were conducted on two publicly available datasets. County-level data for DV/sexual assault agencies were sourced from the North Carolina (NC) Council for Women and Youth Involvement (NC CFWYI). Ninety-eight of 100 counties reported data in 2020. These data were merged with county-level SVI data for 2020.

The dependent variable is the number of calls to the National Domestic Violence Hotline in each county divided by the county’s population in 2020. The four independent variables are the four “summary themes” that comprise the SVI: (1) Socioeconomic status (SES), (2) household characteristics, (3) racial and ethnic minority status, and (4) housing type and transportation. The CDC sourced the data from the American Community Survey five-year estimates (2016-2020).

Stata was used to conduct multiple regression to analyze the relationship between one continuous dependent variable and four continuous predictors. Robust standard errors were calculated to address a violation of homoskedasticity.

Results:

We found a significant relationship between the SES theme and calls to a DV hotline (p < 0.05). Specifically, we found a 0.017% increase in calls to a DV hotline (+/- 0.001) for every unit increase in SES vulnerability. The remaining three predictor variables were not significant.

Conclusions and Implications:

In this study, only SES was significantly related to calls to a DV hotline. This index may be used to assess potential risk and plan for equitable allocation of services and funds for underserved communities. This suggests that policies targeting financial well-being and access to education can contribute to safer and healthier post-disaster outcomes for entire counties, particularly as it relates to communities affected by structural racism.

Importantly, “racial and ethnic minority status” is not a significant predictor of DV hotline calls. This finding suggests either that minoritized groups did not need the hotline, or the hotline was not accessible to them. A significant limitation of this measure is that the concentration of minoritized groups is not an ethical measure of “social vulnerability,” and may leave more questions than answers. Additional research is necessary to understand how to make services available to underserved and minoritized groups during and after disasters and disease outbreaks.