The relationship between drug overdose and child maltreatment in rural, sub-urban and urban counties across the United States: Implications for Social Work

Background and Purpose.

Substance use disorders and poverty predict child maltreatment, but these relationships have not been extensively studied using population-level data. Given the epidemic of opioid overdose in the US, it is imperative to understand how a community’s rates of serious substance use disorder contributes to child maltreatment. The purpose of this study was to examine changes in county child maltreatment rates, predicted by drug overdose death rates and moderated by status on the rural-urban continuum.

Methods.

We conducted a national study of all US counties (n=2,963). We used zero-inflated negative binomial modeling to estimate the impact of drug overdose death rates on child maltreatment report rates.  We examined whether these risks are moderated by rurality, using a three category measure created from the USDA’s rural-urban continuum codes (metropolitan, non-metropolitan, and rural). Our measures were compiled and linked from five sources, including the US Census, CDC, USDA, RWJF County Health Rankings and the National Child Abuse and Neglect Data System.

Results.

As expected, county drug overdose death rates positively predicted maltreatment rates [χ²(1)=16.03 , p<.0001].   Compared to urban counties, the increase in maltreatment reports with increasing overdose deaths is stronger in non-metro rural counties (χ²(1)=3.31, p=.069) and strongest in non-metro suburban counties (χ²(1)=8.60, p=.003). Median family income [χ²(1)=331.18 , p<.0001] negatively predicted maltreatment rates, while % children in single-parent households [χ²(1)=28.61, p<.0001], violent crime [χ²(1)=32.47 , p<.0001] and % non-white population [χ²(1)=137.04 , p<0001] positively predicted maltreatment.

Conclusions and Implications.

Consistent with family-level research, counties challenged by substance use and poverty also experience greater maltreatment report rates. This relationship varies, with non-metro counties demonstrating a stronger relationship. This study’s limitations included issues common to secondary data analysis, including varying data vintages from 2013-15 (common in large studies linking multiple datasets), unavailability of data on some likely contributing factors, and measurement at the county level rather than a smaller area of analysis, due to limits on tract or smaller area levels of measurement.

Future research should explore how access to health care and social support services may mediate the relationship. Access to health care has not yet been measured holistically in social work and public health research, and thus it is difficult to study disparities; this is a compelling area for future research efforts.  There remains a need for social work and public health researchers and professionals to understand and interrupt the relationship between substance use disorders and child maltreatment.  Studies at smaller areas of analysis, as well as qualitative methods, may be warranted to elucidate risk and protective factors, as well as potential intervention strategies.