Child Maltreatment Investigation Risks By Child Poverty Risks Among Whites, Blacks, and Hispanics in the United States

Background/Purposes:

Despite the strong established association between poverty and child maltreatment, there is no national estimate for official maltreatment investigation rates based upon poverty. This study estimates maltreatment investigation risks among US children based upon environmental poverty by total and subtypes (e.g., neglect) with estimates further broken down by race/ethnicity.

Methods:

National Child Abuse and Neglect Data System (NCANDS) data (2009-2013) were linked to Census data at the county level. During linkage, 30% of US children in 80% of counties were excluded because of suppressed county identifiers where having < 1,000 records, a confidentiality safeguard in NCANDS. Three race/ethnicity-specific data sets were constructed through this linkage. To ensure stable counts of maltreatment, counties with < 10,000 children of a given race/ethnicity were excluded from each data set. Finally, the White data set had 23,847,097 White children (61% of US White children) in 515 counties, the Black data set had 7,096,048 Black children (67% of US Black children) in 185 counties, and the Hispanic data set had 13,716,513 Hispanic children (79% of US Hispanic children) in 207 counties. A Generalized Additive Mixed Model was applied to estimate maltreatment investigation risks based upon child poverty risks by race/ethnicity while considering any curvilinearity and the multilevel structure of the data (i.e., nesting of counties within states).

Results:

With increasing within-race county poverty rates (e.g., % poor among White children), within-race county maltreatment investigation rates (e.g., % investigated among White children) increased in an inverse-U shape for Whites but in a linear fashion for Blacks and Hispanics. As White poverty increased from 2.2% to 25.7%, White investigation rates increased from 0.3% to 6.6% for total, 0.0%-4.5% for neglect, 0.2%-1.6% for physical abuse, and 0.0%-0.5% for sexual abuse. When Black poverty increased from 9.1% to 51.8%, Black investigation rates increased from 3.5% to 9.2% for total, 1.5%-6.1% for neglect, 1.1%-2.2% for physical abuse, and 0.2%-0.6% for sexual abuse. With increasing Hispanic poverty from 10.1% to 46.9%, Hispanic investigation rates increased from 2.0% to 4.8% for total, 0.9%-3.3% for neglect, 0.6%-1.1% for physical abuse, and 0.2%-0.4% for sexual abuse. At similar poverty levels, Whites had slightly higher investigation rates in both total and subtypes than Blacks, but the differences were mostly non-significant. Hispanics showed significantly lower rates than other groups in most cases. For example, at the 20% poverty level, the total investigation rates were 5.8% (95% CI: 5.2%-6.3%) for Whites, 5.0% (3.9%-6.0%) for Blacks, and 2.7% (2.2%-3.3%) for Hispanics.

Conclusions/Implications:

Using national data, this study confirmed the strong positive associations of total and type-specific maltreatment investigation rates with environmental measures of child poverty in all race/ethnicity groups. After controlling for poverty, Blacks showed no higher maltreatment risk than Whites in either total reports or specific subtypes, suggesting that the nationwide racial disproportionality in child maltreatment may be mostly due to the disproportional poverty risks between Blacks and Whites. The “Hispanic Paradox” or “Healthy Immigrant Effect” was robustly supported, as Hispanics mostly had lower maltreatment risks while controlling for poverty.