Economic and Racial Disparities in Child Maltreatment Subtypes

Background/Purposes:

Emerging evidence suggests that the large racial disparities often found in child maltreatment data disappear or even reverse after controlling for poverty. Conversely, poverty substantially increases maltreatment risk even when controlling for various covariates. What we currently lack is a good understanding of how these factors operate at the level of subtypes (e.g. neglect, physical abuse…) This study aims to examine economic and racial disparities among subtypes, including neglect, physical, sexual, and emotional abuse, as well as for overall maltreatment.

Methods:

We obtained data from two high-risk samples. The “CA/N” sample included all St. Louis children ≤ age 3 with a first-time maltreatment report in 1993-1994 (N=2,111). The “AFDC” sample included randomly selected St. Louis children ≤ age 3 receiving AFDC in 1993-1994 with no current/prior maltreatment report (N=1,923). We longitudinally traced them within various Missouri administrative records, including child protective services, welfare, birth, medical, criminal, education, and census records. Multilevel logistic growth curve models estimated the likelihood of total and type-specific maltreatment reporting at each age from 1 to 17 as a function of AFDC/TANF receipt, race, and various other predictors.

Results:

In the baseline (1993-1994) maltreatment reports, non-Whites (98% Black) were overrepresented by about a factor of 2 for total reports and within subtypes compared to their representation in the population. In the baseline total and type-specific maltreatment reports, the proportion of families receiving AFDC was about 6 times higher than in the general population at that time. During follow-up (1995-2009) we tracked subsequent child maltreatment reports. The large racial disparity in reports not only disappeared, but often reversed after considering AFDC/TANF receipt and other predictors. Multivariate models estimated that compared with Whites, non-Whites had a 18%-34% lower likelihood of any maltreatment (CA/N-sample: OR=0.82, 95%CI=0.73-0.93; AFDC-sample: OR=0.66, 95%CI=0.54-0.80), and a 27%-41% lower likelihood of neglect (CA/N-sample: OR=0.73, 95%CI=0.62-0.86; AFDC-sample: OR=0.59, 95%CI=0.47-0.74). For other subtypes, power became an issue and statistical significance was not achieved, despite the rates for Whites being generally lower. Conversely, AFDC/TANF receipt remained a strong predictor after controlling for various other predictors. AFDC/TANF receipt increased the report likelihood by 1.69-1.94 times for any maltreatment (CA/N-sample: OR=1.94, 95%CI=1.70-2.21; AFDC-sample: OR=1.69, 95%CI=1.42-2.02), 1.76-1.98 times for neglect (CA/N-sample: OR=1.98, 95%CI=1.69-2.33; OR=1.76, 95%CI=1.41-2.19), and 1.38-1.96 times for physical abuse (CA/N-sample: OR=1.38, 95%CI=1.12-1.69; AFDC-sample: OR=1.96, 95%CI=1.43-2.67). AFDC/TANF receipt increased the likelihood of sexual abuse reporting by 2.19 times only for the CA/N sample (OR=2.19, 95%CI=3.49).

Conclusions/Implications:

Results suggest that the large racial disparity in total and type-specific maltreatment reports may be mostly due to differences in poverty between racial groups. While controlling for economic indicators, Blacks had a slightly lower risk than Whites. This tendency was stronger for neglect, but also observed in other subtypes. The economic disparity in maltreatment reports remained large even among high-risk children while controlling for various risk/protective factors. The economic disparity was larger for neglect, but generally remained substantial for other subtypes as well. This study highlights necessity of addressing poverty and related issues in policy and practice.