Latent Class Analysis of African American Substance Abusing Caregivers in Child Welfare: Implications for Targeted Child Welfare Policy and Practices

Purpose: Although caregiver substance abuse is a serious problem associated with child maltreatment (Larrieu, et al., 2008), little information exists to establish patterns of correlates of caregivers with substance abuse problems associated with child abuse. Research shows that African American children are identified by child protective services as victimized by serious maltreatment (Bartholet, 2009) stemming from maltreatment by caregivers. Unfortunately, less is known about the combinations of factors among caregivers, especially for those involved with child welfare. Using a person centered analytic approach, this study sought to examine patterns of risks among African American substance abusing caregivers. Methods: This investigation analyzed data from the National Survey of Child and Adolescent Well-being (NSCAW). A subsample of 258 African American women with child welfare worker identified substance abuse and whose children were younger than five years of age was analyzed. NSCAW is a national probability, stratified sample of families investigated for child maltreatment, which includes families regardless of the outcome of the investigation allowing for a range of families with and without substance abuse and substantiated child maltreatment. This investigation examined well-established correlates of family maltreatment, including prior reports, mental health issues, recent arrest or detention, low social support, intimate partner violence (IPV), education level, partner living in the home, whether living in urban or rural area, TANF receipt, marital and work status; to determine group membership based on patterns of characteristics identifying individuals with unique similarities. Mplus was used to estimate the model. Starting with all cases in one-class, models were run iteratively and systematically compared to determine the number of classes which best fit the data. The Bayesian Information Criteria (BIC), the likelihood-ratio ÷2 and posterior probability of group membership were used to assess goodness of fit.

Results: A three-class solution was determined to be the best fitting model indicating the presence of three distinct profiles of African American caregivers within the child welfare system. The largest class (59%) was characterized by young, employed caregivers with only low to moderate rates of additional stressors. The second class was comprised of single caregivers contending with multiple, chronic risk factors (13%). Caregivers in this class had prior maltreatment reports, mental health problems, a recent arrest or detention, low social support, recent IPV, were unemployed and were receiving TANF. Membership in the third class included married, older caregivers with low social support (28%). The classes were not differentiated on the communities in which they lived (this was a mostly urban sample) or whether or not they themselves had experienced child abuse or neglect.

Implication: A unique contribution of the study is the differentiation of profiles based on the heterogeneity of the group. Our study suggests three distinct groups of caregivers and the differentiation is meaningful. Interventions that aggregate African American caregivers into one category for treatment without differentiating treatment options based on subtypes may not be in the best interests of the families they serve. Findings from this study can help develop comprehensive yet targeted policies and practice interventions.