Profiling Changes in Caregiver Needs in Child Welfare: Implications for Maltreatment Risk

Background and Purpose: Annually, three million children and families experience contact with the child welfare (CW) system for maltreatment investigations or assessments. Despite this being a critical opportunity for caseworkers to bridge caregivers with resources to mitigate risk factors for maltreatment, evidence indicates that caregivers’ needs often remain unmet. Emerging evidence shows that caregivers involved with the CW system have complex, integrated needs. Yet, few studies have used person-centered models to identify need patterns or link them to maltreatment risk. Accordingly, this study aimed to identify patterns in caregivers’ areas of service need, number of needs, and stability of needs over time; and assess how these need profiles predict subsequent maltreatment among caregivers of toddlers with prior CW contact.

Methods: Data were drawn from the National Survey of Child and Adolescent Well-Being II, focusing on caregivers of children aged 12-36 months at the baseline maltreatment investigation, in which the caregiver and the child’s placement setting did not change across waves (N = 683; White [40%], Black [28%], and Hispanic [27%]). Caregiver needs were measured at Time 1 (T1; baseline) and T2 (18 months post-baseline) via self-report and/or caseworker referral, and categorized into eight indicators: legal, employment, housing, parenting skills, mental health, substance use (six dichotomous), number of needs across timepoints (0, 1-2, ≥ 3), and stability (never needed, unstable [needs at one timepoint only], chronic). Maltreatment was measured at T2 using the Parent-Child Conflict Tactics Scale. Latent class analysis (LCA) was conducted using maximum likelihood estimation with robust standard errors to identify caregiver needs profiles. The Bolck-Croon-Hagenaars method was used to test maltreatment as an outcome of class membership.

Results: The four-class solution was selected as the optimal solution based on fit indices, parsimony, and interpretability: “No Needs” (low probabilities of endorsing needs at any time [25%]), “Changing Needs” (moderate probabilities of legal, housing, and mental health needs, high probabilities for 1-2 needs and unstable needs over time [24%]), “Chronic Multi-Type” needs (moderate-high probability for all needs, high probabilities for ≥ 3 needs and chronic needs over time [25%]), and “Parenting Only” needs (high probabilities for parenting skills need only and chronic need over time [26%]). Class membership was associated with subsequent physical abuse (χ2[3] = 7.92, p < .05), emotional abuse (χ2[3] = 32.05, p < .001), and neglect (χ2[3] = 15.66, p = .001). Pairwise comparisons revealed that compared to caregivers with “No Needs,” caregivers with “Chronic Multi-Type” needs and “Parenting Only” needs (26%) were both significantly more likely to engage in all maltreatment types, whereas caregivers with “Changing Needs” (24%) were more likely to engage in emotional abuse and neglect.

Conclusions and Implications: Involvement with CW presents a key opportunity to provide services that can strengthen families, reduce maltreatment risk, and support family preservation. Our findings suggest that caregivers with multiple enduring needs, those needing parenting skills only, and those with needs that change over time require unique services. Timely, targeted support that addresses chronic and evolving needs is crucial for families impacted by the CW system.