Predictors of Child Welfare Caseworker Referrals to an Evidence-Based Parenting Program

Background and Purpose: Even with the increasing number of evidence-based practices (EBPs) for children and families, child welfare agencies continue to encounter barriers to EBP implementation, including a lack of utilization by child welfare-involved children and families. Since families typically access services through their child welfare caseworkers, a lack of referrals to EBPs by child welfare caseworkers could be contributing to a lack of EBP utilization. Previous research seeking to increase caseworker referrals examined the effectiveness of a training intervention aimed at increasing caseworker knowledge and awareness of local EBPs; results showed no intervention effects on caseworker referrals (Dorsey, Kerns, Trupin, Conover, & Berliner, 2012). Thus, further research is needed to determine additional factors besides knowledge and awareness that impact child welfare caseworker referrals.

This study sought to explore caseworker demographic factors that predict caseworker referrals.  Relying upon tenets of the Theory of Planned Behavior (TPB), a psychological theory to explain the relationship between intention and behavior, this study also examined whether intention to refer predicts caseworker referrals to an EBP.

Methods: The study was conducted at two community-based agencies contracted to provide child welfare services that were implementing the Positive Parenting Program (Triple P), an evidence-based parenting program. Survey data was collected from a convenience sample of 130 caseworkers (N=72) and support staff (N=58) with the ability to refer families. Participants had worked an average of 6.2 years with families and nearly 78% identified as African American. Measures included a demographic survey and a questionnaire measuring caseworker intention to refer families to Triple P based upon Azjen, (2006) TPB model. Researchers collected actual caseworker referral data two months post survey data collection from agency administrators.

A negative binomial regression was conducted to determine the predictors of caseworker referrals, given that it accounts for positively skewed distributions commonly observed with count data (i.e., number of referrals) and allows for overdispersion of the data. Missing data on the dependent variable was accounted for using Full Information Maximum Likelihood.

Results: Results showed that the participants from Agency A were 14.58 times more likely to make a referral to Triple P than participants from Agency B (b=2.68, p<.05). Participants who identified as African American were .05 times less likely to refer to Triple P as participants who did not identify as African American (b=-2.96, p<.05).  No other variables significantly predicted number of referrals.

Conclusions and Implications: Surprisingly, intention to refer did not predict referral behavior, which may have been caused by intervening events (i.e., housing, substance abuse, mental health needs) that took precedence over parenting needs. Providing support to caseworkers in navigating well-being needs amidst emergent crises that arise may mitigate this barrier to caseworker referrals. Findings underscore the need to explore organizational differences that may facilitate caseworker referrals to Triple P. Differences to examine in future research include the impact of physical co-location of Triple P providers in the child welfare agency, as this was the case at Agency A but not Agency B, and differing agency norms around Triple P referrals.