Dissemination of a Reliable and Valid Family Maltreatment Determination Model: 33-Site, System-Wide Replication and Extension of Heyman and Slep (2009a)

The need for standardized criteria in partner and child maltreatment response systems is critical for providing fair decisions, allocating family support, producing reliable research findings, and aiding prevention efforts, among other tasks. The primary goal of this study was to replicate Heyman and Slep’s (2009a) study—whether maltreatment incident determination committee decisions of local sites matched those of master reviewers. This study extended the prior work by testing if specific training techniques (i.e., social worker self-assessments, motivation-enhanced briefs, per-case feedback) increased the master reviewer-site concordance of the Field-Tested Assessment, Intervention-Planning, and Response (FAIR) system. Overall agreement between the master reviewers and the committees was 87%, indicating good agreement but falling just below the 91%–92% achieved in earlier FAIR dissemination studies. Sites were randomly assigned to one of eight training conditions using a multiphase optimization strategy design. The full factorial model was not statistically significant, indicating that the training techniques were not associated with committees’ voting fidelity. The replicated agreement findings indicate that the FAIR system can be implemented successfully on a large scale and produce consistently good agreement on family maltreatment case decisions. Areas for future research are discussed (e.g., testing and usage of evidence-informed training techniques).