The overdose and substance use disorder epidemic has had devasting consequences for families since parental substance use disorders put children at elevated risk of abuse and neglect To address this dual crisis, 53 Ohio county child welfare agencies are implementing Sobriety Treatment and Recovery Teams (START), an effective intervention for improving parents’ access to substance use disorder treatments (including Medication Assisted Treatment), sobriety, and reunification with their children. However, 1/3 of Ohio START teams (mostly rural or Appalachian) stopped serving families because of high turnover among case workers and family peer mentors. Family peer mentors are peer supporters with lived substance use disorder recovery and child welfare histories who help parents navigate across systems, but often leave their position after feeling unsupported and stigmatized within child welfare agencies. Supervision coaching has the potential to improve supportive supervision practices in child welfare agencies, and create healthy work environments that support staff retention, and continued implementation which is needed to address the overdose crisis among families. This project will test the effectiveness of a supervision coaching strategy to promote workforce stability and START implementation while also creating conditions for strategy sustainment in Ohio’s child welfare system. This Hybrid Type III implementation-effectiveness study will test the impact of CrOss System Technical Assistance for Retaining Staff (COSTARS), a supervision coaching strategy for child welfare supervisors focused on supportive supervision practices, destigmatizing substance use disorder, and improving collaboration between child welfare and substance use disorder treatment systems. Building on our earlier research and partnerships, the goal of the R61 phase is to refine and build the COSTARS model. R61 Aim 1: We will tailor COSTARS to local supervisor, family peer mentor and caseworker needs by convening a community workgroup, conducting a supervision needs assessment, and developing a manual and fidelity measure for COSTARS. R61 Aim 2: Through a series of in-person and virtual didactic training, coaching, and fidelity feedback sessions, we will train COSTARS coaches (a peer coach paired with an expert supervision coach) to competency. The goal of the R33 phase is to investigate the real-world effects of COSTARS. Using a stepped wedge cluster-randomized implementation trial design, COSTARS will be delivered to 40 Ohio START teams in 4 waves. R33 Aim 1: Drawing on annual survey data from all staff members, we will examine the effects of COSTARS on perceptions of their work environment, and retention. R33 Aim 2: We will analyze program records from 1300 parents expected to participate in START to examine the effects of COSTARS on service timeliness, implementation fidelity, and parental outcomes. The proposed study is significant because we will build child welfare system capacity to address the opioid epidemic and address treatment inequities in rural and Appalachian communities, where overdoses and child maltreatment are especially high. This study is part of the NIH’s Helping to End Addiction Long-term (HEAL) initiative to speed scientific solutions to the national opioid public health crisis. The NIH HEAL Initiative bolsters research across NIH to improve treatment for opioid misuse and addiction.