Child and Caregiver Outcomes Using Linked Data (CCOULD)

STATUS: Completed Project

BACKGROUND 
Research has demonstrated that parental substance use negatively impacts child health and other outcomes later in life and is a leading determinant of child maltreatment and foster care placement. Child welfare-involved parents with substance use disorders (SUDs) are an especially vulnerable population facing additional barriers to receiving SUD treatment.  

Understanding the correlation among treatments, additional supports provided by the child welfare system, and a return to successful daily functioning and family stability is central to assessing positive patient outcomes for parents receiving treatment. State child welfare agencies generally track the outcomes of children involved in their systems and regularly monitor outcomes as youth transition to adulthood. However, health outcomes for child-parent dyads involved with the child welfare system are rarely tracked regularly and consistently, which impedes the ability to understand how a parent’s SUD affects child-welfare involved youth. Additionally, little is known about the use of Medicaid for treatment of SUD or mental health among parents of welfare-involved children, as adequate data on parents are rarely collected. A major barrier to addressing these evidence gaps is the lack of comprehensive data relating to SUD diagnoses and treatments for parents involved with the child welfare system.

PURPOSE 
The purposes of this project is to enhance data infrastructure and increase data availability for research on a set of parents that have children in the child welfare system and are in need of treatment for opioid use disorder (OUD), other SUDs, or behavioral health issues.

The overall objectives of the project are to:

• Develop datasets that link records from State Medicaid and from child welfare systems in two to four states. These data sets will contain linked patient-level data including Medicaid enrollment, patient diagnoses, services, and claims, along with child welfare outcomes.
• Prepare a single, harmonized research use dataset and develop a process for external researchers to access the data.
• Develop an approach for states to have an ongoing link between Medicaid and child welfare data systems, including integrating Medicaid eligibility, enrollment, and claims data with the new CCWIS. In addition, create a roadmap for other states to follow that documents the process as well as lends insight into lessons learned, challenges and successes.
• Design, conduct, and encourage analyses on the research use linked dataset created in the research use dataset.

KEY IMPACTS 
Improving access and use of data: Linked dataset to study outcomes of welfare-involved families 
The project linked records from state Medicaid and child welfare systems for two states to develop the Child and Caregiver Outcomes Using Linked Data (CCOULD) dataset. This dataset provides a more complete picture of the services families receive that are funded by child welfare agencies and Medicaid.

 Improving data quality: New methods to address missing data and data harmonization 
The project developed a data linkage methodology that allowed for linkage of records that lacked Social Security numbers, which helped address prior data linkage challenges. Additionally, the project team created a common data model (CDM) for the linked CCOULD dataset. The CDM harmonized records across child welfare and Medicaid data systems, making them easier for researchers to use for analyses.

Enhancing analytic resources: Roadmap to support child welfare and Medicaid data linkages 
The project developed a report with guidance and lessons learned for other states interested in developing their own research use linked data across child welfare and Medicaid systems.

PUBLICATIONS 
CCOULD Project Overview. This document provides a brief description of the CCOULD project, including information on the participating states and available project resources.

CCOULD Dataset and User Guide. The de-identified CCOULD dataset is available to researchers free of charge at the National Data Archive on Child Abuse and Neglect at Cornell University. The CCOULD dataset contains linked records on 1,087,763 children and 89,871 caregivers and covers the periods of January 2017 through June 2021 in Florida, and January 2017 through 2020 in Kentucky.

Linking Child Welfare and Medicaid Data: Lessons Learned from Two States. This report provides lessons learned on recruitment, data linkage and harmonization, data privacy, data governance, and sustainability to state, tribal, and local child welfare and behavioral health agencies interested in linking their Medicaid and child welfare data.

Child and Caregiver Outcomes Using Linked Data (CCOULD) Project Common Data Model. This CDM offers potential research questions that may be answered using the CCOULD dataset and provides a common data model that may be of value to other agencies looking to link their data. 

Advancing Research on Intersections of Child Welfare and Medicaid Using Linked Data from the CCOULD Project. This brief aims to increase usage of CCOULD data by researchers, drawing attention to the capacity of the data to answer pertinent research questions, and assist researchers in maximizing the utility of the data and navigating limitations.

Timing of Medicaid Enrollment for Children Engaged with Child Welfare Services in Two States. This brief examines the timing of Medicaid and Children’s Health Insurance Plan (CHIP) enrollment relative to timing of engagement with child welfare services for children in Florida and Kentucky using the CCOULD data.

Project Final Report. The final report summarizes critical background information, major project accomplishments, lessons learned, and key products and publications resulting from the project activities.