Foster Care and Substance Use Survey

Attachment A: Site selection criteria for qualitative study data collection

The Relationship Between Foster Care and Substance Use Study: A mixed Methods study

Overall Study Questions:

• What is the relationship between foster care caseloads and substance use at the national level?

• How much does this relationship vary at the subnational level, and which counties see the strongest positive relationship?

• What are the mechanisms by which substance use may lead to increase foster care caseloads?

• In what ways does opioid use disorder impact foster care caseloads differently than other types of substance use disorder?

• What system and community-level factors contribute to the subnational variation in the relationship between substance use and foster care caseloads?

The quantitative portion of the study has focused on the first three questions while the current, qualitative component, will focus on the last two.

Nine primary criteria informed the selection of our sample of sites. We have defined a site as either a county or a small cluster of contiguous counties.

Selection Criteria

Selection Criteria	Description and Rationale
Relationship between measures of substance use and foster care entries.	We use two measures to triangulate substance/opioid use: Prescription opioid sales (ARCOS) Drug overdose deaths We used AFCARS as the data source for children/youth in foster care. Counties are classified into four categories: high foster care rate and high prescription opioid sales high foster care rate and low prescription opioid sales low foster care rate and high prescription opioid sales low foster care rate and low prescription opioid sales The overall numbers in the foster care system have increased steadily over the past 4 years. Anecdotally we have heard from state officials that the increase in the numbers of children in foster care is due to the rise in substance use. We selected sites in the first three categories to better understand the relationship of the increase in the foster care and substance use.
Child Welfare Systems	Alternative response (AR) (often referred to as differential response) is a child welfare system reform effort that is a formal response from the agency that assesses the needs of the child and/or family without requiring a determination of maltreatment). This response is usually reserved for those reports where the identified child is at a low or moderate risk of maltreatment. This effort is an attempt to ensure the safety of children and at the same time respond more effectively to their developmental needs and the needs of their families. An AR initiative may be implemented at the state or county level. Currently 25 states and the District of Columbia have implemented an AR program. These states were identified as having AR programs because they have submitted data identifying the numbers of children participating in their AR program to NCANDS. It is not known at this time if the AR program has been implemented state wide or simply in certain locales. Five tribal child welfare programs have implemented AR programs. Those tribal programs are not a part of our initial sample. (from: https://www.childwelfare.gov/pubPDFs/differential_response.pdf). Kinship care is where children are cared for by family members (as defined by state statute), rather than non-family foster parents. Kinship care is available in all states. We believe it is likely the use of both AR and kinship care influences how child welfare systems address parental substance use and treatment.
Geographical Diversity	In order to ensure that we have represented the cultural, economic, social, and political circumstances from across the country we have selected that sites reflect this diversity.
Geographical Diversity: Urbanicity	The relationship between involvement in the child welfare system and urbanicity is complex and is also related to poverty, the child’s age, and the presence of mental health disorders (Barth, Wildfire, Green, 2006). Related research has shown a geographic variations in two other notable areas of child welfare. First, the receipt of psychotropic medication varies by geography with those in mostly rural and mostly urban areas being more likely to receive stimulants than those who were living in fully rural areas (from: http://files.eric.ed.gov/fulltext/ED539228.pdf). Secondly, placement in congregate care also varies by geography with urban counties using more congregate care than nonurban counties (https://fcda.chapinhall.org/process-quality-and-capacity-investments/new-report-on-the-use-of-congregate-care/). Using the USDA RUCC Codes, we separated the counties into four groups: large metro, small metro , micropolitan, and rural (https://www.ers.usda.gov/data-products/rural-urban-continuum-codes/)
Presence of substance abuse treatment facility, particularly opioid treatment program	The availability of substance use treatment, particularly medically-assisted treatment in a county is likely to impact how the child welfare makes decisions about foster care placement. The availability of these centers is indicated on the spreadsheet. We are also aware that parents may have access to treatment facilities that are located in neighboring counties. We plan on obtaining that kind of information through the interview and focus group process.
State laws on parental Substance abuse	States have different legal frameworks for how parental substance abuse is handled in the context of child maltreatment. For example, in 2017 24 states consider substance use during pregnancy to be child abuse (from: https://www.guttmacher.org/state-policy/explore/substance-abuse-during-pregnancy).
State or County Administration of Child Welfare	The degree of local control over the administration of child welfare services may play a role in how parental substance use is treated. Child welfare services are administered either entirely or partially at the county level in twelve states (from: https://www.guttmacher.org/state-policy/explore/substance-abuse-during-pregnancy).
Presence of Family Dependency Treatment Drug Courts	Family dependency treatment court is a juvenile or family court docket of which selected abuse, neglect, and dependency cases are identified where parental substance abuse is a primary factor. Although a state may be recognized as having a family dependency treatment court, there may not be one present in the specific county/ies that are selected sites for this study. In 2014, 12 states did not have family drug courts (ND, SD, KS, TX, AR, KY, WV, SC, DE, CT, MA, NH from: http://www.nadcp.org/learn/find-drug-court).
Tribal community	Tribal communities face unique challenges with substance use and child welfare.
State has seen large increase in foster care	AFCARS data from 2015 indicate that the total number of children in foster care nationally has increased for the third year in a row. 71% of the states reported increases. Of the five states with the largest increases, we selected one to include in this study.

In addition to the above criteria, additional factors were considered. These factors are not primary criteria, but were reviewed to ensure adequate representation. These include:

• Percent of population that is a minority. Communities with different racial/ethnic profiles may approach parental substance use differently. Research also identifies systematic differences in child welfare removals by race/ethnicity.

• Historical experience with substance use and child welfare. To get a range of perspectives, communities that have different degrees of experience dealing with substance use – particularly opioid misuse - in the child welfare system could be considered.

• Other state-wide implemented programs. Due to the severity of opioid use in certain states, systems such as public health departments, governors’ councils have been mobilized and have implemented programs that may affect how many children are placed into foster care.