How can we ensure a child’s first placement is with a family? 


Separating children from their families is a traumatic experience. When children must be removed from their families and placed in out-of-home care, one of the ways to mitigate that trauma is to place the child in the best possible setting, right from the start. Across the country, child protection agencies have developed ways to secure the most appropriate first placement for children who come into care: with families. Placement with families reduces trauma for the child and decreases reliance on congregate care settings.

These system changes—moving away from initial placement in crisis nurseries, emergency shelters, diagnostic facilities, or receiving centers, and prioritizing the placement of children with kin or foster families from the start—in turn support placement stability and the achievement of timely permanency.

First Placement…Best Placement…Only Placement

A child’s first placement can have implications for long-term outcomes. Children initially placed in congregate care experience, on average, more placement changes than children placed with relatives or foster parents. At least one study1 found that children initially placed with relatives are the least likely to experience placement changes, less likely to re-enter foster care, and more likely to achieve permanency. Placement instability, in turn, can lead to poor educational and mental health outcomes, and delayed permanency.2 The use of shelter care may, in fact, be based on capacity rather than clinical need, as research indicates that children who are initially placed in emergency shelter care actually have a lower level of behavior problems than their peers in family-based foster care.3

Recognizing the impact of a child’s first placement, agencies in several jurisdictions have asked for strategies to improve their capacity to place children in family settings at the outset rather than in group settings. Interviews were conducted with leadership from six agencies that have made changes in order to support a child’s first placement in family-based settings: Arkansas,4 Georgia,5 Idaho,6 Iowa,7 New Jersey,8 and Washington, D.C.9 Three of these jurisdictions—Georgia, New Jersey, and Washington, D.C.—previously had used receiving centers when children entered placement but since have disbanded them in order to provide family-based first placements. Agency changes were motivated by a desire to reduce trauma for the child, as well as concerns related to the quality of care provided, the length of time children were remaining in receiving centers, and the safety of younger or more vulnerable children when placed alongside older youth with high behavioral health needs.

Strategies that have played a role in improving the agencies’ capacity to place children in families when they enter care are summarized below, as well as lessons learned and advice from agency leaders for child protection agencies seeking to make similar improvements.

Key strategies to support first placement in families

Leadership vision and expectations

Child protection leadership plays a key role in setting forth a strong vision and firm expectations for the use of family-based first placements, and clearly and repeatedly articulating why they matter. In Arkansas, the director of the Department of Human Services and the child welfare director, with support from the governor, teamed up to present a unified vision that children needed to be in a family setting from the initial placement. As part of the effort, the state has reduced the number of children in emergency shelters, especially younger children.

In Washington, D.C., senior leadership articulated urgency for reducing reliance on emergency placements so that placing a child into out-of-home care can be a more planned and thoughtful process. This has helped to significantly decrease the number of emergency and after-hours placements.

In New Jersey, leadership expressed consistently and clearly to staff, providers, and stakeholders: “We’ve moved away from just filling a bed or finding a placement. We’re looking for a family.”

Systemwide behavioral barriers

Changing the way that caseworkers approach placement into out-of-home care can be challenging, especially if receiving centers or shelters are easily accessible. Putting in place systemwide barriers to the use of congregate care is necessary to ensure first placements are family-based placements. The science of human behavior explains some of the challenges at play for agency leadership:

  • People tend to be confronted with much more information than they are willing or able to process.
  • People seek to minimize effort and are disproportionately affected by small barriers to change.
  • People typically stick with the way things are—the status quo.
  • People tend to interpret facts using mental “shortcuts” (rules of thumb or assumptions) that confirm existing views.10

Incorporating business process roadblocks to help change caseworker behavior is also important. In Washington, D.C., for example, approval from the deputy director for program operations is required before a child who is entering care can be placed in a group care setting. Instituting additional hurdles and time-intensive activities—such as requiring written authorization or multi-step approval processes in order to place a child into a non-relative or congregate care setting—is essential for systemwide change.

Planned placements

One of the other primary strategies shared by interviewees for improving a child’s entry into care is a focus on planning for the placement whenever possible. Moving from a system that operates in crisis and responds primarily to emergency placements to a system that plans for placements has multiple benefits. These include:

  • Families are better able to partner in the process.
  • Children’s needs are identified.
  • Service plans can be developed.
  • The most appropriate placements can be located.

As part of the shift to planned placements, leadership in Washington, D.C., consistently emphasizes the importance of communication in this process. For example, investigation staff should be communicating with placement staff as soon as they know that there is a possibility they might need to remove a child, even if they do not have a court order for placement yet. This allows for early identification of appropriate foster homes in the event that a relative placement is not available.

Georgia’s Pre-Removal Staffing, New Jersey’s Pre-Placement Conference, and Polk County, Iowa’s Pre-Removal Conference are all models for meeting with the family prior to removing a child, with the goal of including the family in the removal process. These strengths-based meetings keep the focus on reducing trauma for the child. The agencies note that as a result, relative placements and educational stability have both increased. In the event that a relative placement is not available, these meetings still provide parents with a chance to be involved in decision-making, and provides the child protection agency with as much information as possible to identify a foster home that can best meet the child’s needs. In Polk County, an income maintenance worker attends the Pre-Removal Conference to ensure the identified relative caregiver is connected to all eligible benefits before the meeting ends. The family also leaves the meeting with a visitation plan in place so that both the parents and the child know when they will see each other again. This level of engagement and communication has improved the agency’s relationship with the community in general, and has also been embraced by staff, county attorneys, and the judiciary.

In Georgia, a Safety Resource policy allows parents to voluntarily place a child with a relative for 45 days as parents work to resolve any safety concerns. These planned placements can also extend across state lines. Georgia has worked closely with its border states, such as Alabama, to develop a strong border agreement that allows caseworkers in border counties to quickly place children with a relative across the border. If the safety concerns continue, the child can remain with the relative while the agency takes custody, regardless of state lines.

Relative placements

In addition to prioritizing relatives during planned placements, all of the interviewees shared that relative placements are also prioritized during emergency placements. Though the tools vary from agency to agency—for example, Idaho uses the Code X national background check while Washington, D.C., uses LexisNexis Accurint—interviewees expressed the same goal of quickly identifying, assessing, and approving relatives so that children in need of out-of-home care can be with family. In some jurisdictions, such as Idaho and Iowa, the definition of “family” has been expanded to include fictive kin and other adults who have a relationship with the child, such as a day care provider.

Arkansas has designated resource staff in each region to support emergency placements with relatives, and Washington, D.C.’s child protection agency includes a kin unit dedicated to quickly identifying and engaging relatives in the local area, including relatives across the border in Maryland. Through a border agreement with Maryland and a strong relationship with the National Center for Children and Families (Washington D.C.’s private agency partner), relatives in Maryland can be licensed on an emergency basis so that state lines do not pose a barrier to placing children with extended family.

Identifying and supporting foster parents

While relative placement is the preferred placement whenever possible, having a large pool of foster parents is also key to placing children with families rather than in a receiving center or other type of congregate care. To this end, New Jersey developed a menu of diligent recruitment strategies, such as market segmentation, which involves looking at the broad population of potential foster families, dividing them into subsets that have common characteristics, and then implementing strategies to target them. The recruitment efforts have been so successful that the system currently has 50 percent more foster home beds than children in out-of-home care. Included among these foster homes is an identified set of foster parents ready, willing, and able to be emergency placements.

Georgia, Idaho, and Washington, D.C., also have some emergency foster homes available so that children are placed in family homes, even on weekends, after hours, or in the middle of the night. In Arkansas, an optional mass text-messaging tool allows hundreds of foster parents to be immediately notified when there is a placement need. Those who are interested can reply quickly just by responding to the text. This system has given staff more placement options and helped them make better placement decisions, keeping more children in their communities and with their siblings.

In addition to having a sufficient number of foster homes available, all interviewees emphasized that building ongoing, meaningful relationships with foster parents is vital to the success of placing children in those homes, especially in emergency situations. When caseworkers are familiar with the strengths, skills, and preferences in their agency’s network of foster parents, they can more quickly match children with appropriate homes that meet their needs. Foster parents also are more likely to accept placement of children if they know that staff have engaged in this kind of matching up front, and that they will be supported in caring for the child following placement. In New Jersey, a caseworker follows up with the child and the foster parent the day after placement to troubleshoot any issues, even on weekends or after hours. Washington, D.C., foster parents are supported by the Foster Parent Support Unit, which engages with and advocates for foster parents. Peer support also can be helpful. Both Idaho and New Jersey have foster parent peer mentor programs, and New Jersey operates a foster parent peer support helpline as well.

Supporting children’s behavioral health needs

Behavioral health services can also make a difference in foster parents’ willingness or ability to accept placement of a child, particularly children with higher needs, as well as support placement stability from the start. Idaho provides Community Based Rehabilitation Services workers to help support children with significant developmental or behavioral needs in their home, school, or other community-based setting, while New Jersey deploys a Mobile Response and Stabilization Services (MRSS) worker to the foster or relative home within 72 hours of a child entering out-of-home care or transitioning to a new placement. This initial visit is designed to acknowledge the child’s trauma and plan for any behavioral challenges that might arise as a result. In the event of a crisis, caregivers also can call MRSS around the clock, 365 days a year, and a worker will be on site within an hour to help de-escalate the crisis, assess the situation, and develop a plan with the child and caregiver.

Supporting staff

The process of finding the best placement for a child entering care can be a time-consuming one, especially for CPS investigators who may be juggling multiple cases. Interviewees shared several different strategies for supporting staff during this process, so that staff do not simply resort to the first available placement because of time pressures. For example, in Washington, D.C., the Placement unit identifies available foster homes, while the Kin unit identifies relative caregivers. Similarly, Georgia utilizes regional Resource Development teams to identify available state-licensed foster homes, as well as a Placement Resource Operations unit when specialized expertise is needed to place children who have more acute or special needs. Both teams employ Georgia’s “bulls-eye” approach, looking first for homes in the child’s home community, then the child’s home county, then surrounding counties, then the region, and finally, outside of the child’s region as a last resort.

In New Jersey, a matching tool in the agency’s Statewide Automated Child Welfare Information System is available to help caseworkers match a child’s characteristics with an appropriate foster family. In addition, placements are done using a buddy system, so that while one caseworker is exploring placement options, another is engaging and supporting the child. This helps reduce the child’s trauma and also allows one caseworker to focus on finding the best and most appropriate placement without having to attend to the child at the same time.

In the same spirit, Idaho is piloting an Office Moms and Dads program, which calls on volunteers to come to the office and accompany children who have been removed from their homes while caseworkers research placement options. The response from staff in the pilot region has been so positive that Idaho is planning to expand this initiative statewide.

Assessment tools

Part of identifying the best possible first placement for a child requires having a clear understanding of the child’s strengths and needs, particularly when a relative placement is not available. This often is one of the advantages cited in using receiving centers, given that they also may serve as assessment centers. But interviewees shared that the use of assessment tools, such as the Comprehensive Safety Assessment in Idaho and the Treatment Outcome Package in Iowa, can fulfill the same purpose and support a child’s first placement with a family that best meets their needs.


Data related to the strategies described above was difficult to obtain, as most agencies do not track trend data related to a child’s first placement in care. However, all of the interviewees stated anecdotally that the focus on prioritizing family-based placements is contributing to an increase in relative placements and a decrease in congregate care placements. In Arkansas, where leadership has been carefully tracking trend data, the percentage of children in kinship care increased from 14 percent to almost 30 percent between May 2016 and August 2017, and during the same time period the number of children in emergency shelter placements was reduced by more than 50 percent. Iowa has reduced the overall population of children in congregate care by 46 percent, from 29 percent (1,972 children) in 2005 to 18 percent (1,068 children) in 2015. Washington, D.C.’s reliance on congregate care decreased 75 percent: from 12 percent (297 children) in 2005 to 8 percent (75 children) in 2015.11 Similarly, New Jersey reduced the overall population of children in congregate care by 45 percent, from 12 percent (985 children) in 2009 to about 7 percent (455 children) in 2016. In 2016 New Jersey placed 92 percent of its children entering care (4,042 children) in a family setting. Of those, 33 percent (1,323 children) were placed with a relative.12

Lessons learned

Finding the best possible placement for a child who is entering out-of-home care is not an easy task, and interviewees acknowledged that their systems still struggle at times. They shared advice and lessons that they have learned along the way so that other child protection agencies can benefit from their experiences:

  • If an assessment/receiving center is no longer available, staff will find other ways of placing children in families. Instead of continuing to operate receiving centers or intake shelters, take the strengths of the receiving center model, such as the ability to place children in the middle of the night and provide assessment services, and incorporate these elements into a process that is more child-centered. For example, use assessment tools and develop a robust pool of foster parents who are willing to take a range of children, including as emergency placements.
  • Similarly, if congregate care is easy to access, it will be used—especially if a receiving center is no longer available. If congregate care is harder to access, reliance on that form of placement will decrease.
  • Although children with significant behavioral health needs may be a small portion of the overall child welfare population, they often are the most challenging to place. In moving toward family-based placements, invest significant time in developing sufficient resources for youth with behavioral health needs.
  • Communicate the importance and feasibility of planned placements, especially relative placements. Extend that communication across teams, such as between the investigator and placement staff.
  • Remove logistical barriers to family-based placements. In New Jersey, staff often transport children to school until a transportation plan can be implemented, and in Washington, D.C., emergency child care is available to both relatives and foster parents.
  • Develop border agreements so that state lines are not barriers for families that live along the border, and consider becoming part of the National Electronic Interstate Compact Enterprise to expedite relative placements with non-border states.
  • Start with the data, monitor it regularly, and share it widely. When Arkansas first drilled down its data, leadership was surprised by some of the findings. These findings helped to shape the agency’s priorities, and leadership continues to look at data regularly to ensure that progress is being made and make course adjustments as needed. Sharing data across the agency and with external stakeholders has been invaluable to buy-in, especially with legislators.

1 Chapin Hall Center for Children at the University of Chicago. (2007). Foster care dynamics 2000-2005: A report from the Multistate Foster Care Data Archive. Retrieved from
2 Moore, T.D., McDonald, T.P., Cronbaugh-Auld, K. (2016). Assessing risk of placement instability to aid foster care placement decision making. Journal of Public Child Welfare, 10(2), 117-131.
3 Chadwick Center for Children and Families & Chapin Hall. (2016). Using evidence to accelerate the safe and effective reduction of congregate care for youth involved with child welfare. Retrieved from
4 Arkansas interviews with Mischa Martin, Director, DCFS; Beki Dunagan, Assistant Director, Placement Support and Community Outreach; Tiffany Wright, Program Manager, Foster Care, Placement Support and Community Outreach; Anne Wells, Assistant Director, Mental Health & Preventative Services; and Megon Bush, Program Manager, Specialized Placement, Mental Health & Preventative Services, August 22, 2017.
5 Georgia interviews with Tammy Reed, Placement & Permanency Section Director, Georgia Division of Family and Children Services, December 19, 2017, and Judge Thomas Hammond, Toombs Judicial Circuit, Georgia, and Deputy Commissioner John C. James, Alabama Department of Human Resources, February 5, 2018.
6 Idaho interviews with Sabrina Brown, Child Welfare Program Specialist, Foster Care Recruitment & Retention, Idaho Department of Health and Welfare, January 16, 2018.
7 Iowa interviews with Nicole Button, Des Moines Service Area Lead Social Worker, and Michael McInroy, Service Area Manager, Des Moines Service Area, Iowa Department of Human Services, November 21, 2017.
8 New Jersey interviews with Elizabeth Manley, SOC Assistant Commissioner, New Jersey Department of Children and Families, November 14, 2017, and Lisa von Pier, CPP Assistant Commissioner, Betsy Sunder, CPP Deputy Director, and Michelle Adams, CPP Deputy Director, New Jersey Department of Children and Families, December 15, 2017.
9 Washington D.C., interviews with Stacy Rodgers, Deputy Director of Program Operations, and Angelia Baker, Placement Administrator, Washington, D.C. Child and Family Services Agency, December 14, 2017.
10 Behavioral Insights Team. (2016). Spreading change: A guide to enabling the spread of person- and community-centered approaches for health and wellbeing. Retrieved from
11 AFCARS data for Arkansas, Iowa, and Washington, D.C., made available from the National Data Archive on Child Abuse and Neglect (NDACAN), Cornell University.
12 Rutgers University School of Social Work. (n.d.). New Jersey child welfare data hub. Retrieved from