Since 1994, federal law and regulation have required states to collect case-level information on all children for whom the state child welfare agency has responsibility for placement, care, or supervision and on children adopted under the auspices of the state’s public child welfare agency.
The Adoption and Foster Care Analysis and Reporting System (AFCARS) includes information on foster and adoptive parents. The information required by AFCARS is what a social worker would normally collect during the course of assessment, planning, and service provision, so workers do not need to collect additional information solely for the purpose of meeting AFCARS requirements. The Administration for Children and Families (ACF) uses the data for many purposes, such as responding to requests from Congress and the public for current data on children in foster care or those who have been adopted; policy decisions; budget decisions and state allocations; monitoring; and technical assistance for states.
The information collected and reported via AFCARS is critical to the federal government. The government uses it to determine a state’s level of compliance with the national standards on child safety, permanence, and well-being. In connection with these standards, all states have undergone a Child and Family Services Review (CFSR) and have developed a CFSR-related Program Improvement Plan. The government either has reviewed or will review the automated information systems of states with an operational Statewide Automated Child Welfare Information System, and at some point, expects all states to have an AFCARS Assessment Review.
The following discusses errors identified during the AFCARS assessment review process. This information is intended to assist reporting agencies in improving the quantity and quality of the information that they report via AFCARS.
Automated System Features
Errors Identified During AFCARS Reviews One identified problem is that some states use placement codes that they cannot appropriately map to valid AFCARS values. If they cannot map these codes to any valid AFCARS value, then they must map the codes to “blank” in AFCARS. A second problem is that in many instances, the number of previous placement settings in a removal episode is significantly high. It appears this is due to the design of the program logic, which counts every placement setting. For instance, AFCARS counts moves that occur for reasons of respite or short-term hospitalizations in the count of previous placement settings.
Although some placements settings are included that should not be counted, in some instances, placement settings are missed. For instance, moves that occur within a contracted placement-provider setting are frequently not included in the count of previous placement settings. AFCARS requires capturing and counting these types of placements settings for Foster Care Element 24.
To be accurate, this AFCARS foster care element must correctly calculate the total number of placement settings during the current foster care episode, that is, it does not count runaway and trial home visits and does count the current placement in the total. It should include respite placements in the count of previous placement settings if the child moves to a different placement after the respite instead of returning to the previous placement setting. The following from the Children’s Bureau’s Child Welfare Policy Manual provides further clarification on this AFCARS element: Trial home visits and runaway are the same in that they are recorded as placement settings.
This allows the State to more accurately show the physical setting that the child is in at the time the report period ends. However, they are not counted in the number of placements in order to guard against misleading data. For example, in the case of a child in a group home who runs away for one week and then returns to the group home, if the runaway were counted as a placement it would appear that the child had three placements, when actually the agency had only placed the child once. Likewise, if the child was placed in a trial home visit with the intent that the child would be discharged back to the family the trial home visit should not be considered a placement. The number of placements element is meant to gather information on how many times the agency found it necessary to move the child while in foster care. (45 CFR 1355, Appendix D, Detailed Foster Care, Element 24)
Missing AFCARS Information
For AFCARS extraction and submission purposes, information not collected or not available for a particular client record (for whatever reason) is mapped as all blanks (not all zeros, all 9s, etc.). Information that is missing or not collected should never be mapped to any valid AFCARS value.
Readers may obtain technical assistance from the Children’s Bureau’s National Resource Center for Child Welfare Data and Technology (NRC-CWDT). The resource center can be contacted at its Web page: http://nrccwdt.org. If you wish to request onsite technical assistance from the NRC-CWDT, contact your ACF Regional Office.