Understanding the Adoption and Safe Families Act (ASFA)

The State of the U.S. Child Welfare System and Legislation Before ASFA

Foster care in the United States was modeled after the animal rescue societies of the mid-19th century, and in urban areas evolved into a system of “boarding homes” or “boarding out” with sponsors.[2],[3] In April 1912, Public Law 62-116, advocated by President Theodore Roosevelt and signed by President Howard Taft,[4] created the Children’s Bureau to investigate matters related to child welfare, protecting and caring for “homeless, dependent, and neglected children.”[5],[6] In the four decades after the organization of the American child welfare system, various amendments transformed its operations and breadth.[7] In the 1960s and 1970s, Congress increased funding to support states’ child welfare services.[8]

The Adoption Assistance and Child Welfare Act of 1980 (AACWA), also called the Family Reunification Act,[9] emerged in response to issues that garnered attention in the 1970s, such as the removal of children by insufficiently trained child welfare service (CWS) agents, the phenomenon of foster care “drift” (i.e., lengthy placement away from the birth family),[10] and the failure of CWS to monitor children in foster care or reunify children with their parents.[11] Regarded as a major step in “reforming our languishing child welfare systems,” AACWA was established on the philosophy of permanency planning, which emphasized stable and continuous relationships in a family setting through reunification of a child with their birth family or release for adoption.[12],[13] Among the major and most remembered provisions of AACWA was the requirement that child welfare agencies make “reasonable efforts” toward maintenance and reunification for children and their birth families, a term that neither Congress nor the Department of Health and Human Services clearly defined.[14]

In the decade and a half after the enactment of AACWA, a combination of factors contributed to challenges in the permanency placement goals of CWS. Beyond the uncertain definition of “reasonable efforts” and lack of guidance for judicial findings of such,[15] overworked and under-trained child welfare workers,[16] a paucity of services and material assistance for birth families,[17],[18] the cultural treatment of poverty,[19] an expanding definition and growing awareness of child abuse,[20] and poor enforcement of administrative requirements “crucial for achieving the goals of [AACWA]”[21] have all been proposed as reasons for the continued state of affairs. Moreover, into the latter half of the 20th century, quick-fix-resistant causes for child removal shifted to the foreground of child welfare focus; these included incarceration, homelessness, AIDS,[22] and, most prominently, substance misuse which was attributed to 80% of substantiated abuse and neglect cases.[23]

The foster care population has fluctuated through these decades of legislative reforms. In 1977, it was about 500,000; in 1980, the year of AACWA’s passage, the population was 302,000; in 1990, it rose to 400,000; and, by 1997 was back to 1977 levels.[24] Of the children in foster care in the mid-1990s, about 40% were in care for two or more years,[25] with an average case length for a child being two to three years;[26] about one-fifth of children in foster care would not return to their families due to the termination of parental rights.[27] One national study[28] calculated that from 1977 to 1982, the number of children made legally eligible for adoption but not adopted increased 50%, which some attribute to courts terminating parental rights without judiciously addressing the problem of permanency.[29]

Passage of the Adoption and Safe Families Act

The Adoption and Safe Families Act (ASFA) of 1997 was formulated in response to the aforementioned shortcomings in the child welfare system as well as the reports on the high-profile child abuse cases of Erik Dawood, Elisa Izquierdo, and others.[30] It constituted one piece in achieving the goal of the Clinton administration to double adoptions by 2002, which was announced by the President in December 1996 and accompanied two months later by a report from the Department of Health and Human Services to investigate barriers to and provide recommendations for foster care and adoption.[31] The report was based on 10 principles, the first of which was the assertion that every child deserves to be in a safe and permanent family.[32]

Throughout Congressional discussions of ASFA, the emphasis was on achieving safety and stability for children in the United States. President Clinton affirmed that the “child’s health and safety are the paramount concern of the child welfare system”[33] and that children have a fundamental right to a “chance at a decent, safe home” and “an honorable, orderly, positive upbringing.”[34] Senator John Chafee, a Republican from Rhode Island, celebrated that ASFA represented a “new day for Child Welfare Services in America.”[35] Then-First Lady Hillary Clinton, who helped bring exposure to issues in child welfare and championed the passage of the law, celebrated ASFA as a key legislative victory.[36]

The law itself[37] heralded significant changes to child welfare practices, chiefly the requirement for children in care to have permanency hearings at least every 12 months, and a requirement for states to seek termination of parental rights for children who have been in care for 15 of the previous 22 months (with multiple exceptions).

Additionally, ASFA provided funding for more preventative and support services, expanded subsidies and incentives for adoption, put stricter background checks in place for foster and adoptive families, expanded children’s health care coverage, and much more. Overall, the comprehensive nature of the bill sought to reframe the paradigm of foster care’s purpose as seeking the best interest of children rather than “the current system of always putting the needs and rights of the biological parents first.”[38]

Congressional Representatives on both sides of the aisle supported the Act. At the fore of writing ASFA were sponsor Representative Dave Camp (R-MI) and original co-sponsors Representative Barbara Kennelly (D-CT) and Representative E. Clay Shaw (R-FL).[39] Of the 29 other co-sponsors, 19 were Republican and 10 were Democratic. In the Senate, lead sponsors included “conservative Republican Mike DeWine, moderate Republican John Chafee, and liberal Democrat Jay Rockefeller.”[40] The Congress-wide support of the bill was borne out further by the final votes: 406 in support and only seven in opposition in the House and unanimous consent by the Senate.[41] At the bill signing on November 19, 1997, President Clinton applauded the “truly bipartisan effort” and “extraordinary example of bipartisan cooperation;” Senator John Chafee echoed this sentiment and Representative Dave Camp called ASFA “bipartisan from the very beginning.”[42]

Most of Congress agreed with Representative Kennelly that ASFA was “a significant first step forward on the road to providing protection and permanency to our nation’s abused, neglected, and sometimes forgotten children.”[43]

—

…the bill sought to reframe the paradigm of foster care’s purpose as seeking the best interest of children rather than “the current system of always putting the needs and rights of the biological parents first.”[44]

—

State Implementation of ASFA

The implementation of ASFA has not been uniform across states; this is important to note when interpreting its impact.

The provisions of ASFA fall into two main categories: child safety provisions and timeframe provisions.[45] The child safety provisions maintained the old requirement that reasonable efforts be made to preserve families, avoiding placement in care and ensuring return for those removed, while also specifying exceptions to this requirement. Federally established exceptions included homicide of a sibling of the child in question, felony assault against the child or a sibling, prior termination of parental rights (TPR) to a sibling, and subjection of the child to aggravated circumstances, such as abandonment and chronic abuse. Under these exceptions, TPR could be fast-tracked; hence, this was called the fast-track provision. Another aspect of the child safety provisions was the requirement for all prospective foster and adoptive homes to undergo a criminal background check.

The timeframe provisions of ASFA, which have historically been more contentious, set timelines for states to hold permanency hearings and to initiate TPR proceedings. Specifically, the provisions mandated that, within 30 days of court findings that family preservation or reunification were not required for a given case, the child should have a permanency hearing. For all children in foster care for whom ASFA was applicable, a permanency hearing had to be held within 12 months of entering care. Related to this is the “15 of 22 provision,” or “15/22 provision,” which requires initiation of TPR if a child has lived in foster care for 15 of the last 22 months. Exceptions may apply for cases involving kinship care, agency documentation that TPR would not be in the child’s best interests, proof that the state has not provided sufficient services to the family, and other compelling reasons.

States have varied in consistency and extent of integrating these provisions in their child welfare operations. After passage of ASFA, it was the responsibility of individual states, working with the federal Administration for Children and Families (ACF), to bring their laws and policies into accord with the federal legislation.[46] Since 2001, the Children’s Bureau (part of ACF) has monitored progress and compliance using child welfare data, case reviews, and stakeholder interviews.

In 2003, the states that were able to provide data related to TPR timeframes reported that the fast-track provision was seldom used and not many TPRs were filed for children who satisfied the conditions for the 15/22 provision to kick in.[47] In fact, in eight of the nine states that submitted data, the number of children exempted from the 15/22 provision exceeded the number of TPRs filed for children to whom the provision was applied.[48] Reasons for not filing TPRs in accordance with these provisions included court delays, judges’ hesitation in relieving states from reunification efforts, and belief that such a ruling would not work in the child’s best interests.[49]

Findings were similar in recent rounds of federal child welfare monitoring visits conducted between 2015 and 2018.[50] Across the 50 states and the District of Columbia, only seven regions were evaluated as having strong TPR practices, while the remainder needed improvement. In 13 states, TPR practices were acknowledged to be inconsistent. Other issues were excessive or repeated extensions granted for permanency timelines in 10 states, lack of adequate timeline tracking in eight states, and reluctance among judges to create “legal orphans” in seven states.

Interviews with staff in three states (Illinois, Utah, and Wisconsin) exemplified the range of experiences. In Illinois and Utah, similar proportions of children who entered care in 2013 experienced TPR within five years; however, in Utah, 87.8% of TPRs occurred within 17 months, versus 15.6% in Illinois. In Wisconsin, a smaller percentage of children experienced TPRs within five years and, of these, 31.7% occurred within 17 months of entry into care.[51]

Given this variation, assessing the impact of ASFA nationwide is a considerable task. State-by-state evaluations would require knowledge of both the language of the provisions woven into individual state laws and analysis of individual state data. Such an endeavor has been only partially undertaken in extant research.[52]

Impact of ASFA

Since its passage over 25 years ago, has the Adoption and Safe Families Act achieved its goals of reducing the number of children in foster care and time until permanency for those children already in the system? Assessing the success of ASFA has been challenging given the lack of comparable pre-ASFA and post-ASFA data, variable enforcement of ASFA provisions, and implementation of other state-level policies and shifts in child welfare practices in some states prior to and alongside ASFA. Nevertheless, noteworthy efforts have been made to quantify the impact of ASFA on important child welfare outcomes, based on the data that is available.

—

Assessing the success of ASFA has been challenging given the lack of comparable pre-ASFA and post-ASFA data, variable enforcement of ASFA provisions, and implementation of other state-level policies and shifts in child welfare practices in some states prior to and alongside ASFA.

—

Foster Care Population

A natural measurement for gauging the effect of ASFA is the size of the foster care population. By setting timelines for reunification and termination of parental rights, policymakers hoped to more quickly move children and youth out of foster care and into permanent placements. Assuming a constant or decreasing rate of entry into foster care, implementation of ASFA should have, in theory, reduced the foster care population and, relatedly, reduced overall time-in-care for the average child welfare-involved youth.

As will be discussed in detail in the next section, adoption rates increased across a majority of states following implementation of ASFA. However, the national foster care population did not experience a proportionate, concurrent decrease. Still, after three consecutive years of increased adoptions, data from the U.S. Department of Health and Human Services does exhibit a decline in the foster care population. From a historical high of 581,000 in 1999, “the foster care population dropped—for the first time since 1981—to 556,000 children in 2000.”[53] For the past 15 years, this number has stayed below 470,000, even falling below 400,000 in FY 2011, FY 2012, and FY 2021.[54]

Whether or not ASFA played a significant role in the declining foster care population in the five or so years after passage has only been assessed in individual cities and states. For instance, one study from 2005 assessed the trends in the foster care population in New York City from 1985 to 2002 and the reasons for differences between those trends and those at the national level.[55] Synthesizing data from the Voluntary Cooperative Information System (VCIS), the Child Welfare League of America (CWLA), and the Adoption and Foster Care Analysis and Reporting System (AFCARS), the author created a base model that regressed number of children in foster care, number of entries, and number of exits on the state and year fixed effects. In layman’s terms, a model was designed to identify independent variables that significantly impacted the number of children in foster care, the number of entries into foster care, and the number of exits out of foster care. Fixed effects of state and year were implemented to control for state-level differences and to offset factors affecting all states in the same timeframe, such as demographic trends.

The model was expanded to include population characteristics, such as demographic factors and per capita income, and incarceration and drug arrest rates, followed by the contribution of child welfare policies, including Title IV-E child welfare waivers and ASFA legislation. With ASFA included in the model, year effects became significant with regard to increases in foster care exits, implying that nationally, ASFA itself did influence foster care exits, which rose from under 200,000 to over 250,000 from 1997 to 2002. This was accompanied by a decline in the foster care population.

—

…nationally, ASFA itself did influence foster care exits, which rose from under 200,000 to over 250,000 from 1997 to 2002.

—

A six-state analysis from 2003 based on data recorded in AFCARS from 1997 to 2000, which spans the year of ASFA implementation and the three years after, reveals that while the foster care population declined nationally, individual state trends differed.[56] The study covered California, Illinois, Maine, Missouri, Vermont, and Washington. Among the outcomes assessed was rates of foster care entry. Considering numbers alone, while foster care rates increased in California, Missouri, and Vermont, they decreased in Illinois, Maine, and Washington over this period. Moreover, a study from the year before that focused on Illinois found that, controlling for the Department of Child and Family Services’ budget, state birth rate, number of indicated reports of maltreatment, and ratio of foster care case openings to case closings, the decline in the foster care population in the last years of the 1990s could be tied to some degree to ASFA.[57] In fact, the report claims that the foster care population in Illinois dropped by about 760 children in foster care per month due to the implementation of ASFA in the state.

Why the differences? Several hypotheses could be suggested. For one, the populations served by foster care vary widely across states.[58] State demographic differences impact which children enter care and how quickly they then achieve permanency. One example is that states with a “disproportionate number of older youth in foster care will likely have lower rates of TPR,” so the foster care population would not decrease as quickly as in states where more infants and young children are served.[59] Another explanation is that states that experienced greater decreases already had family-supportive social policies in motion. In Oregon, for instance, ASFA-like regulations existed prior to passage of ASFA and considerable work was already being undertaken to improve communication and cooperation among the substance treatment providers and the state’s child welfare and court systems.[60]

Rate of Adoption From Foster Care

What about the change in the number of children adopted from foster care? A main goal of ASFA was to expedite the process for children who were unlikely or unable to reunify with their birth families to be made eligible for adoption. Thus, whether the policy did, in fact, significantly influence the adoption process and rates has been an important question for researchers to assess. Evaluations of the change in the adoption rate have been conducted both nationwide and at a state level, usually relying on statistics provided by AFCARS and local administrative data.

Looking at raw data, we do see an increase in adoptions after passage of ASFA. In FY 2000, 48,680 foster youth were adopted, an increase of 57% from the 31,004 adoptions out of foster care recorded in FY 1997.[61] Moreover, nationwide from FY 1998 to FY 1999, 42 states increased the total number of children who were adopted out of their child welfare systems. However, this is not the full story.

A source of difficulty in parsing the impact of ASFA on this measurement is that the number of adoptions had already been on the increase prior to implementation. Moreover, it could take a “relatively long time to know whether the adoption process changes in response to specific policy or programmatic initiatives,” as noted in Chapin Hall’s Adoption Dynamics: An Update on the Impact of the Adoption and Safe Families Act,[62] and, after a sufficient amount of time for assessment has passed, it is likely that influences apart from the policy of interest will have affected the cohort members of interest. The combination of these factors makes it difficult to isolate the effect of ASFA alone.

One tool that researchers have used to interpret data through these difficulties is a particular class of survival models called proportional hazards models. The Cox regression model, the most popular for survival analysis,[63] estimates the effect that certain risk factors have on survival or the time it takes for a specified event, called the hazard, to occur.[64]

One multi-state study covering 13 foster care entry cohorts employed survival analysis to measure the total proportion of children adopted from annual entry cohorts, the length of time spent in care before discharge to adoption, and any contribution that ASFA had to increasing the “risk” of the hazard (adoption from foster care). Drawing on data from the Multistate Foster Care Data Archive,[65] the proportional hazards model in the study discovered that, for children admitted into foster care between 1997 and 2001, the hazard of adoption was more than twice that for the 1990 entry cohort. That is, “over twice as many children per unit of time [were] adopted as in 1990.”[66]

—

Evidence exists suggesting that, while one cannot reasonably attribute all of the change to ASFA, the policy does deserve some credit.

—

How much of this change is attributable to ASFA and how much is attributable to pre-existing or developing state policies and practices? Evidence exists suggesting that, while one cannot reasonably attribute all of the change to ASFA, the policy does deserve some credit. The same study assessing data from seven states found that, for children in pre-ASFA entry cohorts, “period-specific probability of adoption increased once the children still in care came under the influence of the law,” meaning that, relative to other cohorts at similar points in time, children who were in care before and through the implementation of ASFA were more likely to be adopted.[67]

Cox regression analysis conducted on state-specific data also supports the hypothesis that ASFA, separate from other policies and practices, has had a significant impact on the rate of adoption from foster care. A study focused on families affected by substance misuse in Oregon who had at least one child involved in the child welfare system between 1996 and 1998 (the pre-ASFA cohort) or between 1999 and 2001 (the post-ASFA cohort) observed faster permanency placements for individuals in the post-ASFA cohort, with the likelihood of reaching permanency 1.19 times higher.[68] Looking more closely at the permanency placements achieved, the probability of being made legally eligible for adoption increased by a factor of 2.2 for the post-ASFA group.

One subgroup who particularly benefited from the adoption-related provisions under ASFA were children who had special needs as recognized under Title IV-E.[69] A combination of more and/or better services, better record-keeping, and availability of federal funds through ASFA to support special needs adoptions has helped to promote the adoption of these children.[70],[71]

Rate of Reunification

Has ASFA negatively impacted reunification efforts? This is a common concern with ASFA and the last main consideration of this brief review. Critics worried that ASFA would prioritize adoption over reunification, creating an imbalanced, binary choice between the two,[72],[73] and that its set timelines were too short for many parents to improve their family circumstances so as to achieve reunification.

As with foster care population and adoption rate statistics, one cannot, from a glimpse at the raw number of reunifications achieved in a given time period, firmly conclude anything about the benefit or harm of ASFA on this facet of child welfare. Meeting such limitations, one well-designed study, conducted through the Office of the Assistant Secretary for Planning and Evaluation (ASPE) in the Department of Health and Human Services (DHHS), interprets the meaning behind the raw numbers using multivariate analyses and, as in studies on other ASFA outcomes, proportional hazard models.[74]

Analyzing data for nine states from the Multistate Foster Care Data Archive that captured children who entered foster care between 1990 and 1999 – seven years before and two years after the passage of ASFA – the study found that, relative to the 1990 entry cohort, the relative risk of exit stayed constant over the decade. That is, children who entered foster care after 1990 did not, in general, exit foster care significantly more quickly or more slowly than children who entered foster care in 1990. However, looking at exit types individually, the data revealed that, relative to the 1990 entry cohort, later cohorts exited foster care to adoption more quickly and to reunification more slowly. The author emphasizes that this result does not imply that reunification is less likely since ASFA, but that, at least in the time frame studied, there has been a slowdown. It is also important to note that this trend started before implementation of the new legislation.[75] Another study of nine states through DHHS showed that the proportion of foster care exits due to reunification had been declining since at least 1990.[76]

This research covered several years before ASFA and a couple of years after. It may be helpful to look at research that has a little more distance from ASFA, allowing more time for states to adapt to the new environment. One study based in California examined the reunification and foster care re-entry rates of children who entered foster care for the first time in calendar years 1998, 1999, 2000, 2001, or 2002.[77] In this state, the trend over these years was an increase in the proportion of children who were reunified and a decrease in the proportion of children who re-entered foster care after having been reunified. Moreover, the odds of re-entering foster care within 12 months of reunification were on the decline from 1998 to 2002. A review of longitudinal data from 12 states also reported a downward trend in re-entry rate since about 1990 (to 2000).[78]

Legacy and Future of ASFA

Based on the existing research, the Adoption and Safe Families Act has achieved the goals of reducing timeframes for youth in foster care and increasing the rate of adoption. Despite the clear evidence of ASFA’s positive impact on children, there are those who contend the legacy of ASFA is mixed. Though it garnered bipartisan support in Congress at the time of passage, public reception of ASFA has not been uniform. Criticisms include that ASFA incentivized adoption over reunification,[79] disproportionately affected Black families,[80],[81] severed family connections along artificial timelines,[82] and failed to lower the number of emancipated youth (youth who age out of care).[83] Although these points are debated by child welfare researchers and advocates, recent legislation has sought to mitigate these issues.

—

Based on the existing research, the Adoption and Safe Families Act has achieved the goals of reducing timeframes for youth in foster care and increasing the rate of adoption.

—

The Family First Prevention Services Act of 2018 (FFPSA) represents one effort to redress these concerns; one of its primary aims is the provision of services and other assistance for families so that children can safely remain at home.[84] Since most states delayed implementation of FFPSA,[85] it is too early to tell what impact FFPSA, operating alongside ASFA, is having on reunification and adoption in the United States.

As the most significant federal legislation on foster care and adoption in the United States at the turn of the century, ASFA was a notable and influential policy that prompted child welfare departments across the country to create and recalibrate services and processes so as to facilitate reunification within a reasonable timeframe and, where reunification was not achieved, to facilitate placement of the child in stable kinship care or an adoptive home. It continues to impact the child welfare system over 25 years on, as local and state governments and social service departments shape laws and policies that align with ASFA and meet the unique needs of the populations they serve.