New Strategies Long Overdue on Measuring Child Welfare Risk

As The Chronicle of Social Change has been reporting over the past two years, various jurisdictions have been exploring new tools to focus the attention of child welfare systems on the children most at risk of subsequent abuse or neglect. The mainstream media has begun to notice, as demonstrated by CNBC’s recent report on Los Angeles’ contract with software company SAS to develop such a tool for its child welfare system.

These new approaches generally rely on predictive analytics, which means using patterns in data to predict future outcomes. Despite the recent media coverage, there is still some confusion about what is meant by this term, how it differs from current approaches like Structured Decision Making (SDM), and the distinctions between the various new approaches.

To understand these new approaches, it is important to understand how child protective services (CPS) works now. I can describe the process in the District of Columbia, where I once served as a caseworker.

CPS workers in the District use checklists to interview children, parents, teachers, and others about an allegation of abuse and neglect. They generally rely on the verbal answers to these questions, although they do receive access to data from schools and public assistance agencies.

The District of Columbia, like jurisdictions in over 20 states, uses an SDM tool to help social workers decide how to proceed at the conclusion of an investigation. Social workers fill out checklists on the computer. SDM assigns points to each risk factor, such as “primary caregiver has historic or current drug or alcohol problem.”

Based on the worker’s checkoffs, the software spits out a recommendation to remove the child or keep her at home.

SDM has at least two major flaws. First, it can be manipulated to recommend the action that the social worker wants to take. Second, it does not obtain any new data but simply uses what the social worker plugs in. That’s why in my experience in D.C., SDM was treated as a meaningless form to be filled out, not as a tool for making decisions.

A new set of predictive tools, known as predictive analytics, is being developed for governments in Los Angeles, Allegheny County, Pennsylvania, and New Zealand. These tools are designed to produce a numeric risk score for each child assessed. A high risk score would target the child and family for special attention, which might include intensive services and monitoring.

Unlike SDM, predictive analytics tools access data directly from other systems rather than relying on self-report from parents and caregivers. These systems might include mental health, substance abuse treatment and criminal justice, among others. Clearly, the information obtained this way would be more accurate than simply asking the parent or caregiver, who has to rely on memory and might have an incentive to withhold information.

Los Angeles County’s Department of Child and Family Services has contracted with software giant SAS to develop a risk assessment tool called AURA. As described by The Chronicle, SAS identified child deaths, near fatalities and severe injuries (called “AURA events”) in 2011-2012 among children referred to CPS within six months of these events.

Programmers then developed an algorithm that calculated the risk of an AURA event in the six months following a referral based on the characteristics of the children and their families. The predictive power of the algorithm was tested against child abuse reports in 2013 (data that was not used to develop the algorithm.)

AURA proved to be a powerful tool for predicting deaths and serious injuries to children. Among those cases that had at least one referral before the current one, the 1 percent of referrals with the highest AURA scores experienced 57 percent of the AURA events. The 20 percent of referrals with the highest risk accounted for 83 percent of total AURA events.

Florida has chosen to use a different approach. It is using a tool called Rapid Safety Feedback(RSF), developed by a nonprofit called Eckerd Kids for Hillsborough County, which had experienced nine deaths among children in open cases receiving services at home over three years.

Rather than a risk assessment algorithm, RSF is a quality assurance tool that is used to target a set of high-risk cases for special treatment and review. RSF targets cases with characteristics associated with a high probability of serious injury or death to a child.

In Florida, RSF is used only for families with in-home cases. Cases designated high-risk are reviewed quarterly to make sure workers are using specific practices that have been shown to increase child safety.

According to Eckerd, there have been no maltreatment-related deaths of children in open cases in Hillsborough County since implementation of RSF in January 2013. Several states are working with Eckerd to adapt and implement RSF in their child welfare investigations.

The new tools discussed above are in their infancy in child welfare. It is important for policymakers and advocates to understand the nature of the tools that are available and the differences between them before making decisions about which if any to support.

This column was published in the Chronicle of Social Change on February 3, 2016.

Families First Act Disregards Foster Care Shortage, Need for Some Group Homes

Congress appears to be on the right track with child welfare reform, as evidenced by a summary of the new Families First Act, which may soon be marked up by the Senate Finance Committee. The centerpiece of the legislation is the expansion of Title IV-E to provide preventive services to assist children in danger of being placed in foster care.

With the passage of this legislation, federal funding will support keeping children at home as much as it does foster care. It is a change that almost everyone seems to support.

The other major part of the legislation, however, is not without controversy. That is the part that seeks to eliminate most federal support for foster care group homes.

The legislation eliminates funding for placements that are not foster family homes after two weeks, with three exceptions: Qualified Residential Treatment Programs (QRTP’s), facilities for pregnant and parenting teens, and independent living arrangements.

The definition of QRTP has not been released, but it appears that Congress has in mind the types of arrangements that are currently known as residential treatment programs, which are large institutions where clients attend school and receive mental health treatment.

I have several concerns with this provision.

First, the bill fails to recognize the need for a continuum of care depending on the child’s needs. Such a continuum should include family foster care, therapeutic foster care, therapeutic community group homes and residential treatment centers. A child who leaves residential treatment is not necessarily ready for a foster family, and a child who can’t function in a family may not need a residential treatment program. What’s missing from the options provided by the new bill are therapeutic group homes, such as those operated by Boys Town, using the evidence-based Teaching Family model.

Secondly, the bill contains no provisions to increase the supply of high-quality foster care to meet the needs of all the children who will need to be accommodated when group homes close down. Today, there is a critical foster care shortage around the country, with children staying in offices, hotels and emergency shelters because foster homes are not available to them.

To make matters worse, many existing foster homes provide care that is neglectful or even abusive. As a former social worker with District of Columbia foster children, I have writtenabout neglectful foster parents. There was the foster parent who hadn’t been to her foster child’s school in more than a year, and refused to pick her up even when she was vomiting. There was the foster parent who refused to go to a meeting at her foster child’s school, saying, “If I cared I would go, but I don’t care.”

I asked my agency to close these homes, but the request was not granted. With the shortage of foster homes, agencies are reluctant to close homes that are anything short of abusive. And even in the case of abusive homes, we have all heard the stories of children who have been killed or injured in foster care.

Sometimes, as in the case of two-year-old Laila Marie Daniel in Georgia, it turns out that agency staff disregarded multiple reports of trouble in the home. I’m sure that the critical shortage of foster parents is part of the explanation.

Abusive and neglectful foster homes need to be closed, but they must be replaced and added to with a new source of high quality foster homes. How can this be done? The only way is to pay foster parents enough so that one foster parent in each home can stay home with the children, thus attracting a completely new source of foster parents.

In an earlier column, I wrote about the SOS Children’s Villages in Illinois and Florida. They not only pay foster parents a salary but provide houses large enough for six children. The provision of housing is particularly important in order to attract foster parents to cities with high housing costs, where many foster kids live.

Paying foster parents a full-time salary is an expensive proposition. But this expense can be offset by recruiting foster couples (either married couples or two single people living together as house parents) to care for five or six children. By locating them in communities like those provided by SOS Children’s Villages or the Mockingbird Society, foster parents can be empowered to help each other, and services to the children can be provided on site.

It would be irresponsible for Congress to pass the Families First Act without provisions for a therapeutic option between foster homes and residential treatment and for increasing the supply of quality foster care. The Senate Finance Committee should modify its bill by adding a provision for short-term therapeutic group homes and by providing new funding and incentives to encourage states to improve the quantity and quality of their foster homes.

This column was published in the Chronicle of Social Change on January 27, 2016.

Evaluation of Foster Care Agency Should Look Beyond ‘Performance Scorecard’

Accountability systems can be harmful when service providers are evaluated based on outcomes they cannot control. It is common knowledge that schools in low-income neighborhoods are often blamed for the poor performance of their students. This problem occurs as well in child welfare, where a public or private agency can be rated as poorly performing because it works with a harder-to-serve population.

On January 6, 2016, The Chronicle published my column about a critical foster home shortage in the District of Columbia. The court monitor for the child welfare system blamed the crisis in large part on the decision by D.C.’s Child and Family Services Agency (CFSA) to end a contract with a private agency (Foundations for Home and Community) that had cared for many of the oldest and most troubled youth in the system.

In my last column, I discussed the impact of Foundations’ closure on the overall number of foster homes in the District. But the closure had another serious impact—on the children who lost their homes when their foster parents quit the system. I did not know this number until CFSA sent me a long-delayed response to a Freedom of Information Act (FOIA) request I made a year ago and repeated last December.

The numbers I finally obtained were startling. The entire transition process took seven months, from November through May 2015. During that period, some young people aged out of care, attained permanency, or were able to stay in the same home because their foster parents transferred to another private agency. But many foster parents chose not to transfer, often because Foundations used private funds to pay them more than other private agencies in the District would have.

When the smoke cleared, CFSA reported that 40 youths who were previously in Foundations foster homes had moved to new placements during the process of closing the agency. That is a total of 40 young people who had to leave their homes because CFSA decided to shut down their foster care provider. While we don’t know if any of those young people might have moved anyway during the seven-month transition process, it seems fair to assume that most or all of them had to move because of the closure of Foundations.

The new document also sheds light on the reasons behind the decision to close Foundations. It explains that the decrease in CFSA’s caseload due to its emphasis on keeping families together resulted in a decreased utilization of available foster care “beds.” In order to “right-size” the system, the agency decided to close the two “lowest-performing providers.”

I consulted the “Performance Scorecards” that CFSA uses to rate its private providers. Indeed, Foundations did have the worst average score for the first three quarters of FY 2014. (Ratings for the last quarter were not available at the time the decision was made to close the agency.)

But as a veteran of two private agencies, I know that these agencies are rated on things over which they have no control. Most importantly, two indicators of permanency (percentage of youth achieving permanency within the past 12 months and those achieving ‘timely’ permanence’) account for 40 percent of an agency’s total score. Foundations did particularly badly on these ratings.

Yet, agencies that work with older kids who have been in foster care for a long time have little control over permanency outcomes for their clients. Foundations took in the hardest-to-serve youth. Many of these youths had been in the system for years, and all of their family connections had been exhausted. Their age and behavioral problems made them difficult to place. Expecting the agency to miraculously find permanent homes for them was not realistic.

When they made the decision to close Foundations, CFSA leaders may not have realized that it would result in the loss of homes for 40 foster youth and a citywide shortage of foster homes. But as I described in my last column, CFSA knew there was a problem as early as January 2015, two months into the seven-month transition.

Once agency leaders knew that many foster parents from Foundations were leaving the system, they had plenty of time to reverse their decision. Other organizations, like the agency charged with monitoring CFSA for the court, the City Council, and advocacy groups also had advance warnings but did not speak out.

This tale has national implications. First, child welfare professionals should beware of benchmarks that penalize agencies for factors over which they have no control. Second, legislators, advocates, and reporters need to be much more aware of what child welfare executives are doing. They should not rely on rosy reports from leaders who fear the career implications of admitting mistakes.

This column was published by the Chronicle of Social Change on January 20, 2016.

Poor Planning and Lack of Oversight Lead to Foster Care Crisis in District of Columbia

The latest report from the Center for the Study of Social Policy (CSSP), the agency monitoring the District of Columbia’s Child and Family Services Agency (CFSA), contained a startling revelation. For the first time in 15 years, children in the District of Columbia have spent the night in agency offices (and even motels) because there were no foster homes available.

Foster homes are scarce around the nation, but the crisis in the District was a direct outcome of decisions made by the child welfare agency. Like many other jurisdictions, CFSA has historically contracted with private agencies to provide foster homes to District youth. As a social worker in one of these agencies, I watched this placement crisis unfold.

Over the past few years, CFSA has been been ending contracts with one or more private foster care agencies every year. Late last year, CFSA announced it was ending its contract with two agencies that served mostly older youth with challenges. One of them, Foundations for Home and Community, was known for taking almost any youth, no matter how difficult their behaviors.

There is a reason Foundations was able to take the hardest-to-serve youth. Using private funds, the agency paid its foster parents considerably more to care for the most difficult-to-serve youth. We don’t want to think about foster parents as mercenary, and most of them are not. But working with hard-to-serve youth should be treated as a profession, as I discussed in a previous column. It requires time and it is often inconsistent with having a full-time job.

When the contract’s end was announced, Foundations’ foster parents were told that they would have to transfer to other agencies if they wanted to continue fostering CFSA youth. I began to hear that these foster parents were refusing to transfer to other agencies with lower payments. Instead, they were staying with Foundations and becoming foster parents to Maryland youth.

I heard that CFSA was pressuring private agencies to find homes for the young people whose foster parents were leaving the system, but most homes were full or not able to take such challenging youth.

When I left my position last January, I tried to spread the word about the impending crisis. CFSA had not issued a press release about the closures of the two agencies, and there had been no media coverage. I testified before the city council oversight hearing on CFSA in February 2015, but there was little interest and no follow-up from the legislators and advocates who were present.

In the spring, I began to hear about kids sleeping in offices but nothing appeared in the press until CSSP issued its report, revealing that 11 children stayed overnight at the CFSA office and four children stayed in hotel rooms while awaiting an appropriate licensed placement between April and June of 2015. According to the report, the placement problems continued in the second half of the year. After CSSP issued its report, an article appeared in City Paper  but there was no reaction from other media or advocates.

CSSP’s report blamed the placement crisis on CFSA’s decision to end contracts with the two private agencies, which “resulted in a shortage of foster care placements available to serve specific populations, including older youth and those with mental or behavioral health challenges.” At about the same time, entries to foster care increased.

Even without a placement crisis, removing young people from stable placements would be a serious concern. I tried to obtain the number of youth who were displaced but was told that the Freedom of Information Act (FOIA) does not require an agency to answer questions or create documents. So we don’t know how many of our most traumatized youth suffered yet another loss, this time at the hands of the agency tasked with protecting them.

This whole episode raises several questions. First, how was the decision to close Foundations made? In response to another FOIA request, CFSA again refused to provide a narrative response. Instead it sent a report showing the percent of contract slots that were actually utilized by private agencies. CFSA gave no reason why this should be considered as a principal measure in determining which agency should be closed as CFSA does not pay for unused slots. In any case, Foundations did not appear to have the lowest overall utilization rate.

Another question is why CFSA did not reverse its decision once it saw that it was losing foster parents. The lack of transparency and the absence of concern among legislators, the press and the public meant that there was nobody to question the decision or press for it to be revoked. Our children deserve better.

This column was published in the Chronicle of Social Change on January 6, 2015.

Family Planning Can Help Curb Maltreatment Mortality

In my previous column, I quoted Sarah Brown, founder of the National Campaign to Prevent Teen and Unplanned Pregnancy, who said that groups focusing on child and family well-being rarely propose interventions that begin before conception.

Nowhere is this more true than in the field of child welfare. So it is not surprising that the draft recommendations of the Committee to Eliminate Child Abuse and Neglect Fatalities (CECANF) do not include family planning.

But anyone who has worked with children in foster care knows that more of these youth come from larger, closely spaced sibling groups than one would expect in the general population. Indeed, research shows a strong relationship between larger families and child maltreatment.

A study of the 2002 birth cohort in California showed that babies born to mothers who already had two or more children were more than twice as likely to have CPS contact than firstborn babies.

Using earlier data from California, Barbara Needell and Richard Barth found that 59 percent of infants entering foster care between 1989 and 1994 were the third or later child born to their mother, compared to only 29 percent of the general population of infants.

Federal data from the Fourth National Incidence Study of Child Abuse and Neglect shows that one in 48 children in households with four or more children were likely to suffer abuse or neglect. This is double the rate of children in households with two children, which had the lowest rate.

Unfortunately, we don’t know what proportion of child fatalities come from large families. But at least one expert has noticed a pattern. Angela Diaz, who directs New York’s Mount Sinai Adolescent Health Center, testified that during her tenure on a panel that reviewed child fatalities in New York City, the mother of the dead child often had had a child in adolescence, followed by more children in short order.

The correlation between larger families and child maltreatment is not surprising. Parents of larger families undergo more stress due to both the difficulty of supervising more children and the financial burden.

Financial strain can also lead to neglect in the form of lack of supervision when the parent must work, inadequate housing, etc. Moreover, both child maltreatment and larger families may be related to underlying social factors such as poverty and low education levels.

Of course, many people raise large families with no difficulties, and many adults from large families treasure their upbringing and their strong support system. In the past, larger families were more practical for all income levels.

Today, I’d submit that a very large family is financially risky to most adults without a serious cash reserve or income in the top percentile. For parents already involved with child welfare, especially those struggling to make ends meet, it is hard to dispute the benefits of postponing further children.

In the last few years, advances in contraception have led to the availability of Long Acting Reversible Contraceptives (LARCs). But the availability of LARCs and knowledge of their safety and efficacy have not yet caught up.

As I recommended in an earlier column, system-involved mothers could be offered the opportunity to get a LARC immediately. Judges, attorneys and social workers could provide information about LARCs and explain the importance of avoiding further births as a mother tries to regain her children and get the agency out of her life.

CECANF should add to its recommendations that future data collection about child deaths include family size. But the correlation between family size and child maltreatment is compelling enough for immediate action. CECANF should suggest that states explore the possibility of integrating pregnancy prevention with child welfare services.

On a policy level, it is time to discuss ways to limit childbearing by women who lack the financial means to handle more kids. A grant program might even be created in order to encourage states to experiment with different approaches.

Such proposals are difficult to bring up in America, thanks in part to our country’s shameful history of attempting to restrict childbearing by women of color through means including forced sterilization and the promotion of birth control methods like Norplant. But the welfare of children should take precedence over the reluctance to offend adults.

This column was published in the Chronicle of Social Change on December 16, 2015.

Teen Pregnancy Prevention: An Overlooked Weapon Against Child Maltreatment Fatalities

The Commission for the Elimination of Child Abuse and Neglect Fatalities (CECANF) has developed a long list of draft recommendations. One overarching recommendation is to “ensure access to high-quality prevention and intervention services for children and families at risk.”

Many prevention services are mentioned, but one was left out in my opinion: services to prevent teen and unplanned pregnancies.

It is not surprising that pregnancy prevention was not often recommended to CECANF as a strategy to prevent child abuse and neglect fatalities, and that this recommendation did not make its way into CECANF’s draft.

As National Campaign to Prevent Teen Pregnancy co-founder Sarah Brown recently pointed out, groups that focus on child and family well-being rarely propose interventions that begin before conception of a child. CECANF could begin to rectify this omission by including teen pregnancy prevention in its recommendations for reducing child abuse and neglect fatalities.

In her testimony before CECANF, Angela Diaz, director of New York’s Mount Sinai Adolescent Health Center, discussed the connection between teenage parenthood and child maltreatment fatalities. In serving for many years on a child fatality review panel, she noticed that in many of these cases, the mother began childbearing in adolescence, and had more closely spaced children thereafter.

Dr. Diaz cited a national study of deaths of infants born between 1983 and 1991, which showed that “childbearing at an early age was strongly associated with infant homicide, particularly if the mother had given birth previously.”

A second or subsequent infant born to a mother younger than 17 years old was 11 times more likely to be a homicide victim than the first child of a mother 25 or older. A second or subsequent infant born to a mother age 17 to 19 was over nine times more likely to be a homicide victim.

Based on her observations and the research, Dr. Diaz recommended ensuring that adolescents have access to comprehensive sexuality education and reliable family planning methods. I hope the Commission will add this to its recommendations.

It is true that teen pregnancy has been decreasing rapidly. But there were still 249,067 birthsto teenage girls in the United States in 2014, or 24 births for every 1,000 girls. Estimates from 2013 data show that 11 percent of adolescent girls in the United States will give birth by their 20th  birthday.

And rates are still higher among Blacks and Hispanics, girls in poor neighborhoods, and teenagers in foster care. Nearly one third of teenage girls in foster care have at least one child.

Most teens in foster care have been abused or neglected, which makes them statistically more likely to abuse or neglect their children. A large study in California found that 40 percent of children born to teen mothers involved in the child welfare system will be reported for child abuse by age five.

We need to know what proportion of child fatalities involve children who were born when their mothers were teens, as well as those born to mothers who began having kids as teenagers. CECANF should recommend that Congress require the collection of these data.

Even without knowing the proportion of child maltreatment deaths occurring to children of teen mothers, we already know that teen motherhood is a risk factor for child abuse and neglect. CECANF should recommend increased emphasis on teen pregnancy prevention, especially for young women in high poverty areas and those in foster care.

The Commission should recommend that all teens, especially those at higher risk of pregnancy, have access to contraceptive methods and education. Clinics in low income areas and those serving youth in foster care and juvenile justice should provide the full array of contraceptive options including the long-lasting methods that are most effective, along with education and counseling.

Special attention should be devoted to preventing a second birth to a teenage mother by ensuring that she is provided with a contraceptive method at the time of the first birth. The federal Teen Pregnancy Prevention Program, which has been under attack in Congress, should be fully funded or expanded.

 This column was published in the Chronicle of Social Change on December 9, 2015.

Birth Match, Tool to Prevent Abuse and Neglect, Slow to Spread in U.S.

The Commission to Eliminate Child Abuse and Neglect Fatalities (CECANF) has heard from experts and stakeholders around the country and has prepared a draft of its recommendations. Among them is the suggestion that states expand a practice called “birth match,” which can identify at birth children who are at risk of child abuse and neglect.

This recommendation should be highlighted and strengthened.

Richard Barth, a highly regarded child welfare scholar, discussed Maryland’s birth match law in testimony during a CECANF hearing in July of 2014. This law was adopted after several children were killed by parents whose parental rights had been terminated for previous children.

Infants born to parents with a prior termination of parental rights (TPR) are not an insignificant group. Barth found that they made up over 10 percent of entries into foster care in Maryland in 2013.

Maryland’s law requires that birth records be matched against a list of parents who had their parental rights terminated within the last five years due to abuse or neglect. The families thus identified receive a visit from a social worker to assess for safety and risk to the infant.

States have a choice in how to intervene with the families they identify through birth match. In Maryland, these families receive a visit from a social worker to assess the family. If the parent refuses the visit, a case can be opened “only if abuse or neglect is suspected.”

Michigan is more aggressive in dealing with high-risk families. Identification of these families results in the opening of a case unless there is a waiver. That seems a more effective way to ensure a child’s safety than allowing families to opt out of the screening.

Identifying only parents with prior TPRs is a very conservative strategy that does not even capture all parents who lost a child permanently due to abuse or neglect. To identify all of those parents, a state would have to include parents who lost a child to guardianship as well as adoption.

To target a broader swath of parents at risk of harming a child, a state could identify all parents who had a prior finding of child abuse or neglect or a child placed in foster care. It could also target teenage mothers who themselves were victims of child abuse and neglect. One study found that 40 percent of children of California teen mothers who were found to be the victims of abuse or neglect as children will be reported for child maltreatment by age five.

Birth records could also be matched with data other than child welfare administrative records. A groundbreaking study of all 2002 births in California shows that “objective data collected at birth can be used to identify those children in a given birth cohort who are at greatest risk of future CPS contact.”

Among the most important risk factors were a mother’s low education level, a mother’s young age, three or more children in the family, and for U.S. born mothers, Medicaid coverage of the birth.

Despite the promise of this approach, few systems have embraced it. Barth told CECANF that “virtually all states have the option to share birth records with child welfare agencies,” but very few states do; to his knowledge, the only states that do are Minnesota, Maryland and Michigan.

According to Barth, the main obstacle to implementing birth match in other states is the reluctance to trigger CPS involvement with families that may not be maltreating their children.

Clearly, it would be distressing for a family to be contacted by CPS when they have not done anything wrong. And by no means am I suggesting that all, or even most, of the parents identified through birth match are incapable of caring for a child.

However, the opportunity to save a child from serious abuse or neglect should take precedence over the inconvenience to families of undergoing an assessment.

We need more research on birth match programs to identify the outcomes of intervening early in families identified as high-risk. How many families are identified and what is the result? Do child fatalities decrease after the adoption of such a policy?

CECANF’s recommendation that states “expand birth match programs such as those operated in Michigan” should be strengthened and elevated to the federal level. The commission might recommend that the federal government establish a pilot program providing grants to states that adopt and evaluate birth match programs.

Information from this pilot could be used to develop a federal requirement for all the states to match birth and child welfare records.

This column was published in the Chronicle of Social Change on November 30, 2015.