A potentially lifesaving algorithm in Allegheny County, PA

broken arm
Image: New York Times

In August 2016, Allegheny County Pennsylvania (which includes Pittsburgh)  became the first US jurisdiction to use a predictive algorithm to screen every call to the child abuse and neglect hotline. In a brilliant article for the New York Times Magazine,  science writer Dan Hurley clearly explains how the tool works and how it changes current practice. Hurley’s account suggests that Allegheny’s experience is a hopeful one for the county and for children nationwide.

Hurley introduces the Allegheny Family Screening Tool, an algorithm developed by leading child welfare researchers in concert with DHS policymakers. To develop the algorithm, the authors analyzed all referrals made to the county child abuse hotline between April 10 and April 2014. For each referral, the authors combined child welfare data with data from the county jail, juvenile probation, public welfare, and behavioral health programs  to develop a model predicting the risk of an adverse outcome for each child named on each referral. (A more technical description is provided by the authors here.) The end product was an algorithm that calculates a risk score between 1 and 20 for each child included in a referral.

The policymakers and developers chose to use the algorithm to supplement, not supplant, the clinical judgment of hotline workers. Only if the score exceeds a certain threshhold does it trigger a mandatory investigation; below that level the risk score it provides another piece of data to help the hotline worker decide whether to assign the case for investigation.

Among the most important takeaways from Hurley’s article are the following:

  1. Before the development of the new algorithm, Allegheny County had experienced a series of tragedies in which children died after maltreatment reports had been made to the hotline but screened out. The problem was not incompetence or poor training. Hotline workers simply cannot within the 30 minutes to one hour allowed for decision making investigate all the historical data on all family members from numerous agencies with which they may have had contact.
  2. Evaluation data shared with the reporter show that implementation of the Allegheny County Screening Tool resulted in more high-risk cases being screened in and more low-risk cases being screened out. Hurley provides a real case example. A teacher reported that a three-year-old child witnessed a man dying of an overdose in her home. Department records showed numerous reports to the hotline dating back to 2008 about this family, including allegations of sexual abuse, domestic violence, parental substance abuse, inadequate food, physical care, hygiene and medical neglect. Nevertheless, the hotline worker was poised to screen out the case as low risk. The tool, however, calculated a risk rating of 19 out of 20, causing an investigator to go out to the home. Eventually, the mother was found to be unable to care for the children due to her continuing drug abuse, and they were placed with family members, where they are doing well.
  3. County officials were astute in awarding the contract to develop a predictive algorithm. Several other jurisdictions have gone with private companies such as Eckerd Connects and its for-profit partner Mindshare, which has a predictive analytics tool called Rapid Safety Feedback (RSF). The details of RSF are closely held by the companies, and the state of Illinois recently terminated its contract  because the owners refused to share its details, even after the algorithm failed to flag some children who later died. The Allegheny Family Screening Tool is owned by the county. Its workings are public and have been published in academic journals. Moreover, its developers, Emily Putnam-Hornstein and Rhema Vaithianathan are acknowledged as the worldwide leaders in their field, with extensive publications and experience in doing similar work.
  4. County officials were also astute in developing and rolling out their model. They held public meetings before implementing the tool, giving advocates a chance to interact with the researchers and policymakers. Choosing to use the tool at the hotline stage rather than a later step such as investigation made it less threatening as the tool is not being used as input on whether to remove the child, simply whether to investigate. In addition, the county commissioned an ethics investigation by two experts before implementing the tool. The reviewers concluded that not only was the tool ethical but that it might be unethical to fail to implement it. The concluded that “It is hard to conceive of an ethical argument against use of the most accurate predictive instrument,”
  5. Many opponents of predictive analytics argue that it institutionalizes racial bias by incorporating data that is itself biased. Supporters have argued that predictive algorithms reduce bias by adding objective algorithms to subjective worker judgments. Preliminary data from Pittsburgh supports the proponents, suggesting that the algorithm has resulted in more equal treatment of black and white families.
  6. Other jurisdictions are already emulating Allegheny County. Douglas County, Colorado has already commissioned Putnam-Hornstein and Vaithianathan to develop an algorithm and California has contracted with them for preliminary statewide work.

Given the Allegheny County algorithm’s promising results, one cannot help wondering whether a similar algorithm should be used at later stages of a case as well. A similar tool could be very useful in aiding investigators in making a decision about the next step in a case. Such a proposal would of course trigger an outcry if used to decide whether to remove a child from home. But like the Allegheny County screening tool, such an algorithm can be used to supplement clinical judgment rather than replace it. Policymakers need not set any level that would trigger a mandatory removal. However, they could set a risk level that requires opening a case, be it out-of-home or in-home. Many children in many states have died when agencies failed to open a case despite high risk scores on existing instruments. Algorithms can also be used to monitor ongoing in-home cases, as Rapid Safety Feedback has demonstrated. Perhaps if and when predictive algorithms are proven to be effective at protecting children they will be integrated into multiple stages and decision points, like the actuarial risk assessments that many states use today.

Identifying the children most at risk of harm by their parents or guardians has been one of the knottiest problems of child welfare. Allegheny County’s experience, as portrayed by Dan Hurley’s excellent article, provides hope that emerging predictive analytics techniques can improve government’s ability to identify these most vulnerable children and keep them safe.

What’s Behind the Drop in New York City Foster Care Numbers? More than the Commissioner Chooses to Say


On December 12, 2017, the New York Daily News published an exclusive story of a dramatic drop in foster care numbers in New York City. Only 8,966 children were in foster care in Fiscal Year 2017, down  from 2016’s total of 9,926. Moreover, New York City’s foster care rolls have been dropping over the past four years as the nationwide caseload increased.

Administration on Children’s Services (ACS) Commissioner David Hansell awarded his agency most of the credit for the decrease, telling the Daily News that “primarily it has to do with keeping families together whenever we can.” As he told the reporter, instead of immediately removing a child deemed to be at risk, ACS seeks to provide services to the family to ameliorate that risk without removing the child.

Hansell’s remarks to the Daily News can be questioned on two grounds. First, there is evidence that agency policy is not the only factor behind the caseload decline. Second, a simple decline in foster care caseloads is not evidence of progress unless we know the agency is not leaving children to languish in unsafe homes

A closer look at the numbers (contained in a Foster Care Strategic Blueprint Status Report issued by ACS on December 18), compared with population trends in New York City, reveals that more is going on than the Commissioner chose to discuss.

In New York City, according to Census data, the number of children in poverty fell from 553,499 in 2012 to 471,190 in 2016, a 15% decrease. During fiscal years 2012 to 2016 (a period that is off by 6 months from the annual data) New York’s foster care caseload fell from 13,820 to 9,926, a decrease of 28%. (See the table below for the numbers.)

It’s a well-known fact, and well-documented by research, that poor children are much more likely to be placed in foster care than their peers. There are  many reasons why this might be the case, and some critics allege that some children are actually placed in foster care simply because they are poor.

Based on the decline in children in poverty, we could have expected roughly a 15% decline in the foster care rolls. The percentage drop in New York City’s foster care caseload was almost twice that, so agency policy probably did contribute to the foster care decline. But based on the percentages, demographic change may have been equally important.

Hansell did admit that there were “a lot of reasons” for New York City’s caseload to decline while the national caseload went up. But he did not choose to mention any of them.

We can’t be sure of the reasons for the decline in child poverty in New York. However, we do know of the influx of well-to-do people into many previously poor New York neighborhoods commonly described as “gentrification,” often driving poor people out of those neighborhoods.

A similar pattern may be observed in other cities experiencing rapid gentrification. For example, the number of children in poverty in San Francisco dropped from 16,000 in 2012 to 11,000 in 2016, according to Census data. And the number of children in foster care dropped from 1,073 in October 2012 to 811 in 2017 at the same time as the State’s caseload increased, according to the California Child Welfare Indicators Project.

Unfortunately, foster care caseload data is not easily available for most other cities, because the current Administration has instructed the federally funded center that houses large child welfare datasets to stop giving out this data to citizens after 20 years of doing so. However, it is highly plausible that other cities experiencing similar demographic changes also saw significant  drops in foster care

There is another problem with Hansell’s remarks. A decline in foster care numbers is not in itself a reason for celebration. We must remember that the purpose of foster care is to protect children. As an important issue brief from a California child advocacy coalition argues, states which have been cutting their caseloads for years may reach a “bottom’ below which further caseload reduction is not feasible without compromising child safety. ”

New York’s foster care caseload has been dropping since 2007. To the extent that these drops are due to policy changes, the city may reach a point where it is not safe to continue in the same direction.

Hansell cheers about “fewer children removed, fewer families separated and much less trauma experienced by children.” But what about the children who are being traumatized when they are left with abusive or neglectful parents at home?

Hansell admitted that there are some cases where the risks of leaving a child in a home are too high, such as the case of Zymere Perkins, who died at the hands of his mother’s boyfriend after ACS missed numerous warning signs.

We must remember that the purpose of foster care and the child welfare system is not to reduce foster care caseloads. It is to protect children. Its success should be evaluated accordingly.


Children in Poverty and Children in Foster Care, New York City, 2012-2016

Year 2012 2013 2014 2015 2016
children in poverty 553,499 522,992 523,538 508,503 471,190
children in foster care 13,820 12,958 11,750 11,098 9,926


Children in Poverty: US Census Bureau, American Community Survey, http://factfinder.census.gov

Children in foster care: Administration on Children’s Services, Foster Care Strategic Blueprint Status Report, https://www1.nyc.gov/assets/acs/pdf/about/2017/BluePrint.pdf









Predictive analytics, machine learning, and child welfare risk assessment: questions remain about Broward study


On November 30, a major child welfare publication reported on a new study, published in the respected journal Children and Youth Services Review, that tested Broward County, Florida’s child welfare decision-making model against a model that was derived using the new techniques of data mining and supervised machine learning. The researchers concluded that 40% of cases that were referred to court for either foster care placement or intensive services could have been handled “with less intrusive options.” A close reading of this opaquely written paper, as well as conversations with two of the authors, Ira Schwartz and Peter York, reveal a pioneering effort at applying emerging data science techniques to develop a “prescriptive analytics” model that recommends the appropriate services for each child. This research is innovative and exciting but this first attempt at deriving such a prescriptive model for child welfare has serious flaws. These very preliminary results should initiate a conversation but should not be used to support policy recommendations. 

The authors began with a large database of 78,394 children with their complete case histories between 2010 and 2015. They merged datasets from the Broward County Sheriff’s Office,  ChildNet (the local agency contracted to provide foster care and in-home services) and the Children’s Services Council (CLC), which represents community based agencies serving lower-risk cases. The authors primarily used only one year of data on each child after they were discharged from the system. Children without a full year of data were not included. So the authors had a large selection of hotline, investigative, and service data for the children in their database as well as information on whether they experienced another referral within a year. 

In a nutshell, the authors applied machine learning to build a model “based on the segmentation and classification of cases at each step of the reporting, investigation, substantiation, service and outcome process.” The result was the creation of groups or clusters that have a similar combination of characteristics based on hotline and investigative data. Each stage of the modeling process produces progressively more uniform groups. The goal was to ensure that if these groups received different treatments, the difference in outcomes would be due to the treatment and not some other aspect of the children or their situations. Within each group of similar children, the researchers compared those who receive different interventions, namely removal from the home or community-based prevention services.  They used a technique called propensity score matching to control  for differences between members of .each group that might affect their outcomes. The authors use one outcome–whether a child is re-referred to the system within a year of exit–to determine whether each intervention was successful.

Based on this analysis, the authors concluded that many families are receiving services that are too intensive for their needs. For example, they concluded that “at least 40% of the cases that were referred to the court and to Childnet (mainly for foster care) were inappropriate based on the outcome data for children in their cluster group. The authors then went on to claim that these  “inappropriate referrals”  are actually harming children. For example,  “inappropriate referrals” to court were 30% more likely to return to the system after the court referral than they would have been if the referral had not been made.  And “inappropriate referrals” to ChildNet were 175% more likely to return to the system than similar cases that did not receive such a referral.

Finally, the authors present a “prescriptive” model that addresses the question, “Which services are most likely to prevent a case from having another report of abuse an/or neglect [within a year]?” This concept of “prescriptive analytics” is a new one in child welfare if not human services in general. The authors devote only two paragraphs to this model but they note that it would result in a decline in “inappropriate referrals” to court and ChildNet.

Even if we accept the machine learning process presented by the authors as a reasonable basis for estimating risk, several issue remain about the authors’ findings. The first issue is the use of one-year re-referral rates to denote intervention success. Ongoing maltreatment may not be seen or reported for months or years. The authors report that 57% of their cases that received another referral did so within one year. However, that leaves 43% that were referred after a year had passed. These cases were not counted as “failures” by their model. In addition, because the databased covered only 2010 to 2015, the authors did not include any referrals that happened after 2015, including those that are yet to happen. If the authors classified  some cases wrongly as not returning, this reduces the validity of their model.

The second problem stems from that famous social science bugaboo–unmeasured differences between groups. The authors relied entirely on hotline and investigative data on family history and characteristics. Yet, many family issues may not be reflected in the data. These could include unknown histories of criminal behavior, mental illness, violence, or drug abuse. If the authors observed that an intervention appeared to cause harm to certain children, the explanation may not be that the intervention was inappropriate. A more plausible explanation might be that that the matching algorithm failed to correctly assess risk as well as the social workers in the system.  If the cases referred to the court were in fact those that social workers correctly identified as being at higher risk (even though this was not picked up by the algorithm) one might expect higher rates of return to the system of these cases relative to cases that were matched with them by the algorithm but not referred to the courts.  This possibility seems a lot more likely than the possibility that court-ordered services made parents more abusive or neglectful.

A third problem relates to the use of the child rather than the family as the unit of analysis. The family or household is obviously the appropriate unit of analysis here. It was the parents or caregivers that perpetrated the abuse or neglect and they are the main recipients of services. Author Peter York agreed that using the family would be be more appropriate but explained that most of the data in the system were linked to the child and not the family. Using the child as the level of analysis means that the same parents will be counted as many times as they have children in the system. This will obviously weight larger families more heavily, with whatever biases this may introduce.

Finally, it is concerning that the authors reported about the proportions of children that were provided with too-intensive services such as foster care but not the proportion that were provided with services that are not intensive enough. We all know about the worst case scenarios when children die or are severely injured after the system failed to respond appropriately to a report, but there are many more cases in which allegations are not substantiated or interventions are not intensive enough, and the children return to the system later, often in worse shape. Reporting on one type of error but not its opposite invariably raises questions about bias.

The authors should not be blamed for making too much of their findings. In their article abstract, they do not mention the specific findings about over-reliance on foster care and more intensive child welfare interventions. Rather, they  argue that their findings indicate that “predictive analytics and machine learning would significantly improve the accuracy and utility of the child welfare risk assessment instrument being used.” I fervently agree with that statement. But this new approach by Schwartz et al is qualitatively different from the predictive risk modeling algorithms currently being applied and studied by jurisdictions around the country. In particular, the authors used machine learning to identify groups with similar risks but which received different treatments. Their purpose was to assess the effectiveness of distinct treatments of different subgroups. How well this approach will accomplish that purpose remains to be seen. This fascinating study is just the the beginning of a conversation about the utility of this new approach, not an argument for reducing the reliance on foster care or community services.





Secrecy in child welfare: cover up or get better?


Evan Brewer, Caleb Blansett, Adrian Jones: From http://www.crimeonline.com

Clint Blansett’s 10-year-old son had been dead just a few days when a social worker from the state knocked on the family’s door in south-central Kansas . She wasn’t there to offer condolences after Caleb’s death or ask about his sister, Blansett said. She wanted him to sign a form saying he wouldn’t talk about his son’s death or the Kansas Department for Children and Families. No details about contact the agency had with the family before Caleb’s mom smashed his head with a rock while he slept and then stabbed him seven times.

So begins a story by the Kansas City Star entitled Secrecy inside child welfare system can kill: ‘God help the children of KansasIn it. reporter Laura Bauer describes an agency that chooses to protect itself at the expense of fulfilling its mandate to protect kids. Among the examples included in the story are

  • A DCF deputy director resigned after she was asked to shred notes of meetings about critical cases. Furthermore,  her attempt to implement a systemwide review process for such cases was refused because administrators did not want mistakes documented in writing lest they would be used in court against the agency.
  • For a year and a half, DCF refused to release information about its repeated interactions with the family of Adrian Jones, who was killed by his father and stepmother and fed to their pigs. It was only after the murderers were sentenced to life in prison that DCF reduced 2,000 pages of records that were haphazardly thrown together in what looked like a purposeful attempt to baffle readers. The records, once put in order, revealed multiple investigative errors, particularly three that probably cost Adrian his life.
  • A Wichita television station reported that DCF received several reports of mistreatment of Caleb Blansett, beginning in 2012 and continuing in the months before his death. On August 3, 2017, the Star requested information about these calls and any ensuing investigations. Three months later, DCF responded that it did not have the staff to respond to the request.
  • Just this past September, the body of three-year-old Evan Brewer was found in a cement structure outside the house where his mother and boyfriend were living. He had been missing at least since the previous March. His father claims to have made multiple reports to DPS alleging abuse of Evan.  DCF denied a request from a local TV station for the records relating to these reports.

Kansas law requires that “in the event that child abuse or neglect results in a child fatality or near fatality, reports or records of a child alleged or adjudicated to be in need of care received by the secretary, a law enforcement agency, or any juvenile intake and assessment worker shall become a public record and subject to disclosure.” But unfortunately, the law does not define “reports and records.”

To receive federal money under the Child Abuse Prevention and Treatment Act (CAPTA), a state must allow “public disclosure of the findings or information about the case of child abuse or neglect which has resulted in a child fatality or near fatality.” Unfortunately, the vagueness of this language allows states to avoid releasing information necessary to identify how the agency failed. In a report entitled State Secrecy and Child Deaths in the U.S., two child advocacy groups found that all states have some sort of public disclosure policy regarding child abuse deaths. However, the report gave 20 states (including the three most populous states) a grade of C or below on these policies based on a variety of criteria, including whether they were encoded in statute, whether the disclosure is mandatory, and the scope and specificity of the information that must be disclosed.

Kansas was actually in the better half of states. It received a “B” from authors of the state secrecy report, mainly because it has a policy, the policy is encoded in statute, and is mandatory, despite the vagueness of the information that must be released. It is worth noting that only 14 states got higher than a B grade. Moreover, the report’s authors also found that states often fail to abide by their own disclosure policies–as when Kansas claimed to lack staff to respond to the request for information about the death of Caleb Blansett.

New Jersey’s child welfare agency, under the guise of protecting children’s privacy, in 2013 adopted a rule that the child welfare agency must release information only “to the extent it is pertinent to the child abuse or neglect that led to the fatality or near fatality.”  Even the under the new tightened rules the agency should have disclosed information about its past interactions with the family of  JoJo Lemons after he became the third sibling in his family to die while sharing a bed with other family members. His  parents were charged with reckless manslaughter and child endangerment, and each pleaded guilty to a count of child endangerment. Nevertheless, CPS concluded that JoJo’s death was not caused by abuse or neglect. Therefore, the agency was not required to release information about its interactions with the family.

In Cleveland, 5-year-old Tenasia McCloud was beaten to death by her mother and her girlfriend on March 17, 2017.  At the time of her death, the child welfare agency had an open case on the family, according to News 5 Cleveland. A social worker had visited the home eight times, including three days before Tenasia was brought to the hospital in cardiac arrest. The paper tried to find out how the agency did not see that the child was in danger. But Cuyahoga County Children and Family Services refused to provide records of agency contacts with the family, citing a rule prohibiting disclosures that might jeopardize a criminal investigation or proceeding. Only five other states have a similar rule, according to the State Secrecy study, suggesting that it is not a necessary requirement. Moreover, two states conversely allow disclosure only if a person is criminally charged or would have been criminally charged if they had not died.

Congress and the states must strengthen disclosure requirements in the event of child maltreatment fatalities or near-fatalities. Congress should amend CAPTA to define specifically what data states must release in the event of a child maltreatment fatality or near fatality. Until that happens, states should amend their own laws to strengthen the disclosure requirements. These disclosures should be required with no exceptions to any member of the public. The information required to be disclosed should include a summary of all past reports on the family or household, whether these reports were investigated, results of all past investigations and reasons for the determinations made; as well as a summary of all cases opened for the family or household, what services were provided, when the cases were closed and the reasons for closure.

Congress and states should also require that a commission of experts review every death or near-death of a child in a family known to the child welfare system. As I stated  in a previous post, the death or severe injury of a child in a family known to the child welfare system should be treated like a plane crash or the loss of the space shuttle Challenger. All such deaths or severe injuries should be reviewed immediately and exhaustively by experts of the highest caliber with access to all agency records regarding contact with the family or household. The report should include recommendations to avoid similar tragedies in the future and should be released to the public with names redacted when necessary to preserve the privacy of innocent children and adults.

The point of requiring release of information and analysis of case history is not mainly to allocate guilt or punishment, although practitioners guilty of egregious errors should be retrained or let go. Rather it is to identify policies or practices that can save children in the future. As the authors of the state secrecy report put it:

Abuse and neglect deaths represent child welfare agencies’ most tragic failures.        Unfortunately, it is often only through such cases that lawmakers and the public learn of systemic inadequacies in child welfare systems. If improvements and reforms are to be achieved, it is vital that the facts about these cases reach the public in a meaningful way.





Domestic violence and child abuse: a lethal combination


It did not take long for the press to discover that Devin Kelley, the the perpetrator of the  recent mass shooting in Texas, had repeatedly assaulted his first wife and fractured the skull of his infant stepson in 2012. He was court-martialed for those offenses, pled guilty, and was imprisoned for a year.

I could not help noting the parallel to the case that I wrote about in my last post–that of Antoine Flemons, who at two months old was beaten to death by his father, Antoine Petty. The post focused on one aspect of this case–the fact the father was known to have abused many other children.  I argued that baby Antoine might have been protected by a broader policy to identify at birth babies born to parents with such a record.

But the revelations about the Texas shooter reminded me of another important aspect of Antoine’s family that put the baby in grave danger.  In an interview with the Washington Post, Antoine’s grandmother stated that her daughter Geneice Petty loved her son but suffered from “battered women’s syndrome.” In other words, she was a victim of domestic violence.

The connection between domestic violence and child abuse is well-documented. Research suggests that “in an estimated 30 to 60 percent of the families where either domestic violence or child maltreatment is identified, it is likely that both forms of abuse exist.”

In the 40 states providing domestic violence data to the Administration Children and Families for its Child Maltreatment 2015 report, 25% of child maltreatment victims were found to have a caregiver who was either a victim, perpetrator or witness of domestic violence.

Co-occurring domestic violence and child abuse can take several forms. In many cases, one parent (usually the father) abuses both the other parent and the child or children. There are other configurations, such as families in which the abused parent in turn abuses the children.

In baby Antoine’s case,  no information has been released to the public. One can speculate in view of the father’s extreme violence that Geneice Petty was afraid to protect her children and that her husband bullied her into covering up his killing of their son.

The key question is what could have been done to prevent the death of Antoine. Historically, child welfare systems have had not responded effectively to domestic violence. Common and problematic patterns have included ignoring or minimizing the domestic violence and, conversely, giving women an ultimatum to leave the abuser or leave their children–a response which often leads women to fear and avoid child protection authorities rather than seek their help.

One study found that “[Domestic violence]appears to have only a minor role in influencing the decisions of child welfare workers; yet, children exposed to [domestic violence] often have multiple contacts with [child welfare services] due to the higher number of repeat allegations of maltreatment.”

The Children’s Bureau has has published a useful manual about how to handle child maltreatment cases in which domestic violence is present or suspected. The manual’s many recommendations provide alternatives to the problematic practices mentioned above.

Unfortunately, we don’t know if Maryland child welfare workers even identified domestic violence in earlier cases involving Antoine’s parents, let alone how they responded. That’s why, as I have said over and over again about all child maltreatment deaths and serious injuries, there needs to be a thorough investigation, a public report, and a proposal for changes in policy and practice to protect future baby Antoines.



Would a broader birth match have saved Antoine Flemons?

Antoine Flemons 3
Photo from GoFundMe fundraiser by Geneva Flemons

Little Antoine Flemons never had a chance. Prince George’s County Maryland Prosecutors described how his father, Antoine Petty, “dangled the infant by the arm and repeatedly struck him before handing the baby to his mother to feed. When the baby continued to cry, Petty dealt another round of blows, quieting the child forever.”

Antoine’s parents left his body in the car for over 24 hours before burying him, according to police.  The Judge sentenced Petty to 40 years in prison for his son’s murder. Antoine’s mother pleaded guilty to involuntary manslaughter and will be sentenced in December.

Information shared by the prosecutors revealed that Petty, the father of nine, had a long history with Child Protective Services dating back until at least 2007. He was reported for carving a three-inch cross into a five-year-old daughter’s arm, pushing a five-year-old down stairs, giving a ten-year-old a black eye, forcing a daughter to watch him having sex with a girlfriend, and failing to adequately nourish an eleven-month-old. One of his children was found at age 11 months to have rib fractures which were found by a doctor to be ‘not accidental.”

How could this father be allowed to mistreat child after child and this mother to fail to protect them for close to ten years when so many acts of maltreatment were reported to CPS? It would be more appropriate to ask how such a parent can be stopped. When an abusive parent has a new child, there is no mechanism in most states to trigger protection for that child.

Interestingly, Maryland is one of the few states that does have such a mechanism– a “birth match” program. Under birth match, birth records are matched against a list of parents who had their parental rights terminated within the last five years due to abuse or neglect. Parents thus identified receive a visit from a social worker to assess the child’s safety. If the parents refuse the visit, a case can be opened if there is reason to expect abuse or neglect.

But Maryland’s birth match law did not protect little Antoine. It is unlikely that his parents had their rights terminated in the past. Perhaps Antoine would have been protected by a broader law, such as Minnesota’s, which triggers an investigation or family assessment under a broader set of circumstances. These include when a parent has subjected a child to “egregious harm,” has failed to protect a child from such harm, has committed child neglect endangering physical or mental health, and has committed first second or third degree assault among others.

We don’t know if a broader birth match law would have protected little Antoine because no information has been released about the results of the prior investigations against Antoine’s parents.

As I discussed in an earlier post, all deaths of children in families known to CPS should be investigated immediately and the results made available to the public. Only with such an investigation can we know how and why the system failed little Antoine.

There has been a shocking lack of calls for such an investigation from Maryland legislators and child advocates. Only  the Washington Post broke the silence, asking, Could this 2-months old’s death have been prevented? Nobody who cares about children in Maryland should rest until they know the answer, and until measures have been put in place to prevent similar tragedies in the future.




No place for the children: A therapeutic group home closes while foster children sleep in hotels and offices

In my last post, I wrote about Washington State’s critical shortage of foster parents, which is rRuthDykemanCenteresulting in children staying in offices, hotels, and by-the-night foster homes. One of my suggestions was to reinvest in quality group care settings. Unfortunately, the state (along with most of the country) is moving in the opposite direction.

KUOW, Seattle’s public radio station, recently  reported on the closure of a group home that provided therapeutic care to  foster children with “severe behavioral problems and emotional needs.” At the Ruth Dykeman Children’s Center in Burien, Washington 15 children lived in lakeside cottages supervised by staff members, with nurses and psychologists on call.

Unfortunately, foster care ideology has changed and now any family setting seems to be considered better than any  group  setting, regardless of the needs of the child and the quality of the placement. The fact that group settings are more expensive than foster family homes might have something to do with this new bias.

Unfortunately, the type of children that were housed at Dykeman don’t do well in family foster care. Children with behavioral problems and emotional needs tend to bounce from one foster home to another, their behavioral problems worsening with each move.

Nevertheless, group homes have been shuttered around the country. In Washington  state, according to Investigate West, “stagnant reimbursement rates have forced many facilities that contract with the state to reduce capacity or shutter altogether.”

The CEO of Navos, the mental health nonprofit running the Dykeman home, told KUOW that ending the contract for foster care was a source of great anguish to the leadership. But it was not financially sustainable. The nonprofit had been paying more than half the cost of running the home for years.

The Dykeman Center is not closing, but it is now off-limits for foster kids. It will now serve long-term inpatient psychiatric care, which is reimbursed at two to three times the rate, according to KUOW.

Now, the fragile children from the Dykeman Center will be competing with less troubled but still vulnerable foster youth for the dwindling supply of foster homes. Some may bounce from home to home, perhaps spending nights in hotels or pay-by-the-night foster homes where they have to be dropped off late in the evening and picked up early in the morning. Some have already been sent out of state, according to KUOW.

It is hard to conceive of a reality where this makes sense. But in the looking-glass world of foster care, ideology and money-saving work together to trump common sense and common humanity.