Wednesday, February 26, 2020

NCCPR news and commentary round-up, week ending February 25, 2020

● Mike Hixenbaugh of NBC News has done a series of outstanding reports about some so-called “child abuse pediatricians” whose blindness to other explanations for childhood injuries or illnesses can wreak havoc in families.  But what happens when the person everyone turns to as the child abuse expert doesn’t even have the credentials to be a child abuse pediatrician in the first place?  That’s what’s happening in Washington State.

● What happens when hospitals drug test pregnant women and rush to turn them in to child protective services? Pretty much what you’d expect, writes Charissa Huntzinger of the Texas Public Policy Foundation:

Women have described efforts to minimize the risk of arrest or punishment through social isolation, withholding relevant medical information, avoiding prenatal care, skipping treatment appointments, or avoiding treatment all together. This means our efforts to protect children are actually placing them and their parents at greater risk of harm.
● If the Orwellian predictive analytics algorithm in Pittsburgh is really so great, why does the director of the county human services agency have to obfuscate so much about what it is and how it really works?  I have a blog post about the selling of “Hello Baby.”

And there are several interesting stories about kinship foster care:

The Chronicle of Social Change reports that still another study shows still another benefit of placing children with relatives instead of strangers: The children are less likely to ultimately be moved into the worst form of “care” group homes and institutions. One possible reason: Relatives are less likely to give up on their own relatives when those children have behavior problems.  (Similarly, other data show that children living in kinship foster care are less likely to be forced to take potent, sometimes dangerous psychiatric medication.)  Another finding from the study: Children are more likely to be institutionalized if they’re Black.

● Also in the Chronicle former foster youth Georgette Todd writes about what she discovered only after she “aged out” and started looking for family on her own:

When my aunt picked up the phone, we caught up with our lives but then she said something that truly stunned me and still has a deep impact on me to this day. She said, “Georgette, I never knew you were in foster care. No one notified me … had I known, I would have taken both you and your sister in.

Todd argues that, with modern technology, there’s no excuse for failing to find relatives. 

● But the problems aren’t always technological. Sometimes the problem is the bias some child welfare systems still have against extended families.  That bias can be seen in the absurd requirements some states impose before a relative can become licensed to be a foster parent. As this story from Chris Serres of the Minneapolis Star-Tribune makes clear, Minnesota is a case in point.

● The hostility to kinship foster care is just one aspect of a bigger problem in Minnesota: The state’s sky-high rate of tearing apart families, and one of the worst records in the nation for racial bias.  KSTP-TV has a story about a hearing that dealt with these issues. And NCCPR has additional context here.

Sunday, February 23, 2020

Predictive analytics in child welfare: The selling of Hello Baby

If the Orwellian predictive analytics model in Allegheny County is really so great, why does the county’s human services director have to obfuscate about what it is and how it works?

 Read between the lines in some of the comments of those supporting Allegheny County’s Orwellian use of “predictive analytics” in child welfare – now including a plan to try to label every baby with a “risk score” at birth -- and you find something that boils down to this:  Well, yes, it’s potentially dangerous, but we trust Marc Cherna.  Cherna is the longtime director of the Allegheny County Department of Human Services.  

The sentiment is understandable.  Cherna has held the job since 1997.  I served on a search committee that unanimously recommended him.  For a long time I was proud of that. Cherna built an impressive record as a reformer, curbing the number of children in foster care and becoming a leader in the use of kinship foster care, among other accomplishments.  That built him a reservoir of goodwill.

But there are two principal problems with the “trust Marc” argument. First, Marc won’t be there forever.  And second, the way he is selling the program undermines trust.

It’s not the embrace of predictive analytics that raises questions about trust. There are other good, honorable child welfare system leaders who support it (though it appears that most of the leading proponents also are the most extreme supporters of a take-the-child-and-run approach to child welfare).  The trust questions arise because of how the county has sought to sell the first predictive analytics program, the Allegheny Family Screening Tool (AFST), which is used to help decide when to investigate reports alleging child neglect, and the even worse “Hello Baby” program – the one that seeks to slap that risk score on every newborn in the county.

The problems with the sales job are apparent in an article Cherna wrote for the Department of Health and Human Services publication Children’s Bureau Express.

Compare the rhetoric to reality

Cherna writes that the program is “supported by the use of integrated data and a predictive risk model…”  That’s it; there’s no further explanation.  In fact, the “predictive risk model” is the lynchpin of Hello Baby – the whole thing collapses without it.

Under that model, vast amounts of  data are collected, most of it from systems that enmesh primarily poor people. Then the system coughs up a “risk score” for the newborn based on how their parents have interacted with those systems.   Cherna promises that the risk score will be used only to target prevention programs. Those with the highest risk scores will get the most intensive “help.”  But there is nothing to stop Cherna, or his successor, from using the data in any other way they choose. We just have to trust them.

Cherna writes that “Hello Baby is a program of DHS's Office of Community Services and is not connected with child protective services.”  He neglects two key points:

● The helpers – who are being sent into a home precisely because an algorithm says the baby is at high risk of child abuse -- also are mandated reporters of child abuse. 

● The person in charge of both the Office of Community Services and child protective services is – Marc Cherna. He, or a successor or a successor’s successor, are free to tear down that wall of separation.

Cherna writes that the program is “voluntary.”  Technically, yes.  But when most people think of voluntary, they think of something that is offered to them and they then choose whether to take part.  Hello Baby doesn’t work that way.  Hello Baby assumes that you have agreed to surrender your data and have your baby labeled with a risk score at birth unless you affirmatively opt-out.  And it is unclear how much chance you have to do that.

So ask yourself: If Hello Baby is so great, why do they have to, in effect, sneak it past the very people its proponents say are most likely to benefit, instead of being open and aboveboard about exactly what it is, and letting people opt in if they really want it?  If it’s as wonderful as Cherna says, people should be lining up to take part.

Cherna goes on to tout the “ethical analyses” done of Hello Baby.  But he doesn’t mention how the county has stacked the deck for such analyses from the start. 

The first “ethics review” done for the AFST program, was a once-over-lightly review of a few papers, all of which were written by one of the designers of AFST or by one of the “reviewers” himself.   That reviewer is a faculty colleague of one of the AFST designers – in fact, they’ve co-authored papers together.

Even that review said AFST was ethical, in part, because it was triggered only if someone called the child abuse hotline and alleged maltreatment.  They wrote:

[The issue of informed consent] is one of a number of points at which we think that it is ethically significant that the AFST will provide risk assessment in response to a call to the call center, rather than at the birth of every child. [Emphasis added.]

So, now that Cherna is rolling out a use of analytics that does just that, what to do? First, call it “voluntary” by, as noted above, stretching the meaning of that word almost to the breaking point.  Then, commission more ethics reviews!  But make sure they’re from people strongly predisposed to support Hello Baby or apparently unfamiliar with child welfare.

All of which leaves one question: Should we be comfortable giving so much data and so much power to use that data to an agency that sells the program this way?

Tuesday, February 18, 2020

NCCPR news and commentary round-up, week ending February 18, 2020

● Who really gets caught in the net of child protective services? Who pays the price for foster-care panic, a sharp sudden spike in removals of children after a high-profile child abuse tragedy.  Some answers in this outstanding ten-minute video from BRIC TV in New York:

● They “lied, spied and withheld and destroyed evidence…”  That’s how the Everett Herald sums up the behavior of the scandal-plagued CASA program in Snohomish County, Washington.  (You can read more about the scandal here.) Part of the problem:  Judges couldn’t hold the program accountable because the CASA program was administered by the court itself.  Now that’s going to change. But the larger problem with the CASA program is simply that it’s a CASA program. Even when they behave honorably, bias is built into the model.

● A New York State Senate task force examining opioids is recommending a radically different approach to dealing with substance use, including substance use and child welfare.  What’s so radical? It actually says we should follow the evidence concerning what works and what doesn’t.  The Movement for Family Power explains the benefits of the Task Force recommendations in this statement, which includes a link to the report itself.

● A common excuse for using Orwellian predictive analytics algorithms to supposedly predict who might abuse a child (or who is safe to let out on bail, or who supposedly is cheating on welfare benefits, etc.) amounts to: Sure, the algorithms are biased, but humans are biased too!  The Correspondent has a good analysis of that issue. 

● Two Colorado legislators have a column in the Coloraro Sun about their bill to narrow Colorado’s incredibly broad “neglect” law to allow “Colorado parents to be able to give their kids the kind of independence almost all of us over age 30 grew up with – the freedom to walk to school, play outside, come home with a latchkey and so on.” 

Yes, we really do need laws to allow this.  One of the legislators recalled her own experience:

“When I was widowed, I had four kids under the age of 11. Parents know their children best and there are children 8, 9, 10 that are perfectly capable of walking to school by themselves. But the parents load them up and drive them because they’re afraid to be charged with neglect, and I think I overcompensated out of fear, too.”

● Most states have raised the age in which youth are allowed to stay in foster care to 21. A California lawmaker wants to extend it through age 25. In the Chronicle of Social Change, a backer of the original age increase, former foster youth  Akin Abioye, says the new proposal is a mistake. He suggests several better alternatives, including “Stop removing kids from their family of origin.”

And there’s more about the dangerous blunders of so-called “child abuse pediatricians.”

Vivek Sankaran writes in the Chronicle of Social Change about how the kinds of tragedies exposed by Mike Hixenbaugh of NBC News, in particular this case from Wisconsin, stem from lack of transparency, lack of accountability, and a refusal to admit mistakes.

● But Katie LaGrone, an investigative reporter for several Florida television stations, has a story (part of a series) about how Florida’s top child abuse pediatrician has attacked the character of those who dissent from their findings and instructs other doctors on how to counter alternative explanations for children’s injuries.  But don’t worry, he says, those falsely accused in Florida can always get a second opinion – from him!

Tuesday, February 11, 2020

NCCPR news and commentary round-up, week ending February 11, 2020

● Eli Hager of The Marshall Project looks at a particular subset of foster care placements: Each year, an average of 17,000 children are torn from their homes – and then sent back within ten days.  But even these short stays can traumatize children.  And, of course, they raise the question: If you can send the children back home in a few days did they really need to be taken at all?

● Anna Claire Vollers of The Birmingham News looks at the harm done to mothers and newborns by child abuse investigations prompted by false positive drug tests.  It’s another example of harm encouraged by  the so-called "plan of safe care" provision in the Child Abuse Prevention and Treatment Act.  Fortunately, there are journalists such as Vollers reporting the side of the story that the Boston Globe / ProPublica “Spotlight Fellows” left out.

­● Erik Gunn of the Wisconsin Examiner looks at the consequences of legislation that would help turn that state’s child welfare system into the ultimate middle-class entitlement: Step right up and take a poor person’s child for your very own.

● Also in Wisconsin, Mike Hixenbaugh of NBC News reports that doctors at Children’s Hospital of Wisconsin are rebelling against their own hospital administration, and the behavior of the hospital’s  “child abuse pediatricians.”  A series of meetings were held in response to Hixenbaugh’s earlier, excellent reporting.  According to the latest story:

At one internal meeting this week, some Children’s Wisconsin doctors told administrators from the Medical College of Wisconsin — which employs physicians who practice at the hospital — that without swift policy changes, they would hesitate to bring their own children to the hospital following accidental injuries, fearing that a medical mistake or overreaction could lead Child Protective Services to break their families apart.

● Sanne Blauw of the innovative news site The Correspondent has a story that has nothing to do with the American child welfare system – and everything to do with the American child welfare system. It’s about how a Dutch court ruled that an algorithm that targets poor people who might be committing public benefits fraud is a violation of human rights. It’s a lot like the algorithm Pittsburgh is using to target potential child abusers – and, soon, to try to slap a risk score on every child in the county at birth.

● A new report from researchers affiliated with what should properly be called the Penn State Penance Institute is a frightening attack on the minimal due process protections afforded those accused of child abuse or neglect in Pennsylvania.  And that means more children traumatized by unnecessary investigations and unnecessary foster care.  I have two blog posts about it.

● Another NCCPR blog post asks if a New York City foster-care panic in 2016 contributed to a child abuse tragedy last month. (It sure looks like it.)

● And Marie Claire Australia profiles an activist fighting the racism in that countries child welfare system.

Monday, February 10, 2020

Child welfare in Pennsylvania: Dialing up the crazy, part two: Enter the “Penn State Penance Institute”

A new report is a frightening assault on the minimal due process protections meant to stop child welfare agencies from needlessly traumatizing children.

             In yesterday’s post to this blog, I described how the due process protections afforded to people who are listed on Pennsylvania’s statewide blacklist of alleged child abusers or in county child welfare agency records range from minimal to none.  I also discussed how State Auditor Eugene DePasquale launched an attack on the minimal due process protections that exist that was rife with inaccuracy, stunning for its ignorance and even veered toward conspiracy theory territory.

            Well, now it looks like DePasquale has an ally – from part of the latter-day “child saver”* wing of the child welfare research community.  I'm sure they have the best of intentions. But they’ve issued a report that amounts to an assault on due process. The extremism in the document manifests itself in many ways, including a flow chart from the report itself.  This is the lower left quadrant of the chart:
  Can you spot the problem?  I’ll discuss it below, but first, let’s look at the report.

            The report was commissioned by the Center for Rural Pennsylvania, an arm of the Pennsylvania Legislature. They hired a team affiliated with the Penn State “Child Maltreatment Solutions Network.”  Though the report originally was intended to look only at commercial sexual exploitation of children, the authors kept expanding their own mandate, first to all sexual abuse cases and ultimately to broad, sweeping conclusions about the entire system – even though the case sample they examined involved only sexual abuse cases.

            Like DePasquale, the report’s authors express alarm at what they imply is an unusually high rate at which some blacklist decisions are overturned.  They’re also alarmed by the fact that, if a report turns out to be false, the state is not allowed to keep the report forever and use it against a family anyway.

The report acknowledges that blacklist decisions are overturned, by the authors’ own estimate, only 19 percent of the time.  That’s because most of those accused lack the knowledge and/or the resources or both to do anything but lodge a written appeal.  That appeal goes to the same division of the same state agency that runs the state’s child abuse hotline.  At that stage, in 2018 the number of times a child welfare agency decision was overturned was exactly zero. Among those who do have the resources and take it to a real hearing before an independent tribunal, the Board of Hearings and Appeals, about 70 percent are overturned.  (DePasquale mistakenly claimed it was 96 percent.)

            In fact, the 70 percent rate-of-reversal is roughly in keeping with at least three other states. It also makes perfect sense, given the sloppy, one-sided nature of many child abuse investigations.  The Penn State report itself devotes page after page to condemning the poor quality of these investigations. (DePasquale raised similar concerns.) 

The report even admits at one point that “the rate at which cases indicated by [county children and youth services agencies] are overturned on appeal is high, which may both affect and reflect the concerns raised in this research pertaining to the quality, scope, and documentation of investigations.” [Emphasis added.] 

            Yet the Penn State report warns darkly that if decisions by county CYS agencies “are frequently overturned on appeal, this could provide a disincentive to making a finding of indicated on a CYS case.”  This reflects a remarkably dim view of frontline child abuse investigators. It suggests they would rather let child abusers run free than actually work improve their investigations.

Taking out the trash

            Under Pennsylvania law, if the Board of Hearings and Appeals or the courts overturn a listing in the registry, or if the county CYS agency finds it unfounded in the first place, the state must expunge the record so the false allegation can’t be used against a family in the future.

            But the presumption of guilt in child welfare is so strong that many jurisdictions want to keep false reports on the theory that, if there are enough of them, then they can’t be false.  According to Marc Cherna who runs the Allegheny County system, the one that leads the nation in weaponizing false allegations and other junk information against families: “A lot of times, where there’s smoke there’s fire.”

            This is an open invitation for anyone who wants to harass a family – an ex-spouse, an angry neighbor, a school district that wants to collect lunch money, or even an upset dentist – to simply call in more and more reports. But also, in child welfare, where there’s smoke, there’s often just an overwhelmed inexperienced caseworker blowing smoke.

            But in the atmosphere of hysteria that has pervaded Pennsylvania since the scandal over former Penn State football coach, former foster parent and former group home operator Jerry Sandusky, the state now gives permission to counties to keep junk data about families forever.  According to the authors of the Penn State report, even that is not enough. They want to require it, and they want the state to keep all of its junk data as well.

            Their excuse for this recommendation is instructive, but not in the way they had in mind.  First they write that expungement 

…may result in wrongly-accused persons being repeatedly subjected to investigation for allegations that were already investigated when the investigation has been expunged and the exoneration details are no longer available.

            Aside from the fact that no one could read this report and believe that the authors give a damn about “wrongly-accused persons” (as is discussed below, it’s not even clear they think such persons exist) they contradict this argument in the very next sentence:

At the same time, expungement may also allow serial perpetrators of child sexual abuse to go undetected. That is, expungement of unfounded allegations disallows consideration of the accumulation of allegations, patterns, and reports that are often critical to exposing serial perpetrators of child sexual abuse.

            In other words, “Where there’s smoke …”  If they believe their own second sentence then there’s no way a pile of unfounded reports clears anyone – it just makes them more suspect.

            Indeed, this mindset is confirmed, accidentally, by a spokeswoman for the state Department of Human Services – which loves the prospect of being able to keep files on the innocent forever. She told the Daily Item:

“Prior abuse reports, regardless of substantiation, are a strong indicator of risk of future abuse … Counties should be aware of all prior incidents, including unfounded or invalid reports, when investigating a new report.”

But there is no real evidence for this. The research that supposedly shows this actually shows only that prior reports make it more likely that there will be more reports – in other words, it amounts to a self-fulfilling prophecy. Workers who assume that where there’s smoke there’s fire are more likely to check the equivalent of the box marked “fire” when the next report comes in.

            And, of course there is a simple way to solve the problem of someone who is innocent not having a record of a false allegation: Send a copy of the record, minus anything that genuinely violates privacy, to the accused, so they can keep it and show it to the caseworker the next time that disgruntled neighbor, ex-spouse, dentist or school district demanding lunch money files a false report, or the next time a well-meaning reporter is simply mistaken.

            The report’s recommendation in this area appears under the header “Stop deleting data.”  But a key principle in any use of data is “garbage in, garbage out.”  Expunging false reports is not deleting data.  It’s taking out the trash.

Another assault on due process

            You can’t convict someone of murdering a child without proof beyond a reasonable doubt.  But in most states you can tear a child from a loving home and consign her or him to the chaos of foster care, indefinitely, using the lowest standard of proof in American jurisprudence – “preponderance of the evidence.” That means you were able to convince a judge that it is slightly more likely than not that the allegations are true and foster care is necessary.  It’s the same standard used to decide which insurance company pays for a fender-bender.

            That’s part of the reason it’s so easy to do so much harm to children. First, confuse family poverty with neglect, then take away the child (in every state workers can do that on their own authority or ask law enforcement to do it for them). Then, after-the-fact, go to court for a hearing – usually a secret hearing - where the family probably has either no lawyer or someone they just met five minutes before the hearing. The state has had plenty of time to prepare its case – and it barely has to “prove” it at all.

            In theory, Pennsylvania is a little different.  In Pennsylvania, CYS agencies eventually – though not immediately - have to meet the middle standard of proof, known as “clear and convincing.” That’s still lower than “beyond a reasonable doubt” but, theoretically, higher than “preponderance of the evidence.”

            In contrast, all it takes to blacklist the accused is a caseworker’s conclusion – with no hearing or impartial fact finding beforehand – using the preponderance standard.

            So what does the Penn State report say? While never explicitly calling for lowering the standard for holding a child in foster care, it’s clear where the authors come down. They write:

This creates a scenario in which it is known that a child is at risk of harm but no action to protect the child can be taken without the consent of the parent, who may themselves be the identified perpetrator.

           But that is inaccurate.  

            First of all, in Pennsylvania a child can, in fact, be in foster care for up to 13 days or more before this higher standard ever kicks in. That happens – in theory -- at a later hearing to determine if the child needs to stay in foster care.

But more important, the whole point of a court hearing is to find out if the child actually is “at risk of harm” and whether the parent is, in fact, the perpetrator. If a caseworker’s guess based on preponderance of the evidence were enough we could just abolish the entire court system.

The Penn State report’s framing also flunks the balance of harms test. It fails to acknowledge the enormous harm of intervention itself. It ignores the evidence that in typical cases children do better in their own homes even than comparably-maltreated children placed in foster care.  It ignores the high rates of abuse in foster care itself.  One should, indeed, have clear and convincing evidence before putting a child at risk harm by needless placement in foster care.

And finally: The “clear and convincing” standard does not stop Pennsylvania from taking away children at a rate above the national average. And it doesn’t stop Philadelphia from taking away children at the highest rate among America’s largest cities.

So yes, there’s a problem with the “clear and convincing” standard in Pennsylvania law: Judges keep ignoring it.

The bias that permeates the report

            I noted above that it’s not clear the authors believe there’s really any such thing as a false report.  Indeed, the overall message conveyed in the report is: If you’re accused you’re guilty until proven innocent – and no one is really innocent.

That’s based on, among other things, the flow chart I mentioned at the top of this post.  The flow chart is misleading on one count and flat wrong on another.  Take another look: 

            It is misleading in that it suggests, as does the report narrative, that those accused of child abuse have a wondrous array of options before them and chance after chance to prove their innocence.

            In fact, that’s only true if they’re rich, can hire great lawyers, and convince court after court to hear their appeals.  In the real world, there’s the BHA – and often, not even that.  (And remember, in most neglect cases there is no right to appeal at all.)

            The error of fact appears over and over in the lower left quadrant of the chart. That’s where, time after time after time the report lists a tribunal’s options as either “ruling in favor of CYS” or “ruling in favor of perpetrator.” [Emphasis in original.]

            The problem, of course, is that if the tribunal says the allegation was false, then the accused is not a perpetrator.  To claim that ruling that the state could not meet the bare minimum required to “indicate” an abuse allegation equals “ruling in favor of perpetrator” is like saying “the jury found the murderer not guilty.”

Of course sometimes a jury gets it wrong and does acquit a murderer.  And a system as arbitrary, capricious, and cruel as child welfare undoubtedly errs in all directions – including wrongly exonerating some child abusers.  But both in the flow chart and often, though not always, in the narrative, this report labels the accused as perpetrators even after a finding of abuse is overturned.

            The fact that a document purporting to meet academic standards includes this mistake, over and over and over, and no one apparently noticed, speaks volumes about the mindset of the research team.  Indeed, in reading the report, I kept thinking of something said by a prosecutor who had falsely accused large numbers of parents as part of the mass molestation hysteria in the 1980s. Said the prosecutor: “I am sick to death of things like the presumption of innocence.”

Beyond the flow chart

            The whole report has a tone reminiscent of the worst excesses of that era, the time of  cases like the McMartin Preschool, allegations of mass child abuse by Satanic cults, and so on. 

            Particularly frightening is a section concerning non-offending parents in cases of alleged sexual abuse, including parents who were unaware of any alleged abuse. The report contends that any such parent (usually the mother) who does not immediately believe the allegations, turn against the alleged offender (usually the father) and do whatever is demanded by child protective services is herself a child abuser. 

According to the report, agreeing to keep the alleged abuser away while the case is being investigated is not enough.  In classic 1984 style, you must believe the charges, and the belief must be immediate, absolute and unwavering.  Because, apparently, those same caseworkers who do those slipshod investigations can’t ever be wrong when they accuse someone of sexual abuse.

            I know of no studies documenting how many families are destroyed when a parent either accepted these demands or rejected them – and it later turned out the allegation was false.  But having been around during the McMartin-era hysteria I can say that, anecdotally, such cases arose over and over.  Children were not only taken from both parents, but loving relatives were denied custody if they did not convince authorities that they believed the charges.

            The justification for this, according to the report: If a parent is not an instant true believer ‘it is possible” that parent may “[pressure] the child to recant or deny the abuse allegations.”

The report is filled with speculation about what “could” or “may” happen – with no evidence or studies cited. (Recall for example the claim that if the BHA does its job and reverses the results of slipshod investigations it “could provide a disincentive” to indicating a case. [Emphasis added.]) That’s not scholarship, that’s fearmongering.

            But, to put it as they might, it could be that this is only to be expected when the organization doing the research was born out of a quest to burnish an institution’s image.

            The Penn State Child Maltreatment Solutions Network should really be called the Penn State Penance Institute.

            The Network was a public relations tactic; part of Penn State’s effort to rehabilitate its image after the Sandusky scandal.  But for the p.r. strategy to work, Penn State has to show that it’s tougher than anyone on child abuse. That leaves little room for nuance or for recognizing the need to balance the harm that may have been committed by the alleged perpetrator against the harm done to children by needless intervention.

            There is plenty in the Penn State report to suggest that the authors have a genuine passion for easing the plight of abused children.  But there is nothing in it to suggest the ability to keep biases in check that is needed for reliable scholarship.

*-In the 19th Century, advocates for tearing apart the families of impoverished immigrants proudly called themselves “child savers.”  They left a legacy of bigotry that permeates today’s system.

Sunday, February 9, 2020

Child welfare in Pennsylvania: They’re getting ready to dial up the crazy again, part 1

            Every few years since the scandal involving Jerry Sandusky, the convicted child rapist, former Penn State football coach, former foster parent and former group home operator, Pennsylvania legislators churn out a new set of bad laws so they can then churn out press releases about how they’ve “cracked down on child abuse.”

            At least one statewide official, State Auditor Eugene DePasquale, has been quick to exploit the tragedy surrounding this former foster parent and group home operator to promote policies that are only likely to lead to more children needlessly placed in foster homes and group homes.  He also comes frighteningly close to trafficking in conspiracy theories.

            Now it looks likely to happen again.  This time the target is the bare minimum amount of due process that exists for families in which someone is wrongly labeled a child abuser, sometimes based on little more than a caseworker’s guess.

            A new report, from researchers affiliated with a group set up by Penn State to do penance for its role in the Sandusky scandal, is attacking the same minimal due process protections that DePasquale went after last year.  I don't doubt that the researchers have the best of intentions. But their bias is, literally, visible in the report itself.

The background

            As in most states, in Pennsylvania all it may take to be labeled a child abuser is for a caseworker to check a box on a form based on her or his belief that it is slightly more likely than not that the accused abused or neglected a child. 

In cases of abuse and what authorities deem severe neglect, that puts you in a statewide central register and can bar you from all sorts of jobs that involve coming anywhere near children. (Pennsylvania is unusual in that the statewide register doesn’t apply to all neglect cases.)  Being on the blacklist bars you from jobs ranging from teacher, to crossing guard, to school lunch aide, to clerk in a children’s bookstore. It also bars those blacklisted from working in home health care.  And, though not required by law, some employers even bar the blacklisted form working with the elderly.

Many of these are low-wage jobs – first-step-out-of-poverty-jobs. So, as with everything else in child welfare, being wrongly blacklisted does the most harm to poor families. The stress of the listing and potential job loss itself increases the harm to children. The harm worsens if the parent on the blacklist can’t get a job, making it more likely the family will be plunged into poverty that is confused with neglect.

            And, of course, being listed in the registry means the next time the same person is accused, it is that much more likely the children’s suffering will increase exponentially because the worker will be more likely to place those children, needlessly, in foster care.  After all, they were accused before so they must be guilty, right?

            “All too often, indicated reports are based on faulty or incomplete investigations, or on actions or omissions by parents or caretakers that simply do not meet the statutory definitions of child abuse,” writes Janet Ginzburg, a lawyer for Community Legal Services of Philadelphia.  She cites this example:

A woman and her ex-husband shared custody of their 18-month-old daughter. After spending the weekend at her dad's house, the girl was returned to her mother's home with a case of diaper rash.
The dad notified the mom of the rash and gave her a tube of ointment that he had been applying. The mom watched the rash for several days, applying the ointment as directed by the medication's instructions. When the rash did not go away, she took her daughter to the doctor, who found that the rash had become infected and reported the mother to the local child-welfare agency for child neglect.
The mom had done what almost every parent would have under the circumstances - she had treated the rash, watched it closely, and ultimately made the decision that it required medical attention. However, she was placed on the civil statewide Childline Abuse Registry and lost her job as a home health aide. Until she was able to get a hearing and clear her name - a process that can take as long as a year - she was unable to get another job in the profession she had been trained in.

            Note that the mother did not get a hearing before she was blacklisted; she had to fight her way out afterwards.  As Ginzburg explains:

There is no hearing, no opportunity to present evidence. A caseworker checks off a box, and an individual is placed on the registry effectively for life - unless this person successfully appeals within a very short deadline.

            An appeal is the only way out.  That way already is foreclosed in many cases.  And in the latest example of ratcheting up the crazy, it looks like DePasquale and others are gearing up for an effort to make that one way even harder.

Pennsylvania’s weird system 

In Pennsylvania, the child abuse hotline and the central register are run by the state, but almost everything else is the responsibility of individual counties.  That’s unusual, but not unique. New York, for example, operates the same way.

            But here’s what is unique about Pennsylvania: Unlike every other state, Pennsylvania classifies abuse and what they deem the most serious cases of neglect as “child protective services” (CPS).  But other neglect cases are classified as “general protective services” (GPS).  In both types of case, a report can lead to an intrusive, coercive investigation and in both types of case that investigation can lead to consigning the children to the chaos of foster care. 

            The statewide central register applies only to CPS cases.  There is no statewide register for GPS cases – but counties still keep these records. They are not checked by employers, but they are used by county child welfare agencies in assessing future reports alleging abuse or neglect.  (This is especially dangerous in Allegheny County, where they’ve created an Orwellian automated predictive analytics algorithm that relies heavily on these data.)

            This makes it especially important that the accused have a chance to correct the record, since the record can be no more than a caseworker’s guess.

But in GPS cases, because there is no statewide central register, there is no appeals mechanism at all.  The caseworker’s word is – almost literally – law.  The family has no recourse against this stigmatizing and stereotyping – and in Pittsburgh, no recourse against such a false report raising an arbitrary risk score that can haunt not only the parents, but their children for life.

            Needless to say that’s not what has people like DePasquale upset.  Nope. They’re upset with the minimal due process protections that exist in CPS cases.

            In these cases, one can seek administrative review from the same state agency that runs the statewide child abuse hotline. But of the 982 such cases they reviewed for which they reached a decision in 2018 they overturned exactly zero.

            Clearly one’s only hope is to take the case to a neutral arbiter, still a state agency but one outside the human services bureaucracy. In Pennsylvania it’s the Bureau of Hearings and Appeals.  Only 955 cases reached the BHA in 2018. They ruled on 545 cases that year.

            If that name – Bureau of Hearings and Appeals -- is vaguely familiar, it’s because that agency was the subject of a bizarre rant last year by Auditor General DePasquale.  As I noted on this blog last year:

DePasquale claims that in hearing appeals, BHA upholds [county children and youth services agency] caseworker findings only four percent of the time.  As a result, he claims, “the system is potentially sending hundreds of children back into the homes where CYS caseworkers found they were being abused.”  Then he ratchets up the fear rhetoric with this:  “No child should ever have to go back into an abusive situation because bureaucrats have their heads buried in the sand.”
There are two problems with this: The four percent figure is contradicted by readily available data, and BHA decisions have no bearing on whether children are removed from or returned to their homes.

            BHA decisions apply only to whether someone stays on the state’s child abuse blacklist, nothing more.  Whether children are removed from the home and where they wind up is decided exclusively by county child protective services agencies and courts.  DePasquale held a news conference and went on his rant apparently unaware of this basic fact.

            But DePasquale went further.  Veering toward conspiracy theory territory, he even suggested that there is something nefarious in the agency’s name: Bureau of Hearings and Appeals.  The more likely explanation: It’s called the Bureau of Hearings and Appeals because it holds hearings and listens to appeals.

            DePasquale made all these bizarre claims during a news conference announcing an “investigation” into the BHA -- as opposed to doing it the old-fashioned way, in which one conducts the investigation first and then draws the conclusions.

            As for that claim that BHA upholds caseworker claims only four percent of the time, nobody outside DePasquale’s office knows where they got it.  What we do know is, it’s wrong.

            The state Department of Human Services claims that BHA overrules county child welfare agency findings 46 percent of the time.  But I don’t know where they came up with that number either.  In fact, in the relatively few cases that make it to the BHA, it appears that BHA reverses caseworker decisions about 70 percent of the time.  (When you count all cases that are appealed, both to BHA and administratively – where nothing gets overturned – the percentage drops to under 25 percent.)

            The statistic concerning the BHA can lead to two possible conclusions: 

            1. The BHA is made up of a bunch of biased and / or ignorant administrative judges who love coddling child abusers and hurting innocent children!


            2. As soon as the accused, at long last, reaches an independent tribunal, it turns out the case was so flimsy and / or the investigation so slipshod that it isn’t even slightly more likely than not that abuse occurred. 

            Here are some reasons why Option 2 is more likely.

            ● Pennsylvania is not unique. In Kentucky between 56 and 66 percent of cases are overturned in a typical year. In New York and Illinois class-action lawsuits revealed that when people finally make it through the slow, cumbersome appeals process they win 75 percent of the time. (NCCPR’s vice president brought the New York suit.) 

            ● In Pennsylvania, the most serious cases never get appealed at all.  If the case results in a criminal conviction it usually can’t be appealed.  Of the rest, only parents with extraordinary determination can be heard before the BHA. That’s because there’s no right to counsel if you can’t afford a private lawyer.  You have to file your own appeal within a tight timeframe, go to Harrisburg and then you’re all alone against the might of the state.  In fact, as Ginzburg notes, you can’t even see the case file with the evidence against you.

            As Pittsburgh attorney Amanda Green Hawkins put it: “It’s like David taking on Goliath.”

● In North Carolina and Missouri state Supreme Courts found that putting someone in a registry first and then holding a hearing so stacks the deck that it’s unconstitutional. Several judges in Pennsylvania have   expressed similar qualms, quoting the Missouri decision with approval.

            ● As we noted last year, DePasquale has himself expressed alarm at the inexperience, overload, and high turnover among frontline caseworkers. Since these are precisely the conditions most likely to lead to errors in all directions, it’s no wonder the independent BHA finds error so often.

            But somehow I have a feeling that, when DePasquale gets around to issuing the report on the investigation for which he’s already drawn his conclusions, something reportedly expected in the summer, he’ll go with Option 1.

            And he already seems to be getting some support in a report that is biased on its face – specifically on page 13, where you’ll find this flow chart that purports to describe the appeals process. Take a close look and see if you can spot the bias. (Hint: It’s in the lower left quadrant).

            This new report doesn’t just discuss the issue of appeals. It’s a multi-front assault on due process. There's more about that in part two.