How to NOT fix abuse in group homes and institutions in 8 easy steps. Lessons from Massachusetts and elsewhere

Steps already completed are checked: 

_x_Take away children at a rate far higher than the national average, creating an artificial “shortage” of foster homes. 

_x_Due to “shortage,” use group homes and institutions, the worst form of “care,” at a rate that’s also well above the national average, creating conditions for rampant abuse of children in group homes. 

_x_Newspaper does an excellent job exposing rampant abuse of children in group homes but does not mention the high rate of removal. 

_x_People most responsible for, over decades, creating the climate of fear and demonization of families leading to the state’s high rate of removal, are quoted extensively condemning the abuse, while taking no responsibility for creating those conditions. 

__Lawmakers express shock and outrage and promise sweeping reforms and/or hearings (all duly noted in the traditional follow-up story). 

__Reforms consist entirely of increasing inspections, and screening and training, the same promises made the last time this sort of abuse was exposed, and the time before, and the time before that. This time, they might throw in closing the “age of consent” loophole. 

__Attention fades, and the state continues to take away children at a rate well above the national average. The new training, screening and inspections change nothing. 

__The entire cycle repeats. 

That’s the Massachusetts version. Something remarkably similar is playing out in Alaska.

 Of course, there is one way to break this cycle …

The latest AFCARS data are just out: No change nationally, but some bad news from a few states

The federal government has released its annual update to the Adoption and Foster Care Analysis and Reporting System (AFCARS), a database that attempts to track, among other things, entries into foster care, exits, and the number of children trapped in foster care on Sept. 30 of each year – known as the “snapshot number.” The data are labeled preliminary, as they are every year when first released. Though it's rare, in a few states revised numbers can be a few hundred higher or lower.

Nationally, when compared to FFY 2024, there is virtually no change in the entry or snapshot numbers. Either there was a slight increase or a slight decrease. We don’t know which, because this year, for the first time since 2022, Wyoming and Washington State finally got their acts together and submitted data. They were not counted in 2023 or 2024. So whether entries slightly increased or slightly decreased probably depends largely on what the actual numbers for those states were in 2024. The only thing we can be sure of is the “slightly” part. There may, however, have been a real, and disturbing, decline of roughly 10,000 in the number of children exiting foster care. 

But, of course, the national figure hides wide variation among states. 

Here’s some of the bad news: 

They're cranking up the foster-care-to-prison pipeline in Missouri

There was a nearly 19% increase in entries in Missouri, a state that already had a high rate of removal. The increase may be the largest of any state in 2025. It's probably is due to a big change for the worse in leadership. When Darrell Missey ran the state child welfare agency, he tried to reduce entries into care. He was pushed out and replaced by Sara Smith, who has made her fanatical take-the-child-and-run approach abundantly clear. The increase in entries in 2025 wiped out the gains Missey made in 2024. 

There also was a disappointing 7.5% increase in entries in Texas; but entries there still are well below the level before groundbreaking reform legislation took effect. 

And there was a nearly 17% increase in entries in Maryland, probably the result of the usual wretched response to high-profile tragedies. The fact that one of those tragedies involved a foster youth who committed suicide after being taken away and dumped in a hotel doesn’t seem to have given anyone in Maryland government second thoughts about taking away more children. That, unfortunately, is typical. 

And then there’s Kansas. I’m holding off saying anything about this one, because in that state, there’s either been a giant increase in the number of children placed in foster care – or a giant data glitch. I don't know which. The state says it's the latter and the federal figures involve double-counting. I have inquiries out to federal officials, and I will update this post when I find out more.

And in case anyone in these states actually thinks this is something to celebrate, here’s one more reminder of what the research tells us about the multiple studies showing that, in typical cases, children placed in foster care typically fare worse in later life than comparably-maltreated children left in their own homes. And here’s one more reminder of all those studies showing high rates of abuse in foster care itself. 

And the good news 

There are several states that have shown commendable decreases in the number of children torn from everyone they know and love – but I’m not going to highlight them here. That’s because as soon as a state or county becomes known for dedicating itself to sparing children the enormous inherent trauma of placement, and the high risk of abuse in foster care, it’s like painting a target on the backs of the system’s leaders. 

Those wedded to the failed take-the-child-and-run approach that has destroyed so many children’s lives bide their time until the next child abuse tragedy involving a child “known to the system” in that community. Since no jurisdiction can prevent every such tragedy, no matter how few – or how many – children they take, there’s always going to be one. (Indeed, a massive study finds no relationship between how many children are taken away and child abuse deaths.) 

Then those who are sincere but mistaken, those who are just grandstanding politicians, and those who, I suspect, deep down, just don’t want poor people, especially poor people of color, to be allowed to raise their children, come out of the woodwork, fingers wagging, to claim that the tragedy supposedly shows that the system is placing family preservation ahead of child safety. Case in point: Santa Clara County. Journalists who, at best, believe this because it sounds right and it fits their own stereotypes about parents who lose children to the system or, at worst, are just Pulitzer-sniffing, rush to amplify the false claims and shut out dissent. Case in point: Santa Clara County. 

So while anyone, including those who will misuse the data, is free to check the AFCARS database and do their own comparisons, I’m not inclined to assist in that process.

Foster-care panic in Silicon Valley: Tragically, in Santa Clara County, California, children DID die of child abuse before 2023.

A page from the 2018 report of the Santa Clara County Child Death Review Team

● A propaganda campaign by the San Jose Mercury News and a grandstanding member of the Board of Supervisors has been so effective, not everyone seems to know it. 

● Both the very limited available data from Santa Clara County and a massive peer-reviewed national study find no evidence that taking more children reduces child abuse deaths or that taking fewer children increases such deaths. 

See also our previous post about the Silicon Valley foster care panic.

It shouldn’t really be necessary to point this out. But judging by some statements during the public comment period at a recent meeting of the Santa Clara County Board of Supervisors, it appears there may be people who think that children only started to die of child abuse in that county toward the end of 2023. 

The “journalism” from Julia Prodis Sulek of the Mercury News, and the comments of County Supervisor Sylvia Arenas, certainly leave that impression. Though they presumably jumped to their conclusions with good intentions, they have wrongly scapegoated very recent efforts to keep families together for three recent deaths. They are undeterred by the fact that one of those deaths occurred well after those family preservation efforts were abandoned, while entries into foster care skyrocketed as a result of pressure from the Mercury News and Arenas. 

They’ve been aided in their propaganda campaign by County District Attorney Jeff Rosen. In response to the latest tragedy, Rosen has pledged to investigate who, besides the actual killer, “is responsible criminally, civilly, morally, ethically, and systemically for what happened in this case. And I think that we should all be asking questions of county officials at the highest level…”  It turns out Rosen has another problem with at least one county official at the highest level. Rosen wants to take funds desperately needed by the county’s health care system to offset federal Medicaid cuts, and divert some of that money to his office. The County Executive disagrees.

Of course, no one literally said that there were no child abuse deaths before 2023. Instead, those pushing the take-the-child-and-run response to recent fatalities engaged in what should be called “inference peddling.” We were left to infer that in all those years before the county tried to do more to keep families together, child abuse deaths didn’t happen.  Therefore, it is suggested, anyone in charge of child welfare in the county now or whose work brought them anywhere near the three recent tragedies should be fired and, ideally, tarred and feathered. 

But of course, children have died of abuse in Santa Clara County, probably every year going back decades if not forever. That’s true in every other jurisdiction of Santa Clara’s size – nearly two million people – and many that are far smaller. 

We don’t know much more than that because of a process for tracking fatalities in Santa Clara County that is subjective and inconsistent. 

For decades, the county has had a Child Death Review Team. It issues periodic reports, on no apparent schedule, in which it examines child deaths and assesses the causes. But determining whether a death is due to abuse or neglect is surprisingly subjective. 

Consider a hypothetical example: Early one Sunday morning, while his parents are asleep, a three-year-old wakes up, manages to unlatch the back door of the family home and wanders away.  He falls into a body of water and drowns.  Accident or neglect?  The history of American family policing suggests that if the body of water is the pool behind a McMansion, it will be labeled an accident. If it’s a pond behind a trailer park, it will be labeled neglect. 

The subjectivity is magnified for an ironic reason: Each child abuse death is the worst imaginable tragedy; the only acceptable goal is zero. But in most places, the number of child abuse fatalities is low enough that, depending on how even one or two deaths are characterized, it can suggest a pattern that does not exist or obscure one that does. Even random chance can play a role. If in one year, say, a father comes home drunk and shoots his wife and two children, that can create a “spike” in child abuse death statistics. 

Compounding the problem: The Santa Clara County Child Death Review Team sometimes neglects to note its assessment of cause of death. Other times, it does not mention the relationship between the victims and the alleged perpetrators. 

So, below is a graphic representation of what I was able to infer from the available data for most of the years from 2005 through 2023. But please read these caveats: 

● There were no data available for 2013 and 2014.

● The official report for 2010, 2011, and 2012 – one death in each year -- is much to be hoped for, but unlikely.

● From 2017 on, the reports gave totals for two or three years, without breaking them down by year, so the numbers shown in this graphic for those years are multi-year averages. 

Source: Santa Clara County Child Death Review Team Reports

So, everybody see the pattern? 

Exactly. There is no pattern. 

While it might be tempting for family preservation advocates to argue that the reported numbers were larger in the early years, when the number of children taken away was at its highest, the relatively low raw numbers and the unreliable reporting methods don’t allow such a conclusion. In fact, they allow only two conclusions: 

● Yes, children did die of abuse and neglect before 2023.

● There is no pattern concerning deaths and entries into foster care.

That should surprise no one – though given the propaganda campaign waged by Sulek, Arenas and their allies, it probably will. 

Once again, the reason is one for which we all should be grateful. In Santa Clara County in 2023, there were allegations of abuse and, far more often, neglect involving 12,036 children. There were approximately six known child abuse fatalities. That means of all the children known-to-the-system that year, at least 99.951% did not die of child abuse. The odds of finding an impurity in Ivory Soap are greater than the chances of a Santa Clara child “known to the system” dying of abuse or neglect. 

Again, the only acceptable number of child abuse fatalities is zero. But such fatalities are needles in a haystack. Here’s the haystack; look closely for the needles. And even this representation makes the needles easier to spot, because they’re all in one place: 

Sources: Investigations: California Child Welfare Indicators Project;
Fatalities: Santa Clara County Child Death Review Team Reports

By pressuring everyone to make the haystack far bigger, Sulek, Arenas and their allies have only made it less likely that the next child in grave danger will be found in time. 

Nationally, some defenders of the take-the-child-and-run approach have a bizarre response. They argue that these are not needles in a haystack because other causes of death – such as cancer – take fewer children’s lives. Of course, cancer tends to be a disease striking older Americans. I’m sure more children also die of child abuse than die of Alzheimer’s.  More important, the fact that other causes of death take even fewer lives means only that there are fewer needles; it is not a reason to keep growing the haystack. 

The only way to really know if there is any relationship between foster care entries and child abuse deaths would be if one could do a massive, peer-reviewed study looking nationwide over many years. Say, a study of 3.4 million records and more than 24,000 fatalities over 13 years.  Fortunately, that’s just been done. And, unsurprisingly, here’s the key finding: 

“Child maltreatment mortality rates did not appear to decrease with higher foster care entry rates or increase with decreasing foster care entry rates.”

That means, at best, all those Santa Clara County children torn from everyone they know and love in the name of reducing child abuse deaths may well have had their own chance for a loving, secure future taken from them for nothing. At worst, making the haystack so much bigger may have contributed to the latest tragedy in Santa Clara County, the death of Jaxon Juarez, and perhaps others to come. 

There are far better ways to work to reduce fatalities – and all other child abuse and neglect – in Santa Clara County. But step one is to consider the possibility that a “Pulitzer-sniffing” newspaper and a grandstanding politician and their allies are not the most reliable sources. 

See also our earlier post about the Silicon Valley foster-care panic.