Tuesday, October 5, 2010

Dept of DUH: Big new study finds CPS doesn’t work

            Brace yourselves: this is bound to come as a shock.  A big new academic study, published in the Archives of Pediatrics and Adolescent Medicine finds that when you send a child protective services investigator into a home where there’s already a lot of stress simply because the family is poor and then pull the children aside and ask traumatic questions, possibly throw in a stripsearch, and inspect every nook and cranny of the home to see if it passes some caseworker’s white glove test – it doesn’t make things better, and sometimes makes things worse.

            They had to do a study to figure this out?   In fact, as I told Time magazine’s Healthland Blog, this study simply confirms what NCCPR has been saying for years: Child Protective Services won’t be effective until it becomes Child Poverty Services. 

            But contrary to what one of the study authors suggested this morning on WNYC public radio, that doesn’t mean you have to eliminate poverty to eliminate child maltreatment – though whoever does the first will come closer than anyone else to doing the second. You can make enormous strides simply by ameliorating the worst effects of poverty. Some examples are in our publication Thirteen Ways to do Child Welfare Right. (The WNYC interview is worth listening to for the comments of the other guest, Mike Arsham, executive director of New York’s Child Welfare Organizing Project.)

            Of course, the scenario I described above in which the investigator interrogates, stripsearches and then goes away, is far from the worst that CPS can do.

            In other cases, the parents will be forced to jump through a series of hoops either to keep their children with them or to get them back.  It’s almost always a cookie-cutter “service plan” almost always requiring lots and lots of “counseling” and “parent education” while the actual problems of poverty are ignored.  So the “services” only add more burdens to this family.

            At worst, of course, the caseworker leaves with the child, throwing that child into foster care, and probably doing at a minimum, serious psychological damage to that child.  Then, if the child is returned, the family has to heal from this, while it is still just as poor and just as stressed out as it was in the first place.

            So why, exactly, is anyone surprised that this doesn’t work?  It is a testament to the love and resilience in many poor families that only one of the many variables measured in this study got worse.


            One of the more feeble attempts to defend CPS came from one Janice Warren, DSW
at the University of Virginia in Charlottesville, who told MedPage Today:

"Many children are taken out of very dangerous living situations permanently based upon investigations conducted by CPS.  CPS serves an essential front line of protection for the most vulnerable children in our society,"

            In fact, very few of the 250,000 children taken from their parents every year are taken from “very dangerous living situations.”
            That is clear from the results of many other studies, most notably two that compared more than 15,000 typical cases seen by CPS workers.  Those studies found that children left in their own homes typically fared better even than comparably-maltreated children placed in foster care. That was true even when CPS agencies provided families with little or no help. 

            In other words, when CPS traumatizes a family and then goes away, it doesn't do any good.  When CPS traumatizes a family and compounds the trauma with needless foster care, CPS goes from doing no good to doing a great deal of harm.

That doesn't mean no child ever should be taken from his parents; it doesn’t even mean CPS should be abolished.  We need an agency to protect the relatively few children in real danger.  Rather it means that foster care is an extremely toxic intervention that must be used sparingly and in very small doses.

It also means that the threshold for initiating an investigation should be higher than an anonymous call to a child protective hotline.  After all, if a big study of the fire department found that, after they left a house, it almost always still was on fire and sometimes the fire was worse, wouldn't we at least try to do more to curb false alarms?


            The study was accompanied by an editorial which recommended some solutions.  The editorial argued that law enforcement should handle allegations of “abuse” on grounds that all abuse is a criminal act, and public health nurses should investigate “neglect.”

            But there are problems with arbitrary distinctions between abuse and neglect.  For starters, you’re going to ratchet up the largely pointless sideshow debate about corporal punishment if you start sending the cops in every time a parent is accused of spanking a child too hard.  Conversely, though they are very rare, there are forms of neglect, like deliberately starving a child, which are criminal – and much more serious than some forms of abuse.  And what do we do when there are allegations of both, either against the same child, or even an allegation of abuse against one child and neglect against a sibling?

            As for turning over some cases to law enforcement, nearly a decade ago, the State of Florida asked County Sheriffs departments to take over the investigative function of CPS workers.  A few counties agreed.  But nothing changed.  The Sheriffs investigators made the same mistakes as the CPS workers, and rates of child removal generally didn’t change. 

            But there is a better way to achieve the goals the editorial is getting at: “Differential response” in which when the allegation of maltreatment, whether abuse or neglect, is less serious, the response is an offer of voluntary help.  Differential Response is discussed in this previous post to this Blog.

Differential response is somewhat different from what the editorial proposes.  The editorial proposes using public health nurses, apparently based on the Nurse Family Partnership program.  That is a very good program, but it is a primary prevention program, and it is strictly voluntary.  In the cases discussed in the editorial, there’s already been a report to a child protective hotline, and it is not clear if the editorial contemplates families having a right to say “no” when the public health nurse comes to the door.


            There also are some disturbing errors in the article itself.

            ●Right at the outset, the article declares that “A CPS investigation, regardless of outcomes, signals a household at risk.”  No, it doesn’t.  It might mean the household is at risk, it might not.  Given how easy it is to start an investigation it may just signal a household being harassed by a neighbor or a spouse harassing a spouse.  Given the way calls to hotlines spike after a high-profile case is in the news, it might simply be that there was such a case in the news, and some well-meaning neighbor, or someone who saw the family in a store and didn’t like what he saw, overreacted.

            Indeed, this University of Minnesota study (another study showing that comparably-maltreated children fared worse in foster care than when left in their own homes) also suggests that there may be very little if any difference between the impoverished families who come to the attention of CPS via neglect allegations and those who don’t.  It may be little more than random chance.

            ● There also is a claim that “Family preservation … services after investigation are not associated with reductions in repeat maltreatment or foster care placement.” That claim is flat wrong.  Indeed, few interventions have a stronger “evidence base” than real Intensive Family Preservation Services programs.  That evidence of success is summarized in our Issue Paper on that topic.

That the claim of ineffectiveness even is still around reflects the persistence of bias in child welfare scholarship discussed previously on this Blog here and here.