Source: U.S. Department of Health and Human Services, Child Welfare Information Gateway, 2018 |
The State Assembly holds a hearing Oct. 9 that could be a first step in the right direction
Last March, the New York City Family Policy Project, an essential resource, and not just for New Yorkers, published a comprehensive report on the harm done by the state’s child abuse hotline, which is run by the state Office of Children and Family Services (OCFS). The report found that New York screens out false reports at a far lower rate than most states. That causes havoc for families wrongly investigated, and deluges workers with so many false reports they have less time to find the few children in real danger.
Responding to the Family Policy Project report, the State Assembly Committee on Children and Families announced it will hold a hearing on the issue on Oct. 9.
I hope they will zero-in on one nearly unique element in the New York system that makes it particularly destructive: a built-in incentive for what should be called mutually-assured buck-passing. Only one other state has the structural problem New York has built into its process. I also hope the legislature will take a look at some relevant history, both recent and ancient.
The structural problem
In most states, the family police agency (a more accurate term than “child welfare” agency) is a state agency. That state agency runs the hotline and that state agency does the investigations and takes away the children. In 10 states,* the investigating and family-separating is a local government function. But in five of those states, including California and Minnesota, the localities typically run the hotline as well, while in three more, localities can screen most or all reports sent from the state hotline.
Only one other state, North Dakota, does it in the awful way New York does it:
In New York, all calls alleging child abuse and neglect go to the statewide hotline run by OCFS. They then send the reports they screen-in to localities to investigate. The localities have no choice – if the state hotline sends it, the locality must investigate it.
And it sure seems like, at least in New York City, they want to keep it that way. The reason for that has to do with safety – no, no, not safety for the children, safety for the city family police agency, the Administration for Children’s Services (ACS).Over and over again, when ACS needlessly investigates and
traumatizes an innocent family, the
agency’s commissioner, Jess Dannhauser,
says something like: We didn’t want to wreak havoc on this family; the state
made us do it!
The state, for its part, has an incentive to send huge numbers of reports to the localities. After all, if the hotline wrongly screens out a report and later there’s a tragedy, the state agency gets the blame. This may well be a key reason the New York hotline wrongly screens in so many cases.
Meanwhile, ACS speaks often of the need for better training so mandatory reporters don’t phone in false reports, (but since they’re mandatory reporters, they still may be afraid not to report) and ACS is piloting some supports in schools to encourage those reporters to seek alternatives to calling the hotline. But Commissioner Dannhauser has never publicly called for the Legislature to simply give ACS the power to screen out false reports sent to his agency by the OCFS hotline.
And no wonder: If ACS were to get that power, screen out a report, and then there’s a tragedy – well, you know.
Hence, mutually-assured buck-passing.
I figure Dannhouser is likely to testify at the Oct. 9 hearing. I hope one of the lawmakers asks him this question: You keep blaming the state hotline for forcing you to investigate reports you know are b.s. Why have you never asked us for a law allowing you to screen out those reports yourself? (And if the legislators don’t, I hope a reporter covering the hearing will.)
As I said, only North Dakota does it the way New York does it. There also are two hybrid states. In Colorado, there’s a state hotline and local hotlines. Most calls go directly to the local hotlines which, of course, can screen reports. Pennsylvania has a state hotline that passes on calls to local family police agencies. But, unlike New York, those local agencies have discretion to screen out almost all reports alleging “neglect” – which are, of course, the overwhelming majority of reports. Pennsylvania localities are not allowed to screen out reports alleging abuse or “severe neglect” – but this still is an improvement over New York’s approach.
Pennsylvania also teaches something else: Giving localities the authority to screen won’t work miracles. Both Philadelphia and Pittsburgh take away proportionately far more children than New York City – and Pittsburgh decided to screen in the worst possible way – using “predictive analytics.”
So why make the change? For starters, just taking away ACS’ excuse and making localities accountable for launching needless, harrowing investigations of families would be worth it.
But also: right now, New York localities have to investigate 100% of the reports sent to them by the hotline. If they get the authority to screen and wind up investigating 90% of the reports instead of 100% that’s still an improvement. And there’s nothing to stop OCFS from using predictive analytics at the state level should it so choose.
Even without a law, it is possible to change regulations to allow for a kind of “circuit breaker” in which localities would have the authority to stop an investigation at a very early stage if the report never should have been sent to them in the first place.
Also, there’s a more encouraging lesson from a county in Upstate New York.
The relatively recent history
At one time, two Upstate counties, Monroe (metropolitan Rochester) and Onondaga (metropolitan Syracuse) ran their own hotlines. In Monroe County that was true until 2015 when, in the wake of a child abuse death, the state took that authority away. Showing notably more courage than their New York City counterparts, Monroe County officials actually asked for that authority back. But, in a remarkably nasty response, the state OFCS said no – because, they said, when Monroe had its own hotline they screened out too many calls.
Well, OCFS sure “fixed” that problem! Once the state took over, the number of screened-in calls skyrocketed.
The ancient history
All the way back in 1987, the (long-since defunct) New York State Legislative Commission on Expenditure Review gave hypotheticals to 23 hotline operators, 12 at the state hotline in Albany, eight in Onondaga County and three in Monroe County. In each case, they asked: Screen-in or screen-out?
The results: Big differences between Monroe County, Onondaga County and state hotline screeners (at that time the Monroe County operators were far more likely to screen-in reports) and no consistency among the state hotline operators.
In one case, for example, all the Monroe County operators said screen it in, almost all the Onondaga County operators said screen it out, and the 12 operators at the state hotline split: 8 yes, 4 no.
There is no reason to think the hotline is any less arbitrary, less capricious or less cruel today.
Improving that will require a better screening tool and yes, operators will have to be trained in how to use it, and others will have to be trained to be sure they are using it correctly. But beware of the use of “training” as a copout. Because in addition to being invoked when it’s needed, it’s also invoked – endlessly – as a way to avoid real change. Training is no substitute for due process.
So if you’re in Albany on Oct. 9, please be extra careful on the roads. Because if anyone suggests a drinking game based on how often some family police official at the hearing uses the word “training,” you can expect a lot of drunk driving.
*-Nine states are fully local. I also count Wisconsin, since the state runs child welfare in only one county, albeit the largest, Milwaukee. I do not count Nevada, where the state runs the system everywhere except the largest county Clark County (metropolitan Las Vegas)