Overview of a key 2011 story

That story would turn out to be the template for Innocents Lost

Sometimes headline writers get it wrong.  But in the case of the Miami Herald story of February 27, the headline writer conveyed perfectly the thesis reporter Carol Marbin Miller sought to foist on readers.  The headline reads: “State steps in less, and more kids die.”

Miller uses the horrific death of Nubia Docter, and the near death of her brother, to try to prove her point.  But the story (which appears in full, with our commentary here) is so rife with inaccuracy, conflated figures and outright distortion that it’s hard to know where to start.

But how about starting here: Nubia Docter’s birth parents did not kill her.  She was not killed because she was left in a dangerous home in the name of family preservation.  She was killed after she was removed from her home, after she was placed with foster parents and after those foster parents adopted her.

So how is it that Carol Marbin Miller seems to think the problem is that DCF hasn’t been taking away even more children?

As for the data:

● If you want to know if a state has a policy to take more or fewer children away from their parents, and if that policy is being carried out, you have to look at entries into care, that is the number of children taken away from their parents over the course of a year.  Miller chooses to ignore this figure in favor of what is known as the “snapshot number” - the number of children in foster care on any given day, a figure which can rise or fall for a variety of reasons unrelated to attempts at family preservation.  For example, a lot of the reason for the decline in the snapshot number in Florida has been an aggressive push to get more foster children adopted.

We believe Miller ignores the entry figure because it refutes her thesis.  For starters, she doesn’t go back far enough in comparing entries to fatalities.  Elsewhere on this site we present those data all the way back to 1995.  They show that back when Florida was taking even fewer children than it takes now, there were fewer deaths of children “known to the system.”

● Entries into foster care soared by 50 percent in Kathleen Kearney’s first year running DCF, 1999.  It was the worst statewide “foster care panic” we’ve ever seen. And removals stayed at that obscenely high level all the way through the years of Kearney’s first two successors, Jerry Regier and Lucy Hadi.  So while Miller uses Regier’s rhetoric to claim Florida embarked on an effort to take fewer children starting as early as 2003, she ignores the reality.  It didn’t happen.  Entries remained at roughly the same obscene level all the way through 2006.   The actual figures are two mouse clicks away.  This table gives the figures for entries from 2002 through 2009.  This one gives the figures from 1999 through 2001.

So those increases in deaths “known to the system” Miller keeps talking about in her story started when the foster-care panic started under Kathleen Kearney, and they kept on increasing all through the years Florida was taking away more children.  Again, full details are available here.

● The other crucial change occurred in 2006.  The Florida Child Abuse Death Review Committee effectively broadened the definition of a maltreatment-related death, exerting pressure on local law enforcement and DCF offices to label every drowning death as maltreatment.  That led to a sharp increase that year and every year since.  That’s why data show what seems to be a big increase in 2006.

But in 2006, Florida still was taking away children at the same high rate as 1999.  The reforms didn’t start until 2007.  And, as a table accompanying Miller’s own story shows, there is no pattern in death rates from 2006 through 2009.  Indeed, deaths of children known to the system in Florida declined by 12 percent in 2009 compared to the previous year.  During that same year, entries into foster care declined by 10 percent.

Although Miller quotes from the latest Review Committee report when it suits her, she ignores the fact that the Review Committee itself acknowledges that such changes make it impossible to accurately compare death numbers in the way Miller does in her story.  Here’s what the review Committee itself said, but Miller declined to share with her readers:

Definitions and procedures may change within a state over time, as well as statutory requirements for reporting, as they have in Florida, resulting in data that is difficult to compare across years or states.

“Florida has a very active interdisciplinary State Committee administered by the Florida Department of Health. Committee members proactively work within local communities to educate the public and law enforcement on the importance of reporting to the Florida Abuse Hotline child deaths that may not have previously been regarded as the result of maltreatment.”

In short, there is not a shred of evidence that recent reforms have increased child abuse deaths – and some evidence that the foster care panic from 1999 through 2006 did increase such deaths.  However –

● Child abuse deaths are, in fact, an unreliable measure of safety, for a reason for which we all should be grateful.  Though each is the worst form of tragedy, thank God the number remains low enough to rise or fall due to random chance or subjective judgments (was that drowning an accident or neglect?)  That can have an enormous impact on rates of maltreatment.  Indeed, the one and only pattern seen across state after state is that deaths of children “known to the system” tend to increase in the wake of foster-care panics – because workers are so overloaded with children who don’t need to be taken away that workers have less time to investigate any case carefully.

So how do you know if a system is getting safer?  The best method is to have independent evaluators looking over everyone’s shoulder making a neutral assessment.  Florida has that.  Florida’s waiver from federal funding rules requires such evaluations.  And those evaluations have found that, under the reforms, child safety has improved.

● It is, in fact, impossible to compare child abuse fatality rates across the nation.  In part that’s precisely because of the reasons DCF mentions in the brief space it is given to respond in Miller’s 2,300 word story: different states use different definitions of when a fatality constitutes maltreatment – and those definitions can change from year to year, just as they did in Florida in 2006.  And, as noted above, the Florida Child Abuse Death Review Committee agrees with this assessment.

In addition, some states do a far better job of reviewing child maltreatment deaths and some states have far better systems of medical examiners or coroners to determine a cause of death accurately.  Phony attempts to compare states only penalize those states that are aggressive in investigating child fatalities and careful about determining cause of death.

Even the national group that has tried to exploit child abuse fatalities to encourage a take-the-child-and-run approach, the group that calls itself Every Child Matters, has flat out admitted that an accurate comparison of rates of child abuse deaths is impossible.  As ECM’s director, Michael Petit has said:

“We emphasize all over the place that it is impossible to compare states because of the different definitions, and we’ve been encouraging congress and HHS to establish measurable and comparable standards between the states.

See also the report from the Center for Public Policy Priorities, a liberal think tank in Texas with a long history of supporting more removal of children from their homes.  Even they admit that such comparisons are impossible, and they say, the traditional responses of agencies like DCF - screening in more calls, substantiating more cases and taking away more children - do nothing to reduce child abuse fatalities.  What does work?  More child abuse prevention programs and more efforts to curb teen pregnancy.  Their report is available in a link from this post to the NCCPR Child Welfare Blog.

● Miller compounds the error when she uses this table in a 2008 federal report to attempt to compare the proportion of child abuse fatalities among children “known to the system” in each state.  She neglects to mention that 17 states, including New York and California, did not report this figure to the federal government at all.  (Reporting for this particular federal document is voluntary, it is subject to error by workers in each state who do not need to give this task the same priority as reporting to a separate federal database for which reporting is mandatory).  But as long as Miller insists on using this chart, she has some explaining to do.   According to the chart, Illinois, which takes children away at roughly half the rate of Florida had no deaths of children known to the system at all in 2008.

● Miller also resurrects a series of claims about the Florida child abuse hotline.  Claims we refuted at the time on the NCCPR Child Welfare Blog.  Those claims also were rife with error, including a basic, but crucial, error of fact concerning a child’s age, which Miller still has not corrected.

● And finally, back to the actual case of the children involved in the latest tragedy.  If you want to know what really contributed to this tragedy, including the role played by Florida’s history of taking away far too many children needlessly, check out the far better reporting in the Palm Beach Post.