The Future of Fatality Prevention

As most readers likely remember, CECANF was created from a  2013 bill that empowered the commission to conduct listening sessions around the country and develop recommendations on how to prevent maltreatment-related deaths.

It became quite the contentious process, with two of the commissioners refusing to sign off on the final product. Among the primary CECANF recommendations:

• A five-year, state-by-state review of child deaths, followed by improvement plans, all supported in part by a $1 billion increase in funding for HHS.
• Passage of the Family First Prevention Services Act, a bill that would permit states to tap into federal IV-E funds to serve birth families while limiting state access to federal funds for congregate care.
• Improving the ability of child welfare, health, education and law enforcement to share information about children and families.

Templeman noted that the Department of Health and Human Services had endorsed many of the executive recommendations, including the featured proposal of the feds working with states on a five-year review of fatalities. The goal there would be to develop state-by-state plans to prevent maltreatment fatalities.

Youth Services Insider’s read of HHS’ response was that it agreed such a review could be done, but that the agency hedged a bit on how valuable it would be. From the HHS response:

Because child maltreatment fatalities are a low incidence event, the development of a national standard is problematic. States frequently have significant year-to-year swings in the number and rate of fatalities. In small states, a single incident rather than a systemic issue can dramatically affect annual statistics. In addition, in small states an analysis of data from the past five years … would include too few cases to draw definitive conclusions.

There was significant wariness in the room about the CECANF report, which opened with a statement from CECANF chairman David Sanders that “we believe we can reverse the assumption that some children will die from abuse or neglect.”

On its surface, it seems like the kind of thing you see written in lots of child advocacy products. But Sanders’ opening salvo presented the thesis that it is possible to build a systemic response that can prevent child fatalities caused by maltreatment.

There are 100-plus CECANF recommendations, and they really add up to this idea: create a different systemic response for kids deemed at risk of fatality than for those kids who are deemed to be at risk for being abused and neglected.

And the recommendation is that a different response be crafted through the lens of a five-year review, which is probably why HHS threw some caution out on the subject. The reviews could shape a very significant new subset of actions in child welfare, and they might be based on reviews of a very small size of events.

There is little evidence at the moment to suggest this would work. As CECANF’s report notes, half of the kids who die of maltreatment are not known at all to the system, which makes it hard to imagine a systemic response that could eradicate maltreatment death.

The potential to save lives makes it compelling to try. On the other hand, those wary of such a process rightly fear the potential inclusion of families in child protection for hotline calls that might be screened out under current practice.

A few of the Wicked Problems attendees made known their concerns with that notion, and most boiled down to this: It’s dangerous to say you can eradicate something, and then have it happen again.

“In New York, I could see someone picking up [the CECANF report up] and saying, ohhh … why aren’t you doing that?” said one attendee from New York, whose name YSI did not catch.

Jim Purcell, head of New York’s Council of Family and Child Caring Agencies, argued that distinguishing between accident and maltreatment is itself a folly that presumes maltreatment is intentional.

“These are stories of multiple-injured people doing unpredictably dangerous things,” Purcell said. “Firefighters don’t go around saying, we’re preventing all fires. We go around saying it, then people are like, ‘This commission is saying you can do it.’  We need to be cautious in our data, and our communication.”

Warren Ludwig, an attendee who once led a fairly large system with about five child fatalities in an average year, suggested something YSI had not heard yet: why not include near-fatalities with child fatalities, and review that?

Earlier in her presentation, Templeman said that CECANF had heard estimates that for every fatal hospitalization due to abuse, there are 10 non-fatal cases.

If that multiple is accurate, Ludwig, said, “then that gives you large enough number. You could begin to see whether things you thought should work are making a difference or not.”

Templeman said that since the release of the report in the spring, “we have heard a lot of energy on doing something differently” including near-fatalities because “it’s more countable.”

There is no doubt such an approach would build a more legitimate data bank. But one key issue that would need to be resolved is the integrity of near-fatality classifications. When a child dies, that is a finite and indisputable event. And even then, one doctor or child welfare administrator might disagree with another about whether the death was accidental or related to maltreatment.

Disagreement on what constitutes “near-fatal” is surely murkier. So to build and then act on a multi-year review that included them, a system would have to develop a firm litmus test for what counts.

If and when fatality reviews provided local patterns of factors in child maltreatment deaths (and perhaps near-deaths), the next step would be to figure out how to respond. Former Illinois child welfare director Ernie McEwen, now with Casey Family Programs, made it clear what the response should not be: more investigations and removals.

McEwen cited CECANF’s embrace of researcher Emily Putnam-Hornstein’s finding that regardless of case outcome, substantiated or not, a single report to child protective services is the best predictor of child fatalities. There were 5.9 million such reports made last year.

“We can’t do CPS on that many families,” McEwen said. “We can’t think about old logic, we have to start to figure out, what does it look like now when the commission says we need eyes on kids?

CECANF’s biggest finding echoes the impetus for these Wicked Problems sessions: the commission found only one service model that demonstrably reduced child fatalities. That would be the Nurse-Family Partnership, a Colorado-based model that has proliferated in the past 10 years and pairs expectant new moms with nurses.

In the Chicago session, there was wide agreement about another idea not addressed by CECANF: a massive public campaign about co-sleeping, aimed at preventing rollover deaths. Sleep-related deaths are squarely in the murky box when it comes to fatalities; some systems track them as maltreatment deaths, others don’t.

“With rollovers, why is that a child protection issue?” one attendee mused. “Why isn’t that a public health issue? They’re not bringing a baby into bed to kill them. “They’re uneducated” about the danger.