HRSA/MCHB 2005 FEDERAL/STATE PARTNERSHIP MEETING

PUBLIC HEALTH ACROSS THE LIFESPAN

Translating the Findings from Child Death Review and Injury Prevention Planning

MARY OVERPECK: Great. Thank you, Terri and Stephanie. I'm going to talk, give an example of how we have used data from child death review teams to address infant deaths occurring in the sleep environment. Contributors as you see here from children safety network and also from the teams all across the nation. And the reason we did this was that we had an appeal from our SIDS program in the Maternal and Child Health Bureau to bring some evidence to the table for a preconference at the SIDS alliance meetings related to the potential for a national crib campaign to reduce sleep‑related infant deaths and they needed risk factor data to assess factors pertinent for targeting appropriate audiences for crib campaigns and also to assess the potential for the success in reducing the deaths.

So why not use infant mortality rates? For one thing, the rates are based on a death certification by over 2,000 medical examiner and coroner systems in the United States. They do inconsistent investigations and cause classification of sleep‑related deaths, the cause classification from death certificates use international classification of diseases or the ICD, and the ICD, information on sleep‑related deaths are absolutely incomplete, inaccurate and not specific enough for us to address what we need to know for a crib campaign.

There's also other things that have been happening on death certification across the United States over the last ten years. To begin with, more jurisdictions are requiring mandatory autopsies of unexpected infant deaths, where they were not done before. And before they may have just been called SIDS but now medical examiners are beginning to be more careful and they're actually shifting designations to categories other than SIDS.

And this is partly because of the recognition that there are multiple mechanisms that are occurring in deaths that were previously considered to be SIDS. These may be infectious diseases. These may be congenital conditions such as cardio respiratory conditions or they may be metabolic conditions. And all these conditions, as standing alone, something is meningitis, can kill an infant. It has nothing to do with a sleep‑related environment. But if you have a child who is at higher risk for death partly because of prematurity, lack of development of the lungs or some congenital condition or from a metabolic problem, if you put them in an unsafe sleep condition or it could be just a transitory condition, and an unsafe environment, that particular circumstance may lead to some form of suffocation or strangulation that might not otherwise have occurred.

There's ambiguity in SIDS as a cause of death. Partly. If you really look at what the definition is of SIDS it's an unexplained cause after a thorough autopsy and investigation, the latter really doesn't happen very often.

There has been a shift to something that is now being referred to as a SUDI, Sudden Unexpected Infant Death. It was mentioned as an off‑shoot to SIDS in the ICD10 that was brought in in 1999. But essentially it recognizes the diagnostic shift away from SIDS to undetermined causes of death or unknown cause or over into suffocation deaths. So I think one of the things that particularly in maternal and child health world where we have looked at SIDS as the leading cause of post neonatal mortality in the United States over all these many years and the success of our back to sleep campaign, is that SIDS and SUDI have been generally classified as natural deaths while asphyxiation and suffocations are classified as traumatic and injury events. And we haven't talked to each other. The folks that work in natural diseases think that's the place we will need to be, aren't talking to injury folks. The injury folks aren't thinking about natural deaths so we are just looking at different circumstances but frequently they're actually the same causes of death.

So this slide actually shows what's happening since 1991 in the SIDS deaths going from the ICD 9 to the ICD 10 classifications, and my laser isn't working on this again. But there was a paper by Dr. Michael Malloy that was published in May of this year in which he showed that if you look at all unexpected infant deaths that have occurred since 1992, there actually has only been a 10% decrease in the number of deaths. I'm going to repeat that. There's less than a 10% decrease in the number of unexpected infant deaths. And a lot of what's happened over this time period is that the medical examiners have begun to shift their diagnosis into other categories, either the injury deaths or into undetermined, unknown or SUDI classifications.

If you go to the normal sources from death certificates on in terms of the cause and mechanism of death for infants, about 60% of the deaths have been classified as suffocations. If you look over here, it's not showing up very well on the bottom. But next slide is transport‑related and the following two slides are either assault or abuse and neglect. But within the assault and abuse and neglect we have absolutely no information about what the actual cause or mechanism was, it's just a classification we don't know how often the medical examiner or coroner determined that the neglect may have been the result of unsafe sleeping conditions.

So what's going on if we look at issues related to suffocation and as asphyxiation which is what we're considering when we talk about unsafe sleep conditions. This is based on the death certificates unintentional injury deaths have shown a definite increase since 1992. Those that were considered intentional have perhaps begun to decline and that's probably due to the diagnostic shift, at the same time those that have been considered undetermined from cause have increased. This again having to do with the diagnostic shift.

The real issue for sleep related deaths is that almost all of the sleep‑related deaths are occurring or all of the SIDS and SUDI deaths are occurring prior to the fifth month of life. We're talking here developmental issues for infants, these are infants particularly in the first couple of months where that peak really goes up, they're the ones who cannot really lift their heads, roll themselves over and remove themselves from an unsafe sleep environment because of their development stage.

So what we need to address issues in these environments is looking at the supportive investigations, including the infant death scene investigations that Terri mentioned. Also scene reenactment for infants death occurring in the sleep environment. There's very few jurisdictions doing these scene reenactments, but it's what we need. And of course the child death review process that brings all the information to the table from other sources. These are pictures that part of death reenactments that were offered to us. These are parents that say you can use our pictures but they're confidential.

Here's a mother who had her arm laying across the infant, slowing down the respiration in the sleeping environment. The infant stopped breathing. We have a mother here who held her baby as it was sleeping, it was sleeping, the infant's face into the pillows and blankets.

Again, a mother sleeping with the infant. The infant got caught up in between the mother and bedding; a lay over death where both parents were sleeping with the infant and the father laid back on the infant as he sleeps. This is a breast feeding mother. She had been breast feeding and fell asleep on her chest and the infant is found then having rolled off and is caught between her and the couch.

Infant that has been pushed off of the adult mattress during sleeping and the couch. Now, here's a safe sleeping situation. We're talking about very young infants, many of whom haven't moved yet into larger cribs, but here is a bassinet that is considered to be safe because it has measures around the edges it allows for breathability. So here's an infant in a safe bassinet, but because of blankets in the environment, the infant ends up being in an unsafe sleep environment. These are pictures from death scene reenactments of actual infants who have died.

So when Paul Rosinco called and said that there was going to be a preconference at the SIDS alliance, he said what evidence can you bring to the table? And so I immediately went to the Child Death Review Center and said can you ask them to send what they have based on information that we had said we wanted to know in this new data review tool. And of course it's in its beta testing, but we do have a standardized report format for sleep‑related deaths as far as that tool. In addition to those participating in beta testing that other CDR teams that review infant deaths and document sleep‑related deaths. So we have a mix of teams who reported to me. But we asked for this information in August, and by Labor Day we had 12 teams across the states who replied.

We asked them if they could provide information on items related to the sleep‑related surfaces in use at the time of death, the location, in other words, whether it was the home or someplace else besides the home where the infant was found dead. Who was the primary caretaker for the infant at the time of the death, and what were some of the multiple risk factors for sleep‑related deaths, the 12 states that replied to us in the period of about three weeks they just did great stuff. I was stunned at that ‑‑ they really wanted to address this problem and be part of this.

The questions that we asked the CDR teams on sleep‑related deaths were how many infant deaths were reviewed in your state in 2001, 2003. Now, this is how many were reviewed. Some of them represent these years 2000 and 2003 but some have reviews that occurred as much as two years later. So the deaths could have been from previous years. Some of them had really good data on this for maybe two years because they had begun to be involved in the data tool development.

Out of about 6,400 deaths they do have these varying review time periods and also completeness of records among the teams related to this topic. So different states reported on different issues. But we define sleep‑related deaths as those occurring while the infant was sleeping or in a sleeping environment. If you think about an infant less than five months of age, most of those are usually in a sleeping environment. So of those sleeping infant deaths reviewed how many were sleep‑related. They came back about a third were sleep related, and probably the number is even higher because of the age of the infants.

So the questions: Of the deaths being sleep‑related, what was the cause by age and we asked the teams to, the responses were affected by the review criteria. Most review all infant deaths from SIDS, suffocation, strangulation, other traumas and deaths undetermined from cause and intent. Fewer review the other medical conditions that I mentioned at the beginning that can be contributing factors to the sleep‑related deaths. 12 states reviewed SIDS. 11 states reviewed suffocation and strangulation. Only five reviewed medical conditions. Eight reviewed deaths undetermined for cause and intent; and seven reviewed the others.

What about the age of death? Nine states reported that about a third of the infants died during the first month of life. 40% died from the second to third month. So we have about three‑fourths of the deaths occurring in the first three months of life. The questions raised from that are most deaths occurring before the age when most infants are placed in cribs. Are there more found in bassinets and other surfaces in younger ages than would be found if you looked across the whole year of an infant's life. What sleep surface was being used when the infant was found dead? 39 percent of those deaths were on bed mattresses. 23% were in a crib or a bassinet. And 15% were in a couch or a chair.

Where did the death occur? 78% occurred in the infant's own home, and because of the different ways the states collected the data, 22% occurred in other places of which could either have been a relative home, a grandparent or in a child care environment. Although they didn't have the information on what type of child care.

What was the relationship of the caretaker at the time of death? And 80% of the cases it was a biological parent; 5% was another relative, and 11% was a child care provider.

How about co‑sleeping, bed sharing? This terminology is not clear out there. We're beginning to understand that the issue is bed sharing, not co‑sleeping, which may mean they're sleeping in the same room, but as we sent this out, people are using the term somewhat interchangeably and not without a good definition. But at least 46% of almost the 1500 deaths in the nine states involved either co‑sleeping and/or bed sharing, and they didn't know in the remaining deaths.

Then we asked if they were co‑sleeping, were they sleeping with adults? And about 80% of the infants were sleeping with at least one or more adult.

And this result is definitely a result of the fact that some of information is collected only for SIDS diagnosis and bed sharing rather than for co‑sleeping. And I have them mixed in here. So this can also, the first question on the co‑sleeping bed sharing may be infants who are simply sleeping in the same room but not sleeping with an adult.

So the final thing that we looked at was the multiple risk factors. You saw the bedding and, of course, the Back To Sleep Campaign addresses the prone position. Now, for an infant, a prone position may be either where the fact that the position, the infant was laid down face down, but if the infant is strong enough it will roll itself over. By three months some infants can extract themselves and they can be found in a prone position at the time they die.

We have blankets and pillows involved in the deaths in 30% of the cases. And we have, in almost half the cases, the infant was sleeping with others. And, of course, you have here, as you saw in the pictures, that they're sleeping on an adult mattress, you have pillows, soft mattresses, you have bedding. So you can have any number of these factors working together to produce an unsafe sleep environment.

So in summary on this, the death certificate information is not accurate for assessing the number of SIDS or sleep‑related deaths of infants. So we turn to the child death review teams to provide new information on the sleep‑related deaths and the risk factors. And we really need more information from the infant death scene investigations and death scene reenactments to assist the death certifiers prior to the notification. We need to know it prior to the medical examiner and corner decide what to call the death.

But in conclusion, in terms of crib‑related evidence, the data systems aren't sufficient to evaluate how often the cribs were available but not used in the home. And there are a number of jurisdictions, either states or localities, who are already distributing cribs of various sorts, either at the hospital when the parents leave the hospital or through other programs; but we do not know why the parents were not using the cribs. We don't know how often it was a breast feeding parent who in order to encourage breast feeding is more likely to have the infant in the bed with them.

So we don't know that. We also don't know the locations of the deaths when the cribs were not available. So if they weren't available, we don't know where else the infants were sleeping.

And we don't know who we need to target for the use of cribs, bassinets and safe alternative services, but we're beginning to see that definitely parents, relatives, caretakers, these are definitely folks we need to target. So there have been ‑‑ there will be recommendations coming out of the preconference that was sponsored by the Maternal and Child Health Bureau at the SIDS Alliance meeting, and I'm sure we're going to hear more about that. But most of all I think what the real point of this is that we can gain a lot of information from our child death review teams. They're really anxious to do it and prevention in their own community is going to come from the fact that they have this information to take back into the community to support the use of a safe sleeping surface, perhaps a crib or a bassinet.

And thank you.