AMCHP 2005 ANNUAL CONFERENCE
DELIVERING RESULTS, IMPROVING PREGNANCY & BIRTH
February 19-23, 2005
MONIQUE SHEPPARD: Again, I would like to say that Dr. Overpeck is sorry she could not be here today. She is recovering, and as Stephanie said my name is Monique Sheppard. You've already heard presentations this afternoon about what is Child Death Review and how does that process affect us. You just saw the presentation with scene re‑enactments looking at the history and previous studies dealing with SIDS and suffocation.
Now I'd like to talk to you about the national data, particularly injury data related to this topic and focus on sleep‑related deaths.
SIDS, Sudden Infant Death Syndrome, as Dr. Kemp already stated, the original definition in 1969 did not include death scene investigation. That definition was changed in 1989, to include that component. And it is so critical, because it's the death scene investigation that allows us to differentiate between this cause of death and suffocation or asphyxiation deaths.
But the ambiguity still exists as Dr. Kemp and Ms. Tackitt alluded to in their presentations about who goes out to the death scene, who is looking at this data, how is this data recorded back to the medical examiner.
To fix some of these problems, in 1999, when the ICD‑10 provision came into existence in the United States , or came into use, a new category for infant death was added called sudden unexpected or unexplained infant death, or Sudden Unexpected Death in Infancy. SUDI, and this category was supposed to capture cases that may suggest SIDS but the information was not definitive to classify the death as a SIDS case.
So this would include roll‑over cases. This would include cases where a non‑lethal injury was present in the infant. This would include cases where previous sibling had died of SIDS. And so when you look at infant deaths, you realize that they are divided into two categories. Natural deaths, which are the SIDS or the SUDI deaths and traumatic deaths which are asphyxiation or suffocation deaths.
And so what we have seen over time is that there's been a shift in the diagnosis on death certificates at the national level. And we believe that this is due because more jurisdictions are requiring a complete autopsy. And many of those autopsies include information on the death scene.
Also, people who certify death certificates are starting to move from saying that these cases are SIDS deaths, they're starting to look into these other common causes, and everyone is recognizing that multiple mechanisms are involved. And two of those mechanisms that are very common are suffocation and strangulation.
When you look at the national data from 1991 to 2001, for all causes of infant injury deaths, you see that natural causes, the blue line, which excludes SIDS cases which are the red line and SUDI cases which are is yellow line are decreasing, which as well acids cases are decreasing overtime.
I pulled out the natural deaths out of this slide so you could really focus in on what was going on with SIDS, SUDI and general injury; and you can see the decline in the SIDS cases over this 11‑year period. You can also see, since the SUDI category was added in 1999, it's been steadily increasing, as well as injuries have been increasing since about 1998. Now the question is: Are these SUDI deaths, are these injury deaths a reclassification? Is it a shift in diagnosis from SIDS cases? And I'm not saying that this is the case, but if it were the case, the decline in SIDS is much greater than the increase we see in SUDI and in these injury deaths.
When we only look at injury deaths, and we see that you have unintentional intentional and undetermined deaths, we see that the intentional deaths are flat‑line. The undetermined deaths, which are the intent is not determined, it was not purposely, it is unknown whether it was purposely inflicted or not, has a slight rise in the time period we want to focus on. But we see unintentional deaths have been on the rise since 1998. So we want to focus in on these unintentional deaths to help explain these shifting patterns.
The remaining data slides are going to focus on the period from 1999 and 2001. This is because in 1999 the ICD‑10 was implemented. So what we're looking at will not be due to a diagnostic shift in coding. This will be due to the cases and the miss classification or possible miss classification between SIDS and injury deaths.
When we look at SIDS deaths, SUDI death and injury deaths, we see there's similar developmental patterns among these infants that die. Between the second and fourth month is the peak of all of these types of deaths. We also see that SIDS deaths decline around the ninth month, where unintentional injuries become the most prominent cause after the ninth month. We also see that during the first week of life that you see blips in intentional injuries as well as SUDI causes.
When only looking at the injury deaths over the course of the infant's first month in life, we can again see the pattern that we spoke of. We see that during the second month of life, even if you combine the unintentional and for the intentional causes they're homicide because infants can't self‑inflict injuries on themselves, even if you combine those causes at one week and one month of age, they would not be as great as the death rate that you see in infants for unintentional injuries at two months. The pattern is similar between homicide deaths and undetermined deaths, where you have a peak, not a peak, but you have a high‑rise at the first week, then they dip down during the first month, a rise again between the second and fourth months and then a decline over time.
Also for unintentional deaths you see that during the last, during the 12th month of life there's another slight rise. This is due to a child becoming a toddler, and those are mobility issues.
So trying to examine these deaths to try to get at nationally our picture of what is going on. If we take intent out of it and we just look at mechanism of death, what we see is that 60% of these injury deaths are due to suffocation. And that makes so much sense in light of what you just heard from Pat Tackitt and Jim Kemp, what is going on with these infants and their process of death. We also see that for transportation is the second leading mechanism of injury deaths, followed by assault, not otherwise specified, or NOS. That means there's no further information available from the death certificate to be coded into the data. Abuse and neglect, and drowning. We know for drowning deaths that those are related to a care giver having the baby in proximity to water. But we want to examine those suffocation deaths, which may or may not be related to SIDS or to SUDI deaths.
So we'll break them out again by intent. And what the data allow us to see, that for the top eight causes of unintentional injury deaths over this time period, four of them are related to suffocation. 30% of them are due to strangulation in bed. 12% of them are not otherwise specified. Again, the death certificate has no more information to help us tell what's going on. Eight percent are due to obstructive inhalation or suffocation, and the scene re‑enactment slides you saw help you see what those were about.
And then six percent are due to other suffocation, which is choking. For the intentional causes, we see that although child abuse and assault are the leading top two causes, suffocation is the third leading cause for intentional deaths in infants as well. We know throughout time since about the 1980s that undetermined deaths have typically been about four percent of all infant injury deaths.
So in summary, we see that there has been an increase in the overall injury rates in infant deaths. This increase may be due to diagnostic shifts in SIDS or it may reflect a true increase in infant injury deaths. However, our current ICD categories which are derived from the death certificate, do not provide enough information to let us know how to prevent these deaths. We don't have enough information that we have with Pat Tackitt and Jim Kemp talked about nationally. We don't know those things.
Better systems are needed to help us explain this at a national level. Child Death Review and the kind of information that it provides would help us to explain what is going on with this youngest group that we have.
There are national efforts that are moving us toward this. The CDC is revising its guidelines for sudden unexpected infant death investigations and on their website there's a sudden unexpected infant death information report form which is an eight‑page form which allows you, leads you through the death scene investigation, shows you what kind of information you should get, who you should interview that you should have scene enactment, re‑enactment, that you should get the child's medical history. There's a diagram so that you can draw the baby and how the baby was laying. Also, the National Violent Death Reporting System, also sponsored by the CDC, is looking into violent deaths for all age groups, including infants; but not only do they look at homicides but they look at undetermined injuries. Even though this group has only been four percent, having more information on this small group would help shed light on some of these other causes, perhaps the not otherwise specified categories.
And HRSA, the Maternal And Child Bureau is funding a Child Death Review resource center. And that resource center is another key way to help us find out more information. The data tool that Dr. Sloane spoke about that is being piloted in those 19 states will allow us to gather the kind of information that's not available nationally so that we can say if what's going on in St. Louis and Michigan is happening everywhere.
And speaking of what's available in states and how it allows us to talk about and learn from this, I'll now turn it over to Doug Paterson.