Ninth Annual Maternal and Child Health Epidemiology Conference / December 10-12, 2003
HAROLD WEISS: This movie brings into focus several points that we’re going to look into today. One, the problem of serious injury during pregnancy is very real and seeing these cars I hope illustrates that to you. And it occasionally affects thousands of families every year. Secondly, you’ll note that throughout those slides we didn’t know much about what happened to the fetus and that reflects the real world and that these reporting systems don’t tell us what happens to the offspring. I’m sorry, that was the other thing I wanted to tell you. So not only are the fetal injury and the mortality hidden, but often the maternal mortality is hidden because of the way, as we’ll talk about today, the definitions have been applied to maternal mortality that exclude injury. And because it’s been hidden, the intersection of the study of injuries in reproductive sciences is really an area that hasn’t gotten much attention from public health people. Other health sciences that study human exposures to toxic, infectious, and radiologic agents pay a great deal of attention to this critical period of human development.
But with a few exceptions, many of which you’re going to hear about today, the injury field has lagged behind in research, practice, and surveillance. This inattentiveness continues even as injuries have become the leading cause of maternal death during pregnancy and estimates that suggest as many as eight percent of all pregnancies incur an injury that result in a hospital visit. To counter this lack of attention we’ve assembled this special session today and the objectives of this session are available to you, I’m just going to list them here quickly but not read them. To accomplish this you’re going to hear from several different speakers today. I’m going to begin by discussing some of the background issues and barriers to the topic, followed by a separate presentation of a multi-state project that looked at the prevalence of risk and pregnancy associated injury hospitalizations using a direct approach using hospital data. We’ll then hear from two other university based researchers who use somewhat similar yet still different types of data linkages and strategies and slightly different targets to look at hospitalized maternal injuries and birth outcomes.
All injuries in the first study reported by Dr. Melissa Schiff from the University of Washington and just motor vehicle injuries in the study reported by Larry Cook from the University of Utah. Our last speaker and panelist, and in many ways perhaps the person most involved in this area is Ms. Monica Randall from Washington State. Her introduction to this area occurred on May 27th, 1995, when her car was hit head-on at a high rate of speed when her son Ben was 26 weeks in utero. Ben was born full term with a large, irregular shaped piece of skull missing. On work up for reparative cranial plasty at the age of two years and eleven months, that’s almost three years old, he was found to have right cerebella hyperplasia, believed to have been caused by an infarct that occurred at the time of the crash. Her main emphasis, and her husband’s, Paul, this last six years, has been on learning about what happened to him, researching and reading about brain development and remediation strategies along with helping Ben recover.
Ms. Randall has worked in the past in child support for state and county governments and since having children has worked as an in home medical transcriptionist. I met Ms. Randall through her poignant letters to me about the experiences that she and her family and her son have gone through since their crash. So when Monica comes up, please give a special warm welcome to her because this is I think the second time she’s been in front of this kind of audience. But let me thank all of the panelists for their willing to share their research and tell their own stories in their own language. They all saw the importance of coming together not just to share their research, but to be real advocates, I think, in the best tradition of public health for more attention to this area by you, the MCH community and to present the various approaches for the first time in front of an MCH audience of ways and techniques that we think can make a difference, and can be done, to track and improve the visibility of this important but neglected maternal and child health issue.
It’s now my job to give you the big picture of maternal fetal health, so let’s begin by looking at some of the familiar indices of maternal and child health, the maternal mortality ratio. The World Health Organization defines the maternal mortality as the death of a woman while pregnant or within 42 days of termination of pregnancy, you’ve seen this before. But I notice that some of the speakers didn’t include that bottom line that’s also included in the definition by WHO, because it excludes accidental or incidental causes. I’ve yet to actually figure out why that’s in there, but it’s in there and it’s been followed by the National Center for Health Statistics ever since. Now, we’ve also seen this slide in several of the sessions, the incredible progress in maternal mortality over the last hundred years.
For many years maternal mortality has been the benchmark in maternal health and it still serves important public health purposes in developing countries. For example, the lifetime risk of dying from pregnancy and childbirth in developing countries is 1 in 60, whereas due to the incredible strides in the United States and other developed countries is probably down to less than 1 in 3,700. This is clearly one of the great successes in women’s and child’s health. But due in part to the reduction in the numerator, but also due to increases in non pregnant related causes of death during pregnancy, in developed countries the maternal mortality ratio more and more fails to describe major risks to women and their offspring. Unintentional injuries are by far the leading cause, three-quarters of which are related to motor vehicle injuries. Also in the top five are death due to homicide and suicide. If we were to list maternal mortality as a separate cause of death on this leading cause of death list it would rank far below the major causes. Of course most of these deaths occur to women who are not pregnant and thus somewhat out of the focus of maternal mortality.
But how many of them are pregnant at the time of their death? Well, actually we don’t know for sure. Most studies have not been population based, or have been hampered by methodological limitations. But if we assume for argument’s sake right now that the distribution is roughly the same for women of reproductive age, regardless of pregnancy status, the following emerges. In the U.S. more pregnant women probably die from intentional and unintentional injury than all maternal mortality related conditions combined. I want to emphasize that this may not always have been the case. The reason can be found in the trends of the largest group of maternal injury deaths, motor vehicles. Over the last three decades there has been a marked trend of more women of reproductive age becoming drivers.
Furthermore, women of reproductive age who do drive are driving more miles. This sets the stage for what I call the hidden epidemic, one that the injury and maternal and child health community I think has totally missed the last 30 years, and that’s the epidemic of maternal crashes. As more and more women entered the workforce and adopted lifestyles and transportation patterns very different from their counterparts of just a generation before, they also adopted the increase risk of serious trauma. So when the definitions of maternal mortality were created, a long time ago, it was before this epidemic and my point is that things have changed and we need to realize that. The result is that today’s women and babies are much more exposed to the dangers of an in utero motor vehicle crash and perhaps some other injuries than all previous generations.
So take home lesson number one is not only has the absolute risk of serious trauma to women of reproductive age increased over the last several decades, but the contribution of trauma to women’s overall risk of dying in the reproductive years has increased dramatically, making it an important maternal and child health issue. Yet, we hardly track the issue. James *Marux said that, “What gets measured gets done.” This statement is particularly germane to the topic of maternal and fetal injury because for far too long the public health system has not been measuring it very well. Here are some of the major data sources that we use in injury and in maternal and child health and we really do a very poor job. The death certificates for the most part don’t have a good pregnancy indicator and tell you nothing about the outcome. For the fetal death certificates there are no E codes, the external cause of injury codes that we have for the other injuries that occur, again because of some WHO regulations on how the coding should be done.
So at best we may know that the baby died because of a maternal injury, but we have no way by looking at the codes what that cause was. Here’s where the opportunities lie, not by looking at the data sources by themselves, but perhaps through some of the linkage opportunities and these are the things we’re going to hear about in more detail today. However, the result of the current system is that none of the important national injury data publications, nor even in the reports from CDC on fetal mortality do we find any numbers pertaining to maternal fetal injury. And because the numbers aren’t there, the public health agenda at the national and state level is embodied in the healthy people framework and CDC research agendas have not addressed the issue. So an important lesson is that the problem has not been addressed and that the problem is not being tracked very well. To get a better handle on fetal related mortality one can look at fetal death registries. Fetal death systems are maintained, as most of you know by most, but not all states. But they generally only include fetal deaths of 20 weeks or more gestation, so they tell us nothing about fetal loss, but it’s a good place to start. Plus, because they are ICD9 based, they do not code the mechanism of injury for the reasons I talked about. However, all though the E codes aren’t there and thus one can’t tell from the computerized data bases what the injury causes are, it turns out that the fetal death certificates themselves usually have the narrative that may explain what happened.
So if one is willing to go through those fetal death certificates that have the code maternal injury as a cause of death for the fetus, one by one by one, you can get a handle on the picture. And this is what I did for 15 states representing about half the U.S. population from 1995 to 1997. this study published in JAMA two years ago showed that motor vehicle crashes accounted for 80 percent of the known causes of traumatic fetal death. I also used this data to extrapolate to the country as a whole, making some adjustments for missing data. In just looking at the crash related deaths, one can make the interesting and provocative comparisons shown here. In terms of frequency the motor vehicle occupant fetal death cases far out number the other causes of fetal childhood injury that receive a much larger amount of attention by public health practitioners. So lesson three, fetal motor vehicle injury deaths alone account for many more deaths than several other leading cause of childhood injury and many more times that of infants dying in crashes because they take on the risks of the mother.
Next, let’s discuss some of the issues surrounding non-fatal outcomes. And this will lead to the three research presentations from today’s panel. So far we’ve only focused on the outcome of fetal death. But due to the unique situation of the developing fetus, there are many more fetal threats and adverse outcomes to be concerned about. Some of these are listed here for example. We can have direct fetal injury in which an exposed body part is directly injured by the forces of the crash or injury. And anybody part can be affected. Obviously when it’s the brain or the spinal cord or vital organ it’s much more important. But it can lead to obstetric complications, placental injury, uterine rupture, amniotic rupture and trauma related, elective, and therapeutic abortion. Notice that most of these things in terms of obstetric complications, they’re not injury conditions. What gets coded in the hospital records often and what may get coded in the fetal death certificate are these conditions, not the underlying injury that triggered the cause of events.
And to the baby we really don’t know much, so some of this is speculative. We certainly know it can lead to neonatal death and you’ll also hear some of the quantifiable aspects from the studies today talking about prematurity and low birth weight. We don’t know much about the adverse development to the neural and other organ damage because it hasn’t been studied on a population basis. We don’t know for sure about mental retardation and ADHD and these other neurological conditions because again there is no study that has followed up crash victims for any period much beyond birth. So remarkably there are very few population based injury studies. Two have focused on motor vehicle injuries, Wolfe and Hyde, but only one, the one you’re going to hear about today, has focused on all injuries, from Melissa Schiff and you’ll hear from the authors themselves.
Even more disturbing, the number of population based injury studies looking at child outcomes after birth, again, is about zero. No long term follow up on the impact of different types of trauma on child development has ever been done. So lesson four, fetuses are at unique risk for a variety of adverse outcomes, but research is needed to quantify these and other long term risks. Let me take this issue in lack of attention a couple of steps further. If we think of injury as a complex environmental exposure, one involving the trauma itself, the stress of the trauma to the mother, the treatment that’s done to the mother to protect her, which should be paramount, we can see that it has dimensions that make it seem as big as an order of magnitude of some of the more traditional concerns. And you can see how it fits in here in roughly the same order of magnitude.
Perhaps a more specific comparison is in order and I’ll admit this is somewhat speculative. In this case, fetal alcohol syndrome in which the rate of fetal alcohol syndrome is pretty well established and I think the rate of maternal vehicle related fetal injury is not quite as well established. So what I did is we did have a good handle on the fetal death rate and I just assumed from my other work in injury that there are roughly 10 injuries, serious injuries, for every death. If you do this you get about 2.5 per thousand births for the number of injuries, which again is on the same order of magnitude as the rate of fetal alcohol syndrome. This may be lower, it may be higher, but what I’m trying to bring home is that it’s on the same order of magnitude. Yet when you look at the literature as a rough measure as what has been done to research and address the issue, you see this kind of imbalance, 155 papers versus 15,000 papers and most of these don’t have to do with outcome, they have to do with treatment of the mother.
So here are the take home lessons. More and more women and fetuses are exposed to trauma, there’s significant gaps in surveillance of pregnancy related injury that hide the problem. Fetal mortality represents a very large proportion of childhood injury, mortality. We are just beginning to measure the adverse fetal outcomes due to trauma and no one has measured the long-term impact of non-fatal fetal injury among children. In conclusion, maternal and child health professionals should not opt out of the responsibility to characterize or reduce the impact of maternal trauma on fetal and child health. To do anything less is to ignore the youngest and most vulnerable members of the population and leave in peril far too many members of the next generation. If we don’t count fetal injuries we can’t measure them. If we don’t measure them the problem’s going to continue to be ignored, but ignoring the problem has never made it go away.