AMCHP 2006 ANNUAL CONFERENCE
EARLY CHILDHOOD: BUILDING THE FOUNDATION FOR LIFELONG HEALTH
March 4-8, 2006
LORETTA FUDDY: Now, Bill is a little taller than I am, so I'll have to go a little lower here. I think I go with our size of Hawaii being small and short, as well.
Good afternoon. I'm Deliana Loretta Fuddy, and I am, as he said, the head of Family Health Services Division for the Hawaii Department of Health. And I've been involved with the SIM collaborative from the beginning, I have several other team members in the audience that have participated in this effort. Dr. Gigliola Baruffi, just wave your hand, who is with the School of Medicine, the Department of Public Health, our Maternal Child Health training program. And Mark Eshima, who is our statistician with the division, who has done most of the data analysis. And also in the audience is Momi Kamau, who is my Maternal Child Health branch chief. So I want to acknowledge all of their efforts in this.
As Bill said, we entered into this effort by wanting to look at some of the causes related to infant mortality, and pardon the typo there. My portion of the panel presentation is to really look at partnerships and how partnerships have played a role in our look at infant mortality. And basically, if we're going to effectively analyze and address the issues of infant mortality, it does require partnership from both the private and the public sector.
And this is basically to look at directing the activities in the collaborative, because as Bill said, each state is very different, and Hawaii certainly has its unique characteristics, and we do need to pay attention to that.
Also as you begin to look at the data, how do you access that data. What are your data sources, and do you need someone to help you negotiate getting access to that data. Fortunately, most of the data we used were already within the Department of Health. But also helping us to drive what are the right questions to ask. Certainly we've had lots of support from the national team, but again, to look specifically how do we define the right questions.
As we began to look at the quantitative data, it became clear that that's not the whole story. And that if you are really trying to dig down deep to some of the reasons for the rising rates of infant mortality, then you also need to look at qualitative data. And there, certainly we need our partners to help us access that portion of the analysis.
And of course, if you're going to effect change, you really need to look at it from a policy perspective. And being of course division chief, I'm more concerned about the policy aspects and how do we change the system overall. And in order to do that effectively, we do need to involve our partners out in the larger community.
Our representatives on the SIM collaborative were standard at the beginning, with the Hawaii Department of Health, the School of Medicine. We are fortunate to have a very close working relationship with the Hawaii chapter of the March of Dimes, it's one of the national sponsors, but has been intimately involved in this arena, as well.
But we extended our partnership to include the University of Hawaii School of Nursing, because certainly, the nursing staff are clear deliverers of services in the perinatal and the post-neonatal period. Once we began to look at some of the service delivery issues, we needed to involve the pediatricians and the obstetricians. So we've had good working relationships with Kapiolani Medical Center, which is our tertiary facility -- one and only in Hawaii; we're small. And of course from an advocacy perspective, Healthy Mothers, Healthy Babies. And as I said, we did need to reach out to the private physician. Pediatrics, obstetrics, perinatology and neonatology. So we've been lucky to have good collaboration on all levels.
The two objectives that we were looking at was to first identify the major risks and protective factors associated with poor birth outcomes. And of course, the long view, then, is to look at disseminating that information, and how do we look at effecting change as it relates to either maternal characteristics, health behaviors, or the health care delivery system.
Under the first objective, we're still in our phase one, it takes awhile before we're -- we're hoping to move into phase two, but we still need to do more data analysis. Is to identify, as I said, those major protective and risk factors associated with poor birth outcomes.
We were going to look at and identify changes in population characteristics. Look at Hawaii's economic environment, was there any kind of correlation between good times and reduction in infant mortality, and poor economic times and rise in infant mortality.
Another issue for us, as in many other of the coastal states, immigration patterns. We do have many immigrants into Hawaii, and was that affecting our infant mortality rates. Lifestyle and health behaviors. And then the overall community environment.
And as you can see from the long view, looking back from the 1900s to 2000, we've had a major reduction in infant mortality. But the one area that was of concern, and why we became part of the SIM collaborative, is that blip going up towards the end, looking at 2000, 2001, 2003. And so that rise of the trajectory is what brought us into the SIM collaborative.
So while we were making good gains, and oftentimes we exceeded the 2010 objectives, we saw a rise, and we're moving away from that.
So the lesson learned, the second lesson for us, is that data analysis just generates more questions. And that the more we dug into it, the more we didn't find one silver bullet, we really needed to do further analysis. And what you'll find is that infant mortality is multi-faceted. And for us, the vital data trend analysis and analysis of surveillance information didn't really decipher for us, it didn't tease out any key determinants.
So that led us on a path of saying, well, maybe we need to look at some qualitative data analysis. As it relates to the delivery system, or community-wide issues.
The kinds of data that we looked at I'm sure are all the information that other states would look at. Bill talked about the vital statistics. We also have, we're fortunate to have PRAMS in Hawaii, so we utilized our Hawaii pregnancy risk assessment monitoring system. And we did some unique analysis there. And that led us to do perinatal periods of risk analysis, to point us in a direction. That also pointed us in another direction of saying let's look at our child death review information, can we tease something out from that.
Now we're thinking at this point, we also then went to the Healthy Start program in Hawaii, which is known as malama a ho'opili pono program, to look specifically at those women to see their infant mortality rates that were getting some quality care, was that different than the rest of the population on the Big Island. That led us to look at medical chart review. And tease out, again, more information.
At this point, we're going to look at focusing on some hospital discharge data, as well.
Again, review of all of these data sets have not led to an answer for us. It just generates more questions. The trend data, looking at demographic characteristics, labor and delivery complications, medical risk factors, doing perinatal periods of risk, looking at fetal infant mortality review, and again, these unique population comparisons, just generated more and more questions for us.
The development of qualitative data sets certainly required partnership, and this was very much partnership-driven. In order to access chart reviews, we really needed the cooperation of the hospitals in the community, and the physicians. We worked with our medical consultant, who is also on faculty with the University of Hawaii, and with the Kapiolani Medical Services.
So it's nice to have all these connections, was really able to have direct access to the neonatal records and to the perinatal records, and involving those physicians in some of our discussion groups. March of Dimes has an NICU parent support group, so again, working with them, to see how we could access some of that data, was important.
In looking at trends of provision of care, this is where we were able to partner with the School of Nursing. One of the faculty members there, Joan Dotson, helped us develop an instrument that we were able to distribute to -- at conferences, either OB-GYN conferences or perinatal nurse conferences, to get a feel for were there any differences in the way they delivered services over the past 10, 20 years. Did we see -- did they see differences there. We also worked with our lend project.
So again, we got the involvement. If it weren't for the involvement of the professional groups, of the physicians on the ground, and the university, we would probably not be able to look at some of the qualitative data piece. And we're still in the process of trying to develop and refine that instrument. We only field-tested it at the conference.
The third lesson for us is Hawaii is a small state. And really, size does -- may have its influence on our rates. They present unique challenges for us.
As you look at that variability from year-to-year, geographic variability, ethnic, or maternal characteristics over time, with your small numbers it may not really be significant. And that's one of the issues that we're struggling with.
And to illustrate that, Mark had put together this very nice chart for us, and you can see the infant mortality rates in Hawaii has jumped up and down. And of course, the pink line going across the way is our 2000 objective, and then the 2010 is the red line.
And what brought us to the infant mortality collaborative was that increase, 7.6, and then again going up to 7.3. Most recently, though, if you look at 2004, our rate has dropped down to 5.4. Certainly below what was the 2000 objective, but not reaching the 2010 objective.
But again, the question arises, is what we're seeing a result of small numbers? Is it that our community is so small that we can't really tease it out? And for us, we're not satisfied with 5.4, we want to push the envelope. So as a state with a low infant mortality rate comparatively, what is it that we can do to push that rate even lower. And that's something that we really want to look at.
And again, to show that this may not be truly significant from year to year, these are the confidence intervals. And you can see from year to year, they're not quite significant. But if you look over time, then certainly, they are.
And again, this is a map of Hawaii, so you can see that we're fairly unique, we're an island state. So as you begin to look at numbers and you try and break it down by county, those numbers are going to get even smaller. And to illustrate that, this is the island of Hawaii. This is where we have the Healthy Start program. And as a result of doing the perinatal periods of risk, we identified that there was a cluster of infant deaths on the Big Island that was of concern, and it made us look at infant deaths below 750 grams. And these are the ones that we did specific chart reviews, and looked at the prenatal records and looked at the hospital records.
And you can see that each of the pink dots, pink squares, is an infant death. So when you're looking at numbers, these are very, very small, and you're just talking about a handful of deaths. One of the curious things, but again we don't know if it's significant, is as we looked at the psychosocial factors around these families, that many of them were high risk families. They had been known to the child welfare system, and there were issues around that. So maybe that's a marker. But again, the numbers are so small, we need to investigate this further. And if you try and tease it out by ethnic group, this gives you a feel that Hawaii is very diverse ethnically.
So again, if you look at the native Hawaiian population, that only makes up about 27 percent of our births. And Hawaii has less than 20,000 births a year, I believe this last year was about 18,000. At one point the high was around 20,000, it dipped down to 17.
So within the range of 17 to 20,000 a year. And if you break it down by each ethnic group, you're going to have -- and then break it down by island, because part of it is to say is it part of the delivery system on the neighbor islands that's contributing to this. We didn't find that.
So again, the lesson for us was that it's not the usual suspects. When we began to look at the maternal characteristics, you would think that ethnicity might have an impact, nutrition, health care coverage. Hawaii is very fortunate, we have employer based insurance coverage. So women coming into the system, only about five percent of them are without health care coverage. So we're getting good coverage.
If you look at tobacco, substance use, everyone says that we have a very high methamphetamine problem, ice in Hawaii. That was not contributing into it. Entry into prenatal care, we didn't see any variability there as well. Delivery system, was it the babies that were all being transported to our neonatal intensive care unit at Kapiolani? Didn't find any difference there, they seem to have been transported in a very timely manner.
Is it the economic environment? Good times, bad times, are we seeing up and down? Didn't find anything there, as well. Immigration patterns, didn't yield anything there.
So really, where do we go from here? And we're right back around to looking at our qualitative health care data system. So again, trying to look at the delivery system, and to look at being creative in our data comparison. Right now, we think we see that the emphasis will be on our post-neonatal deaths, and that we're seeing a difference between the '80s and the '90s. So our research question right now is why is our post-neonatal death rate higher in the '90s than it was in the '80s? Wouldn't you think that the delivery care system should have improved? Is it that we're just delaying the inevitable, that infants are being born much younger, and are surviving? We don't know.
So we're going to be really looking at expanding our partnership again, to go back to perhaps the -- we have an organization called Hawaii Health Care Information Corporation that collects all of the hospital discharge data. So form a partnership there, and really look at, one, was there a high readmittance rate, when were they readmitted, for what purpose, what was the cause of death, what was the length of stay.
And then to really look at some of the issues around that stay in the hospital. One issue might be an issue of overcrowding. That sort of cropped up when we met with the neonatologist.
So again, that will bring us right back around to some of the policy issues, and how do we begin to address that.
And all of this, as I said, would not be possible without the wonderful collaboration that we have with so many partners in Hawaii. And I think that's my last slide.