AMCHP 2005 ANNUAL CONFERENCE
DELIVERING RESULTS, IMPROVING PREGNANCY & BIRTH
February 19-23, 2005
JO ANN WALSH DOTSON: --came to the State Health Department, I came as a nurse consultant that was responsible for the home visiting program up there, the Public Health Home Visiting. We call ours Miami . Go figure. Somebody came up with that as Montana ’s Initiative for the Abatement and Mortality in Infants. If you ever really want to bore your legislatures, call it something like that. But at any rate, I became very interested in pregnancy outcomes and fascinated at the variability. In, lets see, in the 1990’s, the early actually 1990’s we started hearing more and more about the PRAMS Projects, which stands for the Pregnancy Risk Assessment Monitoring System. And a wonderful lady that I worked with at the time, kept saying Jo Ann we need to apply for that, PRAMS. We need to apply for that, PRAMS.
And PRAMS is a surveillance project of the Centers for Disease Control and Prevention and is in partnership with State Health Departments. PRAMS collects state specific population based data on maternal attitudes and experiences. So a good number of you are involved in PRAMS in your states, correct? And the thing that I was most interested with is, as I said, was initiated, you know, in the 1980’s, but it was partly initiated due to the fact that their infant mortality rates were no longer declining. We were starting to hit that plateau. And the instance in the low birth weight rate had changed little of the previous 20 years. So it was apparent that we needed better data and proved data to figure out what was happening. The goals of the PRAMS project were to improve the health of mothers and infants by reducing adverse outcomes, and that’s what’s a really important distinction.
It wasn’t only to gather the data but to allow both at a national level and state level the information and the drive that we needed to allow us to actually make changes in policy. In my doctoral work one of the things I’ve definitely discovered about myself is I don’t seem to terribly interested in theory development. I want to, I want to be able to identify something and apply it immediately. Now, I have been informed by my faculty that sometimes I need to slow down a little bit, because I will skip right to the chase even if it really isn’t very well determined sometimes. But it’s that application of health policy that’s of most interest and import to me. The PRAMS does provide the state specific data for planning and assessing health programs. And the importance is that the sample is chosen from all women who have had a live birth recently. I’ll talk a little bit more about the process in a few minutes. So the findings can be applied to the states entire population of women.
Many of you involved in research or not involved in research recognize that much of the, many of the data sources that we have are based on a very specific population from a specific clinic, from a specific hospital, from a specific area. The advantage of the PRAMS is, is that it does generalize and get a broad enough sample that we do have something that’s generalizable to the state level. It allows us, because of the consistency of the data collection, it allows us to monitor changes over time. And of the greatest interest, I believe, to a lot of us, and the reason that this wonderful person I worked with kept saying Jo Ann we need to apply. Jo Ann we need to apply, is so we were going to be getting data that was comparable to other states. We had and you have, many of you have participated this in the MCH performance indicator, a lot of you were gathering infant mortality data or teen pregnancy was a favorite. You know, some of us had teen pregnancy data on 13 to 17, others had 15 to 19, we had all of, so we had apples and oranges and we never quite knew how comparable we were to other states.
The advantage of the PRAMS is that we ask the same core questions the same way. So when I look at the larger data sets, I can ser where we’re at and that’s a huge advantage to me in policy development and it’s a great advantage as I go to try to explain it to other policy makers about what this means for the surrounding states of the nation as a whole. The states participating in PRAMS, there’s 31 states and New York City , I always love that. You know, we don’t have as many people in the state as, even close. But it’s the 31 states and New York City participating. It is important to note that 62 percent and that’s probably slightly higher recently, of the births occurring in the nation are captured in the present PRAMS data collection. So you’ve got a good data source that you’re working with. This is sort of pulled out of context. I hope that many of you have had the pleasure of meeting with and working with Roger Roche who has blessed Montana by getting so he likes us.
So he comes up and works with us periodically. He put, he delivered this information so I shamelessly stole it but I did recognize him down below. PRAMS is shared in a lot of areas and in a lot of settings. These are some of the ways that PRAMS is shared. You can read those yourself. Reports, cyper, presentations, thesis dissertations is important to me because that’s what I’m working on. But this is, of many of the ways that the PRAMS data is shared and of specific import, you can see that over 30 percent is looking at prenatal care, unintended pregnancy factors, and all the way down. Factors closely associated with pre-maturity. We can’t say causations of, but they are associated with. I’m going to do a very, it’s not even a 101. It’s like 50 and a half of the PRAMS process because so many of you know it already. It is a survey of women with live born infants in the previous 12 months. There’s standardized core questions with a state option of additional questions. We actually chose not to do additional questions and I’ll talk about that more later on, we’re a point in time.
But a number of the states have developed core questions and that’s nice to because there’s more and more core questions. You have a library to draw on instead of inventing wheels. The sample is drawn from the birth certificates using a stratified random sampling technique. And there’s a number of ways participating state and cities have options of doing additional stratification. Much of stratification is looking at birth weight, maternal race, ethnicity, maternal education. We in Montana chose to stratisfy just on maternal age. We were specifically interested in our teen pregnancies. And in North Dakota they looked at Medicaid, which number of pregnancies, were paid for in Medicaid. And I don’t know what you guys ended up looking at. That’s an option. You don’t have to go with stratification, although they do suggest that. One thing that may seem apparent and we struggled with this, is well why didn’t we go with the minority births, which for us is our Native American births.
Native Americans make up about six to seven percent of our population, about 12 percent of our births, and about 16 percent of our infant mortality. So it’s evident that there is, that’s an interest population and it’s a rapidly growing population as evidenced by the high percent of the births. The challenge that we faced in that this was a point in time study. And we have seven reservations that are home to nine sovereign nations. It was, it’s very difficult to get approval for, through the tribal councils. And it probably would have taken us we were to do data collection in a year. It probably would have taken us that long to get permission from all the various tribal councils. And the idea was that we weren’t trying to specify or narrow to any particular tribal affiliation. So we do have information about Native American births obviously in the data set, but we didn’t choose to stratisfy an over sample in specific populations having to do with the Native American population, something that we may look at in the future and we will do so in conjunction with the tribal leaders on that.
I didn’t say, usually most states do data collection end up sampling between 12 and about 3500 people or women in order to have a representative sample. We did about 1300, so we were obviously on the low side, which is because we don’t have very many people, and the larger the community or the larger the population, the larger the sample. The data collection is mixed mode. It is a self-administered questionnaire that’s mailed to this specified sampling. It’s a fairly specific sampling technique that’s been described many times in many other publications about how that process goes. If you’re doing it, you know. If you aren’t, you find that out. And I always remember when they said, well, we needed a full time person to do the PRAMS survey, I was like we can probably do this with just you know, a little bit of a person. It’s pretty time intensive to just do the mailings and the tracking and doing everything. That is difficult for us because it is it’s also just 1300, person sample.
It’s kind of tough to go into the legislature and go, well, I need an FTE to do a survey of 1300 people, because they look at you like you’re not very efficient. But it is something, the process if fairly intensive. We also do follow up with a telephone survey in order to assure that we have at least a 70 percent, PRAMS does require a 70 percent response rate in order to assure that you’ve got a generalizable data set. And the self-reported survey data is linked to specific, selected birth certificate data, which is the real beauty of it that you don’t lose that data, you can link it. The core question topic areas were of very, of great interest. The attitudes and feelings about most recent pregnancy, the content and source of prenatal care, you can read the rest of them, substance use. One that I didn’t put up here that has been of great interest is physical abuse before and after pregnancy, that’s another specific area. And what you’ll find is that there’s a series of questions regarding each one of these areas.
The total questionnaire count depends on if the state adds in additional questions, but you have a survey that’s typically about 80 questions. It takes a little time for people to respond to. This is not a five minuter where they can get through pretty quickly. So a lot of states, and I won’t be going into that in this presentation, have done some incentive mechanism, some incentive program to encourage people to participate in the survey. What we’ve found and what I’m seeing in many of the reports is the people, the women like the opportunity to share the birth experience as many of you know. There’s a need to speak that out or to share that. In a lot of ways the qualitative or the narrative comments we get in is, this is so great. And one of the areas that’s really opening up, this isn’t brand new but it’s over the last few years, is the qualitative analysis, potentials with the PRAMS data set. There’s huge qualitative data available in these. And it’s a gold mine.
I want to talk briefly, you guys know most of you know this too. But, you know, you just have to go over it a little bit. The pre-maturity in the U.S. as you well know, is defined as birth prior to 37 completed weeks. It has increased by 20 percent in the last decade. This is not unlike other industrialized nations in the world. So we’re not completely unique. There are a few exceptions where there have been better there continue to be better outcomes in pre-maturity. But many industrialized, and one thing I might note now is that’s a quote by Joseph. I did put one of, an abbreviated listing of some of the references that I’ve used in the one page front and back form. So if you wanted that, it’s just an easy place to find them. If you want about 12,000 more, just give me a call, because I have those in my office at home. The pre-maturity accounts for 11 percent of all the births in the U.S. , which means that that’s about 470,000 premature babies in the U.S. annually or about 1200 per day.
Of that 1200 per day, 12 babies die. And, those of you, that worked in ICU’s, or just are working with the clients all the time, that’s what it always gets down to is you know that some of these kids aren’t making it and what the families are going through. That’s what drives us. Two thirds of the infant deaths in the U.S. occurred in infants who were low birth weight or premature and you know well that that’s not an exact link. And out pre term infant cost at least 13 times the cost of a full term new born. Lockwood notes that about half of the 25 billion that is spent on infant hospital care is spent on that 11 percent that are the preterm births. So it’s a really significant financial driver as well as ethical and other concerns. Fifty percent of the pre-maturity that we know of results from, this is one thing we can say, spontaneous labor. We don’t know what caused it but it does spontaneous labor.
Thirty percent from spontaneous rupture of membranes and 20 percent was due to assisted delivery because of there was crisis in either the mother or the infant. Now all the causes for those things happening, that’s where we’re struggling, with. We do that factors associated with pre-maturity include all of these things. A few that I didn’t write down there also are diabetes and uterine anomaly, and placenta previa. These are no surprises to any of you that are involved in healthcare area. Factors associated with pre-maturity, the social structures, depression, domestic violence, psychiatric disorder, homelessness, and I wrote, and multiple residencies. I don’t know if any of you used terminology on that. Up north in Montana we have what is considered a very low homeless rate. That’s because it’s too cold. And, you know, I’m really not trying to be glib, what happens very often is the multiple residencies, every time they’ll find a family member or a friend or a fellow substance user to crash with, they stay there until the get kicked out or the next rent is due and then they move again.
So one of the things we found early on is the homelessness, a client wouldn’t respond and say I’m homeless because I have a place to stay. But they were moving seven, eight, nine times during the course of a short pregnancy. So they’re moving every few weeks. The stresses involved with that are horrendous and very often it’s the same that you see in homelessness although they may not have been physically on the street at the time or in a shelter even. We don’t have that many shelters that house women in Montana . So people just aren’t, we’re not picking up what may be evident in other states. Substance use, tobacco, alcohol and other drugs, both crack and meth, those of us from rural areas, it’s certainly not less of a problem in the urban settings. But rural areas and frontier areas seem to be great places for meth cook joints because they’re, nobody can find them, they’re like out, there’s nothing around. And they can essentially just abandon places as they move around so we have a lot of old sheds and things like that that nobody realized that were meth.
We also have a lot of trailers and things that are being used as meth cook joints and they pop up and down. The multiple births, this is one that you folks have all heard about. This is big in the news. This is of real interest to people, all the more reason we have more pre-maturities. We have more multiple births. We do have more multiple births and it is true the number of the multiple are premature. That doesn’t fully explain the premature rate. Not even close so but it is a concern. Lets see, one of the things that we do know, environmental hazards, there’s exposure. Those are something that we’re just starting to get a handle on and one of the things I’m real interested in. Analysis in our area is not only looking at rurality, impacts of rurality or relationships or rurality, but those places where we know that there’s been environmental hazard. We have a lot of super fun sites. We have a lot of other things. And we haven’t done a lot of linkages about impact done, prenatal outcomes.
Some of the things that we’re pretty clear on are maternal age. We do know that young women 13 and 14 year olds have a 2.7 to 4.1 more likely incidence of having preterm birth than a 25 year old that’s in the same ethnic and socioeconomic group and that’s by Akimbo. The point is that the knowns that we know don’t sometimes hold out very well. I was sitting in a meeting this last week in Montana with our fetal infant child mortality review and I had a wonderful physician that I worked with that somebody asked the question about pre-maturity and I said, well I’m just getting going and I’ll see if I get smarter when I come back. Maybe I’ll know a lot more about pre-maturity when I get back. But he made the statement about, well we know that if there’s prenatal care, there’s early prenatal care then you’ll have improved pregnancy outcomes. And I went well, not really. And you know, that’s a shock to many of us. She was, Lois you were good enough not mentioned when I graduated from my Masters and my Bachelors.
That was a known to us when I was growing up that if you got people into early and continuous prenatal care, you were going to have improved pregnancy outcome and decreased low birth weight and pre-maturity. And we were so confident of that and it was kind of a shocker when you find out that was wrong. But, and one thing I don’t know if I put on that particular thing, I would refer you to Michael Cogan. He did such a nice job of summarizing that, the changing pattern or prenatal care utilization in the U.S. And Michael Cogan published this in (inaudible) and this is old, this is ’98 but it’s still a goldie. You know, he kind of put it all out there. Lets see, and one of the things that we do know is this is more and lots of other people have published pretty much the same statement. A full half of the women have preterm labor, or preterm births, we have no idea why they did. Zip. We don’t know.
That tells us that there’s research needed. There’s areas that we really need to look at. The purpose and uses of PRAMS related to pre-maturity. PRAMS can be used in foreign policy makers about the incidence of risk factors in their state. Again, it’s comparable from state to state so it’s a good place to use. The characteristics of the women with specific risk factors, the geographic, demographic clues to risk factors, this is of real interest to me. The knowledge deficits, existing in pregnant-women that’s one of the beauties again. This is a self-reported and there’s lots of criticism of self-reported surveys, but the flipside of that is, you do actually find out what women believe and what their experience is and then you can link those to the pregnancy outcomes. And you can find the successes or failure of efforts in the state to address the risk factors, although we don’t like to admit that sometimes, but some of the stuff we do probably, that go forth and do good approach doesn’t work very well sometimes. The analysis, there is a basic analysis.
I’m going to be sharing a little bit of that briefly with you. It does provide us valuable information regarding comparative incidents. It can direct more in depth analysis. It can tell us where we need to shoot for and what we need to look at. And it’s useful and legislative analysis. Legislators love this because they can see, from state to state, what else is happening. The true associations or relationships, you’re probably going to get into logistic regression. And informs the association and related instances with things, I just put some examples down, maternal age, cigarette use, and pre-maturity incidents. You know, how can you look at that? What’s the relationship? I would like to say I completed mine in four-hour dataset, yet, I have not, but that’s what I’m working on. I did want to tell you a few of the pieces that have.
There’s lots already published with PRAMS. So you can walk away today and if you have a legislator going what do we know about cigarette use? What do we know about this? There is a lot published already and you can pull it out and tell them this analysis isn’t of your state but it is of similar areas, if that’s the case. And you can just look to see where, which states were involved and provide them similar information and then decide whether or not it’s worthwhile for you to do that analysis on your own. Specifically I reported in the MMWR in ’99. There was smoking during pregnancy. This particular sample was from Maine , hi Maine , and to determine the incidence and impact of smoking during pregnancy. The findings were that the smoking during the last months of pregnancy, decrease in women 20 and over and the women less than 20 did not have a significant decrease in incidents of smoking. That’s important information. When you’re doing interventions, when you’re doing targeted services, you need to know that the women that you’re really not getting to are those under 20. They’re continuing to smoke at pretty rapid rates and I’ll talk about that just a little bit here later.
Another study is Beck et al. Beck included 17 states in their analysis and this was published in 2002 and examines the impact of behaviors in socioeconomic indicators on outcomes. We find attendance ranging from 33 to 52. How many, have you guys, well, everybody has decided by this point on how you’re measuring unattendedness. But you know when I came like in the early 1990, everybody was using different numbers about where, you know, intendedness was all over the place. Well, the fact is, intendedness is all over the place. But it’s nice as we have more and more consistency from state to state about intendedness. We don’t have to have apples and oranges again. The later no prenatal care, we have, that ranges from 16 to 29. One of the things I will tell you in some of our rural areas, we find that we have people getting into prenatal care fairly early. They don’t have a large number of visits.
They, you know, may not have an adequacy using the coddle check index, but they get in and that’s simply because physicians realize they’re going to see them now or they’re going to see them in the delivery room. They might as well see them earlier and whether or not there’s a good payment mechanism. Smoking ranges from 6.2 to 27 and I’ll talk a little bit more about that. Abuse, this is probably still way under reported. Breast feeding initiation, but that breast-feeding duration. Many of you recognize these as things that we’re also monitoring for the performance measures for MCH, so it’s a good way to inform other funders. I like this one. I forgot to put the findings up here, so this is, you can make up your own as you go. This is prevalence in patterns of physical abuse. This is by Martin and Mackie et al in 2001, incidents of physical abuse around pregnancy.
This data source was actually from the PRAMS in North Carolina . Their sample was two point well, 2,600 and some. The findings were really between the relationship between abuse, socioeconomic characteristics and healthcare practices, use of well baby care. What they found is what many of us kind of know but it’s nice to have some place to go back to, is that abuse is happening across the spectrum. It’s not only seen in low income. It is seen in others. And there are clues to earlier repeated abuse and ongoing. That’s one of the things that we did find evidence of if there had been history of reported abuse previous to the pregnancy. You’re going to see it later on. And again, that’s something to know when people are working with these clients in home settings or whatever. If you’ve been abused before, you’ve got a higher incidence.
So lets think about what kind of interventions are in place. Do you know about battered women’s shelters? Do you know about interventions available? Another one that I didn’t put up a slide on was, Louelia in ’01. That was the folic acid awareness. This was kind of interesting. You know how we’re always thinking that we’re doing great shakes. Actually what we found out is the folic acid awareness in WIC programs and others isn’t any higher than anywhere else. And that kind of tells you something. It’s like okay, lets see. We have a captive audience here, lets see if we kind bump things up a little bit. Or you know, there’s a lot more analysis that can be done on that one paper, perhaps it’s not a good use of space. My time of PRAMS, I need to get done here. Mothers who had live birth in Montana within the last year. These are pretty much standards. We didn’t do spontaneous intended abortions, no stillbirths. And we didn’t do Montana mothers who had their babies outside of Montana, which is a concern in our state because we have no level three, true level three facilities or level four some of you call them, which people always find. We do have four nunitologists in our state now, which we’re quite proud of because when I started, we had none.
So, but we simply don’t so we airlift out to Salt Lake , to Seattle , to elsewhere for any of our truly high risk. We were a point in time survey. I want to tell you real quickly, we only did data collection one year. We will be applying this year to see if we can get ongoing. I continue to struggle. And anybody from CDC here? Okay. So I’ll do my little shot from up here. One of the things we’d really like to do is do, be able to do data collection one year and analysis the next, data collection and analysis. It’s really difficult for us to continue to do the data collection and not to have the time to really do the analysis back and forth. And we simply don’t have, you know, I’m trying to run the WIC and family planning program and figuring how to analyze. We kind of need a little bit of we’d appreciate a slower time. Also, our numbers are low enough. We simply don’t have that rapid a change in our state, so that’s something that I keep going for so every time they see me, it’s like, oh she’s going to do the point in time thing again. But they did do the point in time option for rural North Dakota and us were the only states that did respond to the point in time. What I keep looking at is could we do it over. So they said they were going to do a point in time option on the RFP so we’ll see what comes up.
The point in time projects, were allowed and these were over a three-year period. What I’m saying is we could do a little bit. And I also said that we could take a little less money if they’d let us do it that way. But anyway, I’d better stop. Yeah lobbying, okay, yeah. We don’t do that. We educate. I’m just going to really quickly let you know, we’re the yellow one. And we’re not doing that great for what is a basically homogenous population that’s isolated from a lot of things, you’d think that we were doing a lot better on a lot of things. We have a high incidence of unintendedness in our state. We have a fairly, we have prevalence of late after first trimester or no entry into prenatal care. As I said, people will see a doctor but they don’t worry about getting into the very first trimester. Prevalent, we have a very high incidence of physical abuse. We have done some follow up surveys on this. We do know that this is an extreme issue. It’s associated with our alcohol use. Not proud. But we have a very, very high incidence of alcohol use. This is something that we need to be aware of with all of our young women that are getting pregnant. We have women drinking.
We have women drinking a lot. And we have, we’re doing a lot of work trying to moderate or provide some prevention efforts for fetal alcohol syndrome and you can see that this was a pretty important piece of data for us to have when we were going in for FAS Prevention Grants. We say we need this, desperately. We also have lets see, one of the things that North Dakota found and this is very similar to our numbers is greater than 30 percent of the moms reported that the healthcare providers did not talk about the importance of tobacco or alcohol use on pregnancy. They’re just not mentioning it at all. When we have this kind of incidence in alcohol use and nobody talking about it, we’ve got a problem. And that’s something again, we can take to the bank with, you know, this is a problem, something that we need to be working on.
For those of you that picked up my handouts, I took one slide out there because I couldn’t decide how to explain it. So if you’re seeing a bar graph there just ignore it, X it out. But we do have extreme alcohol use in three months before pregnancy. Eighty two percent on people are saying they’re drinking before pregnancy. That’s a big issue we’ve got to deal with. We also have a lot of people smoking in our state, and it’s the Marlboro man or women, this is only smoking too. We’re not talking about smokeless tobacco. We’ve got a lot of chew being used. And I will tell you that in Montana we’ve got 28.5 percent. North Dakota reported 29.4 percent so they’ve got even a little bit higher. In our high risk pregnancy group, the ones we’re doing the home visiting on, we have 45 percent of the women that we’re working with that are smoking. And it’s probably higher than that but, you know, we didn’t catch them. So smoking is a very big issue and all Montana mothers do you’re smoking now, yes, 21 percent.
But less than 20, 43 percent are smoking during the pregnancy. They’ve already got issues. We don’t need to add to that. Well, I’m going to sort of wrap up here and hand it over to Nan . I will emphasize that that qualitative analysis is just a fascinating data source. If you haven’t had a chance to look at the qualitative data available off of your PRAMS, see if you can grab it. It’s amazing in what it’s going to tell you. They’re just trying to develop some standards and some recommendations for qualitative mechanisms. So, okay.