MCHB/EPI Miami Conference — December 7 - 9, 2005

Maternal Age: In the Tails of the Distribution

SARAH NABUKERA: Good morning. Thank you all for coming today to listen to my presentation. And I would like to acknowledge my co-authors, my partners in crime while we're doing these studies. More especially Dr. Greg Alexander, who is having open-heart surgery as we speak.

The study I'm presenting to you today, we were looking at first time birth among women 30 years and older in the United States , particularly we wanted to revisit issues related to patterns and risks for adverse outcomes. And I must confess I have a personal reason for wanting to look at women who are older and just beginning, you know, having families. My presentation today is going to focus on these areas. The background, our study objectives, and the methodology we used, findings, some brief comments on implications for future research.

Over the past two decades much of the literature is showing that there is an increasing number of women, particularly in developing, developed nations, the U.S. inclusive, who are having children at the age of 30 and older. However, there is conflicting information regarding pregnancy outcomes in this group of women. While some studies have shown that, you know, advanced age is not really a risk factor for adverse outcomes in both the mother and the infant, other studies have shown, you know, adverse outcomes. And partly the reason could be because of different study sample which have been used. Some of the studies are very small, very small sample sizes or very specific, from specific hospitals or looking at specific population, either highly educated women are women in the work force. So, you know, and this has contributed to the differences in the study. So we really wanted to revisit this study using a more large population base and we wanted to look at what are the trends currently over the last 20 years of women who start late initiation of child birth, to describe the characteristics of these women, and to determine maternal and infantile counts by the age of the mother, and to determine if there's indeed any risk factor as related to the age of the mother when they start having children.

To do this we decided to use a national database. We used a database from the National Center for Health Statistics. More specifically, they linked birth, infant and death files, cohort files, the mortality files and the fetal death files. And just so to give us a more comprehensive picture we had to use multiple data sets. And our analysis was limited to singleton, index, pregnancies. The first pregnancies. And for obvious reasons we had to eliminate multiple pregnancies because they do have their own unique problems and too we figured that in this population of women, probably a high proportion of them would be having multiple pregnancies. So we decided to eliminate that. We compared proportions by the age of the mother when they started having children. And we did multiple logistic regressions to actually determine the risks.

What this one shows, this gives us the information of the trends over the last 22 years and what it shows is that over the last 22 years, women 20 to 24 years old, and still while they're the largest number of women having children in the United States there is a decline over the years. Same to 25 to 29 years. On the other side, women it their 30s are showing an increasing pattern of having children, starting to have children much later on.

Looking at the maternal characteristics, it's clear that more women who are having children later are white mothers and more highly educated as opposed to women in the other age groups. Perhaps what I need to mention here is that because they are, we figured that women who were in their 20s are very, very similar we decided to collapse the 20 to 24 and 25 to 29 into one age group and compare, use that as our comparison group for women who start when they are 30. And all these ones we did p-values for them and because they're such a large data set, almost 7.1 million birth over this 5 year period we had, so most of the p-values were statistically significant, less than .05.

Marital status, too, most of them are married and they demonstrate adequate and intensive use of prenatal care for women who start having children much later. Particularly those who are in their 40s.

Now looking at our counts, these are specifically for maternal outcomes pregnancy induced hypertension, pre-existing hypertension, and gestational diabetes. All these parameters are high for women who start having children when they are 30, but more so for those who are in their 40s. Same goes for pregnancy anti-partum, hemorrhage, abruption, previa, all of them are much higher, a small proportion of pregnancies have this complication, but it is higher in women who start having children much later, particularly for those who start when they are 40. Caesarian section delivery, which has been increasing steadily in the United States and as you can see older women are no exception, perhaps are even having more operative deliveries as opposed to other categories of age groups.

Now, switching gears to look at the birth outcomes, we looked at very pre-term, pre-term, very low birth weight, low birth weight, and SGA. And we defined very pre-term as less than 32 weeks of gestation. And pre-term is, we define it as less than 37 weeks of gestation. Very low birth weight was less than 1500 grams and low birth weight was less than 2500 grams. SGA we used less then the 10 th percentile based on the gestational age. And we find that for, compared to women who are 20 to 29, higher rates are seen in women in their 30s and above, more so for those who are in their 40s when they start having children.

We looked at fetal mortality, neonatal mortality, post neonatal mortality, and infant mortality. And fetal mortality, we pretty much found the same study my colleague who presented first from Utah found that women who are much older do still have a higher rate of mortality, both neonatal and infant mortality as well. Post neonatal they're not very much different and we were thinking that this perhaps is because, you know, most of the older women are highly educated professional women and, you know, are probably some of their effects are being mitigated here, that they can be able to take care. I guess, have better access to care for their children, so much of the post neonatal effects are almost not there.

To try out, we expanded this further like I said to look at the risks for women who are having children and really we wanted to see is age per se as a variable on its own is it really a significant predictor of adverse outcomes in women. If you had the mother coming to you and she's 35 years old, should you be concerned if you're taking care of her or if you're giving her any form of support, that she's, you know, at high risk for anything in regards to the pregnancy. And we looked at the risks, we looked at where pregnancy induced hypertension, hypertension, and gestational diabetes and in this we did the multi-variant logistic regression.

We used the 20 to 29 year age group as our reference category and we controlled for education status, race, marital status, and prenatal care use. Mainly the socio-demographic factors. And you find that in, you know, while all of them are high for pregnancy induced hypertension, really the effect is very more great, almost not there for the ones who are 30 to 34. When you go to pre-existing hypertension you have a two-fold risk for women who are 40 and above. Gestational diabetes you have a two-fold risk, almost 3 times as much for women who are 40 who start having children in their 40s. So there is a significant risk for gestational diabetes in women who delay childbearing. Placental abruption, biggest concern is in 35 to 40. And placental previa, two-fold risk, almost 5 times as high for women who start when they are 40. And C-section as we mentioned previously, also the risk is much higher for women who delay, who start having children when they're in their 30s.

Regarding birth outcomes for very pre-term, pre-term, low birth weight, very low birth weight, and SGA. The risks are more great for those who are in their 30s, highest for those who are in their 40s where you have almost two-fold increases for very pre-term, pre-term, low birth weight, and very low birth weight. And those are perhaps the ones, which are full of areas of more concern.

Fetal mortality, fetal mortality, the risk for fetal mortality is almost 3 times as high for women who are in their 40s when they start having children. Neonatal mortality too, they have a two-fold risk and infant mortality. Like I said, post neonatal mortality they don't really appear to have an increased risk for that. And in this first model we just controlled for socio-demographics, but of course we know that mortality has a lot to with gestational age and birth weight. So we did put those two into consideration as well. And while the effects were still present, the odds ratio are slightly reduced, but you still see that for fetal mortality you still have a two-fold increase for women who are 35 and older when they do have children.

So what is our take home message here? Is that age per se as it is is indeed a predictor for adverse outcomes, maternal and infant outcomes for mothers who start, particularly for those who start when they are 35 and older. I mean we may see that those who start in their 30s may be of low risk, but the ones who are 35 and older, that's just a group of people, population of women you would want to pay attention to as they are increased risk for complications, both infants and maternal.

Okay. And there is need to, you know, this is an area where we need to, I guess, give more preconceptual care is vital for this category. And needless to say women need to be educated to make informed choices, because, I know, we can't dictate to them that you have to have your children before you're 30, but if they're going to do that then at least they have to be aware of what risks they probably are going to be facing.

And what are implications for future research? Like I've, the data I've presented didn't look at anything by race groups, but it will be interesting to know are there any differences between different race categories of women who start delaying child bearing. And we put it at 35 simply because we found in this study that women at 35 and older are the ones who are perhaps at more risk than women who begin when they are 30 to 34.

And what happens when women, you know, who start late still go on to have subsequent pregnancies. Current research shows that women, even though they start to have children late, many of them don't want to have one child, they do want to have a subsequent pregnancy, and at the same time, they're all being bombarded with information that, you know, you're at increased risk for that or the other so, you know, they try to do it in a more rapid fashion. What is the outcome in those subsequent births, if women do have children in such a rapid fashion? There isn't much in the literature right now to give us an insight to how those pregnancies go, but it will be interesting to find out. Thank you very much. Any questions?