Ninth Annual Maternal and Child Health Epidemiology Conference / December 10-12, 2003

Comparison of three Prenatal Care Indices and Their Association with Small Gestational Age

LAJEANA HOWIE:  Good morning, and as Nancy stated, my name is LaJeana Howie, and the topic of my presentation is the "Comparison of Three Prenatal Care Indices and their Association with Small for Gestational Age."  And before I begin, I'd first like to acknowledge my co-author, Kenneth Schoendorf from the National Center for Health Statistics.  Just to give you some background on prenatal care, prenatal care is a frequently used preventative health service in the United States and is viewed as advantageous to pregnant women and their infants.  Prenatal care was designed to aid in the identification of women who are at risk of having pregnancy complications with the expectation of reducing the incidence and adverse of pregnancy outcomes.  When utilized adequately, prenatal care may be associated with the reduced risk of preterm birth, low birth weight, and fetal growth restriction.  Small for gestational age, which I refer to from now on as SGA, is a measure of fetal growth restriction and is generally defined as the birth weight less than the 10th percentile for gestational age.  SGA has been associated with adverse health outcomes in infancy, childhood, and adulthood. 

There are different methods in measuring prenatal care, and they usually use the Analysis of Vital Statistics data, and these methods are created from the timing of the initial prenatal care visit, the number of visits, and gestational age at the time of birth.  So all this information led us to the following research objective, which was:  Compare the associations between three common measures of prenatal care utilization with SGA.  Although this study does not attempt to measure the content or quality of prenatal care but rather how it commonly measures a prenatal care utilization are associated with one pregnancy outcome.  Our data was obtained from the 2000 U.S. (inaudible) file compiled by the Centers for Disease Control Prevention and National Center for Health Statistics.  Our study population was restricted to singleton births delivered between 37 and 42 weeks of gestation.  Birth records missing information on birth weight, gestational age, timing of prenatal care initiation, or the number of prenatal care visits were excluded from our analysis because these items are necessary to calculate either SGA or the three prenatal care measures. 

Our final study population consists of approximately three million births.  For our study, we chose three commonly used methods in measuring prenatal care.  The first was the Trimester of the First Prenatal Care Visit, the Kessner Index, and the Adequacy of Prenatal Care Utilization, which I refer to from now on as the APNCU.  The Trimester of the First Prenatal Care Visit Index is based on the trimester that the mother began her prenatal care, and this was separated into the following categories:  first trimester, second trimester, third trimester, or no prenatal care.  The Kessner and the APNCU indices combined the timing of the first prenatal care visit, the number of visits, and gestational age at the time of birth to characterize a quantity of care as either adequate, intermediate, or inadequate.  In addition to the adequate, intermediate, and inadequate category, the APNCU defines an intensive prenatal care category comprising women who receive more than 110 percent of the recommended number of visits by the American College of Obstetricians and Gynecologists. 

This category was intended to identify women receiving extra care for conditions associated with adverse pregnancy outcomes and thus may represent a group of women at relatively high risk for adverse outcomes.  Other variables consisted of the following.  Fetal growth restriction was dichotomized as either SGA or non-SGA based on the 10th percentile for gestation using published criteria.  To examine prenatal care categories by maternal race and ethnicity, a composite variable was created:  Hispanic, non-Hispanic black, and non-Hispanic white.  (Inaudible) women of other race ethnicity combinations were not included in this analysis because a number of small births to these women were not sufficient for analysis.  Maternal age parity, maternal education, and marital status were also examined.  So now on to results.  The first three graphs that I'll show you describe the distribution of three prenatal care indices by maternal race and ethnicity, and in these first three graphs, the height of the bars represent the level of prenatal care. 

So this first graph shows the percent distribution of the trimester of first prenatal care visit by maternal race and ethnicity, and as you can see, nearly 90 percent of non-Hispanic white women (inaudible) to prenatal care in their first trimester while non-Hispanic black and Hispanic women were more likely to receive no prenatal care.  This graph shows a percent distribution of the Kessner Index by maternal race and ethnicity, and again, you see that non-Hispanic women were more likely to receive adequate prenatal care while non-Hispanic black and Hispanic women were more likely to receive inadequate prenatal care.  This graph shows the distribution of the APNCU Index by maternal race and ethnicity, and again, you see that non-Hispanic white women were more likely to receive adequate prenatal care than non-Hispanic black and Hispanic women.  In this slide, you can also see that the percentage of women receiving prenatal care in the intensive category, which is the light blue bars, did not vary substantially by race and ethnicity. 

These next three graphs shows the percent of SGA by their prenatal care indices, and the height of the bars represents the percent of SGA infants.  So this graph shows the percent of SGA by the trimester of the first prenatal care visit.  And as you can see, white women who received care in the first trimester, only about seven percent of their infants had SGA, compared to white women who received no prenatal care, about 16 percent of their women were SGA.  Women who received no prenatal care had the greatest risk of SGA and it's especially apparent in non-Hispanic black women.  This graph shows the percent of SGA by the Kessner Index and maternal race and ethnicity, and as you can see again, women who receive inadequate prenatal care had the greatest risk of SGA.  This graph shows a percent of SGA by the APNCU index, and again, you can see women who received inadequate prenatal care had the greatest risk of SGA.  Based on the last three slides, overall, non-Hispanic white women generally had the lowest scores of SGA for any given level of prenatal care.  These tables demonstrate that, in general, the risk of SGA increases with the later initiation of prenatal care unless adequate prenatal care, regardless of maternal race and ethnicity.  This table summarizes the last three graphs that I've shown you when it's the relative risk of SGA and 95-percent confidence integrals by prenatal care index and maternal race ethnicity. 

I just want to point out three main points about this table.  The first is that the relative outcome, which is SGA, among the three prenatal care indices, were consistent within the three race ethnicity groups.  Second, that the risk of SGA increases with the late initiation of prenatal care or less adequate prenatal care.  And third, that the risk of SGA was strongest among white women in the intensive group than for non-Hispanic black and Hispanic women, and that's the highlighted group in blue.  To explore potential reasons underlying the relative differences in SGA risk among the intensive prenatal care groups, the distribution of demographic factors potentially related to a pregnancy outcome were examined, and this table presents the percentage of mothers by race ethnicity receiving adequate or intensive prenatal care by other demographic factors.  And in here, you see I have the three race ethnicity categories under each.  It's less than 20 years of age, less than high school, or unmarried.  And among the non-Hispanic white women, the intensive prenatal care group was more likely to include women less than 20 years of age and to have lower educational attainment than the adequate prenatal care group.  Conversely, among the non-Hispanic black and Hispanic women, the intensive care group was slightly less likely to include adolescent mothers and more likely to include women with higher education or who were married than did the adequate prenatal care group. 

This last table shows the adjusted odds ratio and 95-percent confidence intervals for SGA by the APNCU Index, and I'm showing the intensive category only.  So as you can see, after adjusting for parity, education, marital status, and age, the odds ratio for SGA by the APNCU group for non-Hispanic white women in the intensive category have remained slightly higher than for non-Hispanic black and Hispanic women.  In summary, it is already known, and our analysis found, that non-Hispanic black and Hispanic women had a higher likelihood of receiving less-than-adequate prenatal care than non-Hispanic white women.  Non-Hispanic white women were more likely to receive adequate care or enroll into prenatal care in their first trimester.  Associations between SGA and the receipt of prenatal care were similar across the three race ethnicity groups examined in this analysis.  In general, the risk of SGA increased with the decreasing adequacy or timeliness of care, and this relationship was strongest among white women.  However, non-Hispanic white women with intensive prenatal care, as defined by the APNCU Index, were at slightly increased risk of SGA compared to women with adequate care.  Non-Hispanic black and Hispanic women were at slightly decreased risk. 

This study had a few limitations I want to point out.  The first:  that the trimester of care does not take into account subsequent prenatal care visits.  So for example, a woman may start care in her first trimester but may never return for follow-up appointments but will be classified as starting care in the first trimester.  Second, the APNCU and the Kessner Index are based on data that are not potentially well collected on birth certificates, such as the number of prenatal care visits by a health professional and gestational age.  In addition, none of these indices provide information on the quality or content of the visits, but the implication is that with early starting of prenatal care and a reasonable number of visits that the care is continuous.  The last limitation is the lack of medical and clinical information.  So from this study, we cannot determine why women had more than the recommended number of visits and why the characteristics of these women varied by race and ethnicity.  Though the differences in relative outcome were small, additional understanding of the relationship between medical risk and additional prenatal care may help guide efforts to minimize racial and ethnic disparities on pregnancy outcomes. 

In conclusion, despite these limitations, our study demonstrated that prenatal care measured by the different indices had similar associations when predicting SGA.  This study found that the more complex indices did not provide additional information when predicting adverse pregnancy outcomes.  All these indices provided similar results when predicting SGA except within the intensive care group.  So therefore, special attention should be taken when making predictions for women who fall into the intensive prenatal care category.  Thank you.