Benefits and Limitations of Prenatal Care: Title and subTitle BreakFrom Counting Visits to Measuring Content

In this issue of THE JOURNAL, Kogan and colleagues¹ identified trends in the patterns of prenatal care utilization in the United States from 1981 through 1995. Four measures of utilization were examined. First, the authors examined patterns based on the trimester that care began and the Institute of Medicine (IOM) Index of Adequacy of Care (Kessner Index).² Clearly, the trimester that care began is an inadequate measure of prenatal care use because neither the total number of visits nor the number of visits relative to the length of gestation is incorporated. While the IOM Index does consider the number of visits, it has been criticized for failing to incorporate the full length of gestation in considering the adequacy of the number of visits. Second, because both these older indices had limitations, the authors also chose to examine patterns of prenatal care utilization using 2 newer measures: R-GINDEX³ and Adequacy of Prenatal Care Utilization (APNCU) Index.^{4 - 5} Both of these measures include a category of "intensive utilization" to categorize women who receive more than the recommended number of visits. Based on these newer indices, the authors report that utilization of prenatal care, especially intensive utilization, has been increasing steadily since 1981. The same relationship was not identified using the 2 older indices, with utilization just beginning to increase, only slightly, during the 1990s.

As noted by Kogan et al,¹ increased and early utilization of prenatal care is the goal of several national programs and policies. Underlying this approach has been the assumption that prenatal care is an important and even primary means of improving perinatal outcomes. However, based on their study of 54 million births, Kogan and coauthors¹ assert that trends toward earlier and more intensive utilization of prenatal care during the last 15 years have not yielded the expected improvements in perinatal outcomes.

Assuming that the initiation and intensity of utilization of care (as shown by the number of visits adjusted for initiation and length of pregnancy) are appropriate indices of prenatal care utilization, it is important to question why rates of preterm delivery and low birth weight have not declined significantly during the same period in which utilization of prenatal care has increased. To understand this discrepancy, 2 overlapping issues must be considered. First, does the increase in utilization include women who are most at risk for poor perinatal outcomes? Second, should early and more intensive utilization of prenatal care alone be expected to make a difference?

Social and biomedical factors can be used to identify women at risk for poor perinatal outcomes. Despite advances in understanding the biological processes underlying these outcomes, social risk factors, such as African American race,⁶ low income status,^{7 - 8} or adolescent age group,⁹ explain much of the variability in pregnancy outcomes in the United States. Considering that the majority of deliveries that result in a preterm or low birth weight infant involve women who are socially at high risk and for whom there are no apparent "medical" risk factors, it is important to determine whether prenatal care utilization increased for women who are socially at high risk.

Kogan et al¹ found a trend toward increased early initiation of prenatal care among adolescents (48.4% in 1981 vs 61.8% in 1995)¹ and that intensive utilization, by both the R-GINDEX and the APNCU Index, also increased for adolescents. The authors do not present comparable data for African American or low-income women. However, analyses of the factors associated with intensive prenatal care utilization (Table 5 of the article¹ ) shed some light on this issue. In 1981, African American women were slightly more likely to receive intensive prenatal care. But by 1995, the relationship between this marker of social risk and prenatal care utilization had weakened for both indices. Measuring utilization by the APNCU Index, African American women were less likely to receive intensive amounts of prenatal care. Using education as a crude proxy for socioeconomic status, the disparity with social risk is even more striking. In 1981, compared with women having less than a high school education, women who had graduated from high school and women who had at least some college were slightly less likely to have intensive utilization, as measured by the R-GINDEX, and slightly more likely, as measured by the APNCU Index. By 1995, these more highly educated women were much more likely to have intensive utilization regardless of the index used. Thus, it appears that 2 of the groups that are socially at high risk, African American and low-income women, did not share in the trend toward increased utilization of prenatal care at the intensive levels. If the groups most at risk are not experiencing increases in utilization, it is not surprising that outcomes have changed little.

It is also unclear whether early and more intensive utilization of prenatal care reduces risk to the extent many have assumed it will, especially among those most at risk. The content of this care is not captured in any of the indices. Prenatal care is clearly not a homogeneous entity. One previous study of national data found differences in the content of care received by African American women as compared with white women.¹⁰ Measuring prenatal care utilization reveals little about how prenatal care addresses factors that are looming more prominent in the causation of low birth weight and preterm delivery, such as bacterial infections of the genitourinary tract. The health of a woman prior to pregnancy may also strongly influence her risk of adverse outcomes. A wide range of chronic diseases has been reported to increase the risk of adverse pregnancy outcomes and preterm delivery specifically, including asthma,¹¹ diabetes,¹² hypertension,¹³ and renal disease.¹⁴ Furthermore, chronic diseases are more prevalent among the same groups of women who are most at risk for poor outcomes, namely African American^{15 - 18} and low-income^{16 ,19 - 20} women. Yet chronic diseases are difficult to address effectively with prenatal care. Increasing the utilization of prenatal care will not improve a woman's preconceptional health. Social factors also have strong influences, and it may be that changes in the woman's environment and resources are more important than medical services provided through prenatal care. Prenatal care serves as the entry point for Medicaid and select social services for the pregnant woman, but indices of utilization provide little information on this issue. Without these additional nonmedical services, prenatal care may have limited benefits for women who are socially at high risk. The stressful and impoverished environment in which many minority and low-income women live may be a fundamental factor that influences pregnancy outcomes but cannot be overcome easily with narrowly targeted interventions. Improvements in perinatal health may depend on the commitment to address underlying societal inequities.

Kogan et al¹ suggest that insensitive measures of prenatal care utilization have been used for too long.¹ Two better measures of prenatal care utilization are now available, but the authors do not indicate under which circumstances each might be used. What is the criterion standard? How can public health agencies at the federal, state, and local levels adopt these new measures and eliminate the older ones? The National Center for Health Statistics and state health agencies should ensure that all agencies have the software to calculate these new indices. The findings from the study by Kogan et al¹ provide baseline rates for these new indices, from 1981 onward, against which future rates can be compared to evaluate emerging trends. However, clinicians and researchers must continue to push forward with the development of still more informative indices of prenatal care utilization. In a commentary on the quality of vital perinatal statistics data, Kirby²¹ noted a willingness to accept the status quo regarding information collected on the birth certificate. Rather than modifying the birth certificate to allow collection of data on key indicators of prenatal care, such as content and timing of components of care, clinicians and researchers have been content with perfecting an index of adequacy based on available vital statistics data. Kirby²¹ suggests the need to "start the quest for the broader set of indicators of the content of prenatal care, in terms of diagnostic tests, risk assessments, procedures, health promotion/education and counseling, all in relation to their timing in pregnancy."

While prenatal care clearly has some benefits, the public health and clinical community may have oversold the idea of increasing prenatal care utilization as a way of decreasing low birth weight and preterm delivery. There is little reason to believe that simply increasing utilization of prenatal care would have the desired effects on pregnancy outcome. Without carefully looking at the characteristics of the pregnancies and the content of the care used to address those characteristics, it is somewhat naive to think that change will take place. Future investigators must perform more detailed and carefully designed studies examining how quality of care before and during pregnancy relates to perinatal outcomes. These studies also will require more detailed data beyond that on timing and number of prenatal visits than currently are collected and reported in vital statistics reports.