The Effectiveness of Prenatal Care in Preventing Maternal and Delivery Complications:

This study utilizes Birth Certificate Data from a large urban county in Tennessee for 2004, 2005 and 2006 to evaluate the effectiveness of prenatal care in preventing potentially avoidable maternal complications (PAMC) among different racial and ethnicity groups.  The complications include premature delivery, deliveries complicated with anemia, intrauterine death, complications associated with drug, alcohol or tobacco abuse and premature rapture of membranes which all can be reduced through effective prenatal care. 

Most previous studies on the effectiveness of prenatal care have appeared in medical and clinical journals.  They have mostly used a one-stage, single-equation estimation approach and treated prenatal care as exogenously determined, thus giving rise to the problem of endogeneity.  Consequences might have included biased estimates of the effects of prenatal care utilization and possible missing of important differences in health care access among ethnic groups (non-Hispanic black, non-Hispanic white, non-Hispanic other race, and Hispanics). 

This study applied a simultaneous equation model that included predicted rather than reported use of prenatal care to correct for the inconsistency and biasedness that can result from the one-stage estimation approach.  A key finding of our simultaneous equation estimation was that Black mothers were twice more likely than Hispanics to receive prenatal care and they were also less likely to experience maternal complications and adverse birth outcomes, a surprising result.  We also found that mother’s age and pre-pregnancy BMI significantly increased the likelihood of experiencing PAMC and these were consistent with the findings of previous studies.

               To improve the model and further examine the interrelationships among prenatal care, race and ethnicity, and PAMC, we merged the Hurdle estimation approach into the simultaneous-equation 3SLS model and estimate PAMC and prenatal care simultaneously in a two-stage, Hurdle process.  This combined approach confirmed the inverse relationship between prenatal care and PAMC, with estimates of the effects of prenatal care greater than those found in previous studies in medical and clinical journals.  Significantly, the Black race, women ages 30-35, and women with lower years of education were now less likely to receive prenatal care and more likely to experience PAMC.  We finally applied the Blinder Oaxaca decomposition technique to account for observed and unobserved factors causing the disparities among the different racial and ethnic groups in the receipt of prenatal care and in the likelihood of experiencing PAMC.  Results demonstrate that PAMCs are more relevant for minority women than for whites.  Our results suggest, among other things, that public policies that encourage greater utilization of prenatal care by minority mothers are likely to produce greater returns to investment.