The changing association between prenatal participation in WIC and birth outcomes in New York City: What does it mean?

Abstract

Journal of Policy Analysis and Management, Vol. 24, No. 4, 687–690 (2005) © 2005 by the Association for Public Policy Analysis and Management Published by Wiley Periodicals, Inc. Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/pam.20132 Many previous studies of WIC conclude that “WIC works,” but in their article in this issue, Ted Joyce, Diane Gibson, and Silvie Colman come to the opposite conclusion (Joyce, Gibson, & Colman, 2005). They arrive at this position even though their actual estimates are consistent with those of previous studies. For example, like our largescale study of women on Medicaid in 19 states (Bitler & Currie, 2005), they find that prenatal participation in the WIC program reduces the incidence of low birth weight and preterm birth. We also find that WIC participation is associated with decreases in birth weight adjusted for gestation, the nights an infant or woman was in the hospital at delivery, and positively associated with weight gain during pregnancy, gestation, birthweight, and use of prenatal care during the first trimester. Why then are Joyce, Gibson, and Colman’s conclusions so different than those of previous authors? The crux of the matter is that Joyce et al. reject the idea that WIC could have any effect on preterm birth, while neglecting discussion of other possible positive impacts of WIC. Hence, they reason that any estimated effects of WIC on preterm birth reflect selection effects in the data and ought to be discounted out of hand. Further, they argue that WIC should only affect birth weight through effects on fetal growth, and they do not find any effects on fetal growth (except in their subsample of Black twins). We think that this line of reasoning is flawed on several counts, and perhaps too narrowly focused. First, Joyce et al. overstate the degree of medical consensus that exists over the question of whether or not prenatal intervention can affect the probability of preterm birth. They cite several studies describing specific clinical interventions that were not effective in preventing preterm births. But the clinical literature also suggests, for example, that smoking causes preterm labor, so that interventions that were effective in reducing smoking might be expected to have an effect on prematurity (compare Werler, 1997).1 Similarly, gestational diabetes is a common complication of pregnancy that can lead to preterm birth, but it is easily controlled if it is detected early (Xiong, Saunders, Wang, & Demianczuk, 2001). Maternal infections are thought to account for up to a third of preterm births, and some trials have shown that treatment of high-risk women for specific infections can increase the length of pregnancy (Gibbs & Eschenbach, 1997; Goldenberg, Hauth, & Andrews, 2000; Locksmith & Duff, 2001), while others have shown a possible positive impact of treatment for wider groups of women (Lamont, Duncan, Mandal, & Bassett, 2003; Kiss, Petricevic, & Husslein, 2004; Ugwumadu, Manyonda, Reid, & Hay,