TY - JOUR
AU - Harris,Jeffrey E.
AU - Sosa-Rubi,Sandra G.
TI - Impact of "Seguro Popular" on Prenatal Visits in Mexico, 2002-2005: Latent Class Model of Count Data with a Discrete Endogenous Variable
JF - National Bureau of Economic Research Working Paper Series
VL - No. 14995
PY - 2009
Y2 - May 2009
DO - 10.3386/w14995
UR - http://www.nber.org/papers/w14995
L1 - http://www.nber.org/papers/w14995.pdf
N1 - Author contact info:
Jeffrey E. Harris
Department of Economics, E52-422
MIT
77 Massachusetts Avenue
Cambridge, MA 02139
Tel: 617/253-2677
Fax: NA
E-Mail: jeffrey@mit.edu
Sandra G. Sosa-Rubi
Centro de Investigación en Sistemas de Salud
Instituto Nacional de Salud Pública
Cuernavaca, Morelos 62508, Mexico
E-Mail: srubi@insp.mx
AB - We employ a latent class model to assess the impact of Mexico's Seguro Popular ("SP") program on the number of prenatal visits in a cross-sectional sample of 4,381 women who gave birth during 2002-2005. We specify an ordered probit model to permit a pregnant woman's probability of membership in one of three latent classes to depend on observed covariates. In the ordered probit model, enrollment in SP is explicitly treated as an endogenous variable. We model the number of prenatal visits, conditional upon membership in a particular latent class, as a Poisson regression. We employ the EM algorithm to reduce the computational burden of model estimation. At any iteration of the algorithm, the parameters of the model of latent class membership can be estimated separately from the parameters of the model of prenatal care utilization. We find that enrollment in SP was associated with a mean increase in 1.65 prenatal visits during pregnancy. Approximately 59 percent of this treatment effect is the result of increased prenatal care among women in the first latent class, that is, women who had with little or no access to care. The remaining 41 percent of the treatment effect is the result of a shift in membership from the second to the third latent class, which we interpret as increased recognition of complications of pregnancy prior to labor and delivery. Our model has a better fit and predicts a larger impact of SP than alternative models that relax the assumption of endogeneity, do not impose ordering on the latent classes, or incorporate only two latent classes. Our findings are consistent with prior work on the favorable impact of SP on maternal health (Sosa-Rubí, Galárraga, Harris 2009).
ER -