Prenatal care and birthweight in Mexico

ABSTRACT

We estimate the marginal impact of prenatal care (PNC) on birth outcomes in Mexico using nationally representative data on about 14 million births from 2009 to 2014. Given the possible self-selection into PNC, we attempt to identify the causal impact of PNC on birth outcomes by estimating an instrumental variable model. We find positive impacts of increased prenatal visits on birthweight (BW), length and apgar score of the newborn. Moreover, the impacts of PNC on birth outcomes differ by mother’s education, development level of the municipality where the mother resides and BW distribution. We also find suggestive evidence that PNC visits affect birth outcomes through the reduction in pre-term births.

KEYWORDS:

• Prenatal care
• birthweight
• neonatal outcomes
• Mexico

JEL CLASSIFICATION:

• I10
• I12
• J13

View correction statement:

Corrigendum

Acknowledgement

This work benefitted from valuable comments by several anonymous referees, as well as Don Freeman, Patricia Triunfo, Prasun Bhattacharjee, Troy Quast, seminar participants at the Sam Houston State University, and numerous participants at the 2016 SEA conference, 2017 EEA, and 2017 International Economic Association conference. We thank Stephanie Luk Bounsawat for excellent research assistance.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ The World Health Organization (WHO) considers newborn infants as LBW and very low birthweight (VLBW) if the birthweight is less than 2500 g and 1500 g, respectively (regardless of gestational age). The World Bank estimates that more than 20 million infants are born every year as LBW babies.

² The apgar score is a composite index of a child’s health at birth and take into account Activity (muscle tone), Pulse (heart rate), Grimace response (reflex irritability), Appearance (skin colour) and Respiration (breathing rate and effort). Newborns are usually evaluated at one and five minutes after birth. The maximum value for each component is 2 points. The score ranges from zero to ten with higher scores indicating better health. We use the 5-minute apgar score in this study.

³ Angrist and Pischke (2009) affirm that marginal effects from binary choice models are same as linear probability model estimates.

⁴ The Mexican government recommends at least five PNC visits, while WHO recommends four PNC visits before delivery.

⁵ CONEVAL (2007) provides a detailed explanation on how the development level of each municipality is computed.

⁶ Removing these municipalities reduces the analysis sample by 1.25%.

⁷ We further add mother’s employment as an additional covariate to explore the channel of increased resources; however, our results are similar to the main findings in Table 3 (results not shown here but available upon request).

Additional information

Funding

We gratefully acknowledge financial support of the EURECA-FAST grant from the Sam Houston State University.