Maternal pre-pregnancy BMI, gestational weight gain, and infant birth weight: A within-family analysis in the United States
Introduction
The last two decades have seen a rapid increase in the prevalence of overweight and obesity among American women of reproductive age (National Center for Health Statistics, 2014). Among the women who had delivered live infants, the rate of pre-pregnancy overweight and obesity increased from 23.2% in 1993–1994 to 44.8% in 2009 (Kim et al., 2007, Fisher et al., 2013). As another important concern on pregnant women, excessive gestational weight gain has become increasingly common, even for the women with high body mass index (BMI) before pregnancy (Frederick et al., 2008). In contrast, while maternal obesity is on the rise, pre-pregnancy underweight and gestational inadequate weight gain due to deficient nutrient intake or supplementation remain a significant health problem among low socioeconomic status women. A recent report by the Institute of Medicine (IOM) shows about 32% of the American low income underweight women had inadequate weight gain during pregnancy in 2007 (IOM, 2009). The high prevalence of unhealthy preconception body weight and inappropriate gestational weight gain among pregnant women is of great concern for the public health community. This is because unhealthy BMI before pregnancy and inappropriate weight gain during pregnancy have been linked to poor birth outcomes especially high birth weight (HBW, birth weight >4000 g) and low birth weight (LBW, birth weight <2500 g).
Both HBW and LBW have lasting negative impacts on one's health, education, and socio-economic outcomes. HBW babies are at a greater risk for overweight, obesity, diabetes, cancer, and other disorders later in life (Hjalgrim et al., 2003, Danielzik et al., 2004, Harder et al., 2007, Wei et al., 2007, Mandl et al., 2009). Moreover, HBW negatively affects cognitive function, learning, and school performance (Richards et al., 2001, Kirkegaard et al., 2006, Cesur and Kelly, 2010). LBW is also a costly birth outcome. The short run excess hospital costs of all the LBW singleton births in 1989 were estimated to be at least 1.07 billion (Almond et al., 2005). In the long term, LBW leads to poor health (Currie and Hyson, 1999), developmental problems (Hack et al., 1995), low educational attainment (Conley and Bennett, 2000), and adverse labor market outcomes (Behrman and Rosenzweig, 2004). In the United States, reducing the incidence of adverse birth outcomes has been an important goal of several large-scale social programs. In this sense, findings from investigating the effects of pre-pregnancy BMI and gestational weight gain on newborn birth weight will have strong policy implications.
This research is also related to the empirical studies on Barker's fetal origins perspective (Barker, 1990). By Barker's theory and its extension, under-nutrition and over-nutrition during pregnancy due to adverse fetal environment presage some of the most common health disorders in adult life. Recent studies further show the “fetal origins” effects can extend to a broader range of lifetime outcomes and in particular, poor health at birth is a key pathway through which deficient or excessive nutrient intake during fetal development exerts persistent effects (Almond and Currie, 2011). This study will shed new light on such a pathway, since gestational weight gain is a good measure of nutrition in utero and both HBW and LBW are crucial indicators of poor infant health. Furthermore, exploring the impact of pre-pregnancy BMI on birth weight is important too, because the corresponding estimates capture the intergenerational returns by promoting healthy preconception weight among women. Such often-ignored returns should be incorporated into the cost-benefit analysis of various weight management programs for women of reproductive age (Siega-Riz and Laraia, 2006).
To date, the relationship of pre-pregnancy BMI and gestational weight gain to infant birth weight has not been well established. Some studies only look at the association between preconception BMI and birth weight (Abenhaim et al., 2007, Gilboa et al., 2008, McDonald et al., 2010). Lacking gestational weight gain data, these studies are unable to address how pre-pregnancy BMI affects birth weight net of its impact that operates through influencing gestational nutritional intake, an important policy-relevant issue. Their results are quite mixed. The second body of the literature evaluates the effects of gestational weight gain only, without controlling for preconception body weight (Hediger et al., 1989, Abrams and Selvin, 1995, Hickey et al., 1996, Ludwig and Currie, 2010, McDonald et al., 2011). However, the omitted pre-pregnancy BMI clearly biases the parameter estimate on weight gain, since women in various BMI categories differ systematically in gestational weight gain. The third literature investigates the combined effects of the two birth weight inputs (Doherty et al., 2006, Rode et al., 2007, Frederick et al., 2008, Crane et al., 2009, Bodnar et al., 2010). Past studies on this topic, like the previous two strands of research, rarely deal with the unobserved generic, mother, or family level factors. Such factors, when not controlled for, will confound the relationship between the two inputs and birth weight. Indeed, the results of these studies are varying and inconsistent. Finally, the datasets used in the whole literature above are often selective, cross-sectional, and lack of important socioeconomic control variables.
This study uses a very large dataset of about 0.3 million sibling births to thoroughly evaluate the effects of pre-pregnancy BMI and gestational weight gain on infant birth weight. This dataset has several noteworthy advantages. One, it is constructed from the natality records of all the live births in two states, while much of the prior research uses convenience samples. Two, the dataset records maternal weight and height before pregnancy plus weight at delivery, which makes it feasible to create indicators of both pre-pregnancy BMI and gestational weight gain. Three, it provides a rich set of infant, mother, and family control variables. Four, this panel dataset allows a comparison of birth outcome of singleton births to the same mother. This within-family design can remove the bias due to unobserved mother heterogeneity. Five, the large sample size gives more precise estimates, relative to the previous estimates in the literature.
With this unique dataset, mother fixed effects models are applied to address the effects of maternal preconception body weight and gestational weight gain on the mean birth weight and two tails of the birth weight distribution. The benchmark results suggest preconception overweight, preconception obesity, and excessive gestational weight gain all significantly increase the risk of having a HBW baby, while underweight before pregnancy and inadequate gestational weight gain lead to elevated risks of delivering a LBW infant. Then, a series of sensitivity checks or extensions are performed, where the new results are highly consistent with the baseline estimates.
Section snippets
Data
The dataset of sibling births is constructed from vital statistics natality records in the states of Pennsylvania and Washington1. These natality records in general cover all the live births every year. However, this
Methods
To examine the relationship of maternal pre-pregnancy body weight and gestational weight gain to infant birth weight, the following linear model is used in the benchmark analysis:where is a birth weight measure of infant born to mother ; is a vector of three pre-pregnancy BMI categories; is a vector of two gestational weight gain indicators; and is a rich set of infant, mother, and family level control variables (Voigt et al., 2004).
Results
Table 1 reports the summary statistics of the three primary birth weight outcomes, infant characteristics, and the key parental control variables11. Column (1) focuses on the full sample of sibling births. It shows 11% of all the infants are HBW and 5% of them are LBW. About 21% of the babies were delivered by obese women with pre-pregnancy
Conclusion
Poor birth outcomes such as HBW and LBW impose a huge burden on families, the healthcare care system, education, and social services. As such, improving the health and well-being of infants has been a crucial public health goal in the United States. Both pre-pregnancy BMI and gestational weight gain are two key determinants of infant health. However, the exact relationship between the two inputs and newborn birth weight has not been well understood. This article uses a large sample of sibling
Acknowledgements
This paper benefits from helpful comments by Josh Berning, Jeremy Bray, Karen Conway, Anca Cotet, Angela Dills, Zheng Fang, Gregory Gilpin, Jennifer Graves, Tiffany Green, Robert Pollak, Kathleen M. Rasmussen, Anna Maria Siega-Riz and participants of Association for Public Policy Analysis and Management Research Conference, Eastern Economic Association Meeting, Southern Economic Association Meeting, and the Work, Families and Public Policy Workshop in Washington University in Saint Louis.
References (50)
- et al.
Maternal weight gain pattern and birth weight
Obstetrics & Gynecology
(1995) - et al.
Severe obesity, gestational weight gain, and adverse birth outcomes
The American Journal of Clinical Nutrition
(2010) - et al.
The effect of gestational weight gain by body mass index on maternal and neonatal outcomes
Journal of Obstetrics and Gynecology Canada
(2009) - et al.
Pre-pregnancy body mass index and pregnancy outcomes
International Journal of Gynecology & Obstetrics
(2006) - et al.
Investing in early human development: timing and economic efficiency
Economics & Human Biology
(2009) - et al.
Is obesity still increasing among pregnant women? Pre-pregnancy obesity trends in 20 states, 2003–2009
Preventive Medicine
(2013) - et al.
Use of spline regression in an analysis of maternal pre-pregnancy body mass index and adverse birth outcomes: does it tell us more than we already know?
Annals of Epidemiology
(2008) - et al.
Prenatal weight gain patterns and birth weight among nonobese black and white women
Obstetrics & Gynecology
(1996) - et al.
Maternal hormones linking maternal body mass index and dietary intake to birth weight
American Journal of Clinical Nutrition
(2008) - et al.
The association between pregnancy weight gain and birthweight: a within-family comparison
The Lancet
(2010)
High gestational weight gain and the risk of preterm birth and low birth weight: a systematic review and meta-analysis
Journal of Obstetrics and Gynecology Canada
Implementation of the new institute of medicine gestational weight gain guidelines
Journal of Midwifery & Women's Health
Weight gain in women of normal weight before pregnancy: complications in pregnancy or delivery and birth outcome
Obstetrics & Gynecology
The relationship between maternal characteristics, birth weight and pre-term delivery: evidence from Germany at the end of the 20th century
Economics & Human Biology
Effect of pre-pregnancy body mass index categories on obstetrical and neonatal outcomes
Archives of Gynecology and Obstetrics
Pregnancy weight gain: still controversial
The American Journal of Clinical Nutrition
The costs of low birth weight
The Quarterly Journal of Economics
Human capital development before age five
The fetal and infant origins of adult disease
British Medical Journal
Returns to birthweight
Review of Economics and Statistics
From the cradle to the labor market? The effect of birth weight on adult outcomes
The Quarterly Journal of Economics
Excessive early gestational weight gain and risk of gestational diabetes mellitus in nulliparous women
Obstetrics and Gynecology
From cradle to classroom: high birth weight and cognitive outcomes
Forum for Health Economics & Policy
Is biology destiny? Birth weight and life chances
American Sociological Review
Is the impact of health shocks cushioned by socioeconomic status? The case of low birthweight
American Economic Review Papers and Proceedings
Cited by (43)
You'll never walk alone – The effects of walkability on pregnancy behaviors and outcomes
2023, Economics and Human BiologyIs WIC effective in improving pregnancy-related outcomes? An empirical reassessment
2022, Economics and Human BiologyWhen children have children: The effects of child marriages and teenage pregnancies on early childhood mortality in Bangladesh
2020, Economics and Human BiologyMaternal prepregnancy BMI and size at birth: race/ethnicity-stratified, within-family associations in over 500,000 siblings
2020, Annals of EpidemiologyCitation Excerpt :Similarly, contextual factors that may buffer or strengthen maternal BMI–outcome associations (e.g., social support, built, and chemical environments) also vary by race/ethnicity, but such effect measure modifiers are typically unmeasured. A handful of studies use within-family study design to control for unmeasured confounding by family-level factors [13,14], comparing outcomes among siblings born to the same woman who had varying BMI prior to each pregnancy [13–19]. Despite possible racial/ethnic differences in confounders and effect modifiers, within-family associations of prepregnancy maternal BMI on size at birth have not been examined within racial/ethnic subgroups.