Infants born small-for-gestational age (SGA) have an increased risk of perinatal morbidity and mortality [1] and later life health problems such as cardiovascular disease, obesity, type 2 diabetes and hypertension in adulthood [2-5], when compared with infants whose weight is considered appropriate-for-gestational age. Research examining socioeconomic position (SEP) at the time of pregnancy in developed countries shows that SEP is inversely associated with the prevalence of adverse birth outcomes, including SGA infants [6]. For example, using family income to represent SEP, Joseph et al. [7] found that women in the lowest family income group in Canada had an SGA rate 34% higher than that in women from the highest family income group. In a more recent study Shankardass et al. [8] used multiple measures of income and found the risk of delivering an SGA or spontaneous preterm birth infant increased with each decreasing quantile of family income. Studies examining relations between SEP and SGA infants using other single measures such as maternal education, neighborhood poverty and occupation status have also reported inverse associations [6].

The “American Dream” represents the belief that one can start out at a lower SEP as a child in the United States (US) and climb the socioeconomic ladder through hard work and by taking advantage of available opportunities [9-12]. This concept of climbing the socioeconomic ladder from childhood to adulthood is called socioeconomic mobility; it can be defined as the difference in income, wealth or occupation in adulthood from that of one's family's when he/she was a child [13]. Cross-country comparisons show that the higher the income inequality the lower the socioeconomic mobility across generations [12-14]. Given that the US has higher income inequality than other developed countries children living in Scandinavian countries and Canada have a greater chance of attaining the “American Dream” than children living in the US [9,12-14].

For women there are multiple theories about how climbing the socioeconomic ladder might or might not affect health and reproductive outcomes. In one framework, exposure to social disadvantage during critical times of growth, i.e. in utero or early childhood, regardless of later exposures, could raise the risk of delivering an SGA infant. [15]. A different framework describes exposure to social disadvantage as having a cumulative or additive effect that is ongoing [15]. And yet another framework, the pathways or synergistic model, posits that exposure to poor socioeconomic conditions in later years are probabilistically linked to one's exposure to poor socioeconomic conditions during early years; together these exposures jointly influence the probability of having an adverse birth outcome [15].

While these different frameworks have been debated for some time only a handful of studies have examined associations between maternal socioeconomic mobility and birth outcomes in part due to limitations in accessing data on socioeconomic measures at multiple time points during a woman's life [15]. Consequently, the majority of socioeconomic mobility studies published to date [16-28] are European-based where there exists the ability to link vital and other administrative records over women's lifecourse. Studies conducted in the US [16,18-22,26,29] predominantly use birth files linked across generations and ecological socioeconomic measures obtained from US Census data [19,20,22,29,30]. These US-based studies overall show that women who reside in an impoverished neighborhood during their childhood, but move upward to a wealthy neighborhood experience better birth outcomes than women with lifelong residence in an impoverished neighborhood. This relationship may vary by race/ethnicity according to some studies.

Investigations of socioeconomic mobility and adverse birth outcomes have focused mainly on low birthweight and preterm birth [16-21,23-27]. In our review of the literature we found only one study that examined fetal growth. In this study, Love et al. [22] examined the concept of maternal “weathering” -early deterioration in women's physiological health due to cumulative social disadvantage [31,32]- in the context of neighborhood economic environment over women's lifecourse. Results showed that among African-American women, the risk of delivering an SGA or LBW infant significantly increased as length of time living in an impoverished neighborhood increased. The risk of delivering an SGA or LBW infant decreased as length of time residing in a non-impoverished neighborhood increased.

The rising concern over decreased socioeconomic mobility and the limited information on the relationship between fetal growth and socioeconomic mobility motivated our study's goal to assess whether changes in women's SEP from childhood to adulthood are associated with the risk of delivering an SGA infant. This study expands on previous work in the area of socioeconomic mobility and birth outcomes by using multiple individual-level socioeconomic measures to construct a composite score representing SEP at childhood and adulthood.

In this sample 8.2% of the 2,463 women delivered an SGA infant. Approximately a quarter of the sample was non-Hispanic Black (25.9%), the mean maternal age was 26.7 years (SD=5.7) and 41.7% were first-time mothers. Table 2 presents the SGA rate by childhood SEP, adulthood SEP, socioeconomic mobility and maternal characteristics for the POUCH Study participants. One fourth of the women moved upward in their SEP, one fourth moved downward and a little more than half stayed at their childhood SEP. The rate of delivering an SGA infant was highest among non-Hispanic Blacks (13.9%), women under 20 years of age (15.6 %), first time mothers (10.4%) and women with normal BMI (9.9%). Among non-Hispanic Whites the rate of delivering an SGA infant was 6.9%.

We found that among women of reproductive age higher SEP and a history of upward socioeconomic mobility were both associated with a lower likelihood of delivering an SGA infant. In further exploration, race-stratified results showed that both non-Hispanic White and non-Hispanic Black women may receive some benefits from upward socioeconomic mobility. These findings are particularly relevant in the context of the US's current political/economic realities. The US is seen as a “land of opportunity” where children have the option to move-up the socioeconomic ladder as adults [9-12]. Our findings suggest that among women who move-up there are positive health consequences for their offspring. Unfortunately, recent studies by Chetty, Hendren, Kline, Saez, and Turner [9,10] show that rates of socioeconomic mobility within the US have been stagnant over the past few decades while income inequality has increased and the size of the middle class has decreased. These authors concluded that the parents a child is born to may be more important in today's world than yesterday's [9,10]. This concept takes on additional meaning if the impact begins in-utero and manifests in alteration of fetal growth.

There are several potential explanations for our findings. Using the cumulative stress or weathering framework [31,32] we might infer that women who move upward in SEP do not accumulate the “wear and tear” to their body's allostatic load [37,38] that is prevalent among women who do not move-up. In turn, lower levels of “wear and tear” create more optimal inutero environments for the growing fetus. We also noted that downward mobility from upper/middle to lower class was associated with a lower risk of delivering an SGA infant when compared to the SGA risk of the SEP group joined. Upon further inspection, our race-stratified results suggest this relationship was only seen among non-Hispanic White women. One interpretation of our main results is that women retain a health advantage or that a health advantage coincides with a stronger start conferred by the higher childhood SEP. Studies with prospectively collected health indicators beginning in childhood could help shed light on the biological basis of critical periods that later translate into maternal impact on fetal growth.

This study contributes to the current literature by investigating the influence of SEP, measured at the individual-level, on fetal growth and extending the inquiry to include socioeconomic mobility up through the period of pregnancy. The majority of studies examining the impact of socioeconomic mobility on pregnancy outcomes have used a single indicator to denote SEP. Our study's use of multiple indicators to denote SEP during childhood and adulthood may have minimized misclassification bias that can occur with temporal changes in the SEP assigned to any single indicator such as occupation [39-41]. While all our SEP indicators were gathered through self-report, and therefore there may be some bias, mothers offered this information in mid-pregnancy before knowing if the infant was or was not SGA. Finally, this study enrolled a socioeconomically diverse population thus permitting us to observe a full range of socioeconomic mobility.

Despite the many strengths discussed above, there are limitations in our study that merit consideration. As mentioned in our description of the POUCH Study sample, the percentage of non-Hispanic Black women over 30 years of age was lower than that in the communities from which POUCH Study participants resided; this may limit the generalizability of our results to this group. While our overall number of women enrolled in POUCH was quite large, our study lacked power in our race-stratified analyses. Hence we were not able to draw meaningful conclusion in some comparisons for non-Hispanic Black women. Our study, as most studies in this area, used birthweight standardized growth curves, which can be biased at early gestations [42,43]. SGA (<10^th percentile) was used as an indicator of poor fetal growth, a common approach in similar studies. Not every infant in this distribution tail experienced poor fetal growth, some are constitutionally small. We used a straightforward approach to building SEP composite measures with equal weighting for each indicator. Also, we broadly grouped measures of SEP into two time points, childhood and adulthood; women's SEP during each of the two periods could have fluctuated. While this level of misspecification is likely to be small and minimally related to fetal growth (non-differential), our approach might have underestimated true effect sizes. Socioeconomic indicators used to create the composite measure for childhood SEP were collected from the POUCH Study participant and not directly from her parents. Though this is a limitation, at least one study [44] examining proxy reporting of SEP showed reasonable concordance in parent and child responses. Perhaps most importantly, causal inferences from socioeconomic mobility studies are challenging in part due to the possibility of indirect selection [45], confounding due to unmeasured individual factors established early in life that may influence both health and SEP across the lifecourse. Studies examining socioeconomic mobility are rarely if ever be randomized; that leaves us with associations from observational studies and uncertainty as to how much effect is due to selection and how much is explained by benefit/harm of changes in SEP [46]. While causation is difficult to infer, the patterns in observed and descriptive data on fetal growth, SEP and socioeconomic mobility can guide medical and public health resource allocation and levels of medical surveillance or intervention during pregnancy.

Our findings invite future studies that might ask, what are the pathways (e.g., biological, behavioral, psychosocial, social structure) through which upward social mobility could lead to improved birth outcomes for both non-Hispanic White and non-Hispanic Black women? If indirect selection plays some explanatory role for the upward mobile effect, what is it that these women do/experience that provides an advantage for fetal growth? Why do White women who experience downward mobility, on average, retain an advantage when compared to women in the social class they join? Why is this same pattern not observed for Black woman? Would investments in public education and income stability, major contributors to socioeconomic mobility, improve birth outcomes for disadvantaged women? These questions point to fertile areas for future research. Answers to these questions could lead to more effective interventions aimed at reducing poor fetal growth.