How does high socioeconomic status affect maternal and neonatal pregnancy outcomes? A population-based study among American women

Abstract

Objectives

The purpose of this study was to evaluate the effect of high SES on multiple pregnancy outcomes, while controlling for confounding factors.

Methods

Using the Healthcare Cost and Utilization Project Nationwide Inpatient Sample (HCUP-NIS), the largest American medical database including 20 % of annual hospital admissions, we studied the years 2004–2014 inclusively. We conducted a population-based retrospective cohort study consisting of women from different median household income quartiles throughout the United States. Women in the highest household income quartile were compared to those in all other lower income quartiles combined. Chi-square and Fischer exact tests were used to compare demographic and baseline characteristics. Univariate and multivariate regression analyses were carried to adjust for confounding factors, including ethnicity, pre-existing conditions, smoking status, obesity, illicit drug use and insurance type.

Results

Among 5,448,255 deliveries during the study period with income data, 1,218,989 deliveries were to women from the wealthiest median household income. These women were more likely to be older, Caucasian, and have private medical insurance (P < 0.05, all). They were less likely to smoke, have chronic hypertension, pre-gestational diabetes, and use illicit drugs (P < 0.05, all). They were less likely to develop complications including gestational hypertension (aOR 0.87 95 %CI 0.85–0.88), preeclampsia (aOR 0.88 95 %CI 0.86–0.89), eclampsia (aOR 0.81 95 %CI 0.66–0.99), gestational diabetes (aOR 0.91 95 %CI 0.89–0.92), preterm premature rupture of membranes (PPROM) (aOR 0.92 95 %CI 0.88–0.96), preterm birth (aOR 0.90 95 %CI 0.89–0.92), and placental abruption (aOR 0.89 95 %CI 0.85–0.93). They were less likely to have an intra-uterine fetal death (IUFD) (aOR 0.80 95 %CI 0.74–0.86), but more likely to deliver neonates with congenital anomalies (aOR 1.10 95 %CI 1.04–1.20).

Conclusions

Higher SES predisposes to better pregnancy outcomes, even when controlled for confounding factors such as ethnicity and underlying baseline health status. Efforts are required in order to eliminate health disparities in pregnancy.

Introduction

Socio-economic status (SES) refers to the economic and social factors that determine what position an individual or group holds within society [1]. Though a complex entity encompassing many aspects, SES is usually measured by income, occupation, education, or a combination of these [2]. SES is considered one of the most important determinants of health disparities, which are preventable differences in the burden of disease or opportunities to achieve optimal health usually experienced by socially disadvantaged peoples [3]. Lower SES is associated with various adverse health consequences such as cardiovascular disease, diabetes, and cancer [4], [5], [6] and worse outcomes, when adjusting for the baseline health status [7], [8], [9], [10].

Most research has explored the effect of low SES on pregnancy outcomes [7], [8]. In some studies, median household income based on residential ZIP code was used as a proxy for maternal SES [9], [10]. Other surrogate measures used included parental education level, occupation or income assessed via questionnaire [7], [11], [12], census [8], [13], [14] or tax records [15]. These studies establish an association between low SES and adverse pregnancy outcomes, specifically an increased risk of gestational diabetes (GDM)[7], [11], preterm birth (PTB) [13], small for gestational age (SGA) [15] and congenital anomalies [12].

Few studies have appraised the role of high SES on adverse pregnancy outcomes. These studies vary in size and location, and the specific outcomes compared. Several Canadian studies have been published. Bushnik et al. evaluated 127,694 women and found preterm birth to be associated with decreased maternal education, while Joseph et al. analyzing a population of 92 914 subjects found the risk of SGA to be inversely related to SES [13], [15]. Two smaller Chinese studies enrolled 6886 patients and evaluated 17,659 patients respectively[7], [11]. Both these studies found higher education and income were associated with decreased GDM risk. A European study consisting of 227 696 patients found PTB to be inversely associated to maternal education [8]. It is clear from this literature search that the studies are relatively small and lack American representation. Therefore, the purpose of our study is to assess the effect of high SES on multiple maternal, fetal and pregnancy outcomes, in the hopes that this knowledge will aid in recognizing health disparities for American pregnant patients in a very large database of deliveries.

Materials & methods

Study design and setting

This retrospective cohort study used all 5,448,255 births in the United States from 2004–2014 inclusively with complete SES information in the Healthcare Cost and Utilization Project, Nationwide Inpatient Sample (HCUP-NIS) [16] Database. During the study period, there was a total of 9 million deliveries, the rest of which lacked SES data. The HCUP-NIS is the largest database of healthcare inpatient encounters that includes over seven million hospital-stays per year in the United States. This database, including patient demographics, comorbidities, procedures, hospital stay, discharge diagnoses and deaths, is systematized according to the International Classification of Diseases, ninth edition, Clinical Modification (ICD-9 CM). It is comprised of inpatient stays submitted by hospitals from 48 states and the District of Columbia. The data represents 20 % of US hospital admissions annually and 96 % of the American population geographically.

By using ICD-9-CM codes for delivery-related discharge diagnoses (650.xx, 677.xx, 651.xx-676.xx; fifth digit is 0, 1, or 2), and birth-related procedural diagnoses (72.x, 73.x, 74.0–74.2), we limited our study group to admissions that ended with delivery or maternal death to exclude multiple admissions with the same pregnancy.

Study variables

Our primary outcomes were pregnancy, delivery and neonatal outcomes. Pregnancy outcomes included pregnancy induced hypertension, gestational hypertension, preeclampsia, eclampsia, gestational diabetes (GDM) and placenta previa. Delivery outcomes included preterm premature rupture of membranes (PPROM), preterm birth (PTB), placental abruption, CD, maternal infection, chorioamnionitis, postpartum hemorrhage (PPH), hysterectomy, and venous thromboembolism (VTE). Neonatal outcomes included small for gestational age (SGA) defined as birthweight less than tenth percentile for gestational age, intrauterine fetal death (IUFD), and congenital anomalies.

The independent variable is the income quartile. After developing a birth cohort for 2004–2014 inclusively, we divided this cohort into household income quartiles based on the patients’ ZIP codes as representations of household income according to the US census. This procedure allowed us to define four quartiles of SES ranging from lowest to highest. The lowest income quartile (Q1) represented an estimated median household income of up to 38,999$. The second (Q2) and third quartile (Q3) represented an estimated median household income of 39,000–47,999$ and 48,000–62 999$, respectively. For the sake of this analysis, these three groups were combined. Finally, the highest income quartile (Q4) represented an estimated median household income of at least 63,000$. For this study, the highest income quartile (Q4) constitutes the cases; all other quartiles combined constitute the controls.

Demographic characteristics, maternal baseline characteristics, and delivery and neonatal outcomes of all deliveries were identified using ICD-9 codes. Baseline maternal characteristics included patient age, race, medical insurance type, hospital type, previous caesarean delivery (CD), multiple gestations, tobacco smoking, obesity (body mass index (BMI) ≥ 30 kg/m²), pre-existing hypertension (HTN), pre-existing diabetes, pre-existing thyroid disease, in vitro fertilization (IVF), and illicit drug use.

Statistical analysis

Analyses were performed with the SPSS version 25.0 statistical software package (IBM corporation, Chicago, USA). Chi-square tests were used to compare the demographic and baseline characteristics among both groups. A p value < 0.05 was recognized as statistically significant. Univariate and multivariate logistic regression analysis was performed to determine associations between household income grouping (high vs other) and obstetrical and neonatal incomes through the estimation of unadjusted and adjusted odds ratios (aORs) and 95 % confidence intervals (CIs), respectively. Any of the baseline characteristics (Table 1) with difference between the groups (p < 0.05) were selected as potential confounding factors and adjusted for in our multivariate logistic regression analysis. According to the Tri-Council Policy Statement (2018), institutional review board approval was not required since this study used data that is publicly available and de-identified.

Table 1.

Maternal Characteristics- Highest income quartile vs all others.

Characteristics	Highest income quartile N = 1,218,989	Other N = 4,229,266	P-value
Maternal age (years)			<0.0001
<25	230,046 (18.9 %)	1,768,986 (41.8 %)
25–34	697,486 (57.2 %)	1,944,196 (46 %)
≥35	291,1267 (23.9 %)	515,775 (12.2 %)
Race			<0.0001
White	683,727 (61.9 %)	189,7782 (50.6 %)
Black	81,034 (7.3 %)	605,972 (16.1 %)
Hispanic	156,640 (14.2 %)	904,706 (24.1 %)
Asian and Pacific	121,452 (11 %)	139,672 (3.7 %)
Native American	4577 (0. 4 %)	33,348 (0.9 %)
Other	56,694 (5.1 %)	170,698 (4.5 %)
Plan type			<0.0001
Medicare	4892 (0.4 %)	31,737 (0.8 %)
Medicaid	250,187 (20.5 %)	2,106,334 (49.9 %)
Private including HMO	906,288 (74.4 %)	1,834,299 (43.5 %)
Self-pay	25,596 (2.1 %)	116,972 (2.8 %)
No charge	890 (0.1 %)	7272 (3 %)
Other	29,802 (2.4 %)	124,982 (3 %)
Hospital type			<0.0001
Rural	4466 (1.4 %)	203,978 (17.5 %)
Urban	318,095 (98.6 %)	964,714 (82.5 %)
Previous CS	209,749 (17.2 %)	703,429 (16.6 %)	<0.0001
Obesity	41,935 (3.4 %)	219,627 (5.2 %)	<0.0001
Tobacco Smoking during pregnancy	25,113 (2.1 %)	269,910 (6.4 %)	<0.0001
Chronic HTN	20,487 (1.7 %)	90,918 (2.1 %)	<0.0001
Pregestational DM	8062 (0.7 %)	41,979 (1 %)	<0.0001
Illicit drug use	7985 (0.7 %)	73,882 (1.7 %)	<0.0001
Thyroid disease	52,488 (4.3 %)	102,873 (2.4 %)	<0.0001
IVF	5859 (0.5 %)	4580 (0.1 %)	<0.0001
Multiple gestations	27,053 (2.2 %)	65,399 (1.5 %)	<0.0001

Results

The HCUP-NIS database contained 9096,788 deliveries during the study period. We included 5448,255 deliveries, with 1486,733 (27.2 %) deliveries in the lowest quartile, 1387,004 (25.4 %) in the second quartile, 1355,529 (24.8 %) in the third quartile and 1218,989 (22.3 %) in the wealthiest income quartile.

The clinical and demographic characteristics of the higher SES cohort compared to all other income cohorts combined can be found in Table 1. Women from the wealthiest income quartile were more likely to be older, Caucasian and have private insurance. They were less likely to be smokers, have chronic hypertension, pre-gestational diabetes, and use illicit drugs.

The effects of higher SES on pregnancy and delivery outcomes are shown in Table 2, Table 3, respectively. Confounding factors which were controlled for are listed under the table. Women with greater SES were less likely to develop pregnancy complications including gestational-HTN, preeclampsia, eclampsia, and GDM. They were also less likely to develop delivery complications such as PPROM, PTB, placental abruption and require transfusion. No significant difference was observed in the risk of CD, instrumental delivery, VTE, PPH, hysterectomy, wound complications or maternal death.

Table 2.

Pregnancy outcomes ^a - Highest income quartile vs all others.

Outcomes	Highest income quartile N = 1,218,989 (%)	Other N = 4,229,266 (%)	Crude OR (95 % CI)	Adjusted OR (95 % CI)	Adjusted p-value
Pregnancy induced hypertension	82,245 (6.7 %)	341,302 (8.1 %)	0.871 (0.861–0.881)	0.867 (0.853–0.882)	<0.0001
Gestational hypertension	38,681 (3.2 %)	153,419 (3.6 %)	0.871 (0.861–0.881)	0.872 (0.852–0.893)	<0.0001
Preeclampsia	38,834 (3.2 %)	162,966 (3.9 %)	0.821 (0.812–0.830)	0.876 (0.855–0.897)	<0.0001
Eclampsia	596 (0 %)	3289 (0.1 %)	0.628 (0.576–0.686)	0.812 (0.666–0.989)	0.038
Preeclampsia or Eclampsia superimposed on pre-existing HTN	5322 (0.4 %)	26,817 (0.6 %)	0.687 (0.667–0.708)	0.862 (0.808–0.920)	<0.0001
GDM	80,944 (6.6 %)	255,883 (6.1 %)	1.104 (1.095–1.113)	0.907 (0.892–0.923)	<0.0001
Placenta previa	8826 (0.7 %)	21,749 (0.5 %)	1.411 (1.376–1.446)	1.004 (0.952–1.06)	0.874

HMO: health maintenance organization; CS: caesarean section; HTN: hypertension; DM: diabetes mellitus; IVF: in vitro fertilization; GDM: gestational diabetes mellitus; SVD: spontaneous vaginal delivery; PPROM: preterm premature rupture of membranes; PPH: post-partum hemorrhage; DVT: deep venous thrombosis; VTE: venous thromboembolism; DIC: disseminated intravascular coagulation; SGA: small for gestational age; IUFD: intra-uterine fetal demise.

^a Pregnancy Outcomes: Adjusted for Race, medical insurance Plan Type, Hospital location, Age, Obesity, Illicit Drug Use, Tobacco Smoking during pregnancy, Previous CS, Chronic HTN, Thyroid Disease, Multiple Pregnancy, Pregestational DM and IVF.

Table 3.

Delivery outcomes ^b - Highest income quartile vs all others.

Outcomes	Highest income quartile N = 1 218,989 (%)	Other N = 4 229,266 (%)	Crude OR (95 % CI)	Adjusted OR (95 % CI)	Adjusted p-value
PPROM	13,499 (1.1 %)	49,056 (1.2 %)	0.954 (0.936–0.973)	0.916 (0.877–0.955)	<0.0001
Preterm delivery	74,879 (6.1 %)	299,295 (7.1 %)	0.859 (0.852–0.866)	0.904 (0.887–0.922)	<0.0001
Abruptio placenta	11,084 (0.9 %)	16,692 (1.1 %)	0.822 (0.805–0.839)	0.892 (0.853–0.934)	<0.0001
Chorioamnionitis	23,199 (1.9 %)	77,942 (1.8 %)	1.033 (1.018–1.049)	1.079 (1.046–1.113)	<0.0001
SVD	738,005 (60.5 %)	2,647,260 (62.6 %)	0.917 (0.913–0.921)	1.002 (0.991–1.012)	0.759
Operative vaginal delivery	79,980 (6.6 %)	262,641 (6.2 %)	1.061 (1.052–1.069)	1.002 (0.983–1.021)	0.845
CS	418,409 (34.2 %)	1,376,936 (32.6 %)	1.083 (1.078–1.087)	1.0 (0.989–1.012)	0.932
Hysterectomy	1199 (0.1 %)	4046 (0.1 %)	1.026 (0.962–1.095)	0.928 (0.812–1.061)	0.273
PPH	34,873 (2.9 %)	125,448 (3.0 %)	0.963 (0.952–0.975)	1.025 (1.0–1.051)	0.055
Wound complications	4611 (0.4 %)	14,003 (0.3 %)	1.143 (1.105–1.182)	1.021 (0.944–1.104)	0.605
Maternal Death	50 (0 %)	311 (0 %)	0.558 (0.414–0.752)	0.554 (0.276–1.110)	0.096
Transfusion	10,965 (0.9 %)	49,923 (1.2 %)	0.758 (0.743–0.774)	0.948 (0.909–0.990)	0.015
Maternal infection	26,591 (2.2 %)	93,415 (2.2 %)	0.987 (0.974–1.001)	1.055 (1.025–1.086)	<0.0001
VTE	704 (0.1 %)	2462 (0.1 %)	0.992 (0.912–1.079)	1.1 (0.921–1.314)	0.293
DIC	2838 (0.2 %)	9640 (0.2 %)	1.021 (0.979–1.065)	0.934 (0.862–1.033)	0.206

^b Delivery Outcomes: Adjusted for Race, medical insurance Plan Type, Hospital location, Age, Obesity, Illicit Drug Use, Tobacco Smoking during pregnancy, Previous CS, Chronic HTN, Thyroid Disease, Multiple Pregnancy, Pregestational DM, IVF, Pregnancy Induce Hypertension, Gestational hypertension, Preeclampsia and Preeclampsia Eclampsia superimposed Hypertension.

Table 4 displays the effect of higher SES on neonatal outcomes. Women with greater SES were less likely to have an IUFD, but more likely to deliver neonates with congenital anomalies. No significant difference was observed in the risk of SGA.

Table 4.

Neonatal outcomes ^c - Highest income quartile vs all others.

Outcomes	Highest income quartile N = 1 218,989 (%)	Other N = 4 229,266 (%)	Crude OR (95 % CI)	Adjusted OR (95 % CI)	Adjusted p-value
SGA	26,923 (2.2 %)	104,087 (2.5 %)	0.895 (0.883–0.907)	0.988 (0.960–1.016)	0.383
IUFD	3837 (0.3 %)	18,748 (0.4 %)	0.709 (0.685–0.734)	0.801 (0.743–0.863)	<0.0001
Congenital Anomalies	7029 (0.6 %)	21,444 (0.5 %)	1.138 (1.108–1.169)	1.096 (1.042–1.152)	<0.0001

^c Neonatal outcomes: Adjusted for Race, medical insurance Plan Type, Hospital location, Age, Obesity, Illicit Drug Use, Tobacco Smoking during pregnancy, Previous CS, Chronic HTN, Thyroid Disease, Multiple Pregnancy, Pregestational DM, IVF, Pregnancy Induce Hypertension, Gestational hypertension, Preeclampsia and Preeclampsia Eclampsia superimposed Hypertension.

Discussion

In this retrospective cohort study, women with higher SES appear to start their pregnancy in more favourable conditions with lower rates of risky behaviour. We found that higher SES women had significantly decreased risks of pregnancy-induced HTN, gestational-HTN, preeclampsia, and GDM. This contrasts a Swedish study which found lower SES was associated with chronic-HTN, but not with gestational-HTN or preeclampsia [17]. They approximated SES by parental education and social class from registers, which may explain how this differs from our results.

Several studies have examined the association between SES and GDM. Our results mirror those of an Australian population-based study, in which SES based on residential ZIP code was inversely associated with GDM risk [18]. Fieg et al. found that women from higher-income neighbourhoods were less likely to have GDM [19]. Bouthoorn et al. found that lower maternal education was associated with increased GDM risk [20]. A Chinese study found no association between household income and GDM, but a strong association between higher education and decreased GDM risk [11]. Liu et al. showed that women from the middle-, high-income and tertiary education groups had decreased GDM risk [7]. Women with higher SES, determined by income, education, occupation or living in a wealthier neighbourhood, are likely more health-educated, and therefore are more likely to make health-conscious lifestyle choices and have access to prenatal care, education or public health programs, promoting healthy pregnancies.

The association between SES and CD remains inconsistent in published studies. Several have shown that higher SES women, estimated by income [21] or education [22] were more likely to have a CD. Contrarily, studies from Canada and Finland found that women from highest-income neighbourhoods had significantly lower age-adjusted CD rates [23], [24]. Another study found CD rates to be highest in women with private insurance or a graduate degree, but overall, SES was not significantly associated with CD risk [25]. Our findings show that higher SES does not significantly increase CD risk, though there is a trend towards higher CD rate with higher SES. In the lower SES cohort, CD rate was decreased. Although CD indication is unavailable, this may represent that lower SES women are less likely to undergo elective CD. Older age is a known risk factor for CD [25]. Those with lower SES were younger, yet the lower CD risk occurred even after controlling for age differences.

Prematurity is a leading cause of perinatal morbidity and mortality [26], however the relationship between SES and prematurity is not well studied. Studies from Taiwan [27], New Zealand [28] and Canada [15] reported no association between SES and prematurity while a North Carolina study found increased PTB risk in Caucasian mothers with low education and income [29]. A systematic review demonstrated that income and education were not associated with adverse birth outcomes, except for PTB in Black mothers [30]. Our results demonstrate an association between lower SES and PTB, even after controlling for maternal race. Two Canadian studies reported an association between neighbourhood SES, and iatrogenic and spontaneous PTB [14], [31]. Bushnik et al. found that PTB was inversely associated with maternal education, but not income, after adjustment for maternal age, ethnicity and marital status [13]. Parker et al. found that lower SES, proxied by education, parental occupation and family income, was associated with LBW, but not SGA or PTB, but lower SES was associated with both these outcomes in Black women more than Caucasian [32]. A large retrospective cohort study from Alberta found that women from highest-income neighbourhoods had significantly lower spontaneous PTB rates. Due to the nature of our data, we are unable to distinguish between spontaneous and iatrogenic PTB. However, Wood et al. also used postal codes to determine a woman’s average income [9] and obtained similar results to ours; the lower SES cohort had increased PTB risk, while the higher SES cohort had decreased PTB risk, when controlling for ethnicity and other confounding factors.

Sparse research has examined the relationship between SES and PPROM, placental abruption, chorioamnionitis, PPH, and wound complications. Raisanen et al. found lower SES to increase the risk of abruption in multiparous women only [33], while a Ghanaian study also found low SES to be a risk factor [34]. Noor et al. found that PPROM was more common in patients with low SES, and primary-middle school education [35]. Our study shows decreased risk for these outcomes in women with higher SES. Several studies have found lower SES to be associated with increased risk of PPH [36], [37], [38], a finding not mirrored in our results. However, lower SES women in our study were more likely to receive blood transfusions. The increased transfusion risk could indicate increased anemia during pregnancy or lack of pre-delivery optimization, all potentially requiring transfusions. The nature of the data used does not allow for confirmation. Higher SES women did not have an altered risk of PPH.

Several studies have shown an association between lower SES and increased VTE risk in non-pregnant populations [39], [40]. A Korean study found VTE to be associated with low SES among peripartum women [41]. We found no increased risk of VTE in either cohort.

Kern-Goldberger et al. found lower maternal education to increase the risk for a composite of maternal complications including hysterectomy, uterine atony, blood transfusion, surgical injury, and wound complications [42]. Our study found no significant difference in the risk of wound complications, DIC and hysterectomy for either cohort.

Maternal socioeconomic disadvantage has been associated with SGA infants in various studies worldwide [43], [44], [45]. Joseph et al. found that lower SES according to family income was significantly associated with SGA [15]. A Canadian study found that both higher maternal income and education were associated with decreased SGA risk, suggesting both factors together contribute [13]. We found no significant difference in SGA rates, even after adjusting for confounding factors. This can be explained by the use of ICD-9 codes not specifically collected for this study. SGA rates, usually quoted at 10 % in the general population, were much lower in both cohorts, suggesting we were unable to find a difference that exists due to the nature of our data.

Worldwide, the risk of stillbirth is highest among the least socioeconomic privileged groups [46], [47], [48], [49]. An observational study demonstrated that stillbirth risk in women with lower education was double that of women with tertiary education, with a higher risk in African mothers [50]. A European study found that higher-level education and occupation were correlated with reduced stillbirth risk [48]. This is similar to our results demonstrating that women with higher SES had decreased stillbirth risk, while women with lower SES had a higher risk of IUFD, even after adjustment for maternal age and other characteristics.

Congenital anomalies contribute greatly to long-term neonatal morbidity and mortality [12]. Studies on SES and congenital anomalies have reported either no association [51], [52], [53] or a higher prevalence among lower SES groups. Varela et al. found that respiratory congenital anomalies correlated with lower SES[12]. Vrijheid et al. found that non-chromosomal congenital anomalies increased with decreasing SES, estimated by ZIP code, with a trend towards increased risk with increasing SES [10]. A Glasgow study found a similar trend to our data for increasing congenital anomalies in higher SES groups after controlling for maternal age [54].

Our study has certain limitations. As our data is based on an administrative dataset, some variables of interest are unavailable, such as complications occurring outside the index delivery admission, early pregnancy loss and pregnancy termination data, as well as some neonatal outcomes. Furthermore, the database only allowed for median household income based on ZIP code to be used as an estimation of SES. Though this is generally reliable, we cannot guarantee that there are no discrepancies between incomes within the same ZIP code. We also recognize that household income only represents one facet of the complex entity that is SES. This database does not allow us to use other determinants of SES such as education or occupation. There is no exact method for measuring SES, with these measures being used as proxies. Therefore, the studies pertaining to SES and utilizing these different proxies may not be easily compared.

To the best of our knowledge, this is the largest population-based study to examine the effect of SES on multiple pregnancy outcomes, which allowed us to determine statistically significant differences in risk amongst different SES pregnancies when using estimated income based on ZIP code as a measure of SES. The large sample size allowed us to control for many potential confounding factors, such as race, age, smoking, and other comorbidities. It is important to acknowledge that though our results help paint a clearer picture of the effect of SES on pregnancy outcomes, it cannot fully assess how different health, population and personal factors influence SES and pregnancy outcomes and the intricacies of how these different factors interact with each other in order to impact these outcomes.

Conclusions

Published studies have demonstrated that women with low SES have worse pregnancy outcomes. This study compared women from the highest SES to all other SES groups combined. Our results show that greater SES predisposes women to better pregnancy outcomes, even when controlled for confounding factors, suggesting that access to healthcare or the means to obtain better prenatal care may significantly improve outcomes.

Furthermore, we hypothesize that women with higher SES were more likely to deliver a neonate with congenital anomalies, possibly because they are less likely to abort, given they are more likely to have the financial means necessary to care for such a child. Further studies are required to evaluate this.

Improving health disparities would benefit pregnant women and their offspring, but improving SES alone may not be sufficient to produce significant change for a majority of women. More research is required to determine which improvements in health care provision, patient education and socioeconomic disparities would be most impactful in improving outcomes for all, regardless of SES.

Ethical approval

The local Institutional Review Board deemed the study exempt from review.

Informed consent

Not applicable.

Research funding

None declared.

Author contributions

All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

Declaration of Competing Interest

Authors state no conflict of interest.