Abstract
Background.
In this paper, we test the hypothesis that SARS-CoV-2 infection and the COVID-19 pandemic period had stronger adverse implications for perinatal outcomes among marginalized racial and ethnic groups in California.
Methods.
We used California birth certificate and hospital data from 2019–2021 to estimate marginal risk differences for SARS-CoV-2 infection and the COVID-19 pandemic period in relation to perinatal outcomes for Asian, Black, Hispanic, Multiracial, and White pregnant people using targeted maximum likelihood estimation.
Results.
Among 849,401 deliveries, there were racial and ethnic disparities in the burden of SARS-CoV-2 infection and perinatal outcomes, and in the magnitudes of risk associated with SARS-CoV-2 infection and the COVID-19 pandemic. Hispanic pregnant people had the highest incidence of SARS-CoV-2 infection. Asian and Black pregnant people had the greatest marginal risk differences for multiple outcomes, particularly outcomes already disproportionately experienced by these groups.
Conclusions.
Risks from SARS-CoV-2 infection and the COVID-19 pandemic period on perinatal outcomes were disproportionately experienced by marginalized racial and ethnic groups. Differential burdens of infection and larger risks experienced with pandemic exposures were associated with worse perinatal outcomes for Asian, Black, and Hispanic pregnant people in California compared with those for White pregnant people.
Keywords: SARS-CoV-2 infection, COVID-19 pandemic, Pregnancy, Health status disparities, Premature birth, United States, Racial groups
SARS-CoV-2 infection and the adverse economic and social consequences of the COVID-19 pandemic period have disproportionately impacted marginalized racial and ethnic groups in the United States (US) and exacerbated structural inequities, such as health care access and employment opportunities, rooted in racial discrimination.1,2 Maternal and infant health worsened over the course of the COVID-19 pandemic period; for example, maternal death increased by 25% in 2020 and 2021.3 The impacts were differential by racial and ethnic groups and Black pregnant people experienced the largest increase in the rate of maternal death compared to White and Hispanic pregnant people.3 Despite indications that racial and ethnic perinatal health disparities have been exacerbated by the pandemic, prior research has not examined the contributions of SARS-CoV-2 infection and the COVID-19 pandemic period as distinct exposures in relation to perinatal health outcomes for different racial and ethnic groups. Collectively examining the differential burden of SARS-CoV-2 infection and the differential associations of SARS-CoV-2 infection and the COVID-19 pandemic period with perinatal outcomes by race and ethnicity has potential to illuminate how COVID-19 has exacerbated perinatal health disparities.
Since the beginning of the pandemic, there has been substantial concern about the effects of SARS-CoV-2 infection during pregnancy. SARS-CoV-2 infection modifies immune and inflammatory responses, and during pregnancy, these changes can affect placental function and perinatal health.4,5 SARS-CoV-2 infection has been found to be associated with a range of adverse perinatal outcomes across multiple cohorts and meta-analyses.6–9 In contrast, the COVID-19 pandemic period, which created substantial economic and social stressors10 and delayed health care,11 has not been as thoroughly studied. In our prior work, we found that uninfected infants delivered during the first year of the pandemic (2020) had decreased risk of preterm birth but increased risk of gestational diabetes and hypertensive disorders of pregnancy compared to deliveries in the prior pre-pandemic year (2019) in California.8 To our knowledge, no other studies have examined effects of the pandemic period disentangled from the effects of SARS-CoV-2 infection.
Racial and ethnic disparities in health are shaped by opportunities and restrictions within systems, structures, and social norms. As described by Parolin et al., marginalized racial and ethnic groups face “greater pre-existing health conditions, reduced access to healthcare, lower-quality neighborhood and housing conditions, and unequal exposure to high-risk occupations”,12 and when SARS-CoV-2 arrived to the US in 2020, these structural inequities drove disparities in SARS-CoV-2 infection and subsequent complications from infection.1 Moreover, these same groups disproportionately experienced the adverse economic impacts of the COVID-19 pandemic period and reported substantially higher rates of unemployment,13 with evidence that Black women experienced the greatest likelihood of job loss and food insecurity.14
It is critical to examine how the differential pandemic-related experiences intersected with pre-existing racial inequities to shape downstream perinatal health disparities. In this cohort study, we differentiate the COVID-19 pandemic period from individual SARS-CoV-2 infection. We examine both the population-level burdens of infection, and associations of infection and the pandemic period with perinatal outcomes by race and ethnicity among deliveries in California from 2019 through 2021. We hypothesize that racial and ethnic marginalized groups (i.e. Asian, Black, and Hispanic pregnant people) will have higher burdens of SARS-CoV-2 infection and stronger associations between both exposures (infection and pandemic period) and adverse perinatal outcomes compared to White pregnant people over the study period.
Methods
Study Design, Population, and Data Sources
We conducted a cohort study of pregnant people with a singleton live birth delivery in California from 2019–2021. We probabilistically linked birth certificate records from the California Department of Public Health (CDPH) to the pregnant person’s hospital records (i.e., inpatient discharge, emergency department discharge, and ambulatory surgery encounters) during pregnancy and the year following delivery from the California Department of Health Care Access and Information. During our study period, we matched over 97% of all California births to the pregnant person’s hospital discharge record (see linkage details in eAppendix A). Data from each year were limited to deliveries between April 1 and December 31 for seasonal comparability given the timing of pandemic onset and the introduction of the COVID-19 diagnostic code in 2020. This study was reviewed and approved by the California Health and Human Services Agency and University of California, Berkeley Committees for the Protection of Human Subjects.
We identified SARS-CoV-2 infection using complication codes from the birth record and diagnostic codes for SARS-CoV-2 infection from the pregnant person’s hospital discharge records between pregnancy onset and delivery. We considered deliveries in 2020 or 2021 that did not have a record of SARS-CoV-2 infection to be exposed to the COVID-19 pandemic period only. Deliveries in 2019 were considered unexposed to SARS-CoV-2 infection and the COVID-19 pandemic period.
Primary study outcomes included preterm birth; indicated preterm birth; spontaneous preterm birth; hypertensive disorders of pregnancy, which include preeclampsia, eclampsia, and hypertension during pregnancy; gestational diabetes mellitus (GDM); gestational weight gain (including the categories of inadequate, adequate, and excessive gestational weight gain); and severe maternal morbidity with transfusion. We also examined severe maternal morbidity without transfusion in a sensitivity analysis. Outcomes were coded using birth record codes and/or hospitalization diagnostic or procedure codes during pregnancy or at delivery (see definitions and diagnostic/procedure codes in eTables 1 and 2).
Race and ethnicity
We obtained self-reported race and ethnicity of the pregnant person from the birth record and classified by the CDPH into the following groups: non-Hispanic American Indian or Alaskan Native, non-Hispanic Asian (hereafter, Asian), non-Hispanic Black (hereafter, Black), non-Hispanic Hawaiian or Pacific Islander (hereafter, Hawaiian or Pacific Islander), Hispanic, non-Hispanic Multiracial, non-Hispanic White (hereafter, White), or Other. “Other” was an option offered to the pregnant person when they self-reported their race and ethnicity. In the primary analysis, we could not include American Indian or Alaskan Native, Hawaiian or Pacific Islander, and Other racial groups due to small populations in each exposure and outcome category when stratified by race and ethnicity.
Covariates
We identified individual- and community-level demographic and socioeconomic characteristics that may confound the relationship between SARS-CoV-2 infection or the COVID-19 pandemic period and our health outcomes of interest informed by prior research. Different socioeconomic and age groups have been shown to have varying levels of SARS-CoV-2 infection,15 and there may be individual- and community-level differences between pregnant people who delivered during the COVID-19 pandemic period. These characteristics are also associated with perinatal health outcomes, potentially confounding the relationship of interest.16 Individual-level confounders for the pregnant person included age, parity, education, and insurance type at delivery. Analyses for gestational weight gain outcomes included additional adjustment for gestational age at birth. Community-level confounders included racial and ethnic composition, poverty, median household income, unemployment, vehicle ownership, vacancy rates, and marital status at the ZIP Code of pregnant person’s residence at the time of delivery obtained through the American Community Survey 5-year estimates.
Statistical Analysis
Analyses estimated the independent contributions of SARS-CoV-2 infection and the COVID-19 pandemic period to perinatal outcomes by racial and ethnic group. We assessed associations for SARS-CoV-2 infection independent of the pandemic period by comparing deliveries in 2020–2021 with and without SARS-CoV-2 infection. We isolated associations for the COVID-19 pandemic period independent of SARS-CoV-2 infection by comparing deliveries in 2020–2021 without SARS-CoV-2 infection to pre-pandemic deliveries. We estimated marginal risk differences using targeted maximum likelihood estimation (TMLE). TMLE is a semiparametric doubly-robust substitution estimator that incorporates data-adaptive machine learning in model fitting.17 If identification assumptions are met, the marginal risk difference we examined is the average treatment effect on the treated (ATT). The ATT describes how the risk of the outcome would change if exposed deliveries (to SARS-CoV-2 infection, or to the COVID-19 pandemic) were unexposed (no SARS-CoV-2 infection, delivered pre-pandemic). Assessment of identification assumptions is available in eAppendix B. We used the likelihood ratio test to compare linear regression models with and without interaction terms between the exposure and racial and ethnic groups to assess the presence of additive interaction by race and ethnicity. P-values for the likelihood ratio test were considered statistically significant at a 0.05 value. All analyses were completed using R version 4.2.1.
Results
Characteristics of pregnant people who delivered during the COVID-19 pandemic period (2020–2021) with (n = 14,281) and without (n = 537,559) SARS-COV-2 infection and in 2019 (n = 297,561) are presented overall and by race and ethnicity in Table 1 (see eTable 3 for characteristics of smaller racial and ethnic groups). Among 849,401 deliveries, there were substantial racial and ethnic disparities in perinatal outcomes. Black pregnant people had the highest prevalence of preterm birth (12.5%), hypertensive disorders of pregnancy (19.8%), and severe maternal morbidity with transfusion (3.8%), and Asian pregnant people had the highest prevalence of gestational diabetes (18.8%). Table 2 presents the burden of SARS-CoV-2 infection among deliveries during the COVID-19 pandemic period. Overall, the highest proportion with SARS-CoV-2 infection was found among Hispanic (3.4%) and Native Hawaiian and Pacific Islander (3.5%) pregnant people. Asian pregnant people had the lowest burden of infection (1.2%).
Table 1.
Individual- and community-level characteristics stratified by race and ethnicity for singleton live birth deliveries: California, April to December 2019–2021
| Race and Ethnicity | ||||||
|---|---|---|---|---|---|---|
| Overall | NH Asian | NH Black | Hispanic | NH Multiracial | NH White | |
| N (%) | 849,401 | 125,387 (14.8) | 43,416 (5.1) | 420,329 (49.5) | 21,794 (2.6) | 238,475 (28.1) |
| Exposure | ||||||
| Pre-COVID-19 Pandemic, n (%) Delivery in April to Dec. 2019 |
297,561 (35.0) | 47385 (37.8) | 14996 (34.5) | 144844 (34.5) | 7401 (34.0) | 82935 (34.8) |
| SARS-CoV-2 Infection, n (%) Delivery with infection in April to Dec. 2020–2021 |
14,281 (1.7) | 946 (0.8) | 688 (1.6) | 9482 (2.3) | 315 (1.4) | 2850 (1.2) |
| COVID-19 Pandemic, n (%) Delivery without infection in April to Dec. 2020–2021 |
537,559 (63.3) | 77056 (61.5) | 27732 (63.9) | 266003 (63.3) | 14078 (64.6) | 152690 (64.0) |
| Individual characteristics | ||||||
| Age in years, mean (SD) | 30.3 (5.8) | 33.0 (4.6) | 29.4 (6.0) | 28.8 (5.9) | 30.1 (5.9) | 31.6 (5.3) |
| Insurance type, n (%) | ||||||
| Private coverage | 445502 (52.4) | 92667 (73.9) | 15982 (36.8) | 151790 (36.1) | 13397 (61.5) | 171666 (72.0) |
| Public coverage | 383576 (45.2) | 23843 (19.0) | 26950 (62.1) | 262068 (62.3) | 8153 (37.4) | 62562 (26.2) |
| Other | 3712 (0.4) | 1220 (1.0) | 79 (0.2) | 990 (0.2) | 80 (0.4) | 1343 (0.6) |
| Self-pay | 16611 (2.0) | 7657 (6.1) | 405 (0.9) | 5481 (1.3) | 164 (0.8) | 2904 (1.2) |
| Parity, n (%) | ||||||
| Zero | 336148 (39.6) | 59350 (47.3) | 16073 (37.0) | 147386 (35.1) | 9761 (44.8) | 103578 (43.4) |
| One | 271529 (32.0) | 46320 (36.9) | 12350 (28.4) | 124837 (29.7) | 6713 (30.8) | 81309 (34.1) |
| Two or more | 241724 (28.5) | 19717 (15.7) | 14993 (34.5) | 148106 (35.2) | 5320 (24.4) | 53588 (22.5) |
| Education, n (%) | ||||||
| College graduation or more | 366348 (43.1) | 97527 (77.8) | 13398 (30.9) | 94967 (22.6) | 10474 (48.1) | 149982 (62.9) |
| Less than college | 386459 (45.5) | 25576 (20.4) | 26347 (60.7) | 243170 (57.9) | 10350 (47.5) | 81016 (34.0) |
| Less than high school | 96594 (11.4) | 2284 (1.8) | 3671 (8.5) | 82192 (19.6) | 970 (4.5) | 7477 (3.1) |
| Individual outcomes | ||||||
| Preterm birth, n (%) | 74444 (8.8) | 10585 (8.4) | 5432 (12.5) | 38110 (9.1) | 2072 (9.5) | 18245 (7.7) |
| Spontaneous preterm birth, n (%) | 53403 (6.3) | 7599 (6.1) | 3832 (8.8) | 27452 (6.5) | 1537 (7.1) | 12983 (5.4) |
| Indicated preterm birth, n (%) | 21034 (2.5) | 2984 (2.4) | 1600 (3.7) | 10654 (2.5) | 535 (2.5) | 5261 (2.2) |
| Hypertensive disorders of pregnancy, n (%) | 107838 (12.7) | 12606 (10.1) | 8604 (19.8) | 51424 (12.2) | 3474 (15.9) | 31730 (13.3) |
| Gestational diabetes, n (%) | 101358 (11.9) | 23559 (18.8) | 3930 (9.1) | 49866 (11.9) | 2242 (10.3) | 21761 (9.1) |
| Gestational weight gain, n (%) | ||||||
| Adequate | 280531 (33.8) | 49344 (39.9) | 11835 (28.2) | 133471 (32.6) | 6321 (29.6) | 79560 (34.1) |
| Excessive | 358866 (43.3) | 37640 (30.5) | 20978 (49.9) | 177906 (43.5) | 11258 (52.7) | 111084 (47.6) |
| Inadequate | 189861 (22.9) | 36542 (29.6) | 9198 (21.9) | 97653 (23.9) | 3786 (17.7) | 42682 (18.3) |
| Severe maternal morbidity with transfusion, n (%) | 17524 (2.1) | 2559 (2.0) | 1639 (3.8) | 8441 (2.0) | 552 (2.5) | 4333 (1.8) |
| Severe maternal morbidity without transfusion, n (%) | 8887 (1.0) | 1250 (1.0) | 855 (2.0) | 4138 (1.0) | 291 (1.3) | 2239 (0.9) |
| Community* characteristics, mean (SD) | ||||||
| Percent of population that identifies as Hispanic | 43.0 (24.4) | 29.3 (18.7) | 44.4 (20.2) | 55.2 (22.9) | 33.1 (19.1) | 29.3 (18.8) |
| Percent of population that identifies as Black | 5.9 (7.3) | 4.9 (5.4) | 14.2 (13.1) | 6.2 (7.4) | 6.8 (7.4) | 4.3 (5.1) |
| Percent of population that identifies as Asian | 13.4 (14.0) | 28.3 (18.6) | 11.5 (11.1) | 10.0 (10.9) | 13.4 (12.3) | 11.7 (11.2) |
| Percent of households below poverty line | 14.3 (7.9) | 10.6 (6.4) | 17.3 (8.2) | 16.6 (7.9) | 13.0 (7.3) | 11.8 (7.0) |
| Percent of households on public assistance | 3.8 (2.8) | 2.8 (2.2) | 5.0 (3.0) | 4.5 (2.9) | 3.5 (2.5) | 3.0 (2.4) |
| Percent of households with zero vehicles | 6.8 (5.6) | 6.3 (6.3) | 8.8 (6.7) | 7.0 (5.1) | 6.6 (5.9) | 6.1 (5.6) |
| Owner vacancy rate | 1.1 (1.2) | 1.0 (1.0) | 1.3 (1.4) | 1.7 (1.2) | 1.2 (1.3) | 1.2 (1.3) |
| Renter vacancy rate | 3.4 (2.3) | 3.4 (2.2) | 3.5 (2.2) | 3.2 (2.1) | 3.6 (2.4) | 3.6 (2.7) |
| Percent of population unemployed | 6.6 (2.9) | 5.3 (2.1) | 7.4 (2.7) | 7.3 (3.0) | 6.3 (2.8) | 5.8 (2.7) |
| Percent of population married | 37.3 (8.1) | 34.9 (7.8) | 40.9 (7.8) | 39.5 (7.2) | 35.8 (8.1) | 34.3 (8.4) |
| Percent of population separated | 2.1 (0.9) | 1.7 (0.8) | 2.5 (0.9) | 2.4 (0.9) | 2.0 (0.9) | 1.8 (0.8) |
| Median household income, $USD | 76201.33 (30439.32) | 95685.21 (34490.23) | 65628.24 (24557.75) | 65727.29 (22808.26) | 80293.69 (30405.15) | 85969.11 (32454.45) |
Note: NH = non-Hispanic. Community characteristics are measured at the ZIP Code Tabulation Area (ZCTA)level.
Table 2.
Burden of SARS-CoV-2 infection among singleton live birth deliveries: California, April to December 2020–2021
| Overall | NH AI/AN | NH Asian | NH Black | NH H/PI | Hispanic | NH Multiracial | NH Other | NH White | ||
|---|---|---|---|---|---|---|---|---|---|---|
| April to December 2020 | N | 278021 | 917 | 39397 | 14269 | 1077 | 136899 | 7105 | 281 | 78076 |
| SARS-CoV-2 infection, n (%) | 7150 (2.6) | 17 (1.9) | 377 (1.0) | 318 (2.2) | 35 (3.2) | 5266 (3.8) | 108 (1.5) | <15 | 1021 (1.3) | |
| April to December 2021 | N | 278315 | 845 | 38605 | 14151 | 1102 | 138586 | 7288 | 274 | 77464 |
| SARS-CoV-2 infection, n (%) | 7267 (2.6) | 24 (2.8) | 569 (1.5) | 370 (2.6) | 42 (3.8) | 4216 (3.0) | 207 (2.8) | <15 | 1829 (2.4) | |
|
April to December
2020–2021 |
N | 556336 | 1762 | 78002 | 28420 | 2179 | 275485 | 14393 | 555 | 155540 |
| SARS-CoV-2 infection, n (%) | 14417 (2.6) | 41 (2.3) | 946 (1.2) | 688 (2.4) | 77 (3.5) | 9482 (3.4) | 315 (2.2) | 18 (3.2) | 2850 (1.8) |
Note: NH = non-Hispanic; AI/AN = American Indian/Alaskan Native; H/PI = Hawaiian/Pacific Islander. Cells with values less than 15 are not shown.
Figure 1 and eTable 4 present the associations between SARS-CoV-2 infection delivery and perinatal outcomes. Estimates suggested an increased risk of preterm birth and subtypes, hypertensive disorders of pregnancy, gestational diabetes, and severe maternal morbidity among all racial and ethnic groups. There were statistically significant differences (P-for-interaction < 0.05) in the associations between groups for preterm birth, indicated preterm birth, and severe maternal morbidity. The marginal risk difference associated with SARS-CoV-2 infection at delivery was greatest among Asian pregnant people for preterm birth (49.7 [95% confidence interval (CI) 24.8 to 74.5] per 1000 deliveries) and indicated preterm birth (23.5 [95% CI 8.1 to 38.8] per 1000 deliveries). Black pregnant people had the greatest marginal risk difference for severe maternal morbidity with transfusion (54.6 [95% CI 32.0 to 77.2] per 1000 deliveries).
Figure 1. Associations between SARS-CoV-2 infection and perinatal outcomes among singleton live birth deliveries: California, April to December 2019–2021.
Note: NH = non-Hispanic. Marginal risk differences indicate number of excess cases of the outcome per 1000 singleton livebirth deliveries. Marginal risk differences were adjusted for pregnant person’s age, parity, education, and insurance type at delivery and community-level characteristics included racial and ethnic composition, poverty, median household income, unemployment, vehicle ownership, vacancy rates, marital status at the ZIP Code of the pregnant person’s residence at the time of delivery. We included additional adjustment for gestational age at delivery for excessive and inadequate gestational weight gain outcome models.
The associations between the COVID-19 pandemic period and perinatal outcomes are reported in Figure 2 and eTable 5. Risk of preterm birth and indicated preterm birth decreased among deliveries during the COVID-19 pandemic for all racial and ethnic groups except for Asian pregnant people. There was no clear direction (i.e. harmful or protective) for risk of spontaneous preterm birth. All racial and ethnic groups had elevated risk of hypertensive disorders and gestational diabetes. Additionally, the COVID-19 pandemic was associated with increased risk of severe maternal morbidity with transfusion for all racial and ethnic groups except for Multiracial pregnant people (Figure 2 and eTable 4). There were statistically significant differences in the associations between racial groups for spontaneous preterm birth, gestational diabetes, hypertensive disorders, and severe maternal morbidity with transfusion (P-for-interaction < 0.05). The marginal risk difference associated with the COVID-19 pandemic period was greatest among Asian pregnant people for gestational diabetes (25.5 [95% CI 20.7 to 29.5] per 1000 deliveries) and severe maternal morbidity (3.6 [95% CI 2.0 to 5.3] per 1000 deliveries). Black pregnant people had the greatest marginal risk difference for hypertensive disorders of pregnancy (24.7 [95% CI 17.0, 32.5] per 1000 deliveries).
Figure 2. Associations between the COVID-19 pandemic period and perinatal outcomes among singleton live birth deliveries: California, April to December 2019–2021.
Note: NH = non-Hispanic. Marginal risk differences indicate number of excess cases of the outcome per 1000 singleton livebirth deliveries. Marginal risk differences were adjusted for pregnant person’s age, parity, education, and insurance type at delivery and community-level characteristics included racial and ethnic composition, poverty, median household income, unemployment, vehicle ownership, vacancy rates, marital status at the ZIP Code of the pregnant person’s residence at the time of delivery. We included additional adjustment for gestational age at delivery for excessive and inadequate gestational weight gain outcome models.
The sensitivity analysis examining severe maternal morbidity without transfusion found similar associations with SARS-CoV-2 infection compared to severe maternal morbidity with transfusion. The associations with the COVID-19 pandemic period were less consistent. There was no evidence of differences in the associations between racial groups (P-for-interaction = 0.053).
Discussion
In our cohort study of pregnant people with a singleton live birth delivery in California from 2019–2021, there was a differential burden of SARS-CoV-2 infection and risk of perinatal outcomes associated with pandemic-exposures by race and ethnicity. Our results quantify disparities in how SARS-CoV-2 infection and the COVID-19 pandemic impacted perinatal outcomes across different racial and ethnic groups.
Asian pregnant people had the greatest risk difference associated with SARS-CoV-2 infection for preterm birth and indicated preterm birth and the greatest risk difference associated with the COVID-19 pandemic period for gestational diabetes mellitus and severe maternal morbidity. During the COVID-19 pandemic period, anti-Asian discrimination and violence surged in the US, and nearly one in five Asian and Pacific Islander individuals reported experiencing a hate incident between 2020 and 2021.18 Asian individuals reported increased interpersonal (i.e., verbal racial slurs or jokes and fears of physical attack/harm due to their race) and economic (i.e., decreased business patronage) discrimination. Concurrently, one large study examining over 50 million patients in Epic’s health record system found that Asian patients were more likely to test positive for SARS-CoV-2 in an inpatient setting and require a higher level of care (i.e., oxygen and ventilation) at the time of testing positive compared to other racial and ethnic groups.19 There are several, non-mutually exclusive, mechanisms that could explain these patterns of increased SARS-CoV-2 severity among Asian individuals. The fear of becoming a victim of a hate incident, particularly when multiple high-profile hate crimes were committed against Asian women, may have delayed SARS-CoV-2 testing and healthcare utilization until symptoms were of more severe, compared with other racial and ethnic groups.18 Asian pregnant people might have more severe disease due to lower levels of vaccination connected with similar concerns of facing anti-Asian sentiment at vaccination sites.20 Moreover, experience of racial discrimination has been associated with higher levels of inflammatory stress reactivity among pregnant people, and this stress response might be further exacerbated during a SARS-CoV-2 infection.21 Severe SARS-CoV-2 infection among Asian individuals during pregnancy could trigger processes associated with preterm birth, such as inflammation or exacerbation of underlying medical conditions. Our findings for SARS-CoV-2 infection are supported by a recent study that also found SARS-CoV-2 infection during pregnancy associated with a larger increased risk of preeclampsia/eclampsia and preterm birth among Asian and Pacific Islander pregnant people when compared to non-Hispanic White pregnant people.22 Regarding the pandemic period, overall increased stress and stress-related coping behaviors for those without SARS-CoV-2 infection could increase the risk of gestational diabetes and severe maternal morbidity. To our knowledge, no prior studies have examined the effects of the COVID-19 pandemic period on perinatal outcomes among Asian pregnant people.
While we observed some of the strongest associations among Asian pregnant people, this group had the lowest burden of SARS-CoV-2 infection and experienced fewer perinatal complications. Thus, the health disparities in most perinatal outcomes among Asian pregnant people are likely due to stronger links between pandemic-related exposures and health, rather than disproportionate infection or existing risk of outcomes. Gestational diabetes mellitus is a notable exception because Asian pregnant people have the highest gestational diabetes prevalence and greatest risk associated with the pandemic period. To improve perinatal health among this population and prevent the widening of disparities, we need to understand why the effects of SARS-CoV-2 infection are more severe for Asian pregnant people and to identify the mechanisms of the pandemic period that elevate risk of gestational diabetes.
We observed the greatest risk difference associated with SARS-CoV-2 infection for severe maternal morbidity, and the greatest risk difference associated with the pandemic period for hypertensive disorders of pregnancy among Black pregnant people. In the US, Black individuals experience poorer overall health due to past and present-day structural and interpersonal racism compared to all other racial and ethnic groups.23 In particular, Black pregnant people bear a disproportionate burden of adverse perinatal health outcomes, driving extreme disparities in maternal and infant health. During the COVID-19 pandemic, structural inequities were amplified through job loss, financial strain, and morbidity and mortality related to SARS-CoV-2 infection.1 Black pregnant people were more likely to report negative impacts related to job security, financial concerns, and worries about pre- and post-natal care compared to White and Hispanic pregnant people.24 Concurrently, the US was reckoning with its long history of racial police violence after the widely publicized murders of George Floyd, Ahmaud Arbery, and Breonna Taylor by current or former law enforcement officers in 2020. During this time, Black individuals reported increased depressive and anxiety symptoms and significant levels of stress related to both the COVID-19 pandemic and racial violence and police brutality against Black individuals.25,26 Stress, in tandem with poor health, could influence the association of SARS-CoV-2 infection with severe maternal morbidity. Pre-existing conditions, such as obesity, diabetes, and hypertension, are disproportionately experienced by Black individuals and significant risk factors of severe maternal morbidity. While these comorbidities do not fully explain racial and ethnic differences in severe maternal morbidity, they contribute substantially to these disparities.27 Moreover, individuals with comorbidities have been found to have more severe symptoms of SARS-CoV-2 infection.28 We hypothesize that the strong association of SARS-CoV-2 infection with severe maternal morbidity among Black pregnant people may be driven, in part, by the exacerbation of increased stress and existing comorbidities by SARS-CoV-2 infection. We also found a strong association of the pandemic period with hypertensive disorders of pregnancy among Black pregnant people, and stress has been linked to hypertension among adults through physiological changes in the body and maladaptive coping behaviors.29 Therefore, the increased stress, due to economic and other stressors, disproportionately impacting Black individuals, together with comorbidities that predispose individuals to high blood pressure, may put Black pregnant people during the pandemic period at higher risk of hypertensive disorders. When compared to other racial and ethnic groups, Black pregnant people faced a moderate risk of contracting SARS-CoV-2 but experienced the highest incidence of severe maternal morbidity and hypertensive disorders.
Our findings collectively are consistent with the hypothesis that stronger associations between COVID-19 pandemic exposures and perinatal health, especially for outcomes where burden is already the greatest, drive COVID-19 related health disparities. The high prevalence of perinatal complications within the Black community indicates that Black pregnant people will continue to face a heightened risk of various perinatal outcomes, irrespective of variations in the strength of association with SARS-CoV-2 infection and the pandemic.
Hispanic pregnant people experienced the greatest burden of SARS-CoV-2 infection. This is consistent with other studies that have found a high burden of exposure risk and infection in this group; for example, another California study found that Latinos were 8 times more likely to live in a SARS-CoV-2 high-risk exposure household, defined as a household with at least 1 essential worker and overcrowding, compared to White people.30 In California, many Hispanic and Latino individuals are employed in essential worker roles, primarily in farming and retail sectors. These occupations often lacked adequate safeguards such as personal protective equipment or sick leave, increasing employee vulnerability to SARS-CoV-2 exposure without means to quarantine if infected.30 The magnitudes of associations of SARS-CoV-2 infection and the pandemic period with perinatal outcomes were not notably elevated among Hispanic pregnant people (often similar to the associations among White pregnant persons), however the high incidence of SARS-CoV-2 infection among Hispanic pregnancies indicates substantial health disparities effects through the differential burden of infection. Further, in California, Hispanic pregnancies constitute nearly half of annual births and drive a considerable proportion of the burden of perinatal health outcomes in the total population.
Limitations
We were unable to examine racial subgroups within larger racial and ethnic categories. For example, even though we were able to differentiate between Asian pregnant people and Pacific Islander and Native Hawaiian pregnant people, there is still considerable heterogeneity among Southeast and East Asian populations in the burden of SARS-CoV-2 infection,31 perinatal outcomes,32 and experiences of discrimination 33 that would have been valuable to investigate. Additionally, we did not have an adequate sample size to investigate the association of SARS-CoV-2 infection with perinatal outcomes for smaller racial and ethnic groups (i.e., American Indian or Alaskan Native, Hawaiian or Pacific Islander, and Other racial groups). We were, however, able to report descriptive frequencies for these groups.
We limited our analysis to individual perinatal outcomes and covariates that could be validly assessed using hospital and birth records. While both records are highly complete, we may have misclassification and unmeasured confounding. We were only able to assess SARS-CoV-2 infection through hospital and birth records, and most of these infections (76.5%) were identified solely from the birth record. SARS-CoV-2 infection in the birth record was assessed at delivery by the birth attendant under the section of “Complications and Procedures of Pregnancy and Concurrent Illnesses,” and could be an active infection at delivery or an infection that occurred earlier during pregnancy. Therefore, we are unable to discern whether SARS-CoV-2 infection preceded the outcome. Additionally, although testing at delivery was near universal and well captured in both hospital and birth records,34 there may be pregnancies with SARS-CoV-2 infection but prior to delivery that may be misclassified as unexposed. Perinatal outcomes may also be differentially misclassified with respect to birth year due to limited health care and health care access at the beginning of the COVID-19 pandemic. There may be unmeasured confounding and effect modifiers given the complexity of relationships between pandemic exposures and perinatal outcomes, as well as different containment, economic, and public health policies implemented among California counties. Moreover, deliveries in 2019 may not be appropriate controls for uninfected pandemic deliveries due to the exclusion of deliveries of SARs-CoV-2 infection. These deliveries may be different in ways that are not adequately captured in available covariates (e.g. prior health history); however, we note that less than 3% of total infants delivered were infected with SARS-CoV-2.
Additional limitations include potential selection bias through multiple mechanisms. First, pregnancy intent may have shifted with economic and social changes during the pandemic. Pregnancy intent may have increased or decreased depending on socioeconomic status and other factors (e.g., job stability, remote work). Second, we only examined live birth deliveries and did not capture fetal loss; unfortunately, complete fetal loss data are unavailable, thus we are not able to investigate here this aspect of COVID-19 pandemic effects. There may also be selection bias due to unknown vaccination status. SARS-CoV-2 vaccines became broadly available in Spring of 2021, and may affect both risk of infection, and severity among those infected, yet we do not have information on vaccination status. However, the burdens of infection were comparable in 2020 and 2021 (Table 2) and vaccination levels may have been low during this time, as many pregnant people reported vaccine hesitancy due to the exclusion of pregnant people in the initial vaccine trials.35 Finally, we omitted 3.9% of records omitted due to covariate missingness.
In this cohort study, there was substantial heterogeneity in the burden of SARS-CoV-2 infection and perinatal outcomes, and associated risk of outcomes with SARS-CoV-2 infection and the COVID-19 pandemic period among marginalized racial and ethnic groups. For Asian and Black pregnant people, COVID-19 related disparities in perinatal health were driven by increased risk of outcomes associated with pandemic-exposures; particularly for outcomes where these groups were already at the highest risk. For Hispanic pregnant people, disparities were driven by the greatest burden of SARS-CoV-2 infection. While our primary focus was health disparities, both SARS-CoV-2 infection and the pandemic period were associated with elevated risk of many perinatal outcomes for all racial and ethnic groups, and no populations were unaffected. This study highlights the value of investigating differences in burden of infection and outcome, and associated risk by race and ethnicity to develop a comprehensive understanding of the impact of COVID-19 on perinatal health.
Supplementary Material
Acknowledgments:
The authors gratefully acknowledge Rebecca Baer, MPH, University of California, San Francisco and University of California, San Diego, for sharing her approach to creating the linkage between the birth and hospital discharge records. She was not compensated.
Sources of Financial Support:
This study was supported by the National Institutes of Health (R01HD098138, Dr. Ahern, principal investigator). The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Conflicts of Interest: The authors declare no potential conflicts of interest with respect to the research, authorship, or publication of this article.
Data Sharing:
The linked dataset used is this analysis is not publicly available; however, the birth and hospital discharge data used to create this dataset are available upon request for research projects from the California Department of Public Health, Center for Health Statistics and Informatics, and from the California Department of Health Care Access and Information. Computing code for statistical analysis is presented in the supplemental material.
Works Cited
- 1.Lopez L III, Hart LH III, Katz MH. Racial and Ethnic Health Disparities Related to COVID-19. JAMA. 2021;325(8):719–720. doi: 10.1001/jama.2020.26443 [DOI] [PubMed] [Google Scholar]
- 2.Inequities exposed: How COVID-19 widened racial inequities in education, health, and the workforce. Published online June 22, 2020. https://www.epi.org/publication/covid-19-inequities-wilson-testimony/
- 3.Maternal Health Outcomes Worsened and Disparities Persisted during the Pandemic. United States Government Accountability Office; 2022. https://www.gao.gov/products/gao-23-105871 [Google Scholar]
- 4.Girardelli S, Mullins E, Lees CC. COVID-19 and pregnancy: Lessons from 2020. Early Hum Dev. 2021;162:105460. doi: 10.1016/j.earlhumdev.2021.105460 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cavalcante MB, Cavalcante CT de MB, Sarno M, Barini R, Kwak-Kim J. Maternal immune responses and obstetrical outcomes of pregnant women with COVID-19 and possible health risks of offspring. J Reprod Immunol. 2021;143:103250. doi: 10.1016/j.jri.2020.103250 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Metz TD, Clifton RG, Hughes BL, et al. Association Between Giving Birth During the Early Coronavirus Disease 2019 (COVID-19) Pandemic and Serious Maternal Morbidity. Obstet Gynecol. 2023;141(1):109–118. doi: 10.1097/AOG.0000000000004982 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ferrara A, Hedderson MM, Zhu Y, et al. Perinatal Complications in Individuals in California With or Without SARS-CoV-2 Infection During Pregnancy. JAMA Intern Med. 2022;182(5):503–512. doi: 10.1001/jamainternmed.2022.0330 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jung S, Liu EF, Goin DE, et al. The COVID-19 Pandemic Period, SARS-CoV-2 Infection, and Perinatal Health. JAMA Netw Open. 2024;7(5):e2410696. doi: 10.1001/jamanetworkopen.2024.10696 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Marchand G, Patil AS, Masoud AT, et al. Systematic review and meta-analysis of COVID-19 maternal and neonatal clinical features and pregnancy outcomes up to June 3, 2021. AJOG Glob Rep. 2022;2(1):100049. doi: 10.1016/j.xagr.2021.100049 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bonotti M, Zech ST. The Human, Economic, Social, and Political Costs of COVID-19. In: Bonotti M, Zech ST, eds. Recovering Civility during COVID-19. Springer; Singapore; 2021:1–36. doi: 10.1007/978-981-33-6706-7_1 [DOI] [Google Scholar]
- 11.Czeisler MÉ. Delay or Avoidance of Medical Care Because of COVID-19–Related Concerns — United States, June 2020. MMWR Morb Mortal Wkly Rep. 2020;69. doi: 10.15585/mmwr.mm6936a4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Parolin Z, Lee EK. The Role of Poverty and Racial Discrimination in Exacerbating the Health Consequences of COVID-19. Lancet Reg Health Am. 2022;7:100178. doi: 10.1016/j.lana.2021.100178 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bowman Williams J COVID-19 Widens Disparities for Workers of Color. ABA J Labor Employ. 2020;35(1). [Google Scholar]
- 14.Coats JV. Employment Loss and Food Insecurity — Race and Sex Disparities in the Context of COVID-19. Prev Chronic Dis. 2022;19. doi: 10.5888/pcd19.220024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rozenfeld Y, Beam J, Maier H, et al. A model of disparities: risk factors associated with COVID-19 infection. Int J Equity Health. 2020;19(1):126. doi: 10.1186/s12939-020-01242-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wang E, Glazer KB, Howell EA, Janevic TM. Social Determinants of Pregnancy-Related Mortality and Morbidity in the United States: A Systematic Review. Obstet Gynecol. 2020;135(4):896–915. doi: 10.1097/AOG.0000000000003762 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.van der Laan MJ, Rose S. Targeted Learning: Causal Inference for Observational and Experimental Data. Springer Science & Business Media; 2011. [Google Scholar]
- 18.Jeung R, Horse AY, Popovic T, Lim R. Stop AAPI Hate National Report. Ethn Stud Rev. 2021;44(2):19–26. doi: 10.1525/esr.2021.44.2.19 [DOI] [Google Scholar]
- 19.Rubin-Miller L, Alban C, Artiga S, Published SS. COVID-19 Racial Disparities in Testing, Infection, Hospitalization, and Death: Analysis of Epic Patient Data - Issue Brief - 9530. KFF. Published September 16, 2020. Accessed June 27, 2024. https://www.kff.org/report-section/covid-19-racial-disparities-in-testing-infection-hospitalization-and-death-analysis-of-epic-patient-data-issue-brief/ [Google Scholar]
- 20.Elam-Evans LD. The Association of Reported Experiences of Racial and Ethnic Discrimination in Health Care with COVID-19 Vaccination Status and Intent — United States, April 22, 2021–November 26, 2022. MMWR Morb Mortal Wkly Rep. 2023;72. doi: 10.15585/mmwr.mm7216a5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dawson D, Goodman SH, Granger DA, Laurent H. Associations Between Direct and Indirect Forms of Racism Exposure and Stress-Induced Inflammatory Response and Health in Pregnancy. J Racial Ethn Health Disparities. Published online November 7, 2022:1–12. doi: 10.1007/s40615-022-01442-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mensah NA, Fassett MJ, Lurvey LD, et al. Association between SARS-CoV-2 Infection and Adverse Perinatal Outcomes by Race/Ethnicity in a Large Integrated Health Care System. Am J Perinatol. Published online April 25, 2024. doi: 10.1055/a-2298-4826 [DOI] [PubMed] [Google Scholar]
- 23.Artiga S, Hill L, Published MP. How Present-Day Health Disparities for Black People Are Linked to Past Policies and Events. KFF. Published February 22, 2024. Accessed June 27, 2024. https://www.kff.org/racial-equity-and-health-policy/issue-brief/how-present-day-health-disparities-for-black-people-are-linked-to-past-policies-and-events/ [Google Scholar]
- 24.Gur RE, White LK, Waller R, et al. The Disproportionate Burden of the COVID-19 Pandemic Among Pregnant Black Women. Psychiatry Res. 2020;293:113475. doi: 10.1016/j.psychres.2020.113475 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Eichstaedt JC, Sherman GT, Giorgi S, et al. The emotional and mental health impact of the murder of George Floyd on the US population. Proc Natl Acad Sci U S A. 2021;118(39):e2109139118. doi: 10.1073/pnas.2109139118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Cox JM, Toussaint A, Woerner J, Smith A, Haeny AM. Coping While Black: Comparing Coping Strategies Across COVID-19 and the Killing of Black People. J Racial Ethn Health Disparities. Published online April 26, 2023:1–12. doi: 10.1007/s40615-023-01600-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Leonard SA, Main EK, Scott KA, Profit J, Carmichael SL. Racial and Ethnic Disparities in Severe Maternal Morbidity Prevalence and Trends. Ann Epidemiol. 2019;33:30–36. doi: 10.1016/j.annepidem.2019.02.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Patel R, Kooner JS, Zhang W. Comorbidities associated with the severity of COVID-19, and differences across ethnic groups: a UK Biobank cohort study. BMC Public Health. 2023;23(1):1566. doi: 10.1186/s12889-023-16499-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Spruill TM. Chronic Psychosocial Stress and Hypertension. Curr Hypertens Rep. 2010;12(1):10–16. doi: 10.1007/s11906-009-0084-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Reitsma MB, Claypool AL, Vargo J, et al. Racial/Ethnic Disparities In COVID-19 Exposure Risk, Testing, And Cases At The Subcounty Level In California. Health Aff (Millwood). 2021;40(6):870–878. doi: 10.1377/hlthaff.2021.00098 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Kalyanaraman Marcello R, Dolle J, Tariq A, et al. Disaggregating Asian Race Reveals COVID-19 Disparities Among Asian American Patients at New York City’s Public Hospital System. Public Health Rep. 2021;137(2):317–325. doi: 10.1177/00333549211061313 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wong LF, Caughey AB, Nakagawa S, Kaimal AJ, Tran SH, Cheng YW. Perinatal outcomes among different Asian-American subgroups. Am J Obstet Gynecol. 2008;199(4):382.e1–6. doi: 10.1016/j.ajog.2008.06.073 [DOI] [PubMed] [Google Scholar]
- 33.Okazaki S, Lee CS, Prasai A, Chang DF, Yoo N. Disaggregating the data: Diversity of COVID-19 stressors, discrimination, and mental health among Asian American communities. Front Public Health. 2022;10:956076. doi: 10.3389/fpubh.2022.956076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Torche F, Nobles J. Vaccination, immunity, and the changing impact of COVID-19 on infant health. Proc Natl Acad Sci. 2023;120(49):e2311573120. doi: 10.1073/pnas.2311573120 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Simmons LA, Whipps MDM, Phipps JE, Satish NS, Swamy GK. Understanding COVID-19 vaccine uptake during pregnancy: ‘Hesitance’, knowledge, and evidence-based decision-making. Vaccine. 2022;40(19):2755–2760. doi: 10.1016/j.vaccine.2022.03.044 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The linked dataset used is this analysis is not publicly available; however, the birth and hospital discharge data used to create this dataset are available upon request for research projects from the California Department of Public Health, Center for Health Statistics and Informatics, and from the California Department of Health Care Access and Information. Computing code for statistical analysis is presented in the supplemental material.