key: cord-0694280-ia8rou81
authors: LOKKEN, Erica M.; TAYLOR, G. Gray; HUEBNER, Emily M.; VANDERHOEVEN, Jeroen; HENDRICKSON, Sarah; COLER, Brahm; SHENG, Jessica S.; WALKER, Christie L.; MCCARTNEY, Stephen A.; KRETZER, Nicole M.; RESNICK, Rebecca; KACHIKIS, Alisa; BARNHART, Nena; SCHULTE, Vera; BERGAM, Brittany; K, Kimberly; ALBRIGHT, Catherine; LARIOS, Valerie; KELLEY, Lori; LARIOS, Victoria; EMHOFF, Sharilyn; RAH, Jasmine; RETZLAFF, Kristin; THOMAS, Chad; PAEK, Bettina W.; HSU, Rita J.; ERICKSON, Anne; CHANG, Andrew; MITCHELL, Timothy; HWANG, Joseph K.; GOURLEY, Rebecca; ERICKSON, Stephen; DELANEY, Shani; KLINE, Carolyn R.; ARCHABALD, Karen; BLAIN, Michela; LACOURSE, Sylvia M.; ADAMS WALDORF, Kristina M.
title: Higher SARS-CoV-2 Infection Rate in Pregnant Patients
date: 2021-02-16
journal: Am J Obstet Gynecol
DOI: 10.1016/j.ajog.2021.02.011
sha: 1d57df9419f26a38138b3a08be8463e423098ec4
doc_id: 694280
cord_uid: ia8rou81

Background During the early months of the coronavirus disease of 2019 (COVID-19) pandemic, risks to pregnant women of a severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection were uncertain. Pregnant patients can serve as a model for the success of the clinical and public health response during public health emergencies as they are typically in frequent contact with the medical system. Population-based estimates of SARS-CoV-2 infections in pregnancy are unknown due to incomplete ascertainment of pregnancy status or inclusion of only single centers or hospitalized cases. Whether pregnant women were protected by the public health response or through their interactions with obstetrical providers in the early pandemic is poorly understood. Objective(s) To estimate the SARS-CoV-2 infection rate in pregnancy and examine disparities by race/ethnicity and English-language proficiency in Washington State. Study Design Pregnant patients with a polymerase chain reaction (PCR)-confirmed severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection diagnosed between March 1-June 30, 2020 were identified within 35 hospitals/clinic systems capturing 61% of annual deliveries in Washington State. Infection rates in pregnancy were estimated overall and by Washington State Accountable Community of Health (ACH) region and cross-sectionally compared to SARS-CoV-2 infection rates in similarly aged adults in Washington State. Race/ethnicity and language used for medical care among the pregnant patients were compared to recent data from Washington State. Results A total of 240 pregnant patients with SARS-CoV-2 infections were identified during the study period with 70.7% from minority racial and ethnic groups. The principal findings in our study are: 1) The SARS-CoV-2 infection rate in pregnancy was 13.9/1,000 deliveries (95% confidence interval [CI], 8.3-23.2) compared to 7.3/1,000 (95%CI 7.2-7.4) in 20-39 year old adults in Washington State (Rate Ratio [RR] 1.7, 95%CI 1.3-2.3), 2) the SARS-CoV-2 infection rate reduced to 11.3/1000 (95%CI 6.3-20.3) when excluding 45 cases of SARS-CoV-2 detected through asymptomatic screening (RR 1.3, 95%CI 0.96-1.9), 3) the proportion of SARS-CoV-2 cases in pregnancy among most non-white racial/ethnic groups was 2-4 fold higher than the race and ethnicity distribution of women in Washington State who delivered live births in 2018, and 5) the proportion of SARS-CoV-2 infected pregnant patients receiving medical care in a non-English language was higher than estimates of limited English proficiency in Washington State (30.4% versus 7.6%). Conclusions The SARS-CoV-2 infection rate in pregnant people was 70% higher than similarly aged adults in Washington State, which could not be completely explained by universal screening at delivery. Pregnant patients from nearly all racial/ethnic minority groups and patients receiving medical care in a non-English language were overrepresented. Pregnant women were not protected from COVID-19 in the early months of the pandemic with the greatest burden of infections occurring in nearly all racial/ethnic minority groups. This data coupled with a broader recognition that pregnancy is a risk factor for severe illness and maternal mortality strongly suggests that pregnant people should be broadly prioritized for COVID-19 vaccine allocation in the U.S. similar to some states.

A. Why was the study conducted? To determine the SARS-CoV-2 infection rate in 117 pregnant patients and assess racial/ethnic disparities in a multi-center, retrospective 118 cohort study in Washington State. 119 B. What are the key findings? The SARS-CoV-2 infection rate was significantly higher in 120 pregnant people (N=240; 13.9/1,000 deliveries) compared to 20-39 year olds (7.3/1,000; 121

Rate Ratio (RR) 1.7, 95%CI 1. 3-2.3) in Washington State. When compared to the 122 distribution of women in Washington State who delivered live births in 2018, the 123 proportion of SARS-CoV-2 cases in pregnancy among most racial and ethnic minority 124 groups was 2-4 fold higher. 125

In the early coronavirus disease of 2019 pandemic, risks associated with a 176 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in pregnancy 177 were uncertain. 1 As pregnant patients are typically in frequent contact with the medical 178 system, they can serve as a model for the success of the clinical and public health 179 response during public health emergencies. Outside of U.S. urban centers with high 180 infection rates, studies in the early pandemic reported low SARS-CoV-2 prevalence in 181 pregnant patients undergoing universal screening at admission for delivery. 2-4 182 Population-based estimates of SARS-CoV-2 infections in pregnancy are lacking due to 183 incomplete ascertainment of pregnancy status or inclusion of only single centers or 184 hospitalized cases. [5] [6] [7] [8] [9] [10] [11] Further, a disproportionate impact of COVID-19 on racial/ethnic 185 minorities, including among pregnant patients, has been reported. 5, [11] [12] [13] [14] [15] However, CDC 186 data is missing pregnancy status for 65% of their COVID-19 case report forms making it 187

impossible to estimate infection rates in the U.S. pregnant population. 16 based studies of COVID-19 in pregnancy with comprehensive data regarding race, 189 ethnicity, and language is essential to developing effective interventions for populations 190 disproportionately affected by 192 Washington State provides a valuable case study for evaluating the impact of 19 on pregnant individuals. Washington State was the first state to detect community 194 transmission of SARS-CoV-2 and impose a shelter-in-place order. 17 The ACH-specific and overall SARS-CoV-2 infection rates in pregnancy (per 1000) at 234 WA-CPC sites were estimated using the site-specific infection rate (number of 235 cases/number of deliveries during the study period) and Poisson regression (with 236 95%CI), with clustering by ACH region for the overall estimate. As a comparison group, 237 the SARS-CoV-2 infection rates in all 20-39 year olds (females and males) in 238

Washington State during the study period were calculated using publicly-available 239 SARS-CoV-2 surveillance data for confirmed cases (numerator) and 2019 population 240 estimates for 20-39 year olds (denominator); we were unable to exclude cases in men 241 due to limitations of the publicly available surveillance data. 21, 22 This group served as 242 the best available proxy estimate for the SARS-CoV-2 infection rate for reproductive-243 aged women. While women <20 and >39 years of age are fecund, Washington State 244 SARS-CoV-2 surveillance data were only available in wide categories including 0-19 245 years, 20-39 years, 40-59 years and older categories; neither age groups 0-19 nor 40-246 59 were appropriate comparison groups for approximating infection rates in most 247

reproductive-age women and therefore the 20-39 year old age group was selected for 248 comparison. Rate ratios (RR) and 95% confidence intervals (CI) were calculated 249 comparing WA-CPC infection rates in pregnancy to overall SARS-CoV-2 infection rates 250 among 20-39 years olds in Washington State within each ACH region; an ACH-251 weighted overall RR was also estimated. To assess how infection rates in pregnancy 252 may have been affected by increased access to testing in the pregnant population, we 253 conducted a sensitivity analysis excluding cases of SARS-CoV-2 in pregnancy detected 254 through asymptomatic universal screening prior to procedures or delivery. We were 255 unable to subtract cases in the general population comparison group similarly identified 256 through pre-procedure universal testing. Lastly, WA-DOH provided SARS-CoV-2 case 257 counts among pregnant females aged 18-50 between March 1-June 30, 2020 by ACH 258 region for comparison 23 ; pregnancy status was ascertained through public health 259 department investigation. As a sensitivity analysis, infection rates in pregnancy were Other Pacific Islander, Multi-Racial, and White; Hispanic was considered a mutually 267 exclusive race/ethnicity group to align with WA-DOH categories. 19 For each 268 race/ethnicity category among pregnant patients in the study population, prevalence 269 and exact 95%CI were estimated with clustering by ACH region. Then, we generated 270 ACH-weighted prevalence ratios (PR) and 95%CI comparing race/ethnicity in the study 271 population to the race/ethnicity distribution among women delivering in Washington 272

State in 2018. In addition, we generated prevalence ratios for the King and Greater 273

Columbia ACH regions, which had the highest number of SARS-CoV-2 cases through 274

June 30, 2020. 21 For ACH-specific analyses, race/ethnicity data were repressed when 275 there were <10 cases in alignment with WA-DOH privacy guidelines. In addition, we 276 compared the proportion of pregnant patients in our study a receiving medical care in a 277 non-English language to the proportion of individuals in Washington State in 2017 with 278 limited English language proficiency (individuals >5 years old, who speak English "less 279 than very well") per 2014-2017 American Community Survey data reported by the WA-280 DOH. 20 Each publicly-available data source and how it contributed to these analyses is 281 further described in Table 2 . Table  298 1). Of the WA-CPC cases, 15.8% (n=38) were detected in the first trimester, 27.9% 299 (n=67) in second trimester, and 56.3% (n=135) in third trimester pregnancies, as 300 previously reported. 24 Of these cases, 18.8% (45/240) were diagnosed through 301 asymptomatic screening strategies (pre-procedure and universal screening prior to 302 delivery); this excludes patients who were asymptomatic but tested due to having a 303 known exposure to 305 During the study period, the WA-DOH identified 346 cases of SARS-CoV-2 in 306 pregnancy throughout Washington state, but pregnancy status was missing for 35% of 307 cases in females aged 18-50. 25 The WA-CPC captured an estimated 69.4% (240/346) 308 of the total number of SARS-CoV-2 infections in pregnancy reported to the WA-DOH, 309 ranging from 26.7%-110.0% at the ACH region level (Table 1) . However, direct linking 310 of WA-CPC and WA-DOH cases was not possible so the exact overlap of WA-CPC and 311 WA-DOH identified cases is unknown. 312

The overall infection rate in pregnancy at WA-CPC sites was 13.9/1000 deliveries 315 (95%CI 8.3-23.2). At the ACH region level, infection rates in pregnancy at WA-CPC 316

Sites ranged from 6.2/1000 (95%CI 3.2-11.2) to 33.2/1000 deliveries (95%CI 26.9-40.9) 317 Table 2 ). In the King ACH region, where capture of annual deliveries and of 318 state reported SARS-CoV-2 cases in pregnancy were >90%, the infection rate in 319 pregnancy at WA-CPC sites was 12.9/1000 deliveries (95%CI 10.5-15.8). When 320 compared to the SARS-CoV-2 infection rate among 20-39 year olds in Washington 321 State of 7.3/1000 (95%CI 7.2-7.4), the overall infection rate in pregnancy at WA-CPC 322 sites was a significant 1.7 times higher (ACH-weighted RR 1.7; 95%CI 1.3-2.3; Table 2 ). 323

This equates to an absolute risk difference of 5.4/1000 (95%CI 0.8-10.0). There were 324 significantly higher infection rates in pregnancy in some, but not all, ACH regions (Table  325 2). For example, in the King ACH region, there was a 2.2 times higher rate of SARS-326

CoV-2 infections in pregnant women at WA-CPC sites compared to the 20-39 year old 327 population (RR 2.2, 95%CI 1.8-2.8). In the sensitivity analysis estimating the infection 328 rate in pregnancy using the WA-DOH reported SARS-CoV-2 in pregnancy case counts, 329 the statewide infection rate in pregnancy was similar to that estimated using data from 330 that were detected through asymptomatic screening strategies (pre-procedure and 334 universal testing at delivery) at WA-CPC sites, the overall infection rate in pregnancy at 335 WA-CPC sites was 11.3/1000 deliveries (95%CI 6.3-20.3), which was 30% higher than 336 the infection rate among Washingtonians aged 20-39 years old (ACH-weighted RR 1.3, 337 95%CI 0.96-1.9; Table S4 ). 338 339

Among the 240 SARS-CoV-2 cases in pregnancy detected by WA-CPC, the majority 341 were among racial and ethnic minority groups including 52.5% (n=126) among Hispanic 342 women, 8.3% (n=20) among Black women, and 3.3% each for American Indian/Alaska 343 Native (n=8), Asian (n=8), and Native Hawaiian/Other Pacific Islander (n=8) women 344 (Table 3) . When compared to the distribution of women in Washington State who 345 delivered live births in 2018, the proportion of SARS-CoV-2 cases in pregnancy among 346 most racial and ethnic minority groups were 2.0-3.9 fold higher ( 

In the early months of the COVID-19 pandemic, the SARS-CoV-2 infection rate was 383 70% higher in pregnant patients than in similarly-aged adults in Washington State. This 384 remained 30% higher after excluding pregnant patients whose SARS-CoV-2 infections 385 were detected through asymptomatic screening strategies including pre-procedure and 386 universal screening at delivery. We also detected significant disparities in the proportion 387 

While not considered an immunosuppressed condition, pregnancy is associated with an 399 increased risk of disease severity for some infections and potentially, acquisition risk. 30-400 19 infection susceptibility is challenging. While the increased infection rate in pregnant 403 patients may be largely driven by increased testing, it remained elevated compared to 404 the general population in the sensitivity analysis excluding cases detected through 405 universal testing pre-procedure and at delivery admission. Notably, our infection rate 406 estimate excluding asymptomatic cases was conservative as we were not able to 407 similarly exclude those in the general population whose infections were also detected 408 through universal testing prior to medical procedures. Whether an increased infection 409 rate in pregnancy has a biological basis or is due to other factors, such as increased 410 

This data provides the first evidence that pregnant individuals may have a higher SARS-430

CoV-2 infection rate than a similarly-aged population. Whether pregnant patients are 431 truly at a higher risk is yet unknown and exploring mechanisms for a potentially elevated 432 infection risk will be challenging with limited data currently available. However, this data 433 should lead to a greater public health response to prevent infections in pregnant women 434 and to focus efforts on individuals from minority racial/ethnic groups and with limited 435 pregnancy status was missing in approximately 35% of case report forms for 454 reproductive-aged females; we may have captured cases not reported to WA-DOH, but 455 were unable to estimate degree of non-overlap. The ideal comparison group for the WA-456 CPC SARS-CoV-2 cases in pregnancy would have been non-pregnant reproductive 457 aged females, but data on these women were not collected in our study. Therefore, the 458 best available comparison group for comparing infection rates to reproductive aged 459 females was publicly-available WA-DOH data; COVID-19 surveillance data were 460 available by age (presented in 20 year categories) or gender, but not both, necessitating 461 a comparison to females and males between 20-39 years. 21 In addition, we did not have 462 individual-case data for any publicly-available datasets so were unable to adjust for 463 individual level characteristics. Moreover, pregnant adolescents (<18 years old) were 464 excluded in our study, but included in overall delivery numbers; though, adolescents 465 only account for <1% of births in Washington State minimizing concern for bias. 19 466

Publicly-available WA-DOH data also served as imperfect proxies for the ideal 467 denominators for analyses of racial/ethnic and language disparities. Nonetheless, this 468 study provided statewide and regional assessments of infection rates in pregnancy, 469

including cases from all pregnancy trimesters, and identified pervasive demographic greatest risk for SARS-CoV-2 infection and associated adverse maternal-fetal 483 outcomes. 11, 18, 24, [41] [42] [43] Broader recognition that pregnancy is a risk factor for severe 484 illness and maternal mortality 16, 24, 25 J o u r n a l P r e -p r o o f 

Prevalence of SARS-CoV-2 Among 26

Characteristics of Symptomatic 17. Proclamation by the Governor Amending Proclamation 20-05: 20-25 Stay Home -Stay 31

Increased Risk of HIV Acquisition Among 618

Women Throughout Pregnancy and During the Postpartum Period: A Prospective Per-619

Analysis Among Women With HIV-Infected Partners

Risk of HIV Acquisition During Pregnancy and Postpartum: A Call for 622

Incidence of Maternal Sepsis and 624

Sepsis-Related Maternal Deaths in the United States

What Obstetric Health Care Providers Need to Know 37. COVID-19 Morbidity and Mortality by Race, Ethnicity and Language in Washington 629 State: Washington Department of Health

Addressing Health Equity During the COVID-19 Pandemic: Position Statement. American College of Obstetricians and Gynecologists

What's Our Role?

Clinical Findings and Disease Severity in 642

Hospitalized Pregnant Women With Coronavirus Disease 2019 (COVID-19)

Maternal death due to COVID-645 19

Care of critically ill pregnant

We would like to thank the pregnant patients contributing data to this manuscript, as 489 well as our partners across Washington State that enabled this investigation. We note 490 that we have shown single names for groups, such as "Hispanic" or "American 491 Indian/Alaska Native", which reflected an inclusive approach to naming, but does not 492 capture the spectrum of diversity in ancestry and cultural, behavioral and linguistic 493 differences. We also recognize the differences between sex and gender, noting that the 494 term "women" is not inclusive for biologically born female individuals that identify as 495 non-binary or transgender. Labels and words are imperfect and ethnic, cultural and 496 gender groups are sometimes overlapping or mischaracterized by single words or 497 names. We apologize if offense is taken regarding group names used in the manuscript. 498We thank Ms. Jane Edelson, who provided expert assistance with project management 499 at the University of Washington and was compensated on an hourly rate through the 500 J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f d For the Washington State COVID-19 in Pregnancy Collaborative, data were abstracted from the medical records. The "other" category reflects the patient's self-reported designation of their race/ethnicity to the health care provider. The Washington State Department of Health data does not include an "other" category.