Incidence, Clinical Characteristics, and Risk Factors of SARS-CoV-2 Infection among Pregnant Individuals in the United States

Abstract Background Data about the risk of SARS-CoV-2 infection among pregnant individuals are needed to inform infection prevention guidance and counseling for this population. Methods We prospectively followed a cohort of pregnant individuals during August 2020–March 2021 at three U.S. sites. The three primary outcomes were incidence rates of any SARS-CoV-2 infection, symptomatic infection, and asymptomatic infection, during pregnancy during periods of SARS-CoV-2 circulation. Participants self-collected weekly mid-turbinate nasal swabs for SARS-CoV-2 RT-PCR testing, completed weekly illness symptom questionnaires, and submitted additional swabs with COVID-19–like symptoms. An overall SARS-CoV-2 infection incidence rate weighted by population counts of women of reproductive age in each state was calculated. Results Among 1098 pregnant individuals followed for a mean of 10 weeks, nine percent (99/1098) had SARS-CoV-2 infections during the study. Population weighted incidence rates of SARS-CoV-2 infection were 10.0 per 1,000 (95% confidence interval [CI] 5.7–14.3) person-weeks for any infection, 5.7 per 1,000 (95% CI 1.7-9.7) for symptomatic infections, and 3.5 per 1,000 (95% CI 0-7.1) for asymptomatic infections. Among 96 participants with SARS-CoV-2 infection and symptom data, the most common symptoms were nasal congestion (72%), cough (64%), headache (59%), and change in taste or smell (54%); 28% had measured or subjective fever. The median symptom duration was 10 days (IQR6-16 days). Conclusion Pregnant individuals had a 1% risk of SARS-CoV-2 infection per week. Study findings provide information about SARS-CoV-2 infection risk during pregnancy to inform counseling for pregnant individuals about infection prevention practices, including COVID-19 vaccination.


Introduction
Emerging data from the COVID-19 pandemic suggest that pregnant individuals might be at increased risks for critical illness 1 and some adverse pregnancy outcomes. [2][3][4][5] While these data provide important information about risks to pregnant individuals once infected with SARS-CoV-2, data on community incidence of SARS-CoV-2 infection and risk factors for infection among pregnant individuals are needed to inform infection prevention counseling and guidance. 6 In addition, data that quantify the risk of SARS-CoV-2 infection during pregnancy might inform decisions by pregnant individuals about whether to get the COVID-19 vaccine.
Estimating SARS-CoV-2 infection incidence is challenging because it requires large-scale longitudinal community studies with systematic testing. 6 In addition, a substantial fraction of SARS-CoV-2 infection is asymptomatic. [7][8][9][10][11] Serologic studies might not capture the true incidence and full burden of infections because asymptomatic infection might not consistently elicit antibodies to SARS-CoV-2 and detectable antibody titers after mild or asymptomatic infection might wane over time. 12,13 In addition, serologic studies are unable to precisely identify the timing of incident infections. Thus, studies designed to identify the true incidence and full burden of SARS-CoV-2 infection ideally should include systematic molecular testing for asymptomatic infections. The Epidemiology of SARS-CoV-2 in Pregnancy and Infancy Community cohort is a prospective multisite cohort study of pregnant individuals who participate in intensive surveillance for SARS-CoV-2 infection that includes weekly mid-turbinate nasal swab specimen collection regardless of symptoms and weekly surveillance for illness symptoms. Using data from this cohort, we aimed to estimate incidence rates of asymptomatic and symptomatic SARS-CoV-2 infections as a measure of the overall risk of infection during pregnancy and describe the clinical characteristics of infections. In addition, we conducted an exploratory analysis to examine selected exposures and practices as risk or protective factors of SARS-CoV-2 infection.
A c c e p t e d M a n u s c r i p t 5

Participants and study setting
Participants were enrolled during August 2020-February 2021 at three U.S. medical centers in Salt Lake City, UT; New York City, NY; and Birmingham, AL. Individuals had to be pregnant at <28 weeks gestation to be eligible. See Supplemental Methods/Results for recruitment methods and additional eligibility criteria. During the study period from August 2020 through March 2021, local community mitigation measures varied by study location (see Supplemental Table 1).

Data Collection Procedures
Before data collection, individuals provided written informed consent to study participation.
At enrollment, participants completed web-based or telephone surveys about their socio-demographic characteristics, past medical and obstetrical histories, prenatal care, whether they were told by a healthcare provider they had suspected or confirmed COVID-19 before enrollment and if they had a confirmatory laboratory test. Participants received an orientation about the study surveillance, including how to self-collect and mail mid-turbinate nasal swab specimens. Participants were asked to self-collect swab specimens every week during pregnancy; self-collection of nasal swabs has been previously validated for detection of SARS-CoV-2 infection. 14 They were given prepared kits and materials to ship specimens on ice packs by overnight courier to a central laboratory. Participants were also asked to collect and ship additional swab specimens if they experienced onset of COVIDlike illness (CLI) symptoms.
CLI was defined as one or more of measured or subjective fever, cough, shortness of breath, sore throat, diarrhea, muscle aches, chills, or change in taste or smell. Participants were contacted weekly until the end of pregnancy by text message, phone call, or email to ascertain whether they had CLI symptoms or any other illness symptoms during the preceding 7 days. Every fourth week, participants were asked -In the past month, when you left your home for activities that involved interacting with other people, how often did you use a mask or other covering over your nose and

Laboratory Methods
Respiratory swab specimens were tested by reverse transcription-polymerase chain reaction (RT-PCR) for SARS-CoV-2 using assays previously approved under Emergency Use Authorization: Quidel Lyra SARS-CoV-2 Assay or the ThermoFisher Combo Kit platform with ThermoFisher probes and primers (see Supplemental Methods/Results). 15

Sample size estimates
Sample size estimates were based on achieving precision around incidence rate estimates of SARS-CoV-2 infection by site. Sample size calculations indicated that 280 participants per site would be needed to estimate a cumulative incidence of SARS-CoV-2 infection of 10% with ±4% precision after accounting for 10% cohort attrition.  In an exploratory analysis of risk and protective factors of SARS-CoV-2 infection, the primary outcome was any SARS-CoV-2 infection and exposures hypothesized to be possible risk or protective factors of SARS-CoV-2 infection included employment and telework status, residing in a household with a preschool aged child <5 years or with a school-aged child 5-17 years, and mask wearing when outside the house.

Analytic populations and statistical analysis
Individuals were considered enrolled if they met eligibility criteria, consented, and completed the enrollment questionnaire. Among enrolled participants, baseline characteristics were compared between participants who did and did not participate in surveillance to identify potential biases in the population available for analysis. All subsequent analyses were limited to enrolled individuals who participated in surveillance by submitting >1 weekly or acute illness swab specimen and did not report a diagnosis of laboratory-confirmed COVID-19 before enrollment. Among this analytic population, baseline characteristics were compared between participants who had incident SARS-CoV-2 infections versus those who did not. Underlying medical conditions were defined as those classified by CDC as conferring an increased risk for severe COVID-19. 5 Chi-square, Fisher's exact test, or the Wilcoxon rank-sum test were used to test for statistical significance. A p-value <0.05 was considered statistically significant.
Incidence rates per 1,000 person-weeks for each incident outcome were calculated with outcomes as the numerator and person-weeks at risk for events as the denominator. Person-weeks at risk for each outcome were calculated from the start of surveillance for each individual through the last week in which they participated in surveillance or through first incident case of SARS-CoV-2 infection (based on date of first positive test). To estimate incidences of any and asymptomatic SARS-CoV-2 infection, person-time was discounted for weeks in which participants did not submit a respiratory specimen. To estimate incidence of symptomatic SARS-CoV-2, person-time was discounted for weeks in which participants did not respond to weekly surveillance questionnaires or had symptomatic illness.
A c c e p t e d M a n u s c r i p t 8 Incidence rates were calculated for the cohort overall and by site using data available through March 28, 2021. Age-and race/ethnicity-adjusted incidence rates by site were calculated using negative binomial models. To give a representative estimate for individuals of childbearing age in the United States, an aggregate age-and race/ethnicity-adjusted incidence rate across all sites was also calculated by weighting age-and race/ethnicity-adjusted estimates from each site by the U.S. 2019 census population count of women aged 15-49 years in each state. 16 As a sensitivity analysis, incidence rates were calculated after restricting to periods of increased SARS-CoV-2 circulation at each site (see Supplemental Methods/Results). Additional sensitivity analyses were conducted to examine the impact on incidence rate estimates of including participants with COVID-19 infection before cohort enrollment and COVID-19 vaccination during the study period (see Supplemental Methods/Results).
The asymptomatic fraction of infection was calculated by dividing incidence rates of asymptomatic infection by rates of any infection. Ninety-five percent confidence intervals were calculated for selected frequencies and all incidence estimates assuming a binomial distribution.

Participant characteristics
Overall, 1,413 participants were enrolled in the cohort, of whom, 1619 (83%) participated in SARS-CoV-2 surveillance by submitting >1 weekly or acute illness specimen (Figure 1). See Table 2 Table 3). After restricting analyses to periods of increased SARS-CoV-2 circulation based on local surveillance data, the overall population weighted incidence of SARS-CoV-2 infection was 11.0 per 1,000 (95% CI 6.2-15.8) and site incidence rates adjusted for age and race/ethnicity ranged from 5.7 to 11.8 per 1,000 personweeks (Supplemental Table 4). See Supplemental Methods/Results for additional sensitivity analyses.

Risk Factors of SARS-CoV-2 infection
After adjusting for site and race/ethnicity, being employed and either teleworking or not teleworking (adjusted hazard ratio (aHR) 1.1, 95% CI 0.6-2.0, and 1.4, 95% CI 0.9-2.3, respectively, p=0.35) and residing in a household with a child aged <5 years (aHR 0.8, 95% CI 0.5-1.2, p=0.32) were not associated with an increased risk of SARS-CoV-2 infection (Table 2). However, residing in a household with a child 5-17 years of age was associated with an increased risk of SARS-CoV-2 infection (aHR 1.6, 95% CI 1.0-2.4, p=0.046). Non-adherence to mask wearing when outside the home was not statistically associated with risk of infection (never versus always wearing a mask: aHR Among 36 participants with symptomatic infection who completed illness follow-up surveys, the median symptom duration was 10 days (IQR 6-16), 19 participants (53%) reported missing work due to illness, and 14 (39%) reported seeking medical care for illness. One symptomatic infection was associated with a 2-day hospitalization for dehydration that did not require intensive care. There were no deaths.

Discussion:
Among a cohort of 1098 pregnant individuals at three U.S. sites, the overall age and race/ethnicity adjusted incidence rate of SARS-CoV-2 infection weighted for state populations of women of child-bearing age was 10.0 per 1,000 person-weeks indicating a 1% risk of infection per week during the study period. The cumulative incidence of infection during pregnancy was at least 9% based on infections during the study period. Incidence rate estimates in this cohort are similar to modelled estimates for US adults of childbearing age during February--December 2020 that adjust for case under-ascertainment (4.6-6.2 per 1,000 person-weeks). 17 Incidence rates among this likely largely COVID-19 unvaccinated cohort are also similar to those among unvaccinated healthcare and frontline workers (9.7 per 1,000 person-weeks) 18  In December 2020, the U.S. Advisory Committee on Immunization Practices stated that pregnant individuals should have the option to receive COVID-19 vaccine if they were in a group recommended for vaccination. 19 In July 2021, the American College of Obstetrics and Gynecology recommended that all pregnant individuals receive COVID-19 vaccine. 20 As COVID-19 vaccine becomes more widely available to pregnant individuals, vaccine hesitancy will likely remain a factor in their decision-making about vaccination. An analysis of vaccine acceptance among this cohort during August-December 2020 prior to Emergency Use Authorization of COVID-19 vaccines in the United States found that only 41% of participants would get a COVID-19 vaccine during pregnancy if given the opportunity. 21 Our findings suggest that pregnant individuals have a similar risk of becoming infected with SARS-CoV-2 compared to the general population with almost one in ten individuals in the cohort becoming infected during the study period. Although the SARS-CoV-2 infections in this study were largely self-limited, the long-term effects of SARS-CoV-2 infection and associated symptoms during pregnancy on perinatal and infant outcomes remain unclear. 4,22,23 Information from this study adds to the growing evidence base about SARS-CoV-2 infection during pregnancy that can inform counseling and risk communication for pregnant individuals as they make decisions about infection prevention measures, including COVID-19 vaccination, amidst evolving local guidance and mandates.
We estimated that 35% of SARS-CoV-2 infections during pregnancy are asymptomatic which is consistent with findings from a meta-analysis of studies among adults of childbearing age with longitudinal follow-up (31%, 95% CI 26-37%). 24 This finding underscores the potential risks of SARS-CoV-2 transmission from pregnant individuals with asymptomatic infection to others in their households and community and the potential risks for horizontal transmission to their newborns. In addition, in this study, pregnant individuals residing in households with children aged 5-17 years had a higher risk of SARS-CoV-2 infection than those who did not which may reflect an increased risk of A c c e p t e d M a n u s c r i p t 13 infection from children in the home or differences in community social mixing patterns among individuals with school-aged children.
Strengths of this study include enrollment of a large, demographically diverse community cohort of pregnant individuals, systematic surveillance and testing for asymptomatic and symptomatic SARS-CoV-2 infections, and a follow-up period that included periods of increased SARS-CoV-2 circulation at each study site. Cohort follow-up is ongoing and pregnancy outcomes will be reported once follow-up is complete. However, several study limitations should be considered. First, adherence rates to weekly swab specimen collection varied by site, and individuals who participated in surveillance differed from those who did not with respect to baseline characteristics and potential exposures that might influence risk of SARS-CoV-2 infection. Thus, estimates of cumulative SARS-CoV-2 infection incidence should be considered minimum estimates, and some infections may have been missed. Second, the study sample size was selected to estimate infection incidence, and the analysis of risk factors for infection might have been underpowered to detect small effect sizes. In addition, 30 percent of participants were not included in the analysis examining the association between mask-wearing and infection risk because they did not receive the monthly surveillance question about mask wearing due to a surveillance messaging error or did not respond to monthly surveillance contacts when the question was asked. Thus, the absence of association between potential risk and protective factors and SARS-CoV-2 infection should be interpreted with caution.   designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources.

Disclaimers:
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. Any references to specific commercial products are for identification purposes only and do not constitute an endorsement by CDC. CDC funded this study. CDC-affiliated authors were involved in study design, data collection, analysis and interpretation, report writing, and the decision to submit the paper for publication. The corresponding author had full access to all data used in the analysis and had final responsibility for the decision to submit for publication. a Prescreened includes all individuals who study staff screened for selected eligibility criteria such as age before formal eligibility screening. \ b Ineligible because individual was not planning to deliver at the study site (n=29), not willing to self-collect and mail respiratory specimens (n=49), had unknown gestational age (n=1), or was currently enrolled in an influenza or COVID-19 vaccine trial (n=3).