Attitudes toward COVID-19 illness and COVID-19 vaccination among pregnant women: a cross-sectional multicenter study during August-December 2020

43

Evaluate pregnant women’s attitudes toward COVID-19 illness and vaccination and 44 identify factors associated with vaccine acceptability. 45 Study Design: Cross-sectional survey among pregnant women enrolled in a prospective 46 COVID-19 cohort study in Salt Lake City, UT, Birmingham, AL, and New York, NY, August 9–47 December 10, 2020. Women were eligible if they were 18-50 years old and <28 weeks of 48 gestation. Upon enrollment, women completed surveys regarding concerns about COVID-19 49 illness and likelihood of getting COVID-19 vaccine if one were available during pregnancy. 50 Vaccine acceptability was defined as a response of “very likely” or “somewhat likely” on a 4-51 point Likert scale. Factors associated with vaccine acceptability were assessed with 52 multivariable logistic regression. 53

Of 939 pregnant women eligible for the main cohort study, 915 (97%) consented to 54 participate. Among these 915 women, 39% self-identified as White, 23% Black, 33% Hispanic, 55 and 4% Other. Sixty-two percent received an influenza vaccine last season. Seventy-two 56 percent worried about getting sick with COVID-19. If they were to get sick, 92% worried about 57 harm to their pregnancy and 80% about harm to themselves. Only 41% reported they would get 58 a vaccine. Of women who were unlikely to get vaccinated, the most frequently cited concern 59 was vaccine safety for their pregnancy (82%). Non-Hispanic Black and Hispanic women had 60 lower odds of accepting a vaccine compared with non-Hispanic White women (adjusted odds 61 ratios (aOR) 0.4, 95%CI 0.2–0.6 for both). Receipt of influenza vaccine during the previous 62 season was associated with higher odds of vaccine acceptability (aOR 2.1, 95%CI 1.5-3.0). 63

Although most pregnant women worried about COVID-19 illness, <50% were 64 willing to get vaccinated during pregnancy. Racial and ethnic disparities in plans to accept 65 COVID-19 vaccine highlight the need to prioritize strategies to address perceived barriers 66 among groups at high risk for COVID-19. 67 68 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 4

69 As of March 1, 2021, there have been over 114 million cases of COVID-19 with 70 more than 2.5 million deaths worldwide(1). Of the 1.1 million cases in the United States, 71 there have been more than 73,000 laboratory-confirmed cases of COVID-19 among 72 pregnant women(2). Based on accumulating data, pregnant women appear to be at 73 increased risk for severe COVID-19 disease(3, 4). In a recent report that included 74 400,000 women of reproductive age, pregnant women with COVID-19 were found to be 75 more likely than non-pregnant women to be admitted to the intensive care unit, receive 76 extracorporeal membrane oxygenation, and die(5). Comparisons of rates of SARS-CoV-77 2 infection and risks for COVID-19 disease among pregnant versus non-pregnant 78 women in the United States also suggest that Hispanic women may be at increased risk 79 of infection and Hispanic and non-Hispanic Black women may be at increased risk of 80 severe disease(5, 6). At this time, the risk of transplacental transmission of COVID-19 to 81 the developing fetus appears to be low,(7, 8) and most studies have shown no 82 increased risk of spontaneous abortions or stillbirth(9, 10). However, pregnant women 83 with COVID-19 illness may be at increased risk for preterm birth(11). 84 Although social distancing, hand hygiene, and mask wearing are key non-85 pharmaceutical interventions to slow the spread of COVID-19, widespread safe and 86 effective vaccination is ultimately necessary to control this global pandemic(12). Many 87 COVID-19 vaccine trials are currently being conducted worldwide with many vaccines in 88 Phase III testing and now a few COVID-19 vaccines approved for emergency use (14, 89 15). Pregnant women have not been included in the Phase III trials. However, both the 90 American College of Obstetricians and Gynecologists (ACOG) and the Society for 91 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 5 Maternal Fetal Medicine (SMFM) recommend that pregnant women be included in 92 vaccine trials and offered COVID-19 vaccines (16, 17). In addition, the Centers for 93 Disease Control and Prevention recommends that pregnant women be provided the 94 opportunity to choose whether to receive COVID-19 vaccine under the current 95 Emergency Use Authorization(18, 19). However, the willingness of pregnant women to 96 be vaccinated is unknown. 97 We evaluated pregnant women’s attitudes toward COVID-19 illness and 98 vaccination during pregnancy and assessed factors associated with vaccine 99 acceptability. Based on findings from studies of influenza vaccine acceptance among 100 adults,(20) we hypothesized that women from racial and ethnic minority groups would 101 be less willing to be vaccinated. 102 103

104 We performed a cross-sectional study of pregnant women enrolled August 9–105 December 10, 2020, in the Epidemiology of Severe Acute Respiratory Syndrome in 106 Pregnancy and Infancy (ESPI) Community Cohort, an ongoing prospective longitudinal 107 cohort study conducted at three centers in the United States (Birmingham, AL; Salt 108 Lake City, UT; and New York, NY). The ESPI Community Cohort study is designed to 109 estimate the incidence of SARS-CoV-2 infection, identify risk factors for infection, and 110 characterize the clinical spectrum of infection among pregnant women receiving 111 prenatal care at the three study sites. Centralized Institutional Review Board approval 112 was obtained (IRB-AAAT1906), and informed consent was obtained from all 113 participants. 114 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 6 Women were eligible for participation in the main cohort if they were 18–50 years 115 old and were at less than 28 weeks of gestation to allow for an average of at least 12 116 weeks of surveillance time in the cohort prior to end of pregnancy. Women without a 117 functioning telephone and those who were not willing to respond to weekly COVID-19 118 surveillance questionnaires or to self-collect nasal swabs weekly were not eligible for 119 participation, as these were key components of the ESPI Community Cohort study. 120 Additionally, women who were unable to speak and read either English or Spanish, and 121 those currently enrolled in a COVID-19 or influenza vaccine trial, were not eligible. 122 Women were not excluded based on prior suspected or confirmed COVID-19 infection. 123 Study staff attempted to approach all women who met age and gestational age criteria 124 for enrollment among those receiving prenatal care at the three participating centers 125 during the study period. Participants were recruited over the phone and in person at 126 outpatient prenatal care offices. 127 Women who consented to participate in the ESPI Community Cohort were asked 128 to complete a standardized survey at the time of study enrollment. The survey included 129 questions about maternal demographics and socioeconomic characteristics, past 130 medical and obstetric history, and attitudes towards SARS-CoV-2 infection/COVID-19 131 illness and vaccination in pregnancy. Women were considered fully enrolled in the main 132 ESPI Community Cohort study if they consented to participation and completed eight 133 core questions on the enrollment survey about enrollment date, estimated delivery date, 134 number of gestations, diagnosis of gestational hypertension or gestational diabetes 135 during the current pregnancy, presence of underlying medical conditions, and number of 136 prior pregnancies. 137 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 7 Using a 4-point Likert scale, pregnant women were asked “How worried are you 138 about getting sick with COVID-19?” as well as “If you were to get sick with COVID-19, 139 how worried are you that COVID-19 would harm you?” and “…how worried 140 are you that COVID-19 would harm your pregnancy?” Women were deemed concerned 141 if they answered “very worried” or “somewhat worried.” Women were asked about their 142 most trusted source for receiving information about COVID-19, and women were also 143 asked “If a COVID-19 vaccine were to become available for pregnant women, how likely 144 would you be to get the vaccine for yourself during your pregnancy?” Vaccine 145 acceptability was defined as women who answered “very likely” or “somewhat likely.” 146 Women were considered not willing to get a vaccine if they answered “not too likely” or 147 “not at all likely.” Lastly, women were asked to answer multiple choice questions about 148 the reasons why they would (or would not) get a vaccine; in addition to standard 149 response choices, questions included an option to indicate other reasons with free text 150 entry. Answer choices for reasons for getting the vaccine included: “to protect myself 151 from getting sick with COVID-19”, “to protect my pregnancy,” “to protect others in my 152 family,” and “to protect others in the community from getting sick with COVID-19.” 153 Answer choices for reasons for not getting the vaccine included: “concerns or question 154 about vaccine safety for myself,” “concerns or questions about vaccine safety for my 155 pregnancy,” “concerns or questions about whether the vaccine would work to protect 156 me from COVID-19,” and “I don’t think I need the vaccine.” 157 A complete case analysis was performed including women who enrolled in the 158 cohort study from August 9 to December 10, 2020. Baseline maternal demographic and 159 obstetric characteristics in the population were described in aggregate and by vaccine 160 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 8 acceptability. Descriptive statistics were used to summarize pregnant women’s 161 responses to questions about COVID-19 illness and vaccination during pregnancy. 162 Logistic regression was used to estimate the association between baseline 163 characteristics and vaccine acceptability. Key characteristics assessed included study 164 site, maternal age, self-reported race and ethnicity, education, employment status, 165 employment in a healthcare occupation, household income, maternal medical 166 comorbidities, and prior receipt of influenza vaccination during the 2019–20 influenza 167 season. Race and ethnicity were characterized as Black if women identified as Non-168 Hispanic Black or African American, White if women identified as Non-Hispanic White, 169 and Hispanic if women identified as Hispanic or Latino regardless of race. All 170 characteristics that were significantly associated with vaccine acceptability in 171 unadjusted analyses at p<0.20 were included in the multivariable logistic regression 172 model, except for study site and employment status, which were highly correlated with 173 other covariates. Backward step-wise selection using a p-value cut-off of 0.2 and 174 evaluating for a >10% change in adjusted odds ratios was used to achieve a final, 175 parsimonious model from which adjusted odds ratios and 95% confidence intervals (CI) 176 were estimated. To assess for differences in predictors of vaccine acceptability before 177 and after the first release of phase III COVID-19 vaccine trial results in the United States 178 in early November 2020, multivariable models were run among all women and stratified 179 by women who completed enrollment surveys during August 9-November 14 and 180 November 15-December 20, 2020 using the same explanatory variables selected for 181 the final primary model. Adjusted odds ratios and 95% confidence intervals (CI) were 182 estimated. 183 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 9 The associations between: (1) women’s level of concern about getting sick with 184 COVID-19 and (2) women’s level of concern about harm to themselves or their 185 pregnancy from COVID-19 in the setting of SARS-CoV-2 infection and vaccine 186 acceptance were also assessed using bivariate analysis. 187 SAS Version 9.4 was used for statistical analysis, and statistical significance was 188 set at p=0.05. 189 190

191 During the study period, 1186 women were screened for eligibility for the main 192 cohort study, of whom 939 (79%) were eligible. Among these 939 women, 915 (97%) 193 consented to participate in the main cohort study and were fully enrolled. 194 Among the 915 women included in this analysis, 39% self-identified as White, 195 23% as Black, and 33% as Hispanic. Overall, 64% of women had more than a high 196 school education, 60% were employed, and 20% lived in households with income below 197 the local poverty line. Twenty-eight percent of women had one or more underlying 198 medical conditions. Last season 62% of women reported receiving the influenza vaccine 199 (Table 1). 200 Seventy-two percent (95% CI: 69%-75%) of women stated they were worried 201 about getting sick with COVID-19. If they were to get sick with COVID-19, 92% (95% CI: 202 91%-94%) of women were worried that COVID-19 would harm their pregnancies, and 203 80% (95% CI: 77%-82%) were worried that COVID-19 would harm them (Figure 1). 204 When asked what source of COVID-19 information women trusted the most, the most 205 common answer was their obstetrician/gynecologist (42%, 95% CI: 38%-45%), followed 206 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 10 by their family doctor or primary care provider (28%, 95% CI: 25%-30%), CDC (13%, 207 95% CI: 11%-15%), and other medical professionals (4%, 95% CI: 3%-5%). 208 Overall, 41% (374/915, 95% CI: 38%-44%) of women reported they would get a 209 COVID-19 vaccine if one became available during their pregnancy. When stratified by 210 enrollment month, the proportion of women willing to get a COVID-19 vaccine was 211 similar (data not shown). Among women who were willing to get a vaccine during 212 pregnancy, the most commonly cited reason for vaccine acceptability was to protect 213 their pregnancy (95%, 95% CI: 93%-98%). Other reasons included protecting 214 themselves (85%, 95% CI: 81%-89%), protecting family members (79%, 95% CI: 75%-215 83%), and protecting the community (68%, 95% CI: 63%-73%). In contrast, women who 216 were not willing to get a vaccine during pregnancy most frequently cited concerns about 217 vaccine safety for their pregnancy (82%, 95% CI: 78%-85%). Other reasons included 218 concerns about vaccine safety for themselves (68%, 95% CI: 63%-72%), vaccine 219 effectiveness (52%, 95% CI: 47%-56%), and the belief that they did not need the 220 vaccine (22%, 95% CI: 18%-26%). 221 Pregnant women’s willingness to accept a COVID-19 vaccine varied by maternal 222 race and ethnicity as well as other baseline characteristics. Women who were non-223 Hispanic White were more likely to be willing to accept the vaccine than women who 224 were non-Hispanic Black or Hispanic (Figure 2). For example, 63% (95% CI: 58%-69%) 225 of women who were non-Hispanic White stated that they were either very likely or 226 somewhat likely to accept the vaccine whereas only 31% (95% CI: 25%-38%) of non-227 Hispanic Black and 35% (95% CI: 29%-41%) of Hispanic women were very likely or 228 somewhat likely to accept the vaccine. Stated otherwise, 69% (95% CI: 62%-75%) of 229 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 11 non-Hispanic Black and 65% (95% CI: 59%-71%) of Hispanic women were not likely to 230 receive the vaccine compared to only 37% (95% CI: 31%-42%) of non-Hispanic White 231 women. In addition to maternal race and ethnicity, Utah study site, having a graduate 232 school degree, and getting an influenza vaccine during the 2019-2020 influenza season 233 were also associated with COVID-19 vaccine acceptability (Table 2). However, in 234 adjusted models that included race and ethnicity, educational level, and getting an 235 influenza vaccine during the 2019-2020 influenza season, only race and ethnicity and 236 prior influenza vaccine acceptance were significantly associated with willingness to get 237 a COVID-19 vaccine (Table 2). Women who self-identified as non-Hispanic Black and 238 women who identified as Hispanic had lower odds of accepting a COVID-19 vaccine 239 compared to non-Hispanic White women (adjusted odds ratio [aOR] 0.4, 95% CI 0.2–240 0.6 for each comparison). In contrast, women who reported getting the 2019-2020 241 influenza vaccine had higher odds of accepting a COVID-19 vaccine compare to those 242 who did not (aOR 2.1, 95% CI 1.5-3.0). Model findings were consistent among women 243 who completed the enrollment survey during August 9-November 14, 2020 versus 244 November 14-December 10, 2020. 245 Women who were worried about getting sick with COVID-19 were more likely to 246 say they would get a COVID-19 vaccine during pregnancy than women who were not 247 worried about getting sick (49% vs. 40%, p=0.01). Women’s worry about COVID-19 248 illness harming themselves and their pregnancies in the setting of COVID-19 illness was 249 not significantly associated with vaccine acceptability (p=0.24 and p=0.60, respectively). 250 251

252 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 12 In this cross-sectional study of more than 900 pregnant women from three 253 diverse centers across the U.S., three-fourths were concerned about getting sick with 254 COVID-19 and worried that COVID-19 could harm themselves and their pregnancies. 255 However, less than half of pregnant women in this study said they would be likely to get 256 a COVID-19 vaccine if one were available during their pregnancy, although this was 257 consistent with data from a national survey of the general adult population in the US 258 from the same period(21). Pregnant Black and Hispanic women were less likely to be 259 willing to get a COVID-19 vaccine than White women. Women who received an 260 influenza vaccine during the previous season were more likely to be accepting of a 261 COVID-19 vaccine. Common perceived barriers to vaccination among women who were 262 unwilling to get a COVID-19 vaccine included concerns about vaccine safety for their 263 pregnancies and themselves, and concerns about vaccine effectiveness. Three in four 264 pregnant women identified healthcare professionals as their most trusted source of 265 information about COVID-19, with 41% specifying their obstetrician/gynecologist. 266 Previous studies of influenza vaccination in pregnancy have identified similar 267 perceived barriers to vaccination to those found in our study, including concerns about 268 vaccine safety and effectiveness and similar perceived benefits including providing 269 protection to the pregnancy(22). Healthcare provider recommendation is one of the 270 strongest and most consistent predictors of influenza vaccination in the United States 271 and globally(22, 23). In our study, we also found that the majority of women identified 272 healthcare providers as their most trusted source of information about COVID-19 and 273 prior receipt of influenza vaccine was associated with higher odds of COVID-19 vaccine 274 acceptability. Our findings, coupled with findings related to acceptance of other currently 275 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 13 recommended vaccines during pregnancy, indicate that obstetricians and other 276 healthcare providers will play a critical role in counseling pregnant women about the 277 risks of COVID-19 illness and providing information about the safety and effectiveness 278 of COVID-19 vaccines. Currently, both ACOG and SMFM advocate that pregnant 279 women have the option to receive COVID-19 vaccines, and that shared decision-280 making be utilized by each pregnant woman and her provider regarding vaccination(16, 281 17). CDC also provides guidance and resources for healthcare professionals to discuss 282 vaccination with patients before and as COVID-19 vaccines become more widely 283 available in the United States (https://www.cdc.gov/vaccines/covid-19/hcp/engaging-284 patients.html). 285 In this study, Black race and Hispanic ethnicity were associated with lower odds 286 of COVID-19 vaccine acceptability. These findings must be considered in the context of 287 the well-documented history of unethical medical experimentation among racial and 288 ethnic minority populations in the United States that has engendered mistrust of medical 289 interventions among some communities as well as racial and ethnic disparities in key 290 social determinants of health that may influence vaccine acceptability(24). For example, 291 two recent studies in the United States found that Black women were less likely to 292 report receiving a healthcare provider offer or referral for influenza or Tdap vaccination 293 and had lower rates of receipt of both vaccines during pregnancy(25, 26). In addition, 294 several studies have now documented that SARS-CoV-2 infection rates in the United 295 States are higher among Black and Hispanic communities and identified disparities in 296 COVID-19 severity among pregnant women from these groups(5, 27). For example, in a 297 recent Morbidity and Mortality Weekly Report , non-Hispanic Black or African American 298 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 14 women represented 27% of COVID-19 deaths in pregnancy, but comprised only 15% of 299 infected pregnant women(5). In the same analysis, pregnant Hispanic women 300 experienced SARS-CoV-2 infection at a disproportionately higher rate compared to non-301 pregnant Hispanic women. Our findings that women who face a greater risk of harm 302 from COVID-19 infection are less willing to accept COVID-19 vaccination highlight the 303 need for outreach and communication strategies to address perceived barriers to 304 vaccination among groups that may be less likely to get a COVID-19 vaccine. 305 Our study has several strengths. Our population was comprised of a diverse 306 group of pregnant women from three different centers across the United States. Efforts 307 were made to approach all women receiving prenatal care at study clinics for study 308 participation in the main cohort to optimize representation of the study source 309 population. The survey collected detailed information about common perceived 310 motivators and barriers to vaccination, tailored specifically for the contexts of pregnancy 311 and the COVID-19 pandemic. 312 Several limitations should be considered when interpreting study findings. While 313 the study was designed to maximize diversity and generalizability, only women who had 314 access to prenatal care and who consented to participate in the prospective cohort were 315 included. Thus, our results may not be generalizable to all pregnant women. We also 316 did not enroll a non-pregnant group, and thus we cannot comment about how these 317 findings may differ from a non-pregnant population in the same communities. Given that 318 this study was conducted before a COVID-19 vaccine was available in the United 319 States, we could not assess factors associated with actual vaccine receipt and our 320 findings are based on hypothetical acceptability of a COVID-19 vaccine. In addition, as 321 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 15 with all surveys, our findings may be subject to social desirability bias in which 322 participants are more likely to respond in a manner that they perceive to be socially 323 acceptable. Vaccine acceptance as well as perceived motivators and barriers to 324 acceptance may evolve among pregnant women as additional data about COVID-19 325 vaccine safety and efficacy become available. Additional studies will be needed to 326 monitor trends in vaccine acceptance among pregnant women over time. 327 Widespread vaccination is the most promising strategy to end the current global 328 pandemic, while hand-washing, social distancing, mask wearing and other key non-329 pharmaceutical interventions will remain important mainstays of COVID-19 prevention. 330 The results of this study provide insight into potential racial and ethnic disparities in 331 COVID-19 vaccine acceptability among pregnant women. When a vaccine is available 332 to pregnant women in the United States, obstetricians and other health care providers 333 will play a critical role in counseling pregnant women about COVID-19 illness and 334 offering COVID-19 vaccine to pregnant women. Similar to influenza vaccination, a clear 335 recommendation by obstetric care professionals to pregnant women to take the vaccine 336 will likely increase COVID-19 vaccine uptake. This may be especially important for 337 pregnant women in racial and ethnic groups who may be at greater risk for infection and 338 severe disease and who also appear to be less likely to accept a vaccine at this time 339 based on the findings of our study. In order to overcome these disparities, outreach 340 programs with community collaboration may be important to customize patient 341 education materials that improve communication and shared decision-making in order 342 to achieve health equity in vaccination during pregnancy. 343 344 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 16

345 1. Johns Hopkins University & Medicine. COVID-19 Dashboard by the Systems Science and 346 Engineering (CSSE). 2020 [cited 2020 12/9/2020]; Available from: 347 https://coronavirus.jhu.edu/map.html 348 2. Prevention CfDCa. Data on COVID-19 during Pregnancy: Severity of Maternal Illness. 2020 349 [cited 2020 December 18]; Available from: https://covid.cdc.gov/covid-data-tracker/#pregnant-350 population 351 3. Delahoy MJ, Whitaker M, O'Halloran A, Chai SJ, Kirley PD, Alden N, et al. Characteristics 352 and Maternal and Birth Outcomes of Hospitalized Pregnant Women with Laboratory-Confirmed 353 COVID-19 - COVID-NET, 13 States, March 1-August 22, 2020. MMWR Morb Mortal Wkly Rep 354 2020 Sep 25;69(38):1347-54. 355 4. Centers for Disease Control and Prevention. People with Certain Medical Conditions. 2020 356 [cited 2020 12/3/2020]; Available from: https://www.cdc.gov/coronavirus/2019-ncov/need-extra-357 precautions/people-with-medical-conditions.html 358 5. Zambrano LD, Ellington S, Strid P, Galang RR, Oduyebo T, Tong VT, et al. Update: 359 Characteristics of Symptomatic Women of Reproductive Age with Laboratory-Confirmed SARS-360 CoV-2 Infection by Pregnancy Status - United States, January 22-October 3, 2020. MMWR 361 Morb Mortal Wkly Rep 2020 Nov 6;69(44):1641-7. 362 6. Ellington S, Strid P, Tong VT, Woodworth K, Galang RR, Zambrano LD, et al. Characteristics 363 of Women of Reproductive Age with Laboratory-Confirmed SARS-CoV-2 Infection by 364 Pregnancy Status - United States, January 22-June 7, 2020. MMWR Morb Mortal Wkly Rep 365 2020 Jun 26;69(25):769-75. 366 7. Dumitriu D, Emeruwa UN, Hanft E, Liao GV, Ludwig E, Walzer L, et al. Outcomes of 367 Neonates Born to Mothers With Severe Acute Respiratory Syndrome Coronavirus 2 Infection at 368 a Large Medical Center in New York City. JAMA Pediatr 2020 Oct 12. 369 8. Salvatore CM, Han JY, Acker KP, Tiwari P, Jin J, Brandler M, et al. Neonatal management 370 and outcomes during the COVID-19 pandemic: an observation cohort study. Lancet Child 371 Adolesc Health 2020 Oct;4(10):721-7. 372 9. Cosma S, Carosso AR, Cusato J, Borella F, Carosso M, Bovetti M, et al. Coronavirus disease 373 2019 and first-trimester spontaneous abortion: a case-control study of 225 pregnant patients. 374 Am J Obstet Gynecol 2020 Oct 8. 375 10. Panagiotakopoulos L, Myers TR, Gee J, Lipkind HS, Kharbanda EO, Ryan DS, et al. SARS-376 CoV-2 Infection Among Hospitalized Pregnant Women: Reasons for Admission and Pregnancy 377 Characteristics - Eight U.S. Health Care Centers, March 1-May 30, 2020. MMWR Morb Mortal 378 Wkly Rep 2020 Sep 23;69(38):1355-9. 379 11. Woodworth KR, Olsen EO, Neelam V, Lewis EL, Galang RR, Oduyebo T, et al. Birth and 380 Infant Outcomes Following Laboratory-Confirmed SARS-CoV-2 Infection in Pregnancy - SET-381 NET, 16 Jurisdictions, March 29-October 14, 2020. MMWR Morb Mortal Wkly Rep 2020 Nov 382 6;69(44):1635-40. 383 12. Lerner AM, Folkers GK, Fauci AS. Preventing the Spread of SARS-CoV-2 With Masks and 384 Other "Low-tech" Interventions. JAMA 2020 Nov 17;324(19):1935-6. 385 13. Zimmer C, Corum J, Wee S-L. Coronavirus Vaccine Tracker. 2020 [cited 2020 386 12/11/2020]; Available from: https://www.nytimes.com/interactive/2020/science/coronavirus-387 vaccine-tracker.html?auth=link-dismiss-google1tap 388 14. U.S. Food and Drug Administration. Pfizer-BioNTech COVID-19 Vaccine Emergency Use 389 Authorization Letter. 2020 [cited 2020 12/14/2020]; Available from: 390 https://www.fda.gov/media/144412/download 391 15. Administration USFaD. Moderna COVID-19 Vaccine EUA Letter of Authorization. [cited 392 2020 December 21]; Available from: https://www.fda.gov/media/144636/download 393 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 17 16. Zahn CM. ACOG Statement on COVID-19 and Pregnancy. 2020 [cited 2020 11/30/2020]; 394 Available from: https://www.acog.org/news/news-releases/2020/06/acog-statement-on-covid-19-395 and-pregnancy 396 17. Society for Maternal-Fetal Medicine. Society for Maternal-Fetal Medicine (SMFM) 397 Statement: SARS-CoV-2 Vaccination in Pregnancy. 2020 [cited 2020 12/1/2020]; Available 398 from: https://s3.amazonaws.com/cdn.smfm.org/media/2591/SMFM_Vaccine_Statement_12-1-399 20_(final).pdf 400 18. Centers for Disease Control and Prevention. Pregnancy and Breastfeeding Considerations. 401 2020 [cited 2020 12/14/2020]; Available from: https://www.cdc.gov/coronavirus/2019-402 ncov/vaccines/recommendations/pregnancy.html 403 19. Centers for Disease Control and Prevention. Interim Clinical Considerations for Use of 404 Pfizer-BioNTech COVID-19 Vaccine. 2020 [cited 2020 12/14/2020]; Available from: 405 https://www.cdc.gov/vaccines/covid-19/info-by-product/pfizer/clinical-considerations.html 406 20. Egede LE, Zheng D. Racial/ethnic differences in influenza vaccination coverage in high-risk 407 adults. Am J Public Health 2003 Dec;93(12):2074-8. 408 21. Nguyen K, Srivastav A, Razzaghi H, Williams W, Lindley M, Jorgensen C, et al. COVID-19 409 Vaccination Intent, Perceptions, and Reasons for Not Vaccinating Among Groups Prioritized for 410 Early Vaccination — United States, September and December 2020. MMWR Morb Mortal Wkly 411 Rep 2021;70(6):217–22. 412 22. Yuen CY, Tarrant M. Determinants of uptake of influenza vaccination among pregnant 413 women - a systematic review. Vaccine 2014 Aug 6;32(36):4602-13. 414 23. Myers KL. Predictors of maternal vaccination in the United States: An integrative review of 415 the literature. Vaccine 2016 Jul 25;34(34):3942-9. 416 24. Egede LE, Walker RJ. Structural Racism, Social Risk Factors, and Covid-19 - A Dangerous 417 Convergence for Black Americans. N Engl J Med 2020 Sep 17;383(12):e77. 418 25. Lindley MC, Kahn KE, Bardenheier BH, D'Angelo DV, Dawood FS, Fink RV, et al. Vital 419 Signs: Burden and Prevention of Influenza and Pertussis Among Pregnant Women and Infants - 420 United States. MMWR Morb Mortal Wkly Rep 2019 Oct 11;68(40):885-92. 421 26. Razzaghi H, Kahn KE, Black CL, Lindley MC, Jatlaoui TC, Fiebelkorn AP, et al. Influenza 422 and Tdap Vaccination Coverage Among Pregnant Women - United States, April 2020. MMWR 423 Morb Mortal Wkly Rep 2020 Oct 2;69(39):1391-7. 424 27. Moore JT, Ricaldi JN, Rose CE, Fuld J, Parise M, Kang GJ, et al. Disparities in Incidence of 425 COVID-19 Among Underrepresented Racial/Ethnic Groups in Counties Identified as Hotspots 426 During June 5-18, 2020 - 22 States, February-June 2020. MMWR Morb Mortal Wkly Rep 2020 427 Aug 21;69(33):1122-6. 428 429 430 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 18 TABLES 431 432 Table 1. Baseline demographic and other characteristics (n=915) Total N=915 n (col %) Willing to get vaccine* n=374 n (row %) Not willing to get vaccine* n=429 n (row %) p-value Site Birmingham, AL Salt Lake City, UT New York City, NY 267 (29) 340 (37) 308 (34) 98/245 (40) 196/311 (63) 80/247 (32) 147/245 (60) 115/311 (37) 167/247 (68) <0.01 Age, years 18-34 35-50 741 (81) 172 (19) 295/647 (46) 79/155 (51) 352/647 (54) 76/155 (49) 0.22 Race/ethnicity White, non-Hispanic Black, non-Hispanic Hispanic Other, non-Hispanic 346 (39) 201 (23) 293 (33) 38 (4) 202/319 (63) 58/186 (31) 83/238 (35) 16/30 (53) 117/319 (37) 128/186 (69) 155/238 (65) 14/30 (47) <0.01 Education Less than high school diploma High school diploma Some college/technical school College degree Graduate degree 57 (6) 265 (29) 230 (25) 214 (23) 147 (16) 23/52 (44) 89/225 (40) 63/191 (33) 106/193 (55) 93/141 (66) 29/52 (56) 136/225 (60) 128/191 (67) 87/193 (45) 48/141 (34) <0.01 Household below poverty level* 177 (20) 73/161 (45) 88/161 (55) 0.71 Marital status Married Never married Committed partnership 473 (52) 267 (30) 127 (14) 235/425 (55) 83/224 (37) 39/112 (35) 190/425 (45) 141/224 (63) 73/112 (65) <0.01 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 19 Other 37 (4) 15/33 (45) 18/33 (55) Employed 549 (60) 239/493 (48) 254/493 (52) 0.23 Occupation† Healthcare professional Other 228 (42) 316 (58) 108/203 (53) 130/285 (46) 95/203 (47) 155/285 (54) 0.10 ≥1 underlying medical condition‡ 253 (28) 107/232 (46) 125/232 (54) 0.87 Previously had suspected/confirmed COVID-19 before enrollment 44 (5) 20/37 (54) 17/37 (46) 0.64 Received Influenza Vaccine Last Year 564 (62) 273/497 (55) 224/497 (45) <0.01 *Based on number of household members: https://aspe.hhs.gov/poverty-guidelines. Column % do not include missing. If data are missing for a variable, n included are indicated in parentheses after variable. *Willing to get vaccine defined as survey response of “very likely” or “somewhat likely”. Not willing to get vaccine defined as “not likely” or “not at all likely.” Women who responded “unknown” are not included in the analysis (n=103). †Among all employed (n=549) ‡Defined as asthma, chronic lung, metabolic, hematologic, cardiovascular, renal, hepatic, neurologic, rheumatologic disease or hypertension. Total Missing/Decline to respond/Unknown: age (2, <1%), race/ethnicity (37, 4%), education (2, <1%), households below poverty level (15, 2%), marital status (11, 1%), employed (7, <1%), occupation (5, <1%), underlying conditions (1, <1%), previous COVID-19 (2, <1%), received influenza vaccine last year (21, 2%). Abbreviations: IQR, interquartile range. 433 434 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 20 Table 2. Logistic regression models evaluating factors associated with COVID-19 vaccine acceptance (n=803) OR (95% CI) aOR (95% CI)† Site Birmingham, AL 1.4 (1.0-2.0) Salt Lake City, UT 3.6 (2.5-5.1) New York City, NY Ref Age, years (n=802) 18-34 0.8 (0.6-1.1) 35-50 Ref Race/ethnicity (n=773) White, non-Hispanic Ref Ref Black, non-Hispanic 0.3 (0.2-0.4) 0.4 (0.2-0.6) Hispanic 0.3 (0.2-0.4) 0.4 (0.2-0.6) Other, non-Hispanic 0.7 (0.3-1.4) 0.8 (0.4-1.7) Education (n=802) Less than high school diploma Ref Ref High school diploma 0.8 (0.4-1.5) 1.1 (0.5-2.1) Some college or technical school 0.6 (0.3-1.2) 0.6 (0.3-1.2) College degree 1.5 (0.8-2.8) 1.0 (0.5-2.0) Graduate school degree 2.4 (1.3-4.7) 1.3 (0.6-2.8) Household Below Poverty Level (n=791) Yes 0.9 (0.7-1.3) No Ref Employed (n=797) Yes 1.2 (0.9-1.6) No Ref Health Care Occupation (n=792) Yes 1.4 (1.0-1.9) No†† Ref . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 21 Underlying Medical Condition‡ ≥1 1.0 (0.7-1.3) None Ref Received Influenza Vaccine Last Year (n=784) Yes 2.6 (1.9-3.6) 2.1 (1.5-3.0) No Ref Ref If data are missing for a variable, n included are indicated in parentheses after variable. Missing/Decline to Respond/Unknown answers not included in calculations. *Willing to get vaccine defined as survey response of “very likely” or “somewhat likely”. Not willing to get vaccine defined as “not likely” or “not at all likely.” Women who responded “unknown” are not included in the analysis (n=103). †Final adjusted model includes variables for which aOR reported above. ††Includes women who are not employed. ‡Defined as asthma, chronic lung, metabolic, hematologic, cardiovascular, renal, hepatic, neurologic, or rheumatologic disease or hypertension. Abbreviations: OR, odds ratio; CI, confidence interval; aOR, adjusted odds ratio 435 436 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint 22 FIGURES 437 438 439 440 Footnote: Women whose self-reported race/ethnicity was Other, non-441 Hispanic (n=38) and women with unknown race/ethnicity (n=37) are not shown in 442 responses stratified by race and ethnicity. 443 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 29, 2021. ; https://doi.org/10.1101/2021.03.26.21254402doi: medRxiv preprint