COVID-19 vaccine uptake and attitudes among pregnant and postpartum parents.

BACKGROUND: Pregnancy poses increased risks from COVID-19, including hospitalization and premature delivery. Yet pregnant individuals are less likely to have received a COVID-19 vaccine.
OBJECTIVE: This study aimed to investigate COVID-19 vaccine uptake and reasons for delay or refusal among perinatal parents. STUDY
DESIGN: A total of 1542 eligible parents who delivered between 2019 and 2021 were surveyed through the Ovia parenting app, which has a nationally representative user base. Adjusted and nationally weighted means were calculated. Multivariate logistic regression and survival models were used to examine uptake.
RESULTS: At least 1 dose of the COVID-19 vaccine was received by 70% of the parents. Those with a bachelor's or graduate degree were significantly more likely to have received a vaccine relative to those with some college or less (adjusted odds ratio for bachelor's degree, 1.854; 95% confidence interval, 1.19-2.90; adjusted odds ratio for graduate degree, 2.833; 95% confidence interval, 1.69-4.75). Parents living in rural areas were significantly less likely to have received a vaccine relative to those living in urban areas (adjusted odds ratio for small city, 0.62; 95% confidence interval, 0.45-0.86; adjusted odds ratio for rural area, 0.56; 95% confidence interval, 0.35-0.89); 56% (281/502) of unvaccinated parents considered that the vaccine "was too new." Among those pregnant in 2021, 44% (258/576) received at least 1 dose, and 34% (195/576) reported that pregnancy had "no impact" on their vaccine decision.
CONCLUSION: There was significant heterogeneity in vaccine uptake and attitudes toward vaccines during pregnancy by sociodemographics and over time. Public health experts need to consider and test more tailored approaches to reduce vaccine hesitancy in this population.

Why was this study conducted?

More than 1500 new/expecting mothers who delivered between 2019 and 2021 were surveyed to investigate COVID-19 vaccine uptake and reasons for delay as there are significant benefits to vaccination for this population and their offspring.

What are the key findings?

More than 70% had at least one dose of the COVID-19 vaccine and 56% of unvaccinated parents said the vaccine “was too new.”

Among those pregnant in 2021, 44% obtained at least one dose and 34% reported that pregnancy had “no impact” on their vaccine decision.

What does the study add?

There was significant heterogeneity in vaccine uptake and attitudes about vaccines during pregnancy by socio-demographics and over time.

Alt-text: Unlabelled box

Introduction

COVID-19 vaccines are critical for pregnant people because pregnancy poses an increased risk of contracting COVID-19, being hospitalized, and giving birth prematurely.1, 2, 3, 4, 5, 6 Despite no evidence of safety risks for pregnant people and their offspring, and significant evidence of risks from SARS CoV-2 infection, COVID-19 vaccine acceptance and uptake in the perinatal population has lagged behind the general population. The CDC currently estimates about 67% of pregnant people were vaccinated as of January 2022, whereas nationally, about 77% of women between the ages of 18 and 39 were vaccinated.

Although the first COVID-19 vaccines became available in in the U.S. in 2020 for individuals age 16 and older, strong recommendations regarding vaccination for pregnant and lactating parents came later after analyses of vaccine surveillance data found no safety concerns. Several subsequent studies have provided greater evidence supporting these recommendations.9, 10, 11, 12, 13, 14 In addition, new evidence has shown that infants benefit from maternal COVID-19 vaccination, with lower rates of hospitalization from COVID-19 infection during the first six months of life relative to infants with unvaccinated mothers and greater antibody persistence relative to infants whose mothers had a natural SARS CoV-2 infection.

Under-vaccination in this group may be occurring for several reasons. Prior research on vaccine uptake has emphasized the “5Cs” as explanatory drivers for hesitancy: confidence, calculation of risks, constraints, complacency, and collective responsibility. Confidence reflects individuals’ beliefs about the safety and efficacy of the COVID-19 vaccine, which may have been diminished among this population due to concerns about insufficient testing in pregnant and postpartum individuals. In fact, early guidelines were somewhat equivocal with unclear guidance for pregnant people given their lack of representation in early clinical trials. Related to this, calculation of risks were likely affected both by the confidence in the scientific evidence, but also from the significant misinformation about safety and effectiveness. , There were false claims that the COVID-19 vaccine affected fertility, increased miscarriage risks, and negatively impacted male reproductive organs. , Constraints, including physical access or availability were less likely to be an issue in many communities in the U.S. but, complacency due to perceptions that COVID-19 is “just a flu” and greater emphasis on individualism vs. collectivism in the U.S. are ongoing challenges. ,

Although prior work has documented that a significant proportion of people (including those who work in healthcare settings) are delaying or forgoing vaccination, , we know little about how decisions among birthing parents evolved over time as the pandemic progressed and the key factors in decisions to delay or refuse COVID-19 vaccines. This lack of information is preventing public health officials from more effectively tailoring communications to parents in the perinatal period and from designing interventions to combat misinformation about COVID-19 vaccines as well as other vaccines recommended during pregnancy. In this study, we examined the COVID-19 vaccine uptake and reasons for delay or refusal among perinatal parents who delivered infants from 2019-2021. We examined variation among individuals who were pregnant when the vaccine became available to the public and how pregnancy impacted individuals’ decision to get vaccinated.

Materials and Methods

Study Design and Population

In this cross-sectional study, we surveyed mothers from October to January 2022 across three “birth” cohorts based on the date they delivered their infants: 1) July-December 2019 (infant aged 22 to 26 months at the time of survey), 2) March-May 2020 (infant aged 17 to 20 months at the time of survey), and 3) June-August 2021 (infant aged 2 to 5 months at the time of survey). Parents of multiples were instructed to answer the survey based on the first-born child in a given cohort.

Active users of the Ovia parenting app, which is one of Ovia's suite of parenting and pregnancy apps available in the U.S. for free download on iOS and Android devices, were invited to complete our anonymous survey. Recruitment occurred until we obtained approximately 1500 respondents. Several previous studies have sampled Ovia users, who look similar to the national population of birthing parents with respect to demographic characteristics.26, 27, 28, 29 Participants received a $10 Amazon e-gift card for completing the survey. RAND's IRB approved the study.

To be eligible, users needed to be age 18-45 and the birthing parent of an infant born in one of the three birth cohorts delineated above. The survey instrument included 40 questions on vaccination status, including the month of receipt, several socio-demographic questions (mother's age, race/ethnicity, state of residence, urbanicity of residence, health insurance status, level of education, marital status, and income) as well as questions about breastfeeding. A draft version of the survey was assessed with 6 Ovia app users who gave birth in the spring of 2021. The survey was then revised to improve clarity and flow following cognitive testing.

Measures

Our primary outcome measure was receipt of COVID-19 vaccine based on the respondent's self- report of having received at least one dose as of January 2022. We examined receipt among our full sample and as well as limited to those who were pregnant in the spring of 2021 when the COVID-19 vaccine first became available to the public. In addition, we calculated frequencies of the reasons reported for not getting vaccinated and the role of pregnancy in influencing decisions regarding vaccine receipt.

We examined outcomes in multivariate regression models adjusting for key socio-demographic and delivery measures: birthing parent's age category (18-24, 25-29, 30-34, 35+), race/ethnicity (Black/African American, white, Hispanic, other), education (high school degree or less, some college, college degree, or graduate degree or higher), annual income category (< $25,000, $25,000 to $39,999, $40,000 to $54,999, $55,000 to $79,999, $80,000 or more, not reported) married, health insurance (private/commercial, public, or self-pay), urbanicity of residence (large city, small city or town, suburb near a large city, or rural area), whether the infant had an neonatal intensive care unit stay at delivery, whether the infant was ever breastfed, and weeks of gestation at delivery.

Statistical Analysis

We calculated sample descriptive statistics, including unadjusted means of our outcome variables that were weighted to be nationally representative for comparison to other estimates of vaccination rates among pregnant persons. Weights were calculated using Centers for Disease Control and Prevention data on national births in 2019 using a raking procedure. We used multivariate logistic regression and a Cox proportional hazard model to examine correlates of vaccine receipt (yes/no) and timing (time to event), respectively. Hazard models are appropriate for modeling time to event analyses. In these models, we adjusted for maternal socio-demographics, infant characteristics at delivery, and delivery month and year. We calculated frequencies of reasons for not getting vaccinated stratified by birthing parent socio-demographics (age group, race/ethnicity, and educational attainment). Finally, we examined whether parents who were pregnant in the spring of 2021 reported that pregnancy impacted their decision to get vaccinated. Statistical significance was determined using p-values of < 0.05.

Results

6,184 Ovia users clicked on the advertisement that explained the survey opportunity, and 1,617 (26%) of those were eligible (i.e. delivered during the 3 periods of interest) to participate. After excluding observations with item non-response of our key measures, our analytic sample included 1,542 parents (95%). In Table 1 , we report unweighted sample descriptive statistics for the full sample and stratified by vaccination status. The weighted proportion of parents in the full sample who reported receiving at least one dose of the COVID-19 vaccine was 70% (95% CI: 55-74%) as of January 2022, with an average vaccination date of May 2021 for the first dose. We found statistically significant differences in maternal age, race/ethnicity, education, marital status, income and urbanicity of residence by vaccination status.

Table 1

Sample Descriptive Statistics, Stratified by COVID-19 Vaccination Status.

	Full Sample (n = 1,542)		By COVID-19 Vaccine Status as of January 2022				P-value
			Had at least 1 dose		Had no doses
Delivery Date			(n = 1,105)		(n = 437)
Pre-pandemic (7/2019 - 12/2019)	501	(32%)	337	(30%)	164	(38%)	0.01
Early pandemic (3/2020 - 5/2020)	504	(33%)	362	(33%)	142	(32%)
Late pandemic (6/2021 - 8/2021)	537	(35%)	406	(37%)	131	(30%)
Infant NICU Stay	187	(12%)	133	(12%)	54	(12%)	0.86
Ever Breastfed	1446	(94%)	1039	(94%)	407	(93%)	0.51
Weeks of Gestation at Delivery
23-28 weeks	10	(1%)	7	(1%)	3	(1%)	0.61
29-32 weeks	20	(1%)	15	(1%)	5	(1%)
33-36 weeks	121	(8%)	81	(7%)	40	(9%)
37-38 weeks	410	(27%)	285	(26%)	125	(29%)
39-40 weeks	794	(51%)	583	(53%)	211	(48%)
41+ weeks	187	(12%)	134	(12%)	53	(12%)
Maternal Age
18-24	179	(12%)	102	(9%)	77	(18%)	0.00
25-29	417	(27%)	246	(22%)	171	(39%)
30-34	518	(34%)	395	(36%)	123	(28%)
35+	428	(28%)	362	(33%)	66	(15%)
Maternal Race/Ethnicity
White	881	(57%)	640	(58%)	241	(55%)	0.01
Other	148	(10%)	111	(10%)	37	(8%)
Black/African American	142	(9%)	89	(8%)	53	(12%)
Hispanic/Latinx	371	(24%)	265	(24%)	106	(10%)
Maternal Education
High School Degree or Less	227	(15%)	122	(11%)	105	(24%)	0.00
Some College	425	(28%)	241	(22%)	184	(42%)
Bachelor's Degree	492	(32%)	392	(35%)	100	(23%)
Graduate Degree	398	(26%)	350	(32%)	48	(11%)
Married	1260	(82%)	935	(85%)	170	(74%)	0.00
Has Private Health Insurance	1167	(76%)	906	(82%)	261	(60%)
Has Public Health Insurance	344	(22%)	182	(16%)	162	(37%)
Uninsured	31	(2%)	17	(2%)	14	(3%)
Annual Household Income
< $25,000	155	(10%)	88	(8%)	67	(15%)	0.00
$25,000 - $39,999	237	(15%)	126	(11%)	111	(25%)
$40,000 - $54,999	191	(12%)	121	(11%)	70	(16%)
$55,00 - $79,999	208	(13%)	143	(13%)	65	(15%)
$80,000+	645	(42%)	551	(50%)	94	(22%)
No Answer	106	(7%)	76	(7%)	30	(7%)
Urbanicity
A large city	409	(27%)	301	(27%)	108	(25%)	0.00
Suburb near large city	589	(38%)	476	(43%)	113	(26%)
Small city or town	417	(27%)	256	(23%)	161	(37%)
Rural area	127	(8%)	72	(7%)	55	(13%)

Data are presented as frequencies and column percentages (weighted to match national population estimates) in parentheses. The p-value is based on the chi-square test to test for differences in measures/categories by vaccination status.

In Table 2 , we report the adjusted odds ratios and hazard ratios from the logistic and Cox proportional hazard models, respectively. Across both models, there appears to be an education gradient with parents with a Bachelor's degree or Graduate degree being significantly more likely to have received one or more COVID-19 vaccines as compared to those with some college or less (aOR for Bachelor's degree, 1.854, 95% CI, 1.19, 2.90; aOR for Graduate degree, 2.833, 95% CI, 1.69,4.75; aHR for Bachelor's degree, 1.574, 95% CI, 1.20, 2.06; aHR for Graduate degree, 2.078, 95% CI, 1.56, 2.76). We also found consistent evidence that parents living in rural areas or small cities/towns were significantly less likely to have received COVID-19 vaccine compared to those in urban areas (aOR for small city, 0.62, 95% CI, 0.45, 0.86; aOR for rural area, 0.56, 95% CI, 0.35,0.89; aHR for small city, 0.70, 95% CI, 0.59, 0.84; aHR for rural area, 0.62, 95% CI, 0.47,0.82).

Table 2

Adjusted Odds Ratios and Hazard Ratios for Received Any COVID-19 Vaccine and Month of Receipt.

	Logistic (DV = Vaccinated)		Cox Proportional Hazard Model
	OR	95% CI	HR	95% CI
Delivery Date
Pre-pandemic (7/2019 - 12/2019)	REFERENCE
Early pandemic (3/2020 - 5/2020)	1.36*	[1.01,1.83]	1.20*	[1.02,1.40]
Late pandemic (6/2021 - 8/2021)	1.30	[0.96,1.76]	1.05	[0.90,1.22]
Infant NICU Stay (yes)	1.23	[0.81,1.86]	1.19	[0.96,1.48]
Ever Breastfed (yes)	0.98	[0.59,1.60]	0.93	[0.72,1.20]
Weeks of Gestation at Delivery
23-28 weeks	REFERENCE
29-32 weeks	1.63	[0.27,9.95]	1.18	[0.47,2.96]
33-36 weeks	0.81	[0.18,3.73]	0.94	[0.43,2.06]
37-38 weeks	0.89	[0.20,3.92]	0.97	[0.45,2.09]
39-40 weeks	0.80	[0.18,3.56]	1.02	[0.47,2.18]
41+ weeks	0.83	[0.18,3.78]	0.95	[0.44,2.06]
Maternal Age
18-24	REFERENCE
25-29	0.65*	[0.43,0.97]	0.78	[0.60,1.01]
30-34	0.82	[0.52,1.29]	0.92	[0.70,1.20]
35+	1.29	[0.78,2.12]	1.11	[0.84,1.47]
Maternal Race/Ethnicity
White	REFERENCE
Other	1.26	[0.80,1.99]	0.90	[0.73,1.11]
Black/African American	0.66	[0.42,1.01]	0.73**	[0.57,0.92]
Hispanic/Latinx	1.13	[0.83,1.53]	0.98	[0.84,1.13]
Maternal Education
High School Degree or Less	REFERENCE
Some College	0.86	[0.59,1.25]	0.94	[0.73,1.20]
Bachelor's Degree	1.85**	[1.19,2.90]	1.57**	[1.20,2.06]
Graduate Degree	2.83***	[1.69,4.75]	2.08***	[1.56,2.76]
Married	0.83	[0.59,1.17]	0.90	[0.74,1.09]
Has Private Health Insurance	1.64*	[1.09,2.49]	1.24	[0.96,1.61]
Has Public Health Insurance	0.87	[0.56,1.35]	0.93	[0.71,1.21]
Uninsured	REFERENCE
Annual Household Income
< $25,000	REFERENCE
$25,000 - $39,999	0.71	[0.45,1.11]	0.74*	[0.55,0.98]
$40,000 - $54,999	0.94	[0.57,1.54]	0.92	[0.68,1.24]
$55,00 - $79,999	0.83	[0.48,1.41]	0.95	[0.69,1.30]
$80,000+	1.30	[0.77,2.20]	1.23	[0.91,1.66]
No Answer	0.94	[0.51,1.71]	0.96	[0.68,1.36]
Urbanicity
A large city	REFERENCE
Suburb near large city	1.23	[0.88,1.70]	1.02	[0.88,1.19]
Small city or town	0.62**	[0.45,0.86]	0.70***	[0.59,0.84]
Rural area	0.56*	[0.35,0.89]	0.62***	[0.47,0.82]

***p<0.01, **p<0.05, *p<0.10 All covariates shown were included in the estimation models.

Figure 1 shows the reasons parents reported not receiving one or more COVID-19 vaccines across the full sample of unvaccinated parents (shown in pink) and by racial/ethnic groups where we found significant differences (Chi-square p-value < 0.05). The most common reasons for lack of receipt were related to the newness (56%) and lack of trust in the vaccine (32%) as well as concerns related to harms to fertility (27%) and to the infant (36%). The most frequently reported reason for not receiving vaccine (vaccine was too new) was the same across all racial/ethnic groups; however, 67% [95% CI:61-73%] of unvaccinated White parents vs. 55% of Black/African American [95% CI:41-68%] and 55% of Latinx parents [95% CI:45-64%] reported this. White parents were also more likely to report that they did not get vaccinated because they were concerned that it would affect their fertility (40% [95% CI:34-46%]vs. 19% and 20% of Black/African American [95% CI:8-30%] and 20% of Latinx parents [95% CI:12-27%], respectively or they were concerned it could harm their infant (49% [95% CI:42-55%] vs. 34% of Black/African American [95% CI:21-47%] and Latinx parents [95% CI:25-43%]).

Figure 1

Percentage of Non-vaccinated Parents Reporting Each Reason for Not Getting Vaccinated, Stratified by Race/Ethnicity Figure 1 depicts the frequencies of reasons given for not being vaccinated by unvaccinated parents as percentages of all unvaccinated parents and subgroups by race/ethnicity.

Next, we present results weighted nationally among parents who were pregnant in the spring of 2021 when COVID-19 vaccine first became available to the public (Figure 2 ). Overall, 44% (95% CI: 43-51%) obtained at least one dose of the COVID-19 vaccine by January 2022, with 36% receiving the vaccine during pregnancy and 8% only after birth. There were stark differences in vaccine receipt and timing by educational attainment. Among those with a high school degree or less, only 29% received at least one dose during pregnancy, but an additional 19% received one or more doses after birth. However, among those with some college, we did not observe the same pattern, as only 18% received a dose during pregnancy with an additional 1% reporting one or more doses after delivery. Among parents with a Bachelor's degree or graduate degree, vaccination rates overall were greater than among the other groups, with the large majority of those who were vaccinated choosing to do so during pregnancy rather than delaying until after giving birth.

Figure 2

Percentage of Parents Pregnant in 2021 Who Obtained at Least 1 Dose of the COVID-19 Vaccine, Stratified by Educational Attainment Figure 2 shows vaccination uptake among parents pregnant during 2021broken out by educational attainment. Uptake is reported for receipt during pregnancy and after birth.

Reported Impact of Pregnancy on Decision to Get Vaccinated Figure 3 shows the distripution of responses to the question “What impact did pregnancy have on your decisions about COVID-19 vaccination? Check all that apply.”

Among those pregnant in 2021, we asked about the impact that pregnancy had on their decision-making regarding COVID-19 vaccine. More than one-third (n=195, 34%) reported that being pregnant had “no impact” on their decision regarding vaccination, and 161 (28%) reported that being pregnant encouraged them to get vaccinated earlier than they would have otherwise. Only 40 (7%) reported that being pregnant made them less interested in getting vaccinated as compared to 133 (23%) who reported that being pregnant made them more interested in getting vaccinated. 90 (16%) reported that they delayed vaccination due to their pregnancy.

Discussion

Principal Findings

Among parents using a popular pregnancy and parenting app targeted for sample inclusion, we found about 70% had received at least one dose of the COVID-19 vaccine as of January 2022, but vaccination rates varied considerably by parental education and urbanicity. Reasons for not getting vaccinated were predominately related to concerns about the newness and safety of the vaccine, with White parents reporting these concerns at much higher rates relative to Black/African American or Latinx parents.

Results in the Context of What is Known & Clinical Implications

The increased rate of these concerns among Whites relative to minority parents is somewhat surprising given that prior research has shown higher rates of vaccine hesitancy among some minoritized racial/ethnic groups. , Our findings might be due to the fact that our sample includes a disproportionate share of parents with college or graduate degrees, across all race/ethnic groups, and those individuals are less likely to be vaccine hesitant. , Given that concerns related to the vaccine being new and questions about safety are consistently among the top reasons for not getting vaccinated overall and across subgroups, public health and clinical providers should focus on efforts to address those concerns.

Studies on effective strategies to reduce vaccine hesitancy among pregnant women have found some evidence that providing education and information to expectant parents, providing additional training to medical staff, and leveraging health systems tools (e.g., clinical decision supports) can improve vaccine uptake. , However, the level of misinformation surrounding the COVID-19 vaccine has reached new levels as internet platforms and social media have made it easy for misinformation to spread.37, 38, 39 One approach may be to frame the vaccine educational information preemptively to expectant parents warning about misinformation that they may encounter as they try to decide about whether the vaccine is right for them (incidentally referred to as “inoculation” in the communications literature). , Of course, this requires additional time from already time-constrained providers and resources to stay current on the latest conspiracy theories or falsehoods being shared.

In addition, providers need to be honest with patients in cases where the scientific evidence is unclear so as not to undermine credibility and trust, particularly among racial and ethnic groups where medical mistrust persists due to historical mistreatment both directly within medicine, but also more systemically within U.S. society. For example, early in 2021, there was limited data on vaccine safety during pregnancy, with post-administration surveillance analyses only being published in April 2021 in the U.S. The studies indicated that vaccination did not increase miscarriage risk. Pregnant individuals who endeavored to “research” vaccine safety themselves would have found significant variation internationally in recommendations for COVID19 vaccination during pregnancy. Taken together, patient concerns about unknown safety and efficacy of the vaccine (at least early on) suggest providers needed to acknowledge evidence uncertainty while at the same time presenting the risks from COVID-19 infection during pregnancy. This can improve both patients’ confidence and risk calculations – two key factors in vaccine hesitancy.

There was significant heterogeneity in the reported impacts of pregnancy on vaccine decision-making. Overall, more than half of respondents reported that being pregnant increased their interest in vaccination, including 28% who reported wanting to get vaccinated earlier. However, about 23% were less interested or deliberately delayed getting vaccinated due to pregnancy. This suggests that improving uptake may require addressing different challenges from among the “5 Cs” depending on parental attitudes. Those who have positive feelings about getting vaccinated due to pregnancy but have not yet received all doses may have practical constraints (e.g., difficulty taking time off from work) or as case rates wane, may become complacent. Those who have negative feelings may be exposed to more misinformation, which requires a different approach.

Research Implications

More research is needed to test these different strategies to determine the most effective approaches for improving uptake among these different populations.

Strengths and Limitations

Our study provides insight into the evolution of parental decisions to obtain the COVID-19 vaccine over time and the key reasons for vaccine refusal, which is critical to developing public health and provider approaches to increasing vaccine uptake. We note the following limitations. Although our sample was drawn from a nationally representative user base that has been used in several other studies, eligible parents (delivering in the windows of interest in 2019, 2020, and 2021) had to decide to participate in the study which means the results presented here may not be representative. The advantage, however, of our use of online surveys from a popular platform used by new and expecting parents is that we were able to survey a relatively large and geographically dispersed population of parents. As our study relied on self-reported survey data and did not include items on all measurable parental characteristics, the usual concerns of potential biases from survey research, including recall bias (particularly among those pregnant in 2019), social desirability bias, and omitted variables bias apply.

Conclusions

In conclusion, we found that in our sample of parents in the perinatal period, vaccination rates were lower than among all adults age 18-45 in the U.S., largely due to reasons that are particularly salient for this population (fertility and infant safety). We also found significant variation in the timing of vaccine receipt among those who were pregnant in 2021 and a non-linear correlation with of vaccination status with education. Taken together, public health experts need to consider and test more tailored approaches to reduce vaccine hesitancy among pregnant and postpartum parents who are uniquely vulnerable to COVID-19.

Declaration of Competing Interest

None.