Do Pregnant Persons Want Influenza Vaccines? Knowledge, Attitudes, Perceptions, and Practices Toward Influenza Vaccines in 8 Low- and Middle-Income Countries

Abstract

A substantial body of evidence demonstrates that pregnant persons are at higher risk of hospitalization from influenza during pregnancy. Moreover, persons with severe influenza illness are more likely to have adverse pregnancy and birth outcomes, including stillbirth, preterm birth and infants who are small for gestational age [1–6]. Seasonal influenza vaccination protects their fetus and infant, providing up to 6 months of conferred immunity to infants [4, 7–9]. As a result, the World Health Organization's (WHO) Strategic Advisory Group of Experts on Immunization (SAGE) includes pregnant persons among their priority groups for seasonal influenza vaccination [10, 11]. Despite these recommendations and the well-documented safety of seasonal influenza vaccines during pregnancy [12, 13], some pregnant persons [14–16] and their maternal care providers [17, 18] are hesitant to accept and recommend influenza vaccines. Their concerns center around lack of awareness of influenza illness or perceptions of influenza as a mild illness and lack of knowledge about influenza vaccine safety [17–19].

Influenza vaccine coverage among pregnant persons in low- and middle-income countries remains very low [20], largely due to low access to vaccines. During the past 10 years, the US Centers for Disease Control and Prevention (CDC) and the Task Force for Global Health have worked with ministries of health to support expanded access to seasonal influenza vaccines in some low- and middle-income countries [21]. Considering their hesitancy to accept influenza vaccines, successfully reaching pregnant persons requires interventions that effectively address barriers, raise awareness, and counteract negative perceptions of seasonal influenza vaccines. Surveys that collect knowledge, attitudes, perceptions, and practices (KAP) related to influenza and influenza vaccines have been valuable tools in the past, both to predict uptake among populations [22] as well as to craft impactful awareness campaigns and design strategies aimed at reducing barriers to vaccination. We describe the findings of KAP surveys administered to pregnant persons in 8 low- and middle-income countries to inform actions to improve vaccine uptake among this priority population.

METHODS

We pooled individual-level data from cross-sectional surveys about seasonal influenza vaccine conducted among pregnant persons in 8 countries with existing CDC collaborations to establish or strengthen influenza vaccination programs and/or policies (Albania, Armenia, Cote d’Ivoire, Kyrgyzstan, Lao People's Democratic Republic [PDR], Morocco, Tunisia, Uganda) during 2018–2019. The surveys used a common protocol based on the Health Belief Model [23]. Standardized questionnaires measured perceptions of susceptibility, severity, benefits, barriers, and cues to action for influenza vaccination, with minor modifications reflecting local data needs. Datasets from each country were translated and aligned by matching question and answer choices to create a core analytic dataset.

A purposive, nationally representative sample of health facilities was selected in each country; within each selected facility, a convenience sample of pregnant persons presenting for routine antenatal care visits during any trimester of pregnancy and during the period of data collection were selected for interviews; respondents were representative of urban and rural settings, and ethnic and linguistic groups. Minor differences in study site type and selection criteria reflect local health delivery systems and infrastructure and may include public or private health facilities (antenatal care clinics, general health clinics, or vaccination clinics).

A mix of countries with (Albania, Armenia, Lao PDR, Morocco, Tunisia) and without (Cote d’Ivoire, Kyrgyzstan, Uganda) seasonal influenza vaccine programs for pregnant persons were included in our analysis. Countries reporting introduction of seasonal influenza vaccine among pregnant persons via the WHO/UNICEF Joint Reporting Form [24] for 2 consecutive years, including the year of data collection, were classified as those with vaccination programs. Country income status was assigned using the World Bank lending group classification [25] during the year the surveys were administered.

We analyzed attitudes and perceptions toward seasonal influenza vaccine among pregnant persons, stratified by presence/absence of an influenza vaccination program, to explore differences between country groups (ie, practice vs intent). We analyzed data by WHO region based on the assumption of shared cultural norms and values within a region. Responses to KAP survey questions were provided on 5-point Likert scales. We dichotomized Likert scale responses into 2 categories: responses > 3 were classified as positive (ie, good knowledge, positive attitude, good practices); responses that were ≤ 3 were classified as negative (ie, poor knowledge, negative attitude, poor practices). We created composite ratings (ie, positive or negative) for overall perceived severity, safety, and benefit scores, calculated by summing positive or negative responses, or correct or incorrect responses, to relevant survey questions. Additional details regarding the calculation of scores are available in the associated table.

We performed logistic regression variable selection to identify key factors that were associated with acceptance of influenza vaccines. Variables in the selection process included those related to demographics (eg, age and education), previous pregnancies and pregnancy care (eg, number of pregnancies, hospitalizations, and antenatal visits), knowledge of influenza and/or influenza vaccines, potential barriers (eg, cost and travel time), knowledge gaps, and misinformation (eg, shortage of information and negative rumors about influenza vaccination).

Respondents were also asked how much they would be willing to pay for the seasonal influenza vaccine if it were available for purchase. Responses were recorded in local currency, and converted to US dollars (US$) using the OANDA currency converter [26] as of the last date the survey was administered to approximate the exchange rate at the time of the survey. Willingness to pay data were also presented as a percent of the country's health expenditure per capita (in US$), using a World Bank dataset summarizing data from the WHO Global Health Expenditure database [27]. Data was stratified by World Bank lending group, and whether the country had a seasonal influenza vaccine program for pregnant persons during the study period. These data were analyzed using descriptive statistics and bivariate nonparametric tests (Kruskal-Wallis, Wilcoxon rank sum). Analyses were conducted using SAS 9.4.

Each implementing partner obtained ethical approvals in accordance with local regulations. All participants provided written informed consent.

RESULTS

Study Population

Our analysis included 8556 pregnant persons from 8 countries representing 4 of the 6 WHO regions; 27% (2312) of respondents were from the African region, 27% from the Eastern Mediterranean (2309), 21% from the European (1835), and 25% from the Western Pacific (2100). Sixty percent (5155) of participants were from lower-middle income countries, 23% (1920) from low-income countries, and 17% (1481) from upper-middle income countries.

Five of the 8 countries (63%) had seasonal influenza vaccine programs for pregnant persons at the time of data collection (Table 1), accounting for 69% of respondents (5890); 31% (2666) came from countries with no seasonal influenza vaccine program. All respondents in upper-middle income countries (Albania, Armenia), most (86%) in lower-middle income countries (Cote d’Ivoire, Kyrgyzstan, Lao PDR, Morocco, Tunisia), and no respondents in low-income countries (Uganda) lived in countries where influenza vaccines were available free of charge to pregnant persons at the time of the survey. Twenty-three percent (1943) of respondents in our study were aged between 18 and 24 years, 36% (3062) between 25 and 29, 24% (2009) between 30 and 34, 12% (1024) between 35 and 39, and 4% (352) above the age of 40 years (Table 1).

Influenza and Influenza Vaccine Knowledge, Perceptions, and Information Channels

Most pregnant persons across the countries, both with and without a vaccine program, had heard of seasonal influenza illness (5945, 70%) (Table 2). Over half agreed or strongly agreed that influenza is more dangerous for pregnant persons than nonpregnant persons (5017, 59%); a greater proportion of respondents from countries without influenza vaccination programs (2004, 75%) believed that influenza is more dangerous for pregnant persons than respondents in countries with programs (3013, 52%, P < .001; Figure 1). About one-third of respondents believed that people who have influenza are never sick enough to be hospitalized (2830, 34%) or knew someone who had been severely ill with influenza (3090, 36%). One-quarter (2197, 26%) believed that influenza does not cause a lot of illness in their country, and nearly half were unsure whether influenza vaccine is safe for a newborn baby (3724, 44%), unborn baby (3666, 43%), or pregnant persons (3522, 41%).

Figure 1.

Reported beliefs and perceptions surrounding influenza and influenza vaccines among pregnant persons.

Open in new tabDownload slide

There were similar differences in beliefs about the benefits of vaccination for new or unborn babies: among pregnant persons in countries with an influenza vaccination program 24% and 25% believed their own vaccination could protect their newborn or unborn baby, respectively. In contrast, pregnant persons in countries with no such program strongly believed in the protection for newborn (64%, P < .001) or unborn babies (71%, P < .001) (Figure 1). More than one-third of pregnant persons across all settings were unsure whether their own vaccination helps protect their unborn (3071, 36%) or newborn babies (3265, 38%) (Figure 1). About one-quarter (26%) of pregnant persons believed that influenza vaccine is unsafe for newborn or unborn babies, or for pregnant persons (25%) (Figure 1); however most responded that they did not receive enough information about the safety of influenza vaccines (6307, 74%; Table 3). Although 10% (860) of pregnant persons had heard or read reports in media or social media that made them hesitate to receive influenza vaccines, only 4% (338) knew someone who had experienced a bad reaction to an influenza vaccine (Table 3). Fourteen percent (854) of pregnant persons from countries with influenza program in our study received an influenza vaccine recommendation from any source during their pregnancy, and 13% (738) were discouraged from receiving influenza vaccine (Table 2).

The primary source of information about influenza vaccines for pregnant persons in countries with seasonal influenza vaccine programs was a doctor or other medical professional (1583, 48%; Table 2); most pregnant persons had the greatest trust in information from a doctor or other medical professional (5372, 91%) above all other sources; few trusted information from a family member (457, 8%), media (523, 9%), or friends (69, 1%) (Table 3).

Vaccine-Related Practices and Intentions

Across all countries, most pregnant persons were willing to receive influenza vaccine if it was offered to them for free (6323, 74%) with differences across country groups: among those in countries with a seasonal influenza vaccination program acceptance was lower (3940, 67%) than in countries without influenza vaccination programs (2383, 89%, P < .001; Table 4). The most trusted sources of recommendations were from health care personnel; two-thirds (5751, 67%) of pregnant persons were willing to accept an influenza vaccine recommendation from a doctor, one-quarter from midwives (1950, 23%). and one-fifth from nurses (1495, 18%). Few respondents accepted vaccine recommendations from other sources: 8% (669) would accept a recommendation from a family member and 5% (396) from an international body (ie, WHO). Recommendations from other sources were of minimal importance (Table 4). Most respondents were willing to take time off from work or away from home to receive an influenza vaccine (5949, 70%), and were willing to spend up to 60 minutes (interquartile range [IQR], 15–120) to do so (Table 4). Despite their willingness to accept their recommendation and the existence of a seasonal influenza vaccine program targeted to them, only 11% (659) of respondents recalled receiving an influenza vaccine recommendation from a medical provider during their pregnancy (Table 2).

Predictors of Vaccine Acceptance in Pregnant Persons

Respondents who were willing to take time off from work to receive vaccines were more likely to accept vaccines than pregnant persons who were not, with an adjusted odds ratio (aOR) of 29.0 (95% CI: 22.3–37.7; P < .001) in countries with influenza vaccination programs and an aOR of 9.2 (95% CI: 5.5–15.3; P < .001) in countries without. Pregnant persons who believed that there were better ways to prevent influenza than through vaccination were less likely to accept vaccines than those who did not, with an aOR of 0.7 (95% CI: 0.6–0.9; P = .001) for countries with influenza vaccine programs and 0.6 (95% CI: 0.4–0.9; P = .008) for countries without. Respondents who perceived greater benefits of vaccination (benefit score) were more likely to accept influenza vaccine compared to those with a lower perception of benefits (aOR, 1.2; 95% CI: 1.1–1.3; P < .001) for countries with influenza programs and 1.4 (95% CI: 1.2–1.5; P < .001) for countries without. Respondents who heard or read reports in the media or social media that would make pregnant persons hesitant about vaccination were less likely to accept influenza vaccines than those who had not heard or read such reports (aOR, 0.5; 95% CI: 0.4–0.7; P < .001). Respondents who were willing to pay were 1.9 times more likely to accept a vaccine than those who were unwilling to pay (95% CI: 1.4–2.6; P < .001) in countries with influenza vaccination programs; in countries with no program, those willing to pay were 3.9 times (95% CI: 1.9–7.9; P < .001) more likely to accept a vaccine than those who were unwilling to pay (Table 5).

Willingness to Pay for Vaccines

Respondents were asked how much they would be willing to pay for seasonal influenza vaccine if it was available for purchase. Sixty-one percent of those queried (5221) provided a price (median, US$1.30; IQR, $0.30–$5.20; Table 6). Of those, 3359 (64%) were in countries with influenza vaccination programs for pregnant persons, while 1862 (36%) were in countries without a program. Pregnant persons in upper-middle income countries reported a higher median price (US$9.40; IQR, $4.70–$10.50), compared with lower prices in lower-middle and low income countries (US$1.50; IQR, $0.10–$5.30 and US$1.30; IQR, $0.50–$2.60; P < .001, respectively). The price in low-income countries represented a higher percentage of their country's annual health expenditure per capita relative to the other income groups (P < .001). The amount respondents in countries lacking national influenza vaccination programs were willing to pay represented a higher percentage of their country's health expenditure per capita (median, 4.10%; IQR, 1.60%–8.10%) relative to the price pregnant persons in countries with vaccination programs were willing to pay (median, 1.00%; IQR, 0.20%–3.00%; P < .001; Table 6).

DISCUSSION

We found an overall very high willingness to receive seasonal influenza vaccines among pregnant persons. High acceptance by pregnant persons in countries lacking influenza vaccination programs is perhaps not surprising, as global evidence shows a widespread acceptance among pregnant persons in lower-middle income countries of vaccines for influenza and other vaccines such as for tetanus toxoid and pertussis [28–30].

Our analysis highlights the critical role that health care workers play in patient vaccine uptake. Indeed, health worker recommendation is a significant driver of vaccine acceptance among pregnant persons [15, 16, 31]. Likewise, a recommendation against vaccination by a health care worker is a significant barrier to uptake or intent [32, 33]. A 2020 review by Kilich et al found that pregnant persons who had received a recommendation were 10 to 12 times more likely to receive a pertussis or an influenza vaccine [15], and a 2017 multicountry study found an 8-fold increase in likelihood to accept a vaccine upon receipt of a recommendation [34]. A 2019 Tunisian study found that a health care worker recommendation carried the connotation of vaccine safety, thus facilitating trust in the vaccine [14]. Among pregnant persons in our analysis who were unsure if they would receive influenza vaccine, 80% could not recall receiving a recommendation to be vaccinated from a health care worker (Supplementary Table 1). Nearly half of pregnant persons in our study were unsure if influenza vaccine was safe for infants and themselves (Supplementary Table 2). A 2018 qualitative study in China found that few pregnant persons had received a recommendation; physicians in the study did not recommend vaccines, citing the lack of a national policy to vaccinate pregnant persons [32, 35]. Such hesitancy to make a recommendation often leads to hesitancy on the part of pregnant persons to accept vaccines [36].

Most pregnant persons in our study did not feel they received enough information about the safety of seasonal influenza vaccines, one of the most common barriers to acceptance among pregnant persons observed [16, 37, 38]. Common concerns both in our sample and in the literature included fears about vaccine safety for a newborn or unborn baby, and concerns about the safety for oneself [15, 37, 39]. Health care workers who are aware of the benefits and importance of influenza vaccines for pregnant persons and confident making recommendations [14] may serve to increase uptake in pregnant persons.

High perceived benefits for themselves and/or their babies were predictive of vaccine acceptance in both groups of countries in our study. Similar trends are described in a 2022 review of acceptance in lower-middle income countries [38]. Perceived benefits or lack thereof were paired with awareness of the risks of infection to mother and baby in our survey; pregnant persons with low perceived risk of influenza illness were less likely to accept a vaccine while belief in risks to mother or baby was associated with higher acceptance, consistent with current evidence [38, 40]. While we found generally high awareness of influenza illness among pregnant persons, awareness of the risks associated with influenza was lower, suggesting that raising awareness of personal risk and risk to ones’ newborn or unborn baby is a necessary first step toward increasing uptake among this population [41].

Pregnant persons in countries with no influenza vaccine program were willing to pay more for a vaccine, as a percentage of health expenditure per capita, than those in countries where influenza vaccine is offered for free. Willingness to pay a higher price among pregnant persons in low- and middle-income countries might be influenced by their higher perceived threat of influenza illness, higher perceived benefits of protection, and higher perception of the need to vaccinate. Future assessment of willingness to pay for seasonal influenza vaccine might consider a bidding game question format as it is less sensitive to bias [42]. Our findings suggest that in the absence of a publicly funded influenza vaccination program, a subsidized national program may be an effective strategy to reduce the financial barriers to access.

We are unaware of any equivalent analysis of primary data collected from pregnant persons across such a large group of countries, making our study unique. Our study was subject to several limitations. Fidelity of some questions may have been compromised by minor survey modifications in some countries; we believe that our careful matching of questions and answers across all country datasets was accurate enough to overcome this limitation. Surveys may have been subject to response bias, with respondents providing responses thought to be favorable to interviewers. Our estimates of willingness to seek vaccination based on intent may be higher than actual practice. Our study population may have been subject to selection bias due to data collection in clinical settings; not all pregnant persons seek antenatal care in a clinical setting; however WHO Global Health Observatory [43] data confirms that a mean of 89% of births in our study countries are medically attended. Finally, respondents who had not heard of influenza vaccine may have been ill-equipped to answer questions about their perceived safety or impact of influenza vaccine.

CONCLUSIONS

Our analysis shows that pregnant persons report a strong intent to accept influenza vaccines. Evidence collected since the coronavirus disease 2019 (COVID-19) pandemic variably shows both a decrease [44] and an increase [45, 46] in uptake in maternal vaccinations overall; these findings generally represent data from high-income settings. While the overall acceptance of vaccines may have diminished, this is unlikely to have made a significant change in the need for accessible vaccines for pregnant persons throughout low- and middle-income countries.

We propose several strategies to reduce vaccine hesitancy and remove modifiable barriers to vaccination. We propose actions to reduce health care workers’ hesitancy to recommend vaccines to pregnant persons, including updating vaccine package inserts with current clinical care recommendations [47], and establishing and updating national influenza vaccine policies to include pregnant persons. Importantly, increasing health care worker knowledge and awareness about the risks of influenza infection and safety of influenza vaccine is critical to build confidence in making vaccine recommendations for pregnant persons.

Removing logistical barriers to vaccination, both for pregnant persons and health care workers, may facilitate increased uptake. Offering in-office vaccination during antenatal care visits may reduce time and travel-related barriers, thus increasing uptake among pregnant persons who are unwilling or unable to dedicate time for vaccination [48]. Straightforward service delivery in antenatal care clinics, while ensuring well-coordinated efforts between health services (eg, antenatal care and national immunization programs) may establish influenza vaccination as a standard of care [49].

The pronounced differences we observed in beliefs about protection conferred to newborn and unborn babies via maternal vaccination between pregnant persons in countries with and without influenza vaccine programs are notable. While our survey was not able to explain this difference, our observation does suggest opportunities for greater dialogue between health care workers and their patients during antenatal care visits. Immunization programs in countries with influenza vaccine may consider training antenatal care providers in communicating the benefits of vaccination for pregnant persons and their babies. This may be accomplished via motivational interviewing or similar techniques [50].

After health care workers’ recommendations, the second most important barrier to coverage among pregnant persons in low and lower-middle income countries, where acceptance of vaccines is very high, is access to vaccines. Investments in supporting the review of evidence, and policy development and implementation in countries without an influenza vaccine program for pregnant persons are critical to increasing protection of pregnant persons and their babies.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Acknowledgments. We thank Susan Y. Chu and Saad Omer for their assistance with protocol and survey development; Sarah E. Pallas for guidance on willingness to pay analysis and Michael Jhung for their scientific review; Nissaf Ben Alaya, Afif Binsalah, and Leila Bouabid for their leadership for the Tunisia team; and Michael Daugherty and Natalie Olson for their assistance with data cleaning and merging.

Author contributions. M. M. contributed conceptualization, methodology, analysis, writing, review, and editing. T. S. Y. contributed data curation, formal analysis, writing, review, and editing. C. G. performed data curation. P. M. performed formal analysis, review, and editing. L. M. D. performed formal analysis, review, and editing. M. S. E. contributed conceptualization and methodology. K. E. L. contributed methodology, review, and editing. L. S. I., S. B., J. B., S. D., G. S., I. C., L. A., D. C., J. J. L., A. D., A. N’G., V. K., C. T., C. P., and D. O. performed data collection. P. L. performed data collection, review, and editing. E. A.-B. performed review and editing. J. S. B. contributed conceptualization, methodology, review, and editing.

Disclaimer. The conclusions, findings, and opinions expressed by authors contributing to this article do not necessarily reflect the official position of the US Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, World Health Organization, or the authors’ affiliated institutions. Use of trade names is for identification only and does not imply endorsement by the Public Health Service or by the US Department of Health and Human Services.

Financial support. This work was supported by the US Centers for Disease Control and Prevention and Control cooperative agreements (grant numbers including CDC-RFA-IP16-1604 “Introducing of Expanding the Use of Seasonal Influenza Vaccines in Public Health Programs Outside of the United States,” and CDC-RFA-IP16-1607 “Expansion of seasonal influenza vaccination programs in low- and middle-income countries” to contributing countries).

References

Author notes