Safety, equity and monitoring: a review of the gaps in maternal vaccination strategies for Aboriginal and Torres Strait Islander women

ABSTRACT

Influenza and pertussis infections are disproportionately higher among Aboriginal and Torres Strait Islander women and their infants compared to other Australians. These infections are potentially preventable through vaccination in pregnancy; however, there is a lack of systematic monitoring and therefore knowledge of vaccine uptake, safety and effectiveness in Australia, and specifically among Aboriginal and Torres Strait Islander women. The limited data available suggest there is a lower uptake of maternal vaccination among Aboriginal and Torres Strait Islander women compared to non-Aboriginal and Torres Strait Islander women, and this review seeks to explore potential reasons and the knowledge gaps in this regard. Other key gaps include the equitable access to quality antenatal care for Aboriginal and Torres Strait Islander women; and pregnancy loss <20 weeks gestation. Furthermore, our review highlights the importance of addressing these gaps in maternal vaccination strategies in partnership with Aboriginal and Torres Strait Islander peoples.

KEYWORDS:

• Immunization
• vaccination
• pregnancy
• gaps
• uptake
• Aboriginal
• influenza
• pertussis
• access
• equity

Introduction

Globally, tetanus vaccination in pregnancy lead to a 94% reduction in death from neonatal tetanus between 1988 and 2013.¹ The role of vaccines in pregnancy to reduce infectious diseases has since gained worldwide momentum,² with inactivated influenza vaccine (IIV) being recommended in pregnancy from as early as 1960,^3–5 and pertussis vaccination in pregnancy joining IIV from 2012 in the UK, and other countries subsequently.^6–8 Clinical trials are progressing for respiratory syncytial virus,^9,10 group B streptococcus,¹¹ and pneumococcal vaccine use in pregnancy.^12,13 Recommendations for IIV and pertussis in pregnancy are based on data from a limited number of small, but well-conducted randomized controlled trials (RCTs) that have been complimented by data from large observational studies.^14–21 Data for vaccine safety and vaccine effectiveness (VE) have been hampered by difficulties in enrolling pregnant women in research studies, or more typically, pregnant women have been excluded from industry-sponsored vaccine trials.^22,23

Influenza and pertussis vaccines in pregnancy offer the best opportunity to reduce the burden of diseases caused by these pathogens during two very high-risk periods – pregnancy (when influenza can be particularly severe) and during the first few months of life (when both diseases can be severe).^14,24 Aboriginal and Torres Strait Islander women and their infants (hereafter respectfully referred to as ‘Aboriginal’), like many First Nations peoples, are a key population group with disproportionately higher rates of adverse birth outcomes, and a disproportionately higher burden of respiratory diseases in pregnancy and early infancy compared to non-Aboriginal pregnant women and infants.²⁵ In the Northern Territory (NT) of Australia, average annual infant influenza hospitalization rates per 100,000 NT population have been reported to be over 42 times higher in Aboriginal infants (254 vs 6) compared to non-Aboriginal infants; and pertussis hospitalizations per 100,000 NT population have been reported as 7.1 times higher in Aboriginal infants (142.5 vs 20.2) compared to non-Aboriginal infants.²⁵ Within Australia, there are very few data describing the uptake, safety and VE of IIV and pertussis vaccines among Aboriginal women during pregnancy, even though the disease burden is highest among this population,^26,27 and surveys have indicated a high willingness to be vaccinated in pregnancy if offered.^28,29

Comprehensive reviews and landscape analyses have described many of the global knowledge gaps surrounding vaccination in pregnancy.^30,31 The reasons for sub-optimal uptake (currently IIV <30–50%; pertussis <68%);^32–36 are poorly characterized, and there are gaps in our understanding of the: immunobiology (immune responses in pregnancy, persistence of antibody in infants, blunting of infant immune responses);³⁷ safety of vaccination early in pregnancy and subsequent pregnancy loss;^30,31 safety of repeated vaccine doses in subsequent pregnancies;^30,31 non-specific effects on neonatal health;^30,31 optimal timing for maximum clinical protection;^30,31 and uptake, safety and VE of maternal vaccination in local First Nations and minority populations.^30,38 What is also evident, is that health-care provider recommendation is a key driver of a pregnant woman’s decision to get vaccinated,²⁸ however remaining gaps in the safety evidence may be contributing to health-care provider hesitancy, and more research is required to provide reassurance.

This review highlights several important issues potentially hindering the successful implementation of current maternal vaccination strategies among Aboriginal women of Australia; equitable access to quality antenatal care; safety of IIV early in pregnancy and pregnancy loss <20 weeks gestation; and monitoring and reporting of the uptake, safety and VE of IIV and pertussis vaccines in pregnancy. We focus on issues likely to affect provider and consumer acceptance of current and future maternal vaccination programs.

Equitable access to quality antenatal care for pregnant aboriginal women

Racism and discrimination adversely affect the health and health-seeking behaviors of minority and First Nations groups.³⁹ Although Australia is considered a high-income country,⁴⁰ maternal and infant mortality rates,^41,42 and respiratory illnesses in Aboriginal women and infants remain more than double than that of non-Aboriginal mothers and infants,^25,27 and this is the case in other First Nations populations in countries and territories around the world.^43–45 In Australia, it has been suggested that the lack of culturally appropriate, safe and equitable quality antenatal services contributes to this disparity in disease burden,^46–48 and current services lack engagement with Aboriginal voices, therefore, cultural factors cannot have been given due consideration.^48–50 For example, younger maternal age and ethnicity have been shown to be predictors of lower pertussis vaccination uptake in pregnancy worldwide,^35,51with non-English speaking women less likely to receive pertussis vaccination in pregnancy compared to English-speaking pregnant women.⁵² Aboriginal women birthing their first child are younger than non-Aboriginal women,⁵³ and English may not be the first spoken language for Aboriginal women.⁵³ Without cultural understanding, issues related to the access of equitable, affordable quality antenatal care are all likely factors affecting the uptake of IIV and pertussis vaccination in pregnancy.

Building and strengthening an Aboriginal health workforce that delivers culturally competent and responsive health care, that is holistic, culturally inclusive, and action-oriented and involves respectful and meaningful collaborative partnerships will facilitate access to and utilization of appropriate and acceptable antenatal care.^54–56 The acceptance of, and access to maternal vaccination programs could in turn, markedly reduce the incidence of influenza, pertussis and other respiratory illnesses in the Australian Aboriginal populations.

Safety of IIV early in pregnancy and pregnancy loss <20 weeks gestation

Timing of vaccination in pregnancy is important to ensure the highest benefit for pregnant women and their infants.^57,58 Inactivated influenza vaccine is recommended at any stage of pregnancy,⁵⁹ and the emerging safety data by trimester are reassuring with respect to no increased risk of preterm birth, low birthweight or small for gestational age infants regardless of the trimester of pregnancy IIV was given.^5,60–64 Research on the safety of IIV exposure in the first trimester and the subsequent risk of pregnancy loss prior to 20 weeks gestation (<20) however remains limited,^65,66 and this gap in the evidence has still not been adequately addressed despite being identified by the Global Advisory Committee on Vaccine Safety in 2014.³⁸ Further, adverse outcomes with respect to timing of vaccination in pregnancy, such as pregnancy loss <20 weeks gestation has been acknowledged as challenging,³⁸ and prone to bias if not conducted using time-varying analytic methodologies.⁶⁷

Compared to non-Aboriginal women, there is a higher prevalence of miscarriage, late spontaneous abortions and stillbirths among Aboriginal women.^68,69 Whilst the risk factors for pregnancy loss <20 weeks in Aboriginal women are multifactorial, fever and infection in any pregnancy leads to an increased risk of miscarriage,⁷⁰ with influenza infection recognized among these.⁷⁰ Given that Aboriginal women experience higher rates and complications from respiratory infections in pregnancy compared to non-Aboriginal pregnant women,⁷¹ it is essential to investigate the potential risk/benefits of IIV in early pregnancy against pregnancy loss among this population.

In 2015, two systematic reviews of Northern Hemisphere studies found no significant differences in the risk of spontaneous abortion following H1N1pdm09 influenza vaccine in pregnancy.^72,73 However, most of these studies had a high or unclear risk of selection or attrition bias, and several failed to use time-varying analyses.⁶⁷ One US study found an increased risk in spontaneous abortion in the 28 d after receiving IIV in pregnancy among women who had also received an IIV in the previous influenza season (aOR 7.7 [95% CI 2.2–27.3]), compared to women who were vaccinated in pregnancy but did not receive an IIV in the previous influenza season.⁷⁴ As there was no obvious biological reason why this may have occurred, the results are concerning and much larger sample sizes using time-varying methodological approaches are suggested to further investigate these findings.⁷⁵ Furthermore, to make a valid inference in observational studies such as this, it is important to accurately estimate the background rates of pregnancy loss <20 weeks gestation and to capture data on potential confounding factors such as maternal age, Indigenous status, maternal comorbidities, lifestyle and pregnancy risk factors for analysis in a multivariable model.

In summary, most studies have shown IIV in pregnancy to be safe, but further research is required to investigate the safety of IIV exposure early in the first trimester and subsequent pregnancy loss <20 weeks gestation, using time-varying methodologies and accurately defined background rates of miscarriage/spontaneous abortions in local populations. Robust evidence about the safety of early maternal vaccination is needed to reassure women and health-care providers and ultimately improve uptake of this important preventative strategy, particularly as more antenatal vaccines are introduced.

Uptake, safety and effectiveness of IIV and pertussis vaccines in pregnant aboriginal women

Inactivated influenza vaccines

There are very few studies worldwide that describe the uptake, safety, VE and/or viability of maternal influenza vaccination programs in First Nations women and infants. An Australian study for a group of remote-living Aboriginal women from the NT (N = 697) found IIV uptake to be poor during pregnancy from 2003 to 2011, (n = 20/697, <3%).⁷⁶ Although being one of the most comprehensive studies conducted among Aboriginal women and infants in Australia, due to the low proportion of IIV uptake in pregnancy, this study was without the power to assess safety and VE. Other small Australian studies that have examined IIV uptake in pregnant Aboriginal women have also lacked power.^29,77–80 Uptake of IIV in pregnancy was low in all of these studies: 35% (n = 100),⁷⁷ 17% (n = 53),²⁹ 15% (n = 214),⁷⁸ 39% (n = 1,311),⁷⁹ and 49% (n = 23/47).⁸⁰ Further, limitations included the failure to account for the remoteness of living or detailed regional differences,^77,79 noting that the remote living Aboriginal population is where the disease burden is highest.²⁵ Selection bias likely affected the generalisability of the results in one Western Australian (WA) study,⁷⁷ whilst misclassification bias was probable in the exploratory study from Queensland that relied on the self-report of IIV uptake in pregnancy alone.²⁹ A Central Australian study had a high degree of misclassification bias,⁸⁰ where women were considered to be vaccinated ‘antenatally’ even if they had received IIV prior to the start of the pregnancy, ultimately resulting in an overestimation of IIV uptake during pregnancy (70%, n = 32/46).

Pertussis vaccines

Australian studies examining pertussis vaccination coverage that involved pregnant Aboriginal women have either had; small sample sizes without the power to assess the safety or VE of pertussis in pregnancy; limited ability to describe vaccine coverage by Indigenous status; or had a high risk of bias.^77,79,80 A small WA survey of Aboriginal women (n = 97) was able to verify that 45% (n = 44/97) received a pertussis vaccine during pregnancy,⁷⁷ however, the small sample size limited interpretation and generalisability to the wider Aboriginal population. A small Central Australian study highlighted the very poor coverage of pertussis vaccination among pregnant Aboriginal women in a high-risk, largely remote living population; the self-report of pertussis vaccination uptake was 28% from 2016 to 2017 (n = 13/47), although 50% (n = 23/46) according to the state-based immunization register.⁸⁰ A 2015 study from the NT found low proportions of women who received a pertussis vaccine in pregnancy for both Aboriginal and non-Aboriginal women (23.5 vs 21.6%, respectively).⁷⁹ There is, however, a high chance of bias present in this study with data on vaccination in pregnancy unknown for ~20% of non-Aboriginal women who gave birth in the private hospital. As such, this likely underestimates the proportion of non-Aboriginal women who received a pertussis vaccination in pregnancy. Further, these data did not encompass remote living pregnant women. There are currently no published data comparing the uptake of pertussis vaccination in pregnancy between remote living pregnant women and other pregnant women, despite the clear disproportionate burden of illness from respiratory infections in remote living Aboriginal infants.²⁵ It will be important to establish reliable coverage of pertussis vaccination in pregnant Aboriginal women following an operational period of the program post its introduction in 2015.⁶

Vaccine coverage in pregnancy must be accurately estimated in order to carefully evaluate the safety and VE of maternal vaccination programs in local populations, particularly for Aboriginal mother-infant pairs, where the risk may be differentially higher at different stages of pregnancy.²⁵ The results from two large Australian cohort studies are in progress and will provide valuable uptake, safety and VE data for Aboriginal women and their infants.^81,82

Monitoring and surveillance of seasonal IIV and pertussis vaccination in pregnancy

Although the Australian Immunisation Register was established to record all vaccine encounters over the life course, pregnancy status at the time of vaccination is not recorded.⁸³ As such, there is no systematic mechanism available to describe the national uptake, safety and VE of IIV and pertussis vaccinations in pregnancy, even though they are recommended and funded. Adverse safety signals specifically related to maternal vaccinations, and VE cannot, therefore, be highlighted without using alternative data sources and data linkage projects. The lack of systematic monitoring and reporting of vaccination coverage in pregnancy prevents any dissemination of data regarding rates of IIV and pertussis uptake in pregnancy. This further hinders the evaluation of maternal vaccination programs, and their subsequent impact on influenza and pertussis infections in Aboriginal pregnant women and their infants.

The formulation of IIV typically changes from year to year, with annual estimates of effectiveness being dependent on the match between the IIV formulation and the circulating strain/s of influenza virus.^84,85 Ongoing monitoring of IIV uptake, safety and VE in pregnancy therefore must occur annually to account for the potential variation in IIV formulations and their corresponding match to circulating strains. It is not sufficient to rely solely on the evidence from the broader adult population, the elderly or infants. Past experience with licensed childhood influenza vaccines in Australia and the United States highlighted safety signals that resulted in the worldwide withdrawal of one manufacturer’s product over the 2010/2011 influenza season.^86,87

Conclusions

Gaps in maternal vaccination processes persist due to a lack of systematic monitoring and therefore program evaluation. Identifying and addressing procedural gaps in local populations is critical to influence acceptance and uptake. Program evaluation must be based on the local epidemiology of adverse pregnancy and birth outcomes and respiratory diseases, and knowledge of the formulations of seasonal IIV used in pregnancy. It is particularly important to include representative data for pregnant women from diverse ethnic backgrounds and First Nations peoples, for whom the risk of disease and of even poorer health outcomes can be much higher when exposed to vaccine-preventable diseases. To this end, increasing the uptake of IIV and pertussis vaccination in Australian Aboriginal pregnant women must begin with involving Aboriginal peoples in the development, implementation, and dissemination of health programs and research.^48,50 Aboriginal peoples therefore must be placed at the center of the decision-making process, leading the way toward developing and implementing culturally appropriate vaccine strategies in pregnancy. Establishing a process for engagement and meaningful and respectful communication within a culturally appropriate governance structure that values and privileges Aboriginal peoples’ voices, histories, experiences and perspectives are critical.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Acknowledgments

We acknowledge the traditional owners and custodians of the land and waters on which we live and work as the First Peoples of Australia.

Co-authors Kristy Crooks and Amy Creighton pay respect to Elders, families and friends for their continued guidance, support, encouragement and strength to advocate for cultural inclusion and privileging First Nations voices in health research.

Additional information

Funding

LMc was supported by an [Australian Postgraduate Award] scholarship provided by Charles Darwin University and [Enhanced Living scholarship] provided by Menzies School of Health Research as part of the Doctor of Philosophy higher research degree. KC was supported by the [Australian Partnership for Preparedness Research on Infectious Disease Emergencies (APPRISE)] Centre of Research Excellence as part of the Doctor of Philosophy higher research degree. MJB is supported by a [Career Development Fellowship] sponsored by the NHMRC funded ‘HotNorth’ program - Improving Health Outcomes in the Tropical North: A multidisciplinary collaboration’: [GNT1131932].