Maternal
Immunization
and Antenatal Care Situation Analysis
Report of the MIACSA
project 2016–2019
Maternal Immunization and Antenatal Care Situation Analysis
REPORT OF THE MIACSA PROJECT
2016–2019
Maternal immunization and antenatal care service delivery situation analysis: report of the MIACSA project, 2016–2019
ISBN 978-92-4-000401-6 (electronic version) ISBN 978-92-4-000402-3 (print version)
© World Health Organization 2020
Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo).
Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”.
Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization.
Suggested citation. Maternal immunization and antenatal care service delivery situation analysis: report of the MIACSA project, 2016-2019. Geneva: World Health Organization; 2020. Licence: CC BY-NC-SA 3.0 IGO.
Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris.
Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing.
Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user.
General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement.
The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.
All reasonable precautions have been taken by WHO to verify the information contained in this publication.
However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use.
This publication contains the collective views of an international group of experts and does not necessarily represent the decisions or the policies of WHO.
Contents
Abbreviations iv Acknowledgements v
Executive summary 1
Introduction 1
Objectives and aims 1
Methods 1
Findings 2
1. Introduction 11
1.1 Background 11
1.2 Conceptual framework 13
2. Study methodology 17
2.1 Study design and sources of data 18
2.2 Key definitions and indicators 20
2.3 Statistical analysis 21
2.4 Study limitations 22
3. Results 25
3.1 Maternal immunization performance indicators 25
3.2 Leadership and governance 29
3.3 Antenatal care service delivery 42
3.4 Human resources 46
3.5 Health financing 51
3.6 Information systems 56
3.7 Logistics, infrastructure 67
3.8 Demand creation 72
3.9 Vaccine hesitancy/confidence 73
4. Associations between health system factors and
maternal tetanus immunization 77
5. Discussion, conclusion, recommendations 83
References 91
Annexes 95
Abbreviations
AEFI adverse event following immunization ANC antenatal care
ANC1 proportion of pregnant women who received one ANC contact during their last pregnancy
ANC4+ proportion of pregnant women who received four or more ANC contacts during their last pregnancy
BCG Bacillus Calmette-Guérin CHW community health worker CI confidence interval
DTP diphtheria, pertussis and tetanus
DTP1 first dose of diphtheria, tetanus and pertussis vaccine DTP3 third dose of diphtheria, tetanus and pertussis vaccine EPI Expanded Programme on Immunization
EVM effective vaccine management
HMIS health monitoring and information system LCA latent class analysis
LMICs low- and middle-income countries
MIACSA Maternal Immunization and Antenatal Care Situation Analysis MNCAH maternal, newborn, child and adolescent health
NITAG national immunization technical advisory group NRA National Regulatory Authority
PAB protection at birth TT Tetanus toxoid
TT2+ at least two doses of tetanus toxoid vaccine during pregnancy UNDP United Nations Development Programme
UNFPA United Nations Population Fund UNICEF United Nations Children’s Fund WHO World Health Organization
WUENIC WHO/UNICEF estimates of national immunization coverage
Acknowledgements
This report was drafted by Sonja Merten under supervision of the World Health Organization (WHO) Departments of Immunization, Vaccines and Biologicals, and Maternal, Newborn, Child and Adolescent Health.
It was informed and reviewed by the contributions of the following individuals:
■
■ Expert advisory panel: Mercy Ahun, Martina Baye, Veena Dhawan, Pradeep Haldar, Michelle Giles (co-chair), Matthews Mathai, Flor Muñoz (chair)
■
■ Consultants: Steve Black, Carsten Mantel, Elizabeth Mason, Jayani Pathirana, Sara Rendelll
■
■ Additional national and international experts: Xavier Bosch-Capblanch, Mari Dumbaugh, Langelihle Mlotshwa, Christian Schindler, Ahmadu Yakubu
■
■ WHO: Theresa Diaz, Laure Dumolard, Joachim Hombach, Elisabeth Katwan, Philipp Lambach, Laura Nic Lochlainn, Allisyn Moran, Marc Perut, Nathalie Roos, Dilip Thandassery, Özge Tunçalp, Emily Wootton, Nasir Yusuf.
The World Health Organization thanks the Bill and Melinda Gates Foundation for providing financial support for the MIACSA project under the BMGF grant agreement OPP1156378.
Executive summary
Introduction
Vaccine-preventable diseases are among the main causes of global child morbidity and mortality, particularly in low- and middle-income countries (LMICs). Maternal vaccines given to pregnant women in the second or third trimester have emerged as a promising way to address vaccine-pre- ventable diseases, providing protection to the newborn during the most vulnerable period in life, through the trans-placental transfer of maternal antibodies. Antenatal care (ANC) is generally accepted as the natural entry point for interventions during pregnancy, including maternal immu- nization. But despite progress made in ANC use, the World Health Organization (WHO) estimates that between 2010 and 2016, only 61.8% of pregnant women attended at least four ANC visits, constraining the time points when vaccination might occur. This calls for better understanding of the optimal ways to deliver vaccines to pregnant women and the value of using ANC services as a delivery platform.
A multi-method study, the Maternal Immunization and Antenatal Care Situation Analysis (MIACSA), was conducted between November 2016 and June 2019 (32 months) aiming to explore current and future preparedness to introduce and implement new maternal vaccines. To learn from expe- rience and to prepare for the introduction of new maternal vaccines, the MIACSA study set out to investigate maternal tetanus immunization programmes, which have been in place for the past three decades and are the most widely implemented vaccination programmes in pregnancy.
The project aimed to develop a typology of health systems in terms of how they are delivering vac- cines to pregnant women, to understand what system attributes correlate with high performance in the delivery of vaccines to pregnant women, and to assess the path forward for introducing additional vaccines for pregnant women.
Objectives and aims
The MIACSA project set out to investigate ongoing maternal tetanus immunization programmes in LMICs. More specifically the MIACSA project aims to:
■
■ improve the understanding of the challenges and successes of ANC and EPI services in implementing maternal immunization with tetanus toxoid and other maternal vaccines.
■
■ inform the sustainability strategy of the maternal and neonatal tetanus elimination initiative.
■
■ Contribute to the development of approaches towards the delivery of new maternal vaccines.
Methods
The MIACSA project had four components, each with a distinct methodology:
1. a review of published and grey literature of indicators relevant for maternal vaccination from global databases for 137 LMICs;
2. an online questionnaire sent to all of the 116 non-European LMICs, yielding validated responses for 95;
3. a semi-structured telephone interview with national representatives from 26 selected LMICs; and
4. visits to 10 LMICs, for key informant interviews and health-care facility observations, of high performing countries and potential early adopters were preferentially visited.
To measure the performance of maternal tetanus immunization programmes we assessed the proportion of neonates protected at birth (PAB) against neonatal tetanus. This programme indicator is estimated by WHO and the United Nations Children’s Fund (UNICEF). Unlike coverage of at least two doses of tetanus toxoid (TT) vaccine during pregnancy (TT2+), PAB accounts for women who have previously received protective doses, women who received one dose without any documentation of previous doses received, and women who received doses in tetanus toxoid (or tetanus and diphtheria) supplemental immunization activities when pregnant or not. In addition, girls who have received diphtheria, tetanus and pertussis (DTP) vaccine in their childhood and are entering childbearing age may be protected with tetanus toxoid booster doses. Unlike the new maternal vaccines, tetanus does not have to be given in pregnancy for its intended effect.
Based on the review of data, from 137 LMICs,. Countries were stratified into low- and high-PAB coverage countries (below 90% and at least 90%, respectively) using the median PAB value as the defining variable.
Countries were further differentiated according to their potential to protect mothers and young infants from vaccine-preventable diseases. For this purpose, PAB was combined with indicators of the performance of immunization and maternal, newborn, child and adolescent health (MNCAH) systems, TT2+, third dose of DTP (DTP3), at least one ANC visit (ANC1), at least four ANC visits (ANC4+), neonatal mortality rate, and maternal mortality ratio. Latent class analysis was used to create four country groups having either high (Group 4), moderate (Group 3), limited (Group 2), or very limited (Group 1) potential to protect mothers and young infants from vaccine-preventable diseases. The desk review data from 2017 was used except for ANC coverage, where the latest available data was used.
Findings
Policies and guidelines
Policies and guidelines underpin the quality and scope of health services. In the MIACSA study, the availability of policies influencing maternal immunization among LMICs was assessed. Overall, 92.6% of the 95 countries participating in the online questionnaire had either a written policy or a guideline on maternal immunization and, with the exception of one country, all had a policy on the recommended number of ANC visits. Countries with a target for maternal tetanus immunization coverage of at least 90%, and countries recommending more than four ANC visits were more likely to have high PAB coverage (at least 90%) and to belong to the group of countries with a high potential to protect women and infants from vaccine-preventable diseases.
National immunization technical advisory groups (NITAGs) are multidisciplinary groups of national experts responsible for providing independent, evidence-informed advice to policy-makers and programme managers on policy issues related to immunization and vaccines. The roles and compositions of NITAGs have changed over the last decades; two thirds (67.2%) of the LMICs had a NITAG in place, according to the desk review among 137 countries; yet the proportion of countries without a NITAG was not negligible, and representation from newborn health experts was not consistent. This situation does not necessarily mean a lack of leadership for vaccination programmes – since there are reasonable alternatives to NITAGs – but it was difficult to assess to what extent NITAGs were really effective in driving the immunization of pregnant women in the countries. Nonetheless, having a NITAG in place was more common in countries with high PAB coverage, though without reaching statistical significance (P-value 0.052), and was less common in Group 3 countries (with moderate potential to protect women and infants from vaccine- preventable diseases) as compared to the other groups.
Program responsibilities for the maternal immunization program at the central level (procurement, planning and management, distribution and training and supervision), and the administration of vaccines, were organized differently in the countries included in the online survey, with 95 participating countries. The procurement and distribution of maternal tetanus vaccines were mainly in the hands of the Expanded Programme on Immunization (EPI), while planning and training
were more often organized jointly by EPI and MNCAH programmes. For maternal vaccines other than tetanus (i.e. mainly influenza, but also pertussis vaccine, and in a few cases hepatitis B and diphtheria vaccines), the responsibilities were not necessarily the same as for maternal tetanus immunization and varied from country to country For example, different channels of procurement and distribution may be used. In several South American countries and in Thailand, maternal tetanus immunization management, planning, training and supervision were co-organized between the EPI and MNCAH programmes, while the responsibility for influenza vaccination during vaccination was solely with the EPI programme.
Just over half (54%) provided tetanus vaccination to more than 50% of pregnant women during ANC services. This was less likely if the EPI programme was solely responsible for the management, planning, training and supervision of maternal tetanus immunization. Countries with limited ANC capacity were more likely to provide tetanus vaccination in EPI or immunization clinics, which may reflect a more effective approach for achieving high maternal tetanus immunization coverage in these countries. A minority of countries (11%) provided maternal tetanus immunization mainly through outreach or campaigns. Other factors, such as the role of non-governmental organizations or the private sector, and the level of decentralization of the health system, influenced the way services were organized and demonstrating considerable heterogeneity in the way maternal tetanus immunization was organized among different countries.
The level of interaction between EPI and ANC services increased from the national to the health facility level. At the most local level, staff members were more likely to have responsibilities in both the EPI and MNCAH programmes than staff at less local levels; often one person provided both ANC and EPI services. This appeared to be independent of the level of integration of EPI and MNCAH programmes at the national level.
There was, however, no clear link between different service delivery strategies and PAB coverage.
Only at the level of vaccine administration did cooperation between EPI and ANC correlate with PAB: in countries where it was more common for both the ANC visit and the vaccination to take place in the same facility on the same day (one-stop service), PAB coverage was higher.
ANC services are the most important platform for maternal tetanus immunization administration and a possible platform for new maternal vaccines. For this reason, the functionality and suitability of ANC services for maternal immunization were assessed for the use of ANC services, the ANC service package, ANC outreach activities and the referral system.
Outreach activities can improve access to ANC services for hard-to-reach populations. These were more often organized in countries with the support of Gavi, the Vaccine Alliance (Group 2 countries), but outreach was not correlated with PAB coverage. Distance to the health facility and transport problems were common reasons for organizing outreach activities.
A well functioning ANC referral system enables the identification and appropriate clinical manage- ment of any complications that may arise during pregnancy, including adverse events following immunization (AEFIs). Most countries had guidelines for referral, and health-care providers could count on clinical guidance by phone or radio from the higher-level facility. Feedback was not always given to the referring facility, however. An important challenge to the referral system was transport by an ambulance, which was often not available, leaving patients to rely on private transport, and leading to inequity in access to higher-level services.
Having policies and management processes in place demands the efficient involvement of staff at each step of the health-care delivery processes, including the effective delivery of vaccines to pregnant women. The professional training profiles of the staff delivering vaccinations were consistent with the human resources mix at the primary health-care level and with the fact that vaccines were often delivered in the ANC visits. Overall, registered nurses or midwives were the ones most often providing vaccination to pregnant women in the health facilities of the countries visited. Registered nurses or midwives played an important role at the primary care level especially, while at the secondary and tertiary levels, clinical officers and doctors, together with enrolled
nurses, were mostly responsible for vaccination (enrolled nurses practice under the direction and delegation of a registered nurse or nurse practitioner to deliver nursing care). From the interviews with facility managers and health workers during the 10 country visits (95 facilities), vaccination by registered nurses or midwives was more common in the health facilities of Group 4 countries.
Key informants interviewed by telephone were more likely to report staff capacity constraints as an important factor if they were from a country with low PAB or when they were a staff member of a primary care level facility, but this did not reach statistical significance.
Funding sources for maternal vaccination included domestic and external funds, out-of-pocket payments, and, more rarely, health insurance. The country groups showed a distinct pattern of funding with a linear increase in domestic, and a decrease in external, funding from Groups 1 to 4. All Group 4 countries relied solely on domestic funding, compared with only 21 % among Group 1 countries.. Greater government funding share for immunization programmes was associated with higher PAB coverage. External financing for maternal tetanus immunization by the United Nations Children’s Fund (UNICEF), Gavi, WHO and other donors was reported by 19/24 countries responding to the telephone interview. Countries with low PAB had a greater probability of receiving Gavi funding. ANC programmes were less likely to receive external funding; 12/24 (50%) of the countries mentioned external funding of ANC programs, and 19/24 (79%) received support for activities related to ANC services, such as training programs. External funding for ANC came mainly from the United Nations Children’s Fund (UNICEF) and United Nations Population Fund (UNFPA).
Maternal tetanus immunization funding schemes were quite diverse and complex when taking into account the scope of funding agencies, support modalities and the funding by the governmental sector. The share of insurance in the financing of ANC services was small (below 10%), and one in five country programs relied on out-of-pocket payments. User fees can decrease the uptake of services, as demonstrated in the literature and confirmed by this study, where, in total, one in four countries did not have a policy exempting pregnant women from paying for ANC services or tetanus vaccination – and this was associated with lower PAB coverage.
Information systems on current and past vaccinations, and on intended and unintended health outcomes, enables the planning of vaccination programmes and the identification of coverage gaps, as well as the assessment of the effectiveness of a particular vaccine and the early detection and documentation of side-effects.
WHO recommends the use of home-based records for child and maternal health. A personal ANC record and/or vaccination card held by the pregnant woman was a component of the programme in most of the countries participating in the electronic survey, usually with an associated clinic- held record, which was electronic in some of the countries. The interviews with health facility managers confirmed that in four out of five health facilities visited, women retained their ANC records, a practice that was also more common in countries with higher PAB coverage.
Electronic records for vaccinations and/or ANC were used in 43 % (40 out of 93) of country programs. Electronic ANC records were more common in Group 4 countries (40%). Health facility managers interviewed during the country visits perceived electronic systems as decreasing the workload, because immunization details no longer have to be registered in several places simultaneously. Electronic databases can be searched if home-based records are lost, which was reported to be a common problem.
If pregnant women missed their appointments, phone contacts and contacts through community health workers (CHWs) were the most frequently mentioned follow-up approaches. CHW contacts were more common in countries with PAB of at least 90%; written reminders and phone reminders for ANC attendance and compliance were more common among Group 4 countries. This finding is consistent with the existing evidence on reminders to take up public health interventions.
Making information about maternal immunization available on the child’s record is a priority for the monitoring and evaluation of a maternal immunization programme. Three out of four health facility managers interviewed confirmed that maternal immunization details could be linked with the child’s record, through joint mother-child booklets, mother-held vaccination booklets or though the linkage of the ANC and/or the childbirth registry with the child registry.
Health management and information systems (HMISs) enable the use of national and subnational data for the planning of vaccination programmes. HMIS differed in content and use with regard to indicators and reporting by the private sector. In one third of the countries, only government providers reported to the HMIS, and in two thirds, both public and private providers provided information about tetanus vaccination. Indicators used by countries to monitor maternal tetanus immunization included TT2+ (89.4%) and PAB (26.6%). Besides the monitoring of maternal tetanus immunization, surveillance of birth doses of Bacillus Calmette-Guérin (BCG), hepatitis B and oral polio vaccines also varied between the countries. Countries that introduced and recorded hepatitis B birth doses had higher PAB coverage and usually belonged to Group 3 or Group 4, while countries recording birth doses of oral polio vaccine were likely to have lower PAB coverage and to belong to Group 2. However, multiple deficiencies were noted in the completeness of the data.
Even though there is no requirement to vaccinate against tetanus in a particular trimester, some current and future maternal vaccines may have to be administered in a specific time window during pregnancy. Consolidated indicators and a consistent documentation of the timing of ANC visits over the course of pregnancy were not found. It is thus a priority to strengthen the reporting system and to collect information on the timing of ANC visits in the current ANC delivery system, and information on maternal tetanus immunization administration by trimester.
Disease surveillance was mainly passive. Active surveillance was reported by one in two countries, while sentinel or community-based surveillance was less common – but the latter was associated with higher PAB coverage. Most common were surveillance for neonatal tetanus and maternal death. Surveillance for congenital rubella, neonatal deaths and neonatal sepsis was more frequently reported by Group 4 countries; congenital rubella and neonatal deaths were also more frequently reported in countries with high PAB coverage. Discrepancies in reporting were, however, observed during the country visits, showing that having a surveillance system in place does not automatically imply sufficient quality of the data.
The reporting of AEFIs is another key component of the introduction of new vaccines that was established by most of the countries responding to the online survey. During the country visits, a number of limitations regarding the content and the quality of reporting became evident: no distinction between pregnant and non-pregnant women in AEFI forms, not having any forms available, and reporting zero AEFIs to the district when the true number in the facility was unknown.
Logistics and supply of vaccines are key elements for the delivery of vaccines. National EPI and ANC managers generally estimated the quality of the procurement, distribution and storage conditions of vaccines in their country as satisfactory. Countries with high PAB coverage, compared with countries with low PAB, expressed more satisfaction with the process of transporting vaccines from health-care institutions to vaccination centres, with storage at vaccination sites and with the vaccination sessions. Concerns with their on-site cold chain capacity or with vaccine supply were more often expressed in Group 2 countries. Problems cited included limited maintenance of freezers and refrigerators, poor temperature monitoring of refrigerators and insufficient vaccine supply, leading to stock-outs. Most health facilities used multi-doses and had an open-vial policy.
However, only about two out of three health facilities calculated wastage rates for tetanus toxoid- containing vaccines.
Observations of vaccine administration showed adequate techniques used in nearly all facilities, except for the observation of vaccinated people for 15 minutes, for adverse reactions, which was rarely done. Auto-disabling syringes were used and disposed of in a safe container, and needles
were not re-capped, which was confirmed in nearly all health facilities. Only Group 2 health facilities were less likely to have adequate safe-box needle disposal. If waste was kept outside the building, the area was often not fenced. In many facilities, waste was picked up weekly or monthly and taken to an incinerator off-site.
Demand-creation activities are a critical part in the introduction of a new vaccine, but also in maintaining high vaccination coverage with existing vaccines. In the country visits, 78% of health facility managers reported conducting demand-creation activities in their catchment area, but rarely or never for maternal tetanus immunization. Most of the countries visited partnered with non-governmental organizations for some or most aspects of demand creation, especially for the creation of information, education and communication materials.
In all country groups, demand-creation activities relied heavily on CHWs, the majority of whom were unpaid, as critical links between health workers and the local community. Having a strong CHW system in place and having the support of local community and religious leaders increased trust in the health system and vaccination campaigns.
Conclusions and recommendations for introducing new maternal vaccines
The results from the MIACSA study suggest that some changes would be required before the introduction of new maternal vaccines. Overall, the following considerations are recommended.
Governance and leadership
■
■ Strengthening the national immunization technical advisory group, such as through the inclusion of maternal, newborn, child and adolescent health (MNCAH) experts, to provide guidance for the introduction of new maternal vaccines.
■
■ Increasing the number of contacts with pregnant women by implementing the current WHO ANC guidelines for eight contacts, to increase opportunities for vaccination during each pregnancy.
■
■ Setting targets for maternal tetanus immunization and improving the accuracy of population data in order to have correct information about the denominator, which is needed to measure coverage.
■
■ Aligning and providing information exchange between public- and private-sector activities through policies and practice for record keeping and reporting.
■
■ Clarifying the reporting requirements placed on private providers, and enforcing existing reporting regulations – needed for accurate coverage data.
■
■ Strengthening communication between the EPI and MNCAH programmes at the national and subnational levels for efficient collaboration in the delivery of maternal immunization through ANC.
ANC service delivery
■
■ Recognizing the important role ANC services can play in the provision of maternal immuni- zation services, it will be necessary to improve the use of ANC by pregnant women. This could be done through demand-side measures (e.g. improving the quality of services, reducing the access costs) or supply-side measures (e.g. training more staff or providing additional ANC opening times).
■
■ Increasing frequency, content and coverage of outreach activities for ANC to create demand and provide services, to improve ANC and maternal tetanus immunization coverage.
■
■ Reducing time spent at the facility through increased coordination between ANC and EPI services.
■
■ Offering maternal tetanus immunization at the same time as ANC (one-stop approach) to reduce missed opportunities, and to reduce indirect and opportunity costs for pregnant women.
■
■ Resolving transport constraints and communication gaps in referral systems without shifting transport costs to patients.
Human resources
■
■ Addressing the shortage of staff capacity and the quality of their training prior to the intro- duction of new maternal vaccines.
■
■ Offering refresher training to build skills, including for communication and reporting procedures.
■
■ Improving feedback mechanisms to better use the potential of supervisory visits for strengthening the quality of the services.
Financing
■
■ Providing external financial support to strengthen downstream capacities of services providing maternal immunization and ANC, and for capacity-building to plan, manage, and disseminate upcoming maternal vaccines equitably across low-resource countries.
■
■ Providing financial support for the hiring and training of health personnel to address staff shortages and the quality as well as the quantity of staff.
■
■ Supporting user-fee exemptions for ANC and maternal tetanus immunization for all vulnerable women.
Information systems
■
■ Improving tracking and reporting systems and facilitating links between records for mother and child.
■
■ Building or adapting surveillance and health management and information systems (HMIS) for new vaccines. Additional research around surveillance systems (active surveillance) is needed.
■
■ Consolidating the different modes of record keeping and addressing the challenges of linking records and inaccurate recording.
■
■ Documenting ANC visits by gestational age to improve tracking and to inform introduction strategies for vaccines with specific “vaccination windows”.
■
■ Addressing inaccuracies in the recording of previous immunization details, to avoid over- immunization, and promoting the use of electronic systems for increased efficacy of the recording.
■
■ Aligning reporting by private-sector providers, enabling the calculation of more accurate coverage rates.
■
■ Strengthening the reporting of adverse events following immunization (AEFI) and ensuring that pregnancy is recorded for the monitoring of adverse events following maternal vaccination.
Logistics, infrastructure
■
■ Investing in the EPI cold chain will also benefit maternal immunization as maternal vaccines are usually stored and distributed in the EPI cold chain.
■
■ Strengthening the management of storage and supplies by clarifying responsibilities and training staff (including ANC staff).
■
■ Addressing maintenance and infrastructure problems.
Demand creation and vaccine confidence
■
■ Supporting sustainable community-based activities to facilitate demand creation for new vaccines.
■
■ Assessing specific pregnancy-related perceptions and beliefs to inform tailored demand- creation activities.
■
■ Strengthening the community health worker system to facilitate communication with communities and improve vaccine confidence.
1. Introduction
1.1 Background
Vaccine-preventable diseases are among the main causes of child morbidity and mortality globally, particularly in low- and middle-income countries (LMICs) (1,2). Since the 1990s, public health interventions have more than halved the mortality of children under five years of age; however, progress in the prevention of stillbirths and neonatal mortality (death in the first 28 days of life) has been slower. Infections and preterm births still contribute to nearly half of all neonatal deaths.
Maternal infections are estimated to contribute to between 10% and 50% of stillbirths and almost 25% of all neonatal deaths – and several of these infections are vaccine preventable (3).
Maternal vaccines given to pregnant women in the second or third trimester have emerged as a promising way forward to address vaccine-preventable diseases, providing protection to the newborn through the transfer of maternal antibodies via the placenta. Maternal tetanus vaccination, according to World Health Organization (WHO) estimates in 2015, has led to a 96%
reduction in neonatal tetanus since the late 1980s, providing a proof of concept for the application of maternal vaccination to prevent maternal and neonatal infections in low-resource settings. In 2018, WHO and the United Nations Children’s Fund (UNICEF) recommended shifting to tetanus and diphtheria vaccine rather than tetanus toxoid vaccine (4). In 2012, WHO recommended influenza vaccination in addition to maternal tetanus vaccination for pregnant women in countries that use influenza vaccine. Several countries, including LMICs, have begun work towards implementing this recommendation. Vaccination for pertussis, hepatitis (both A and B), yellow fever, meningococcus, pneumococcus and polio are recommended in specific situations (5).
Vaccines are under development for administration during pregnancy to protect infants against potentially serious infections that occur in early life, such as respiratory syncytial virus and group B Streptococcus (6–8). Other vaccines to protect pregnant women could be developed, such as vaccines malaria, hepatitis E and Zika virus.
1.1.1 Information gap and study rationale
An expansion of maternal immunization against more diseases promises to further reduce vaccine- preventable mortality among neonates and young infants. Despite recommendations by WHO, however, the introduction of influenza and pertussis immunization to maternal immunization programmes in LMICs is rather slow (9). Reasons for a slow uptake of maternal immunization are manifold and include financial and service-delivery constraints on the side of the health system, and vaccine-safety concerns among potential users (10). The latter is in spite of the global evidence for the safety of maternal vaccination (11). Vaccination performance further differs according to countries’ economic situations and external support (12), and by level of health system decentralization (13,14). However, little is known about how health systems respond to these challenges, and whether there are specific service-delivery models that are generally better suited to providing maternal vaccination services than others (15).
Antenatal care (ANC), which is generally accepted as the natural entry point for interventions during pregnancy, such as maternal immunization, provides important opportunities for the prevention and management of diseases affecting pregnant women and their newborns. Despite progress made in the use of ANC, though, it is estimated that, globally, only 61.8% of women from 2010 to 2016 received at least four ANC contacts (16). Since late 2016, the new WHO ANC model recommends delivering ANC services in a more integrated manner. This increases the number of
contacts to a minimum of eight, with more frequent contacts during the third trimester (17). In addition to the challenge of countries meeting the recommended number of ANC contacts, data on the quality of ANC services beyond just the contact is scarce. This calls for a better understanding of the optimal ways to deliver vaccines to pregnant women, and of the value of using ANC services as a delivery platform.
The Expanded Programme on Immunization (EPI) was initiated some 40 years ago, and WHO estimates that immunization saves the lives of around 2.5 million people each year and protects many millions more from illness and disability (18). The EPI provides routine childhood immunization against diphtheria, pertussis, tetanus, measles, poliomyelitis and tuberculosis, with many countries also offering vaccines against rubella, hepatitis B, Haemophilus influenzae type b, rotavirus and pneumococci. In many countries, human papillomavirus vaccine is included for girls and boys, and routine immunization against regionally important diseases such as epidemic meningococcal meningitis, yellow fever and Japanese encephalitis is also offered. Both ANC services and EPI are used as platforms for vaccinating pregnant women against tetanus. The efficiency and effectiveness of ANC and EPI programmes to deliver vaccines to pregnant women deserves review. This will inform the optimal service-delivery strategy for maternal tetanus vaccination in the context of the potential introduction of other vaccines for pregnant women.
The specific health system characteristics and the rationale behind the choice of vaccine service- delivery strategy by LMICs needs to be understood to help to optimize the use of maternal tetanus vaccines across countries, and to generate generalizable information for the introduction of maternal vaccines.
1.1.2 Aims and scope of the study
WHO works with partners to expand the coverage of high-impact interventions that ensure women and newborns survive and stay healthy during childbirth and beyond. As part of ongoing global research in the area of maternal vaccination, WHO departments are conducting an analysis of the present and future delivery of maternal immunization and ANC in LMICs (the Department of Maternal, Newborn, Child and Adolescent Health, the Department of Immunization, Vaccines and Biologicals, and the Department of Reproductive Health and Research).
To learn from previous experiences and to better understand factors that could affect the introduction of new maternal vaccines, the Maternal Immunization and Antenatal Care Situation Analysis (MIACSA) study set out to investigate ongoing maternal tetanus immunization pro- grammes. As the most widely implemented vaccine given in pregnancy, these programmes have been in place for the past three decades. Maternal immunization is contributing towards the elimination of maternal and neonatal tetanus worldwide. In the least developed areas of mainly low-income countries, tetanus cases are often birth-associated following unhygienic deliveries and poor postnatal hygiene, affecting both mothers and their newborns if the mothers are insufficiently vaccinated prior to giving birth.1 Until today, not every country has achieved maternal and neonatal tetanus elimination, pointing to gaps in the implementation of an effective maternal tetanus immunization programme. The MIACSA project aims to improve the understanding of the challenges and successes of ANC and EPI services in implementing maternal immunization with tetanus toxoid and other maternal vaccines. Research outcomes can inform the sustainability strategy of the maternal and neonatal tetanus elimination initiative as well as the development of approaches towards the delivery of new maternal vaccines.
1 Protection from birth-associated tetanus for both mothers and their newborns is assumed if the mother has received six doses of tetanus toxoid-containing vaccines in her lifetime. It is recommended that all pregnant women with fewer than six doses or with unclear vaccination status are vaccinated with at least two doses of tetanus toxoid-containing vaccine, at least four weeks apart and with the second dose given at least two weeks before birth (19).
1.1.3 Research questions
In November 2016, the MIACSA project set out to investigate the typologies of delivering vaccines to pregnant women in LMICs, to assess what system attributes correlated with high performance in the delivery of vaccines to pregnant women. It discussed the path for introducing additional vaccines targeting pregnant women. The following overarching research questions guided the study:
1. Which service delivery models (ANC, EPI, other) exist for maternal tetanus immunization in LMICs?
2. How do service-delivery models described in research question 1 correlate with the perfor- mance of maternal tetanus immunization (measured as protection at birth) in LMICs?
3. How do selected systems factors correlate with the performance of maternal tetanus immunization in LMICs?
4. How do both service-delivery models and selected systems factors correlate with the performance of maternal tetanus immunization in LMICs?
Several secondary questions were formulated, which are presented in Annex 1.
1.2 Conceptual framework
The introduction of additional vaccines for pregnant women requires careful consideration of the current capacities of health systems, the anticipated sustainability of such programmes and pregnant women’s acceptability of additional vaccines (15,20). ANC delivery systems must have the capacity to identify women’s vaccination status and to vaccinate at the right time during gestation. Timely universal access for pregnant women to ANC services is an important pillar for the prevention of maternal and neonatal tetanus, and will be equally relevant for the introduction of new vaccines. Another potential entry point for maternal tetanus immunization is EPI, which is an established part of health-care services in many LMICs.
Overarching factors influencing the delivery and uptake of maternal immunization, and maternal tetanus immunization in particular, draw from the six defined health-system building blocks, including the following major system categories: leadership and governance, service delivery, health workforce, financing, information, products, logistics and infrastructure, and demand (see Figure 1).
Figure 1. Conceptual framework of the determinants of delivery and uptake of maternal immunization
LEADERSHIP AND GOVERNANCE
SERVICE DELIVERY
HEALTH WORKFORCE
FINANCING
INFORMATION
PRODUCTS, LOGISTICS, INFRASTRUCTURE
DEMAND
Political comment Health Regulation
NITAGNRA Legislation Norms and standards
Organization, structure and reform Accountability Programme planning and
management
Political priorities Budgeting
Policy
Partnerships and coordination Interagency coordinating
committees
Demand creation
Delivery modalities ANC/EPIANC
EPI Integration Spatial organization
Access and coverage Physical and geographic
Affordability Culturally appropriate
Equitable access Referral Outreach Record keeping
Tracing Follow-up mechanisms
Partnerships Quality of care
Timeliness Safety
Performance and supervision Supportive supervision
Guidelines
Availability and distribution of staff Workload and work conditions
Number and distribution Role of CHW Training and capacity Pre- and in-service training
Communication Non-discrimination
Technical capacity
Remuneration and satisfaction Incentives
Wages and career
External financing Donor pooling Innovative mechanisms
Budget support Domestic financing
Fiscal space Fund handling
Affordability Sustained financing
Vaccine pricing Service integration
Health markets User fees and exemptions
Financing mechanisms
Disbursement Competing priorities Treatment and hospitalisation
Functioning data collection and reporting
Actionable HMIS Data collection process
Data quality
Monitoring and surveillance Surveillance sites and platforms
Impact monitoring Safety (AEFI) Information campaigns
Demand creation
Data for decision making Data management
Supply chain Procurement Distribution
Cold chain and storage
Cold chain management Vaccine management Stock management Efficiency and wastage
Waste disposal EVM
Forecasting Demand and supply and
forecasting
Socioeconomic factors Socioeconomic status Personal characteristics
Socio-cultural factors Health literacy Knnowledge on vaccination
Mobility and security Culture and religion
Health system interaction Receives adequate information
Distance to facility Cost (direct/indirect/opportunity)
Transport Infrastructure Communication Non-discrimination
Social mobilisation Advocacy Incentives
2. Study methodology
To capture the complexity of maternal tetanus immunization service delivery, the Maternal Immunization and Antenatal Care Situation Analysis (MIACSA) study used a multi-methods approach collecting global data on maternal tetanus immunization and on related health- system factors in low- and middle-income countries (LMICs). Both quantitative and quantitative- qualitative methods were used at different levels, including a global desk review, a written survey of LMICs, a telephone survey with a sub-sample of countries, and in-person country visits. During the latter, key informant interviews and health service delivery observations were conducted.
Data were collected in four phases (see Figure 2):
1. Desktop review of global databases for selected health indicators in 137 LMICs;
2. Quantitative online survey sent to 116 LMICs;2
3. Semi-structured telephone interviews with 26 selected LMICs; and
4. In-depth key informant interviews, focus group discussions and health-care facility observa- tions during country visits to 10 LMICs.
2 Responses were received from all regions, except the World Health Organization (WHO) European region, which decided not to participate in this research project because the maternal neonatal tetanus elimination initiative is not a priority here. For similar reasons, the Western Pacific Regional Office decided not to contact a number of its countries. In Latin America and the Carribean also countries recently classified high
Country selection ensured:
1. Regional balance
2. All vaccine-delivery models represented
3. Performance categories (high and low ANC4+, and protection at birth) represented
Figure 2. Flowchart and timeline for data-ollection phases of the MIACSA project
Desk review (137 LMICs) November 2016 – August 2017
Global online survey (95 LMICs) April – September 2017
Telehone interviews (26 LMICs) August 2017 – April 2018
Country visits (10 LMICs) February – August 2018
Project timeline November 2016 – June 2019
ANC4+: the proportion of pregnant women who received four or more contacts for antenatal care (ANC) during their last pregnancy; LMICs: low- and middle-income countries; MIACSA: Maternal Immunization and Antenatal Care Situation Analysis
The MIACSA project took place from November 2016 to June 2019 (32 months).
2.1 Study design and sources of data
The four phases of the MIACSA study were conducted in sequence, each building on the next. The methodologies used are presented for each study module separately in the following sections.
2.1.1 Desktop review of global databases
As a first step, a desktop review of existing global databases sought to retrieve information about vaccination, antenatal care (ANC), delivery and related policies and programmes. In addition, information about the country income level and selected socioeconomic indicators was included.
The following databases were used to extract data:
■
■ Demographic and Health Surveys (DHS)/Multiple Indicator Cluster Surveys (MICS)
■
■ WHO/UNICEF estimates of national immunization coverage (WUENIC)
■
■ WHO/UNICEF Joint Reporting Form
■
■ Maternal and neonatal tetanus elimination database
■
■ WHO maternal, newborn, child and adolescent health policy indicators
■
■ Gavi, the Vaccine Alliance’s eligibility criteria for support
■
■ United Nations Inter-agency Group for Child Mortality Estimation (UN IGME)
■
■ Trends in maternal mortality: 1990 to 2015. Estimates by WHO, UNICEF, UNFPA, World Bank Group and the United Nations Population Division
■
■ United Nations Department of Economic and Social Affairs, Population Division (UNPD)
■
■ World Bank Group data for economic level and female literacy rate.
Data were available for 137 LMICs. Indicators retained for the analysis as measures of vaccination and ANC performance included coverage of at least two doses of tetanus toxoid vaccine during pregnancy (TT2+), estimated protection at birth (PAB), infant coverage with the third dose of the diphtheria, tetanus and pertussis (DTP) vaccine (DTP3), coverage of completion of one ANC visit during pregnancy (ANC1) and of completion of at least four visits (ANC4+), as well as the year of data collection. Additional indicators were maternal and neonatal tetanus elimination status and year of elimination; tetanus toxoid supplementary immunization activities; type of tetanus vaccine administered (tetanus toxoid or tetanus and diphtheria); vitamin A given in routine immunization; influenza and pertussis vaccines given in routine immunization; number of adverse events following immunization; neonatal tetanus and total cases of tetanus reported; existence of a national immunization technical advisory group; maternal mortality rates; neonatal mortality, infant mortality and stillbirth rates; coverage of institutional deliveries (year, percentage);
availability of policy on minimum ANC contacts; tetanus toxoid-containing vaccination for pregnant women in routine contacts; infant coverage of vaccines – Bacillus Calmette-Guérin (BCG), the first and third doses of the DTP vaccine (DTP1 and DTP3); physician and midwife density; eligibility for Gavi support; female literacy rate; percentage of total expenditure on routine immunization financed by government funds; and World Bank income classification (low, lower middle and upper middle income levels). See Annex 2 for the list of all variables initially available.
The data informed the development of the different study tools and questionnaires and suggested a categorization of countries in terms of health system performance for maternal tetanus immunization and ANC service delivery (see Chapter 2, section 2.3.1). The country selection for in-depth telephone interviews and for the country visits was informed by the desktop review.
2.1.2 Written online survey
The instrument for the written survey included mandatory and non-mandatory questions on vaccination and ANC coverage, on policies and programmes, and quality control measures to prevent false entries to all questions. An online version and a PDF version of the questionnaire were available. The online survey collected information on 18 items structured along four sections (see Annex 3):
■
■ service delivery models
■
■ ANC system capacity and maternal immunization
■
■ vaccine safety surveillance
■
■ other maternal vaccines used.
The questionnaire was answered by 97 countries, two incompletely, leaving 95 countries included in the analysis (Figure 3).
2.1.3 Telephone survey
For the telephone survey, 30 countries were approached and 26 consented to interviews (Benin, Bhutan, Botswana, Cameroon, Colombia, Congo, Costa Rica, Democratic Republic of the Congo, Dominica, Egypt, Ethiopia, Fiji, Gabon, Malawi, Morocco, Nicaragua, Niger, Pakistan, Papua New Guinea, Senegal, Sierra Leone, South Sudan, Sri Lanka, Thailand, Uganda and the United Republic of Tanzania). The telephone interview was developed and piloted in two countries (Sri Lanka and the United Republic of Tanzania), and thereafter adapted based on the comments provided. The telephone interview consisted of 91 questions, which followed and built on the questions posed in the online survey (see Annex 4). This enabled the collection of more in-depth information about service delivery models, ANC capacity, maternal immunization and information systems.
The instrument was developed both as a paper document and in electronic form, the latter using Open Data Kit. The electronic version made the already collected data available to the country visit teams during the third phase of the project. The survey results were exported into a Microsoft Excel file, which for consistency was reviewed in duplicate by two consultants.
Figure 3. Countries participating in the different study phases
Phase IV: country visit Phase III: telephone interview Phase II: online survey
2.1.4 Country visits
Country visits took place in Benin, Bhutan, Ethiopia, Fiji, Gabon, Morocco, Panama, Senegal, Thailand and the United Republic of Tanzania. Structured interview guidelines were developed for the team conducting the country visits. The stakeholders interviewed were national-level managers of the Expanded Programme on Immunization (EPI) or maternal, newborn, child and adolescent health managers, subnational-level managers and health facility managers. Where possible, interviews were also conducted with community health workers. The instruments were made available on an electronic tablet device using Open Data Kit technology, and the responses and observations were directly captured on it, allowing for some standardization of the data collection. Observation guides for ANC services and maternal immunization activities were available. In each country, the ministry of health identified between six and 14 health facilities, and 96 health facility interviews were conducted. In addition, 23 face-to-face interviews were conducted with managers at the national or subnational level, mainly to validate information given in the telephone interviews.
The data obtained from the interviews and observations were retrieved from the Open Data Kit database while the team was still in the country. The data were exported into an Excel sheet to provide an overview of responses from all interviews by question. The information was compiled in a short report and a presentation matrix, and presented to the respective ministry of health for review and validation at the end of the country visit. Figure 3. Countries participating in the different study phases
2.2 Key definitions and indicators
2.2.1 Maternal immunization performance
To measure the performance of maternal tetanus immunization, vaccination coverage assessed by PAB is used as a proxy variable:
PAB is the proportion of neonates protected at birth against neonatal tetanus, by
combining data on the number of tetanus vaccine doses received by the mother by the last baby born, interval between doses, and time since last dose (using card or verbal history).
PAB is a supplemental method of determining tetanus coverage in pregnancy (which could be with tetanus toxoid or tetanus and diphtheria, but tetanus toxoid will be used hereafter), particularly where TT2+ is unreliable and where the coverage of the first dose of the DTP vaccine (DTP1) in infants is high. To monitor PAB during DTP1 visits, health workers record whether infants were protected at birth by the mother’s tetanus toxoid immunization status. The percentage of PAB is then estimated as the number of infants protected divided by the total number of births. The rate will, however, be less accurate if DTP1 coverage is low, which results in missed opportunities to assess PAB. If a child is determined to have been unprotected, the mother should receive a dose of tetanus toxoid during the same visit and should be followed up with a subsequent tetanus toxoid dose if needed for protection. The same applies for mothers whose children were protected at birth but who remain eligible for another tetanus toxoid vaccine dose. Unlike TT2+ coverage, PAB therefore accounts for women who have previously received protective doses, women who received one dose without documentation of previous doses received, and women who received doses in supplemental immunization activities for tetanus toxoid (or tetanus and diphtheria). PAB has thus been described by WHO as a good indicator of health system performance and will also include in the numerator those babies born to women having received all the necessary doses for life-long protection against tetanus, even if not through TT2+. PAB was used as a dichotomous variable: high versus low PAB (respectively, less than 90% PAB or 90% and over). The median coverage was selected as the cut-off point.
A limitation of using PAB as an indicator is the fact that the rate will be less accurate if DTP1 coverage is low, because all babies not coming for DTP1 vaccine are categorized by default as not being protected at birth against tetanus.
2.2.2 Antenatal care performance
To measure ANC performance, the proportion of pregnant women who received one ANC visit during their last pregnancy (ANC1) and the proportion of pregnant women who received four or more ANC contacts during their last pregnancy (ANC4+) were used. Both indicators were collected through the desk review.
2.2.3 Performance of the Expanded Programme on Immunization
To measure EPI performance, data from the desk review on coverage of the third dose of the infant DTP vaccine (DTP3)3 and the third dose of the pentavalent vaccine (Penta3)4 was used as a proxy, using WUENIC estimates and administrative data sources.
2.3 Statistical analysis
All data were imported into Stata (release 15, StataCorp LCC, Texas) for analyses. Data were checked for completeness and consistency. Variables collected using a multiple-answer option were split into individual variables reflecting each answer category. As necessary, new variables were created combining different questions.
All analyses were conducted using non-missing data. Summary measures (proportions, means, medians and standard errors) were calculated for all variables using univariate analysis. To identify maternal tetanus immunization service delivery models and process components favourable to the vaccination of pregnant women, associations of different service delivery process components with PAB coverage and with country groups were established. First, bivariate analyses were conducted to assess the relationship between different variables according to specific hypotheses. The significance of the relationship was tested with Fisher’s exact test. A two- sided P-value of 0.05 or below was considered as significant throughout. In addition, regression analysis and structural equation modelling was used to assess associations of different variables with maternal tetanus immunization. Regression analysis depended on the data structure and included simple logistic regression analysis for binary outcomes of country-level data, multilevel mixed-effect logistic and ordinal regression analysis for individual-level outcome variables, as well as generalized estimating equations models for group-level outcome variables, as in the case of country-level coverage rates as dependent variable with health-facility-level predictors. For the analysis of the country-level data, robust standard errors were calculated.
2.3.1 Latent class analysis
Central to the MIACSA study was the question of whether specific health service delivery factors would be associated with maternal immunization performance. For this purpose, countries were initially grouped into low- and high-performing countries based on PAB coverage (low or high).
Dividing the countries into two groups did not, however, consider the differences that may exist, within the countries of high and low PAB coverage, in the potential of ANC or EPI services as platforms for maternal immunization. A classification that takes into consideration the potential for maternal immunization would have to consider several indicators of ANC and EPI capacity.
Latent class analysis (LCA) is a statistical approach that can be used to group countries based on several indicators. It enables the characterization of an unobserved (latent) variable through the analysis of the structure of the relationship among several observed variables (22,23). In our case, it enables the characterization of a latent categorical variable for the potential to protect women and their young infants from vaccine-preventable infections, from a set of observed indicators related to maternal immunization. LCA allows the inclusion of many indicators to define groups, and assigns group participation based on the highest probability according to the model. LCA
3 Coverage for DTP3 is calculated by dividing the number of children receiving the third dose of DTP vaccine by the number of children who survived to their first birthday (21).
4 The pentavalent vaccine protects against five infections: diphtheria, tetanus, pertussis, hepatitis B and
thus allows the accommodation of complex situations, as in the case of maternal immunization capacity. A further advantage of the approach is the conditional independence of the variables included in an LCA model, which responds to the problem of multi-collinearity.
To enrich the analysis, an LCA was conducted post hoc – after the data had been collected and preliminary analyses done. The selection of variables included in the LCA model was based on theoretical considerations and preliminary analyses of the desk review indicators. LCA was conducted using the doLCA.ado function5 in Stata release 15. The number of country groups was selected based on the model with the lowest Bayesian information criterion. The LCA model provided the probability of each country to belong to a specific category of the latent variable, thus characterizing a country’s potential to protect mothers and young infants from vaccine- preventable infections. Individual countries were allocated to groups (categories) based on the highest probability.
Classifying countries using LCA produces a relative rather than an absolute classification.
Depending on the variables included in the LCA model, the appearance of some countries in a specific group changed. The LCA classification must therefore be interpreted with some caution as it only indirectly measures a latent trait and is not based on classification criteria such as the cut- off value of a specific indicator. The classification serves mainly to identify clusters of countries as an auxiliary means to better understand the way health system characteristics relevant for maternal immunization differ between clusters of countries.
2.4 Study limitations
The study methodology has several limitations. The different data sources used in the MIACSA project mean that inconsistencies between the different databases are likely to be observed for some quantitative and policy-related indicators. This can be due to the different time points by which the information was collected, and the different databases referenced in the desk review and the MIACSA online survey. In other instances, indicators may not have used exactly the same definitions, or the sampling methodology varied. This was the case, for example, for coverage indicators collected during the desk review and national coverage indicators collected through the telephone interviews and country visits. Besides different sources of data, inaccurate reporting cannot be fully excluded despite the efforts made by the countries to validate the information provided to the study team.6 Due to the different data sources and time points of data collection, conflicting information was not updated between the databases. Differences were described and the sources of the data highlighted.
Another limitation was the relatively small number of countries visited, and the number of health facilities by country. Data analysis was done both qualitatively and quantitatively, the latter using conservative statistical approaches, taking into account the clustering of information or group- level information in the choice of statistical approach.
Health facilities, in addition, were not randomly selected but had been assigned by the local ministries of health. Health facilities with various performance were included deliberately to give a broad representation of strengths and weaknesses. As the country selection for the country visits was geared towards best practices, the selection of well performing health facilities is not problematic in principle for the study. Well performing health facilities are not representative of the entire country but rather reflect the potential of the health system in a particular country. Data collected from the 96 health facilities across the 10 countries were also analysed quantitatively,
5 See https://methodology.psu.edu/downloads/lcastata.
6 Conflicting information was confined to a limited number of cases. For example, information on policies on the number of ANC visits was available for 122 LMICs through the desk review, and among these, three countries indicated not having a policy on the number of ANC visits. In the online survey among the 95 countries who responded, all the same three countries indicated having a policy of a minimum of four ANC visits. In contrast, in the online survey, five countries indicated not having a policy on the number of ANC visits in place. Among those, four were listed as having a respective policy in the desk review data.
