6

A Systems Approach to Increasing Vaccine Confidence and Uptake: Opportunities for Community-Based Strategies

The second half of the workshop’s third session focused on how community-based approaches and strategies can effectively and ethically be employed to reduce vaccine hesitancy and increase vaccine confidence. The session was moderated by Rafael Obregon, United Nations Children’s Fund (UNICEF), who highlighted the importance of dealing with vaccine hesitancy and vaccine confidence as parts of a larger ecosystem that includes other systems and services. These issues are inseparable from how communities, parents, and individuals engage with providers, as well as context-specific levels of trust and confidence that are built in those relationships. Mohamed Jalloh, senior behavioral epidemiologist at the U.S. Centers for Disease Control and Prevention (CDC), presented on social mobilization as a strategy to increase vaccine acceptance and uptake. Catherine Jackson, director and World Health Organization (WHO) consultant at Valid Research Limited, described how the COM-B model was adapted and integrated within the Tailoring Immunization Programmes (TIP) approach to increase vaccination acceptance and uptake. Clarissa Hsu, assistant investigator at the Kaiser Permanente Washington Health Research Institute, presented on the Immunity Community, a community-engagement strategy to boost vaccine confidence. Louise Letley, nurse manager for Research, Immunisation Operations at Public Health England, provided an example of engaging with faith communities to increase vaccine acceptance and uptake in North London’s Charedi Orthodox Jewish community. Patsy Stinchfield, senior director of infection prevention and control at Children’s Minnesota, presented an example of engaging with immigrant communities to increase vaccine acceptance and uptake in the

Somali American community in Minnesota. Each speaker presented an example of how trust and confidence can be built under different contexts, the lessons learned, and the models used to address these challenging issues.

SOCIAL MOBILIZATION AS A STRATEGY TO INCREASE VACCINE ACCEPTANCE AND UPTAKE

Presented by Mohamed F. Jalloh, U.S. Centers for Disease Control-Tanzania

Jalloh explored how social mobilization can be used as a strategy for increasing vaccination acceptance and uptake. He noted that experiences with these strategies from low- and middle-income country (LMIC) contexts have crosscutting implications and could potentially inform social mobilization strategies in the United States. Jalloh and colleagues published a commentary on lessons learned from social mobilization for immunization comparing experiences in different LMICs, including Bangladesh, India, and Sierra Leone, among others (Jalloh et al., 2020). The study examined social mobilization across a number of contexts: routine immunization, supplemental immunization activities, campaigns, new vaccine introductions (e.g., human papillomavirus), and outbreak response. Jalloh presented two examples of successful social mobilization campaigns drawn from this commentary to help contextualize issues related to social mobilization and to identify crosscutting lessons learned.

Functional Definitions of Social Mobilization

Jalloh noted that the term mobilization has its roots in the military concept of mass military mobilization, which involves well-coordinated and comprehensive activities that go through a chain of command, and he presented two functional definitions of social mobilization. UNICEF defines social mobilization as “a process that engages and motivates a wide range of partners and allies at national and local levels to raise awareness of and demand for a particular development objective through dialogue.”¹ He noted that the UNICEF definition—which is widely used—emphasizes that social mobilization is a process that involves engagement. In another functional definition, Rogers and colleagues describe social mobilization as “the effort to marshal many people to perform behaviors that impose a net cost on each individual who complies and provides negligible collective benefit unless performed by a large number of individuals” (Rogers

___________________

¹ More information about social mobilization is available at https://www.unicef.org/policyanalysis/42347.html (accessed November 5, 2020).

et al., 2018). Jalloh remarked that this definition highlights the need for collective effort to achieve a collective outcome; it also recognizes that social mobilization imposes a net cost on each individual, be it financial or social. He added that both definitions construe social mobilization as a means to an end, such as the increased uptake of vaccinations. In their commentary, Jalloh and colleagues developed a definition of social mobilization for immunization as “the collective effort by diverse stakeholders to ensure optimal vaccination uptake in a target population by generating and sustaining demand for vaccines using community-based participatory approaches” (Jalloh et al., 2020).

Universal Child Immunization Campaign in Sierra Leone (1985–1990)

Jalloh described the universal child immunization (UCI) campaign implemented in Sierra Leone (1985–1990) as an early example of successful social mobilization. The UCI campaign was launched with the aim of achieving universal vaccination coverage by 1990 for all children worldwide against six antigens: measles, tetanus, whooping cough, diphtheria, tuberculosis, and polio (Mandl, 1985). In 1985, Sierra Leone recognized it was lagging behind in achieving the immunization targets established under the UCI campaign. With only 6 percent of the country’s children vaccinated against the six antigens by 1985, Sierra Leone’s vaccination coverage lagged behind most nations worldwide, including most countries in Africa. In an attempt to improve vaccine coverage, Sierra Leone’s Ministry of Health (MOH) invested in interventions that primarily focused on supply-side components of vaccine service delivery—such as improving and enhancing the cold chain—but coverage did not improve.

By early 1987, Sierra Leone only achieved 10 percent vaccine coverage, said Jalloh. In response, the MOH and UNICEF conducted a survey of knowledge, attitudes, and practice. It revealed that many survey respondents had inadequate knowledge about vaccines and intended to refuse vaccination. For example, at the time, Sierra Leone did not even have an indigenous word for vaccine. Without a word for vaccine, respondents had difficulty in receiving information about vaccines in local languages. Additionally, respondents reported that they did not have information about where and when to access vaccinations. The survey also revealed that many respondents intended to refuse vaccination because of the perception that vaccines were incompatible with religious and traditional belief systems.²

___________________

² The Pew-Templeton Global Religious Futures Project estimates that in Sierra Leone, most of the population identifies as either Christian (20.4 percent) or Muslim (78.5 percent). See https://www.state.gov/reports/2018-report-on-international-religious-freedom/sierra-leone (accessed February 18, 2021).

In response to these survey findings, the government of Sierra Leone, with support from UNICEF and other partners, designed a robust nationwide social mobilization campaign that extended across the national, district, chiefdom, and community levels. Jalloh explained that a primary component of the campaign strategy—and a major factor in its success—was to engage religious leaders, given that many survey respondents believed that vaccination did not align with their religious beliefs.³ Rather than providing religious leaders with messages to disseminate, the government’s strategy was to explain the challenges faced in the vaccination campaign and seek to partner with the religious leaders to develop solutions. To develop familiar and easily understood language to frame conversations about vaccination, the leaders found passages in their religious texts with messages that supported immunization to develop the messaging. Then they trained fellow imams and pastors at the national, district, and community levels. The religious leaders were well organized and widely distributed, noted Jalloh, which enabled them to use their platform and position as influential messengers to promote immunization through faith-based messaging across the entire country. The government also engaged the media, educators, and traditional institutions, as well as recruiting paramount chiefs⁴ as ambassadors for immunization. Within 3 years, Sierra Leone had achieved a dramatic improvement in vaccination coverage, from 10 percent in 1987 to 75 percent in 1990 (UNICEF, 1991).

Jalloh noted that contemporary governments face many of the same issues addressed in Sierra Leone three decades prior. He emphasized that even though these structures might be difficult to keep and maintain, once a country has made investments in social mobilization, those structures can be easily repurposed for purposes beyond immunization. For instance, during the Ebola outbreak in Sierra Leone (2014–2015), the government was able to quickly leverage the religious action groups from the UCI era to scale up widespread social mobilization nationwide.

Improving Vaccination Coverage Among Displaced Rohingyas in Bangladesh (2018)

Jalloh remarked that social mobilization has played a pivotal role in other immunization campaigns—such as the successful eradication of polio in Uttar Pradesh, India—but he focused on how social mobilization was used to improve vaccination coverage among Rohingya refugees who

___________________

³ Based on the majority religions, the MOH and UNICEF organized the country’s religious leaders into two action groups: the Christian Action Group and the Islamic Action Group.

⁴ In a country such as Sierra Leone that is administered by chiefdoms, a paramount chief is the highest-ranking leader in a given region.

arrived in Bangladesh from Myanmar in 2018 (Coates et al., 2013; Jalloh et al., 2019). Even though vaccines were available, people living in the refugee camps were not accessing them for reasons that were not well understood. A rapid behavioral assessment was conducted to better understand why the uptake of vaccinations was low. Many of the Rohingyas reported believing that getting vaccinated would cause a scar that would be considered a tattoo, which is forbidden by Islam and would cause them to go to hell when they die. Data from the rapid assessment were used in real time to inform the next vaccination campaign. By engaging religious leaders, traditional healers, and women leaders in the Rohingya camp, they were able to increase vaccination coverage by 10 percent in the next campaign.

Recurring Challenges in Using Social Mobilization for Immunization

Social mobilization can give rise to many different challenges, noted Jalloh. One such challenge occurs when campaign representatives repeatedly engage communities on vaccine-related topics to the point of fatigue and to the exclusion of other issues related to the well-being of communities, such as social and economic interests that may be a priority for them. However, the majority of those challenges involve human resource constraints, inadequate funding, and weak monitoring and evaluation. For example, social mobilization efforts often have inadequate or nonexistent budgets within vaccination campaigns. As a result of underfunding and lack of foresight, many campaigns encounter problems that could have been solved by properly funding social mobilization at the campaign’s inception. Instead, campaigns too frequently treat social mobilization as an afterthought and fund it on a one-off basis. Furthermore, many vaccination campaigns do not budget enough to adequately staff their social mobilization efforts. The effectiveness of social mobilization campaigns can also be undercut when monitoring and evaluation practices are weak, he added.

Lessons Learned from Social Mobilization Efforts in Vaccination Campaigns

Jalloh highlighted five main lessons learned from past experiences in using social mobilization in vaccination campaigns:

1. Avoid the “spare tire” problem.
2. Do not conflate social mobilization with message dissemination.
3. Get the right people.
4. Use behavioral science insights.
5. Use standards, improve quality, and demonstrate effect.

He explained that the “spare tire” problem arises when campaigns consider social mobilization as an afterthought or they only consider it when problems arise, which can render the mobilization efforts largely ineffective. Instead, social mobilization should be considered one of the four main (proverbial) tires, if not the steering wheel. Jalloh drew an analogy with the military to illustrate the “spare tire problem.” Military mobilization is not a spontaneous response to an emergent problem, but a planned response to an anticipated problem. Furthermore, military mobilizations succeed when there is buy-in at all levels, from the highest-ranking generals to the lowest-ranking soldiers, which is achieved by involving and engaging all parties. Immunization campaigns benefit from treating the campaign as a single, coherent social mobilization effort, he maintained.

Message dissemination is a component of—but not tantamount to—social mobilization, cautioned Jalloh. “It is not just about the messages, [but] about how those messages are delivered,” he said. The messenger is equally as important as the message, and how the message is delivered can be even more important. The messenger’s identity, roles, and relationships within the community determine whether people listen to and engage with the message being delivered, regardless of how clear and powerful that message is. Moreover, messages are best delivered by proactively engaging the right people throughout the process of designing, implementing, and monitoring a campaign. Therefore, immunization campaigns should avoid “talking at” the community. Instead, they should engage communities in the design, implementation, monitoring, and evaluation of the mobilization strategies. He noted that too often, campaigns only engage key individuals upon implementation.

Jalloh also highlighted the value of applying insights from behavioral science when designing social mobilization strategies. Such insights include (1) making the behavior observable, (2) normalizing the behavior, (3) aligning the behavior with how people would like to see themselves, and (4) using existing structures and networks in the community. Because social mobilization and community engagement can be somewhat nebulous concepts, he suggested that it will be important to apply standards to improve quality and demonstrate the effect of interventions that operationalize those concepts. He noted that UNICEF had recently published a document that highlights the minimum quality standards and indicators for community engagement,⁵ which can help shape the design and evaluation of high-quality social mobilization efforts. He suggested that such standards should be used to understand process-oriented issues related to social mobilization and to

___________________

⁵ UNICEF’s minimum quality standards and indicators for community engagement is available at https://www.unicef.org/mena/reports/community-engagement-standards (accessed October 22, 2020).

more deeply integrate quality improvement as a core component of social mobilization efforts. Jalloh concluded by emphasizing that social mobilization is not one size fits all—it must be tailored to specific contexts—and data should continue to be generated and used to inform the next generation of strategies.

ADAPTING COM-B FOR THE TAILORING IMMUNIZATION PROGRAMMES APPROACH TO INCREASE VACCINATION ACCEPTANCE AND UPTAKE

Presented by Catherine Jackson, Valid Research Limited

Jackson described how the COM-B model of behavior change was adapted for the WHO TIP approach to increase vaccination acceptance and uptake.^6,7 She provided an overview of how both models can be tailored for vaccination behavior, and she described how the combined approach was used to inform the design of interventions to increase vaccination among internal migrant families in Kyrgyzstan.

The Tailoring Immunization Programmes Approach

The TIP approach is typically initiated to address below-target or declining vaccination coverage in specific unimmunized and underimmunized populations at the national level or within certain population groups or geographic areas. The approach was developed to guide health authorities through an evidence-based and people-centered approach that acknowledges the complexity of vaccination behavior and the diversity of populations. She emphasized that TIP is not a one-size-fits-all approach—it is a communications-based intervention that is designed to facilitate targeted and tailored solutions within a holistic program view.

COM-B Model of Behavior Change

Jackson explained that the COM-B model was developed as a part of the behavior-change wheel to help advance the science of behavior change (Michie et al., 2011). The model has extensive application across multiple

___________________

⁶ COM-B is the behavior theory model that informs the analysis and intervention design for the TIP approach. More information about the Tailoring Immunization Programmes approach is available at https://www.who.int/immunization/programmes_systems/Global_TIP_overview_July2018.pdf?ua=1 (accessed November 4, 2020).

⁷ The COM-B model of behavior change posits there are three factors—Capability, Opportunity, and Motivation—that together influence Behavior. This model is explained in detail in a following section.

health behaviors and was designed to support practitioners through practical applications, as well as designing and evaluating public health interventions and policies. According to the COM-B model, three factors must be in place for a health behavior to occur: capability, opportunity, and motivation. Capability includes physical and psychological capability, motivation includes automatic and reflective motivation, and opportunity includes social and physical contexts. Capability and motivation are associated with individuals, while opportunities are associated with the individual’s physical and social context. All three factors exert influence on some of the other factors, she noted. For example, an individual’s motivation can be influenced by their capabilities and the opportunities in their environment. All three factors combine to influence and produce a behavior and, conversely, an individual’s behaviors can influence all three factors. For instance, a person’s vaccination history is likely a good predictor of future vaccination behavior. Much research has been conducted using the COM-B model, including work done by Jackson to adopt COM-B for TIP.

Rationale for Using the COM-B Model

Jackson discussed the rationale for applying the COM-B model in the TIP approach. COM-B offers broad perspectives on individual and context determinants, which helps avoid blind spots. For example, vaccination program managers often attribute low uptake to well-educated middle-class parents who refuse to have their children vaccinated, particularly in the case of the measles, mumps, and rubella (MMR) vaccine. However, Jackson explained, using the COM-B model rather than merely relying on the assumptions of any single stakeholder ensures that all explanations for behaviors are considered. The COM-B model is applicable for evaluating the vaccination behaviors of various stakeholders, including parents and health workers. A growing body of global evidence on the determinants of vaccination behaviors has confirmed the relevance of the three factors identified by the COM-B model, she said. Unlike other models, COM-B accomplishes the following:

• Links research findings to interventions and effects,
• Provides clear steps to understanding the barriers and drivers to behavior,
• Designs interventions that target those barriers and drivers, and
• Links the monitoring and evaluation framework back to those barriers and drivers.

Jackson commented that the theoretical basis of the COM-B model has developed considerably since 2011, as new complex theoretical processes

have been added to the existing framework. These new theoretical processes include theoretical domains framework (Cane et al., 2012), mechanisms of action (Connell et al., 2016), and Behavior Change Technique (Steinmo et al., 2015). Because COM-B is simple and easy for people in the field to comprehend and use, Jackson and colleagues continue to use this approach in their work.

Adapting the COM-B Model for Vaccination Behavior

Jackson explained that the COM-B model has been adapted for vaccinations based on lessons learned through fieldwork conducted in several countries as well as the translation of research into interventions. To better suit the vaccination setting, the COM-B model has been adapted by (1) distinguishing between physical opportunity and social opportunity and (2) consolidating the subcategories of capability (physical, psychological) and motivation (automatic, reflective) factors (Habersaat and Jackson, 2020). This results in four factors (individual capability, individual motivation, context-specific physical opportunities, and social opportunities) that interplay to shape vaccination behavior (Habersaat and Jackson, 2020).

Jackson explained the rationale for this adaptation through the example of asking parents about vaccine safety. The original COM-B model would make a clear distinction between thinking about vaccine safety (reflective motivation) and worrying about vaccine safety (automatic motivation). In practice, however, it would be difficult for researchers to maintain such a distinction during interviews and subsequent data analysis. Moreover, such a distinction does not add value to the intervention design process, she added. Similarly, the original model’s distinction between physical and psychological capacity adds little value when applied to health workers’ capacities for administering vaccination, as health workers’ physical ability to administer a vaccine is interlinked with their psychological knowledge about how to do so. She noted that data that are relevant to the factors and sub-factors of the original model are still being collected.

Using COM-B in the Tailoring Immunization Programmes Approach

Jackson described how the adapted COM-B model of vaccination behavior is embedded within the TIP approach. The TIP process involves five distinct steps: planning (pre-TIP), situation analysis (phase 1), research (phase 2), intervention design (phase 3), and implementation, monitoring, and evaluation (post-TIP). In the situation analysis phase, governments and researchers inform their decisions about population focus and population target groups by reviewing coverage and disease data and identifying existing relevant research and reports. If they find evidence for barriers and drivers

to vaccination behaviors in the existing data, they can begin to organize that information using the COM-B factors. In the research phase, governments and researchers explore the barriers and drivers to positive vaccination behaviors among the TIP project’s target population. Interview topic guides and surveys are designed to explore the four factors; the analysis is then organized in terms of the four factors. In the intervention design phase, practitioners identify evidence and theory-informed interventions that relate to the four factors. Finally, a monitoring and evaluation framework is used to assess the effect of the interventions by measuring changes in the targeted factors, she said.

Intervention Design

Jackson explained how the COM-B model is used to help inform intervention design (Michie et al., 2011). During the intervention design phase, Jackson’s team conducts a series of activities to translate research findings into interventions through workshops with key stakeholders. These activities help new practitioners of this method understand and appreciate the link between the four factors and appropriate interventions. Applying the COM-B model helps identify specific types of interventions that can be used to address each of the four (adapted) COM factors of capability, motivation, physical opportunity, and social opportunity.⁸ For instance, if interviews during the research phase reveal that parents are not reminded of their child’s appointment and that there is no formal recall or reminder system in place, the practitioner would designate this as a barrier to the physical opportunity factor. The COM-B model identifies a list of potentially appropriate interventions, which in this case would include training, restriction, and environmental restructuring. In the above case, Jackson suggested that practitioners consider implementing a short message service reminder system and training the primary health care administrators to operate the system.

Using the Tailoring Immunization Programmes Approach in Kyrgyzstan

Jackson provided an overview of the geographic distribution of TIP use. To date, most use has occurred in Europe across a diversity of countries and target populations, covering health workers, caregivers, and parents.⁹ To

___________________

⁸ Types of interventions include information/education, persuasion, incentivization, coercion, training, restriction, environmental restructuring, and modeling.

⁹ The Tailoring Immunization Programmes approach has been used in the following countries and target populations: Armenia (medical experts), Australia (communities with lower coverage in several regions), Bulgaria (vulnerable and Roma populations), Estonia (alternative health views population), Federation of Bosnia and Herzegovina (health

illustrate how the approach is implemented in practice, she described the TIP project in Kyrgyzstan for a target population of internal migrant families, which was selected based on the situation analysis. These families tend to move from rural areas to the cities, where they often live in exurban settlements with poor housing and few amenities. A legislation review revealed an overlooked physical opportunity barrier for parents to have their children vaccinated. Although Kyrgyzstan has passed legislation that enshrined free universal primary health care, other legislation was passed to limit primary health care access to the citizen’s registered location. The primary health care facility registration process is a challenge for many migrant families, she noted. Additional research was conducted to study parents and grandparents of vulnerable internal migrant children to learn more about related barriers. Parents reported difficulties attending facilities without registration and without a record of their child’s vaccinations. Many parents reported that they could not access their paperwork, which was held at their original facility of registration. Parents also reported that their original facility could not transfer the paperwork. Many health workers reported that when vaccination records are missing, they fear “over-vaccination” and prefer not to vaccinate. Moreover, health workers had low knowledge regarding the internal migrants’ constitutional right to vaccination. These are examples of physical opportunity barriers, capability barriers, and motivation barriers, she noted.

Multiple interventions were implemented following the completion of the TIP project in Kyrgyzstan. The first intervention involved collaboration with UNICEF, which was already implementing a TIP-informed social mobilization project with parents in some of the same settlements and communities. Although the TIP project was focused on interventions for health workers, UNICEF used the research findings to augment its work with parents and support its own interventions. A second set of interventions focused on environmental restructuring and training to address gaps around knowledge and capability. UNICEF advocated successfully for a change in legislation that would enable parents to bring their child for vaccination without any documentation and without being registered in an urban health facility. A new ministerial order was issued to clarify internal migrants’ right to vaccination. The government supported this policy change by training health workers on the ministerial order and procedures for vaccination in cases with missing records. Workers also received updated information

___________________

workers and parents), Germany (health workers), Kyrgyzstan (urban migrants), Lithuania (pregnant women), Mauritania (health workers), Montenegro (health workers), Norway (childhood vaccination), Romania (parents and health workers), Serbia (health workers administering flu and routine vaccines), Sweden (Somali community, undocumented migrants, and the Anthroposophic community), and the United Kingdom (orthodox Jewish Charedi community).

about vaccination and undervaccination. Jackson added that the interventions were piloted and evaluated before being scaled up, and they are currently monitoring and evaluating vaccine coverage data to evaluate the effect of the interventions.

IMMUNITY COMMUNITY: A COMMUNITY ENGAGEMENT STRATEGY TO BOOST VACCINE CONFIDENCE

Presented by Clarissa Hsu, Kaiser Permanente Washington Health Research Institute

Hsu presented on the Immunity Community, a community engagement strategy to boost vaccine confidence. The Immunity Community was created through VAX Northwest, a public–private partnership focused on children’s health, and implemented by WithinReach, a Seattle-based nonprofit that connects parents in Washington State with needed health resources. When VAX Northwest and the Immunity Community were formed, there was escalating concern about the increasing rate of vaccine hesitancy in Washington. Between 1999 and 2011, kindergarten vaccination exemption rates increased in most Washington counties. In 1999, approximately half of the state’s counties had rates at or below 2.9 percent, with the highest rates at 5–5.9 percent. In contrast, between 2010 and 2011, the overwhelming majority of counties had rates above 3 percent, with eight counties exceeding 10 percent. Although there was some improvement in 2011, the state passed a new policy regarding kindergarten vaccination exemptions in 2012. This policy was similar to California’s changes to vaccination exemption policies, but not as robust, said Hsu. After the new policy was implemented, Washington’s overall exemption rate dropped to 4.6 percent¹⁰ and has remained at or near that level. Despite this improvement, the exemption rates in Washington State remain among the highest for school-aged children in the United States, she noted.

Development of Immunity Community

Immunity Community was created to address growing concern about the persistent problem of vaccine hesitancy, said Hsu. Early in this process, VAX Northwest conducted a literature review and held parent focus groups, which revealed two key insights. First, although the majority of parents vaccinate, they are a silent majority in that their voices are not often heard on

___________________

¹⁰ For more information on vaccine exemption rates in Washington State, see https://www.doh.wa.gov/Portals/1/Documents/Pubs/348-246-SY2014-15-ImmunizationGraphs.pdf (accessed December 18, 2020).

this issue. Second, parents who vaccinate their children are easily activated when people talk to them about herd immunity and the problem of vaccine hesitancy. Based on these findings, VAX Northwest developed a program to engage parents by working within existing organizations, such as schools and day care centers, to spread positive messages about vaccination. Immunity Community was then created with the help of a community advisory board and BC/DC Ideas, a marketing firm that specializes in social marketing campaigns for nonprofits.¹¹ The campaign developed a large number of materials, including a parent action guide and viral images for social media.¹² The parent action guide was used to train the parent advocates and included (1) information about vaccines, (2) guidance on difficult conversations using techniques such as AIMS and motivational interviewing,¹³ (3) ideas for community advocacy, and (4) updated information resources. Based on feedback from parents in focus groups after the first year, Immunity Community developed a campaign called the Real Parents Line. They created viral images, videos, and other materials—featuring actual parents and their quotes—that were designed for parent advocates to share on social media.

Evaluation of Immunity Community

The Immunity Community program was evaluated using a logic model approach that employs multiple data-collection activities to ensure robust data, said Hsu. Evaluation activities included (1) observation and document review, (2) activity and media tracking, (3) a parent survey, (4) interviews with key informants, and (5) focus groups with parents. As part of activity tracking, parent advocates tracked their monthly activity and returned monthly reports. In the parent survey, the campaign included a pre–post, cross-sectional parent survey to examine the parent’s knowledge and attitudes. For the key informant interviews, the campaign conducted these interviews annually with parent advocates, representatives from schools and other organizations, and other relevant stakeholders. Focus groups with parents were conducted annually through organizations involved in the program. The evaluation demonstrated that the implementation of Immunity Community was successful, so the program was scaled up and expanded from 4 sites and 6 parent advocates in year 1 to 7 sites and 13 parent advocates in year 2, and then to 10 sites and 14 parent advocates in year 3. In years 2 and 3, the campaign also expanded in terms of trainings,

___________________

¹¹ More information about BC/DC Ideas is available at https://www.bcdcideas.com/about (accessed November 2, 2020).

¹² More information about Immunity Community is available at https://immunitycommunitywa.org (accessed November 2, 2020).

¹³ See Chapter 5 for more discussion of the AIMS (announce, inquire, mirror, and secure) method.

kickoff events for parent advocates, enhanced technical assistance, program materials, and web resources.

The evaluation of the campaign revealed that parent advocates had taken action to raise awareness by monitoring site vaccination rates, educating other parents, and generating conversations on social media. In monitoring vaccination rates, parent advocates played multiple roles. Although Washington State requires elementary schools, day care centers, and preschools to collect immunization information on the children who are participating in their program, the state has lacked resources to monitor and hold those organizations accountable. Immunity Community’s parent advocates stepped in to help hold them accountable by providing tools for their organizations to collect data, reviewing their organization’s records, and—on occasion—actively assisting in the collection of records. In educating parents, parent advocates organized question-and-answer sessions with local physicians, had one-on-one conversations with parents, stationed booths at science fairs, passed out merchandise associated with the campaign, and used Immunity Community’s wheel of vaccination game to educate families. Some advocates were active on social media and worked to generate conversations around vaccines.¹⁴

Effect on Parental Knowledge and Attitudes About Vaccines

Hsu presented the results of the pre–post cross-sectional surveys that looked at parental knowledge and attitudes about vaccines before and after the Immunity Community program was implemented, which found that parental attitudes had become more supportive of vaccination after the intervention (Schoeppe et al., 2017). For instance, more parents said they agreed or strongly agreed with statements such as “I am concerned about other parents not vaccinating their child(ren)” (81.2 percent pre-intervention; 88.6 post-intervention). Fewer parents agreed or strongly agreed with statements such as “Vaccines are given to children when they are too young” (31.0 percent pre-intervention; 24.7 post-intervention) and “Individual people are responsible for choosing whether or not to vaccinate their child(ren)” (70.9 percent pre-intervention; 66.9 post-intervention). Parents also expressed more confidence that vaccinating their child was a good decision and less concern about the safety of vaccines, she added. Parental attitudes also changed with respect to vaccine hesitancy, said Hsu. Overall, they observed a decrease in vaccine hesitancy from the first cohort to the second cohort. The proportion of respondents who reported being “very hesitant” decreased

___________________

¹⁴ In one example, Hsu describes a parent advocate, Allison, who had 17 conversations with other parents, sent 11 emails about the Immunity Community program, took part in planning activities, and posted 11 social media posts in 1 year of participation.

slightly, from 3.8 percent to 3.1 percent. A larger decrease in hesitancy was observed among those who were “somewhat hesitant,” which dropped from 18.8 percent to 10.9 percent. This was expected because that group tends to be more easily influenced, she noted. Overall, there was a decrease of 38 percent in the proportion of respondents who reported being “very” or “somewhat” hesitant.

Impact on Policy

Immunity Community was able to effect organizational-level policy changes, primarily related to the collection of immunization information, said Hsu. In one notable policy change, WithinReach and two of the parent advocates worked with the organization overseeing all of the state’s cooperative preschools to adopt changes to their risk management manual, including new information about immunization and reports of immunization. They also established guidelines for collecting immunization data—requiring coops to appoint one to two people to be responsible for collecting immunization data—and provided guidance about how to address a disease outbreak. This policy change alone has the potential to impact 10,000 families in Washington State each year, she noted. Immunity Community also received local and national media coverage, which was another goal of the campaign (The Bellingham Herald, 2012; Rochman, 2013).

Lessons Learned from Immunity Community

Hsu highlighted three factors that contributed to Immunity Community’s success. First, schools provided leadership and support at the organizational level. There were some schools where teachers were resistant to Immunity Community’s messages, which posed challenges for the parent advocates. Second, reports from parent advocates demonstrated the importance of strong technical assistance and support from the campaign. Immunity Community’s program staff was responsive to parent advocates, providing them with resources when they had questions or concerns. Third, a culture of active parent engagement contributed to the success of Immunity Community. For example, parents helped to spread information by using existing virtual spaces, such as online forums, and physical spaces at the school where parents tend to encounter one another. Where there was no space for advocates to have conversations, leave materials, or engage other parents, the work was more challenging. Hsu discussed the challenges they faced around recruitment, implementation, and the issues of replication and sustainability. Recruitment, which accounted for an unexpectedly large proportion of total time spent during the Immunity Community project, is a challenge because it is so time intensive. It involves building trust at the community level and

with schools, day care centers, health departments, school districts, and other organizations; then, individual parents need to be recruited. Replication and sustainability were challenging due to resource constraints. She noted that when Immunity Community attempted to implement low-resource Immunity Community models, they were unable to get past the recruitment phase.

ENGAGING WITH FAITH COMMUNITIES TO INCREASE VACCINE ACCEPTANCE AND UPTAKE IN A CHAREDI ORTHODOX JEWISH COMMUNITY

Presented by Louise Letley, Public Health England

Louise Letley, nurse manager for research and immunization operations, Public Health England (PHE), outlined the TIP project carried out with the Charedi community in North London in 2015–2016 after outbreaks of vaccine-preventable disease originated in the community and spread to other countries. The TIP project in London, one of the first such projects, was conducted to determine effective methods of increasing vaccination uptake. Guidance to the TIP approach has since been updated, improved, and further streamlined.

The Charedi Community in North London

The North London borough of Hackney is home to the largest Charedi community in Europe, said Letley. This community of strictly observant Jews was already established in London in the 1920s, but the population grew significantly during World War II as new arrivals fled the Holocaust. Membership within the community was not systematically recorded in medical records, creating a challenge for the TIP project in terms of estimating population size. At the time of the TIP survey, the community had an estimated population of 25,000–30,000. The Charedi community in Hackney had suffered recurrent outbreaks of vaccine-preventable disease (e.g., measles outbreaks in 2007 and 2013), which indicated suboptimal immunization uptake. Furthermore, the Charedi community suffered a higher burden of disease during those outbreaks than other parts of the borough. Close links with Charedi communities in other parts of the world led to the export of measles from the United Kingdom to other countries, including Israel and Belgium. The perception among health professionals was that religious or cultural objections contributed to lower immunization uptake, she added.

TIP Process in the London Charedi Community

Letley described how the TIP process was conducted to address vaccination uptake among the Charedi community in London, with community

engagement and involvement emphasized at every step of the process (Letley et al., 2018). The first step was to hold an initial stakeholder meeting and determine the TIP focus. This meeting included local commissioners and immunization service providers, public health professionals, representatives from the PHE National Immunisation Team, and experts from WHO. The Charedi community was selected as the focus of an effort to identify immunization barriers and enablers. The next step of the TIP process was to map current immunization services by examining what was already taking place in the community. Mapping revealed that some small projects designed to deliver services in a flexible way had been conducted, but these tended to be short term and had not been evaluated effectively. Next, the project analyzed available surveillance and outbreak data.

The second stakeholder meeting actively engaged members of the Charedi community, including the senior rabbi for health,¹⁵ a Charedi nurse, staff from children’s centers, and representatives from a Charedi support organization. Local commissioners and immunization service providers also continued to participate. Much of that meeting focused on an analysis of the strengths, weaknesses, opportunities, and threats of the current immunization policy, which informed a survey developed for parents. Community members contributed to the design, translated it into Hebrew and Yiddish, and checked it for cultural awareness. The survey was then distributed to parents through children’s centers and general medical practices. After collecting survey results, in-depth interviews were conducted with parents and key informants to get additional information regarding some of the results. Once that process was complete, the stakeholders from the second meeting reconvened to present the results and prioritize findings to inform the recommendations made to commissioners, service providers, and the community.

Analysis of Findings from the TIP Process

Letley described the findings that emerged from the TIP process in the Charedi community. In addition to confirmation that the uptake of immunizations was lower within the Charedi community, data analysis showed that recurring vaccine-preventable diseases were placing a burden on the community, particularly in children under 4 years of age who would be protected by the routine childhood vaccination schedule. Charedi community

___________________

¹⁵ Letley highlighted the efforts of a participant of the TIP stakeholder group, the senior rabbi for health, Rabbi Avrohom Pinter, who died from COVID-19 in April 2020. Rabbi Pinter was pro-immunization from the start and became even more committed and vocal in his views through his work with TIP. She shared Rabbi Pinter’s quote from a 2017 article in the Jewish Chronicle (Kollrin, 2017): “People don’t take immunization seriously because they’ve seen those illnesses and think ‘It isn’t that terrible.’ They don’t realise that it could kill somebody else. We have a responsibility to others as well as to ourselves.”

members tend to have more children than average, she noted. The analysis revealed that the proportion of children under 4 years of age in the Charedi community was larger than the proportion in the general national population. Data from general medical practices serving the Charedi community indicated that children under 4 years of age composed 10–17 percent of the registered community, compared to 6 percent nationwide. Thus, the Charedi community was weighted toward the younger age group, but did not have additional resources to support this population composition. A potential barrier identified was that general practice services were strained in providing immunization services to the high number of children in the community. Furthermore, parents of large families have many responsibilities and may not always prioritize the effort required to take their children to the clinic to be immunized, she added.

Although the survey was available to Charedi parents, all of the respondents were mothers, who tend to make most of decisions around health care in the Charedi community. Surprisingly, the survey did not highlight any barriers stemming from religious or cultural beliefs, noted Letley. Rather, issues related to access, wait times, and facility child-friendliness were prominent. For example, parents mentioned wanting reduced wait times and facilities featuring space to park strollers and a room for breastfeeding. Children’s centers were a popular preference for additional immunization venues. Additionally, community-specific initiatives such as Sunday immunization clinics and Charedi nurse immunization providers were popular.

Data analysis indicated unmet immunization information needs within the community, said Letley. During the in-depth interviews, participants were asked what might improve vaccination uptake or services. Responses suggested providing more information or information that was less biased through up-to-date leaflets, magnets, and wall calendars; email reminders when vaccinations are due; informal information sessions at children’s centers; posted information about the immunization helpline; receiving information from someone within the culture; and religious guidance. To improve access, respondents suggested providing walk-in clinics or clinics where it is easy to book appointments; nearby clinics or home visits; longer clinic operating hours; shorter wait times; comfortable and pleasant surroundings with space for strollers, such as children’s centers; and school immunizations.

Letley highlighted several clear messages that emerged through the TIP process. First, no evidence of any cultural or religious objection to immunization was found. In fact, Charedi community members’ views on immunization were broadly similar to the wider population, Letley said. Resolving service access issues was found to be a priority in improving uptake. Furthermore, improved recording of community membership in health records would aid in monitoring uptake and targeting tailored interventions. Collaborating with community members, including charities and community

and religious leaders, was found to be invaluable. Lastly, community-specific interventions should be evaluated for effectiveness and cost-effectiveness.

Measures Implemented After the TIP Process

Letley remarked that the TIP project among the Charedi community generated many recommended measures for commissioners, service providers, and the community to increase vaccination uptake, several of which have since been implemented. For example, medical practices agreed to send families proactive reminders by text and follow-up phone call. A digital medical record called EMIS Web is used to track patient health care.¹⁶ Children are flagged in this system, which enables providers to remind parents of vaccinations that are due for any of their children, regardless of which child is scheduled for the appointment. If a child is attending a sibling’s appointment, all possible vaccines falling within the UK schedule can be provided. For children not in attendance, reminders and alternative appointments are offered. Other measures include medical practices sending monthly data reports on vaccination activity to NHS England. The immunization providers in the borough of Hackney had a representative on the Hackney Immunisation Group, affording the opportunity to meet with public health professionals and commissioners. The practice payment process was altered to recognize special circumstances, which gave them more resources. Lastly, local authority and immunization providers explored options for community venues for immunizations, such as children’s centers and other locations that might be more accessible to parents than general medical practices.

ENGAGING WITH IMMIGRANT COMMUNITIES TO INCREASE VACCINE ACCEPTANCE AND UPTAKE IN A SOMALI AMERICAN COMMUNITY

Presented by Patsy Stinchfield, Children’s Minnesota

Stinchfield described efforts to engage with the immigrant Somali American population in Minnesota to increase vaccine acceptance and uptake after anti-vaccination messaging affected the MMR coverage rate in that community and outbreaks of measles occurred.

Somali American Population in Minnesota

Drawing from Ahmed Yusuf’s book, Somalis in Minnesota, Stinchfield described the immigration trends of Somali refugees in the state (Yusuf, 2012).

___________________

¹⁶ More information about EMIS is available at https://www.emishealth.com/products/emis-web (accessed November 6, 2020).

Minnesota has historically welcomed refugees and immigrants, its original settlers being German, Irish, and Swedish immigrants. In more recent decades, immigration patterns have shifted to an influx of refugees from the Hmong community in Laos, Liberia, and Somalia. Somali refugees began settling in Minnesota in 1993 and continued to increase because of the availability of jobs, the level of community acceptance, the quality of life, and a strong, welcoming social service system. Stinchfield noted that the Somali American community is an oral society that relies on word of mouth; thus, word spread from community members and religious leaders to Somali refugee camps that Minnesota was a welcoming state. By 2017, Minnesota had the largest Somali American community in the United States, comprising more than 52,000 people.¹⁷ In 2006, the apex thus far of Somali immigration, the state welcomed more than 3,600 Somali immigrants, compared to less than 50 in 2018. Stinchfield noted that the Somali American population is now well-integrated into the community at large, and members of the community have held local and national political offices.

Declining Measles Vaccination Rates and Subsequent Outbreaks

Stinchfield emphasized that historically, Somali immigrant vaccination rates matched those of the general population. For example, in 2004 in Minneapolis-Hennepin County, 92 percent of Somali American 2-year-olds were immunized compared to 88 percent of all county residents. Although these rates have remained steady for the general population, there was a shift in 2007–2008 in the Somali American community. The idea of the MMR vaccine causing autism began to infiltrate the community, and then a local anti-vaccine group began actively promoting that message to Somali Americans. In 2010 and 2011, Andrew Wakefield—who published the now debunked, small study that originally claimed a connection between MMR vaccines and autism—was invited to Minneapolis by the anti-vaccine group (Bhatt, 1998). Rather than inviting Wakefield to meet with the entire community, the group hosted him at a closed-door session with local imams. This resulted in mosques spreading the message that parents should not give their children the MMR vaccine, said Stinchfield. By 2014, the MMR coverage rate in the Somali American population in Hennepin County had dropped to 42 percent.¹⁸ Work is still ongoing to close the vaccination gap between Somali American and non-Somali American children in Minnesota, but the

___________________

¹⁷ More information about 2017 American Community Survey estimates is available at https://www.census.gov/programs-surveys/acs/news/data-releases/2017/release.html (accessed November 5, 2020).

¹⁸ More information about measles vaccination in Minnesota is available at https://www.health.state.mn.us/diseases/reportable/dcn/sum17/measles.html (accessed November 5, 2020).

efforts are currently hampered by the impact of the COVID-19 pandemic on immunization rates.

Children’s Minnesota foresaw the possibility of an immunization rate drop and worked to avert such a scenario in the early 2000s. In 2002, the organization hired a Somali American film crew to make a video with Somali American providers. Children’s Minnesota also had clinical leaders reach out to Somali American families in their own language. Yet, the efforts of Children’s Minnesota and the public health department were not successful in warding off an outbreak. Stinchfield said she talked with many Somali American parents after the outbreak, asking them for their thoughts on what led to the situation. She quoted a mother who did not know why she was not supposed to vaccinate her child, and she later regretted it when her child was admitted with measles in 2017:

In 2011, a study investigated whether Somali American parents were more likely than non-Somali American parents to refuse childhood vaccinations—particularly MMR vaccines—and to determine what factors influenced the decision not to vaccinate (Wolff and Madlon-Kay, 2014). Somali American parents were significantly more likely (35 percent) to believe that autism was caused by the MMR vaccine compared to non-Somali American parents (8 percent). Correspondingly, Somali American parents were more likely to refuse the MMR vaccine than non-Somali American parents. The study suggests that beliefs that the MMR vaccine causes autism contributed to an immunization gap between Somali American and non-Somali American children, noted Stinchfield.

Measles Outbreak in the Somali Population

By April 2017, MMR vaccine rates had dropped dangerously low among the Somali American community in Minnesota, leading to an outbreak of measles. Stinchfield said that young children—mostly preschool-aged Somali American children—began presenting at emergency rooms with fever, rash, cough, conjunctivitis, coryza (runny nose), and had the appearance of a person with a histamine reaction. Stinchfield noted that the rash can be difficult to see on darker skin tones, so providers need to look carefully and palpate for it.¹⁹ A total of 75 cases of measles were reported in the 2017 Minnesota

___________________

¹⁹ Stinchfield stated that with the COVID-19 pandemic affecting vaccine rates, medical professionals should keep measles in the forefront of their minds when examining children.

outbreak, 66 of which were in Hennepin County.²⁰ The vast majority of infected people (71) were children, 61 of whom were Somali American. Of the 75 cases, only 3 individuals were fully vaccinated for MMR and 68 were confirmed to be unvaccinated. Minnesota Children’s hospital treated 53 patients with measles, with a median age of 3 years (range: 10 months–14 years). Twenty-one of these children had to be hospitalized, with length of stay ranging from 2 to 17 days and averaging 4 days. One child had to be re-hospitalized for pneumonia. Fortunately, no lives were lost.

In an effort to publicize the outbreak, Stinchfield asked a health reporter to document the outbreak from the individual families’ perspective. The story received national and worldwide media coverage (Mele, 2017; Sun, 2017). During the same period, England and Sweden were experiencing similar outbreaks in their own Somali refugee communities (Tomlinson and Redwood, 2013). As in Minnesota, the health beliefs among community members were influenced by fear and mistrust, and concerns about autism were linked to decreases in MMR coverage.

Outbreak Response and Interventions

Stinchfield highlighted the value of building trust and leveraging systems communication in addressing a measles outbreak. For example, she suggested partnering with the media and ensuring that both the message and the messenger are carefully considered. Local-level communication should be tailored to social and cultural sensitivities, with information delivered in the community’s language by a trusted messenger or community leader. Location of messaging is another consideration; it is important to determine where the community prefers to receive information, such as a mosque or another meeting space where participants are open to learning. Opportunities for one-to-one communication include the hospital or clinic where a child is being treated for measles.

To respond to the Minnesota outbreak, Stinchfield was part of a team that was formed to go from mosque to mosque to talk to imams. At these meetings, Stinchfield spoke about MMR, the American Academy of Pediatrics chapter president talked about the danger of measles, and another pediatrician spoke about autism. They separated those topics and discussed them with these trusted leaders in the community. In addition, Children’s Minnesota conducted a series of interventions to connect with the Somali American community. The organization developed a Somali American employee resource group that continues to thrive today, said Stinchfield.

___________________

²⁰ More information about the 2017 measles outbreak in Minnesota is available at https://www.health.state.mn.us/diseases/reportable/dcn/sum17/measles.html (accessed November 5, 2020).

Furthermore, they sought feedback from Somali American clinicians from multiple institutions about how to improve their response efforts. They have also worked to build and foster trust with community imams and hosted public Ramadan Eid dinners to engage with the community. Informational materials have been translated to make them accessible to more people and Children’s Minnesota is working to build a hospital staff and leadership that reflect the families the organization serves. In terms of public health interventions, the state health department conducted outreach via Somali American imams and Somali American health care providers, held multiple Somali American community meetings with health leaders and legislators, and ran culturally appropriate ads in Somali American radio and television outlets. Additionally, they ran newspaper ads and were interviewed for articles.

Evaluating the Effect of the Interventions

To evaluate the effect of the interventions to increase vaccination acceptance and uptake in the Somali American community in Minnesota, the state department of health conducted a survey of 300 parents whose children were previously unvaccinated, but received MMR immunization during the measles outbreak.²¹ They were asked what made them change their minds and how the vaccination affected their children. The vast majority of the respondents (95 percent) cited fear of measles as the motivation for vaccinating their children. About 80 percent reported having no concerns about how the vaccination affected their children and, for those who reported side effects, most were mild (e.g., fever, mild rash, pain at injection site, crying after the vaccine) and not long lasting. The majority of respondents said they trusted their health care providers as a source of information regarding vaccines.

Lessons Learned and Ways Forward

Stinchfield described lessons learned through this process of engaging the Somali American community around vaccination. For example, fear and mistrust are major obstacles to immunization coverage. Tapping into the Somali American oral communication method was a valuable strategy, as was using Somali-speaking health care professionals to engage with the community. Imams are leaders in their communities, so gaining their trust by visiting them, listening, and sharing can create powerful connections. Additionally, working with public health Somali American outreach workers can be an effective partnership. Communication modes should be used at all

___________________

²¹ More information about the Minnesota Department of Health’s “Reporting Back to Health Care Providers: MMR Survey” is available at https://www.health.state.mn.us/people/immunize/mmrsurv (accessed November 5, 2020).

levels, from systemic to local/social to individual. Stinchfield remarked that “outbreaks change minds.” The mother of a child who was on a ventilator for 15 days during the 2011 outbreak said she did not know that a child could get so sick from measles. She asked Stinchfield to share a picture of her child in the hospital—connected to the ventilator and other equipment—so parents could learn the danger measles can pose to their children. The mother said, “Please let them know that we must care for them.”

Moving forward, Stinchfield suggested that measles be kept at the forefront of health professionals’ minds in order to intervene early, as well as increasing the use of vitamin A in the management of measles care in the United States. Children’s Minnesota has conducted studies indicating that improving vitamin A status can reduce the risk of serious complications (Stinchfield and Orenstein, 2020). She also noted that as the COVID-19 pandemic is decreasing vaccine rates worldwide, creativity will be needed to safely immunize children and maintain MMR vaccine rates. Otherwise, measles outbreaks could compound outbreaks of both COVID-19 and influenza once international travel resumes.

DISCUSSION

Obregon opened the discussion by highlighting the value of culturally centered interventions tailored to specific contexts and settings to address similar issues across different groups. He added that COVID-19 is posing challenges to other vaccination efforts.

Ensuring Immunization Card Availability and Use

Noting that surveys show large discrepancies between parents’ vaccination recall and card validation, Obregon said that this is largely due to parents not having vaccination records for their own children on hand. Vaccination cards for parents and clients are not sustainably available in many settings, especially those with weak health systems or other challenges. Obregon asked Jalloh how professionals can work collectively to ensure that vaccination cards are available and used. He added that this issue poses particular challenges to migrant communities, such as those moving from Venezuela to other countries in Latin America.

Jalloh responded that this is a complicated issue that warrants both low-tech and higher-tech but simple solutions that can be implemented as appropriate in different contexts. He said that in Bangladesh, the Rohingya population faced issues similar to those facing Latin American migrant communities. Because it was not possible to ascertain the level of immunization in the Rohingya community, a serosurvey was conducted to estimate the baseline immunization coverage of Rohingya refugees coming into Bangladesh. Jalloh

noted that serosurveys are not a sustainable method of tracking immunization coverage rates, so technology-based solutions should be explored. Innovation is already taking place in this area, such as the capability to insert copies of the vaccination card into people’s phones. However, phones can be lost or damaged just as vaccination cards can. Electronic dual-purpose bracelets are a simple and low-tech solution being tested in Nigeria; they can retain the vaccination record and also provide reminders by lighting up when immunizations are due. In addition to exploring such innovations, Jalloh suggested the need for ways to improve the durability of home-based, paper-based cards because once cards become damaged, people are more likely to lose them.

Maintaining Engagement with Religious Leaders

Obregon asked how religious and faith leaders can be engaged in a sustainable way to maintain their ongoing alliance in addressing vaccine resistance and boosting confidence. Noting the positive experience during the TIP project in the Charedi community, Letley remarked that the rabbi who led health efforts was very engaged in the project from beginning to end. However, this rabbi made it clear that having one religious leader who is engaged does not necessarily mean the synagogue will indefinitely maintain the same level of engagement. Rather, it warrants a continuous process of engagement and reengagement over time, which is a time-consuming process that adds a degree of difficulty to ensuring sustainability. Another potential barrier to engagement is the potential for “one-offism”—that is, focusing exclusively on vaccination in communities that face many other issues of concern. Letley said that attaining and sustaining community engagement can be challenging, but it is helpful to engage community leaders from the outset and ensure that they remain involved at all stages. For instance, Charedi community members provided cultural awareness training to health care professionals to educate them about community beliefs, practices, and religious holidays. This cultural awareness enabled the health care providers to increase cultural sensitivity and avoid scheduling appointments on days of religious services or holidays.

Stinchfield said that in working with the Somali American population in Minnesota, her team came to appreciate that imams are not a homogenous group but individuals with diverse opinions and thoughts. When the measles outbreak began and providers began to investigate why half of Somali American parents were vaccinating their children and half were not, the opinions of imams were found to be highly influential. Families whose imams promoted vaccination tended to feel it was important to vaccinate. She added that personal experiences also contribute to vaccination attitudes, because some parents had seen measles in Somalia before coming to the United States and recognized its consequences. Stinchfield surmised that when Andrew

Wakefield was invited to speak with the imams in the Somali American community, their opinions about vaccinations were likely mixed before hearing him. However, Wakefield’s charismatic message was effective in convincing many of them to discourage vaccination. Some of the imams told her that it was the first time a doctor had ever spoken with them. Regardless of whether this was strictly accurate, she acknowledged that many imams likely felt heard for the first time, because Wakefield made them feel special and was billed as a famous person—despite the fact that he was no longer a physician and his claims were not truthful. This demonstrates the importance of having trusted providers within communities to share accurate information, she noted. In addition to working with families individually to build trust and develop one-on-one relationships, broader efforts should focus on engaging the entire community, including both patients and providers, within a larger system of communication. Stinchfield noted that when they began investigating the reasons why measles outbreaks were occurring among Somali heritage communities across the world, they discovered that there is an international weekly phone call for imams in which a variety of topics are discussed. False information about MMR causing autism was being shared and quickly disseminated through this platform to Somali heritage communities worldwide. She suggested that similar widespread communication methods could be used to dispel myths and provide accurate information to help increase vaccination coverage.

Impact of COVID-19 on Vaccination Efforts

Obregon asked whether the COVID-19 pandemic is shaping conversations around vaccines for other infections. Jackson noted the effect it has had on childhood vaccination programs. For example, in Kyrgyzstan, the childhood vaccination program was suspended for 3 months. Bosnia and Herzegovina have had ongoing challenges with low vaccination coverage rates that improved somewhat after a TIP project was conducted there, but rates have plummeted again during the COVID-19 pandemic, she added. Hsu said that Washington State has had a decline in childhood vaccination rates during the pandemic. The overall rates are now recovering, but there is concern that the coverage rates for the large Somali American community in the state will not recover at the same rate. Hsu and colleagues are interested in conducting work to understand community-level responses to the COVID-19 pandemic and vaccination campaign. However, it is challenging to have conversations with specific communities without knowing the shape of the eventual COVID-19 vaccine (e.g., side effects, recovery time). She suggested the possibility of adapting a program like Immunity Community to respond to the concerns of specific communities regarding a COVID-19 vaccine in particular or childhood vaccines more broadly.

COVID-19 Vaccine Outreach

Obregon asked about communication strategies for working with closed immigrant communities in the United States on contact tracing or COVID-19 vaccine hesitancy. Stinchfield suggested having community members who speak the native language be the primary channels of engagement and communication. Ideally, those individuals would be employed by public health—or at least partner with public health—to build trust for these efforts among the community.

Obregon asked how lessons learned from previous experiences can inform efforts to introduce the COVID-19 vaccine. Jalloh responded that work should begin to anticipate issues, start engaging communities, and use existing structures. The “spare tire” approach—that is, only initiating community engagement once a problem arises—can be avoided by initiating planning and engagement efforts as soon as possible. Engagement increases as a function of the extent to which social mobilization structures are engrained and sustained within a community and used to address crosscutting issues, he noted. When health providers only come to a community to discuss specific public health objectives, a top-down method is being used.

The organic, bottom-up approach to engagement that extends beyond a specific epidemic or public health emergency is more robust than a top-down, narrowly focused approach, he added. His work in West Africa highlighted the value of building on existing structures, continuing dialogue, maintaining engagement, and bringing communities into conversations. People often have similar objectives that they approach from diverse perspectives founded on different values, which need to be understood so communication strategies and other efforts can be aligned with those values, he added.

Letley said that in England, they are preparing for the COVID-19 vaccine by developing a comprehensive communication strategy and working with high-risk groups. They are also conducting qualitative work and surveys to understand how people feel about the vaccine, including their fears about potential exposure when receiving the vaccine in a health care setting or other distribution point. Obregon emphasized that engaging communities and parents should be a continuous focus that is embedded in the system, rather than occurring post hoc when a crisis happens.

Reflections on Session 3

Walter Orenstein, professor and associate director at the Emory Vaccine Center, reflected on the workshop’s third session. He commented that in addition to immunization mandates, there are myriad ways that the legal system can be used to enhance access, increase convenience, and remove barriers to vaccination, as demonstrated by the Vaccines for Children program

and efforts to increase vaccine delivery by pharmacists to broaden availability. Noting that no legal challenge to overturn a school mandate law has been successful thus far, he suggested those types of mandates should be used not as the starting point, but as a strategy to catch up on school children whose vaccinations have been delayed. He highlighted several lessons learned from the removal of nonmedical exemptions from vaccine mandates in California, including the importance of (1) assuring valid contraindications, (2) avoiding grandfather clauses that allow unimmunized children to stay enrolled, (3) enforcing those laws, and (4) working with provider groups and other stakeholders to bolster the political will to move these initiatives forward.

Orenstein remarked on strategies that were presented to promote parent engagement and improve parent–provider communication to shape vaccination behavior. Parental incentives are a useful tool, with some early data suggesting that the immunization bracelet being used in South Asia is promising in terms of effectiveness. Rather than assuming that providers will communicate with parents, providers should be appropriately trained in a variety of communication methods (e.g., face-to-face conversations, social media engagement) to help overcome vaccine hesitancy. Orenstein highlighted several strategies for improving communication, including the use of the presumptive approach and motivational interviewing. Rather than instructing parents that they must vaccinate their children, motivational interviewing engages with parents and reassures them that the provider understands and empathizes with them, thus building trust.

Social mobilization through engagement with individuals, communities, and trusted messengers is critical in increasing vaccine coverage both nationally and internationally, said Orenstein. He suggested testing out various communications methods rather than operating on assumptions about what is needed. Jackson’s presentation on TIP demonstrates that services can extend beyond a one-size-fits-all approach, he added. Individuals within communities can be engaged as effective messengers, as Hsu described in Washington State. Letley’s presentation highlighted how reaching out to trusted individuals within a religious community can help engage that community. Stinchfield outlined methods of addressing perceptions via reaching out to individuals and groups who are trusted within a community.

Orenstein concluded that effective vaccines have been developed and recommended, but they are only of benefit when they are administered. During the current COVID-19 pandemic, professionals must work together to advocate not only for resources to develop COVID-19 vaccines but also for resources to ensure these vaccines are delivered to the populations for whom they are recommended. He noted that even vaccinees benefit from high vaccination coverage rates because no vaccine is 100 percent effective; when everyone is vaccinated, it protects people who have vaccine failures.