Bringing the Social into Vaccination Research: Community-Led Ethnography and Trustbuilding in Immunization Programs in Sierra Leone

"Inspiration for bringing the social into vaccine research can be found in anthropological work on local engagements with scientific research, including vaccine trials, in low-income settings....From these perspectives, we can challenge hesitancy frameworks that focus on individual responsibility, looking instead to how social, political, and economic factors affect vaccine acceptance, anxieties and refusal."

Trust is understood as important for understanding the complex social phenomenon of vaccine hesitancy. However, trust is an elusive idea, and efforts to capture its nature and drivers have been criticised as being too narrowly focused on high-income countries. Context is important: Situating concerns and hopes around vaccination in broader social analysis supports engagement with the interplay between structural realities and individual decisions, all of which influences demand for vaccines. It is increasingly acknowledged that interventions to strengthen vaccination confidence and coverage can only be sustainable if they are led by local knowledge, responsive to the experience of affected individuals and communities, and directly shaped by those closely involved in the delivery of vaccines on the ground. This paper presents findings from a project aiming to integrate community-led ethnographic research into the development and implementation of strategies to address vaccine hesitancy and other immunisation challenges in Sierra Leone.

Lying on Sierra Leone's North-Western border, Kambia District has ranked lower than other parts of the country in terms of vaccination coverage. In January 2019, Kambia was affected by a measles outbreak and, like the rest of the country, was adversely hit by the 2014-16 Ebola outbreak. During the latter, the district became the site of an Ebola vaccine (EBOVAC) trial, whose collective research design and implementation ignited a conversation about the importance of identifying reasons behind Kambia's relative lower rates of vaccination and directly addressing the specific issues that may be determining hesitancy in different communities.

It was in this context that the pilot project was conducted in Kambia District between September 2019 and March 2020 through a partnership between the London School of Hygiene & Tropical Medicine (LSHTM), University of Bath, and the Kambia District Health Management Team (DHMT), the local representation of the Ministry of Health and Sanitation. In brief, the project trialled a social science training for community health workers (CHWs), who had been active in supporting the Ebola response by taking a key role in community engagement, amongst other activities. As part of their work, CHWs also carry out social mobilisation and health education around vaccinations. They are the closest health workers to community members, who tend to have trusting relationships with them, making them well suited to act as community researchers exploring citizens' experiences and perspectives of vaccination.

In August 2019, members of the DHMT community engagement and Extended Programme on Immunization (EPI) teams drew up a list of five border towns or villages that had lower coverage levels or where the engagement team had experienced what they perceived to be community avoidance or refusal during immunisation drives. Ten CHWs associated were invited to a one-week training in September 2019, framed as an introduction to qualitative research with an emphasis of ethnographic perspectives and participatory methods such as participant observation, participatory power mapping and rumour tracking, focus group discussions (FGDs), and key stakeholder interviews.

After the training, the CHWs then conducted participatory, ethnographic research on vaccination challenges in five communities and worked with these communities and district public health officials to translate these findings into new community engagement and communication strategies to improve vaccine confidence. In February and March 2020, the CHWs conducted another round of qualitative research to assess community perceptions of the new activities and messages they had trialled the previous month. CHWs then tested key aspects of this strategy and consulted community members on the appropriateness of the recommendations.

The CHWs' research generated a wide range of empirical findings around experiences and perspectives of vaccination in Kambia's border communities. These can be summarised in three broad thematic areas:

1. The impact of prior experiences with the health system on (mis)trust:
   • Listening to communities' concerns, while also being aware of working conditions in the health centres, CHWs were able to spot tensions and communication gaps between healthcare workers and their communities. An example of this was the fact that nurses were volunteers and at times had to supplement their income by "selling" vaccination cards. These structural issues were exacerbated by experiences of poor service, such as patients feeling they were looked down upon and discriminated against based on their appearance or lack of education. Resultant feelings of mistrust in the healthcare system and healthcare workers were directly connected to explanations for avoidance of routine immunisation and vaccination campaigns.
   • Emphasis was placed on expressions of respect when respondents praised individual nurses, citing instances when they would help mothers read vaccination cards and translate appointments into the Islamic calendar. These reflections were illustrative of the possibilities for building trust through interpersonal relations. In sum, in the particular interactions recorded by the CHWs, trustworthiness can be summarised by referring to three characteristics: social proximity, reliability, and respect.
2. Relevance of livelihood strategies and power dynamics for vaccine uptake and access:
   • CHWs' community profiling offered more nuanced descriptions of cross-border relations and movement patterns, highlighting mismatches between borderland livelihoods and the organisation of immunisation campaigns. For example, their research revealed that travel to trade in the nearby Guinean town of Pamlap meant that parents often had to miss dates for vaccination.
   • The power mapping workshops made clear that in each community, different kinds of people had the power to influence opinions and behaviours around vaccination. Having a formal position did not mean that the individual was trusted. Conversely, some of those with most sway over public opinion had no position at all and as such were missed out during mobilisation campaigns. Power was also important to consider at household level, as CHWs observed that while women often took care of their children's health and in some cases wanted to bring them to the clinic to be vaccinated, they were not always able to make final household decisions.
   • Communities' recent histories also mattered. For example, memories of Ebola were invoked in comparisons with immunisation drives, as people criticised public health efforts led by "strangers" from outside the community. CHWs noted that at the beginning of the research process, their questions and observations in villages raised concerns that it may be foreshadowing the arrival of a new disease.
3. The contextual nature of knowledge around vaccines: Contrary to their initial expectations, the CHWs' research showed that most people they interacted with actually found vaccines to be important and valued. Vaccine-specific concerns were primarily linked to fears of side effects, which were tied to broader concerns about the implications of having to take a member of the family to the health centre because of the experiences and mistrust noted above. These insights undermined assumptions, shared both by CHWs and district public health officials, that rumours and concerns or avoidance of health centres were caused primarily by a lack of knowledge. In contrast, their findings showed the importance of considering other challenges for uptake, including those that did not reflect vaccine hesitancy per se, but rather more contextual or structural issues, such as discrimination associated with accessing healthcare.

In addition to generating new understandings of what vaccination challenges were in the first place, the research process strengthened CHWs' research and writing skills and brought them closer to the community where they worked, facilitating different kinds of interactions. One CHW said that while he knew the community where he worked very well and lived close by, the relationship he had with them was changed by the nature of participant observation and that this had helped him build trust with residents.

The pilot also highlighted the practical challenges to meaningfully co-designed research. For instance, CHWs encountered difficulties in arranging FGDs and in navigating their role of representing both the interest of their fellow residents and those of public health officials. In some instances, the CHWs reported being called "journalists" by their communities, expressing concerns they were there to "gossip". The process of building trust through daily encounters and participation was key in encouraging respondents to open up to the CHWs.

In December 2019, the team considered how some of the findings from the CHWs' research could be translated into actionable changes, presenting recommendations to the DHMT, including, for example:

• Conduct power mapping activities prior to a vaccination campaign to ensure the right leaders are involved in community engagement efforts.
• Consider the timing of campaigns and outreach (both time of the day and time of the year) around the farming cycle and trading commitments and to consider sending vaccination teams to key border crossings.
• Provide stipends for volunteer health workers to do outreach and tackle the problem of frequent drugs stock outs, as part of larger efforts to improve community members' health-seeking experiences.

The second set of recommendations were more directly targeted at short-term efforts to improve community engagement. The paper focuses on two of these strategies as case studies. In brief:

• To address the issue of a lack of trust in the health system, the CHWs proposed, and then facilitated, interface meetings between community members (identified through the power mapping) and health staff. The CHWs emphasised they wanted community members to have an opportunity to voice their anxieties around visiting the health centres, while also encouraging the health staff to put their own challenges on the table. Interviews with both health staff and community members after the meetings suggested they were found to be valuable on both sides, and respondents proposed that these kinds of meetings should be convened more frequently. While many of the structural issues raised in the meetings could not be addressed at the community or even District level, simply offering a forum to have frank and respectful conversations was a starting point for efforts to rebuilt trust.
• In light of the research finding that traditional healers or birth attendants were shown to be relatively trusted, the second strategy was focused on leveraging local concepts, knowledge, and experience to reframe discussions around vaccination. For example, CHWs began using the concept of "tarma" (or adapted it to other terms that specific communities were more familiar with like "mabukor"), a local word to refer to processes of protecting oneself from witchcraft and its manifestations through rituals that are common in Kambia District. CHWs used this notion to leverage existing notions of prevention to start a conversation about how vaccines can help prevent infectious diseases, or "hospital sicks".

The paper's discussion section divides lessons from this project into two categories: (i) substantive findings that contribute to more contextual understandings of the social dynamics of vaccine confidence and hesitancy, including an insight into local drivers of trust (this category responds to calls for deepening qualitative evidence on these dynamics and to continue expanding the geographical focus to include experiences from the Global South); and (ii) process findings about the role of CHW-led research in strengthening community engagement strategies for vaccination.

The researchers note: "Whilst increased uptake is an undeniably important goal, we argue that the most significant lessons from this project emerged from the process of directly involving CHWs as community focal points and public health officials in knowledge production through immersive, participatory research. The project was co-designed from the start", which enabled community-led data collection, analysis, and operationalisation processes that informed activities that were responsive to the experiences and concerns of residents of the areas where engagement was to take place. "The research process itself created spaces to stage the kind of dialogue that we found to be necessary to start rebuilding trust between communities and health workers."

In conclusion: "There is scope for broader application of a community-led ethnographic approach will help redesign programming that is responsive to local knowledge and experience. Involving communities and low-cadre service providers in generating knowledge and solutions can strengthen relationships and sustain dialogue to bolster vaccine confidence."