Discussion
Evidence suggests that community leaders and other community helpers can strengthen social connections within communities and act as a key conduit between official government services and community members.35 36 These dynamics have historically been understood through a social capital lens, where trust is shaped by social networks, prior interactions and ongoing engagement within communities.37 Within this framework and what we identified in our work, we found that trust in community leaders and organisations is not assumed but develops through repeated contact, familiarity and perceived reciprocity. Despite its wide utility, the social capital theory has notable limitations, including conceptual ambiguity, measurement difficulties and concerns about causality. Furthermore, social capital can have unintended negative effects, such as reinforcing social exclusion or privileging already well-connected groups, which is particularly relevant when considering diverse and marginalised communities.38
Our findings indicate that trust in community leaders, CBOs and government agencies is closely linked to individuals’ likelihood of seeking help or information and to the frequency of interaction during a health crisis. Rather than being inherent to specific actors, trust appears to emerge through engagement, suggesting that individuals who are already connected to community structures are more likely to view these actors as credible and trustworthy. This aligns with diffusion of innovations theory, where trusted individuals and organisations act as key channels through which information and behaviours spread within social networks.39 This finding also reflects similar results to research in civic engagement in Australia and social capital in Norway, where group membership, specifically the number of group memberships per person and to a lesser extent intensity of involvement, are both important predictors of civic engagement.40 41 This is an important finding and a reminder to governments that they cannot assume that multicultural community members align with community organisations or leaders simply because an emergency is occurring. It also reinforces the need for CBOs to invest in outreach and engagement activities both before and during emergencies, including efforts to better understand the priorities and needs of the communities they serve. Such engagement is central to building the relational foundations required for effective crisis communication.
Perhaps unsurprisingly, we found that arrival in Australia ≥20 years ago was a significant predictor of higher trust in CBOs, which may reflect the accumulation of social capital over time. Longer-term residents may have had more opportunities to establish networks, engage with community organisations and develop familiarity with local systems. This finding highlights the importance of targeted engagement strategies for newer arrivals, who may have fewer established connections and therefore lower baseline trust. To ensure meaningful and relevant responses during emergencies, the WHO advocates for early and sustained community engagement, including fostering partnerships that enhance trust and support public health measures.42
Among participants, there were varying levels of trust in different information sources, with most reporting the highest trust in the Australian Government and medical professionals and the least trust in religious leaders. Only among Pasifika participants did we observe a different pattern, with lower trust in doctors and higher trust in religious centres and leaders. Notably, trust in community groups and local councils was a significant predictor of trust in both CBOs and community leaders. From a social capital perspective, this suggests that trust may transfer across interconnected networks, where confidence in one local actor reinforces trust in others within the same ecosystem. We acknowledge that our study was conducted in 2023 and not at the height of the COVID-19 pandemic, which may have influenced responses. However, these findings align with previous research and broader evidence indicating consistently high levels of trust in healthcare providers across settings.35 36 These findings also align with a recent systematic review, which examined 41 studies on previous pandemics, epidemics and global outbreaks in the 21st century. While the authors identified inconsistencies in public trust towards the government and the media across multiple countries over time, trust in healthcare providers was generally reported to be high, with a few exceptions.43
There is limited COVID-19-specific research examining patterns of trust and the factors shaping community confidence in CBOs and community leaders. Existing qualitative and quantitative studies on information preferences during the pandemic have largely centred on the roles of government agencies, healthcare providers, social media and family or friends. Although many accounts describe CBOs as having played a critical role in supporting the COVID-19 response, there is comparatively little empirical evidence to substantiate these claims. For example, Dada et al highlighted the role of community partners, including Black faith-based leaders, in the COVID-19 vaccine programme as part of a review of strategies to promote COVID-19 vaccine equity. Still, there are no data to support the statements being made.44 Similarly, in a paper exploring the roles of religious communities during the early stages of COVID, the authors highlighted that a possible solution to vaccine hesitancy was to involve religious authorities in health awareness and vaccine promotion. Again, there was no evidence cited about the level of impact of these leaders (even in events before COVID) that supports the strong statements about their role in promoting vaccine uptake and reducing misinformation.45 In bringing attention to this issue, the point is not to criticise the work that has been done, but rather to ensure we do not make assumptions about emergency planning or implementation that do not align with community needs. In our study, community leaders or faith leaders may not play as critical a role as other actors or organisations in some ethnic minority communities. For example, the Chinese in particular do not necessarily see themselves as having ‘community leaders’ at all. However, for other groups, it seems clear that faith leaders are recognised as instrumental to communication efforts during an emergency, especially for those aged 35–44 who had the strongest trust in community leaders. Understanding these nuances across different cultural and linguistic groups is critical. Otherwise, governments, health organisations, or emergency management organisations may rely on community actors who may not have the hoped-for significant influence in their communities. It also means that in the period between emergencies, it’s important we move beyond assumptions about which actors diverse communities trust, and endeavour to bring to light perceptions, lived experiences, and other reliable data to support decisions about which actors might effectively assist during crises. Future studies should explore this finding in greater depth and across a broader range of population groups to confirm it across cultures.
As Wong et al suggest, CBOs and their staff are ‘well-equipped with the cultural acuity, language capacity and familiarity with local norms to improve structural gaps’.46 Beyond positive reports about CBOs, findings from this study also show that trust in a community worker who speaks the same language or is from the same community group predicts higher trust in CBOs. They lead in providing culturally responsive, in-language resources, often filling a void left by governments. Their hallmark is often the flexibility they can take to support community outreach. However, it is important to recognise that these organisations may face the same issues related to community pushback and hesitancy that governments and health officials experience. To reduce these issues during future health emergencies, CBOs must establish themselves as credible response partners during non-emergency periods. One suggestion to assist this process is to conduct a Data Walk.47 In this event, CBOs can hear from their communities about the challenges experienced during COVID-19 to elicit feedback on how the CBO supported efforts and to brainstorm solutions to enhance communication and engagement during future emergencies. Having discussion prompts and parameters for the conversations can facilitate this process. Not only will an event like this identify the strategies that the CBOs need to consider in the future but it will also help build authentic relationships and trust with the community as they will see these activities as promoting bidirectional communication. To support these organisations in preparing for future emergencies, governments must invest in soft infrastructure during non-emergency conditions to ensure CBO’s capacity to support emergencies/crises.48
Early in the pandemic, Gilmore et al reviewed community engagement strategies used for infectious disease prevention during epidemics and highlighted three key issues in high-resource settings: weak pre-existing community engagement structures, a limited definition of leadership roles and inadequate training and resources for community leaders and helpers.49 Picking up on the issue of definition, in this study, we used ‘community leader’ to explore the interaction between community members and those postulated to play a role in the distribution of information and public health messages. This term was adopted in accordance with the nomenclature used in Australia’s pandemic response plan and international plans, including those of the WHO. A ‘community leader’ is traditionally understood as someone who is seen as the ‘model representative’ of any given community. However, based on conversations about our findings with stakeholders and peer researchers, it has been suggested that in the scope of the response to COVID, the use of the term ‘community leader’ was too narrow and may not have sufficiently allowed for the capture of all the actors within a community who played a role in supporting their communities. This issue has probably influenced the data collected regarding the proportion of community members who identified as engaging with a ‘community leader’. We likely recorded a low engagement rate because of this narrow term. Subsequent conversations have highlighted that some communities may not necessarily have identifiable leaders within their communities (which may be defined by geography) or that others in the community may have been playing a role. Based on these conversations, our team has decided to use the term ‘community connector’ in future discussions and research to capture the broader scope of actors who could be assisting during a health emergency. It is important to capture the most relevant term, as it has implications for future activities during pre-emergencies and emergencies. Focusing first on preemergency periods, our findings suggest that changes in the language in national pandemic plans (or strategies for other emergencies) may be needed to ensure that the focus is not just on the potential role of so-called community ‘leaders’ but instead highlights the need for governments to engage with, train, fund and support a broader group of actors. Future research should be conducted to validate these implications and inform future changes.
Our study has several other limitations. Participant recruitment was conducted through peer researchers and CBOs, which may have limited the survey’s geographic scope to those within known networks. To reduce non-representative sample bias, the survey was available in the language to support a wider range of participants completing the study. It was available via hard copy and online. However, we acknowledge that the survey results are likely not fully representative of the six communities we focused on, as we were unable to achieve proportional quota sampling for each group. Because of this, some findings yielded wide CIs, which may limit generalisability. Future research should be conducted with more participants to provide stronger evidence of the relationship between trust and actor. Additionally, the findings may not be generalisable to other ethnic minority communities in Australia or elsewhere. In the framing of ‘community’, there was a focus on geography and on attempts to represent shared values, customs and assets. However, defining community was more challenging in the larger cities and urban centres due to a lack of cohesion, high density, informal settlement conditions and high mobility.50 For example, Arabic communities range in country of origin, cultural beliefs, religions, etc.