Exploring the Role of Community Engagement in Improving the Health of Disadvantaged Populations: A Systematic Review

"The findings suggest that CE models can lead to improved health and health behaviours among disadvantaged populations if designed properly and implemented through effective community consultation and participation."

Although community engagement (CE) is widely used in health promotion, components of CE models associated with improved health are poorly understood. This is the premise of a study that aimed to examine: the magnitude of the impact of CE on health and health inequalities among disadvantaged populations, which methodological approaches maximise the effectiveness of CE, and components of CE that are acceptable, feasible, and effective when used among disadvantaged populations. Disadvantaged populations studied here included those of low socio-economic status (SES), ethnic minorities, sexual minorities, culturally diverse populations, indigenous groups, and marginalised groups such as people with disabilities and the homeless. Such groups often experience health inequalities and bear a disproportionate burden of disease as a result of structural, social, and cultural barriers. "CE has been advocated as a tool for providing a 'voice to the voiceless' and therefore is considered valuable for tackling health inequalities....However, health interventionists tend to use CE approaches that have worked among non-disadvantaged populations for disadvantaged groups, often resulting in failure to achieve the desired outcomes....Current evidence shows that disadvantaged populations are not adequately approached or effectively engaged in the efforts taken by service providers and health interventionists to improve their health..."

As explained here, there are several CE models being used in health studies, including the Social Ecological model, the Active Community Engagement Continuum, Diffusion of Innovations, and community-based participatory research (CBPR), which aim to initiate population-level changes in health through the active involvement of the community. CBPR is often used synonymously with participatory action research (PAR) and action research, which include participatory approaches to health research. In contrast to the other CE models, CBPR has sought to bridge the gap between research and practice through equitable engagement of the community to eliminate disparities in population health. CBPR has achieved this by addressing power imbalances and enabling knowledge exchange, resulting in its wide uptake as an appealing CE approach across various cross-cultural, diverse, and disadvantaged settings. In addition, Rapid Assessment Response and Evaluation (RARE), a component of PAR, has been used as a public health research tool, particularly among ethnic populations, and incorporates the use of datasets, community participation, and evaluation. K. Staley's (2009) review identified key areas where CE can positively impact health research, including agenda setting, ethical conduct, programme design and delivery, involvement of the public in a project, and academic partnerships.

Of the 24 studies included in this review of studies published between January 1995 and June 2015 (see the article for inclusion and exclusion criteria), 17 were conducted in the United States (SU), and there was one each in Canada, Bangladesh, Africa, China, the United Kingdom, Iran, and India. The studies used various designs, including randomised controlled trials (RCTs) (n=11), quasi-experimental (n=2), longitudinal (n=2), qualitative (n=4), and mixed methods (n=5). The sample size varied from 23 to 3,986, with study populations having a variety of ethnic backgrounds, including African American, Hispanic, Indian, African, Chinese, Iranian, and indigenous First Nations communities. Thirteen studies focused on the improvement of health behaviours, 4 studies examined maternal/neonatal health outcomes, 2 studies focused on breast cancer, 2 studies examined mental health, one study examined sexual health among homosexual men, one study examined childhood asthma, and one study was on influenza pandemic planning. Out of 24 studies that met inclusion criteria, 21 (87.5%) reported improvements in health behaviours, public health planning, health service access, health literacy, and a range of health outcomes (see Supplementary Table 1). Fourteen studies were found to be of good quality (58%), 6 studies (25%) were moderate in quality, and 4 studies (17%) were poor in quality.

The researchers identified 11 categories of CE initiatives in this review, including CBPR, which was used in 12 (50%) studies, and community-partnered participatory research (CPPR), a variant of CBPR, which was used in 2 studies. The other 9 categories include the community health worker (CHW) model, community empowerment model, community action cycle, youth development model, the Well London model, participatory action cycle, the FOCUS (Families in Our Community United for Success) model, the Culturally appropriate Diffusion Communication (CDC) model, and ANGELO (Analysis Grid for Elements Linked to Obesity), all of which were used in one study each. Apart from CBPR (the most commonly used CE model), the researchers identified 6 other CE models that have successfully addressed health disparities among disadvantaged populations: FOCUS, ANGELO, CDC, community empowerment, the CHW model, and participatory action cycle. Although most of these models share similarities with the CBPR model, they lack 3 components that were key drivers of success in the CBPR model: (i) engagement of community partners in all stages of research development, including dissemination of findings, (ii) facilitating knowledge exchange between community and academic partners, and (iii) achieving balance between research and action.

The findings from this systematic review showed that the CE approaches in 21 out of 24 studies that met inclusion criteria led to improvements in health behaviours, public health planning, health service access, health literacy, and other health outcomes. Fourteen studies (58%) in this review showed that CE-informed research led to reductions in health inequalities by showing improvements in health behaviour and outcomes among disadvantaged populations bearing a disproportionate burden of disease compared to the mainstream populations. The results for each of the 24 included articles are presented in Supplementary Table 2. In short, 8 studies reported positive impacts of CE on health behaviours, including healthy eating, physical activity, breastfeeding, and condom use. One-quarter (n=6) of the studies reported positive impacts of CE on health outcomes, including reduction in obesity, improvement in mental well-being and quality of life, and reduction in neonatal mortality. Three studies reported increased awareness and improved knowledge of health issues among participants, and 2 studies also demonstrated improved participation in health screening programmes. Four studies reported community-level changes, including improvements in community empowerment, community-level health initiatives, public health planning, and the use of public parks. (These are selected findings; see the full article for more.)

For CE interventions that had positive impacts, components closely associated with improved health and health behaviours included incorporating the voice and agency of indigenous and ethnic communities in the research protocol, real power-sharing, bidirectional learning, and needs assessment. In contrast, CE models that did not improve health behaviours were affected by lack of community involvement in formative research and inadequate needs assessment.

To elaborate a bit, the available literature on CE states that there is currently an evidence gap in understanding which CE components contribute to successful study outcomes. This review examined CE levels along a continuum from informing communities to empowering them and found a link between low levels of CE (information-sharing and consultation) and poor study outcomes in 3 studies. On the other hand, studies achieving high levels of CE such as collaboration, partnerships, and empowerment showed positive study outcomes. A number of studies have found that CHWs can be successful in addressing health disparities among ethnic populations. This review showed that using CHWs among ethnic communities improved programme feasibility and impact by enhancing the relevance of health promotion messages, fostering improved health behaviours, overcoming cultural and access barriers, and encouraging participant engagement. Several studies in our review used a combination of CE approaches such as CBPR to develop collaborative partnerships and CHWs to deliver health interventions. Adopting such an approach had a two-fold benefit, where the community partners facilitated recruitment and training of CHWs, while CHWs accessed "hard to reach" participants experiencing health disadvantages and enabled their retention. Another CE indicator of study success found in the review was collaborative partnerships, which facilitated an improved understanding of traditional tribal and ethnic health beliefs among academic and other partners, enabling the development of locally relevant health policy initiatives for these groups. The review found that a range of CE tools such as surveys, forums, and photovoice enabled the establishment of these partnerships. It showed that new partnerships between community, government, and academic stakeholders and the use of existing infrastructure such as faith networks, park authorities, and tribal agencies were responsible for the post-intervention sustainability of programmes. Studies conducted among ethnic and tribal communities have shown that post-programme intervention effects were directly related to their cultural acceptability, the existence of a historical collaborative partnership, and the engagement of an influential community partner such as a government organisation or tribal agency in all stages of the research.

A further finding was that identifying needs unique to each ethnic community during the formative research phase was directly responsible for positive outcomes. Examples of these needs among culturally diverse and tribal communities include lack of childcare, traditional barriers to hygienic birth practices, barriers to health information access among homosexual men, fear of dying, fear of talking about cancer, and traditional beliefs preventing healthcare utilisation. The bidirectional translation and uptake of cultural concepts not only enabled community-specific needs to be identified and addressed in the intervention design, but were also responsible for programme satisfaction and retention.

The systematic review identified several non-health-related positive outcomes of CE, such as building of social capital, community capacity building, and empowerment of community members leading to community championship. The researchers found that CBPR enabled external partner organisations to achieve their goals by facilitating trust-building between native and academic communities. CE was found to facilitate referrals to social services, increase the quality of local services, and enable linkages with community resources. Contrary to the evidence stating that CE participants experience emotional distress and stress, the findings suggest that the majority of CE participants were empowered and improved their social networking and self-efficacy skills.

The article explores several challenges associated with implementing CE models and their uptake. Although CE is useful in reducing health inequalities, it is labour-, cost-, and time-intensive, and its effectiveness varies according to the type of intervention and CE model used. High-quality CE is often compromised by a lower-quality research methodology; in addition, due to several associated methodological challenges, RCTs are not the most effective approach to evaluate CE interventions. Several gaps in the current measurement of CE in heath intervention studies suggest the need for development of innovative frameworks and approaches to demonstrate the effect of CE on health outcomes in a comprehensible rigorous way. Furthermore, particularly among ethnic and indigenous communities, evidence shows that empowering the community without corresponding changes in the system's infrastructure can compromise the trust and relationship-building purpose of CE initiatives, resulting in poor individual health outcomes. The social hierarchy experienced by socially disadvantaged groups remains another significant challenge. For example, the CDC model, which was used to reach socially marginalised groups, received minimal support from the hierarchical health education system in China. Similarly, the participatory action model, using non-health workers to deliver the intervention among tribal women in India, has the potential for being unrecognised by the Indian health systems. One study showed that although the lay health advisor model is a sustainable approach, due to the inherently low health literacy levels of the rural lay advisors, it compromised the intervention quality, resulting in a short-duration low-intensity programme. Available evidence shows that the lay health advisor model as a primary intervention strategy has limited benefits in achieving health outcomes and has better potential in combination with other health promotion approaches.

In conclusion: "We have found that CE improves the health of disadvantaged populations and enhances health programme participation and retention within ethnic minority, indigenous, and immigrant communities who are usually excluded from research and innovative programmes....[H]owever, due to the lack of tools to accurately measure CE, the quantitative relationship between elements of CE and health outcomes could not be determined. In order for researchers to be able to accurately demonstrate the direct impact of CE initiatives, psychometrically robust tools measuring the dimension of CE in existing models are needed. Given that there is no 'one size fits all' CE model, health interventionists using CE models should include measurements of CE in addition to other variables in data analysis to demonstrate its relationship to the outcome variables."