1 Introduction

Neglected Tropical Diseases (NTDs) are common among the world’s poorest populations. The World Health Organisation (WHO) estimates that nearly 1 in 10 of the world’s population lives in extreme poverty (< $1.9 per day), with a number of them infected with one or more NTDs [1, 2]. The WHO has set an ambitious target of eliminating NTDs by the year 2030 [3]. Efforts to achieve this include the London Declaration of 2012 on NTDs, which aims at preventing morbidity and avoidable deaths associated with NTDS through global action to control the diseases [4]. In West and Central Africa, the burden of Buruli Ulcer Disease (BUD) remains high [5]. BUD is an NTD caused by Mycobacterium ulcerans that affects the skin [6]. The clinical presentation is diverse and usually includes oedemas, ulcers, nodules, and papules [3, 5, 7, 8]. Patients who get healed may still suffer significant scarring leading to contractures that limit their functional abilities, especially if they do not receive urgent medical intervention. Despite advancements in research and medical practice, the mode of transmitting BUD is still unknown [3, 5]. As a result, its prevention requires greater attention to reduce morbidity, mortality and disability.

The incidence of BUD is still high among the most endemic districts or communities in Ghana. Since 1993, there have been more than 11,000 Buruli Ulcer cases reported in Ghana [9]. Although recent evidence shows a significant improvement over the last few years, there are, however, a number of endemic districts or communities still struggling to control the spread of the disease [3]. In Sub-Saharan Africa, community engagement has been described as a ‘critical enabler’, especially in response to reducing NTDs and other communicable diseases [10]. Typically, for under-resourced health systems such as Ghana, these initiatives are key to building the capacity of communities to effect change in their behaviour and improve the overall outcomes of health interventions [11,12,13]. Community engagement, also described in other studies as Patient-Public Engagement [14], Community Participation [15], and Community Empowerment [16], is defined as “the process of working collaboratively with and through groups of people affiliated by geographic proximity, special interest, or similar situations to address issues affecting the well-being of those people” [17]. It has been identified by a number of studies as an effective approach to achieving the best outcomes in community-based health interventions [13, 18, 19]. In this study, we investigate the significant role of community engagement in the eradication of BUD in the Asante-Akim North District, which is the second most endemic district in Ghana with an average of 131.5 per 100,000 cases [9]. The findings of this study particularly provide important lessons for implementing community-based health interventions to improve health outcomes.

2 Methods

2.1 Study settings

The study was conducted in the Asante-Akim North District in the Ashanti Region–which is the second most endemic district in Ghana for BUD [3, 20]. The Asante-Akim North District occupies an area of 1089km2 with a total population of 85,788 as at 2021 population census [21]. The district is an agrarian economy with most people (72.7%) being employed in the agriculture sector. The Twi language is the most spoken in the district. In terms of religious affiliation, 79.8% of the population are Christians, 10.2% are Muslims, 1.2% belong to the African Traditional religion, 0.7 belong to other religious groups and 8.2% are people without any religious denomination [21]. The Asante-Akim North District is also a tropical forest zone with an average temperature of 26 °C.

The District Directorate of Health Service (DDHS) is mandated with the responsibility of supervising the delivery of all healthcare activities in the district. It is tasked with the responsibility of providing accessible and affordable primary and secondary health care services in accordance with national health policies. The DDHS also ensures effective and efficient management of healthcare resources.

The Asante-Akim North District has a total of 88 communities and 22 electoral areas or Community-based Health and Planning Services (CHPS) zones. The district has 12 health facilities made up of 1 hospital, 4 health centres, and 7 CHPS compound. Even though the district has the second largest hospital in the Ashanti Region of Ghana, it still does not have the adequate number of health facilities for most of its remote communities. Many people in the district travel on very deplorable roads or cross rivers to be able to access healthcare. The current doctor to population ratio is 1:7586 and the nurse to population ratio is 1:236 [21]. Epidemiologically, malaria continues to be one of the highest morbidities in the district. In addition, the district continues to record an increasing incidence of Neglected Tropical Diseases (NTDs) such as Buruli Ulcer, Onchocerciasis, Guinea worm disease, Schistosomiasis, Trachoma among others [22].

2.2 Design and sampling

This study used a qualitative research design to specifically explore the role of community engagement in a Buruli Ulcer eradication programme in Ghana. This was part of a broader qualitative study undertaken to understand Patient-Public Engagement (PPE) in Ghana’s health system improvement. A total of 22 participants were purposively selected because of their involvement in the BUD activities. They included Community Health or Public Health Nurses, members of the Community Health Management Committees (CHMC), Assemblymen,Footnote 1 Community Health Volunteers (CHVs), and Traditional leaders (Chiefs/Queen mothers) via focus groups and in-depth interviews. The community-based participants, particularly the residents were selected on the basis of their past or present experience with the BUD activities.

The focus group and individual interview guide adapted Rifkin’s spider‐gram methodological tool to explore the extent of community engagement in the implementation of the Buruli Ulcer Eradication Programme in the district [23]. The Spider-gram tool considers five key indicators in relation to lay participation in the implementation of health programmes: Needs assessment, Leadership, Organisation, Resource Mobilisation and Management. The first indicator, ‘needs assessment’, assesses the process of identifying the health problem and relating this to the health needs of the community. Leadership examines the authority of the programme and how it functions in the interest of the serving community’s, especially in terms of representing the poor and marginalised groups. The third indicator explores engagement from the point of ‘organisation’–that is, how successfully the new health programme integrates with other pre‐existing community structures. The fourth, ‘Resource mobilisation’ assesses a community’s commitment to the health programme through their contribution of resources (in various forms) and involvement in the allocation of these resources to support the health programme. The last indicator, ‘Management’ explores the decision‐making structures and processes of the health programme [24].

Thus, nine broad areas were explored, including participants’ views and understandings of the BUD programme, the extent of community involvement in the selection, design and implementation of the BUD health intervention, the community’s involvement in decision‐making and feedback processes and the extent of integrating existing community structures, if any into the BUD programme. Finally, participants were asked to rank their community’s involvement in the health programme on a spider‐gram scale of 1–5 (with 1 being the lowest and 5 being the highest involvement).

2.3 Data collection

The first author conducted in-person semi-structured individual interviews and one focus group interview between July and December 2021. An interview guide was designed (for both the focus group and individual interviews) to elicit participants’ views, experiences and opinions on the role of community engagement in the Buruli Ulcer eradication in the Asante-Akim North district. Prior to the interviews, an information sheet on the study was provided and discussed with participants, and informed consent was obtained. A Twi language translation of the study information sheet and informed consent form was available for those who could not communicate fluently in English (both English and Twi are acceptable languages of Ghana). Both the focus group and individual interviews were conducted in convenient locations of participants–mostly close to their homes or offices. Average duration of the individual interviews was 45–60 min and the focus group interviews lasted about 1 h and 30 min. Data collection continued until saturation was reached, that is, a point where additional interviews added no new thematic insights.

2.4 Data analysis

All interviews were audio-recorded and transcribed by the first author. Those conducted in the Twi language were directly transcribed into the English language. Transcripts were de-identified, and all participants were identified by their positions and assigned unique identifiers for analysis [25]. The research data were analysed and grouped into themes according to Braun and Clarke’s six phases for thematic analysis [26, 27]. These include: (1) becoming acquainted with the data; (2) generating codes; (3) identifying themes; (4) reviewing the themes; (5) defining and naming themes, and (6) writing about the findings [27, 28]. Transcripts were coded within NVivo Version 12 Plus [29]. The process of collecting and analysing these research data was systematic and this was duly documented in the fieldwork diary and the broader study [30]. That is, second, third and fourth authors verified the quality of data coding and reviewed all interview transcripts to ensure data consistency and interrogating the analysis process to minimise bias, especially in relation to any preconceived ideas that may have influenced the analysis.

2.5 Reflexivity statement

This research was conducted by an inclusive and diverse research team, with authors from varied disciplines transcending health systems, health policy, epidemiology and infectious disease. SAO and MDN are both early career researchers while SEA and EK are mid-career researchers. The rest are senior researchers with several years of experience. Most authors are experienced qualitative researchers.

2.6 Ethical approval

This research was approved by the University of Otago Ethics Review Committee (20/002). Additionally, approval was obtained from the Ashanti Regional Health Directorate of the Ghana Health Service and the Asante-Akim North District Health Directorate. All participants consented to the study. Data collected were not linked to individual participants. The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

3 Results

The study findings are presented under the five spider‐gram dimensions. As shown in Table 1, 22 people participated in the study (focus group interviews–9; Individual interviews–13). The focus group participants were made up of Traditional leaders (n = 1); Residents (n = 3); Community Health Nurse (n = 1); CHMC–CHVs (n = 2); CHVs only (n = 2). The individual interviews were made up of traditional leaders (n = 1); Resident (n = 2); Community Health Nurse (n = 1); Public Health Nurse (n = 1); CHMC-CHVs (n = 2); Assemblyman/Woman (n = 1); CHVs only (n = 4); Health Administrator (n = 1); Programme Coordinator (n = 1) (Table 1).

Table 1 Distribution of participants
Full size table

Overall, across the focus group and individual interviews, most participants (67%) were male and 33% female. In addition, 40% of participants had received tertiary education, 20% secondary and 30% junior high school and 10% had no education. The average age of participants was 40 years. The focus group interview was conducted in a local language (Twi)–which was the preferred language for the participants.

3.1 Buruli ulcer disease in the Asante-Akim north district-insights from participants

Findings from the study revealed that traces of Buruli Ulcer infections were discovered in the Asante-Akim North District in the early 1980s. Although research has shown that the cause of BUD is unknown [3], our study participants expressed the opinion that the disease was mostly prevalent in the rural farming communities which also had lots of water bodies and swampy environment. A report from the Presbyterian Hospital, located in the Asante-Akim North District also revealed that about 50% of all surgical cases in the hospital, as at the year 2002, were mostly related to wound debridement, skin grafting and amputations resulting from Buruli Ulcer infections [31]. The worsening cases of Buruli Ulcer infections had resulted in almost every household in the endemic communities having at least a family member suffering amputation as a result of Buruli Ulcer infection. This led to most residents fleeing the community for fear of being infected. A greater majority of them also attributed this to spiritual causes. As a result, the affected communities, particularly, did not initially consider the programme as a solution to the problem.

Our community did not accept the Buruli Ulcer programme because we knew it was a curse that had been inflicted on us by the gods of our land. Our traditional chief priest mentioned that someone offended the gods and that they did some rituals, but the gods were still angry. So, it was difficult to now go back and tell the people our chief priest was lying but rather this white man’s approach is the solution to our local problem (CHV 1, FG).

3.1.1 A Traditional leader also commented

The people had perceived that the hospital people were more interested in cutting people’s hands and legs. So, bringing the Buruli Ulcer programme here made us more scared for the fact that it was going to rather worsen the situation by cutting many people’s hands and legs. (Traditional leader, FG).

With these perceptions in the minds of the community, the Buruli Ulcer programme, therefore, employed a community-based approach, which was considered as more efficient and effective in achieving the aims of the programme.

The initial resistance from the community made us even approach the issue more differently. Although there were other factors like inadequate number of nurses, and lack of resources, but most importantly, we adopted the community-based approach, particularly using the CHVs to ensure we could come down to their level and use their own people to detect and treat the disease faster. Our programme focused on controlling the spread of the infection as well as eliminating it completely if possible (BU Programme coordinator).

3.2 Extent of community engagement in the implementation of the buruli ulcer disease programme

The spider-gram or pentagram dimensions were used in this assessment. These focused on: Needs assessment, Leadership, Resource mobilisation, Organisation and Management.

3.2.1 Needs assessment

All participants agreed the selection of the health programme was solely done by the health professionals rather than the community. However, they mentioned this was crucial to control the spread of the infection which had rendered a number of young people in the community disabled. Although the community did not directly call for this health intervention, it was an important and urgently needed to stop the devastating effect of Buruli Ulcer infection in the community.

The health professionals realised they needed to intervene and help us with this problem. It was a huge problem for us. The disease chased us out of the community. Although we did not immediately accept that this was a medical cause because of what our traditional chief priest told us, in the end the health experts were right (Traditional leader, FG).

A representative of the Buruli Ulcer programme also commented during an individual interview:

The health needs were identified based on the data we had. The patients who reported to the hospital with Buruli Ulcer disease were coming from particular communities. In fact, 97% of the cases were from this community. With that data available, we needed to intervene and not wait for them to tell us. As health workers, we needed to be proactive (BU Programme coordinator).

The responses from the informants also revealed that although this was an important health need which resonated well with the community’s need, however, the programme design did not include community members or their representatives. Responses from the CHVs involved in the programme implementation revealed that they did not make inputs into the programme design and their role was also more dictated for them by the health professionals who led the implementation of the programme.

We were not involved, and we did not have any idea when they were planning the programme. They only told us what they expected from us, and what we needed to do. We made some suggestions to them but those were not considered so we also just followed their plan (CHV 2, FG).

The community participants ranked the community’s involvement in ‘needs assessment’ at point 1 which indicates narrow or no involvement. Whilst this indicate low involvement, the community, however, accepts it was an important health need, although their input in the design of the programme would have made it more successful.

3.2.2 Leadership

The research findings also revealed the Buruli Ulcer programme was implemented through CHVs who were selected by the community. The participants also mentioned that although the CHVs were representative of the various community interest groups, it did not have female representation. The health workers indicated they preferred men to women due to the difficult nature of the task as it involved travelling to the hard-to-reach village communities which sometimes involved swimming across streams and rivers at night. Notwithstanding, the participants expressed the view that the interest of all minor and major groups including women were considered in the design and implementation of the programme.

The disease affected everyone, both the youths and adults. It did not discriminate against age or ethnicity, so we ensured the CHV selection was representative. But, due to the nature of the work…. because it involved going to these villages around the Afram Plains River. There are many risks involved including swimming or travelling across these places in the night, so we preferred men to women. It was not discriminatory, but we needed to ensure their safety. But it was the existing CHVs in the community we used just that we did not include the female CHVs for this programme (Community Health Nurse, FG).

The findings also revealed the CHVs did not have decision-making power as their role was more subjected to the rigid design of the programme which ensured all decisions emanated from the Buruli Ulcer programme coordinators. However, the community-based participants, involving a traditional leader, a resident/opinion leader and CHVs mentioned that few of their inputs were factored in decision-making especially during the programme implementation. Although the community did not have absolute control over decisions-making, the community’s inputs were given some attention and decisions taken on their behalf by the Buruli Ulcer programme managers were generally sensitive to the community needs and it represented their interest.

As my colleague mentioned earlier, this was not a programme that had more room for certain things. The programme has a decision-making structure that everyone from national to the community follows. The CHVs are important but the community’s decisions had to still go through the decision-making structure of the programme. So yes, I can say they did not have so much autonomy in terms of making decisions, but we did consider their suggestions (BU programme coordinator, FG).

3.2.2.1 A traditional leader commented

We as a community are pleased with the decisions they still took on our behalf. There were some suggestions we gave them which they also considered. Not many of them were considered but I think it was just satisfactory (Traditional leader, FG).

Overall, community engagement in the ‘leadership’ dimension was ranked at point 2–indicating restricted or limited community involvement.

3.2.3 Resource mobilisation

We also found that the Buruli Ulcer programme was externally funded which limited the community’s control over how resources were allocated. The community did not also mobilise any financial support for the programme. The Buruli Ulcer Programme Coordinator revealed that the affected communities were very poor, as many of those infected with the disease had difficulty affording the cost of transportation to the hospital. As a result, the programme provided all needed resources, including paying for their cost of transportation through the CHVs. It was also indicated the CHVs were provided with the needed logistics and financial incentives to motivate them in their work.

This programme received funding support from the Agogo Presbyterian hospital and the WHO, so we tried to manage with that. These are poor communities and the last thing you want to attempt is to subtly tell them to raise money to support this programme. That will be difficult to even attempt. For now, their time and cooperation are enough for us (BU Programme Coordinator, FG).

3.2.3.1 A CHV also commented

The programme provided everything free for us. From taking the infected person to the hospital, they paid for the transport. As CHV, when the person is confirmed as having Buruli Ulcer, I am also given some money to motivate me to bring more patients. They gave us bicycles, and quarterly training for which they paid us after each training. The community did not provide any resource. They did it all by themselves (CHV 1, FG).

Thus, participants unanimously ranked the ‘resource mobilisation’ dimension at point 1 on the spider-gram scale indicating the community’s involvement was narrow.

3.2.4 Organisation

Our study findings also revealed the Buruli Ulcer programme was effective in integrating all pre-existing community structures into the programme. Participants opined that the programme implementation integrated all existing CHVs in the community and also involved the leadership of the community, particularly the traditional leadership in the selection of additional CHVs for the programme. The findings also revealed a unique approach adopted by the Buruli Ulcer programme in integrating the traditional healers into the programme. Several traditional healers in the community were included as CHVs for the Buruli Ulcer programme, whilst those not selected as CHVs were also provided with financial incentives for referring Buruli Ulcer cases to the hospital. In cases where a referred patient from herbalist tested negative for Buruli Ulcer, the patient was still referred back to the traditional healer to decide whether to treat or refer again for hospital-based treatment. These ensured maximum support from the traditional herbal practitioners in the community as it largely contributed to early detection of the disease. A Buruli Ulcer programme coordinator elaborated:

One of the strategies we adopted was to identify the herbalists who managed such cases. We involved them in our meetings and briefed them and talked to them nicely with pictures about the disease. For them, we decided to engage few as community volunteers, but for the majority, we told them about our expectations from them. We told them if they suspected Buruli Ulcer, they should refer to us and the amount we gave to the CHVs for identifying positive case, same amount will be given to them. In case the patient is negative for Buruli Ulcer, we still referred back to them in the community; either they will manage it, or patient will prefer to do hospital medicine. Because of that, their participation was good (BU Coordinator, FG).

The participants unanimously ranked this spider-gram dimension at point 5 indicating the community engagement in terms of organisation was wide and excellent.

3.2.5 Management

The research findings revealed that although the selection of the CHVs was democratically done, the Buruli Ulcer programme was not independently managed by the local community people or the CHVs. The Buruli Ulcer programme managers entirely managed and decided how the service was managed. A participant highlighted this in the excerpt below:

As we have mentioned earlier in the discussions, the Buruli Ulcer manager was in-charge of everything. The role we played as a community and CHVs was to support them because the programme was for our own good. The creation, management, supervision everything was in their hands. We only followed their direction and how they wanted us to support in this programme. There were times we also provided some suggestions and they listened (CHV 2).

In relation to post implementation sustainability of the programme, the research findings revealed there were limited efforts to ensure the community was able to continue with the Buruli ulcer control activities after the programme had ended. The programme design did not have clear pathway towards sustaining the gains made in the control of Buruli Ulcer in the community. With funding for the Buruli Ulcer programme ended, the programme relied on the generosity of the Presbyterian hospital to cater for the cost of patient treatment. This support, however, excluded the other community-based activities such as providing logistical and financial support to CHVs, and other educational activities. A member of the Buruli Ulcer programme elaborated:

…. the sad reality is that the programme has now run out of funds, and we rely on the benevolence of the Presbyterian hospital here to provide free treatment of healthcare to our confirmed Buruli Ulcer patients. Now, it is difficult supporting the activities of the CHVs and even other community surveillance and health promotion activities (BU Programme Coordinator).

The participants however expressed the view that the difficulty in sustaining the programme presented an important lesson for future programmes with the call for programme managers to ensure the building of capacities of local communities through their active involvement across all stages. This would ensure communities have the capacities to sustain the programme post implementation.

As the saying goes, prosperity does not last forever. That was why it was important they involved the community in all that was going on. We all knew from start of the programme that there will be such a time the programme will end. They were supposed to empower the community to take up so many things even in their absence. They run almost everything from start to finish, so it becomes difficult to sustain the momentum after they left. They themselves have learnt a good lesson from this (Assemblyman).

Overall, the participants ranked the leadership role of the community at point 2 on the spider-gram scale which also indicates restricted or small community engagement. Table 2 and Fig. 1 provides a summary of the findings under the spider-gram framework.

Table 2 The Pentagram dimensions and PPE rankings (in bold) Buruli Ulcer Control Programme at the Asante-Akim North District
Full size table
Fig. 1
figure 1

Spider-gram diagram depicting extent of PPE in the implementation of the Buruli Ulcer programme

Full size image

4 Discussion

This research examined the role and lessons of community engagement in eradicating Buruli Ulcer Disease in Ghana. The spider-gram framework was adapted to examine the extent of community involvement in the BUD programme [23]. The Asante-Akim North District, being one of the most endemic districts for BUD cases in Ghana was selected for this study. Our study findings were categorised in the following dimensions of the spider-gram framework: needs assessment; leadership; resource mobilisation; organisation; and management.

In our needs assessment, we noted that the Buruli Ulcer programme was not a request from the community but rather an initiative by the health authorities to stem the spread of the disease and its severe effect on the people. As argued by Draper et al. health needs can be identified, either by the community or the health authorities [32]. However, in situations where health authorities find a good justification for a health intervention to be implemented in a community, this must not be imposed unilaterally on the community without engaging or involving them in the programme design and implementation [24, 33]. Evidence from this study revealed that although there was a strong justification for the Buruli Ulcer programme implementation in this community, the people were not involved in the programme design. As a result, their inputs and feedback were lacking in the entire programme design and implementation. This, however, was considered contrary to the effective implementation of such community-based health interventions as their involvement in the design of the programme was key to its success [13, 14, 24, 33, 34].

On leadership, the Buruli Ulcer programme was effectively implemented through the CHVs in the community. As noted by Abass et al. implementing such community-based health interventions through CHVs provides appropriate local strategies that ensures effective implementation of the programme [20]. Thus, also confirming Campbell and Jovchelovitch’s theory of psychology of social participation which argued for ‘local knowledge’ as key element to any successful implementation of health programmes [35]. In this study, we similarly found that the CHVs had minimal control on decision-making regarding the programme, contrary to recommendation by Charles and DeMaio and other scholars on the need to involve lay representatives in decision-making [36,37,38]. Nonetheless, in certain key situations such as facing resistance from the community, the CHVs’ understanding of the local socio-cultural environment played crucial roles in getting the support of the community. For instance, we noted from the study, how the involvement of the CHVs was key in not directly telling the people that the Chief Priest lied to them about the spiritual cause of the disease as this could have negatively impacted on people’s acceptance and participation in the programme. We again noted the selection of CHVs was not gender-sensitive and hence representatives as female CHVs were not included in the programme implementation. Although the health authorities indicated the involvement of females in the programme was not ideal for the nature of the work, this study, however, found that this was largely the preference of the health authorities and not the community. Similar to the findings of Baatiema et al. this study also found that the elimination of female CHVs affected some aspects of the programme implementation [39]. For instance, the male CHVs were largely restricted in physically examining suspected female Buruli Ulcer patients as this was interpreted to mean inappropriate ‘touch’. However, this could have been avoided if such patients were physically examined by the female CHVs. Thus, in designing such community-based health intervention programmes, gender consideration and cultural sensitivity need to be prioritised.

Regarding the community’s involvement in resource mobilisation, the study found that the Buruli Ulcer programme was solely funded by the National Buruli Ulcer Control Programme with support from the Presbyterian Hospital in the district. As a result, apart from the voluntary involvement of the CHVs, and the support given by the community and the traditional leadership, there were no financial or logistical support from the community. However, as argued by some scholars, a community’s resource mobilisation support cannot always be measured in terms of logistical or financial contributions, but also other intangible support which are also key to the success of the programme [24, 32, 33, 39, 40]. Therefore, although the community did not contribute financially and logistically towards the programme, the study found their invaluable support for the programme, including the voluntary sacrifices of the CHVs as significant [41]. However, the inability of the community to influence allocation of resources means, the extent of PPE in relation to resource mobilisation was considered narrow or limited. Importantly, one key factor that worked for the Buruli Ulcer programme and the community’s interest in the programme, was the ability to provide the needed financial and logistical resources for the work of the CHVs. The CHVs were provided with financial support to cater for their transportation when sending suspected Buruli Ulcer patients to the hospital. In addition, they were provided with bicycles for their routine surveillance trips and even some financial rewards when a suspected patients they referred are confirmed positive for Buruli Ulcer disease. Some CHVs were also rewarded with employment into the health sector as a form of motivation. These largely boosted their morale for the programme and also encouraged them in the work [20]. As noted by Alhassan et al. and Ankomah et al. although the work of CHVs is voluntary, providing them with the needed logistical and financial support are key to achieving the overall aim of the programme [41, 42].

Again, examining the findings of the Buruli Ulcer programme on the spider-gram framework in relation to ‘organisation’ showed a very wide and excellent PPE. Whilst integrating pre-existing community structures has been mentioned by many scholars as essential for the effective implementation of community-based health interventions [33, 39, 40], anecdotal evidence from Ghana, as revealed in earlier sections also indicates not doing this effectively can result in conflict situations between health workers and community members. In this study, integration of pre-existing community health structures into the Buruli Ulcer programmes was also ranked excellent. This could indicate most community-based health programmes learnt lessons from the regular conflicts that mostly occur between health workers and communities during implementation of such programmes. The Buruli Ulcer programme provided a unique approach to integrating the traditional herbal healers into the programme. For instance, few traditional herbal healers were nominated and approved by the community into the programme as CHVs, whilst the others were also supported with training and other logistical and financial resources for early case detection and referral. That is, these traditional herbal healers were provided with financial rewards for referring confirmed Buruli Ulcer case, and they were also re-imbursed their cost of transporting cases to the hospital. These largely ensured that Buruli Ulcer patients, who ordinarily would have been kept with the traditional healers till their prognosis went bad were now referred early to the hospital for medical treatment. As was noted by Abass et al. and also confirmed in this study, the successful integration of the existing CHVs, and traditional herbal healers into the Buruli Ulcer programme were instrumental in the successful implementation of the programme [20].

Lastly, on management, the findings from this study confirmed other similar studies conducted in Ghana where CHVs or CHMCs were mostly not allowed to independently manage a community-based health programme [43, 44]. Although other scholars such as Bate and Robert have suggested a co-designed and co-managed health programmes between the community and health authorities [45], the implementation of the Buruli Ulcer programme was, however, contrary to these as the programme was independently managed by the Buruli Ulcer coordinators with limited inputs from the community. This had also greatly affected the sustainability of the Buruli Ulcer programme after the programme funding ended. As was found by Ankomah et al. [24, 46], not involving users in the design and implementation of a community-based programme largely affects its sustainability beyond the implementation phase. Thus, this study found that the implementers of the Buruli Ulcer programme were struggling to sustain the programme after the funding ended. Therefore, as also recommended by other scholars [47, 48], communities must be empowered and involved in the design and implementation of a programme to ensure its continued sustainability beyond the implementation phase.

5 Limitation of the study

The use of the spider-gram framework to analyse the extent of community involvement in a Buruli Ulcer eradication programme may not have addressed all aspects important to the success or failure of the programme. This study recognises that the spider-gram dimensions may not be exhaustive in assessing the fundamental role of PPE in health programmes as certain areas may not have been included in the five dimensions. However, as already mentioned, the key objective of this study was not to measure the ultimate success or failure of the programmes, but to investigate the specific role of community engagement in programme implementation. Thus, this study recognised that the spider-gram provided a workable framework to assess the community’s role in the Buruli Ulcer eradication programme as previously employed by other scholars in different community-based programmes [33, 39, 40].

6 Conclusion

Overall, this study notes that the extent of community engagement in the implementation of the Buruli Ulcer programme was largely limited–in what Arnstein would describe as ‘tokenism’ [47, 49]. However, there were key areas such as integration of pre-existing community structures into the new Buruli Ulcer programme, and provision of adequate logistical and financial support to CHVs which were instrumental in the successful implementation of the programme. This, particularly, provides important lessons for similar future implementation of community-based health interventions to consider especially in ensuring communities are involved in the design and implementation of health programmes as well as ensuring a wider involvement in all the five key dimensions of the spider-gram framework.