In the first phase of implementation as well as post intervention, formative research [33, 34] was conducted and qualitative assessment were performed. In-depth interviews, focus group discussions, participant observations, and informal conversations were conducted with several levels of stakeholders and actors at the community level, including community health workers (CHWs), health centre chiefs, school directors, monks, priests, teachers, farmers and members of the local education office as well as students.
Qualitative data collected has been used to facilitate community dialogues and inform workshops focused on messages and material’s design within the aforementioned COMBI strategy. These workshops, in the form of group discussions, were held with key community members, volunteers and district stakeholders in schools included in intervention arms. During these meetings, participants agreed that vector control tools needed to be adapted to the community context and integrated into the schools’ health education revised curriculum. In fact, there was a general sense of agreement towards strengthening the project’s “adaptive capacity building/empowerment/educational” aspects and the qualitative dimensions of the project. An emerging question that arose during these community gatherings is how can we use vector control tools that most community members agree are efficient (i.e. guppies) to revamp the health education curriculum and integrate these tools for routine communication in community settings. In summary, there was a general demonstration of interest by community actors to improve ownership of the vector tools and a strong participation towards adapting tools and methods to the cultural context and local socio-economic level. This indicates that SI, a processes through which social change grounded in local realities is generated, has emerged from the engagement of stakeholders during the numerous meettings, workshops, focus group discussions and interviews. Together with SI, several social innovation products were developed, representing critical project outputs towards building community adaptive capacity and project outcome sustainability. These SI products are described below.
Through regular visits and collaborative trainings, women’s groups were capable of producing 3228 medium size traps (MST) and 6300 small size traps (SST), a total production of 9528 traps. These traps which replicate autocidal gravid ovitraps [35] in principle (Fig. 1), were placed in 20 implementation villages with 3 traps (1 MST + 2 SST) deployed per household (HH) in each of the 3158 households, and 2 traps (2 MST) in each of the 161 rooms in 16 schools. Beyond the impact on entomological indicators, the process of developing the trap design with the women’s group during workshops and subsequent follow-ups contributed to a community-owned innovation and an increased sense of ownership of the product and its use (Additional file 1). This was demonstrated through the ongoing transformation of the women’s group from casual gatherings to civil society organizations and social enterprises. These more formal organizations are being created with the intention to improve the design and distribution of the traps as well as to develop outreach strategies for continuing impact. In addition to that, the traps, made of recycled plastic bottles, generated income for the participants and contributed to increase awareness regarding solid waste management and effectively recycling plastic waste. Together these actions and outcomes encourage positive change and offer new opportunities.
Locally made adult mosquito traps replicating autocidal gravid ovitraps (AGO). A Finished product, B Schematic design. The AGO trap consists of five basic components: 1) black polyethylene cylinder that serves as the trap entrance (12.8 cm in diameter) and transparent capture chamber; 2) sticky surface covering the interior of the capture chamber that is coated with 155 g/m2 of a nonsetting polybutylene adhesive 3) screen barrier at the bottom of the capture chamber to prevent adult mosquitoes from moving between the capture chamber and the infusion reservoir. It also prevents any mosquito emerging from the infusion to escape from the trap (occasionally, eggs from captured females may be washed by rain into the infusion reservoir and develop into adult mosquitoes); 4) black polyethylene pail with drainage holes to allow excess infusion to drain from the trap 5) infused water
The team facilitators worked closely with the Ministry of Education and the School Health Department of the Ministry of Health as well as school directors, officers and teachers to collaborate on the revision of the health education curriculum to incorporate elements of Dengue transmission, mosquito biology and ecology, biocontrol with guppies, waste management to minimize breeding sites as well as mosquito collection procedures (Table 2). The project team facilitated training sessions that also included pedagogy, learning and teaching style focusing on hands-on transformative learning experiences. These training sessions provided the basic material, know-hows and inspiration for teachers to subsequently implement the revised curriculum and hands-on activities with maximum engagement of the students. In total about 100 teachers, school directors and officers participated in these training sessions and over 500 students were involved in receiving and in turn communicating this novel curriculum content. For students, part of the transformative learning experience was related to their contribution to community-based “education” sessions whereby students could showcase their acquired understanding of dengue, its significance and how to address the problem in their communities. Students participated in 40 of these sessions during which knowledge transfer across generations was meant to augment current community sense of ownership of dengue and its control (Additional file 2).
The cross-sectorial collaboration and transdisciplinary action that took place during the school-based sessions together with the strong engagement of students in activities of knowledge sharing in communities, led the department of school health of the ministry of education to incorporate the co-designed dengue curriculum into the national school program with 1 h per week allocated to dengue and its community-based integrated control.
An essential and very effective vector control tool in this project is the use of guppy fish in water storage tanks as well as smaller containers that are commonly found around households. The efficiency and acceptability of guppies in reducing vector abundance has been demonstrated in several projects including previous community-based dengue trials [28]. The project team together with school partners, community leaders and community health workers established guppy fish banks in schools (3 jars × 16 schools), in community settings (6 jars × 20 communities) and at health centers (20 jars × 6 HCs). Students were involved in guppy fish distribution to their households and community health workers were responsible for distributing guppies to community guppy banks.
Community members could also directly collect guppies from the health centers. A total of 22 400 guppy fishes were distributed in the first 6 months of the project. Training sessions have been facilitated to empower 100 school teachers, 94 CHWs and staff from six health centers providing knowledge on how to rear, maintain and distribute guppy fishes (Additional file 3). Overall there has been a strong consensus of the relevance and ease of use of guppies as “decentralized” vector control tool as described in the following accounts:
“The guppy fish is not complicated; it is no need to take care of it for often. It can eat all the larvae from the water containers.” Male CHW in Kraloang village, 49 years old.
“We distributed three fishes, two females and one male to students, right now there is much fish still in the jars. We still give to students who lost their fish when they ask from us.” Teacher, 20 years old.
The iterative community engagement initiatives regarding guppy use and distribution has led to a dramatic increase of guppy presence in households, from 11% of HH using guppies in August 2018 to 42% in August 2019 with about 1260 households now having guppies. Observations or anecdotal reports that were received from community members indicated that guppies were informally distributed outside interventions areas, suggesting knowledge transfer, cultural acceptability, strong feasibility of scaling up and project outcome sustainbility. Discussion during intervention follow-up sessions as well as during the November 2019 research uptake meeting highlighted the value of further operationalizing the guppy distribution system. Among the ideas exchanged, it was mentioned that the development of a phone application would offer a flexible interface for communication among distributors and household members or guppy banks in the communities regarding stocks and refill needs and create another opportunity for social innovation.
Focus group discussions and key informant interviews in school and community settings enabled constructive discussions and significant engagement of stakeholders. Participants generally considered that working with schools was a good strategy to introduce knowledge on guppy rearing and care, as well as to bring that knowledge to villages’ homes through involved pupils. However, participants pointed out that working at the village level is equally important:
“I think we have to do both. The school is the place to grow the human resource for present and the future because they are young ( … ) it is good for them to receive the knowledge. But for the adult people who live in the community, they don’t get any knowledge from the children because some children can learn but they cannot explain to their parents ( … ) so we have to do both.” Monk, 38 years old.
Recommendations about relevant sites for the diffusion of information at community levels included pagodas, commune halls, health centers and private medical practices. The mobilization of health workers during vaccination campaigns was also seen as reference—and potential strategy—for the diffusion of health-related information. Similarly, monks have proposed ceremonial occasions at the pagoda as acceptable moments to communicate dengue related control knowledge or procedure, provided that they previously receive education on dengue control.
In relation to communication channels, women and grandparents were identified as responsible for decision-making and implementation of prevention activities in relation to dengue control at the household level in the past. Participants agreed that women and grandparents should be mobilized as key actors in current and future dengue interventions, particularly regarding enabling knowledge and action to flow from schools to communities through their privileged relationships with their children. Village health workers are also generally trusted as sources of information at the community level, particularly in contexts where health centres’ collaboration with local schools is highly dependant on staff’s availability.
Content wise, most participants were aware that dengue fever is caused by Aedes, locally known as ‘tiger’ mosquitoes. Interviewees generally stated that guppy fish, ABATE (Temephos) larvicide and environmental cleaning around their settlement can be useful methods to eliminate mosquitoes breeding sites. Playing spaces around mango trees were referred as potentially high-risk sites for transmission.
Initial communication material was developed during “high level” stakeholder meetings whereby official representatives of government bodies as well as community leaders met to prepare the planning of intervention activities, monitoring and evaluation, and to help to mobilize local resources and give logistical support. It was further adapted during 40 health education sessions where students presented their own versions of the posters and banners. The students communicated their material to an audience of between 20 and 45 villagers in each of these sessions with the support of CHW for the design of specific messages.
An important medium of engagement was the co-creation of community maps spatially representing local perception of breeding sites locations, zones of contact with mosquitoes, frequency, extent and timing of people movement, significant infrastructure enabling mosquito presence and general epidemiological data. About 650 villagers, particularly women, participated in Particpatory Epidemiology Mapping (PEM) sessions and were actively involved in the identification of the dengue transmission arena boundaries (Additional file 4).
During PEM sessions, higher participation from local people contributed to increase local mobilization in reducing breeding sites, leading to reducing the adult mosquito population (manuscript in preparation). The maps created could then be used to focus subsequent vector-control effort and better understand dengue transmission overall. Participants involved in PEM have significantly developed new relationships between their experiences and the knowledge shared during the sessions. Participants could then compare the map to the real infrastructure elements or processes in their village.
“PEM could help local people to identify and manage the mosquito breeding site in the village. People will be aware of mosquito breeding place around the house and in the village.” Krasaing Pul village chief, 60 years old.
Participants indicated that PEM was a useful tool for them to know how to identify breeding sites and locate them as well as to help CHWs improve control initiatives in the village. The outputs, mat mapping or paper maps (Additional file 4), can also support primary schools teaching capacity helping students understand the local transmission locations.
The process of mapping and the discussions around it also contributed to highlight knowledge gaps. Most of the participants for instance were still confused and surprised that mosquito larvae aquatic habitats could be found in and around lakes, ponds, or streams. Some people also were not aware of the breeding sites around their house. The majority of participants were female and elders while young men were at work and during participatory session only few outspoken individuals mostly contributed
“Even there are many participants in PEM but only a few people expressed their idea in the meeting” Male, 35 years old.
However, as information was discussed the message was heard by all and consensual spatial representations were made.
It was observed that, through the series of community sessions, participants gradually acquired a stronger sense of ownership and the capacity to become the stewards for their own vector control responsibilities as the maps took shape session after session (manuscript in preparation). As such, the process of spatially representing epidemiological information and infrastructure, create a forum for community members to strengthen the community relevance and practicality of dengue and its control. Control and surveillance intervention then become grounded in community context and therefore ownable and actionable. Doing so means that community-based trapping scheme (for surveillance) or school science approach for dengue mosquito monitoring can support government-led dengue vector surveillance and control by providing insight in vector species distribution and dengue transmission local patterns. From our observations, we anticipate that the data generated via these approaches are relevant for the planning, implementation and evaluation of vector control activities by NDCP. The involvement of local schools or communities in the science of mosquito ecology is expected to lead to more sustainable solutions for dengue control. This approach presents opportunities to bring down institutional barriers, such as low level of community involvement in vector control, limited financial resources for mosquito surveillance and the current exclusion of more remote areas in mosquito monitoring which are known to remain critical impediments to sustainable vector control.
