Initiating a network to support engagement between health researchers and schools: recommendations from an international meeting of schools engagement practitioners held in Kilifi, Kenya [version 1; peer review: awaiting peer review]

Engagement between health researchers and local schools, or School Engagement, has become incorporated into the engagement strategies of many health research institutions worldwide. Innovative initiatives have emerged within Wellcome Trust-funded African and Asian Programmes (APPs) and elsewhere, and continued funding from the Wellcome Trust and other funders is likely to catalyse further innovation. Worldwide, engagement between scientists and schools is well-described in the scientific literature (1-4), however, engagement between health researchers and schools is much newer, particularly in Africa, and rarely documented in the academic literature. In November 2018 the KEMRI-Wellcome Trust 1,2 1 3,4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Reviewer Status AWAITING PEER REVIEW 18 Nov 2019, :180 ( First published: 4 ) https://doi.org/10.12688/wellcomeopenres.15556.1 18 Nov 2019, :180 ( Latest published: 4 ) https://doi.org/10.12688/wellcomeopenres.15556.1 v1 Page 1 of 13 Wellcome Open Research 2019, 4:180 Last updated: 26 NOV 2019


Reviewer Status
Invited Reviewers version 1 18  We thank the Wellcome Trust's Provision for Public Engagement (grant number 203077) for funding and supporting this work, Grant information: and would like to acknowledge the role of the Global Health Bioethics Network (Wellcome Trust Strategic Award 096527) for nurturing the strong international working-relationships essential for this meeting. KWT was supported by the NIHR Southampton Biomedical Research Centre. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Background
Community engagement with health research is argued to be important to enhance the ethical conduct of health research [5][6][7] . Goals for engaging communities range from protecting communities from harms and exploitation, empowering autonomy for individual research decision-making, and enhancing the benefits of research participation to communities, to incorporating community views into research implementation 8 . Previous discussions with community engagement practitioners have revealed a broader list of goals such as improving health care, contributing to community development, and raising awareness of research for recruitment 9 . This broad range of goals have precipitated a range of engagement approaches in Africa, from community bazaars to deliberative discussions and community advisory boards 10 . More recently, health research institutions have explored engagement between researchers and local schools as a means of addressing some of the goals of community engagement 11 .
The term 'school engagement' is frequently used to describe the degree of investment a student has in their school 12 ; however, in this article, 'School Engagement' with health research describes a range of activities which facilitates learning interactions between health researchers and primary and secondary school teachers and students. School Engagement approaches have included: attachment schemes and pre-university summer schools for students 11,13,14 ; establishing long-term involvement of individual scientists with schools to enhance science lessons 1,15 ; in-school education resources, coupled with school visits to institution laboratories to facilitate interactions between researchers and students 11, 16,17 ; and Young Persons' Advisory Groups (YPAGs), which can potentially facilitate the incorporation of the unique perspectives of young people into research implementation [18][19][20] . The goals of these approaches vary considerably from raising student awareness of health and research and gaining their insights into research, to stimulating an interest in science and research careers and demystifying the work of scientists. The purpose of this paper is to describe the range of School Engagement activities and evaluation approaches used by participants and their institutions, to map the range of goals for School Engagement from the participants' perspective, and to explore practitioner perspectives on the potential usefulness of a School Engagement practitioner network and means of sustaining it.

Workshop description
The workshop was funded through the Wellcome Trust's Provision for Public Engagement. The Wellcome Trust is a major global health research funder with a funding portfolio of £4.3 billion in 2019, £56 million of which is invested in public engagement. Given their prominence in funding community and public engagement with health research, the core of practitioners working in Asia and Africa were identified through Wellcome Trust-funded partners working in engagement. In addition to Wellcome Trust-funded practitioners, other delegates were invited based on previous and potential future collaborations and interest in engagement between schools and health scientists. All participants and their affiliations are included in the author list and Acknowledgement.
This participatory collaborative workshop comprised a combination of presentations of School Engagement approaches and evaluation; video sharing sessions; learning sessions on applying theory of change to School Engagement; small group discussions and plenary reflections. The latter two activities were used to share practitioner views, experiences and perceptions, in a similar way to a previous workshop, held in Kilifi, for community engagement and informed consent scholars 9 .

Mapping of engagement activities, and evaluation approaches.
The range of School Engagement activities and evaluation approaches were mapped through a combination of participant presentations and follow-up summaries provided by email. The range of activities is shown in Table 1.

Mapping the goals of School Engagement.
In small groups, workshop participants were tasked with brainstorming goals for engagement between health researchers and schools based on their experience. These goals were shared in the plenary and group reflection sessions then captured and added to the list of resulting goals. Following this activity, the small groups were asked to select their most important goals and further interrogate the purpose of these goals using a modified approach to the 'five whys' technique 21 , using the tool illustrated in Figure 1.  The purpose of this exercise was to explore 'higher level' or overarching goals for engaging schools with health research. In this exercise, participants placed their most important goal in the lower box, and asked themselves, "why do we want to achieve this goal?" The goal yielded from this question, arrived through group discussion and consensus, was subsequently written into the 1 st box up. This process was repeated up to four further times, or until a range of 'higher level goals' were achieved across the groups. Reflections about these higher-level goals were shared in plenary. Following this exercise, a goal map was drawn and further reflections shared among the group through email.

Mapping participant views about an School Engagement practitioner network.
Similar to the initial brainstorming of engagement goals, in order to gather participant views on what they could share and draw from a School Engagement network, participants in small groups were asked to discuss the following questions: • Why do we need a School Engagement network?
• What skills and resources would you like to draw on and share in the network?
• How can we share materials and resources?
Group responses were shared in plenary allowing all participants to reflect further on the potential utility of a School Engagement network.
Capturing participant experiences and views. All views and reflections were captured on flipcharts and through note-taking. The notes and flipchart data were then summarized and consolidated post workshop. Participants were given an opportunity to reflect on the summaries through email.

Workshop findings
The findings are presented in four areas: School Engagement activities; evaluation approaches; the goals of School Engagement; and the value of a network of practitioners.

School Engagement activities.
Participants described a range of engagement activities facilitated mainly in Africa and south East Asia, but also in other parts of the world including the UK, ranging from in-depth engagement with a few school students over a prolonged period of time, to short one-day visits for larger groups, to large-scale engagement through popular media and the internet. These activities are summarized below and described in more detail in Table 1.
• Student visits to laboratories and exhibitions visits for interactive/practical activities and discussions with scientists; • Fully resourced modules of work, linked to the school science curriculum, taught in school and supported by a hands-on practical activity day at separate purposebuilt educational facility; • Follow-up school sessions and lesson suggestions with related support materials; • Scientists visiting schools -to give health, career and motivational talks to students; • On-line engagement on a platform called "I'm a scientist, get me out of here" where students can ask questions and chat to scientists through text over two-week events; • Primary science clubs where schools are supported to establish in-school science clubs supported by scientist visits and support material; • YPAGs -which involves regular meetings with a group of 10-20 young people in order to facilitate discussion into practical and ethical aspects of research; • Science competitions -this can involve a large number of schools in on-line or face-to-face interactions such as quizzes; • Regular features in popular (national) magazines for children; • Teacher trainings and workshops -to support good science-teaching practice; • A health qualification (related to national UK GCSE level qualifications); • Using comics, videos, hip-hop music and storytelling to stimulate dialogue and learning; and • Creating and performing village based dramas to stimulate discussions about health.
The way in which these engagement activities were rolled out also ranged from single short-term under 2-year projects involving less than five schools, to large longer-term, larger school programmes comprising multi-engagement approaches established over 5-10 years. Clearly, these diverse approaches to School Engagement had a correspondingly broad range of goals.

Evaluation approaches.
Participants described a wide range of evaluation approaches and methods, ranging from clusterrandomized controlled trials 26 to mixed method evaluations combining quasi-experimental designs with qualitative and participatory approaches, such as participatory video 27,28 . With a few exceptions, most projects reported that they had not published their findings in peer reviewed journals, but instead used evaluation to report back to funders and to contribute to learning about the weaknesses and strengths of the projects. This would inform future development of the activities/programmes. The full range of evaluation approaches and methods is described in Table 1.

The goals of engagement between researchers and schools.
These goals of engagement between researchers and schools shared during the workshop can be grouped into four broad overlapping and interrelated categories: community engagement goals; science education goals; research/science capacity strengthening goals; and health promotion goals. Each of these goal categories were articulated as contributing ultimately to two overlying aims: facilitating development in LMICs through encouraging the uptake of science careers among young people; and facilitating ethical research to improve health. The pathways through which the categories of goals contribute to the main overlying aims can be represented as a goal map illustrated in Figure 2.

Research related community engagement goals.
Some of the goals in this category resonate with goals described for community engagement with health research described elsewhere 9 . Goals which aligned to the broader goals of engagement were described in three ways: firstly as contributing an 'operational goal' of helping to facilitate research to influence health policy, through fostering an informed and supportive public; secondly, to address 'intrinsic' engagement goals, such as respecting local culture and creating fora where members of the public/community could express their views about research; and thirdly to address ethical principles of research, through supporting individual informed consent through a better understanding of research, and providing benefits to students/schools in the form of enhancing science education and promoting health. For many workshop participants, engaging school students as members of the community/public was deemed important in its own right but was also described as important in terms of the students' potential to become 'agents of change' influencing wider understanding of and attitudes towards health research among their families, friends and wider communities.
Creating opportunities for discussion, dialogue and debate were described as means of facilitating the students' learning about research, but also in some cases, for example through YPAGs, to facilitate the learning in the other direction: to feed student perspectives into research implementation and nurture researchers' understanding of public/community perspectives.
The underlying purpose of all of these research related community engagement goals were primarily articulated as a means of nurturing a public/community who value, support, participate in and contribute to research. Nurturing a community/public and support for research was described as a contributor to facilitating research to influence health policy, which would in turn, contribute to better health.

Goals aligned to enhancing science education.
Educational goals of interactions between scientists and schools have been widely described in science education literature elsewhere (see for example Braund and Reiss 29 ) and similar goals were shared for engagement between health researchers and schools. These goals included: improving students' understanding of scientific processes (improving scientific literacy); offering opportunities for students to relate school science to 'real-world science'; promoting positive attitudes to science; de-mystifying scientists; and strengthening science teachers' capacity and confidence through offering opportunities to improve their understanding of science.
For most workshop participants, School Engagement was described as a means of drawing from research institutions' scientific or arts backgrounds and resources, both human and material, to contribute to enhancing school students' science education experiences. Through contributing to local education, research institutions could provide a benefit to local communities/publics, which in itself can be also be thought of as addressing a goal of Community Engagement. This benefit, widely described as a being a valid and important output of engagement in itself, was also described by some workshop participants as a means of nurturing a community supportive of research implementation. This highlights overlaps between the categories of goals.

Goals aligned to research capacity strengthening.
Providing opportunities for students to interact with researchers, as well as contributing to addressing science education goals, was viewed as a means of raising students' interest and aspirations for pursuing science and research-related careers. Though overlapping with science education goals, these goals were deemed important contributing factors in themselves for attracting talented young people to science/research-related careers and therefore potentially important in strengthening capacity for future research. Strengthening science/research capacity was primarily described as a means of ultimately improving health, but also as a means of promoting development through science specifically in LMICs.

Health promotion goals.
Underlying health promotion goals was a notion that health research institutions, because of their obvious expertise in local prevalent disease and health, were well placed to support health promotion initiatives such as: empowering young people to make healthier choices to reduce their risk of developing chronic diseases (such as heart disease and diabetes) later in life; raising awareness of sexually transmitted infections including HIV, and strategies to prevent infections; minimizing the risk of water-borne diseases through good hygiene; malaria prevention and childhood vaccination; and antimicrobial resistance. In a similar way, enhancing science education, health promotion were described in terms of both benefiting individual students and in terms of improving community/public health.
Networking. Participant feedback for the workshop evaluation highlighted that learning from others about the range of goals and engagement approaches, and sharing practical and ethical challenges faced during implementation felt very enriching and helpful. As summarised in Table 2, group discussions, subsequent plenary presentations and workshop evaluation feedback yielded four broad categories of motivations for establishing a network of School Engagement practitioners: to draw on network expertise for skill-sharing and capacity strengthening; to facilitate discussions to share experiences; to facilitate collaboration and support for fundraising; and to share resources. These categories and their associated objective raised during the meeting are listed below. Overwhelmingly, delegates felt that regular workshop meetings would provide the best opportunities for achieving goals such as capacity and skills development, sharing experiences, problem solving and brainstorming. Many delegates felt that though several other networks presented opportunities for sharing experiences, the growing diversity of School Engagement approaches merited its own network meeting.
Though delegates felt that face-to-face workshops could not be replaced, several other modes of communication were suggested to supplement this. The online Mesh network was felt to be potentially useful for sharing resources such as theories of change, evaluation documents and activity curricula. It was also felt that group and one-to-one communication could be facilitated through Skype, email, and social media.

Discussion
We acknowledge that this is by no means an exhaustive description of all projects facilitating School Engagement with health research. However, it provides a snapshot of some of the diverse activities conducted in parts of sub-Saharan Africa, South East Asia and the UK. This broad range of activities has diverse goals which can be categorised into four main overlapping groups: research-related community engagement goals; goals aligned to enhancing science education; goals aligned to research and science capacity strengthening; and health promotion goals. More fundamentally, in LMICs all these goals aim at nurturing a country's development through supporting science and facilitating ethical research to improve health.
As well as ranging in terms of approaches and goals, engagement activities also varied tremendously in terms of project 'depth' and 'width'. Holliman and Davies 30 point to the importance of considering the 'depth' of engagement, in terms of relative ratios of students to participating researchers and duration of engagement activity, as a means of "moving beyond the seductive siren of reach". Comparing a range of engagement activities between universities and schools, they argue that the deeper the engagement per student, the more 'meaningful and direct' the interactions will be, but at greater resource costs per student (including researcher time). At our workshop, while 'deep' activities such as YPAGs 19 and attachment schemes 11 , involved prolonged engagement over an extended period of time with low student to researcher ratios, wider activities such as one-off lab-tours for considerably larger student groups (see for example Davies, Mbete et al. 27 ) and the OUCRU Viet Nam National Science magazine initiative (see Table 1), involved much larger numbers (the latter 40,000 students) for considerably shorter interactions. Some activities described in this article could be argued to be both wide and deep simultaneously. For example, the LifeLab project, through a combination of teacher professional development training, student activity days at a hospital facility, and 10 follow-up classroom-based lessons 16 , could be described as both deep and wide interaction. This illustrates that school engagement activities lie along a continuum of deep to wide engagement.
Given that there is a need, particularly in LMICs, to attempt to balance meaningful 'deep' engagement, with the need to distribute potential school engagement benefits to the largest possible number of schools, we feel that both 'deep' and 'wide' forms of engagement are equally important, but arguably serve different purposes. For example, whilst a deeper engagement through regular YPAG meetings, can potentially facilitate the incorporation of unique youth perspectives into research implementation in contexts where their voices are

Overall goals Specific goals and activities
To draw on network expertise for skillsharing and capacity strengthening for school engagement: • Sharing engagement approaches/practice -to learn about new approaches/best practice; • Sharing evaluation/research methodologies -quantitative, qualitative and participatory methods to strengthen the evidence and argument for School Engagement; • Provide opportunities for practitioners to develop presentation and science communication skills; • To offer opportunities to learning about important skills such as stakeholder engagement; and • To create a database/directory of different skills within the School Engagement practitioner group.
To share experiences and facilitate discussions on School Engagement: • To nurture mutual-encouragement through offering participants an opportunity to have their work appreciated; • To provide a platform to share successes, failures and lessons learned (to avoid duplication of mistakes); • To provide a platform to share share, discuss and offer guidance on ethical and practical challenges emerging in School Engagement activities; and • To provide opportunities to brainstorm, share and develop new engagement activities.
To facilitate collaboration and support for proposal writing and fundraising • To draw on credible advisors/consultants/experts for funding applications from the pool of experts within the network; • To better understand the funding landscape and strengthen funding applications and collaborative awards; • Collaboration towards generating new cross-programme School Engagement ideas; and • To grow the network -identify new partners/collaborators, and opportunities.
To share resources • Evaluation tools and documents; • Published literature and meeting presentations; • Training resources (e.g. teacher and researcher); and • Guidance/advice for capacity strengthening and trainings. taken seriously by research leads 20 , there is growing evidence of the benefits of larger-scale short institutional visits to students knowledge, attitudes and aspirations 27,28,31,32 .
During workshop discussions, a great emphasis was placed on the importance of engagement being of 'benefit' to students and schools, and of identifying and minimising any potential inconveniences or disruptions. Regarding benefits, research conducted in LMICs has the capacity of generating collateral benefits for research participants and communities hosting research 33 . While individual research participants may acquire direct 'collateral benefits' such as access to medicine or healthcare, which may not usually be available 34 , communities may also benefit as a result of research institutions' presence in communities through for example, improved healthcare, access to health education materials or the provision clean water and boreholes 35,36 . From the individual researcher or research institution's perspective, contributing to local education can be seen as a collateral benefit. This resonates with previous findings from Kenya where researchers felt that motivating school students in science provided an appropriate means of 'giving back' to local communities in return for their careers in science being supported by local communities 27 . Whilst the importance of engagement being beneficial for students in terms of, for example, inspiring positive attitudes towards science or promoting healthy behaviours cannot be understated, engagement practitioners were also aware that any such benefits must outweigh potential disruption to children and curriculum delivery. Further research is needed to document any unintended perverse outcomes and to strengthen the evidence that engagement is beneficial to students in terms of some of the overarching goals of School Engagement listed in this article: enhancing science education; promoting health behaviour; and upstream science capacity strengthening. Further, though researchers, teachers and students may perceive School Engagement overall as beneficial, it is essential that there is a continued engagement with representatives of broader communities on their views about School Engagement: what are the most valued approaches; and how does School Engagement weigh up against other potential forms of collateral community benefit?
Whilst community engagement with health research has limitations in terms of its capacity to address all community needs 9 , we argue that School Engagement can make a substantial contribution to addressing goals of community engagement, capacity strengthening and health promotion, while at the same time benefitting host communities through supporting local education. The workshop has illustrated the value of networking across a range of projects to highlight similarities and differences between projects and to understand the nature and range of engagement approaches. Participants at this initial School Engagement meeting were mainly restricted to engagement practitioners attached to research institutions, educationalists and science communicators. Whilst questions about future network participants remain, for example: a sole focus on engagement between health researchers and schools or a broader outlook on engagement with science; or, should the network extend to science teachers and children? participant thirst for sharing, learning and collaborating highlighted the value of further developing this practitioner group.

Conclusion
The workshop identified four main School Engagement goals and several School Engagement approaches and evaluations. Participants identified the value of a School Engagement practitioner's network to share resources, lessons and experiences. We argue that given the potential benefits of School Engagement, there is a need to strengthen its practice and research across LMICs.

Data availability
No data are associated with this article. community engagement goals', which is not quite the same thing. In this section, the important point that the categories of goals overlap is rather lost in the middle; all the categories overlapped, so I suggest this could have been accorded more clarity.
The tensions and questions around the competing claims of depth and width in engagement form a sizeable section of the Discussion but are only brought in in this section. Given the wide range of activities summarised in Table 1, this dichotomy could have formed a useful discussion point for the participants, which would make its presence in the Discussion more relevant.
I would like to have seen more evidence from the data gathered during the workshop (the authors say there are no data but the flipcharts, recordings, etc. are all data sources). Hearing the authentic voices of the participants would greatly strengthen the conclusions and recommendations. However, there is no mention of ethical approval from any institution, so perhaps this is why no data are offered.

Does the article adequately reference differing views and opinions? Partly
Are all factual statements correct, and are statements and arguments made adequately supported by citations? Yes

Is the Open Letter written in accessible language? Yes
Where applicable, are recommendations and next steps explained clearly for others to follow? Partly No competing interests were disclosed.

Competing Interests:
Reviewer Expertise: Engaged research, public engagement with research, qualitative methods, facilitation, science communication.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. This is an engaging and accessible article clearly showing learning from a diverse range of practitioners on processes, goals and learning from school engagement activities with health research, across a range of contexts from sub-Saharan Africa, East Asia and the UK. It is an interesting and important read given the growing interest in this area and the relative lack of published literature on this; as stated in the article, most of those involved in school engagement activities had not published their work, but rather used evaluations to inform reports to funders.
The abstract is detailed on the workshop, methods and participants and covers the four emerging broad goals for schools engagement including contributing to science education; capacity strengthening for health research; contributing to goals of community engagement; and health promotion. However, there is less detail on the learning and key points in the discussion (e.g. trade offs between deep and wide approaches and on the depth of engagement) which we felt would be useful to include in the abstract too.
In the background it would be useful to know more about where work on school engagement has taken place, for example, which contexts and where does the literature cited come from? Throughout the paper, there are some inconsistencies in the language used in describing where the work has taken place; for example, the authors refer to work in "Africa" but it would be good to share where in Africa, such as in the references in the background. However, the term sub-Saharan Africa is used in the discussion, which provides more specificity, and Table 1 clearly details where in Africa the work has been done.
It may also be useful to comment on whether there were any differences between the different contexts in prioritisation of the goals concerning school engagement. For example, in Table 1, it seems there was more engagement with art, such as cartoons, story-telling and theatre, in LMICS compared to the UK contexts; while in South Africa and Zimbabwe, there was more of a focus on gender in encouraging girls into STEM careers. Table 1 is a comprehensive summary of engagement approaches, goals, audiences, methods for evaluation and demonstrates the wide range of initiatives and perspectives that were included in the workshop. This is a very useful resource for others who may be interested in taking forward school engagement with health research. Figure 1 and 2 are interesting and illustrate the tools used to map the goals. However, it would be good to expand on the discussions and dilemmas pertaining to these and whether there are any links between the two figures. If so, the title of Figure 2 could be more descriptive if this summarises the themes generated from the workshop with the network of practitioners. The tension and importance of both deep and wide approaches to school engagement are interesting in the discussion and we wondered whether these could also be displayed visually?
Throughout the Open Letter, school engagement is positioned as a benefit to individuals, communities and researchers or institutions. Whilst the authors note that stakeholders highlighted benefits of school engagement activities should outweigh disruptions to children and curriculum, there was little discussion presented about any of the challenges of school engagement activities, especially who is engaged and who is potentially missed. However, the Open Letter helps facilitate a discussion on how school engagement activities can support wider community engagement strategies and highlights the appetite amongst practitioners to support mechanisms which will enable sustained shared learning within this very engagement activities can support wider community engagement strategies and highlights the appetite amongst practitioners to support mechanisms which will enable sustained shared learning within this very important field. There was also an interesting focus on students becoming agents of change and the role they can play in community engagement. It is clear from this letter that establishing a network in school engagement can help facilitate shared learning; now it is important to have a dialogue about how this could be sustained and what stakeholders could contribute. A collaboration such as this could have some important and innovative outputs.

Does the article adequately reference differing views and opinions? Partly
Are all factual statements correct, and are statements and arguments made adequately supported by citations? Yes

Is the Open Letter written in accessible language? Yes
Where applicable, are recommendations and next steps explained clearly for others to follow? Yes No competing interests were disclosed. Competing Interests: Reviewer Expertise: We are 3 reviewers Sally Theobald, Helen Piotrowski, Shahreen Chowdhury with expertise in health systems strengthening, community engagement and equity.
We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.