Engaged scholarship and public policy decision-making: a scoping review

Background Engaged scholarship includes the coproduction and use of research by partnerships that blend research, policy and/or practice perspectives. This way of doing research attempts to bridge-the-gap between knowledge and its application. Recent reviews have described practices that support engagement and involve the community in research and patients in healthcare but there is less known about how to engage individuals working to inform public policy. Aims and objectives The purpose of this research was to articulate the actions and context that support the coproduction and use of research to inform public policy decisions. The study focuses on partnerships between researchers and stakeholders working in public policy across different levels and sectors of government. Methods A scoping review methodology was used. Relevant articles were identified from six electronic bibliographic databases of peer-reviewed literature. Findings A total of 9904 articles were screened and 375 full-text articles were assessed for eligibility. The included 11 studies were from research partnerships internationally and described actions and contextual factors contributing to the coproduction and use of research to inform public policy. Key actions included facilitating frequent interactions with public policy stakeholders, joint planning for research, and collaboration to execute data collection and analysis. Contextual factors included clarity in responsibilities, prior relationships, and mutual respect for partner priorities and perspectives. Conclusions Key actions and contextual factors were identified in this review and warrant further study to strengthen research–policy partnerships and their outcomes.


Background
Engaged scholarship (ES) offers a collaborative approach to research where academicians and non-academicians (e.g. community, policy-makers and other partners) engage in the coproduction and use of knowledge. This way of doing research attempts to bridge the gap between knowledge and action by blending the perspectives of those who produce knowledge and those who use knowledge as part of the research process [1]. Forms of ES have emerged across university institutions to affirm the commitment of academia to the scholarship of engagement and address complex population and social issues [2][3][4][5]. The characterisation and experiences of ES have focused on research, instruction, service and commercialisation and academic experiences across disciplines of health research, social science, engineering, computer science, information technology and business [2,[4][5][6][7]. As a result of the diversity, ES experience and involvement is variable and shaped by discipline, language, institutional culture, individual roles and understandings [3].
Within the domain of ES for research, the focus is on discovery and inquiry in collaboration with a broad set of community partners that are affected by the issues (e.g. community members, patients, professionals, organisations) and/or by decision-makers (e.g. policy-makers, leaders, managers) who may apply research findings [4,8]. With roots in social sciences, ES emphasises multidirectional learning where different expertise is valued and shared to inform better quality and more relevant research [9]. This is distinctly different from a traditional biomedical knowledge transfer paradigm, where researchers are responsible for doing research and communicating the results to end users [9]. Other terms have been used to describe more recent efforts to engage stakeholders in the co-development of research (e.g. integrated knowledge translation, knowledge mobilisation, dissemination and implementation) and each has its own paradigmatic influences and underpinnings [10]. Biomedical roots in health may limit the extent to which traditions from the social sciences and humanities, like ES, are used to understand and improve research partnerships.
Recent reviews involving ES have explored the roles and implications of community partners in research. These reviews demonstrate variability in practices that facilitate engagement and influence community mobilisation and empowerment as a result of the practice of ES, including trust, linkages, training, resources, institutional processes and sustainability [8,[11][12][13][14][15]. Studies have also highlighted factors influencing the use of research by policy decision-makers such as the perceptions of evidence, culture and competing influences, and practical constraints [16,17]. In population health research, engaging stakeholders working in public policy in research is particularly important to ensure that evidence is produced in a timely manner to influence policy decisions [18,19]. ES provides an opportunity to bring together explicit or codified knowledge that is typically represented in scientific literature, with tacit or experiential knowledge that is based on professional expertise and involvement with local communities [20], which can be important in the process of programme planning and decision-making [21]. However, little is known about how those working in public policy, such as government decision-makers and staff, should be engaged in research to bridge the research-to-policy gap. This includes conditions and actions of ES partnerships to optimise public policy and is the focus of the present study.
An extensive and diverse literature addresses questions about research-policy partnerships, particularly in relation to factors that facilitate the use or uptake of research knowledge [22,23]. Yet, it is unclear what conditions and actions of ES partnerships best support the desired public policy outcomes [24]. Scoping reviews offer a review method to summarise the extent and range of literature on a given topic and to identify gaps in the literature [25]. The purpose of this research was to conduct a scoping review to articulate the actions and context that support the coproduction and use of research knowledge to inform public policy decisions. It was expected that this review would identify strategies for stakeholders working in ES partnerships to enhance their collaborative research efforts and ascertain key areas that require future investigation.

Methods
The review was guided by the scoping review methodological framework by Arksey and O'Malley [25] and by further recommendations by Levac et al. [26]. The review followed five key steps, including (1) identifying the research question; (2) identifying relevant studies; (3) study selection; (4) charting the data; and (5) collating, summarising and reporting the results.
The contextual sensitivity of a realist lens was used to inform the research question to identify what ES actions work for whom, in what circumstances, in what way and how [27]. We used the word 'action' rather than 'mechanism', which is traditionally used in realist approaches as the review is not about theory-building related to underlying (invisible) mechanisms [28]. We adapted Denyer and Tranfield's approach to developing a wellformulated question for use in a social science or organisational context, which is also informed by a realist approach and uses the CIAO acronym (context, intervention, actions and outcomes) [29]. Our final research questions were -Under what conditions does ES influence the coproduction and use of research to inform public policy? What are the actions that contribute to sustained ES partnerships between researchers and government stakeholders working in public policy? For the purpose of our review, we defined public policy as a government action or inaction that includes formally approved and implemented goals and regulations, practices and programmes [30]. This research question informed the inclusion and exclusion criteria, which were refined iteratively throughout the search process.
The research question informed search concepts to identify relevant studies, with searches performed using the terms 'Policy Makers' AND 'Partnership' AND 'Research'. The team identified example articles within the target subject area. A search strategy was then developed for use in Ovid MEDLINE and tested for retrieval of the target articles. Once finalised, the search strategy was translated to five other databases (Ovid MEDLINE, Embase, Web of Science, ABI/INFORM Global, PAIS Index, ERIC). Additional file 1 provides a summary of the search translation for all databases, which includes the literature up until May 2017. Following deduplication of identified articles, an iterative process was used for study screening. Two reviewers (KM/BB and JLM) independently reviewed titles and abstracts, followed by the full articles within the Covidence platform. Reviewers met to discuss discrepancies and refine the inclusion/exclusion criteria (Table 1) as needed [26]. For example, as studies were screened, it became clear that it would be important to focus the screening on collaborations that were enduring to ensure they reflected the intent of a ES partnership, rather than isolated meetings without a clear long-term collaborative purpose.
Data from the final articles were charted to extract information according to a priori characteristics, including title, authors, year, journal, country, public policy issues, type of ES partnership described, methods used to describe or evaluate, stakeholders (research, public policy, other), initiator, funding, duration, actions (including duration, frequency, timing), contextual factors (barriers and facilitators), and public policy outcomes. From the extracted data, we used a thematic analysis approach [31] to code data from the included study and identify themes that related to the key conditions and actions to summarise and report the results.

Results
A total of 9904 articles were screened ( Table 2). Of these, 375 full-text articles were assessed for eligibility (see Additional file 2 for PRISMA flowchart).
Articles were excluded during full-text screening if they did not provide details on the actions contributing to their ES process, if the partnership was not between researchers and public policy decision-makers, and if the partnership was not an explicit and enduring research collaboration aiming to inform public policy. For example, some studies were excluded as they were a project without a long-term collaboration or because they focused on participatory policy-making. A final 29 articles were initially included but 18 of these were excluded from further analysis as they were deemed to be structured as a 'research network' with a shared domain of interest and collective outcomes rather than an explicit 'research partnership' that was focused on carrying out specific project activities within a scope of agreed-upon outcomes. The characteristics of the final 11 studies are described in Table 3.
NR none reported, NSW New South Wales The included studies were from research partnerships in various countries, including Australia (n = 3), Canada

Context
Includes researchers and individuals working in public policy in any level of government (local/ municipal, provincial/state, federal/national) across low-, middle-and high-income countries and descriptions of their environments (socio-cultural, political, economic); may also include other stakeholders (e.g. community, providers/ practitioners, patients/public) Does not involve researchers or individuals working in government (e.g. community, organisational, administrative or clinical); no description of environments Intervention (in this study meaning the engaged scholarship partnership) Description of experiences working together toward the same end goal; focused on collaborative coproduction and use of knowledge (e.g. involvement in teams, developing research/policy questions, designing and conducting methods, disseminating results); enduring collaboration that goes beyond one project/meeting Researchers working independent of decisionmakers and vice versa; only translating information to public policy decision-makers, focusing only on uptake of evidence or policy relevance; a network without an enduring mutual purpose; an isolated meeting or project without a long-term collaboration; focus on participatory policy-making rather than on using research to inform policy; focus on a network without information about the specific use of research to inform policy-making

Actions
Clear and concrete examples of processes/ steps that outline what was done to engage and support collaborations Not a direct account of an experience of engaged scholarship (no or vague details on the process of how it was implemented such as a commentary or descriptive paper on the topic of engaged scholarship)
None of the included studies explicitly used the term 'engaged scholarship' but were deemed to fit the inclusion criteria for an ES partnership. Various terms were used to describe the partnerships, including action research, alliance, research or academic policy/practice and collective research. Various public policy partners were engaged in research, including those from national, state/provincial and local governments or authorities. Seven studies also engaged other stakeholders such as those from non-government or community organisations, research councils and practitioners. Six included studies were descriptive case studies, without evaluation methods used to systematically collect information on the partnership process or outcomes. Four studies used process evaluation methods to collect information on the partnership, including participant observations, surveys, process diaries, reflective dialogues and meeting notes. One study used a framework to analyse the characteristics of case studies using a framework for research-policy linkages. Various actions and contextual factors were identified in the included studies as contributing to ES partnerships. With respect to actions, the importance of frequent interactions with public policy stakeholders and co-planning and executing research was discussed as being important in all partnerships. Common methods for interactions included regular email communications, inperson meetings or teleconferences. One study also described weekly visits to government offices which allowed for ongoing and face-to-face discussion with policy actors [38]. Several partnerships held capacitybuilding for stakeholders such as training for data analysis [33], conferences with stakeholders [34] and a national dissemination workshop to share findings of the research [41]. Steering groups, coordinating or management committees and working groups were discussed in several studies as facilitating partnership development [34][35][36]. Bates et al. [36] further developed a governance structure with members responsible for administrative and operational activities of partnerships and a communication plan. Several studies also discussed appointing team leaders/champions [34,[36][37][38]42] or expert/technical personnel from stakeholder groups [32,36] to support partnered research. Communication planning and priority-setting were other reported actions within the included studies [37,39].
Clarity in responsibilities and respect for partner priorities were key contextual factors that influenced the partnerships in the included studies [33][34][35][36]38]. One study discussed the lack of clarity of their action research methodology, which required more meetings to guide the management team [37]. Tran et al. [41] commented on the differences in perspectives that may create challenges for collaboration, whereas other studies discussed the importance of flexibility in partnerships to allow for acknowledgement and appreciation of cultural differences among stakeholders [35,38]. Prior experience or relationships among stakeholders and credibility of the researcher were reported to also facilitate partnership development [34,38]. Contextual limitations related to timing or readiness of stakeholders and continuity of policy actors was also discussed in one study [39].

Discussion
This study used a scoping review methodology to articulate the actions and context that support the coproduction and use of research knowledge to inform public policy decisions. The purpose was to identify strategies for researchers and policy stakeholders working in ES partnerships to enhance their collaborative research efforts and ascertain key areas that require future investigation. We intentionally used the term 'ES' to connect it with an engagement paradigm, compared to a knowledge transfer paradigm, with its biomedical roots and tendency to focus on communication and dissemination by researchers [9]. In contrast, ES focuses on        [1]. Notwithstanding this intent, none of the included studies defined their partnership as ES. However, they all met the inclusion criterion that operationalised ES, namely a partnership where researchers were working with individuals working in public policy toward the same end goal of coproduction and use of knowledge. Key actions included facilitating frequent interactions with public policy stakeholders and joint planning and execution of data collection and analysis. Critical contextual factors influencing partnerships were clarity in responsibilities, prior relationships or experiences, and respect for partner priorities and perspectives. The application of an ES approach is expected to result in more relevant research through coproduction between researchers and policy actors, thereby increasing the likelihood that the evidence will be used to inform policy decisions. Although the included studies discussed how the results would lead toward public policy decisions, the outcomes reported in the included studies focused on academic outputs, including research papers, dissertations and conference presentations with some additional policy-relevant products such as presentation at local forums, individualised reports or local discussions with stakeholders. None of the included studies specifically articulated how the research led to a policy change; however, this could be due to the nature of timing of the publication and that policy outcomes from ES may take more time to achieve.
The partnership, actions and context were often incomplete and inconsistency was reported. In particular, we were interested in understanding the 'embeddedness' of the researcher in the policy environment. 'Embedded research' is one action that may facilitate ES as it situates a researcher within a policy setting to conduct evaluation and research as a member of the host organisation [43,44]. This enables the researcher to be independent from the host organisation but familiar with the policy context, providing an opportunity to engage in critical analysis and bridge knowledge-to-action gaps. For example, Newman et al. [38] described weekly visits to government offices allowing for face-to-face discussion with key policy actors and joint work. The study referred to key elements critical to the success of researcher and public policy partnerships, including developing the relationship and creating the process and knowledge together. These findings reflect the potential importance of greater embeddedness of research within policy, although the other studies in the review did not provide sufficient detail on the context of the partnership to determine the extent to which researchers were embedded in the policy environment. Further, half of the included studies did not use an evaluation method in their study (including Newman et al. [38]), meaning that the actions and context were not systematically collected and may be biased through the accounts of the authors. As a result of these limitations in the literature, it is not possible to determine the relationships between actions, context and outcomes for ES. This is a similar finding by Gagliardi et al. [14] who advocated for more longitudinal approaches to determine the impact of partnerships in research and relationships between approaches and outcomes. The results from this scoping review suggest that further study is needed to more systematically understand the relationships between contexts and actions. It is also recommended that future studies articulate the nature of ES partnerships, including the embeddedness of researchers within policy environments, and use evaluative methods to collect perspectives from the various stakeholders involved to reduce the bias of researcher accounts.
This study is a novel contribution to the literature by focusing on partnerships between researchers and public policy decision-makers on a variety of topics. Previous reviews have explored healthcare research partnerships [14,45,46] and community-based participatory research [8]. The scoping review identified public policy issues across multiple sectors, including early childhood, mental health, health promotion, healthcare, social exclusion and safety. This transdisciplinary approach to the review provides an opportunity to learn about partnership approaches across sectors. An additional 18 studies were excluded due to their structure as a 'research network'. This boundary was necessary to keep the review focused on partnerships that carried out specific project activities with agreed-upon outcomes rather than broad collectives that engage researchers and public policy stakeholders with shared domains of interest but without tangible activities. However, the research network studies were charted in the preliminary stages of the review and similar actions and contexts were reported. These studies also represented multiple sectors, such as health policy, healthy aging and long-term care, public education, regional water issues, wildlife conservation and family violence. One noted difference in the 'research network' studies included further actions related to stakeholder capacity and relationship building, such as workshops and large assemblies. Additional contextual challenges were also mentioned that related to tension and conflicts working within a large research network and balancing the needs of different stakeholders.

Conclusions
In conclusion, more systematic study of the conditions and actions that influence the coproduction and use of research is needed for a diversity of public policy issues. ES partnerships could be further explored beyond the focus on public policy issues to determine the use of this partnership approach within practice-oriented literature. Future research could build on the inherent limitations of the scoping review methodology given their exploratory nature and uncertainty about interpretation due to the lack of quality appraisal [47]. More comprehensive descriptions (including the duration) and evaluations of research partnerships are also needed to elucidate the contextual and process factors that contribute to the research-policy partnerships and to determine the longterm outcomes of its use, given the limited available evidence [24,48,49]. This advancement in the literature would help to strengthen research-policy partnerships and their intended outcomes.