Developing domestic violence primary prevention capacity through a community of practice project: Learnings from Alberta, Canada

Abstract Domestic violence practitioners and community organizations often lack the capacity to engage in primary prevention activities. In part, this limited capacity exacerbates the gap between evidence-based research and practice, necessitating innovative initiatives specifically aimed at user uptake. Using a community of practice (CoP) model within two distinct communities in Alberta, Canada, we sought to translate research knowledge on domestic violence prevention and build primary prevention capacity with practitioners. One hundred twenty professionals from various sectors attended CoP sessions, with 20 attending all six sessions. Data was collected using in-depth semi-structured telephone interviews. Interview findings include that face to-face learning was effective for deeper understanding and building networks across sectors, as well as supporting new aspects of prevention work that had not been previously considered. Findings also indicate that skilled facilitation increased CoP effectiveness, particularly where community context was considered in relation to the topics presented. Impacts include changes to discourse, priorities, and resource allocation to support primary prevention. Areas for improvement include a slower pace of information delivery, and increased focus on policy and system changes. Overall, using a CoP model seemed to support knowledge translation and practitioner capacity building in the area of domestic violence prevention. Considerations for future utilization of this model are explored.


PUBLIC INTEREST STATEMENT
The importance of prevention approaches to domestic violence has been recognized, although there remains a gap between research and practice. Our research findings show Community of Practice (CoP) models to be a promising way in which to build individual practitioner primary prevention capacity. Findings specifically highlighted the ways in which the CoP model built common language and develop shared meaning between practitioners in a variety of sectors as well as changes in resources allocation to support more primary prevention activities. More time between CoP sessions and building explicit content on policy and systems change were deemed important improvements to future iterations of this model. These findings are of particular interest to organizations working in the area of domestic violence primary prevention, researchers seeking to support primary prevention research translation and knowledge mobilization, and system leaders seeking to implement research-based findings in their local context of domestic violence prevention.

Introduction
Domestic violence (DV) in Canada is a serious and prevalent issue, representing more than a quarter of all violent crimes reported to the police in 2013 (Statistics Canada, 2015). Alberta has the third highest rate of self-reported spousal violence in the country, with rates remaining higher than the national averages (Statistics Canada, 2016).
Recognizing lack of practitioner capacity to practice primary prevention as a significant barrier, this paper evaluates a Community of Practice (CoP) model to increase capacity for, and knowledge translation on, primary prevention approaches to DV among stakeholders in two regions of Alberta, Canada. CoPs can be defined as "groups of people who share a concern or passion for something they do and who interact regularly to learn how to do it better" (Canadian Health Services Research Foundation, 2005, p. 15). In this way, CoPs are a promising way to build individual prevention capacity (i.e. motivation, awareness, knowledge and skills; Spoth et al., 2004) and knowledge (i.e. by effectively moving knowledge from academic research to community implementation). applicable research question, and produce knowledge that is neither timely nor adequately communicated to practitioners (Allen, Peckham, Anderson, & Goodwin, 2007).
Given this limitation, researchers increasingly identify interactive, community-centered models as a promising method to translate research into practice (Estabrooks, Thompson, Lovely, & Hofmeyer, 2006;Flaspohler et al., 2008;Miller & Shinn, 2005;Wandersman, 2003;Wandersman et al., 2008). In order to facilitate this reciprocal translation, Mitton, Adair, Mckenzie, Patten, and Perry (2007) suggest that the following occur: (a) face-to-face exchange between decision-makers and researchers; (b) education sessions for decision-makers; and (c) the development of networks of community stakeholders. In addition to this overall structure, the research to be translated must also be understandable and accessible so that laypeople and key stakeholders (i.e. practitioners, service providers, system leaders and policy makers) can apply it to their local contexts (Macdonald et al., 2008;Miller & Shinn, 2005;Thomson, Schneider, & Wright, 2013;Wandersman, 2003;Wandersman et al., 2008).

Knowledge translation and connection to communities of practice
CoPs represent one method to create reciprocal knowledge exchange among research and practice, and are a rich way of sharing best practices and creating new knowledge (Andriessen, 2005, Cambridge, Kaplan, & Suter, 2005Estabrooks et al., 2006;Thomas, Kellogg, & Erickson, 2001). Key CoP concepts are that learning is social (Estabrooks et al., 2006;Li et al., 2009), that CoPs are the product of an interaction between people and the systems in which they participate (Macdonald et al., 2008;Wenger, 2000), and that CoPs come together to problem solve, share experiences, learn, and build professional capacity (Andriessen, 2005;Canadian Health Services Research Foundation, 2005;Hoadley, 2012;Lathlean & Le May, 2002;Thomson et al., 2013). Given their role in building capacity and translating knowledge effectively, a CoP model was deemed ideal for the described project. Through socially situated learning through interaction and dialogue with others in a given community (i.e. practitioners working in the area of DV), the gap between what practitioners know and what practitioners do could be addressed (Thomson et al., 2013). CoPs can promote knowledge transfer not only within networks, but also across networks, where participants represent a variety of roles, such as practitioner, policy-maker, researcher, etc. (Thomson et al., 2013). This potential for CoPs to facilitate both locally-created and globally-created knowledge across networks of diverse participants and strengthen working relationships is particularly salient in relation to knowledge translation.
CoPs are usually self-organized, but are sometimes initiated and facilitated by sponsoring organizations (Jennings Mabery, Gibbs-Scharf, & Bara, 2013). It has been argued, however, that bottom-up strategies (participants in the CoP sharing and disseminating evidence-based practice) tend to be more successful than top-down strategies (issuing guidelines and protocols; Braithwaite, Runciman, & Merry, 2009). Ensuring members have the ability to define the topics and domains of interest is another key aspect to a bottom-up approach to CoP design (Thomson et al., 2013).

Profile of the two communities for the CoP pilot
Two communities in Alberta were selected for the pilot. Both had strong DV networks, and were willing to participate in a primary prevention CoP. CoP #1 had an existing relationship with Shift, while CoP #2 contacted Shift to request capacity building assistance.

Description of CoP #1: Grande Prairie and Area -Northwest Alberta
Grande Prairie is the largest city in Northern Alberta. The population is over 55,000, with 180,000 people living in the surrounding area. From January to October 2010, 830 calls were made to the local police DV unit and in 2013, Victim Services opened 429 DV files (J. Suddaby, pers. comm., 8 September 2014).
The Grande Prairie and Area Prevention of Family Violence Network was informally established in the early 1980s as an information sharing and planning body. Their current mission is to end family violence through cross-sector education; stakeholder engagement and feedback; provision of resources, services, and supports; increased public awareness; and, advocacy. The network comprises 43 members representing 23 community service providers with a variety of professional backgrounds (e.g. social workers, child care workers, mental health professionals, etc.) who meet monthly (H. King [Grande Prairie Domestic Violence Coordinator] pers. Comm., 10 September City of Grande Prairie, 2014).
In total, 64 people from both the DV network and the community participated in one or more CoP sessions, representing a range of community organizations and systems, such as the police, the municipal government, health services, and school districts. Community stakeholders were invited to participate. A core group of 12 network members attended all six sessions, while eleven individuals participated in two sessions and forty-one attended a single session. In 2013, over 230 individuals and organizations from the Southeast participated in the Kind Communities Alliance (KCA), a "collaborative, action focused leadership coalition committed to enhancing our community's ability to respond to and prevent family violence and bullying" (Rieger, 2013, p. 1). The goals of the KCA were to end family violence in the region, prevent future incidents from occurring, support healthy and positive relationships, and work towards changing the stigma of family violence (Rieger, 2013).

Description of CoP #2: Medicine Hat and Southeast region
For the CoP pilot, 56 individuals from the KCA participated in one or more sessions, representing sectors such as health, education and social services. Of these 56 participants, a core group of eight individuals attended all six sessions; and an additional five attended all but one session.

Development of the community of practice model
Principles for the CoP model used in this project to increase capacity and knowledge translation were: • Practitioners require enhanced capacity in DV primary prevention to increase effectiveness.
• The best way to build capacity is to support practitioners to access and understand research findings and to apply findings to their work, local contexts and communities.
• This model will support changes not only at the individual practitioner level, but also through the organizations, systems, and policies at play in the community.
For the purposes of this project, the knowledge translation framework developed by Jacobson, Butterill, and Goering (2003) was used to support the development of the CoP model. The framework for knowledge translation outlines five domains to consider for success, specifically: (1) the user group, (2) the issue, (3) the research, (4) the researcher-user relationship and (5) dissemination strategies (Jacobson et al., 2003). Each domain includes considerations when developing models for effective knowledge translation. This is intended to help researchers and knowledge brokers identify the perceptions, characteristics and needs of their particular user group. Given that social networks within CoPs are not always easy to build and sustain, this project was careful to consider that the right people were brought together and that certain roles were filled. Of particular relevance for this project were considerations around: • where the group is situated and how it functions (e.g. local and historical contexts, decisionmaking processes, structure, etc.), • relationship to the issue of DV (e.g. women's emergency shelter workers, RCMP, addictions counselors, health practitioners, etc.), • group's orientation to research (e.g. preferences regarding the use of research, perception of research applicability to practice, etc.), • researcher-user relationship (e.g. basis of trust and rapport with the group, agreement on desired outcomes for the initiative, etc.), and • preferred dissemination strategies (e.g. high levels of communication and interaction, customization according to practice context, etc.).
These domains and reflections by the research team regarding each domain, were instrumental in developing a model that was flexible to the local context and communities of the two CoPs. Through understanding the importance of user context in relation to knowledge translation, key information could then be tailored and disseminated to the needs and perspectives of practitioners. The ability of the local communities to define the topics of interest and ensure that potential participants could connect to the issue at hand is a key quality of CoPs (Wenger & Snyder, 2000). Furthermore, the researchers believed this bottom up approach to CoP design and structure to be a critical element in local communities to support interest in, and capacity for, DV primary prevention activities in Alberta.
Prospective participants received an information package describing the CoP opportunity, requirements (e.g. pre-reads, meetings etc.), and potential benefits (e.g. increased awareness of primary prevention, increased capacity to engage in primary prevention activities, increased networking). A survey of possible topics was distributed as a "pull" strategy (Macdonald et al., 2008, p. 8) in order to incorporate members' interests including, (a) the role of policy and legislation; (b) engaging men and boys; (c) the role of alcohol outlet density and consumption; (d) children exposed to or experiencing DV; (e) how to improve healthy relationships for youth/young people; (f) role of a coordinated community response and connection; (g) working with ethno-cultural communities; and (h) working with LGBTQ communities in order to incorporate members' interests.
Based on member responses to the "pull" strategy, six full-day, face-to-face sessions were developed and delivered over an eight-month period. Sessions were held in Grande Prairie and Medicine Hat, with practitioners from the more rural areas travelling in. Of the six sessions delivered at both sites, four were on the same topics. Knowles Adult Learning Principles (Knowles, 1984;Knowles, Holton, & Swanson, 2005) were used as guidelines for these sessions, namely: (a) adults are internally motivated and self-directed; (b) adults bring life experiences and knowledge to learning experiences; (c) adults are goal oriented; (d) adults are relevancy oriented; (e) adults are practical; and (f) adult learners like to be respected.
Shift's Director traveled to both regions facilitated the CoP sessions. Content specialists were brought in to further explore topics with the CoP group, and provided additional readings and resources. Social opportunities for learning and sharing-a critical feature of CoPs (Li et al., 2009;Thomson et al., 2013)-were provided through extended coffee and lunch breaks. The CoPs also facilitated group discussion, in order to encourage participants to critique and integrate the day's information at both individual and community levels.
Evaluation goals were to document and test this model for knowledge translation and capacitybuilding in DV primary prevention, and to identify improvements for subsequent CoPs. Evaluation data was collected from post-session, investigator-designed surveys. This was supplemented with information collected from in-depth semi-structured telephone interviews with a sample of participants after the conclusion of the last session of the CoP and again six-months afterwards. Interview questions were designed based on two domains from Cambridge et al. (2005): learning and developing knowledge, and developing relationships. Surveys and interviews aimed to capture the quality and relevance of each information session. This paper presents the interview data, as the focus was on gaining feedback of the CoP model and evidence for capacity change.

Interview recruitment
CoP members were provided with information about the evaluation, and interview sign-up sheets were circulated at the beginning of sessions three, four and five, in order to allow those who might be interested in participating enough advance notice to their calendars. Those who expressed interest in being interviewed were asked to provide contact information and to select an interview date and time to take place within two weeks of the final CoP session. A follow-up interview took place six months after the first interview.
In total, 18 members expressed interest and participated in interviews (seven from Grande Prairie and area, or 11% of participants, and eleven from Medicine Hat/Brooks and area, or 20% of participants). All of the participants attended four or more of the CoP sessions. Participants represented municipal government, non-profit organizations, provincial family violence and bullying bodies, faith organizations, education and early childhood initiatives, and health services. Three participants were men, and sixteen were women. Two of the participants could not be reached for the follow-up interview.
This project was funded by the Social Sciences and Humanities Research Council (SSHRC) Public Outreach Grant. University of Calgary Conjoint Faculties Research Ethics Board approval was sought and obtained. Both written information and verbal explanations about the project study were provided, and informed consent was verbally provided before telephone interviews began. Participants were free to withdraw at any point during the interview process. Information obtained during the evaluation was anonymized and treated as confidential. Approval for this project was granted by the University of Calgary Conjoint Faculties Research Ethics Board, File #7371 10 July 2012.

Data collection
Telephone interviews were approximately 45 min in length for the post-project interview, and 15-20 min in length for the follow-up. Telephone interviews were used as a matter of convenience since participants were located in areas of the province three to eight hours away from the researchers. Interview questions reflected evaluation framework items based on the six components identified in the description of the evaluation design, and sought to capture both the immediate effects of participating in the CoP (e.g. relationships built, increased knowledge and awareness of primary prevention), as well as perceived longer-term impacts (e.g. changes in resource allocation, organizational priorities, policies, or practices; see Tables 1 and 2). Interviews were audio-recorded and conducted by Shift researchers (first and third authors). Notes were taken during each interview, and the audiorecordings were transcribed verbatim.

Data analysis
A directed content analysis approach (Hsieh & Shannon, 2005) was chosen for this project. This method tests anticipated patterns and relationships to extend a framework or theory (i.e. using CoPs to support knowledge translation and build capacity in DV primary prevention; Hsieh & Shannon, 2005;Berg & Lune, 2012). Only transcribed audio-recorded data was analyzed and transcripts were organized and reviewed manually, allowing the researchers to remain close to the data.
The six components of the evaluation framework were used to categorize and code interview content (model, engagement, learning, relationship-building, reach, and impact). Transcripts were assessed for concordance with the codes, with subcategories developed as needed. For example, model elements were broken down into format, flexibility, and facilitation. Codes were grouped into themes by the researchers who both re-examined the transcripts and each code in detail, cross-checking with each other for consistency on each code and theme. In this article, the focus was on the overarching themes expressed by members of both CoPs on the effectiveness of the model emerging from both post-session data and themes emerging from the follow-up interviews on longer term CoP impacts.

Results
Themes that emerged from the analysis related to the six evaluation components are detailed below.

Format
As both CoPs were in rural and/or remote regions where professional development opportunities are less accessible, participants appreciated that CoP sessions were conducted face-to-face and that content experts came to them, as opposed to them having to go to the content experts. As stated by one participant: … in person was the best and we really appreciated that extra sort of people coming into our community instead of having, oh we just have to go somewhere else that costs us a lot more money.

Facilitation
Participants felt that the facilitators chosen for this project enhanced knowledge translation and capacity building. As noted by one participant: I think [she, the facilitator] was very sensitive to the fact that each community has unique needs and will implement things in their own, within their community's context. Right? It highlighted we're going do things a little bit different maybe but here's some of things that other people have done in our communities if you need to tweak it a bit so it works with yours better you're ok to do that. That's the message given. (Domestic Violence Sector worker, Medicine Hat) Given the comment above from the CoP member, trust in the facilitator appears to be a key factor in supporting knowledge translation and capacity building.

Engagement and learning
Most interview participants from both regions felt that the learning experience was motivational, enhanced their awareness, and supported their development of primary prevention knowledge. They reported increased knowledge in areas such as engaging men and boys in DV prevention, school-based efforts to promote healthy relationships and prevent dating violence, and bystander interventions.
Several CoP members from both regions also described that in their experience, DV initiatives are generally focused on treatment and intervention. The opportunity to focus on primary prevention https://doi.org/10.1080/2331205X.2017.1333697 was, thus, helpful, especially as practitioners can get "stuck" in a crisis/response model. One health care system worker stated: We do have the tendency in our community to focus on treatment and intervention…I feel like even within our organization there has been a lean towards treatment. So this focus on [primary] prevention, and definitions so that we have a common understanding of what is prevention and strategies, you know, was amazing. (Health care worker, Grande Prairie) Interviewees indicated that an increased understanding of primary prevention helped them to think differently about their work. The director of a non-profit organization said: It was great, really made me rethink some of the things we were doing and even help identify some gaps that we didn't even realize we had missed in our work. (Nonprofit organization director, Grande Prairie) For participants, reflecting on their current DV prevention work helped them consider assumptions, patterns and gaps in activities, and to move forward in an informed way. Nevertheless, respondents found the pace to be challenging at times. One stated: Need to slow down -too much information in 1 day. It is an efficient model, but is it as effective as it could be? Not enough time for conversation, and need to pull the different segments of a region together. (Local funder, Medicine Hat) Based on the above feedback, a slower pace may provide practitioners with the time they need to adequately reflect on their own practice, and begin implementing the new knowledge to fully change their practice.

Relationship building
Interviewees reported that the CoP enhanced existing relationships, and developed new, strategic relationships with a range of stakeholders. As noted by one participant: It's been useful in terms of interacting with the other agencies and groups in town that are wanting to tackle the same issue. Cuz you know, now at least we're all speaking from the same platform in terms of we've seen the presentations together, we've taken the information in. (Local funder, Medicine Hat) Despite coming from the same region, many participants did not typically work together. One participant working in early childhood intervention stated: The amount of people in the room that normally aren't together…I thought that was really useful. I'm with an early intervention program so I don't typically work with kind of high risk in the domestic violence sense, but because I work with a large number of moms in the community, I felt that it was important that I should be there. (Nonprofit organization program manager 1, Medicine Hat) Thus, these CoPs facilitated new connections and introduced new aspects of prevention work.

Reach of information
New information travelled through participants' organizations and networks. As one participant remarked: I came back and shared things with my staff almost after every meeting. Just about "did you know this was going on?" or did you know, or some of the research…so it was a good opportunity for me to share certain information with my staff too.

Impact
Participants noted changes to views on and conversations about DV prevention in their respective communities. As one interviewee commented, "now we have foundation for common language. Everyone's starting point is now the same-current prevention knowledge." Information and resources for school-based approaches to healthy relationships and dating violence prevention also led to the implementation of the evidence-based 4th R (Relationship) program in both districts. As one participant explained: The other quick win of course was the school district buying into the 4th R and hopefully that's going to continue to roll out across the region. We will certainly support that. (Nonprofit organization program manager, Grande Prairie) The follow-up interviews suggested that information from the CoP was still percolating six months later. Although no concrete actions or changes had occurred in many areas, members were keeping the ideas in the forefront of their minds. As shared by one participant: If we're gonna be doing anything we should be tackling the alcohol accessibility you know. That's the biggest bang for the buck, based on what was presented there and there are a few others feeling the same way so we are planning on getting together a little bit more to see what we can do as a community to address that. (Municipal government representative, Medicine Hat)

Discussion
This evaluation suggests that the CoP model used in this project was of value to the participants, both in terms of individual learning, and in developing relationships and seeding wider scale change.

Use of CoP model to support knowledge translation and capacity building
While there is mounting interest in capacity building for the primary prevention of DV, less is known about how this capacity should be developed (Feinberg, Greenberg, Osgood, Anderson, & Babinski, 2002;Flaspohler et al., 2008). Participant interviews suggested that our CoP model was an effective way to build elements of individual primary prevention capacity, such as building awareness and increasing knowledge. Providing opportunities to connect with others working in different areas of prevention and reflecting on gaps in current prevention efforts was perceived as a significant benefit of participating in the CoP process. Thus, this CoP model appears to be a promising practice to translate knowledge in a way that builds capacity at both the individual practitioner level and the local community level.
Situated learning and opportunities for relationship building are also important aspects of knowledge translation and capacity building through CoPs (Lees & Meyer, 2011;Li et al., 2009;Urquhart et al., 2013;Wenger, 2000). Our findings strongly suggest the benefits of a face-to-face format, enhancing networking opportunities and deepening understanding by learning from each other. Several studies have examined the use of web-based technology and video-conferencing to support professional development with practitioners. In a study of practitioner networks done by LaMendola, Ballantyne, and Daly (2009), participants privileged face-to-face communication over online communication, suggesting that online platforms worked only as a supplement to face-to-face contact. A 2010 study by Church et al. found mental health practitioners to be somewhat dissatisfied with the use of technology to support interprofessional learning, given the challenges to building facilitator and participant rapport. Comments from participants in the CoP appear to concur with findings from other studies. Certainly, the groups felt that the opportunity to interact face-to-face supported community capacity building around DV prevention in way that technology would not have supported. This format may be more applicable to practitioners with a strong verbal culture (such as social workers; Harrison, Hepworth, & De Chazal, 2004). It may also be more important in early stages of the CoP in order to help groups familiarize and build trust (LaMendola et al., 2009).
Social networks are a foundational aspect of CoP models (Li et al., 2009;Thomson et al., 2013;Urquhart et al., 2013;Wenger, 2000), and CoPs can support boundary crossing and access to other networks (Lees & Meyer, 2011;Probst & Borzillo, 2008). While there may be an assumption that rural practitioners are well connected and networked with each other, research indicates this may not always be the case (Church et al., 2010). The CoP model provided participants with the opportunity to connect and interact with professionals in a variety of organizations and sectors, allowing individuals to find out about others' work and areas of expertise. These intra and inter sectoral connections were perceived to be particularly valuable in thinking about DV prevention work (Table 2).
Research indicates the importance of facilitation in the success of a CoP (Lees & Meyer, 2011;Li et al., 2009;Probst & Borzillo, 2008;Walczak & Mann, 2010), and this factor was found to be important to our CoP participants. Church et al. (2010) found that the relationship between the facilitator and participants to be a central element in the potential success of interprofessional continuing education models.

Changes as a result of participation in the shift CoP
Of particular interest to Shift was whether enhanced capacity and the uptake of knowledge supported changes in discourse, organizational practices, systems, and policies in DV prevention. Participants indicated a marked shift in discourse within their communities. This, in turn, supported CoP members to use common language and develop shared meaning that may facilitate the exchange of ideas and knowledge around DV prevention objectives and activities (Thomson et al., 2013).
In terms of practices, systems and resource allocation, school-based healthy relationship programming was the most notable change. School districts in both communities connected with CoPs for information about violence prevention and subsequently implemented the 4th R program (Crooks, Wolfe, Hughes, Jaffe, & Chiodo, 2008). Both CoP regions requested training on the 4th R to implement in their school jurisdictions. Respondents also provided examples of practice and priority changes within their own organizations as a result of CoP participation. For some, DV primary prevention had not previously been part of their work, while for others, recognizing the importance of dating violence prevention with adolescents propelled them to build strategies in this area. Still others spoke about the need to move organizational resources to support prevention activities. Thus, the impacts of the CoP went beyond individual participants and led to changes in their agencies and organizations (Jennings Mabery et al., 2013;Schenkel & Teigland, 2008).

Areas for improvement
The number of topics was too ambitious for some participants given the eight-month CoP timeframe; this may have also been a reason for the lower than expected number of practitioners who attended more than one session. Time for reflection has been shown to be a component in supporting positive changes to service delivery (Church et al., 2010). Perhaps the number of topics in the designated timeline was too rapid a pace to support adequate reflection, and so it is possible that fewer topics over a longer length of time would be beneficial to the CoP model.
There was also limited evidence of systems and policy changes as a result of CoP participation. Although change of this type is gradual and slow, the sessions could be more explicit on the potential for policy and systems change, in order to help participants start this work.
Through follow-up interviews, findings revealed a decreased pace of momentum by the CoP members in addressing ideas and initiatives that had been raised during the CoP sessions. In some cases, participants wanted researchers to schedule a "booster" session with the CoP in order to keep momentum and energy going. Unfortunately, this was not planned for in the project timeline or budget, meaning that researchers could only connect via telephone and email after the sessions were completed.

Conclusion
Capacity and knowledge translation have been highlighted as critical aspects of creating the conditions for the primary prevention of DV. To support DV stakeholders in these areas, we presented a CoP model that was tested in two remote and rural Alberta communities. Evaluation interviews emphasized the importance of a face-to-face format and good facilitation that fosters engagement in the learning and opportunities for reflection. Impacts included changes in discourse, priorities, and resource-allocation, as well as information sharing with other stakeholders and enhanced social networks. Based on results of this pilot study, future versions of this CoP model will cover fewer topics. This is a significant change to the model, given the research suggesting that a slower pace provides practitioners with the opportunity to more deeply reflect on the professional development material. Building more purposeful content around policy and systems change is another important shift to future versions of this CoP model, especially since the current models lacked explicit discussion on the potential for policy and systems change in the province. Another consideration for change would be to have researchers connect with the CoP members intermittently after the official topics had been covered in order to support group momentum. This may be an important consideration for future knowledge translation and capacity building efforts, as participants clearly valued the social and intellectual contributions of researchers to support momentum in their communities. Overall, however, the CoP model appears to be a promising way in which to support DV primary prevention knowledge translation and practitioner capacity building.