for Opioid

Forming equity-based community-academic partnerships focused on recovery research is a time-consuming and challenging endeavor, but one well worth the care and effort required. Through building trusting relationships, vital research collaborations emerge, which are driven by expressed community needs and supported with university resources. This article describes the stakeholder engagement process utilized by a university-based and opioid-focused initiative entitled Innovations in Recovery through Infrastructure Support (IRIS). IRIS developed a diverse and representative network of clinical providers, peer recovery workers, academics, and other behavioral health leaders. The process was informed by community-based participatory research (CBPR) practices and principles aimed at creating equitable partnerships. Lessons learned include the need to reshape the relationship between research and the community through an acknowledgment of harms committed by academia, as well as the importance of maintaining an approach of humility, accountability, and patience with the partnership process. Concrete benefits that go beyond the long-term promise of change, including compensating partners financially for their time, help ensure equity. A commitment to always asking “Who’s missing?” and then filling those gaps builds a broad network inclusive of the various constituencies that make up the recovery support system. As IRIS builds on these lessons learned and plans next steps, we share our experience to support others engaged in forming community-academic partnerships through deep stakeholder engagement and use of participatory approaches within and outside of recovery research.

Community-academic partnerships can help to close the gap between research and practice in health care by actively engaging community stakeholders and leveraging the firsthand insights they bring to pressing issues (Drahota et al., 2016;Michel et al., 2013). The opioid epidemic is an example of an urgent health issue that cannot be addressed without the active and equitable engagement of multiple stakeholders, including those with lived experience, community providers, policy leaders, and academic researchers.
The opioid crisis has been devastating, with over 550,000 lives lost to opioid overdose in the United States between 1999 and 2020 (Hedegaard et al., 2021). The COVID-19 pandemic has only exacerbated matters; in all U.S. states, rates of resumed use and drug overdose deaths have increased or spiked (American Medical Association, 2021). The Maryland Opioid Operational Command Center (2021) reported that fatal overdoses in the state increased nearly 20% between 2019 and early 2021, with approximately 90% of these being opioid related. Black Marylanders have borne a disproportionate and growing share of these fatalities. Baltimore, where our research and practice are based, has experienced one of the highest opioid overdose rates in the country (Irwin et al., 2017). Lopez (2019) described Baltimore as "ground zero" in an opioid crisis disproportionately killing Black Americans, a group that makes up 62.5% of the city's residents (U.S. Census Bureau, n.d.). Bringing more recovery research and resources to cities like Baltimore is imperative because while much attention has focused on White, middleclass people outside major metropolitan areas who use opioids, the harm inflicted by opioids on urban Black populations has been to a great extent unrecognized (James & Jordan, 2018) outside of impacted communities and their allies.
Lack of attention to the impact of opioid use among urban Black residents must be seen within the sociohistorical context of systemic racism. In Baltimore, this includes the Inner Harbor playing a leading role in the slave trade; the city's deep history of housing bias, including through segregationist redlining practices; as well as a pattern of unconstitutional police policies and practices targeting Black communities, which led to the 2015 murder of Freddie Gray (Badger, 2015;Childress, 2016;Clayton, 2000). Hart and Hart (2019) pointed out that during the opioid crisis, criminalizing policies were adopted to further oppress Black Americans who used or sold drugs. However, when public perception shifted to the impact of the opioid crisis on White Americans, a more compassionate approach of empathy and support was adopted.
Recognizing how health inequities are distributed in ways which disproportionately impact Black people and other marginalized groups-and understanding the structural racism that undergirds these inequities-is one important step academics committed to anti-racist research can take as they seek to collaborate with community-based organizations (Muhammad et al., 2018). As a guiding approach to effective and equitable community-academic partnerships, community-based participatory research (CBPR) offers numerous principles and strategies such as this one to facilitate strong collaboration . Wallerstein and colleagues (2018) offer the following definition: "CBPR embraces collaborative efforts among community, academic, and other stakeholders who gather and use research and data to build on strengths and priorities of the community for multilevel strategies to improve health and social equity" (p. 3). Within this context, it is important to recognize that in response to racial oppression, Baltimore's Black communities have long histories of resilience and struggles for justice (Arnold-Garza & Gadsby, 2017), which many community partners involved in this manuscript and the initiative it examines continue through their work.
This article describes collaborative work conducted through a four-year 2020-2024 National Institutes of Health (NIH)-National Institute on Drug Abuse (NIDA) R24 grant under NIH's Helping to End Addiction Long-Term (HEAL) initiative. The project described in this paper seeks to build research infrastructure and capacity through community-academic partnerships focused on opioid recovery support services. Because this is not an intervention study, no specific research methods were delineated in the funding proposal. The grant outlined these aims: to build community-academic partnerships that identify critical recovery research questions, to build data and research infrastructure to answer these questions, and to disseminate best practice strategies to measure impact and otherwise advance the science on opioid recovery support services. Toward these ends, proposed activities included training workshops, funded pilot research projects, and a research fellowship. The project, Innovations in Recovery through Infrastructure Support (IRIS; https://www.iris. ssw.umaryland.edu/), is located in Baltimore at the University of Maryland School of Social Work. IRIS is an affiliated project of the university's broad-based Behavioral Health and Well-Being Lab (BHWell Lab; https://www.bhwelllab.ssw. umaryland.edu/), which employs an equity lens to guide research focused on behavioral health and well-being among emerging adults and adults.
IRIS is built on long-standing partnerships between community-based organizations and researchers from the University of Maryland who hold the shared goal of collaboratively combating the opioid epidemic. The university seeks to deeply engage the surrounding community, including through agency-based student internships, faculty member service on boards of directors, and funded research partnerships such as the IRIS project. From the original partners who envisioned this initiative, wrote the grant, and helped the principal investigators (PIs) to launch the project, the IRIS network has expanded and diversified greatly.
During IRIS's first year, 2020-2021, we utilized CBPR as a foundation from which to engage stakeholders and establish partnerships. This approach was based on significantly overlapping principles, values, and goals of CBPR and social work-including a focus on community strengths, self-determination, and pursuit of social justice (Branom, 2012). We intended to establish a foundation of deep trust upon which we could build our partners' capacity to answer important research questions and ultimately improve opioid recovery outcomes.
The overarching aim of IRIS is to facilitate the wide dissemination of effective opioid recovery practices while supporting innovation to develop new strategies to address the opioid crisis. These practices include medications for the treatment of opioid use and the utilization of peer recovery workers. This "Research from the Field" article describes lessons learned from the project's Year 1 stakeholder engagement phase, including aspects identified by our academic team and community partners as strengths, mistakes, and positive next steps. Our paper has 13 coauthors representing university and agency partner leadership, many of whom identify as peers and/or come from racial and ethnic minority backgrounds or other structurally marginalized groups. Although the process described here was implemented within the opioid recovery support sector, lessons learned may resonate with researchers across disciplines who aim to deeply engage stakeholders and develop equitable community-academic partnerships.

Community-Academic Partnerships and CBPR in Recovery Work
Similar to views shared by Helm and colleagues (2017) in their paper on a Native Hawaiian community-academic partnership for drug prevention, IRIS takes a critical stance toward traditional ideas of researchers as experts and those with lived experience as passive recipients of education and services. Just as Helm et al. saw youth as leading healers and change agents, we see our partners, particularly those with lived experience, as community health experts in possession of critical knowledge that, when paired with academic resources, produces important practice-based research and researchbased practice.
Though we use CBPR as our guiding approach, many other related orientations influence our work, including community-engaged research, participatory action research, and communitydriven participatory research. Adams (2020) described a community-engaged research process wherein a diverse set of stakeholders developed a map of opioid recovery pathways that served as a resource for collaborative efforts to address the opioid epidemic. Importantly, the pathways map was a tool identified by nonacademic stakeholders as helpful to meet their needs, and not imposed on them by academic researchers.
Zimmerman et al. (2020) demonstrated how, through the action phase of their participatory process, researchers supported stakeholder groups' identification of community priorities to address the opioid crisis, which they then implemented collaboratively. Serving as a resource for this Virginia-based initiative, as well as for our Maryland-based IRIS project, is the Patient-Centered Outcomes Research Institute (PCORI; https://www.pcori.org/). PCORI stresses how those with lived experience and other health care stakeholders should be treated not as research subjects but rather as equitable partners. A call for true partnership between community members and academics is stressed by Urban Survivors Union activists, the national drug users' union, in their journal article outlining the need for a community-driven participatory research approach (Montoya & Kent, 2011). This process is characterized by community-initiated research questions, leadership capacity development, and joint data ownership (Simon et al., 2021). The deep involvement of those with lived experience makes research more relevant and helps ensure results are disseminated and utilized more extensively.

The IRIS Project and Stakeholder Engagement
IRIS was initially conceived before the start of the COVID-19 pandemic. The PIs envisioned that in Year 1 of the grant we would convene a large in-person conference with stakeholders to assess needs, establish priorities, and create a solid foundation built on trust for ongoing partnership. When the project launched in the fall of 2020, meeting in person was no longer a viable option. Though we discussed with our community partners the option of holding an online conference instead, it quickly became clear that this was not a desirable alternative.
Our initial IRIS network skewed toward academics, representatives from larger agencies, and people of a White racial background, and we recognized the need to diversify. The project's PIs therefore assigned two team members (a faculty member and a clinical research specialist who is also a PhD student) as leads in the stakeholder engagement process. The goal was to connect with more individuals and organizations that would balance our current composition with additional community agencies, smaller organizations, people with lived experience, and people from racial and ethnic minority groups. This was particularly important because our internal faculty and staff team was composed mostly of people from non-Hispanic White backgrounds (one team member/author is an Asian American woman, and one is a Hispanic man). Our team also lacked representation of peers or others who identified as being in long-term recovery. Recognizing that our positionality could adversely affect our ability to partner around relevant supports to populations served in our community, we conducted a wide outreach process as we worked to diversify our internal team. During an intensive 9-month period from October 2020 to June 2021, we held over 50 meetings with over 120 individuals. There were 30 organizations represented, 24 of which were community based (not university based or affiliated), and five of which were collaboratives comprised of 10-20 different agencies.
From October 2020 through September 2021, we held four quarterly meetings with our leadership committee (LC). The LC is distinct from a community advisory board because it includes a mix of academics and representatives from agencies outside the university. However, the LC serves a related purpose in providing one CBPR mechanism for members of the IRIS network to have representation in our various research initiatives (Newman et al., 2011). Lavallee and colleagues (2012) have identified accountability as a key criterion for assessing effective stakeholder engagement. The IRIS LC helps us to stay accountable to community partners and responsive to their inputs.
IRIS has encountered many barriers to stakeholder participation in research. These include the urgent need to address opioid overdose deaths, the far-reaching impact of the COVID-19 pandemic, and community partners' skepticism of researchers' motivations based on past harms. Despite these obstacles, through strategies like leveraging existing relationships, developing trust, using accessible language, having a focused agenda, and facilitating in a collaborative manner (Rush et al., 2012), our LC meetings have been productive and positively evaluated in stakeholder interviews and online surveys. For example, one LC partner from the peer recovery movement reported that she appreciated being seen as a leader within these meetings, describing an atmosphere in which all inputs are valued, it feels like an equal playing field, and power is shared.
During the first year of IRIS, we also launched our Recovery Research Program (RRP), which provides funding for community-academic partnerships between agencies and our University of Maryland academic team or other universities. During the first RRP cycle, we funded 13 proposals and are now distributing nearly $325,000 from our grant directly to the community. This resourcesharing represents the important CBPR principle of equitable partnerships attending to social inequities (Ward et al., 2018). By distributing a portion of our grant funds to community groups, IRIS aimed to demonstrate a commitment to sharing power and providing immediate benefit (Cain et al., 2014).
To assess and describe the state of our community partnerships since IRIS's initial launch, we conducted interviews with five key community partners and also engaged them as manuscript authors. This type of early evaluation and documentation of a project's efficacy in implementing CBPR is an important tool that helps collaborative initiatives improve their processes and redouble their efforts to achieve intended outcomes (Israel et al., 2005). For IRIS's stakeholder interviews, we invited partners to participate based on the depth of our engagement with them and because they identified as peers, led peer-operated agencies, and/or directed community-based organizations focused on vulnerable populations such as the unhoused. The partners' input was supplemented by interviews with five IRIS academic team members and written comments submitted by two others. Each person who contributed to the lessons learned and next steps sections of this article responded to the following questions: 1. What have been the strengths of IRIS's efforts to engage leading community agencies and individuals working to support those affected by opioids in Maryland? 2. How would you advise to build on these strengths moving forward? 3. What have been shortcomings in IRIS's efforts to engage leading community agencies and individuals working to support those affected by opioids in Maryland? 4. How would you advise to address these shortcomings moving forward? How may they have been avoided? 5. Overall, what lessons have been learned by you, your agency, and the populations you represent about forming communityacademic partnerships?

Lessons Learned
To establish core themes that emerged from these discussions, IRIS's two stakeholder engagement leads reviewed interview notes, independently grouped responses, discussed similarities and differences in identified themes, then collaborated to finalize key lessons learned as presented below. A third IRIS team member, who is a PhD candidate, also contributed by identifying and writing up themes. Key quotes are included in this Lessons Learned section as well as the Next Steps section that follows to bring themes to life and to offer detailed representations of thoughts and feelings expressed. The main themes of both sections are summarized in Table 1.

Strong Team, Ample Resources, and Allocated CBPR Staff
IRIS's PIs strategically played to the strengths of two academic team members with significant experience in community-engaged and participatory approaches to practice and research. These individuals actively engaged stakeholders and applied participatory approaches to the work. One IRIS academic team member indicated: "It was so good that the PIs sought out to recruit PhD students and faculty that were interested in stakeholder engagement and gave them an opportunity to do the work, to develop an agenda." Another agreed, stating, "Having them take the lead has been important so it gets the focus it needs." This successful strategy of allocating specific project staff for community engagement mirrors more macro-level findings of Weerts and Sandmann (2008), whose study demonstrated the benefits to universities that have community engagement offices to lead these efforts. Besides work done by IRIS PIs and engagement leads, however, numerous interviewees cited the academic team's collective commitment to building a diverse network as a key asset. One academic team member stated that IRIS's greatest strength was our dedication to stakeholder engagement, "really trying to involve community partners at every step in the process. " In addition to strong human capital, having significant financial capital has also been beneficial to IRIS's community engagement. Being a well- Dedicate time and effort to developing equitable communityacademic partnerships, as the robust collaborations and outcomes which emerge make it well worth the investment Continue making time for process, acknowledging expertise brought by community and academic partners, and evaluating project's progress in partnership with stakeholders funded and multiyear grant project made it easier to allocate research staff to focus on stakeholder engagement. It was also an intentional choice by the PIs to frame Year 1 as a time for needs assessment and relationship building toward a broad and inclusive IRIS learning collaborative. The NIH R24 funding stream supports this approach, as it is specifically designed for academics to partner deeply with community organizations. Our experience underscores how important it is for funders to invest more deeply in stakeholder engagement and participatory processes, and for academics to prioritize community engagement within timelines and budgets, whether through research infrastructure building grants or more traditional intervention proposals. Being a wellresourced and longer-term initiative also enabled IRIS to offer benefits (described below) that a project with less resources would have had more difficulty fitting into its budget. Still, research initiatives of any budget, staff size, or timeframe may choose to redistribute funds to the community, even if on a more limited basis.

Benefits for Academics and Community Partners
Though research has been criticized for failing to provide communities with a fair share of study benefits (Schulz-Baldes et al., 2007) beyond abstract knowledge gain and the potential for longterm progress on pressing community issues, our community partners cited numerous short-term benefits of working with IRIS. These included stipends for participation in our LC, RRP funding, and subcontracts for training workshop delivery. By immediately resourcing community partners with needed funds, skill-building workshops, and support to conduct research on issues of importance to them, IRIS is facilitating the creation of concrete products and offering the type of tangible community results that are essential for successful university-based initiatives (Carney et al., 2011). Through this process, IRIS is building trust with our partners and creating comfort to engage more deeply in research infrastructure building activities.
Some community partners also saw partnership with IRIS as a means of longer-term capacity building to use research to address the opioid epidemic through improved services and enhanced advocacy. One partner appreciated how, with IRIS funding, her organization is now empowered to develop and pilot test an opioidfocused anti-stigma training while gaining research skills to examine data and hone their in-house evaluation procedures. Another partner stated that with a new outlook on the benefits of community-academic partnerships, she could not only champion participation in recovery research but also address COVID-19 vaccination hesitancy in her community. With numerous benefits, IRIS had transformed in her eyes from a project to "an opportunity." For IRIS academics, our deep stakeholder engagement process came from an acknowledgment that for researchers without the lived experience of opioid addiction, there is a limited scope of what we know. We as a team recognized that without care and focused attention, our academic training and research experience could contribute to losing perspective, adopting inaccessible language, and becoming disconnected from community voices that need to be heard. Only through intentional and sustained conversation with communitybased organizations could we formulate the right research questions, which are already embedded into their daily work. "When we talk to providers, there are books and books and books of questions, and we should be funding those questions, not coming up with them on our own. We're failing if we don't draw from this expertise, " declared one of IRIS's PIs.

Ongoing Commitment to Representativeness
At the outset, our LC was skewed toward academics and did not adequately represent the racial and ethnic diversity of IRIS stakeholder communities or people with lived experience in recovery. By constantly asking our partners "Who's missing?" and "Who else should we talk to?" and then immediately following up, we diversified the LC and overall network to include more grassroots organizations, peers (who bring needed perspective as both staff and persons with lived experience), and members of racial and ethnic minority groups. This process, which mirrors snowball sampling, is ongoing. Besides soliciting suggestions from partners, we also conducted frequent internal audits of our LC to identify areas of need in our representativeness, which then generated ideas for targeted recruitment from our internal team and external partners.
It is worth noting that simply outreaching to groups recommended by previous contacts would not have yielded the positive results we saw without the faculty research team leveraging prior trusting community partnerships built over years before this grant started. In turn, these individuals facilitated introductions and followed up with prospective partners to encourage them to give IRIS a chance, even if there was hesitancy based on negative past experiences with research. We were also encouraged to engage with less formal networks like peer recovery social media groups and harm reduction listservs; to contact not just traditional provider organizations but also peer training, certification, and advocacy initiatives. When an online training or weekend community recovery event came up, we prioritized attending not only for outreach but also to demonstrate our commitment to showing up, learning, and relationship building. As each community partner was integrated more deeply into IRIS, we were able to leverage these connections to develop new ones. "Oh, you work with ________!" people would say smiling and nodding, indicating IRIS was benefiting greatly from trust in these respected leaders as we worked together to shape our initiative.

Reshape Research's Relationship to Community
For authentic engagement, IRIS researchers were clear that historical harms and current forms of oppression, including systemic racism, needed to be acknowledged. Just the word "researcher" was triggering for some partners. One community partner who is a peer leader said: The whole word "research" means you're going to be used as a guinea pig, and so automatically we put that wall up because of the history. So the whole word "researcher" puts fear in us because that means you're coming to use us as lab rats. You're coming for something, but we will never know what you did.
It was clear from our discussions that these valued stakeholders needed to be partners in the research process rather than participants.
Community partners reported that they had felt underappreciated and overlooked during past collaborations. When researchers needed information, they came and took it, then were gone without even reporting on results. As described by one partner, this "one-way" extractive process of "information mining" falsely presumes that knowledge will "trickle down" after researchers analyze and write up their results. It is a top-down approach, as if researchers are here to educate the community on issues they already understand very deeply. Thus, another theme that emerged from our discussions was the importance of sharing results obtained by research with the community partners participating in the work.
One partner distinguished that the IRIScommunity relationships were not collaborations, which they said were extractive, but rather mutually beneficial partnerships in which "I need you, and you need me. " Another partner cited our assetbased approach that affirmed that communities already have networks and resources. This stood in contrast to a typical deficit-oriented model to prove something is not going well. In the "two-way" approach to scientific inquiry, it is understood that researchers and community partners both bring vital strengths to collecting data, analyzing and contextualizing results, and sharing findings. With these approaches, IRIS was able to move beyond a "check the box" approach in which community engagement is paid lip service toward one where voices are equally valued and time for authentic engagement is invested.

Approaching the Work With Humility and Accountability
As social work researchers, IRIS team members have had extensive training and deep experience in strengths-based approaches that lead to powerful alliances. Some characteristics and strategies cited by staff and partners that facilitated this process were "patience, praise, calm, affirming, passion, commitment, personal touch, sincere interest, delicacy in asking questions, listening with genuine curiosity, and laughter. " Instead of coming to partners with our own research questions, the nature of our infrastructure building grant and our own participatory approaches led to numerous meetings in which we asked about community needs. With a group of peer recovery leaders whom we met with consistently (including every two weeks for months leading up to their successful submission for RRP funding), it was important to stress when they asked us what we wanted to study that we did not come with preset research priorities. Instead, we were looking to them to identify issues in need of investigation.
Even with the best of intentions and the above attributes, IRIS researchers made mistakes. The benefit of having established trusting relationships, however, meant that these mistakes were discussed, and we were given the opportunity to apologize and work through challenges. One such moment occurred when we as researchers gave the impression to peer recovery leaders that, besides in-kind staff time to support joint research, IRIS funding for their projects was guaranteed.
We then had to explain that all partners would need to apply for funding within the competitive RRP application process. Though financial resources were eventually secured through a close collaboration on the proposal, this initially created some tension.
Another lesson learned through reflection and stakeholder feedback from our peer partners was to ensure they had the chance to digest and respond to requests for different types of collaboration, such as pursuit of RRP funding and inclusion of their organizations in other funding proposals. Although each of these project components was part of the IRIS academic team's daily activities and paid roles, our community partners were not inherently funded to participate. Even if a chance of financial resources was attached to these activities, they still amounted to extra and unpaid work for our partners, and we needed to be sensitive to that. As Teeters and Jurow (2018) pointed out, within community-engaged research aiming to achieve equitable partnerships, awareness of and responsiveness to these dynamics is key. As academics, we learned to proactively create space for discussion to process issues like these before moving forward, and to take on more of the workload (including scheduling meetings, note-taking, conducting background research, and drafting funding proposals), at least until resources were secured for partners' time.
Even with these conflicts, however, our partner relationships remained strong and actually got stronger. "Nothing was avoided, no matter how big or small, " said one peer partner describing how we dealt with issues that arose during initial meetings. We made time at the outset of every meeting and checked in with partners individually between meetings, all to make sure it was a safe and encouraging space for voicing concerns. This gave the partner a feeling that our relationship was "more a circle than a box" and helped take down walls that had been put up based on negative past experiences with research.

Engaged Research Is Challenging, but Ultimately Well Worth It
Both IRIS team members and community partners recognized that community-engaged research has a different pace compared to traditional academic research and on-theground service provision. Whether related to communication and logistical challenges or issues of rapport and trust building, the work of forming equitable community-academic partnerships is challenging and moves slowly. Even within our IRIS academic team, though we share a common goal of "walking the walk, " we have had to make time to examine how our varying conceptual and operational frameworks around engaged research inform the road we take to get there. There is consensus, however, that dialogue around these differences is important, and that diversity makes us stronger. Even if our research process requires more time, one IRIS staff member captured our collective sentiment when they stated, "It will result in more robust collaborations and outcomes, " an assertion we have seen come to fruition in Year 1 of the IRIS project.

Next Steps
In their consideration of ways to build on strengths and address challenges, IRIS's academic team and community partners largely focused on continuing the work described above. There was a common theme that this aspirational work is never-ending because there is always someone else to reach, and there are always ways to engage in a deeper manner. The ideas collected for this article and presented below have already begun to inform IRIS's work. This process of pausing for reflection, then taking corresponding action towards transforming our world to be more equitable is part of what Freire (1972) called praxis and has become central to efforts grounded in CBPR like ours. One IRIS team member noted that researchbased knowledge is needed to transform reality. By rigorously helping to evaluate partners' work, we can uplift effective and promising practices in a way that provides data for program expansion and policy advocacy.

Bringing Stakeholders Together Around Common Interests
Though our quarterly LC meetings provide an opportunity for academics and community partners to gather (remotely up to now), there was a call for deeper engagement. One partner cited the need to "break down silos" and form "learning labs around common themes which could lead to other partnerships and problem solving. " Toward this end, IRIS recently held our first LC retreat. This retreat was planned to be held in person but moved online because of severe weather.
One retreat agenda item was planning an IRIS research fellowship centered on peer integration into the recovery workforce. The need for this initiative emerged during our Year 1 stakeholder engagement process and presented an opportunity to bring together peers, clinicians, agency managers, and academics around using research to pursue this shared interest. One partner cited an important goal in bringing together these constituencies: "to leave the space with mutual understanding" around the role and use of peers within recovery support settings. This 10-month fellowship will offer opportunities to build relationships, trust, and research infrastructure through day-long monthly sessions wherein learning will be applied through agency-based recovery research projects.
Another set of convenings requested by a community partner was for IRIS to gather its funded RRP projects for networking and to share lessons learned from their research. Although our initial plan to hold a needs assessment conference was abandoned due to COVID-19 and partners' negative reaction toward holding it online, we now intend to hold an in-person convening in the fourth year of the grant, so that IRIS and its funded RRP partners can present the results of our work together to a wider audience.
A community partner brought to light one benefit of continuing to bring together a broad network of representatives from the recovery support system: if you don't know each other, you are prone to misunderstanding. Although she had some prior negative experiences with researchers, she felt her perceptions about academics were based in part on misinformation and was now pleased to develop a more positive impression. By spending more time together within a strengths-based social work approach and CBPR framework, we expect that researchers and community partners will continue to build mutual understanding and move toward our collective goal of building effective opioid recovery practices.

Share Benefits of Research With Broader Community
Many of IRIS's community partners expressed that they now see research in a more positive light, and they desired to share this message along with the benefits research can provide One partner agency stated they could hold a training for peer-run organizations on lessons learned from research in a way that makes it more practical and accessible, particularly for smaller groups. Others expressed a desire for dissemination tools beyond academic publications, including social media content, blogs, and engaging visuals like infographics. This request for capacity building will be answered as part of an upcoming IRIS-delivered "Measuring What Matters" workshop on data collection, analysis, and presentation. IRIS community partners also identified policy implications as a clear benefit of research, requesting that we more directly link our work to urgently needed actions within the criminal legal system and around issues like overdose prevention and consumer rights. The IRIS e-newsletter and website provide forums for information sharing about these issues. Our fellowship, which will provide financial compensation to participating organizations, will also offer opportunities for learning and collective action around research-policy links. Numerous IRIS partners have advocated for Maryland to implement Medicaid reimbursement for peerdelivered recovery services, and results from various IRIS research projects could provide valuable data on the essential role peers serve and how best to resource them.
What is important, our partners point out, is that the research is done in a way that builds trust instead of further eroding it. As one community partner said: Data shapes policy. If you've got the data, you can get the dollars. If you can't prove what you're doing is successful, you can't get the funding. IRIS can help inform partners on ways to get data that don't diminish the people or the process.

Continue Expanding IRIS to Be More Representative
Although we have made great progress in creating a broad and diverse LC and greater IRIS network, our partners called for further development in these areas. Currently underrepresented groups in IRIS include the LGBTQ+ community, the Latinx population, youth and emerging adults, policy-makers, people from rural areas, and those with lived experience who are not working as peers. This latter group can offer unique perspectives on how the overdose and COVID-19 epidemics may be impacting people who use opioids.
Our outreach period to generate applicants for a next round of RRP-funded projects provides an opportunity for a second wave of deep stakeholder engagement to address these gaps. We are also engaged in targeted recruitment with organizations and individuals who represent these groups. In our outreach, we will continue to prioritize smaller grassroots organizations that might not hear about us naturally because they are busy with "boots on the ground" and not as connected to large academic or medical systems.
Within our IRIS academic team, there is also a need for greater diversity around racial and ethnic identity and lived experience with recovery that we aim to address through future staff hires, including a peer recovery research specialist; our next wave of graduate research assistants; and additional university faculty.

Continue Approaching Work With Humility
Though IRIS has been committed to acknowledging how research has harmed the community and how current forms of oppression continue to create inequity, we must redouble our efforts to build trust and remain accountable. Toward this end, the IRIS academic team has been working with our LC on a set of values statements that will make it even clearer to the community who we are and how we approach our work. We also made time on our retreat and LC meeting agendas to explore how we can better act as antiracist researchers engaged in anti-oppressive research practice. We have also begun to integrate acknowledgments of past and current harms done by research into IRIS-sponsored workshops and will continue this through our culminating conference.
In order to remain humble and responsive to feedback, both IRIS academics and community partners called for frequent check-ins to minimize misunderstandings and recover from mistakes. Recent feedback from peer partners revealed the need for greater awareness around language we use to describe IRIS and opioid treatment and recovery. In social work we may use the term "providers" for all those who serve people in recovery and "clients" for those we serve. However, our peer partners have expressed that these terms are clinical and do not represent them or the support they offer. Instead, they prefer that we distinguish between clinical providers and peers and use terms like "participant, " "member, " "person in recovery, " or "person with lived experience. "

Continue Making Time for Process
During periods when COVID-19 cases dropped, restrictions eased, and face-to-face meetings were safer, we began making site visits to partner agencies, and we will continue to allocate time for this activity as the state of the pandemic allows. Having face-to-face instead of remote communication will help us remain connected. Though more time is needed for travel, it is a worthwhile investment to see people where they work and learn more directly from their experience. For researchers who prioritize site visits and in-person meetings during the pandemic, however, it is important to recognize that face-toface interactions remained the full-time work of countless frontline workers, particularly peers, who continued to invest heavily in this type of relationship building and support despite the associated health risks. Their firsthand knowledge of recovery is invaluable, and as researchers we must do all we can to uplift their expertise.
In general, both the IRIS academic team and community partners felt we needed to stay patient as our partnerships build over time and across differences. As one partner noted, "What seems common sense for you may be a revelation to someone else. " We need to acknowledge each other's expertise and slow down the mutual learning process.
As part of this commitment to deep engagement and participatory research, IRIS will continue assessing its work and the impact we are having. Along these lines, community partners have advocated for IRIS to not only support evaluating other organizations' work but also evaluate our project, host school, university system, and overall research community. For IRIS, other researchdriven initiatives, and academic institutions to continually develop, we have to constantly check, "Did we get what we were expecting?" and "What were those outcomes?" Patton (2015) put it simply: "Research seeks to prove; evaluation seeks to improve" (p. 532). The IRIS evaluation process cannot be one-way. Academics may know more about specific scientific methods used to evaluate, but community partners overwhelmingly know more about what is being evaluated and how to apply results to forward their work. IRIS will ensure our evaluation is ongoing so we can maintain the CBPR principles that have guided our initial phase. As we collect evaluation data, we will engage partners in interpreting results, which is one aspect of transformative participatory evaluation recommended by Wiggins and colleagues (2018). Toward this end, we have reviewed suggestions made during stakeholder interviews not only to produce content for this article but also as a part of IRIS's Year 1 internal process evaluation on ways to deepen engagement and strengthen participatory practices.

Conclusion
The IRIS project aims to build research infrastructure through effective and equitable community-academic partnerships. During our first year, we focused on stakeholder engagement using principles and practices from engaged and participatory research. With a strong grounding in CBPR, we endeavored to form trusting relationships with a diverse array of partners by being humble in our approach and acknowledging past and current mistakes made by researchers. We prioritized expanding our network to include smaller organizations and peer-delivered services. We worked to ensure our LC was representative racially of our home city and the people in recovery we ultimately served. The process was not without challenges, including its time-intensive nature. However, given the great value added by community partners through their own time spent on the project, this was a very worthwhile investment. IRIS partners appreciate "learning more about research, why it's so important, and how it helps, " including the community benefit derived when "recovery organizations are supported by academics to obtain implementable, evidence-based answers to their own immediate questions. " IRIS will continue to engage community partners within the recovery support system in mutually beneficial research activities. This paper presents lessons learned up to now and next steps we intend to take to continue and advance our community-academic partnerships. Our hope is that these lessons have relevance for others engaged in forming community-academic partnerships through deep stakeholder engagement and using participatory approaches within and outside the realm of recovery research.