Promising Patient‐Engagement Practices Used by National Research Networks at the US Veterans Health Administration: Perspectives of Veterans, Staff, and Research Leaders Authors: Tracy L. Sides, Jennifer Wisdom, Malloree M. Argust, Erin C. Amundson, Diana Burgess, Laura A. Meis, Agnes C. Jensen, Erin E. Krebs Published: 2026 ABSTRACT ABSTRACTIntroductionActive involvement of patients is an essential feature of learning health systems and patient‐centered research. Research networks and healthcare organizations lack guidance on best practices for developing and sustaining patient engagement infrastructure. We sought to examine perspectives on promising Veteran‐engagement infrastructure and practices used by US Veterans Affairs (VA) research networks.MethodsWe solicited expert opinion from three separate panels of Veterans, VA engagement staff, and VA research leaders using a modified Delphi process comprising two rounds of online surveys and a set of virtual group meetings or individual interviews. In a prior study, we identified promising engagement practices in use by VA national research networks. We analyzed survey data with predefined rating categories to determine which practices the three expert panels consistently endorsed. We used thematic analysis to compare experts' perspectives across panels and to identify emergent cross‐cutting themes.ResultsVeterans, VA engagement staff, and VA research leaders consistently endorsed 66% (29/44) of the research networks' promising Veteran‐engagement practices. Summarized, these practices formed five recommended engagement‐related responsibilities for VA research networks. Some network engagement practices were inconsistently endorsed (13/44, 30%), and two practices (4%) received overall ratings of neutral or unsure by at least one expert panel. Network practices not consistently endorsed varied by expert panel, but panelists' comments reflected three overarching themes: constraint of resource availability, commitment to respectful and meaningful engagement processes, and concern for potential burden on partners.ConclusionsPanelists' perspectives highlight challenges regarding how to best invest limited patient‐engagement resources within research and learning health system organizations. Research network engagement infrastructure can help expand the capacity and culture of patient‐centered research and care. Organizations seeking to function as learning health systems need to devote planning time, skilled staff, and funds to develop and sustain patient‐engagement infrastructure tailored to their context. Introduction Active involvement of patients, families, and communities is an essential feature of learning health systems and patient‐centered research [1, 2, 3]. Growing evidence demonstrates that this engagement in research can lead to more feasible, efficient, and relevant health studies, care, and services [4, 5, 6, 7, 8], and form the basis for equitable partnerships between community‐based partners and researchers [9, 10]. There is also a growing body of literature documenting examples of infrastructure supporting patient engagement in research networks and healthcare organizations [11, 12, 13, 14, 15]. However, there is a lack of guidance on best practices for developing and sustaining such engagement infrastructure. Understanding perspectives of different stakeholders on emerging patient engagement infrastructure and approaches in different contexts can help bring learning health system theory and models into actual practice [3, 16]. Easterling and colleagues systematically developed a practical framework for conceptualizing the bodies of work and enabling conditions required of an organization to function as a learning health system, one of which involves engaging clinicians, patients, and other stakeholders [3]. However, effectively engaging patients presents unique challenges such as potential differences in social and institutional power, education, and communication norms relative to clinicians and other internal health care stakeholders that must be addressed to realize human‐centered, data‐driven learning health systems [7, 12, 17, 18, 19]. Few investigations have examined the feasibility, impact, or desirability of engagement activities in the context of research or healthcare organizations that historically have not required patient engagement throughout the research process. The U.S. Department of Veterans Affairs (VA) operates the largest integrated health care system in the country and has a 100‐year history of funding intramural research focused on the health and healthcare of Veterans [20]. Over the past 10 years, VA leadership has placed a growing emphasis on Veteran involvement in the research process [21, 22]. As of 2024, all VA research involving human subjects requires consideration of patient engagement in research proposals [23, 24]. These efforts have increased Veterans' involvement in both individual studies and in the activities of local research centers and national research networks [12, 22, 25, 26, 27]. Previously, we conducted an environmental scan using the PCORI framework of Cope and colleagues [11] to describe the patient engagement activities and infrastructure among seven national research networks funded by the VA Health Systems Research program [12]. Each network was established to enhance collaboration and build capacity for research and implementation of evidence‐based health care in a particular Veteran health or health care delivery priority area (suicide prevention, pain/opioids, virtual care, long COVID, access and community care, electronic health care record modernization, and women's health). Our scan identified 129 engagement‐related activities among the seven networks. The purpose of the current study was to examine perspectives among Veterans, VA engagement staff, and VA research leaders on the feasibility, impact, and desirability of promising practices used by VA research networks to facilitate Veteran engagement in the research process or support network engagement infrastructure. Our goal was to facilitate learning across research networks and organizations. Questions of Interest This project addressed the following research questions: Which VA research network Veteran‐engagement practices do Veterans, VA engagement staff, and VA research leaders consistently endorse as generally feasible, impactful, and desirable?How do perspectives on Veteran‐engagement practices in use by VA research networks compare across Veterans, VA engagement staff, and VA research leaders?What key considerations emerge from participants' comments about research networks' promising Veteran‐engagement practices? Which VA research network Veteran‐engagement practices do Veterans, VA engagement staff, and VA research leaders consistently endorse as generally feasible, impactful, and desirable? Which VA research network Veteran‐engagement practices do Veterans, VA engagement staff, and VA research leaders consistently endorse as generally feasible, impactful, and desirable? How do perspectives on Veteran‐engagement practices in use by VA research networks compare across Veterans, VA engagement staff, and VA research leaders? How do perspectives on Veteran‐engagement practices in use by VA research networks compare across Veterans, VA engagement staff, and VA research leaders? What key considerations emerge from participants' comments about research networks' promising Veteran‐engagement practices? What key considerations emerge from participants' comments about research networks' promising Veteran‐engagement practices? Methods Overview The study design was a modified Delphi process conducted separately with each of three different panels of VA engagement experts. The Delphi technique is a highly structured process that employs a series of “rounds” to gather and compile information from expert stakeholders to gain consensus regarding a complex phenomenon [28, 29]. In this study, a Delphi technique was used to establish recommended Veteran‐engagement practices for VA research networks. It was modified from the traditional structure in two ways. First, Delphi studies typically are conducted with a single panel of participants [29], but in this study, we conducted the same Delphi process with three different expert panels. Second, although typically a Delphi process asks participants to rate items for inclusion or exclusion, we also asked participants to provide ratings on additional attributes of each practice (i.e., feasibility and impact). Participants were asked to rate a list of promising Veteran‐engagement practices identified in our previous study of VA research networks [12]. Two rounds of independent ratings with each expert panel were conducted using online surveys. After the first rating round, participating experts were asked to complete a virtual individual interview or group meeting to clarify their perspectives on lower‐ranked practices and Veteran engagement in VA research more generally. Data were analyzed to address the three research questions. The study was approved by the Minneapolis VA IRB with a waiver of written informed consent on December 29, 2023. Participant Recruitment Potential participants were people with interest in and knowledge of Veteran engagement in VA research, with at least one of the following three types of expertise: (1) Veterans, with experience participating as a patient partner in a research engagement group; (2) engagement staff, with experience managing research engagement activities; and (3) research leaders, with decision‐making authority in a VA‐wide research line or program. We used our professional networks to identify and recruit purposeful samples of participants with diverse sociodemographic and geographic characteristics and research center/program affiliations [30]. We organized participants into three panels, based on their type of expertise. When a participant had multifaceted expertise, we based panel assignment on their primary role. Table 1 shows details about the recruitment approach for each panel. TABLE 1Interest group definitions and recruitment methods.Interest groupDefinition, recruitment methodVeteransVeterans who serve as a member of an active VA research Veteran‐engagement group.We contacted two existing VA research Veteran‐engagement groups to schedule consultations, one coordinated by the Minneapolis VA Health Care System, Center for Care Delivery and Outcomes Research and the other by the VA Pain/Opioid Consortium of Research.Engagement staffIndividuals who have experience implementing Veteran‐engagement activities in VA and are: (1) a member of the VA National Veteran Engagement Workgroup, (2) a Veteran‐engagement group liaison or coordinator, (3) a VA employee and an author of a published article about patient‐engagement methods or evaluation, or some combination of the three.Based on our professional networks and criteria for diversity, we identified a convenience sample of engagement staff, who had not been interviewed as part of the environmental scan associated with this study, and emailed each an invitation to participate in this study.Research leadersVA Office of Research and Development (ORD) employees in leadership positions.A VA ORD leader agreed to participate in an interview and nominate other VA research leaders to participate in this study. T.S. contacted the leader, in August 2024, and received a list of nominated leaders for this study. T.S. emailed each nominated leader, except one who had been interviewed during the environmental scan associated with this study, with an invitation to participate in this study. Interest group definitions and recruitment methods. Rating Procedures For use in the present study, T.S. compiled a consolidated list of 44 “promising practices” based on those reported or documented as successful from the 129 total engagement activities identified among seven VA research networks in our previous work [12]. These activities and resulting promising practices spanned 11 engagement topic areas across three broad domains, including facilitating Veteran‐engaged research, network engagement infrastructure, and building and maintaining relationships with partners (Box 1). Two reviewers (M.A., E.A.) independently checked the list for completeness and accuracy using environmental scan records from our previous study. The two reviewers convened with T.S. to reconcile the list through an iterative discussion of practice descriptions. Once consensus on the list was reached, five members of the research team (T.S., E.A., M.A., J.W., and D.B.) met to review and revise the 44 practice descriptions for clarity and brevity (Appendix S1). Domains of promising engagement practices. Facilitating Veteran‐engaged research. Approaches and resources to incorporate or promote the incorporation of Veterans as partners in the research process (e.g., research network representatives establish a standing Veteran engagement group or co‐present with Veterans to the research community about effective engagement practices). Network engagement infrastructure. Practices that include enduring structures, frameworks, services, or staff maintained by a research network to carry out and evaluate its patient engagement work, or approaches to incorporate Veterans in research network decision‐making. Building and maintaining relationships with partners. Approaches and resources related to recruiting, onboarding, or preparing partners (i.e., Veterans or researchers) for engagement or related to maintaining effective relationships and interactions with Veteran partners. Measures Initial surveys included questions about panelists' background, including demographics, military service, and years involved with VA research. Both rounds of surveys asked for ratings of the feasibility, impact, and desirability of each promising practice using a five‐point scale, anchored by “1—Strongly Disagree” and “5—Strongly Agree.” Feasible practices were defined as engagement practices that a VA research network would be capable of implementing. Impactful practices were defined as those practices that help shift VA research culture to be more Veteran‐centered. Desirable practices were defined as those that, taking everything else into consideration, VA research networks should adopt. We developed the rating criteria based on the Learning Health Systems Consolidated Framework, which describes four enabling conditions for stakeholder engagement and other bodies of work carried out by a learning health system [3]. Our feasibility criterion aligns with the three enabling conditions related to sufficient financial, human, and technical resources to carry out engagement. Our impact criterion aligns with the fourth condition, a supportive organizational culture including norms that support the work of engagement. We included desirability as a summary criterion to capture each expert's overall considerations for endorsing a practice. Panelists were instructed to provide ratings based on ideal circumstances. An open comment field on each item invited respondents to provide an explanation for any ratings or add other comments. We created semi‐structured interview guides tailored for each of the three panels to explore perspectives on (1) attitudes/beliefs regarding lower‐rated engagement practices used by research networks to engage Veterans in the research process, (2) what is going well and what could be going better regarding Veteran engagement in VA research, and (3) what is needed to advance Veteran‐centered research culture in VA. Data Collection Procedures We conducted Delphi procedures separately with each panel to build group‐specific consensus and identify how perspectives on the feasibility, impact, and desirability of Veteran engagement practices by VA research networks compare across the three panels. In advance of the first round of ratings, we emailed panelists an overview of the environmental scan and the resulting list of research network promising practices arranged by domain and engagement topic area. Panelists were then invited by email to complete an online survey to independently rate the feasibility, impact, and desirability of each promising practice. We sent email reminders to panelists who had not completed the survey 3 days and 1 day before their interviews. We leveraged existing virtual meeting structures of two standing VA research Veteran engagement groups to explore the perspectives of our expert panel of Veterans; except for three Veterans who were unable to attend their engagement group's meeting and participated in individual interviews. We chose existing group meetings for Veterans to decrease the burden of participating in the study. To mitigate dominance and conformity pressures, we purposely selected two Veteran engagement groups known for diverse membership characteristics and use of inclusive meeting facilitation methods that did not seek group consensus. All engagement staff and research leaders participated in virtual individual interviews. At least 1 week prior to each interview or group meeting, each panelist received preparatory materials (project summary, list of research network promising practices, and main discussion questions). All panelists provided verbal consent prior to the start of the meeting or interview. The Veteran engagement group meetings were facilitated by their respective, experienced moderators known to include all group members' voices in discussions. In accordance with the Veteran engagement groups' norms, we did not record these meetings; instead, two members of the study team attended and took detailed (M.A. or E.A.) and thematic (T.S.) notes. T.S. facilitated the individual interviews with engagement staff and research leaders, which were conducted virtually and recorded with transcription using the Microsoft Teams platform. A second study team member attended as a thematic notetaker. Study staff edited transcripts and meeting notes for accuracy in preparation for coding and analysis. In preparation for distributing the Round 2 survey, we compiled an individualized Round 1 report for each participant. The report contained a list of the promising practices with the individual panelist's Round 1 ratings alongside their respective panel's attribute averages (for feasibility, impact, and desirability) and overall average for each practice (possible values 1.00–5.00). We edited the promising practice descriptions for clarity based on participants' collective feedback in survey comments or interviews that indicated misunderstanding of a practice (see Appendix S1 for Round 1 and Round 2 practice descriptions). The order of promising practices, by engagement domain and topic area, remained the same across the two survey rounds. After all individual interviews and group meetings were completed, panelists were invited to complete a second round of ratings. We emailed each panelist a link to their respective expert panel's Round 2 survey and attached their individualized Round 1 report. We provided the report to support independent rating and facilitate anonymous group‐level feedback to panelists [28, 29]. We sent email reminders to those who had not completed the survey at 7, 14, and 21 days after the initial invitation email. All surveys were conducted using the Qualtrics survey platform and accessed by panelists via a URL provided in an email. Data Coding and Analysis Quantitative Data Analysis A priori, we defined “endorsed with consistent agreement” for a practice as an expert panel overall average rating of 3.41–5.00 (i.e., the calculated Likert scale interval spanning overall average ratings “agree” and “strongly agree,” as described by Pimentel [31]) and no more than one panelist rating 1 or 2 (“strongly disagree” or “disagree”) for any attribute, to reflect low variability. “Endorsed with inconsistent agreement” for a practice was defined as a group average rating of 3.41–5.00 and two or more panelists rating 1 or 2 for any attribute, reflecting higher variability. “Neutral or unsure” about endorsing a practice was defined as a group over average rating within Likert scale interval 2.61–3.40. A “not endorsed” practice was defined as a group average rating of 2.60 or below (Likert scale interval spanning overall average ratings of “disagree” and “strongly disagree”). To aid in determining endorsement, any Round 1 practice rated as “not endorsed” and without any individual ratings of 4 or 5 (“agree” or “strongly agree”) for any attribute within an expert panel was eliminated to reduce content. None met these criteria. Qualitative Data Analysis Interviews were analyzed using principles of thematic analysis [32] and consensus coding [33, 34]. T.S. and J.W. created an initial codebook with a priori codes related to key topics of inquiry and themes from the PCORI engagement measures framework [35]. Under the direction of T.S. and following consensus coding guidelines, two reviewers (E.A. and M.A.) independently coded interview transcripts and Veteran group meeting and individual interview notes. Relevant text excerpts were extracted and imported into a database for analysis. Next, the reviewers convened with T.S. and J.W. to compare coding decisions, discuss discrepancies, and refine code definitions until shared interpretations were reached. This consensus coding process strengthens analytic rigor by making coding decisions explicit, reducing idiosyncratic interpretation, and supporting a stable, transparent codebook for subsequent analysis [33, 34, 36]. Once consensus in the coding was reached, three members of the research team (T.S., M.A., and J.W.) met for an iterative discussion to review comments and identify concepts for each expert panel and overarching themes. The process of analysis for concepts and themes was repeated until consensus was reached. Results Participants Table 2 shows the number of participants for each study procedure among the three expert panels. Approximately half of participating Veterans were female (Table 3). Most Veterans (14/19, 74%) reported 1–5 years of involvement with VA research. More than half of engagement staff and research leaders were female (Table 4). Two engagement staff panelists were also Veterans. On average, engagement staff and research leaders had more than 10 years of experience both in VA research and in efforts to engage patients, within or outside VA, as partners in the research process. Most engagement staff (12/14, 86%) and one‐third of research leaders (5/15, 33%) reported involvement in three or more patient‐engagement activities, such as serving on a national or local Veteran‐engagement professional workgroup, leading engagement efforts for a national study, providing technical assistance to others in VA, or publishing patient‐engagement research. Collectively, the scope of leaders' research focus spanned the translational spectrum from basic science to health systems and rehabilitation research. TABLE 2Participation among Veterans, engagement staff, and research leaders.VeteransEngagement staffResearch leadersRound 1 survey191415Interview19 (100%) a 14 (100%)15 (100%)Round 2 survey17 (89%)13 (93%)14 (93%) a Includes 16 who participated in one of two engagement group meetings and 3 who completed an individual interview. Participation among Veterans, engagement staff, and research leaders. Includes 16 who participated in one of two engagement group meetings and 3 who completed an individual interview. TABLE 3Veteran characteristics self‐reported in Round 1 survey, n = 19.SexMale9 (47%)Female10 (53%)Race a White8 (42%)Black/African American7 (37%)Asian2 (11%)Native Hawaiian/Other Pacific Islander2 (11%)Native American1 (5%)EthnicityHispanic or Latino/Latina2 (11%)Not Hispanic or Latino/Latina17 (89%)Military branch a Army10 (53%)Navy4 (21%)Air Force4 (21%)Marine Corps1 (5%)National Guard2 (11%)Era of service a Vietnam (1961–1975)4 (21%)Post‐Vietnam (1976–1990)7 (37%)Gulf War (1991–2001)10 (53%)Post‐9/11 (2001–Present)14 (74%)ResidenceRural3 (16%)Urban/Suburban16 (84%)Current life role(s) a Employed for pay (full or part‐time)9 (47%)Retired12 (63%)Student1 (5%)Caregiver or parent b 8 (42%)Volunteer worker10 (53%)Unemployed1 (5%)Years involved with VA researchMean (SD)4.6 (2.2)Range1–10 a Participants were allowed to select more than one answer, responses total more than 100%. b Includes being a parent, guardian, and/or caregiver to an adult or child. Veteran characteristics self‐reported in Round 1 survey, n = 19. Participants were allowed to select more than one answer, responses total more than 100%. Includes being a parent, guardian, and/or caregiver to an adult or child. TABLE 4Engagement staff (n = 14) and research leader (n = 15) characteristics self‐reported in Round 1 survey.Engagement staffResearch leadersSexMale4 (29%)5 (33%)Female10 (71%)8 (53%)Did not answer0 (0%)2 (13%)Years involved with VA researchMean (SD)15.9 (9.8)15.2 (10.7)Range4–40< 1–30Years involved with efforts to engage patients as partners in researchMean (SD)12.1 (5.2)11.8 (9.8)Range5–25< 1–30 Engagement staff (n = 14) and research leader (n = 15) characteristics self‐reported in Round 1 survey. Which VA Research Network Veteran‐Engagement Practices Do Veterans, Engagement Staff, and Research Leaders Consistently Endorse as Generally Feasible, Impactful and Desirable? Table 5 summarizes the 44 Veteran‐engagement practices identified as “promising practices” by representatives from one or more of the seven VA health services research networks during the previous environmental scan. Practices spanned three domains (facilitating Veteran‐engaged research, network engagement infrastructure, building and maintaining relationships with partners) and 11 total topic areas. TABLE 5Forty‐four Veteran‐engagement promising practices, across three domains and 11 topic areas (A–K), used by one or more VA HSR research networks and consistency of expert panel agreement to endorse after two rounds of rating.P.Promising practiceTotal Netw.Consistency of agreement to endorse Expert Panel : Comments related to 17 promising practices endorsed with inconsistent agreement to endorse (+) or received “Neutral or Unsure” as overall rating (+/−)Veterans (n = 17)Eng. staff (n = 13)Res. leaders (n = 14) Domain 1: Facilitating Veteran‐Engaged Research Topic Area A. Engaging Veterans in the Research Process 1.Network establishes a national standing Veteran engagement group, relevant to its respective priority area, that is available to consult with VA researchers during a virtual meeting.5++++++2.Network includes Veterans in designing and carrying out a conference to identify VA research priorities (e.g., health services research “State of the Art” Conference).1++++++3. Network refers researchers to existing Veteran engagement groups of other national or local research networks/centers, as mutually agreeable. Engagement Staff: Potential to overload other engagement groups; Other groups may not have relevant lived experiences 3+++++4. Veteran engagement group process/resource: Network staff use a set of standard procedures, with defined roles and responsibilities (e.g., facilitator, co‐facilitator, notetaker), to help efficiently coordinate Veteran engagement group meetings.4++++++5. Veteran engagement group process/resource: Network staff share example acknowledgment language that a researcher can tailor to describe how the researcher's project was influenced by the network's Veteran engagement group and insert into grant proposals, presentations, or publications. Engagement Staff: Use with caution to avoid tokenism 2+++++ Topic Area B. Promoting Engagement Practices to Others 6.Engagement team (network staff and/or Veterans) publishes or presents about their engagement practices to research audiences within and outside of VA.3++++++7.Engagement team (network staff and/or Veterans) publishes or presents about their engagement practices to Veteran/public audiences.1++++++8.Engagement team (network staff and/or Veterans) shares their engagement practices or resources with other Veteran engagement teams (i.e., provide technical assistance to others in VA).2++++++ Domain 2: Network Engagement Infrastructure Topic Area C. Network Decision‐Making 9.The research network's strategic planning process includes the network's Veteran engagement group as a key partner.5++++++10. When considering national VA policies related to the research network's priority area, network leaders consult with an engagement council composed of Veterans employed at very high levels of VA administration, leading Veteran Service Organizations, or similar. Veterans: Concern about missing the voices of “regular” Veterans; Requires Veterans to have specialized knowledge to meaningfully participate Research Leaders: Concern about missing the voices of “regular” Veterans 1++++11. Network includes Veterans in designing and participating in a conference to identify the network's research priorities. Veterans: Requires Veterans to have specialized knowledge to meaningfully participate 1+++++12. As part of the research network's grant‐awarding process, members from the network's Veteran engagement group participate as reviewers of grant proposals using a proposal rating form designed to capture their perspectives on the value and acceptability of the proposed project to Veterans. Veterans: Uncertain feasibility, impact, or desirability Research Leaders: Resource intensive to do well; Depends on topic, scope of review, Veterans' backgrounds; Veteran involvement is more valuable elsewhere in the research process 1++++/− Topic Area D. Network Engagement Support 13.Research network has dedicated paid staff to carry out engagement activities and provides referrals to other Veteran engagement groups, as needed.5++++++14. Research network has no dedicated paid staff for engagement activities. Instead, network refers researchers to existing Veteran engagement groups or resources. Veterans: Networks should dedicate paid staff time to engagement activities; Assumes other engagement groups have capacity Engagement Staff: Quality referrals take time; staff should be paid for this activity; May not be an authentic engagement strategy Research Leaders: Networks should dedicate paid staff time to engagement activities 2+/−+/−+ Topic Area E. Evaluation of Network Engagement Activities 15.Network staff develop an evaluation plan using a conceptual framework and intended outcomes of engagement activities.1++++++16.Network staff offer informal check‐ins with Veteran engagement group members on a regular basis.2++++++17. Network staff regularly meet as a group to review feedback from Veterans and researchers about their experiences with the network's Veteran engagement group meeting processes, and consider need for any changes.2++++++18.Network staff periodically devote a network Veteran engagement group meeting for group planning and maintenance (e.g., reviewing evaluation data and considering changes to membership agreements or group processes).1++++++19.Network staff offer Veterans an opportunity to provide feedback following each meeting of the network's Veteran engagement group.4++++++20. Network staff conduct a post‐meeting survey of the researcher's experience consulting with the network's Veteran engagement group. Research Leaders: Depends on process; Concern about survey fatigue 4+++++21. Network staff conduct routine informal interviews with each researcher 6–12 months after consultation with network's Veteran engagement group to learn how the Veterans' perspectives made a difference in the researcher's project, thinking, or actions. Research Leaders: Concern about network staff capacity 1+++++22. Network staff conduct occasional surveys of the experiences and perspectives of Veteran engagement group members. Research Leaders: Concern about survey fatigue; Uncertain validity of assessment method 1+++++23. Network staff track actual compared to intended output (e.g., number of meetings) and outcomes (e.g., levels of Veteran and researcher satisfaction with each consultation) identified in the network's engagement evaluation plan. Research Leaders: Uncertain value for effort; Unclear outcomes 1+++++ Domain 3: Building and Maintaining Relationships with Partners Topic Area F. Veteran Recruitment for Engagement 24.Network staff use recruitment materials (e.g., recruitment flyer and “group member” description) and a multifaceted outreach strategy to generate interest among Veterans in joining a research network's Veteran engagement group.3++++++25.Research network prioritizes a diverse membership, when recruiting and selecting Veterans to serve on the network's Veteran engagement group.5++++++26. Network staff use a consistent set of questions to interview Veterans (or caregivers) to serve on a network's Veteran engagement group. Veterans: Concerns about inflexibility Research Leaders: Depends on recruitment process and engagement group structure and function 3++++27.Network staff describe to potential Veteran engagement group members the process and criteria for selecting Veteran engagement group members.3++++++28.Network staff share other VA engagement opportunities with members of the network's Veteran engagement group.3++++++ Topic Area G. Veteran Preparation or Training for Engagement 29. Network staff provide Veteran engagement group members an orientation or training prior to their first consultation with researchers. Veterans: “Orient” rather than “train” engagement group members 5+++++30.Network staff send meeting materials to Veteran engagement group members in advance of each meeting.5++++++31.Network staff schedule a learning activity or training, on a topic requested by members, as part of regularly scheduled Veteran engagement group meetings.1++++++ Topic Area H. Researcher Recruitment 32.Network staff use multifaceted outreach (e.g., direct email, network newsletter, presentations, recruitment flyer) to promote the network's Veteran engagement group to researchers.1++++++33.Network staff maintain a webpage on the research network's website with information and directions for researchers to learn more or schedule a consultation with the research network's Veteran engagement group.2++++++34. Prior to scheduling a researcher to meet with the network's Veteran engagement group, network staff email the researcher an “FAQ” document that describes the network's approach to Veteran engagement and provides an overview of the consultation process. (FAQ = Frequently Asked Questions) Veterans : No comments noted 2+++++35. Prior to a consultation meeting with the network's Veteran engagement group, network staff email the researcher a document that describes the composition of the Veteran engagement group. Veterans: May create preconceived ideas about engagement group members; Unclear value to researchers 1+++++ Topic Area I. Researcher Preparation for Engagement 36. Prior to a consultation meeting with the network's Veteran engagement group, network staff email a presentation file template to assist a researcher in preparing slides and questions. (Network staff require a researcher's slides and questions in advance to provide feedback and guidance to the researcher.) Veterans: Concern about burden on researchers 1+++++ 37. Prior to a consultation meeting with the network's Veteran engagement group, network staff meet with each researcher to (1) review the consultation intake form completed by the researcher and again to (2) walk through the facilitation guide and meeting materials developed by network staff. Veterans: Concern about burden on researchers; Value contingent on funded staff to execute 2+++++ Topic Area J. Network Staff Training 38.Network coordinates with other VA resources to offer engagement skills training to staff.1++++++39.Network arranges for one‐on‐one mentoring to build staff's engagement/facilitation skills.1++++++ Topic Area K. Relationship Management 40.As part of joining the network's Veteran engagement group, Veterans receive a membership agreement, from network staff, to sign which describes what they can expect and what is expected of them as a member of the group.2++++++41.The network pays Veterans for their time serving as a member of the network's Veteran engagement group.5++++++42.Network staff communicate with Veteran engagement group members between meetings (e.g., to send reminders about upcoming meeting dates, to share other engagement opportunities with members for themselves or others in their networks).4++++++43.During a Veteran engagement group meeting, network staff recap feedback provided by members at the previous meeting and how the researchers plan to use it.3++++++44. During a Veteran engagement group meeting, network staff report back evaluation results from informal interviews staff conducted with researchers 6–12 months after the researcher consulted with the group. Veterans: Feasibility concern (staff and researcher time) 1+++++ Note: P., practice number; Total Netw., number of the seven networks that reported use of practice; ++ endorsed, consistent agreement; + endorsed, inconsistent agreement; +/− “Neutral or Not Sure” overall average rating. See methods section for complete rating definitions. Forty‐four Veteran‐engagement promising practices, across three domains and 11 topic areas (A–K), used by one or more VA HSR research networks and consistency of expert panel agreement to endorse after two rounds of rating. Network refers researchers to existing Veteran engagement groups of other national or local research networks/centers, as mutually agreeable. Engagement Staff: Potential to overload other engagement groups; Other groups may not have relevant lived experiences Veteran engagement group process/resource: Network staff share example acknowledgment language that a researcher can tailor to describe how the researcher's project was influenced by the network's Veteran engagement group and insert into grant proposals, presentations, or publications. Engagement Staff: Use with caution to avoid tokenism When considering national VA policies related to the research network's priority area, network leaders consult with an engagement council composed of Veterans employed at very high levels of VA administration, leading Veteran Service Organizations, or similar. Veterans: Concern about missing the voices of “regular” Veterans; Requires Veterans to have specialized knowledge to meaningfully participate Research Leaders: Concern about missing the voices of “regular” Veterans Network includes Veterans in designing and participating in a conference to identify the network's research priorities. Veterans: Requires Veterans to have specialized knowledge to meaningfully participate As part of the research network's grant‐awarding process, members from the network's Veteran engagement group participate as reviewers of grant proposals using a proposal rating form designed to capture their perspectives on the value and acceptability of the proposed project to Veterans. Veterans: Uncertain feasibility, impact, or desirability Research Leaders: Resource intensive to do well; Depends on topic, scope of review, Veterans' backgrounds; Veteran involvement is more valuable elsewhere in the research process Research network has no dedicated paid staff for engagement activities. Instead, network refers researchers to existing Veteran engagement groups or resources. Veterans: Networks should dedicate paid staff time to engagement activities; Assumes other engagement groups have capacity Engagement Staff: Quality referrals take time; staff should be paid for this activity; May not be an authentic engagement strategy Research Leaders: Networks should dedicate paid staff time to engagement activities Network staff conduct a post‐meeting survey of the researcher's experience consulting with the network's Veteran engagement group. Research Leaders: Depends on process; Concern about survey fatigue Network staff conduct routine informal interviews with each researcher 6–12 months after consultation with network's Veteran engagement group to learn how the Veterans' perspectives made a difference in the researcher's project, thinking, or actions. Research Leaders: Concern about network staff capacity Network staff conduct occasional surveys of the experiences and perspectives of Veteran engagement group members. Research Leaders: Concern about survey fatigue; Uncertain validity of assessment method Network staff track actual compared to intended output (e.g., number of meetings) and outcomes (e.g., levels of Veteran and researcher satisfaction with each consultation) identified in the network's engagement evaluation plan. Research Leaders: Uncertain value for effort; Unclear outcomes Network staff use a consistent set of questions to interview Veterans (or caregivers) to serve on a network's Veteran engagement group. Veterans: Concerns about inflexibility Research Leaders: Depends on recruitment process and engagement group structure and function Network staff provide Veteran engagement group members an orientation or training prior to their first consultation with researchers. Veterans: “Orient” rather than “train” engagement group members Prior to scheduling a researcher to meet with the network's Veteran engagement group, network staff email the researcher an “FAQ” document that describes the network's approach to Veteran engagement and provides an overview of the consultation process. (FAQ = Frequently Asked Questions) Veterans : No comments noted Prior to a consultation meeting with the network's Veteran engagement group, network staff email the researcher a document that describes the composition of the Veteran engagement group. Veterans: May create preconceived ideas about engagement group members; Unclear value to researchers Prior to a consultation meeting with the network's Veteran engagement group, network staff email a presentation file template to assist a researcher in preparing slides and questions. (Network staff require a researcher's slides and questions in advance to provide feedback and guidance to the researcher.) Veterans: Concern about burden on researchers Prior to a consultation meeting with the network's Veteran engagement group, network staff meet with each researcher to (1) review the consultation intake form completed by the researcher and again to (2) walk through the facilitation guide and meeting materials developed by network staff. Veterans: Concern about burden on researchers; Value contingent on funded staff to execute During a Veteran engagement group meeting, network staff report back evaluation results from informal interviews staff conducted with researchers 6–12 months after the researcher consulted with the group. Veterans: Feasibility concern (staff and researcher time) Note: P., practice number; Total Netw., number of the seven networks that reported use of practice; ++ endorsed, consistent agreement; + endorsed, inconsistent agreement; +/− “Neutral or Not Sure” overall average rating. See methods section for complete rating definitions. After two rounds of rating the feasibility, impact, and desirability of each practice, Veterans, engagement staff, and research leaders consistently endorsed most practices (29/44, 66%, Table 5). Thirty percent of practices (13/44) received inconsistent endorsement by one or more panels. Two practices received ratings of “neutral or not sure” by a panel. Appendices S2a–S2c contain comparisons of Round 1 and Round 2 rating results for each expert panel. When summarized, the 29 network practices consistently endorsed by all three panels formed five recommended responsibilities for VA research networks related to engaging Veterans in the research process (Table 6). TABLE 6Recommended responsibilities for VA research networks related to engaging Veterans in the research process. (1) Establish and maintain a standing Veteran engagement group available to consult with VA researchers (provide a Veteran‐engagement service) Compensate Veterans for their time Prioritize diverse membership Use a multifaceted recruitment strategy with a transparent selection process and criteria Provide orientation/onboarding for group members Establish membership agreements Send meeting materials in advance Use consistent group meeting procedures and involve members in adapting them over time Communicate with members between meetings Recap Veteran feedback provided at previous meeting and share guest researchers' plans for using it (2) Involve the standing Veteran engagement group in network's strategic planning (3) Facilitate Veteran involvement in VA process to identify organizational research priorities (4) Promote a culture of engagement Provide orientation and guidance to prepare researchers for engagement Share other VA engagement opportunities with Veteran group members Provide technical assistance to other engagement teams Partner with Veteran engagement group members to share experiences and lessons learned with research community and public audiences (5) Invest in engagement infrastructure Dedicate paid, skilled network staff to carry out engagement activities Establish an evaluation plan for network engagement activities Recommended responsibilities for VA research networks related to engaging Veterans in the research process. Compensate Veterans for their time Compensate Veterans for their time Prioritize diverse membership Prioritize diverse membership Use a multifaceted recruitment strategy with a transparent selection process and criteria Use a multifaceted recruitment strategy with a transparent selection process and criteria Provide orientation/onboarding for group members Provide orientation/onboarding for group members Establish membership agreements Establish membership agreements Send meeting materials in advance Send meeting materials in advance Use consistent group meeting procedures and involve members in adapting them over time Use consistent group meeting procedures and involve members in adapting them over time Communicate with members between meetings Communicate with members between meetings Recap Veteran feedback provided at previous meeting and share guest researchers' plans for using it Recap Veteran feedback provided at previous meeting and share guest researchers' plans for using it Provide orientation and guidance to prepare researchers for engagement Provide orientation and guidance to prepare researchers for engagement Share other VA engagement opportunities with Veteran group members Share other VA engagement opportunities with Veteran group members Provide technical assistance to other engagement teams Provide technical assistance to other engagement teams Partner with Veteran engagement group members to share experiences and lessons learned with research community and public audiences Partner with Veteran engagement group members to share experiences and lessons learned with research community and public audiences Dedicate paid, skilled network staff to carry out engagement activities Dedicate paid, skilled network staff to carry out engagement activities Establish an evaluation plan for network engagement activities Establish an evaluation plan for network engagement activities How Do Perspectives on Veteran‐Engagement Practices in Use by VA Research Networks Compare Across Veterans, Engagement Staff, and Research Leaders? Respective practice entries in Table 5 summarize written or verbal comments by expert panel members about the 15 individual practices that were not consistently endorsed for VA research networks by all three panels (i.e., either endorsed with inconsistent agreement or rated as neutral or unsure). Engagement staff consistently endorsed more practices (40/44, 91%) than did research leaders (36/44, 82%) and Veterans (33/44, 75%). All practices within two topic areas received consistent endorsement across all three panels (Promoting Engagement Practices to Others [P.6–P.8] and Network Staff Training [P.38 and P.39]). Involving Veterans in strategic planning was the only practice in the topic area of network decision‐making that was consistently endorsed by all three panels. Research leaders were neutral or not sure about involving Veterans as reviewers in the network's grant‐awarding process (P.12). All three panels consistently endorsed networks having dedicated paid staff for engagement activities. In contrast, Veterans and engagement staff were neutral or not sure about a research network referring researchers to standing Veteran‐engagement groups elsewhere in VA, rather than having dedicated paid staff for engagement activities (P.14). Research leaders were the only experts to inconsistently endorse evaluation practices of the networks. Comments about individual practices focused on dependencies that could make a practice less feasible, impactful, or desirable. Seven of the eleven (7/11, 64%) practices that were inconsistently endorsed by Veterans fell under the domain of building and maintaining relationships with partners. In this domain, engagement staff consistently endorsed all practices and research leaders all practices but one. Veterans' comments on most of the seven practices (4/7, 57%) focused on potential burden or uncertain value of a practice to other partners (i.e., researchers or network engagement staff members). What Key Considerations Emerge From Experts' Comments About Research Networks' Promising Veteran‐Engagement Practices? Overall, panelists' comments showed they had positive opinions about the promising Veteran‐engagement practices in use by VA health services research networks. Experts voiced interest in reading about the engagement work occurring among national research networks. As one expert expressed it, “Reading through the practices was thought‐provoking and led me to reflect on my own engagement work and thinking about trying this or that practice.” Another expert noted, “It's great to hear about the amount of work that's going on.” Although 95% of the promising practices were consistently or inconsistently endorsed by all three expert panels, comments revealed three key considerations about challenges in applying promising Veteran‐engagement practices in VA's organizational context. The first theme was the constraint of resource availability, both at the VA research network and organizational levels. Experts' comments about the practice of engaging Veterans as reviewers in a network's grant‐funding process exemplified the challenge at the network level. One Veteran noted, “If extra training were involved so that Veterans were taught how to review proposals, or how various proposals can impact the greater VA structure, this [practice] would be more desirable and impactful.” Similarly, an engagement staff panelist noted, “There are a lot of logistics, time, and effort that go on with this. …I think it is definitely more complicated than what [research network leaders] might want to do to carry this out well.” “We all agree it's really important to have Veteran input into research, but at what stage [when resources are limited]? I think involving them to identify research priorities is more valuable.” Similarly, experts commented on the broader availability of engagement resources and infrastructure within the VA research enterprise. One expert reported, “I think [meaningful Veteran engagement] is really hard to accomplish and it takes a lot of time and it does take resources, and I think it takes resources outside of the individual project.” As another expert framed the VA organizational context, “PCORI [has] so many millions of dollars in infrastructure. They have so many… expert engagement staff members to help facilitate things, so things happen. And [patient engagement] was part of their constitution, when they were created. And, in comparison, we [in VA] are always operating with [fewer funds for engagement].” Experts' perspectives revealed a second theme of commitment to respectful and meaningful engagement processes. As one expert explained the challenge of operationalizing promising engagement practices, “More is not always better… a lot is wrapped up in the execution. How you actually do it makes a big difference to the quality of [partners'] engagement experiences.” Similarly, another expert explained, “It can be very easy to make [a Veteran engagement practice] something that is feasible and feels very… surface. That's not meaningful, authentic engagement.” Another expert reflected this theme in their concern that “Some researchers feel pressed to do what's expedient. In my experience, there's been this desire to try to shorten the time it takes to create relationships in Veteran engagement. Particularly, if a researcher is relatively new to the requirement of Veteran engagement… then they're like, ‘How do I get this done? What do I do?’ and there's pressure to try to turn it into a check box.” Another aspect of this theme was experts' emphasis on the importance of activities that support respectful and meaningful engagement such as ensuring the Veteran engagement group membership reflects the diversity of the Veteran population of interest (e.g., geography, race/ethnicity identities, enrollment in VA healthcare), providing orientation for Veteran engagement group members, and closing the feedback loop with Veterans about how their perspectives were integrated into a project. Attention to the potential burden on partners was the third theme that emerged. For example, comments about individual practices within the Evaluation of Network Engagement Activities topic area often included questions about the “potential for survey fatigue” among Veterans and researchers, whether network staff have capacity to execute evaluation efforts, or if the value returned from tracking engagement output and outcomes over time was worth the invested effort. As one expert summarized, “determining when to engage [Veterans], how often, what the… burden is, etc. is important. We need to look at the priorities for where Veteran involvement is most valuable to best use their time.” Panelists recognized network engagement staff as partners and noted concerns for potential burden on them, such as responsibility for engagement activities without protected time to carry them out. As one panelist stated, “The practice of referring people [to other engagement groups] takes time. You need to ask them about their project and try to find where they can go. You have to do a soft handoff. Anytime I see no protected time for [engagement activities], I get wary because making effective referrals takes a lot of effort.” Discussion We sought to examine perspectives among Veterans, VA engagement staff, and VA research leaders on the feasibility, impact, and desirability of promising practices used by VA research networks to facilitate Veteran engagement in the research process. We found that all three expert panels consistently endorsed most of the VA networks' engagement practices, viewing them as generally feasible, impactful, and desirable. Experts identified three key considerations when selecting practices: availability of resources for engagement, commitment to respectful and meaningful engagement processes, and concern for potential burden on partners. The list of promising practices used by VA national research networks to engage Veterans is generally consistent with an environmental scan of PCORI research networks' engagement activities by Cope and colleagues [37] and with general guidelines for patient engagement in research [9, 38, 39, 40, 41], which likely explains the expert groups' high level of endorsement across promising practices. Our study found unanimous endorsement of practices focused on promoting and building capacity for engagement within VA. These areas were not identified by Cope et al. in their documentation of PCORI practices. The different organizational contexts likely explain why these engagement practice topic areas emerged. PCORI was established in 2010 with mandates to operate a patient‐centered research culture and to provide support and resources that enable patient and consumer involvement in the research process [42]. In contrast, VA's research program, established 100 years ago, began promoting and building capacity for involving patients in the research process relatively recently [21, 43]. Recognition and endorsement by Veterans, VA engagement staff, and VA research leaders of practices that build VA's organizational capacity for Veteran‐engaged research and care is consistent with calls by other longstanding research funders to build partner engagement capacity and facilitation skills alongside shifts in policies to require engagement in proposed research [9, 44, 45] and with operationalizing theoretical learning health system frameworks [3, 16, 46]. The five recommended engagement responsibilities for VA research networks we found highlight specific ways that research infrastructure can contribute to expanding the capacity and culture of both patient‐centered research and learning health systems. Access to a standing engagement group has enhanced many VA investigators' capacity to involve Veterans' perspectives in their research [47, 48] and enabled leaders to involve Veterans in processes to identify organizational research priorities [27, 49, 50] and inform clinical practice guidelines and quality improvement initiatives [47, 48] among other contributions [51, 52]. Expanding the frame of actions and impacts that research engagement infrastructure can provide beyond individual projects to include promoting a culture of engagement may help organizations operationalize not only the concept of engaging patients in learning health system work but also of cultivating the conditions that enable that work to happen [3]. The “people” or relational components of learning health system models and operations are underexplored; our findings add to others' examples of uses and value that engagement infrastructure can bring to organizations [13, 15]. Although research leaders recognized the importance of networks monitoring and assessing their engagement activities, they consistently endorsed only about half of the networks' evaluation practices. This lack of consistent agreement could reflect a lack of common engagement goals, metrics, and assessment methods for standing Veteran engagement panels in VA research or generally in the field of engagement [7, 22, 35]. Numerous panelists in our study emphasized the need to establish practical ways to define and measure success in both quality and outcomes of standing engagement groups, and to close the feedback loops to Veterans about how their perspectives made a difference. Although some detailed reports are available of approaches to and results of evaluating engagement infrastructure within health research networks [14, 47, 53], centers [26, 40, 52], and programs [54, 55], more work is needed to clarify best practices. Experts' perspectives described in this study corroborate others' work, drawing attention to the dilemma of how to best invest limited resources for partner engagement within research and learning health system organizations that historically have not required or funded patient engagement throughout the research process [44, 56, 57]. For example, research leaders expressed uncertainty about whether conceptual benefits outweigh the costs of meaningfully involving patients in study proposal review, consistent with the findings of den Oudendammer and colleagues in the Netherlands [58]. To help prioritize engagement activities, the Canadian Institutes of Health Research articulated a strategy for patient‐oriented research including a supporting framework to guide action [59, 60]. In the U.S., funders such as NIH and VA could support applicants and the research field by collaboratively developing their own strategic frameworks to help advance patient‐centered research within their organizational contexts. Our study expands upon previous work to describe research network engagement infrastructure and activities [11, 12] by gathering organizational stakeholders' perspectives on identified promising engagement practices among intramural research networks. Our findings highlight potential approaches for addressing the dilemma of how to best invest limited engagement resources within research and learning health system organizations that historically have not required, built capacity, nor funded patient engagement throughout the research lifecycle. Integrating patient‐engagement responsibilities into the infrastructure developed by research networks is a potential approach to expand and sustain the involvement of patients not only in research but also quality improvement efforts of healthcare organizations seeking to function as learning health systems. Other integrated health systems may want to replicate this work to identify commonalities and further our understanding of how patient‐engagement infrastructure can best be deployed in different settings. This project has limitations. First, we may have misinterpreted “promising practices” during the compilation of “successful” practices from among those reported by research networks. This was minimized by the two rounds of review with respective network representatives to describe their engagement practices and broadly including practices that representatives perceived or documented as “successful.” The list of engagement promising practices also was limited to those in use by research networks at the time of study, and we did not explore perspectives on engagement approaches outside that list. Second, by choosing to prioritize ease of participation for Veterans and using existing engagement groups' meeting schedules and processes rather than individual recorded interviews, there were greater chances for omissions or inaccuracies in capturing Veteran discussions and for differential group influence compared to the other two expert panels. This was mitigated by working with well‐established engagement groups known for inclusive facilitation practices and having two notetakers. Third, we did not include a separate expert panel of field‐based researchers without programmatic leadership authority. This is mitigated in part by some leaders and engagement staff having dual roles as VA researchers. A strength of our study was the balance we attained in the diversity of each panel to achieve a broader perspective and generalization of consensus with homogeneity, which supports reliability of findings [61]. Further, panelists' voting was anonymous to each other to reduce dominance and group conformity. Future research is needed to efficiently describe and benchmark patient‐engagement infrastructure in different settings, and identify meaningful and practical ways to measure and improve its quality and outcomes. Health care and research organizations seeking to function as learning health systems need to devote planning time, skilled staff, and funds to develop and sustain patient‐engagement infrastructure tailored to their context. Funding This work was supported by Project Award Number SDR 24‐121 from the United States Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Systems Research. Ethics Statement The study described in this manuscript was approved by the Minneapolis VA Institutional Review Board with a waiver of written informed consent on Dec. 29, 2023. We obtained verbal consent, prior to each interview or meeting, for the use of potentially identifiable information collected from each panel member (participant) and affirmed that each participant was aware of the context of such use. Conflicts of Interest The authors report the following potential conflicts of interest regarding this project: Dr. Tracy L. Sides reports grants from VA Health Systems Research during the conduct of the study. Dr. Jennifer Wisdom reports personal fees and other income from the Veterans Health Administration during the conduct of the study. Ms. Malloree M. Argust reports grants from VA Health Systems Research during the conduct of the study. Ms. Erin C. Amundson reports grants from VA Health Systems Research during the conduct of the study. Dr. Diana Burgess reports grants from VA Health Systems Research during the conduct of the study and grants from the National Institutes of Health and the Department of Defense, outside the submitted work. Dr. Laura A. Meis reports grants from VA Health Systems Research during the conduct of the study and grants from the Department of Defense, outside the submitted work. Ms. Agnes C. Jensen reports grants from VA Health Systems Research during the conduct of the study. Dr. Erin E. Krebs reports grants from VA Health Systems Research during the conduct of the study; funding from VA Rehabilitation Research, the National Institutes of Health, the Patient Centered Outcomes Research Institute, and Origin Editorial, outside the submitted work; and during the past 3 years, Dr. Erin E. Krebs has received travel expense reimbursement from the American College of Physicians, the Foundation for Opioid Response Efforts, and the US Association for the Study of Pain. Supporting information Appendix S1: Research network promising engagement practice descriptions for survey Rounds 1 (left) and 2 (right); yellow highlights show changes. Appendix S2a: Specific and overall levels of agreement on the feasibility, impact, and desirability of 44 patient‐engagement promising practices, across three domains and 11 topic areas (A–K), reported by one or more VA HSR research networks: Expert Panel 1—Veterans (Round 1 (R1), n = 19, Round 2 (R2), n = 17). Appendix S2b: Specific and overall levels of agreement on the feasibility, impact, and desirability of 44 patient‐engagement promising practices, across three domains and 11 topic areas (A–K), reported by one or more VA HSR research networks: Expert Panel 2—Engagement Staff (Round 1 (R1), n = 14, Round 2 (R2), n = 13). Appendix S2c: Specific and overall levels of agreement on the feasibility, impact, and desirability of 44 patient‐engagement promising practices, across three domains and 11 topic areas (A–K), reported by one or more VA HSR research networks: Expert Panel 3—Research Leaders (Round 1 (R1), n = 15, Round 2 (R2), n = 14).