Engaging veteran stakeholders to identify patient‐centred research priorities for optimizing implementation of lung cancer screening

Abstract Background Patient engagement in research agenda setting is increasingly being seen as a strategy to improve the responsiveness of healthcare to patient priorities. Implementation of low‐dose computed tomography (LDCT) screening for lung cancer is suboptimal, suggesting that research is needed. Objectives This study aimed to describe an approach by which a Veteran patient group worked with other stakeholders to develop a research agenda for LDCT screening and to describe the research questions that they prioritized. Methods We worked with Veterans organizations to identify 12 Veterans or family members at risk for or having experience with lung cancer to form a Patient Advisory Council (PAC). The PAC met repeatedly from June 2018 to December 2020, both independently and jointly, with stakeholders representing clinicians, health administrators and researchers to identify relevant research topics. The PAC prioritized these topics and then identified questions within these areas where research was needed using an iterative process. Finally, they ranked the importance of obtaining answers to these questions. Results PAC members valued the co‐learning generated by interactions with stakeholders, but emphasized the importance of facilitation to avoid stakeholders dominating the discussion. The PAC prioritized three broad research areas—(1) the impact of insurance on uptake of LDCT; (2) how best to inform Veterans about LDCT; and (3) follow‐up and impact of screening results. Using these areas as guides, PAC members identified 20 specific questions, ranking as most important (1) innovative outreach methods, (2) the impact of screening on psychological health, and (3) the impact of outsourcing scans from VA to non‐VA providers on completion of recommended follow‐up of screening results. The latter two were not identified as high priority by the stakeholder group. Conclusions We present an approach that facilitates co‐learning between Veteran patients and providers, researchers and health system administrators to increase patient confidence in their ability to contribute important information to a research agenda. The research questions prioritized by the Veterans who participated in this project illustrate that for this new screening technology, patients are concerned about the practical details of implementation (e.g., follow‐up) and the technology's impact on quality of life. Patient or Public Contribution Veterans and Veteran advocates contributed to our research team throughout the entire research process, including conceiving and co‐authoring this manuscript.


| INTRODUCTION
Research funds are limited, making it important to use them in ways that maximize impact on patients and providers. Historically, research priority setting has been controlled by scientists 1,2 rather than individuals affected by the conditions being studied. 3,4 Increasingly, patient engagement is seen as a strategy to ensure that research leads to effective and responsive healthcare service delivery. [5][6][7] Thus, those developing research priorities have begun to seek out the voice of affected individuals in developing research agendas. 1,8 There is some evidence that involving patients throughout the research process, starting with setting the research agenda, helps ensure that the research conducted is patient-centred, 9 while without patient representation, research agendas may not align with information needs of greatest importance to patients. 10 One area where patient input has been limited is lung cancer screening research. The National Lung Screening Trial (NLST) and other studies have demonstrated that screening with low-dose computed tomography (LDCT) reduces lung cancer mortality among adults at high risk of lung cancer. 11,12 Consequently, the US Preventive Services Task Force now recommends lung cancer screening for such individuals. 13 However, these screenings cause harms (anxiety, unnecessary surgeries, radiation exposure, etc.) as well as benefits (decreased lung cancer mortality) for patients. Research suggests that decisions regarding lung cancer screening are difficult for many patients. 14,15 It is thus not surprising that uptake has been slow and uneven, threatening to worsen existing disparities in lung cancer outcomes. 16,17 Consequently, there is considerable interest in research regarding how to optimize LDCT implementation to maximize population benefit. [18][19][20] Lung cancer is of particular importance to US military Veterans as they are often at increased risk for lung cancer due to high rates of smoking and exposure to carcinogens (e.g., Agent Orange). Higher rates of lung cancer have been observed in Gulf War Veterans 21 and among Marine ground troops who served in Vietnam. 22 Not only is their incidence of this cancer higher but their survival rates are also lower than among civilian populations. 23 With three-fourths of the nearly 22 million Veterans receiving some or all of their medical treatment outside the Veterans Administration (VA), 24 engaging Veterans in research agenda setting for LDCT implementation can help guide research both within and outside VA.
In this paper, we describe an approach by which a Veteran patient group developed a research agenda for LDCT screening in collaboration with other stakeholders. We then present the patientcentred research priorities around optimizing LDCT implementation; this was generated through this collaboration.

| Study design
This study reports an approach for identifying patient-centred research questions around optimizing LDCT implementation through facilitating a process of two-way co-learning between two groups of stakeholders: the Patient Advisory Council (PAC) and Stakeholder Group (SG). Co-learning is one of the Patient-Centered Outcome Research Institute (PCORI) principles of patient and stakeholder engagement. The project focused on Southeast Wisconsin, where the research team has deep connections to Veteran groups. Figure 1 shows the project's organizational structure and design, which is consistent with PCORI engagement principles, 25 including (1)

| Recruitment
We recruited 12 PAC members from various racial backgrounds based on (1) fitting the NLST entry criteria (age 55-74 years, 30 or more pack-years of cigarette smoking history and smoking within the previous 15 years), 13,27 (2) perceive themselves as being high risk for lung cancer, but not eligible for the NLST, or (3) having a personal or family 'lived experience' with lung cancer. Only Veterans or their family members were included. Although two members were not Veterans (they were family members of Veterans); we will refer to this group as 'Veterans' throughout the manuscript. We recruited PAC members through direct outreach to individuals known to our Veteran community partners, placement of electronic and printed recruitment materials in prominent locations within the community and word-of-mouth referrals from Veterans we approached about participation.
We recruited 10 SG members to represent lung cancer providers (VA and academic radiologists, thoracic surgeons and pulmonologists), researchers, health system administrators and F I G U R E 1 Project organizational structure and the design PAC and SG members were paid based on meeting attendance in accordance with PCORI's best practice. 28 Some SG members were unable to accept payment as VA employees.

| Meeting principles
Meetings were based on an evidence-based design called the 'building block approach' for consensus building. 29 This model suggests that groups have time to discuss separately and independently when 'potential conflicting issues' arise. As PAC members have the potential to defer to the SG as 'experts', and the SG may expect this deference, the PAC and SG met separately for initial meetings. To develop the PAC's confidence in their value to the research team, they were asked to review and provide feedback on research tools such as recruitment strategies, data collection instruments and outreach letters. After each PAC meeting, we emphasized how their experience provided perspectives that the research team could not have acquired in any other way through a summary of 'How the PAC is making a difference' based on input that they had provided during that meeting. We also coached the SG members regarding the importance of gaining Veteran perspectives and warned them against dominating interactions. All PAC meetings occurred in person in a reserved VA conference room, except the last, which was virtual due to COVID-19.
We arranged two joint meetings for PAC and SG members in year two, adhering to the PCORI engagement principle of co-learning. During the meetings, we used moderation techniques to ensure that there were opportunities for both groups to speak.
Despite these efforts, after each joint meeting, we discussed how to improve the dialogue with both groups (PAC and SG), so that our moderation strategy evolved over time. Additionally, PAC feedback caused us to orient PAC members more fully about the planned joint meeting activities before that meeting. A designated staff took minutes for all meetings. The project staff also summarized each meeting immediately after the meetings and sent the meeting recap follow-up emails to all PAC and SG members.

| PAC and SG meetings
PAC members met quarterly, and each meeting lasted approximately 2 h, with lunch provided. Between June 2018 and December 2020, the PAC had seven meetings: one kick-off, two joint meetings with SG and five PAC meetings. PAC meetings had two major objectives: (1) share personal experiences with lung cancer screening, diagnosis and/or treatment and (2) develop a Veteran-centred research agenda in collaboration with the SG. Responding to PAC requests, learning sessions to address questions related to LDCT from the patients' perspective were incorporated into meetings.
The SG met less regularly, and their meetings were scheduled to coordinate with PAC needs. The SG met three times per year (the last meeting was cancelled due to COVID-19) for a total of five meetings, including one kick-off, two joint meetings with PAC and two SG meetings. The main objectives of the SG meetings were to: (1) translate the PAC's ideas into research questions and (2) generate a sustainability plan to maintain the PAC for future research engagement.

| PAC and SG joint meetings
The PAC and SG convened as a single group two times (joint meetings). These learning collaborative 2-h sessions were designed for colearning. The third joint meeting, a project celebration, was cancelled due to the COVID-19 pandemic. The goals of the first two joint meetings were to develop a research agenda and identify research questions. The project team had a synthesis responsibility during the 'reaching consensus' section on the research agenda, combining the agendas developed by the two groups. The two sets of agendas were revised and negotiated at each joint meeting and presented back to members at the individual meetings (of SG and PAC) in between those two joint meetings.

| Research agenda development process
The research agenda development process involved several steps, including PAC and SG independent meetings, PAC and SG joint meetings and a final voting. The voting process was anonymous following the REPRISE priority setting reporting checklist formulated by Tong et al. 30 Figure 2 shows the sequence of PAC and SG meetings and the workflow of developing the research agenda as a result of these meetings. First, PAC developed a list of eight topic areas that emerged during key discussions around personal experiences in early PAC meetings. Second, PAC members voted on their top three priorities, from among these topic areas ( Table 1). The top three topic areas were as follows: (1) 'How does insurance impact lung cancer screening?' (2) 'What is the best way to do outreach to Veterans for screening?' (3) 'What happens after people get screened?' Third, in the first joint meeting of the PAC and SG, questions were generated under each topic area. After the joint meeting, the project team (the PIs and project staff) edited the list of questions for clarity. The PAC and SG discussed these questions during independent meetings, and then the project team shared the topics discussed across the groups. Fourth, during the second joint meeting, the PAC and SG continued to revise, refine or add questions. This led to 20 questions, which the project team again edited for clarity. Finally, the PAC ranked their top three priority research questions in the final review. Table 2 shows the final PAC-approved agenda with the top three research questions in bold (Table 2). These research questions were as follows: (1)

| Characteristics of PAC and SG members
Among 12 PAC members we recruited, one-third were women, one was African American and one was Hispanic. The 10 SG members

| Research agenda
The agenda reflects Veterans' prioritized research topics influenced by their life experiences. These topics were different from and sometimes viewed as less significant than the topics identified by researchers.

| Priority topics
The PAC prioritized their top three topics of interest by their third meeting (  The second topic focused on the best way to do outreach to encourage screening. Much of the concern on outreach revolved around those Veterans who were 'falling through the cracks': the young, uninsured, isolated and marginalized (e.g., dishonourably discharged or rural Veterans

| Research agenda and priority questions
In the three prioritized topic areas, two cycles of feedback generated 20 research questions ( (2) Do screening results affect psychological health?
This was a prioritized topic early on in PAC meetings and remained so through multiple iterations of the research agenda. Most PAC members expressed concern over waiting: waiting to see a specialist, waiting for results and waiting for further care.
The emotional burden and stress of uncertainty were considered potential triggers for Veterans who are 'already psychologically fragile'. The stress of waiting for results or having to make a decision based on complex health information can compound mental health issues. Some PAC members felt that this added anxiety might be a cause for screening avoidance. for Veterans) may play a bigger role in leading up to the current situation (i.e., chronic diseases, medical non-adherence, higher mortality rates). Only by overcoming this challenge will patients and providers each be able to adjust the angles of their respective lenses so that their vision can come into common focus. 36  we ensured that the PAC made decisions independent of the stakeholders who advised them, we cannot be sure if endorsement by respected professionals caused some questions to be assigned more weight than a purely patient-driven approach. However, the PAC saved their highest rankings for questions that had not been suggested by the stakeholders. Thus, since the objective of the study was to generate informed patient-centred research questions and priorities, we believe that this approach achieves an appropriate balance.
We anticipate that the next steps are twofold. First, policy makers, practitioners and decision-makers within local VA and healthcare systems need to be informed about our findings. The research team has already started the dissemination process. For example, our findings were disseminated to the statewide Veterans service organizations. Second, final priorities will be disseminated with researchers so that they can utilize this information to guide the design of future research projects. In fact, our PAC members already provided additional consultation to several VA researchers outside the study. Together, we plan to investigate many of the priorities outlined here with large comparative effectiveness research studies.
Patient engagement has the potential to enrich our understanding of patient priorities for research. Given the current focus on developing patient-centred research questions, we suggest future studies to evaluate various patient engagement approaches and determine what approach works best, under which circumstances. Successful approaches will build trust in the patient-researcher partnership, ensure that patients are meaningfully engaged throughout the process and include diverse patient experiences and perspectives. We present our patient engagement approach as one option for groups interested in facilitating collaboration among multiple stakeholders to identify shared research priorities. The study highlights a hypothetical counterfactual-how different would research priorities be if selected solely by scientists versus this diverse patient advisory council. We noted that the research priorities were not the same as those initially identified by the scientists/health professionals, which points to the outcome of this hypothetical counterfactual. This study calls for future studies to assess whether lung cancer screening actually becomes better optimized by targeting research priorities identified by scientists or priorities identified by patients through a patient-centred engagement process. While we likely will not truly know the answer to that experiment, the current study does provide valuable guidance around how to conduct a high-quality patient advisory council.