A protocol for stakeholder engagement in head and neck cancer pragmatic trials

Abstract

Meaningful engagement with stakeholders in research demands intentional approaches. This paper describes the development of a framework to guide stakeholder engagement as research partners in a pragmatic trial proposed to evaluate behavioral interventions for dysphagia in head and neck cancer patients. We highlight the core principles of stakeholder engagement including representation of all perspectives, meaningful participation, respectful partnership with stakeholders, and accountability to stakeholders; and describe how these principles were operationalized to engage relevant stakeholders throughout the course of a large clinical trial.

Background

Approaches to patient involvement and stakeholder engagement have blossomed in the past decade as important ways to improve quality of health care and research [1,2,3,4,5,6]. While patient refers to individuals living with an impaired health condition, stakeholder includes any individual who could be affected by or have a connection with a topic area as a clinician, knowledge user, caregiver, or decision-maker [7]. Stakeholder engagement has roots in the Community Based Participatory Research (CBPR) movement of the 1980s and 1990s, in which all individuals involved in the outcome of a community-based intervention have a say [8,9,10]. Early efforts towards expanding stakeholder engagement into clinical care focused primarily on quality improvement efforts where clinicians involved patients as partners in their own plan of care and engaged them as self-care or self-management agents [11, 12]. These efforts expanded to including patients as partners in co-designing new programs and clinical care models aimed at improving patient experience and the quality of care [13]. Stakeholder engagement is now a key strategy in designing, investigating, and implementing person-centered philosophies or approaches in health care.

More recently, researchers have incorporated a range of stakeholders, including patients, in their initiatives [14,15,16]. For research endeavors, stakeholder engagement aims to increase the likelihood of investigating research questions important to patients and utilizing patient-centered research methods. Researchers have also included stakeholders to accomplish several objectives during the research project implementation: (1) effective participant recruitment, (2) collection of data relevant to stakeholders, and (3) and interpretation of results from multiple stakeholder perspectives [14]. Additionally, stakeholder engagement has also been used to ensure successful dissemination, uptake, and implementation of research findings. Ultimately, the impact of research could be enhanced by involving relevant stakeholders throughout the life cycles of the research effort (e.g., planning, study conduct, dissemination/implementation) rather than participating in only a limited part of the process (e.g., endorsement of a trial concept).

Literature describing how researchers engage stakeholders has grown steadily [17, 18]. Specifically, levels of engagement range from consulting on priorities for investigations before research begins, to serving on steering or advisory committees, to full membership on a research team [19, 20]. Likewise, the timing of engagement can vary. It may occur early, prior to proposal design with a workshop to identify research questions or, during the proposal development and writing. It may also occur only once funding has been awarded and implementation commences or after study completion to discuss findings and plans for action. Various challenges have been identified as researchers seek to engage stakeholders resulting in recommendations to avoid tokenism, to budget sufficient time and energy for each step in the engagement process, and to provide specific training and resources to stakeholders to support engagement [3, 4]. One significant observation is the challenge for all involved to hold non-judgmental, open attitudes about engagement and find ways to interact respectfully and collaboratively. Another is the challenge of finding strategies to overcome the inherent power imbalance that can exist in groups where patients and healthcare clinicians are together [21].

For engagement work with clinical populations with distinct needs, based on our assessment of the existing literature, it is unclear whether general principles of stakeholder engagement are useful across all research initiatives or if special approaches are required for particular disease groups, such as certain types of cancer. Protocols demonstrating how to engage stakeholders successfully, what constitutes success for various levels of engagement, and systematic evaluation of engagement models remains sparsely reported in cancer trials overall [4, 22]. Within cancer research, head and neck cancers (HNC) represent a unique population who experience functional changes that distinguish their experience from other survivors (e.g., changes in fundamental activities like eating, talking, and their appearance) – this distinction is empirically relevant to the composition and operationalization of stakeholder engagement for this cancer survivor population. Yet, there is only one published experience to guide investigators interested in conducting stakeholder engaged research in the HNC populations [23].

No uniform model exists for stakeholder engagement overall or in specific clinical populations, and few presented in the literature have evolved from empirical work [6]. There is growing agreement that it is important to involve stakeholders as early as possible in the research process and engage them in on-going evaluation and feedback about the process of engagement. Therefore, an urgent need exists for protocols that detail population specific models for stakeholder engaged clinical research.

The aim of this work is to describe the protocol for engaging stakeholders in the implementation of a pragmatic trial in a specific HNC clinical cancer population. [24] We will operationalize the model structure, the principles that will guide the model and the overall engagement plan across the various stages of the trial.

Methods

As the Engagement Team for an international, multi-site pragmatic trial aimed at comparing the effectiveness of swallowing interventions during radiotherapy (RT) for HNC (PRO-ACTIVE), we adopted four core principles of stakeholder engagement based on our interpretation of national guidelines for stakeholder engagement, like those from the Canadian Institutes of Health Research (CIHR) [7], National Institute for Health and Care Research (NIHR), [25] and in the existing literature [1, 4,5,6,7,8, 26,27,28,29]. We paid special attention to the stakeholder engagement guidelines provided by our funder, the Patient-Centered Outcomes Research Institute (PCORI) [26, 27]. In particular, we focused on principles emphasizing that all involved in the outcomes should be represented, that mutuality in stakeholder relationships be established, that ongoing relations between the trial team and various stakeholders be reciprocal, and that our plan be guided by opportunities for mutual learning between stakeholders and trial leadership. We also needed to meet the complex and varied needs of a pragmatic, multi-site, randomized clinical trial for a behavioral intervention to be conducted in both the USA and Canada [24]. We adapted these principles for our purposes and intended outcomes in our plan for engaging with stakeholders throughout the duration of the trial in a HNC population.

The PRO-ACTIVE trial

Dysphagia is a major problem facing individuals undergoing RT for HNC. Questions remain about the most effective timing and intensity of swallowing interventions during RT. PRO-ACTIVE is an international multi-center 3-arm pragmatic randomized controlled trial (RCT) which aims to compare the effectiveness of PRO-ACTIVE versus RE-ACTIVE behavioral swallowing interventions among 952 patients with HNC scheduled to undergo radiotherapy [24]. Before commencing RT, eligible patients will be randomized to receive one of three therapy models delivered by oncology-specialized speech language pathologists (SLPs) during the 6–7 weeks of radiotherapy. Regardless of the Arm, all patients will receive a swallow assessment 3 and 12 months after RT completion. The 3 intervention arms are:

• Arm 1 (RE-ACTIVE): Reactive intervention started promptly if/when dysphagia is identified;

• Arm 2 (PRO-ACTIVE EAT): Early low intensity proactive intervention started before RT commences;

• Arm 3 (PRO-ACTIVE EAT + EXERCISE): Early high intensity proactive intervention started before RT commences.

Dysphagia is common following RT, and swallowing function is of high priority for patients, caregivers, and therapy providers [30,31,32]. There are, however, differences in perspective on the manner in which dysphagia impacts health. Patients prioritize issues of depression/anxiety whereas caregivers and clinicians prioritize pneumonia/malnutrition [30, 33]. The perspective of the payer and policy maker has not before been explored, but given that tube dependency affects costs, we expect payers and policy makers to endorse reduction in tube feeding utilization as a priority issue for quality improvement in the area of dysphagia [34]. Clearly, patients, caregivers, providers and (we suspect) policy makers have varying perspectives on the most relevant issues for HNC survivors with dysphagia.

This trial is focused on a behavioral intervention which is applied during a notoriously arduous course of treatment and a time when many patients are known to experience a number of toxicities including increasing difficulty swallowing and maintaining their nutritional status [35]. Numerous physicians and allied healthcare providers work closely with patients and their caregivers during this treatment period. Our approach to stakeholder engagement needed to ensure that these diverse perspectives would be optimally included, and that the complex issues relevant to supportive care in HNC in two distinct healthcare policy environments (Canada and the US) be considered. We also want to ensure policy/payer perspectives are included as future implementation of a successful intervention may include cost implications and require policy change.

The PRO-ACTIVE engagement protocol

Engagement principles

We designed the PRO-ACTIVE stakeholder engagement model structure and plan for engagement around four key principles derived from the literature and from our collective understanding of the needs of the HNC patient population (see Table 1).

Table 1 Key principles of stakeholder engagement adapted for a pragmatic trial

The engagement model

Our engagement protocol was designed to ensure that the key principles of representation, meaningful participation, respectful partnership, and accountability are integrated into all our exchanges with stakeholders. How we operationalized these principles or organized the structure and environments to integrate them in our approach is described below.

Representation

• US and Canadian engagement sites: Because of the important differences in healthcare delivery, reimbursement approaches, and patient experiences between the US and Canada, we designed a comprehensive stakeholder engagement processes to occur at one site in each country. The two sites will work in tandem, with complementary stakeholder panels, under the leadership of the Engagement Team (CM, MF, KH, RM, TM).

• Core and Periphery sites: In the US, where a significant proportion of cancer treatment occurs at non-academic settings, stakeholders representing stakeholder and satellite care facilities will be invited. The proximity to a major medical center is also a crucial factor in the feasibility and possible future uptake of therapies. Therefore, representatives from remote, rural areas will be included in each stakeholder panel.

Stakeholder selection

Identifying relevant stakeholders for this investigation requires thinking about both the conduct of the trial itself and the future uptake and utilization of the trial results in an interdisciplinary complex clinical environment. In our situation, relevant stakeholders will include patients/caregivers, physician providers, allied health providers, and policymaker/payers. (See Fig. 1 for target distributions for stakeholder samples.)

Fig. 1

Target distributions of characteristics for each USA and Canadian stakeholder research partner panel

Meaningful participation

In order to ensure meaningful participation among all stakeholders, we will structure the engagement process to equalize differences in understanding of research processes, and to ensure that all stakeholders have a voice in each phase of engagement.

• Training: In recognition that stakeholders have different backgrounds related to research, the Engagement Team will provide training regarding research methods as part of the early engagement process. Facilitators will have extensive experience in working with low-literacy stakeholders as well as with professional stakeholders and will be able to provide this training as needed at the initial focus group sessions.

• A modular structure based on closed-loop communication: our engagement protocol is based on a series of five task-based modules, with all stakeholders brainstorming solutions to research-related problems, the SAB prioritizing those suggestions and recommending an approach to the research problem, and finally the Trial Executive Committee deciding on stakeholder recommendations and reporting back to stakeholders detailing the research decisions before the start of the next module. (See Fig. 2)

Fig. 2

Proposed Workflow for each six-month long stakeholder engagement module

• Homogeneous Brainstorming Panels: Forming homogeneous stakeholder focus groups will empower the various participants to voice their unique perspectives in a safe and comfortable environment by ensuring that participants meet in sessions comprised of those with similar experiences of HNC. We will utilize the homogeneous groups for brainstorming sessions around research-related questions, allowing participants to contribute more freely and to build on common experiences more easily than is possible in mixed groups. Each site will convene four separate stakeholder panels (i.e., patients and family members, MDs, allied clinicians, and policymaker/payer groups). Each patient/caregiver, MD and AH group will meet once in each module to generate ideas around a common topic.

• Heterogeneous Stakeholder Advisory Board: Following each brainstorming session, we will convene a stakeholder advisory board composed of 2–3 individuals from each of the homogeneous groups to review the recommendations from each of the brainstorming sessions and discuss their merits. They will develop consensus recommendations based on the discussions of the homogeneous panels which will ultimately be presented to the Trial Executive Committee.

Respectful partnership

• Stakeholder Engagement throughout all phases of research process: Stakeholder consultation and collaboration are intentionally built into all phases of this study, from conception through dissemination. The last module will also include planning the future implementation of the successful dysphagia intervention, including stakeholder involvement. Stakeholder panels in both the USA and Canada will work in parallel with close collaboration by the facilitators who will share agendas, inquiry approaches (questions for the stakeholders to discuss), and experiences with their groups throughout the work. Five modules (See Table 2) will guide the dialogue with stakeholders and mirror the phases of the research study (see Fig. 3 for Workflow for Stakeholder Engagement Modules).

• Fair compensation: Based on the guidelines outlined by PCORI, stakeholders will be offered stipends commensurate with their usual rates of pay in addition to compensation for travel and accommodations, and meals at engagement sites [27].

• Meetings at neutral sites: To ensure a free exchange of ideas, stakeholder meetings will be held at non-medical, neutral sites, and scheduled to maximize participation. Members from more distant, rural areas will be given the option of participating via remote video conference if they are unable to join in person.

• Independent,Professional facilitation: All stakeholder meetings will be facilitated by professional research methodologists, who specialize in focus group and stakeholder engagement facilitation. They will be skilled in ensuring that all voices are heard and accounted for. To ensure free and frank discussion, facilitators should be independent of the trial leadership. All meetings will be audiotaped, de-identified, transcribed and summarized by the facilitators following each session [36].

Table 2 Modules for stakeholder engagement meetings (5 modules across 5 years)

Accountability

• Member validation: De-identified notes from each stakeholder panel will be shared among all stakeholders for their validation and further input. Feedback will also be shared from the SAB and the Trial Executive decisions back to the stakeholder panels to ensure closed-loop communication [37].

• Investigator feedback: Once the SABs reach consensus, the Engagement Team will collate the recommendations, prepare a report on that process and recommendations for presentation to the Trial Executive Committee. The Executive Committee members will utilize the report to determine how to incorporate stakeholder recommendations into the trial. Once the final decision has been made, the facilitators will disseminate an accountability report to all stakeholders.

Fig. 3

Model of information flow between stakeholders, facilitators, and trial leadership

Discussion

This paper details a protocol for stakeholder engagement developed specifically for a pragmatic trial of a supportive care intervention during RT for HNC. A model structure and engagement plan were developed for the investigators’ trial design and an anticipated clinical environment for future implementation. Involving stakeholders in clinical trials and clarity about how stakeholder engagement principles will be operationalized in trials is increasingly necessary as funding agencies and investigators recognize the merit of this process by requiring stakeholder engagement in proposal submissions. This tailored protocol offers the first available template for other researchers designing clinical trials with an oncological patient population.

Stakeholder engagement in research is still an emerging phenomenon, especially in terms of how to operationalize engagement principles [3, 38]. Questions remain regarding how best to organize and incorporate principles of fair and fruitful engagement of stakeholders in cases of specific trial designs in the context of specific clinical populations [11]. In planning for a pragmatic trial of a behavioral intervention, we posited that engagement of stakeholders would benefit the conduct of the trial as well as future dissemination and uptake. We anticipated the future uptake will require not only the evidence from the trial regarding the effectiveness of the intervention but also consideration of what is important to patients, how patients and caregivers are able to manage the intervention, and who would support utilization of the trial findings as clinicians or as payers. Therefore, it was important that we include clinicians in all potential sites of care across U.S. and Canada, as well as representatives of payer organizations in both national policy domains.

Meaningful stakeholder input needs to be defined in relation to the type of trial, the practice environment, and both the clinician and patient characteristics. Protocols are needed to guide the uptake and utilization of research results that outline the range of considerations for successful implementation [6, 11]. A primary consideration for stakeholder engagement is the careful identification of who would be considered relevant as a stakeholder given the context of the work. Criticism has been made about using generic panels of individuals who are not experienced in the selected clinical setting or care intervention. Engaging individuals who have experience undergoing the treatment processes under consideration can offer unique insights regarding how interventions are most apt to be successful [39,40,41]. Behavioral interventions will require understanding both patient and clinician behaviors and motivations as well as the environment where the intervention will be taught and performed.

The clinical care environment for this HNC patient group is complex, with a range of clinicians who hold diverse, and often divergent, perspectives about supportive care and treatment priorities. Also, the behavioral intervention would need to be enacted by patients who are already burdened with increasing pain and difficulty swallowing. They may also be limited by socio-economic burdens and uncertainty. Their capacity to engage with the successful behavioral intervention itself is of paramount importance in the future. It is therefore essential that patients who have undergone RT for HNC and their family members advise on all aspects of the trial. Payers also must be consulted because supportive care interventions may not be seen as central to curative care and they may require additional costs and policy changes to implement the successful intervention in current health services models. This stakeholder engagement protocol, therefore, represents a very specific context of supportive care in cancer but may have relevance to clinical populations with similar interdisciplinary considerations and competing priorities at play.

Central to the our protocol was the composition and manner of engaging stakeholders to fit the context of the trial and follow best practices for engagement. By engaging the relevant stakeholders in the research itself and providing an environment where opinions can be shared respectfully without feelings of intimidation, and meaningful relevant input can be obtained to set the stage for success in the future implementation of study findings. Creating such environments requires intentional organization. Drawing from research in other arenas (e.g., psychology, social work, nursing) offers ideas about working with stakeholders effectively [42,43,44]. For example, grouping of individuals with similar experiences (i.e., homogeneous groups) and paying close attention to power dynamics (e.g., who is perceived to be influencers) are helpful. Especially in the world of health care, power dynamics can influence feelings of safety and open dialogue when patients, physicians and/or other health care professionals are grouped together. Separating the stakeholders into homogeneous groups for open brainstorming of ideas, but then organizing a way to build consensus through skilled facilitation, is one approach to overcome this challenge. However, we have made a commitment to report divergent perspectives if consensus cannot be reached.

It is also important to have an independent, trained facilitator to manage the group sessions, who understands the trial environment but does not express a preference for a particular trial outcome [45]. An added concern for the HNC patient population is the reality that some will have challenges speaking (e.g., because of tongue injury or dry mouth) while others may find it difficult to describe their experiences easily, or in a public situation [36]. Some may require added time to express their views clearly which could contribute to frustration within a group situation.

Finally, accountability to the stakeholders is not only respectful but also promotes added engagement. That accountability needs to be about how their input was utilized as well as asking them routinely about how the processes for engagement are working. Designing expectations for stakeholder roles and rules of engagement at the onset is an important step to establish the approaches for accountability. Taking the time for all to understand how the engagement principles are to be handled can truly enrich the research process, its outcomes, and the impact of the work.

Conclusions

Stakeholder engagement requires specific strategies to ensure relevant and meaningful engagement for all concerned. We designed a protocol of stakeholder engagement operationalizing basic principles and incorporating it as an integral component within the design of an ongoing clinical trial. Given that understanding how to operationalize these principles within specific contexts is still emerging, our protocol could be helpful to other research teams who wish to incorporate stakeholder engagement in meaningful, relevant ways in a pragmatic clinical trial particularly in the context of interdisciplinary care of head and neck cancers.