Deliberative stakeholder consultations: creating insights into effective practice-change in family medicine

Deliberative stakeholder consultations are a qualitative descriptive study design that uses sequential group debates as the method of data collection. The method is intended for use by family medicine healthcare providers and researchers who are interested in developing practice-change interventions. This method was derived originally in the public policy field to emphasize citizen input over policy leadership input (1). Within the field of family medicine, deliberative stakeholder consultations have the potential increase the relevance and utility of the evidence produced in the family medicine context and enhance the development of practice-change interventions (2,3). Deliberative stakeholder consultations embody the following key elements: (i) use of informed debate in a group setting; (ii) little to no interactions with the researcher (or moderator) during group deliberations; (iii) attention to group dynamics through multi-stage data collection and ethnographic participant observations; and (iv) fewer participant numbers and length of debate to minimize participant burden. The uninterrupted debate aspect is particularly critical with participants being informed about the topic of interest, encouraged to actively discuss and consider other diverse opinions while weighing the merits of competing arguments and arriving at a considered judgment or producing a set of recommendations for action (4). Translating evidence from the world of research to the clinicians’ office is a well-documented challenge in family medicine, given the complexity of clinical practice (5,6). Family medicine may be more susceptible to problems in translating evidence into practice as the interactions with the patients are far more complex as is the healthcare environment (7–9). The complexity includes the fact that family physicians interact with families across age groups and developmental stages; they are often gatekeepers to the health system and they manage a wealth of different conditions rather than one (10). In addition, research may be generated in overly controlled settings that do not reflect the populations seen in family medicine (11,12). As a result, innovations in family medicine practice often experience significant time lags for implementation (11) and, consequently, less than half of North Americans receive the recommended preventative, acute and long-term care (13). To address this implementation lag, evidence that is intended to change clinical practice needs to be adapted and tailored to develop practice-change interventions (11,12). In the knowledge-to-action framework, Graham et al. propose an ‘action’ or evidence application cycle by adapting the evidence to the local context, assessing barriers to the use of evidence, followed by the selection and tailoring of interventions to implement evidence (14). The major assumption in Graham’s framework is that the researcher will have or be able to obtain relevant information through knowledge synthesis to be able to make adaptations to the local context, either before or at the same time as they are assessing barriers to using the evidence. This assumption can be very challenging to operationalize, therefore, we developed a method to engage stakeholders in identifying knowledge translation barriers and guiding the development of practice-change interventions. Although the actual methods for conducting the deliberative stakeholder consultations share a certain commonality with well-designed and well-conducted focus groups, the stronger emphasis on debate encourages a high level of engagement while addressing power dynamics and ethical issues (15–18). In the method brief, we will first provide some of the context that has driven the development of the deliberative stakeholder consultations followed by a description of the study design, data collection, analyses, strengths and weaknesses. We will conclude with a discussion on how this method may improve the knowledge-to-action cycle and ensure our study methods match the patient-centred principles essential to clinical practice. Deliberative stakeholder consultations are not intended to recruit a statistically representative sample, as evidence implementation has precise research objectives that are applicable only to a subset of ‘stakeholders’. Participants, therefore, are selected based on epistemic (i.e. knowledge) diversity. Put simply, participants’ unique lived experience(s) and diverse perspectives are brought to bear on the topic under study (4). The selection of individuals of similar statuses, in this first phase of deliberations, is intended to minimize the effects of group polarization. Group polarization occurs when there is a movement of opinion towards an extreme due to an imbalance in arguments favouring one opinion over another (19). As only a small number of relevant stakeholders are purposively selected, recruitment is feasible even for heavily solicited health professionals. Basic demographics may be collected and reported descriptively to qualify this epistemic diversity. Deliberative stakeholder consultations are held in two phases and the process is illustrated in Figure 1. We will be explaining these two phases in the following paragraphs. Deliberation stakeholder consultation format In Phase I, relevant stakeholders or knowledge users for evidence implementation are identified and invited to participate in individual group debates on implementation barriers. Each stakeholder group is invited to attend a preliminary, small-group deliberation of 1.5 to 3 hours in duration with 6–10 participants to encourage discussion among colleagues with similar ‘power’ statuses (i.e. similar roles or positions—patients, family physicians, administrators, etc.) (20). This is a very short length compared to most public deliberations which are often held over several days (21). The topic or questions that start the debate are based on available empirical evidence on implementation barriers or intervention context. For the initial small-group deliberations, a brief presentation (less than 10 minutes) is provided by a knowledge expert to help frame the discussion. Deliberants can ask questions that are requests for clarifications of the information provided or definition of terms included in the presentation; however, requests for the experts’ opinions on the topic should be avoided. After the question and answer period, a moderator initiates the group discussion using a priori defined questions to incite debate. These questions pre-impose a deliberation structure to frame the debate around the intended focus of the discussion. The moderator, however, remains neutral to promote an environment where deliberants feel they are able freely elaborate on their opinions and depart from the initial questions. The moderator and researchers avoids injecting themselves into the discussion—a crucial point to ensure sufficient quality of debate and a significant difference from focus groups where moderators provide regular input in terms of following pre-defined discussion prompts (22). Although the researchers or moderators may have a great deal of knowledge about evidence in scientific literature, they deliberately avoid influencing the content of the discussion. The moderator can be the researcher or someone experienced with qualitative research methods. In Phase II, a ‘mixed’ consultation is conducted with a subset of participants who are invited by the research team and chosen to maintain diversity from all Phase I groups and follows the same format as the single group consultations from Phase I. At least one member from each group from Phase I should be represented in the Phase II deliberation. Participants are invited to deliberate on areas of disagreement or outstanding questions raised during the first phase. The description of how the data is analyzed from phase one to create deliberative outputs that are ratified by the participants and used to develop the Phase II research questions is described in the paragraphs that follow. The goal of the mixed deliberation is to facilitate agreement on potential implementation barriers for evidence use (18). In some cases achieving full consensus may not be feasible or even desirable, especially when there are marked value conflicts (23). Longo et al provide an example of the use of deliberative stakeholder consultations in the family medicine context (2). All sessions in Phase I and Phase II are audio recorded and transcribed. Note takers produce detailed reports of small-group and large-group deliberations. Participants are asked to complete surveys at the end of each session to assess quality of the debate. Deliberative stakeholder consultations embrace divergent views with supporting arguments, and hence strive to produce ‘working agreements’ rather than ‘forced consensus’ (23,24). Working agreements identify the important concerns surrounding the implementation or intervention, the shared underlying reasons for persistent disagreements, and the positive courses of action or recommendations that should be pursued post-deliberation. These working agreements as well as points of disagreement are considered the deliberative outputs of deliberative stakeholder consultations (25). The deliberative outputs are done by qualitative content analysis of the transcribed debates that are verified by the attending research team members. These are then sent to all participants in a respective group, usually by email, to be ratified. The analyses required for created deliberative outputs occurs at the end of Phase I and the end of Phase II. In addition, after Phase I the transcripts and notes are analyzed for themes that highlight contentious issues arising from the deliberations that inform the debate question for Phase II that is arrived at through consensus agreement amongst the research team. In addition to deliberative outputs, at the end of Phase II, analytic outputs are generated using thematic content analysis. The content of the discussion and the structure of argumentation are examined to provide greater depth in understanding about what points were raised and what was the logic underlying the conclusions reached following the discussion. These analytic outputs are particularly valuable in revealing the arguments or propositions with which there was disagreement, and that influenced the deliberants’ ability to achieve consensus. The deliberative outputs are used to inform future implementation strategies or practice-change intervention development while the thematic analysis is used for more academic dissemination such as peer-review journal publications. The quality of the deliberations for both phases are evaluated in several quantitative and qualitative ways. First, descriptive statistics are derived from a validated survey completed by participants on their perception of the debate, that is, whether they felt their opinions had been represented and/or respected, if they felt heard, etc. Second, the transcriptions are used to assess turn taking and proportion of words spoken for each participant once the debate started (26,27). Finally, ethnographic participant observation conducted by one of the note takers is assessed for types of interactions, particularly as it relates to power dynamics and non-verbal communication (28). Certain individuals or groups tend to have an influence on others’ opinions or willingness to debate particular issues in a way that is not discernible from the written transcripts. Therefore, participant observation with emphasis on power dynamics provides greater depth in understanding how particular deliberative outputs were achieved. One of the strengths of the deliberative stakeholder consultations follows from exposing deliberants to the current state of evidence, while limiting researcher influence on framing the problem of interest, as well as its potential solutions and future implications. The deliberants are not asked to ‘validate’ the researchers’ preconceptions about the importance of certain barriers, for instance; instead the deliberants inductively generate, through debate, their own contextually relevant and experienced implementation barriers, unimpeded by external categories or indicators from other settings. In every deliberation consultation we have completed, the debate began with deliberants putting forward many of the surface-level issues or proposed solutions often identified in the literature and presented as current evidence (e.g. communication with patients). These discussions then rapidly progressed to more complex issues that either nuanced, or led directly to the identification entirely new barriers (i.e. the role of conflicting professional cultures) and potentially innovative solutions. Thus, without interference from the researchers or moderator, the deliberants moved beyond the empirical literature to reveal more complex, contextually relevant issues and areas of interest. In addition, stakeholders often have different and multiple agendas. This method enables each stakeholder to feel ‘heard’ which promotes engagement in the topic under debate thus creating a research tool for development of practice-change interventions that follows the clinical principles of shared-decision making (29). There are some limitations to this method. The first relates to power dynamics. If there are small groups who are identified as key stakeholders who have historically been disenfranchised, the members of this group may not be comfortable presenting their views in the mixed session. This may be the case when patients, particularly those with a stigmatizing condition, are combined with health care providers or others seen to be in a position of power. To minimize this possibility, a series of complete stakeholder deliberations can be held. When completing a set of deliberations with their families of terminally ill children and the health care professionals responsible for treating these children, we completed one set of deliberations with the health care professionals then completed a second set with the parents. In the first deliberation, our small groups were different health care professionals (i.e. one group of paediatric oncologists and one group of paediatric palliative-care physicians for Phase I who then mixed for Phase II). The second complete deliberation was with the parents where one group was bereaved parents and the other was parents with children in treatment. We used the same initial question for the debate but the healthcare providers did not mix with the families. In the family medicine context, this might be healthcare professionals and patients with stigmatizing conditions such as obesity. The second limit is based on different stakeholder’s ability and willingness to participate in debate. For example, some patient stakeholders may not be physically well enough to debate with a group for the required time. Another example is when debating is not culturally appropriate. In this situation, a different method may need to be considered or our method may need to be adapted as we did for Kanien’kehá:ka (Mohawk) community with the use of Talking Circles (30). In this study, we were supported by the Kahnawà:ke Schools Diabetes Prevention Project to engage community stakeholders in the evaluation of the Kateri Memorial Hospital Centre Health Education Program for Diabetes Prevention. The main goal was to engage with teachers, principals and parents to identify and understand facilitators and barriers to the program delivery. Although the deliberative stakeholder consultations were an excellent method to work with a small group where it was unclear what the issues were as was the case in this study, the element of debate or argument does not align with the Kanien’kehá:ka cultural values. Indigenous researchers suggested changing the debate element to Talking Circles that fit well with the consensus-building model of the Haudenosuanee and are considered more respectful to participants. Each member of the Circle states their personal belief about the topic presented before passing to the next person in the circle. This continues until the participants feel they have no more opinions to state. With this adaptation of the method, opinions were stated without interruptions thus preserving an important aspect of deliberative stakeholder consultations. We used the same method of data collection and analyses for this modified method as described in this brief. Deliberative stakeholder consultations provide a forum for meaningful debate, among key stakeholders and knowledge users, around the translation of evidence into practice, while ensuring that researchers’ influence is minimized. They are a valuable means of exposing deeper, and often more problematic issues, while simultaneously generating new courses of action. They can deepen our understanding of issues surrounding care delivery processes and the culture of care. More importantly, the new insights revealed in these debates sometimes challenge or call upon us to question the relevance or applicability of existing conclusions drawn from published evidence. Finally, the deliberations achieve significant stakeholder engagement, a critical aspect of practice-inspired evidence generation and translation. With ongoing use of deliberative stakeholder consultations, this will be research that matters to family medicine stakeholders and improves the implementation of evidence into practice. Conflict of interest: The authors have no conflicts of interest to declare. This work was supported by Genome Canada/Quebec and Canadian Institutes of Health Research (CIHR) through a grant entitled ‘Biomarkers for Pediatric Glioblastoma through Genomics and Epigenomics’ with GB as the GE3LS principle investigator for this study. JG and SP are funded with CIHR Doctoral Scholarships and CL is funded with a FRQ-S Doctoral Scholarship. VR is a CIHR Vanier Canada Graduate Scholar.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.