Introduction
Homelessness is a major societal concern given its prevalence and personal, health, social, and economic consequences in major Western cities (Homeless World Cup Foundation, 2019; Laird, 2007; Latimer et al., 2017; Latimer & Bordeleau, 2019). Persons experiencing homelessness tend to use emergency response services when they need to contact the health network and when they are in crisis or under the influence of psychoactive substances (Abramson et al., 2021; Charette et al., 2011; Doran et al., 2018; Durant & Fahimi, 2012; M. Moore et al., 2019). Homelessness, substance use, and mental health problems are frequently associated, which can lead to many perceived inappropriate behaviors and other consequences for those who experience them, such as agitation, incivilities, risk-taking behaviors, aggressiveness, and suicidal ideation (Baggett et al., 2013; Kidd et al., 2017; Noël et al., 2016; Nolan et al., 2014; Richer et al., 2013). Homeless persons’ increased visibility in public settings may lead to more frequent calls to emergency services (Parazelli et al., 2013). This situation may also be explained by the extent and complexity of their needs and of the barriers hindering their access to care (Argintaru et al., 2013; Chambers et al., 2013; Doran, Vashi, et al., 2013; G. Moore et al., 2011).
Current emergency response is often stigmatizing and insufficiently adapted to the needs of homeless persons, including answering their basics needs or responding when they are under the influence of psychoactive substances. As a result, the responsibility of caring for these people shifts from one ill-equipped organization to another, especially between the emergency services themselves such as police, paramedics, and hospital services. This has led to the phenomenon of revolving doors and the emergence of collaboration issues (Chang et al., 2014; Choi et al., 2011; Doran et al., 2014; Doran, Ragins, et al., 2013; Doran, Vashi, et al., 2013; Farmer, 2012; Maharaj et al., 2013; Raven et al., 2010).
Our study thus aimed to contribute to the improvement of the emergency responses to substance users experiencing homelessness. Given that the needs differ from one context to another, it was essential to focus on concrete needs and recommendations that would be specific to our local context. We therefore opted for a need analysis to intersect the perspectives of multiple stakeholders, especially persons who have experienced homelessness and addiction, emergency response actors (i.e., police and professionals from hospital emergencies), and other actors who are involved in some way in this response (professionals from specialized services and community-based organizations; people from the political and research communities).
We considered deliberative democratic evaluation a particularly interesting avenue in this regard. This collaborative method aims for the inclusion of all the people concerned by an issue and for a consensus when different perspectives are presented (Duplin, 2007; House & Howe, 1999, 2002). Nonetheless, it is still little used despite the growing popularity of participative methods in qualitative research, and few papers propose concrete ways to apply it (House & Howe, 1999; Hreinsdottir & Davidsdottir, 2012; Ryan, 2004; Ryan & Johnson, 2000). The goal of this article is to demonstrate the relevance of combining deliberative democratic evaluation with a concrete way to apply it, the World Café, for addressing complex issues, through an example in the field of addictions.
Deliberative Democratic Evaluation
Democratic evaluations are participative evaluations that are positioned to serve the community by promoting education and empowerment (Howe & Ashcraft, 2005; MacDonald, 1974; Ryan, 2004), which makes them a good fit with qualitative research. They are used as an instrument of negotiation, social justice, and empowerment. To do this, democratic evaluations aim to involve all interest groups, rebalance the relationships of power in decision-making, and produce information that is useful and used (House & Howe, 1999; MacDonald & Kushner, 2004; Théberge, 2011). Deliberative democratic evaluation also meets these objectives, but with an emphasis on: including the more vulnerable and marginalized persons in the decision-making; co-constructing knowledge by distributing objective and credible information; and formulating valid, practicable recommendations (House & Howe, 1999).
Conducting this evaluation requires adherence to three basic principles: inclusion, dialogue, and deliberation (House & Howe, 2002; Howe & Ashcraft, 2005; Ryan, 2004; Théberge, 2011). Inclusion encourages consideration of the most relevant viewpoints, that is, those of the people concerned by a given issue, program, or policy (House & Howe, 1999, 2002, 2003; Hreinsdottir & Davidsdottir, 2012). This means that all stakeholders must be included. In the field of addictions, active substance users should thus be included, which is consistent with the recent move to integrate users within research (Damon et al., 2017; S. Greene et al., 2009; Neale et al., 2017; Shippee et al., 2015). Dialogue is the exchange process where objective information is provided and experiential knowledge is shared (Boyko et al., 2012; House & Howe, 1999, 2002). There is thus a co-construction of knowledge and the development of a common understanding (Boyko et al., 2012; House & Howe, 2002; Hreinsdottir & Davidsdottir, 2012). Deliberation consists in thoroughly examining and debating different conceptions, such as values and social representations (Duplin, 2007), in order to make collective decisions and formulate recommendations (House & Howe, 2002; Howe & Ashcraft, 2005; Théberge, 2011). According to several authors, it is not the ultimate goal that matters, but rather the effects of the deliberation on the participants (e.g. respect and openness) (Chambers et al., 2013; Duplin, 2007).
Evaluators and researchers are important players in this evaluation (Contandriopoulos & Brousselle, 2012; House & Howe, 1999; Ryan, 2004). They must structure the activities to favor adherence to the three principles and to produce valid conclusions (House & Howe, 1999, 2002; Ryan, 2004). They must anticipate potential power imbalances and implement mechanisms to reduce them both upstream (e.g. choice of methods) and during the activities (e.g. reframe the overflow of emotions, encourage the more discreet persons to express themselves) (Hreinsdottir & Davidsdottir, 2012; Théberge, 2011). It is thus recommended that evaluators and researchers be neutral and independent during the collection and analysis of data (Contandriopoulos & Brousselle, 2012; MacDonald & Kushner, 2004; Ryan, 2004). Otherwise, this type of evaluation fits well with pragmatism and transformative-emancipatory values as an epistemological position: choosing methods that are best for meeting research goals, creating useful knowledge and democratizing it, seeking ways to diminish the effects of oppression and discrimination, and integrating the research into an iterative and continuous learning process with social justice values (Biddle & Schafft, 2015; Long et al., 2018).
Although it is primarily used in education, this evaluation is easily transposable to other contexts, especially those with top-down management, generalized concerns, and potential participants who are curious and open to other viewpoints (Contandriopoulos & Brousselle, 2012; Hreinsdottir & Davidsdottir, 2012; MacDonald & Kushner, 2004; Ryan, 2004). For its implementation, the authors recommend using an innovative combination of methods and qualitative designs, taking the local context into account (House, 2001; Hreinsdottir & Davidsdottir, 2012; Ryan, 2004).
The World Café
The World Café is a dialogical, creative, and collaborative method. It aims to create a network of exchanges between people with different viewpoints on a given theme (MacFarlane et al., 2017; The World Café, 2008). Its use is growing in research (Schiele et al., 2022), especially in qualitative, health, and education fields (e.g., Alunni-Menichini et al., 2020; Banfield et al., 2021; Decker-Lange et al., 2021; MacFarlane et al., 2017; Restall et al., 2016).
Its implementation is flexible and can be adapted to various contexts and several recommendations are proposed. Good preparation is suggested, such as identifying in advance the theme (e.g. issues of interest), objective(s), potential participants, and parameters of the activity (e.g. three rounds of discussion periods). A pleasant atmosphere should be created, where all participants are encouraged to share and combine their ideas through dialogue, written communications, or drawings. Organizers are encouraged to give participants the opportunity to change tables (i.e., rotations) and meet new people to cross-pollinate knowledge. At the end of the round, a host remains at the table and summarizes the previous discussions at the next round. Thus, participants can enhance and add nuance to the comments previously reported by others. Finally, it is recommended to end the activity by identifying the themes that form patterns (Brown, 2010; MacFarlane et al., 2017; The World Café, 2008). These exchanges can be recorded and a thematic analysis can be done afterward (Alunni-Menichini et al., 2020; Khong et al., 2017).
Combining the World Café and the Deliberative Democratic Evaluation: The Example of Our Project
Our study was designed to fulfill two needs: 1) identify avenues for concrete solutions to improve the emergency response to substance users experiencing homelessness in Montreal; and 2) identify strategies to help emergency response stakeholders improve collaboration (Farmer, 2012; Parazelli et al., 2013; Théberge, 2011).
We opted for an innovative evaluative study combining the needs analysis (Altschuld & Watkins, 2014; Watkins et al., 2012) and the deliberative democratic evaluation (House & Howe, 1999, 2002) with a qualitative design (Miles & Huberman, 2003). The deliberative democratic evaluation allowed us to focus on: 1) the local context (House, 2001; Hreinsdottir & Davidsdottir, 2012; Ryan, 2004); 2) the inclusion of all emergency-response stakeholders, including substance users experiencing homelessness, while striving to reduce the inequalities of power (House & Howe, 1999, 2002); and 3) the effects of the evaluation (Duplin, 2007). For the purposes of our study, the World Café seemed interesting as a data collection activity. First, it was recommended in the needs analysis (Altschuld & Watkins, 2014; Watkins et al., 2012). Secondly, it has the potential to follow the principles of deliberative democratic evaluation (House & Howe, 1999, 2002). Finally, the World Café seemed a relevant way to collect rich qualitative data: data resulting from a crossing of the perspective of people who usually have little opportunity to exchange with this dynamic and collaborative process. As this was the first attempt at combining the World Café and deliberative democratic evaluation in qualitative research, we wanted to demonstrate the relevance of this combination of participants’ perspectives.
Methods
World Café Participants
We received authorization from an ethics committee to conduct this study. Every participant was recruited through expert sampling, i.e. administrators of the organizations involved and the members of our committee of experts (Baker, 1996). Thirty-four people participated in the World Café, excluding the note-takers and organizers. There were ten police officers, two of whom specialized in mental health or homelessness; five professionals from addiction- or homeless-related services in health and social services system, including specialized outreach teams; one professional from a psychiatric hospital; six community-based stakeholders in the field of addiction, mental health or homelessness, two of whom had previously experienced homelessness; three representatives of the homeless persons who also had experiences of homelessness and addiction; three political actors at the municipal and provincial level; and six research community members with expertise in addiction, mental health, or homelessness.
The activity began with the participants signing a consent form, as requested by our ethics committee (100%; n=34/34) and handing in the sociodemographic questionnaire (97%; n=33/34). Of the professionals who responded to the questionnaire (97%; n=30/31), most were women (63%), had a university degree (78%), and had worked in their organization for at least five years (63%). Most professionals (77%, n=23/30) interacted regularly to frequently (more than five times per month) with persons experiencing homelessness. However, interactions with emergency response professionals seemed to vary depending on the type of stakeholder (see Table 1).
The three people who had previously experienced homelessness responded to a questionnaire concerning substance use, the ASSIST V3.0, in addition to the sociodemographic questionnaire. In summary, they were three men whose average age was 53 years, with education levels varying from high school to university and with a history of poly-drug use. In the three previous months, they had all lived in an apartment and had continued to use three of the following substances: tobacco, cannabis, cocaine, sedatives, and/or alcohol. Two were in a methadone treatment program. Only one mentioned having had health, legal, or financial problems related to his cocaine use in the previous three months.
Finally, due to challenges associated with the ethics-related application and recruitment, health and social service professionals were less well represented at the World Café. Individual interviews were conducted to ensure adherence to the principle of inclusion. Individual interviews were also conducted with the substance users experiencing homelessness to obtain their viewpoint in a context free from power-related issues. However, the results of these interviews are not presented in this article.
World Cafe Process
To promote adherence to the three principles of deliberative democratic evaluation, our team identified and implemented several strategies upstream, during, and after the World Café (see Table 2 and the World Café Participants section). For example, we recommend that community workers accompany the representatives of persons experiencing homelessness and addiction to promote inclusion. We added conferences to provide objective information and promote the principle of dialogue. Furthermore, we encouraged participants to prioritize three themes at the end of each round to promote the principle of deliberation.
The World Café lasted one day. There were six theme-based tables concerning the dimensions of our needs analysis (see Table 3; Alunni-Menichini et al., 2020). A “neutral” host and a “neutral” note taker from the research community were assigned to each table. Upon their arrival, the participants were invited to join the table of their choice, having been asked to sit with stakeholders in different categories from their own (assignment of one color per category of actors).
The World Café process is illustrated in Figure 1. The day began with a brief presentation of the objective of the day, the process, and the guidelines (e.g. be respectful, listen to others, be open, summarize one’s thoughts) (The World Café, 2008) by the facilitator of the day. For the purpose of the needs evaluation, the activity was divided into two sessions (Watkins et al., 2012): 1) current emergency response; and 2) ideal emergency response.
At the beginning of the first session, all participants were encouraged to respond to a general question. Next came three rounds of discussion concerning issues specific to the thematic tables. Before the rounds ended, the participants were asked to prioritize three elements to retain and write them on a board (see Figure 1). Each round was followed by a rotation and began with a summary by the host, to encourage more in-depth discussions. The second session began with a general question also but, for feasibility reasons, it included only one round on the thematic questions. At the end of each session, a synthesis of the elements prioritized by each of the tables was presented by the hosts to all the participants, who were then invited to comment on the results. Throughout the day, participants were invited to write their ideas on the boards during the rounds or the breaks (The World Café, 2008). Although the participants could not join all the thematic tables, the syntheses and the boards gave them access to all the content. Furthermore, four brief conferences were given during the day (see Figure 1). In our opinion, the small-group discussions and large-group activities made the knowledge acquisition and pollination possible.
Each exchange was recorded and a thematic content analysis was made. The results of this thematic content analysis, combined with that of individual interviews, are presented in another article (Alunni-Menichini et al., 2020) as it is not the focus of this methodological article. In summary, the data collected was the result of a discussion and a debate of ideas, and they provided a systemic perspective on a complex issue.
Participants’ perspectives of the World Café: Data collection and analysis tool
The participants filled out a satisfaction questionnaire immediately after the World Café to evaluate: 1) satisfaction levels (reception, venue, atmosphere, first session, and second session) using a Likert scale from 1 (very dissatisfied) to 5 (very satisfied); 2) knowledge acquisition using one dichotomous question (Did you acquire new knowledge?) and three open-ended questions (Which ones, how will these acquisitions influence your practice, and how will these acquisitions influence your relationships with the other stakeholders ?); 3) networking with other key actors using a Likert scale from 1 (strongly disagree) to 5 (strongly agree); and 4) adherence to the principles of deliberative democratic evaluation according to Corus and Ozanne’s (2012) criteria with five questions (In your opinion, to what extent did this World Café: include all emergency response stakeholders, ensure unconstrained and uncoerced discussions, allow sufficient time for the presentation of objective information, allow sufficient time for deliberation, and respond to your needs and expectations?) based on a Likert scale from 1 (strongly disagree) to 5 (strongly agree). When choosing these questions, we considered the possible effects of a deliberative democratic evaluation according to Corus & Ozanne (2012), the ingredients that facilitate collaboration (Bentayed, 2010; Guerrero et al., 2014; INESSS et al., 2016), as well as the evaluation of positional clarification events hosted by the Montreal cross-training program in mental health and addiction (Perreault et al., 2020). Positional clarification events were chosen because their characteristics resemble those of our World Café (e.g. intersectoral exchanges concerning a particular issue to improve networking). Finally, there was an “Other Comments” section. We performed descriptive statistics and a thematic content analysis. All the participants’ comments presented in this article are translations from French.
Results of the Satisfaction Questionnaire
Twenty-three participants (68%, n=23/34) completed and returned the satisfaction questionnaire. This low response rate may be explained in part by the time it was administered—that is, at the end of the World Café. The following results are the evaluation of: 1) the participants’ general satisfaction; 2) the effects of our World Café in terms of knowledge acquisition, transformation, and networking; and 3) adherence to the principles of deliberative democratic evaluation, from the participants’ perspective.
General Satisfaction
The respondents were “satisfied” to “very satisfied” with the activity (see Figure 2), in terms of reception (x=4.48/5, n=23/23), venue (x=4.57/5, n=23/23), atmosphere (x=4.52/5, n=23/23), and the two sessions (x=4.50, n=22/23 and x=4.57/5, n=21/23, respectively). Seven of the 23 respondents (30%) commented on their general appreciation of the day (n=9 comments) in the “Other Comments” section. All the comments were positive. For example: “Well done; it was my first World Café and I found it extremely relevant and interesting to be able to hear the stakeholders’ different points of view” (P14). One suggestion was made: “It would have been interesting to spend more time on the solutions, on what to implement [second session]” (P9).
Effects of the World Café from the participants’ perspective
Knowledge acquisition
Nineteen participants (80%) indicated having acquired new knowledge and twenty-one (91%) wrote comments in this regard (n=23 comments). Eight participants (36%) mentioned gaining a better knowledge of the resources, the people who work there, and the mechanisms for accessing certain services, for example: “that there are other actors involved in the community network for homeless persons who are in crisis and use substances” (P19). Eight participants (36%) wrote that they had better knowledge of the other actors’ realities, their roles, and the limitations of their work, for instance: “several aspects of the actors’ various jobs were unknown to me” (P14). One specified needing further clarification. Three participants (14%) indicated having a better knowledge of the current situation and of the issue. Two participants (9%) mentioned noticing common views and objectives with other stakeholders: “We agree on the problem and on the solution” (P2). One person (4%) learned more about substance users and another one (4%), about the legislative procedures.
Transformation: Influence of the acquisitions in practice, representations, and relationships with others
Thirteen participants (57%) commented on the immediate or potential influence of their acquisitions on their practice and their relationships with other stakeholders (n=16 comments). Most of the comments (84%, n=16/19) concerned the elements that could help improve collaboration between the key actors. Seven participants (30%) mentioned a change in how they perceived the other stakeholders and wanting to approach them in a more open, flexible, and consistent manner. For example: “I think that it will slightly change the way I address police officers in the future” (P5). Four participants (17%) said they wanted to decompartmentalize the services by improving communications between the services, improving the intersectoral work, or filling out more information-sharing forms. Three participants (13%) indicated being able to better refer persons experiencing homelessness according to their needs, “Use of my new knowledge of the other resources to better guide people with respect to their needs.” (P16).
Finally, three participants mentioned other effects of this activity: “participating in improving the system as a peer supporter and in providing more quality care” (P7), wanting to transfer the new knowledge to his or her environment, and to develop new research projects.
Networking
According to the eighteen participants who responded to this question (78%), the World Café helped develop relationships with other stakeholders (x=4.06/5). Five participants (22%) wrote comments along these lines: “[it] enabled me to widen my network” (P9) and “exchange of contact information” (P20).
Compliance with the principles of deliberative democratic evaluation
Most of the respondents indicated that they agreed or strongly agreed that the activity adhered to the principles of deliberative democratic evaluation. As shown in Figure 3, this activity included all the stakeholders (x=4.14/5, n=22/23), was free of constraint and coercion (x=4.77/5, n=22/23), allowed sufficient time for the presentation of objective information (x=4.27/5, n=22/23) and for deliberation (x=4.24/5, n=21/23), and met their needs and their expectations (x=4.36/5, n=22/23).
The principle with the highest agreement score was an absence of constraint and coercion (x=4.77/5). One participant wrote this comment about this principle: “The responses to the various questions were given without judgment and with an understanding of the lives of persons with addictions and experiencing homelessness” (P14).
The principle with the lowest agreement score was the inclusion of all the stakeholders (x=4.14/5). Consistently with these results, three participants (13%) indicated that some actors were missing: decision-makers and health professionals, such as physicians, nurses, and paramedical technicians. One participant considered the presence of decision-makers to be essential in a change process: “Decision-makers must be involved; nothing will change without them.” Two participants (9%) wrote positive comments: “The stakeholders were well chosen” (P14) and “Being able to discuss with service users was very enriching” (P8).
Discussion
We were interested in the contribution of the World Café and, more specifically, as part of a deliberative democratic evaluation. We will thus discuss the results of the satisfaction questionnaires, the benefits of this combination for both the deliberative democratic evaluation and the World Café, its relevance to qualitative and participatory research as well as the challenges and limitations of its implementation.
Thirty-four participants involved in different areas of emergency response gathered during our World Café. The twenty-three participants who responded to the satisfaction questionnaire were “satisfied” to “very satisfied” with the activity. Most of the respondents indicated having acquired new knowledge, notably concerning the resources, the other stakeholders’ realities, and the current situation. Slightly more than half indicated that this new knowledge influenced their representations and their future practices and professional relationships. These results are consistent with those from Terry et al. (2015) and Hafford-Letchfield et al. (2021). More specifically, participants indicated: having changed the way they will perceive and approach the other actors in the future, wanting to decompartmentalize the services through joint activities like the World Café or information-sharing forms, and being able to better refer the people according to their needs. Most of the respondents said that this activity also helped improve networking, as is the case in similar activities (Hafford-Letchfield et al., 2021; Perreault et al., 2020). The World Café thus combines several ingredients considered by several authors to help improve intersectoral and interorganizational collaboration (Bentayed, 2010; Guerrero et al., 2014; Perreault et al., 2020).
Moreover, the World Café integrated well with the deliberative democratic evaluation because, according to most of the participants, it enabled adherence to the required values and principles (Corus & Ozanne, 2012; Duplin, 2007; House & Howe, 2002; Théberge, 2011). The participants also emphasized the constraint- and coercion-free atmosphere of the World Café. We believe the strategies implemented to reduce the power imbalances and the sufficient time allotted for discussion contributed to this result (Contandriopoulos & Brousselle, 2012; Hafford-Letchfield et al., 2021; Hreinsdottir & Davidsdottir, 2012; MacDonald & Kushner, 2004). Nonetheless, there were either too few or none of certain types of actors whom the participants considered important (e.g. decision makers).
So, the World Café is a concrete, accessible application for researchers and evaluators interested in using deliberative democratic evaluation; it is short in duration and has the capacity to comply with the principles of inclusion, dialogue, and deliberation. Moreover, deliberative democratic evaluation helps enrich the World Café by ensuring that the viewpoints of all the stakeholders are integrated in the decision-making process, including those who are still insufficiently consulted (House & Howe, 2003; Hreinsdottir & Davidsdottir, 2012; Terry et al., 2015). This is the case for active substance users, who are still not sufficiently integrated in research or evaluation processes (Lazarus et al., 2014; Neale et al., 2017). In our opinion, deliberative democratic evaluation facilitates their integration by encouraging evaluators to think about inclusion of all the stakeholders in a program or an issue, anticipate potential power inequalities, and reflect on the strategies that should be implemented to reduce them (Corus & Ozanne, 2012; House & Howe, 1999, 2002). This is, in our perspective, an important consideration in any participatory research that includes vulnerable and socially excluded people. Furthermore, the deliberative democratic evaluation contributed to better structuring the World Café and facilitating its implementation. For example, we allocated five minutes at the end of the rounds to deliberate and prioritize the three most important reported ideas, which ultimately also facilitated the development of the syntheses and the data analysis.
Finally, the combination of deliberative democratic evaluation and the World Café is interesting, and it can be easily used in qualitative research, since they are dialogical methods. It promotes the intersection of different perspectives, a debate of ideas and the development of consensus among various groups of key actors gathered during a single activity. It allows co-construction of knowledge and a learning transformation (Brown, 2010; Hafford-Letchfield et al., 2021; Mezirow, 2000; Terry et al., 2015), because it combines pollination of the experiential knowledge of the World Café (Brown, 2010) with access to objective information from the deliberative democratic evaluation (House & Howe, 2002). It also helps provide a systemic view of an issue, not only part of it like too many studies tend to do (Alunni-Menichini et al., 2020; Hafford-Letchfield et al., 2021). For these reasons, it has the potential to provide rich qualitative data, data that is different from the other data collection methods. This is particularly relevant when complex issues are at play and for which the involvement of several groups of stakeholders is required—for example, with co-occurring disorders (Perreault et al., 2020). Considering that health problems are increasingly complex (e.g., entangled with physical, mental, relational, or legal problems) and in the presence of adverse conditions (e.g., health and social inequities, stigma, or violence), these types of methods must be conducted more often, in consistence with syndemic and systemic approaches (Center for Behavioral Health Statistics and Quality, 2016; Hein & Small, 2006; Rouleau, 2020; Singer et al., 2017). In addition, it allows researchers to conduct dynamic research close to the communities and even have an impact on them, as participatory studies aim to do. The results that come from this combination are, in our opinion, unique and cannot be replicated in a usual data collection activity as suggested by Schiele et al. (2022). Moreover, other researchers have identified other interesting uses of the World Café, such as its combination with a Delphi method (Jolly et al., 2021) or in other formats, such as virtual mode (Banfield et al., 2021), which opens several promising avenues.
Although this combination is interesting, some challenges and limits persist. Including all the stakeholders was a major challenge. We found it difficult to identify the key actors beforehand, undertake complex ethical processes, and find a time slot that was convenient for all potential participants (J. C. Greene, 2000). We believe that the complexity of our issue made it more difficult to identify the stakeholders and that a pre-evaluation (for example, through a survey), could be beneficial (House & Howe, 1999). In both our project and the literature, having a sample aimed at a theoretical representativeness proved to be a promising strategy for adhering to this principle (House & Howe, 1999; Howe & Ashcraft, 2005). A selection bias is nonetheless possible because we used expert sampling for the recruitment, the participants were volunteers, and a small number of people represented an entire group. To reduce this bias, we diversified each of the sub-groups of participants, by organization (or sector), and by expertise. A bias in group representativeness was also possible due to the different number of participants in each stakeholder group in the World Café: some might have been better represented than others. This also applies to user representation and the challenges associated with their recruitment: recruiting users with whom no relationship had previously been made and ensuring that possible power issues were reduced, considering that few strategies were available in the literature for this type of approach. To minimize it, we implemented various strategies as shown in Table 2, and encouraged diversity within each discussion table. In addition, the planning and implementation of a World Café in terms of time, logistics (e.g. a neutral venue), and budget (e.g. coffee and snacks) are non-negligible, although this method is more accessible than other examples of deliberative democratic evaluation (House & Howe, 1999; Howe & Ashcraft, 2005). Finally, some challenges and limits were also meeting during the analysis. There is a lot of material to analyze (i.e., boards, recordings of tables and synthesis), so we have focused on the syntheses, i.e., the result of the consensus and the contents to be retained according to the participants. It was not possible to contrast perspectives across stakeholder groups, however, this was consistent with our goal of developing a common understanding and consensus. It was also difficult to integrate the qualitative data of the World Café to other qualitative data (i.e., individual interviews), because they did not have the same analytical perspective and objective. We used qualitative analysis software to integrate the thematic content analysis of different data collections during interpretation.
Lessons Learned
The combination of deliberative democratic evaluation and World Café was a good choice to bring together people from different backgrounds and reach our objective (i.e., needs evaluation with a qualitative design). In addition, it has had some impact on our communities, here on our participants: it improves participants’ knowledge and attitudes as well as the intersectoral collaboration. These effects are interesting in a context where there are complex problems, where there is often a need to exchange and few possibilities to do so, which is often the case in large urban centers (Doran et al., 2018; Feral-Pierssens et al., 2016; G. Moore et al., 2011).
While it is important to consider the location, food, coffee, and financial compensation, it is possible to organize a World Café at a relatively low cost. The budget of the current study was limited; we had snacks and coffee for everyone, but no lunch. Our team offered financial compensation for homeless participants, but not for professionals.
The integration of substance users experiencing homelessness into our World Café was pivotal because they are key actors concerned by the emergency response and several actors appreciated their integration. However, the recruitment of substance users experiencing homelessness was challenging. To improve their recruitment, we could have spent more time in the field. Regarding their integration into the activity, the strategies implemented by our team seem to have facilitated their integration (see Table 2). Pairing with a trusted person and the presence of “neutral” hosts seem good options when vulnerable people are integrated in a World Café. Nonetheless, further studies are needed to develop best practices for recruiting and integrating service users into collaborative activities such as World Café.
In the current study, we also modified the traditional World Café. We had thematic tables and we added, among other things, a prioritization at the end of each round and a synthesis by the hosts at the end of each session (see Figure 1). The prioritization at the end of each round seems to have facilitated consensus building, writing summaries (end of each round), and synthesis (end of each session). The addition of two syntheses seems to have facilitated knowledge transfer since it was limited during the activity (people did not speak during the group discussion, nor read boards during the breaks). These syntheses also help data analysis and results validation, since they provide access to prioritized results of the entire day in a short time (two minutes by tables) and allow participants to mention their disagreements about these results.
As a lesson and recommendation, future organizers should not be afraid to adapt the activity to their context and participants’ needs in order to achieve their goals. Among the possible adaptations, there is the activity duration, the number and duration of each round (more than four rounds is not recommended), questions for all tables or thematic tables, strategies for mixing participants with different backgrounds and for integrating service users. In our opinion, the essential elements are to be well prepared (e.g. schedule), identify a person who will lead the activity and manage the time, have trained hosts if it is part of the research, explain the World Café guidelines to the participants (World Café Community Foundation, 2015), and have at least one rotation. To facilitate the collection of data, recording tables and adding a synthesis of the exchanges at the end of each round are recommended. Finally, these methods are particularly interesting in contexts that are favorable to the use of research results, for example with people who are motivated to change their visions and practices (Contandriopoulos & Brousselle, 2012).
Conclusion
The World Café is a collaborative method that is concrete, accessible, and of short duration for researchers and evaluators in the qualitative and participatory fields. Together, the World Café and deliberative democratic evaluation enable all key actors to meet, comply with the participants’ values, and have a systemic view of an issue. This is particularly interesting in complex health problems, such as addiction and its numerous comorbidities. Our findings suggest that it can potentially improve intersectoral collaboration and practices, through results like reducing stigma. Our activity responds to a glaring need for different actors to meet and speak about a common concern in a neutral and pleasant environment when complex issues are in play. It is therefore an innovative combination that provides a new way of doing qualitative research, that is, through both the collection of rich qualitative data and the impact on communities.
This study also highlights the importance of including service users in the research and decision-making processes (House & Howe, 2003) and, even more so, vulnerable populations (e.g., active substance users) who are still rarely included in the research processes (Lazarus et al., 2014; Neale et al., 2017). Although these persons are vulnerable, it is possible to implement strategies to facilitate their integration, even in interdisciplinary activities. However, it remains important to more thoroughly document the strategies that could be used by research teams wishing to better integrate service users.
Finally, there is opportunity to go further in the study of these methods to favor success in the qualitative and participatory fields as well as to identify indicators for evaluating these methods.