Ask us! Adjusting experience‐based codesign to be responsive to people with intellectual disabilities, serious mental illness or older persons receiving support with independent living

Abstract Introduction Experience‐based codesign (EBCD) is a valuable tool for participatory quality improvement. However, the EBCD process needs to be adjusted to make it suitable for long‐term care. The focus of the improvement process needs to shift to the care relationship, as this is an important part of the quality of care in these settings. Furthermore, the EBCD process needs to be made more accessible to vulnerable populations. Methods Through a participatory research approach, EBCD was adjusted to long‐term care. The research was conducted in two care organisations: one supporting people with serious mental illness and intellectual disabilities in independent living and one providing homecare services for older persons. Results The participatory research resulted in the development of ‘Ask us!’—a method for critical reflective codesign. The research furthermore provided valuable lessons for participatory projects with vulnerable clients. A common problem with participatory research in long‐term care is ensuring the involvement of clients and informal carers. We report on various strategies developed to include experiences of a diverse set of services users, such as combining interviews with participant observation, photo‐voice and involving experts‐by‐experiences as co‐ethnographers. In close collaboration with an inclusive theatre company, these experiences were translated into 42 short videos on complex situations in the care relationship from the perspective of clients, professionals or informal carers. These videos instigate critical reflection and accelerate the participatory quality improvement process. Moreover, practical tools were developed to overcome barriers regarding the involvement of people with disabilities. These include the use of photo‐elicitation to enable participation of clients with disabilities in heterogeneous group discussions and involving experts‐by‐experience as proxies to share experiences of clients for whom participation in the ‘Ask us’ method remains inaccessible. Conclusion The result of a robust participatory process, ‘Ask us!’ is a promising method for participatory quality improvement in long‐term care. The research furthermore generated lessons for involving vulnerable populations in participatory research and codesign. Patient or Public Contribution Clients were involved as informants, sharing their experiences with the care relationship in interviews, photovoice and observations. They were also involved as consultants, helping to analyse input for the film scripts during data validation sessions.


| INTRODUCTION
Experience-based codesign (EBCD) is a promising method for involving clients, professionals and family members in improving the quality of care (see Box 1 for an overview of the EBCD process). [1][2][3][4][5] Applying this method in long-term care settings, such as supported living for people with intellectual disabilities (IDs), serious mental illness (SMI) or older persons however raises several design challenges.
First, a shift in focus of the improvement process to the quality of the care relationship is needed. EBCD often focuses on specific aspects or moments in service delivery that impact the experiences of service users. However, in long-term care, the care relationship is central to the quality of care. [7][8][9][10] The focus of EBCD in these settings should thus be on the care relationship. This care relationship is complex as it is often laden with value tensions. 11,12 Policy trends and care models, such as person-centred care, 13 recovery-oriented care, 14 active ageing 15 and rehabilitation 16 assign clients a more active role. In this context, values such as self-determination, independence and community participation have become prominent. Putting those values into practice is complex as these values can be specified in different ways and may conflict with other values such as preventing harm. Self-determination, for example, may mean respecting a client's decision to neglect standards of cleanliness or motivating them to clean their homes. Moreover, informal carers, clients and professionals can have different perspectives on these values, further complicating the care relationship. 12,17 To improve the quality of care in long-term care, the EBCD process thus needs to shift focus to the ethical tensions in the care relationship.
Second, the EBCD process needs to be made more accessible to service users in long-term care. Previous research has highlighted how the EBCD process can also produce vulnerabilities. 18,19 For instance, related to service users having to process rapid information flows or express themselves in heterogeneous group deliberations. 20 Moreover, the length of the EBCD process also poses barriers to involvement as motivation declines or drop-out occurs for other reasons. 4 Previous adaptations of EBCD have accelerated the process by using existing films from a national archive to trigger responses in group sessions with clients, thus skipping the initial phase of interviews and observations. Using existing trigger films can have the advantage of making the process less threatening or challenging. A possible trade-off could be staff engagement is aversively impacted. However, such an effect was not found in accelerated EBCD projects. 3,4,21,22 A key design challenge is therefore to BOX 1. EBCD EBCD (see Figure 1) is a process in which clients and professionals reflect on the quality of care and codesign improvements together. Informal carers are sometimes included in this process. 6 The method consists of several phases. The first phase is to collect care experiences through interviews and observations. Of the interviews conducted with professionals, informal carers and clients, those with clients are videotaped and edited into a trigger film showing the various 'touchpoints' where clients experienced the service in a way that impacted them emotionally. The use of films is an important feature of EBCD, allowing emotional engagement in the codesign process. The trigger film also helps to create a level playing field in which client experiences receive enough attention and engage others towards change. Second, clients, professionals and sometimes informal carers reflect on their experiences in peer-homogenous focus groups and identify areas for improvement. These group dialogues are facilitated through emotional mapping and, in the client group, by the client trigger film. Having peer-homogenous groups gives the participants a safe space to share their experiences with peers and find common ground. Third, the separate groups come together to watch the client trigger film, deliberate the issues raised in their group dialogues and set common priorities for quality improvement. Smaller codesign groups representing all stakeholders are assembled for each priority and meet several times to develop organisational practices. The EBCD process concludes with a celebration that highlights the successes of the quality improvements. [1][2][3] HEERINGS ET AL. | 2247 better enable clients to participate in the dialogue sessions and to accelerate the process while fostering engagement. However, even when engaging in relational strategies and offering creative means of participation, such as photovoice, to include services users well beyond 'the usual subjects', certain groups of service users still often are excluded from participation. This results in specific experiences not being taken up in the codesign process. 19,20,23 This raises the question how such otherwise excluded experiences can be included, in order to ensure a diverse range of experiences informs the quality improvement process.
In this paper, we report on a participatory project adjusting EBCD to long-term care settings. This resulted in the method 'Ask us!'. While a formal evaluation of 'Ask Us!' has not been conducted yet, the design process generated valuable lessons for participatory quality improvement with vulnerable populations, which are further explicated in this paper.

| Setting
We conducted participatory research in two care organisations situated in the same urban area in the Netherlands. The first was a community care organisation (CCO) providing supported housing (group homes and supported independent living) to people with IDs or SMI. The second provided homecare and other services to older persons (HO).

| Project group for designing the instrument and developing the trigger films
We set up project groups for both the CCO and HO organisations to coproduce the research and the method for participatory quality improvement (see Table 1 for an overview). The size of the two project groups and their level of involvement differed, with fewer meetings in HO and less involvement of policymakers. This reflected differences in the two organisations, with HO having a much smaller policy/management layer. The number of meetings in HO could be reduced as the data collection and analysis was less complex, covering only one client group (older persons) instead of two (people with ID or SMI). Moreover, the participatory research in HO was conducted after the CCO project had finished and several decisions, for instance regarding the production of the videos, were already set.
Members of HO were involved in the data analysis leading up to the content of the videos.  Table 2 for an overview).
All three CCO and HO teams included professionals whose experiences were collected through interviews, participant observation (CCO) and shadowing (HO). Professionals were further involved through group discussions on the analysis of the interviews and observations serving as content for the trigger films. CCO professionals consisted of social care workers, while HO professionals included a community nurse, registered nurses, nurses in training and aides. The professionals of each team were asked to participate after introducing the research during a team meeting (CCO) or were asked by the nurse leading the team (HO). Reasons for declining included leaving the care team or being too busy.
Client recruitment proved more complex. Two things are specifically worth mentioning. First, it was difficult to recruit a diversity of clients with SMI because the affliction itself prevented them from participating. For example, clients refused interviews because they did not want to talk to 'strangers' or said their 'voices' did not allow it.
Moreover, professionals who asked clients about being interviewed reported negative responses; in some cases, their already fragile care relationship was impacted. In response, we developed relational strategies to involve clients, for example by getting acquainted with them during coffee moments where they socialized. 20,24 The clients who did participate were not, however, a representative in terms of openness to contact. We, therefore, relied more on participant observation and informal conversation during these observations and developed further strategies to include the experiences of people unable to be interviewed. This included interviewing experts-byexperience about situations they had witnessed during their peersupport work involving care relationships in the context of promoting self-determination, independence and community participation.
For HO, we excluded people with dementia from interviews because they had difficulty recalling experiences with care professionals. We were, unfortunately, unable to include a sufficient diversity of clients from minority backgrounds, resulting in an underrepresentation of these groups. Partly these issues were resolved by including data from shadowing professionals.
It also proved difficult to interview informal carers in CCO.
Service users acted as gatekeepers for contacting informal carers and their relationships were often complex. 20 Moreover, of the limited number of informal carers that we were able to contact, several declined for various reasons. We resolved this problem by conducting additional interviews with family support workers, who elaborated on their own experiences with the services and that of other family members they supported. It was less relevant in HO to include informal carers because the older persons often managed their own care (see Table 2 for an overview).

| Adjustments to the dialogue sessions
We organized a workshop to modify the method used to foster dialogue in both the peer-homogenous groups and the joint EBCD event. The aim was to mitigate vulnerabilities by making the method more inclusive for clients who have difficulty processing information and speaking up in deliberative sessions and to shift the focus to the complex care relationship (see Table 1). The input for the workshop came from the researcher and the member of the patient advocacy organisation, who suggested ways of adapting the various EBCD phases. Their suggestions were based on interviews (n = 2) with  Interviews  12  12  4  8  8  3  9  13  5 Photovoice Group consultation on analysis (no. of participants)

| RESULTS
This participatory research resulted in the 'Ask us!' method. It consists of the 42 trigger films and a process for the various group and codesign sessions (see Figure 2). We first elaborate on the participatory process of developing the trigger films and reflect on lessons learned to include vulnerable groups in participatory research. We then describe the developed method for participatory codesign: 'Ask us' and explicate the lessons for participatory researchers aiming to involve vulnerable populations.

| Trigger movies
The 42 trigger films are based on client, professional and informal carer experiences and are meant to accelerate the EBCD process and focus it on the complex long-term care relationship. The films' length ranges between 1 and 3 minutes. The spoken language is Dutch; the films are made accessible to an international audience through English subtitling.
For an overview we refer to Table S1, which lists the main themes for each film; for the full content, please visit www.eur.nl/en/eshpm/ research/if-you-would-ask-us. The process we undertook to produce these trigger films differed from previous accelerated EBCD strategies. 21,22 Because many clients with SMI chose not to be interviewed on camera (similar to previous EBCD projects with this client group), 23 we collaborated with theatre artists and developed scripts based on different data collection methods. As this also fitted our aim to accelerate the method, this format was used to produce films reflecting experiences of services users with SMI or ID, older persons and experiences of professionals and informal carers as well.
After discussing various possible formats, including documentary, digital storytelling and animation, the project group decided to give the films a realistic feel. The CCO members suggested an inclusive theatre company as a possible partner, as many of its actors had disabilities and received care from the CCO. They could contribute their own experiences to the creative process, making the films even more 'real'. The theatre company's director suggested a mis-en-scene of close-up monologues, to which the project team agreed. The format was therefore already set when we started the HO participatory process. The rest of the process ran similarly in CCO and HO.
Each show a complex situation from every perspective but also possible differences between clients, informal carers and professionals and, consequently, the need for deliberation, which is part of 'Ask us!'.
The collaboration process involved translating the data of service users' experiences to the actual films in a series of steps. 25 Throughout, we adhered to both a qualitative research logic and an artistic logic to ensure that the trigger films were both grounded in empirical research and able to engage audiences in reflection on their own experiences. Below, we describe the participatory process and translations steps leading up to the films.
The teams and the project group in each organisation were involved in developing content for the trigger films. We followed the logic of qualitative research by using ethnographic methods to collect data on the complex care relationship (see Table 2 for an overview).
We used the same semi-structured interviews for clients, professionals and informal carers, opening with the question 'What is F I G U R E 2 Ask us! redesigned experiencebased codesign important to you in the care relationship you are involved in?' Each named value prompted narrative exploration of situations in which this value was appropriately practised or proved difficult. When values such as self-determination, an independent lifestyle or community participation were not named, the researcher provided prompts in the second half of the interview. Interviews were audio-recorded and transcribed verbatim. The photovoice method included one session to explain how to operate the camera. Participants were then asked to take photos of what they did or did not like about where they lived. After 2 weeks, the photos were examined in interviews for themes related to the complex care relationship, 26 with notes taken during the interviews being elaborated shortly after.
Different strategies were developed to include the experience of service users who chose not to partake in interviews or photovoice, or for whom these data collection methods were inaccessible. 27 A first strategy included participant observation in the group homes part of CCO or shadowing of professionals in HO. This allowed us to observe the 'enacted appreciations' of clients, or their reactions to care practices without them having to vocalize these. 28 Moreover, participant observation and shadowing allowed for many informal conversations with both clients and carers about care moments, shortly after they unfolded. These informal conversations were often much more accessible to clients compared to formal interviews and proved very valuable in collecting their experiences with care.
As a second strategy, we included the experiences of vulnerable clients in data collection by conducting interviews with experts-byexperience in their role as peer support workers. In the CCO organisation, experts-by-experience were part of care teams as peer support workers. As a result, they had witnessed many care moments and had many conversations with clients about their experiences with care. Similarly, the family-experts-by-experience could share many experiences of carers, which complemented the interviews that were conducted. The interviews with these (family)-experts-byexperiences followed the same format as the other interviews, although the focus was not on their own experiences but on those they had witnessed as part of their peer-support work. These strategies allowed for the experiences of clients who chose not to partake in interviews or photovoice sessions to become part of the input for the trigger films. This is important as these clients seemed often more vulnerable and may have different experiences compared to clients who are willing and able to be interviewed.
To translate the different data of individual stories to common themes related to the complex care relationship, the first author conducted a thematic analysis. The analysis involved inductive coding of the data using Atlas-ti software, identifying similarities and differences, and using axial coding to develop the themes. To refine this analysis, the researcher edited quotes per subtheme into comprehensible narratives and discussed them in various sessions with the project group and with clients, professionals and family support workers (see Tables 1 and 2).
The narratives were then adjusted based on these discussions.

| 'Ask Us!' dialogue sessions
We now turn to the details of the different dialogue session part of the participatory quality improvement method 'Ask us!' and explicate the lessons learned, which are of interest to a wider audience of participatory researchers aiming to involve vulnerable populations.
The first phase of the method focuses on collecting care relationship experiences and reflecting critically on the tensions between values associated with clients' playing an active role. In this phase, clients, professionals and informal carers engage in separate group dialogues in which they share experiences and reflect critically after viewing selected trigger films specific to each peer-homogenous group. The main issues for making the EBCD more accessible to vulnerable populations were first to keep the input manageable and prevent information overload; second, to use visuals over textual information.
For the purpose of managing information overload, we reduced the time of the trigger film to a selection of 3-5 films, lasting no more than ten minutes rather than the usual 30-min trigger film in EBCD. The selection is made in consultation with an expert-by-experience and professional familiar with the team engaged in the 'Ask Us!' method.
After watching the selected videos, the participants reflect on their own experiences triggered by the films. We modified the EBCD emotional mapping method to make it more accessible to a diverse group of clients and to focus on the complex care relationship. For this, we developed various visual materials. Clients are asked whether watching the film evoked a recent experience in their care that still moves them emotionally or that they think about at times. They are asked to choose a worksheet belonging to one of the films that triggered this memory (see Another key aspect of participatory research with vulnerable populations, which emerged during the process of redesigning the various dialogue sessions, is making sure their inputs are responded to in ways that encourage further participation even when their input is not further incorporated within the participtory quality improvement process or reasearch. This requires designing additional processes in which these concerns can be addressed. In EBCD and in 'Ask Us!', this for instance is important in the process of selecting themes for quality improvement.
This is a delicate process as it can be overwhelming for some clients to voice a concern or area for improvement in the peer-homogenous group and they may feel hurt or demotivated if their concern is not selected. An additional procedure was therefore designed in which experts-by-experience coach clients to address these concerns in an appropriate setting, for instance in their individual care relationship, in the client council or in a team meeting of professionals. This procedure is an addition to the more generic support available for clients before, during and after the sessions. 23,29 An practices so not to reinvent the wheel. In those cases the co-design process consists of adjusting these best practices to the specific needs and context of the clients, professionals and informal carers. A key point is these best practices should be offered to inspire participants, while making sure participants remain ownership over the co-design process.

| DISCUSSION
In this participatory research project, clients, professionals, informal carers, experts-by-experience, family support workers, researchers and policymakers collaborated on developing a method for participatory quality improvement of the complex care relationship in F I G U R E 4 Worksheet after watching trigger films long-term care focusing on self-determination, an independent lifestyle and community participation.
EBCD was a valuable source because it involves a process whereby client, professional and informal carer engage in reflection, deliberation and codesign, but it needed to be redesigned for long-term care by: 1. Accelerating the process and focusing on the complex care relationship: 42 short trigger films were developed addressing the dilemmas and burdens experienced by clients, informal carers and professionals in the care relationship in supported housing for people with IDs or SMI and homecare services for older persons.
2. Reducing existing and preventing new vulnerabilities: the various group dialogues central to EBCD were modified to make the method more accessible for clients with SMI or ID or for older persons.
The resulting method, 'Ask us!', brings together critical reflection, deliberation and codesign in a comprehensive process that has the potential to improve services.
'Ask us!' allows organisations to involve professionals in critical reflection, helping them to better handle complex situations, and to engage clients, professionals and informal carers in redesigning their services so as to empower service users. It should be noted, however, there has yet to be a formal evaluation, which will be an important next step in the refinement of this method for participatory quality improvement. Collaborating care organisations and audiences at presentations and film viewings-including client representatives and professionals-however, have responded positively to the method.
Both clients and professionals expressed that the content of the videos was recognisable to them and the videos could be useful in fostering conversation on quality of care. 30 The participatory process through which this method was developed yields several lessons to researchers aiming to involve vulnerable populations in participatory research. Previous research on involving vulnerable populations has highlighted how ensuring participation requires creating relationships with service users before engagement, including creative methods, to enable them to share experiences and involving service users at their own pace. 19,20,23,31 These strategies however proved insufficient in our research to ensure diversity in the experiences included. While it remains preferable both from a methodological and ethical standpoint to ensure direct involvement of vulnerable groups, alternative strategies might be needed. When methods of direct involvement such as interviews, focus groups or creative methods like photovoice prove inaccessible to certain clients or family members, their important experiences are excluded from the research or quality improvement process. To overcome this barrier, such methods of direct involvement can be supplemented with other methods, such as observations of enacted appreciations 28 ; informal conversations during participant observations and involvement of (family) experts-byexperience as co-ethnographers.
Another important lesson is to develop a process through which the input of clients that is not part of the further codesign process is still taken up and responded to within the care organisation. This is an important addition to current research on involving vulnerable populations in codesign processes in which the need of having counselling available during and after sessions is often highlighted. 23,31 While this is an important part of caring for participants' well-being, responding to their concerns regarding the quality of care that were not selected in the design process is also needed.
Lastly, the traditional EBCD process already uses creative means to lift some of the power imbalances in mixed-group sessions: the trigger films showing clients' experiences. We added a photoelicitation method to further enable clients to take a stance and voice their concerns and professionals and informal carers to hear these.
These lessons emerged in our process of redesigning EBCD to fit long-term care settings. They are valuable as well to other researchers in involving vulnerable populations in participatory research or codesign projects.