Patient and public involvement facilitators: Could they be the key to the NHS quality improvement agenda?

Abstract Objective Research into patient and public involvement (PPI) has not examined in detail patient and public involvement facilitators’ (PPIFs) roles and activities. This study analysed PPIFs’ roles using qualitative data gathered from three different UK health‐care organizations. Design Thematic analysis was used to examine cross‐sectional data collected using a mixed‐methods approach from three organizations: a mental health trust, a community health social enterprise and an acute hospital trust. The data set comprised of 27 interviews and 48 observations. Findings Patient and public involvement facilitators roles included the leadership and management of PPI interventions, developing health‐care practices and influencing quality improvements (QI). They usually occupied middle‐management grades but their PPIF role involved working in isolation or in small teams. They reported facilitating the development and maintenance of relationships between patients and the public, and health‐care professionals and service managers. These roles sometimes required them to use conflict resolution skills and involved considerable emotional labour. Integrating information from PPI into service improvement processes was reported to be a challenge for these individuals. Conclusions Patient and public involvement facilitators capture and hold information that can be used in service improvement. However, they work with limited resources and support. Health‐care organizations need to offer more practical support to PPIFs in their efforts to improve care quality, particularly by making their role integral to developing QI strategies.


| Setting
Three separate organizations were studied in this cross-sectional thematic analysis research: a mental health trust (MHT), a social enterprise (SE) providing community care and an acute hospital foundation trust (AHFT). This approach was used to capture both the common and context-specific elements of PPIFs' roles. Each organization differed in terms of size, budget and remit. MHT employed around 6000 staff and provided a range of inpatient and community mental health services; SE employed around 2000 staff and provided community health services; and AHFT employed around 8000 staff and provided acute hospital services across two sites.

| Design and data collection
A mixed-method approach was used. Some of the PPI activities used were identified through conversations between the lead author and the organization's named PPI lead (identified from a Strategic Health Authority's website). PPI activities were then observed by the lead author during visits over a period of 1 year between June 2013 and July 2014, with detailed observation notes collated.
Patient and public involvement activities observed included focus groups, project groups and governance meetings (see Table 1). These activities were classified as 'PPI activities' because at least one layperson or patient representative was involved.
Twenty-seven interviews were conducted with health-care managers, PPIFs, laypeople/patient representatives and health professionals (see Table 2).

| Data analysis
Thematic analysis was conducted using an iterative process of coding to identify key themes. 53 This technique allowed for the identification of patterns both within and across sources of data. No set of pre-defined themes was applied to the data. A variety of factors were identified during a literature review, but the analysis was inductive (see Figure 1). This allowed for more freedom to identify novel and nuanced findings. 54

| What is a PPIF?
Our study showed that a PPIF is principally responsible for acting as a link between health-care organizations and patients and the public. 76 They facilitate and support patient and public access to PPI activities. This ranges from providing them opportunities for giving their views through surveys and formal complaints, to securing PPI in service improvement projects, and facilitating attendance at governance meetings. These activities are widely referenced in the literature, 5,24,47,51,[77][78][79][80][81][82] but with little reference to the role that PPIFs played in supporting these activities.
In interviews, many PPIFs reported that they were responsible for providing reports to senior management through governance meetings and that they supported service managers and clinicians in involving patients in service improvement projects. They interacted with a diverse range of stakeholders: current and former patients, the general public, charities, interest groups, health pro-

| Formal position in the organization
According to interview data and initial conversations with PPIFs, they were usually based within central corporate functions, reporting to senior managers and the Board (directly or via their line manager).
Many were members of teams of up to six people, and in AHFT, there was no specific team at all, with PPI work spread across multiple corporate functions (such as public membership office and formal complaints). As identified through interviews, these roles did not appear to confer any formal decision-making powers. However, PPIFs' position in the hierarchy (formal power 83 ) indicated that they could advise service managers and clinicians (expert power 83 ) about necessary service improvements following feedback from patients. Sometimes, this advice was not acknowledged or utilized, rendering the PPI activity an ineffective use of time and resources, as illustrated in the quote below.

| Context-specific differences
A range of context-specific differences appeared to have an influence over PPIFs' roles and responsibilities, and the way in which they perceived support for PPI in their organization.

| Practical delivery of services
Both MHT and SE had services spread across multiple sites including in health centres and small community hospitals, meaning they had a physical presence in the community. In contrast, AHFT was primarily based on one hospital site. PPI activities often involved laypeople coming to the hospital, rather than PPIFs going out to the community. The physical space within which these activities took place was, therefore, very different. PPIFs in MHT and SE went out to communities rather than requiring the public to come to them.
This difference is highlighted in the quote below:

| Cultural differences
Staff in both MHT and AHFT expressed a view that mental health providers are better at PPI and cited this as being as a result of an organizational culture that sits more comfortably with the principles of PPI. In acute hospital care, the patient is often seen quickly to treat a physical problem and then sent home as soon as possible, limiting the time the patient spends with staff. In a mental health setting, the relationship between health professionals and their patients is generally much longer. Many mental health patients will be accessing those services for the rest of their lives. As such, greater partnership working occurs at the clinician-patient level, arguably making PPI a more natural approach for mental health staff than acute hospital staff. This is demonstrated in the two quotes below.

| Organizational strategy
All three organizations had small teams responsible for PPI, but a key difference was noted in AHFT which had no specific team or budget allocated to PPI. Both MHT and SE had a designated PPI lead, whereas there was no overall lead at AHFT. Both MHT and SE also had a PPI strategy document. As such, there seemed to be less coordination of PPI activities in AHFT. The primary focus for AHFT appeared to be managing formal complaints and meeting targets (ie 'transactional' functions). A lack of strategic direction was cited as problematic by some staff interviewed at AHFT (example below).
Oh, and the other thing was we never had a strategy [

| Gatekeeper
In this role, they were determining laypersons' access to PPI activities. They appeared to hold informal power pertaining to the nature of the access laypeople were granted. A large part of this role was in building and maintaining relationships between stakeholder groups.
PPIFs were usually the primary contact for patient and public representatives and groups. As such, the access they have is at least initially determined by the PPIFs. Tenbensel 52

| Mediator
Patient and public involvement facilitators frequently facilitated conversations between staff and patients/the public in arenas such as governance meetings or committees. These interactions often led to heated debate, as observed by the lead author and reported by PPIFs in interviews. As such, PPIFs were central in mediating that conflict and ensuring the various stakeholders were able to make a fair contribution to conversations, as illustrated in the quote below.

| Tensions that limited PPIFs' roles
Data analysis identified three tensions influencing PPI processes and the relationships between major stakeholders: top-down vs bottomup management; individual vs collective needs; and patient experience vs patient involvement.

| Top-down vs bottom-up management
Many of those interviewed indicated that they believe PPI is most effective when it involves front-line staff and/or service managers (ie those in organizations who are responsible for delivering services). This is similar to findings that PPI works best at a 'grass roots' level. 27,64,85,86 All PPIFs in our study suggested that greater front-line staff engagement was needed in PPI processes. PPIFs felt they had to lead on all PPI work because they believed the workload and its complexity placed too many burdens on front-line staff.  88 and locally (bottom-up or 'decentralization'). 88 PPIFs reported being frustrated by the difficulties they encountered in their efforts to engage front-line staff. Their view was that PPI would work best through a participatory bottom-up approach but that this was difficult to initiate and maintain. Decentralization can, in theory, motivate front-line staff to take on these responsibilities by giving them greater autonomy, 88 and this appeared to be the argument presented by the PPIFs in this study.

| Individual vs collective needs
There was also evidence of tensions between individual stakeholders' needs and the perceived collective needs of the organization.
There were some occasions where PPIFs had to curtail discussions that they identified as being too personal. Data analysis suggested that personal stories were valued but had the potential to detract from the core objectives of PPI, as indicated in the two quotes below.

| Patient experience vs patient involvement
The interchangeable use of the terms 'patient experience' and 'patient involvement' created confusion for stakeholders about the purpose of PPI. In Arnstein's ladder of participation, 90 PPI methods are defined according to levels of involvement, with co-design and codelivery of services at the top of the ladder and providing information and tokenistic involvement at the bottom. PPIFs expressed a desire to use a variety of methods but with a preference for those nearer the top of the ladder. They felt that surveys could not accurately be labelled as 'involvement' but rather a measure of patient experience.
In AHFT, some PPIFs suggested that there was pressure from senior management to report against KPIs and statistics by conducting surveys and managing complaints, rather than what they deemed to be 'proper' involvement (see quotes below). Some PPIFs reported that front-line staff would sometimes conduct surveys and describe this as 'patient involvement', therefore 'ticking the box' on their obligation to involve patients in service changes. PPIFs had a more nuanced interpretation of PPI and appeared to value different PPI methods in hierarchical terms, much like Arnstein's ladder of participation. 90 They suggested surveys would be at the bottom of the ladder and would prefer to aim for more involvement of patients and the public in decision-making processes.
All we do is a patient satisfaction, and in fact, we're going more that way than proper engagement because I've seen it move towards numbers…so we're doing the lowlevel engagement on "are we doing a good service?" We're not even doing the level above that -the "what would make a better service?"…not anything more engaging.

AHFT: PPIF (ID C003)
Patient and public involvement facilitators also commented that senior management and front-line staff did not always fully comprehend the breadth and depth of PPI activities and the potential benefits (see quote below). Some indicated that this was because of a lack of deep understanding of PPI theory and practice. Moreover, some PPIFs felt that PPI knowledge was tacit and thus difficult to transfer.

Sometimes I feel that they [the Board] don't always quite understand what's actually involved in it…I hear them
talk about the patient experience programme, they'll get the name wrong and you'll think "Ah, you don't really understand".

| How these tensions impact on PPIFs
There was evidence that the tensions between top-down and bottom-up management were influencing decision-making processes and power relationships. PPIFs regularly needed to simultaneously obtain the support of senior management for strategic organizational change, as well as the support of service managers and front-line staff at the local level. The frequently reported and observed tension between individual and collective needs was also a considerable challenge for PPIFs to balance when it came to recommending service improvements. Senior managers needed to manage both individual and community-wide needs simultaneously. Whenever an individual's needs conflicted with wider organizational priorities, the latter prevailed.
The influence of these tensions is illustrated in Figure 2. PPIFs were often reduced to simply sharing information with senior managers and service managers in the hope that it would lead to service improvements, but not having the power themselves to enact changes.

| The mirrored characteristics of laypeople and PPIFs
There were similarities between the reported motivations, beliefs and characteristics of PPIFs and those of the laypeople involved in PPI. For example, all laypeople had experienced or witnessed poor care, and this reportedly motivated them to take part in the PPI process. During interviews, they often claimed to be motivated by wanting to give other people a voice and improve services as a result. These experiences and motivations seemed to be shared by the  And it has never been a priority, in any of the NHS organizaƟons that I have worked for; even when there's been Board-level commitment. The challenge comes, always, no matter how commiƩed the Boards are, to what if the community want to make a decision that you don't agree with? And it always falls down at that point. Or they'll even agree it, commit to it, and then something else will change in the system. Because it changes so much because we have so much involvement from the SHA, naƟonally or whatever. We'll commit to a parƟcular process; we'll even go with communiƟes someƟmes on a parƟcular decision; and then it will get pulled.
But it would hardly ever get to that stage anyway because we'd already made the decision on what we were consulƟng on. And the whole Ɵme; and I've been in terrible commissioning and service development things; where I'm just thinking "the fact that you're arguing back with the public that they're making the wrong decision means that you've already made your decision so why are you even engaging with them?" What you're talking about is a PR exercise, which is fine, but what communiƟes really don't like is being set up.
PPI by Milewa et al 80 If health professionals view PPIFs as using PPI to performance manage them, this may create further conflict. In this study, PPIFs reported that some front-line staff and service managers resisted being involved in PPI initiatives. One of the reasons for this resistance could be their perceptions of PPIFs' motivations and allegiances with patients and the public.
Furthermore, adding to these conflicts of legitimate, expert and referent power 83 was the prevalence of an 'in-group'. There was a limited pool of people who PPIFs invited to join projects, groups and committees. This resulted in the same group of people being included in multiple activities. The finding supports research that has noted this 'in-group' phenomenon in PPI. 55,59,86,91 This potentially could exacerbate power conflicts with health professionals and managers who may argue that this type of group is unrepresentative, therefore justifying questioning PPIFs' recommendations.
Laypeople defined their own role as a result of being motivated by seeing or experiencing poor care. This confirms findings from existing literature on layperson motivations and identities. 32,49,63,68,79,[92][93][94] In addition, the PPIFs identified themselves as advocates more than employees. This study indicates that often the motivations and identity perspectives of laypeople were mirrored by PPIFs.
These conflicts and the complexity of the PPIFs' roles appeared to be very challenging and involved considerable emotional labour and cognitive workload, as they often worked in isolation, as reported by Staniszewska. 12 One of the major recommendations for future practice is to improve support for PPIFs in their daily work.
Patient and public involvement facilitators indicated that their role was constrained by its level within the organization, in that they neither had the power to make decisions nor did they have the power to enact service changes. It was only within their power to try to influence service changes through sharing intelligence. Figure 2 shows the groups of stakeholders PPIFs are required to work with and how information is exchanged or brokered by PPIFs. It highlights that actual power for decision making (and therefore service changes) rests with service managers and senior clinical staff, and/ or senior management.

| Implications for practice
The main recommendations from this study focus on supporting PPIFs to balance the highlighted tensions. As PPIFs are key in all PPI processes, from being a gatekeeper for patients and the public, to facilitating stakeholders' conversations, to making recommendations for service improvements, it appears important to provide them with adequate instrumental and emotional support. These individuals may begin to feel ineffective and under-valued if these are not provided.
As there was evidence that PPI was not embedded in decision-making processes, senior managers may need to be clearer about the value of PPI in their organization. If the purpose is not clear, or PPI conflicts with wider organizational priorities, it is unlikely to improve services. In order to prevent PPI conflicting with organizational strategy, we suggest PPIFs should be involved in strategic decision making. Furthermore, PPI activities should be aligned with said priorities so that it directly feeds into organizational decision making, for example by involving patients in developing cost-saving initiatives when under financial pressure.
Patient and public involvement facilitators should also be supported in more practical and tangible ways, such as through training and mentoring, as well as through ensuring access to adequate financial and physical resources. This has been suggested elsewhere in the context of PPI in research. 95 This could facilitate greater variance of PPI activities, including more targeted work with hard-to-reach groups, and the ability to hold activities in multiple venues.

| Study limitations and future research
A limited number of PPIFs were interviewed from three health-care organizations. As such, we cannot be confident that the perspectives of the PPIFs presented here would be found in other organizations. As Barnes et al 49 suggested, PPIFs' motivations, beliefs and perspectives should be given equal consideration with those of the patients and public. We propose that further research is carried out which focuses on PPIFs' perspectives. This could be done through surveys, in-depth interviews and/or focus groups, in order to gain both rich insight and patterns of personal engagement.
As there were conflicts between PPIFs, front-line clinicians and service managers, it may also be pertinent to assess the perspectives of staff not closely associated with PPI work. This would give greater insight into why PPIFs encounter barriers with these stakeholders and provide further recommendations for practice in the future.
Another potential limitation of the study may have been that,

| CON CLUS I ON: WHAT C AN PPIFS CONTRIBUTE TO Q I?
This study has demonstrated how important PPIFs are in PPI processes. With some exceptions, 12,49,50,52 there is a considerable lack of acknowledgement and appreciation of the role of PPIFs in the literature. Our results indicate that PPIFs can feel isolated in their roles when attempting to influence organizational cultures and conduct QI.
There was evidence that they felt that senior managers did not appreciate the benefits of PPI in general but also their specific role in PPI processes. PPIFs' sense of isolation was further exacerbated by this lack of shared understanding and appreciation of PPI with others.
Our study also identified the difficulties that PPIFs face in balancing tensions that impact on their work. This balancing role further demonstrates how PPIFs are potentially an underutilized resource in health-care organizations. These three tensions (topdown vs bottom-up management, individual vs collective needs and patient experience vs patient involvement) have implications for the QI agenda. Therefore, it is argued that health-care organizations need to offer more support to PPIFs to ensure that their valuable contribution can be realized.
Finally, as QI was an explicitly cited driver for PPI, one would expect PPI interventions to influence service developments. The tensions underlying PPI work can inhibit PPI directly influencing QI. In addition, the power conflicts between stakeholders and PPIFs limit the ability of PPIFs to influence QI, as decision making primarily lies within the remit of managers and clinical teams. Incorporating the role of PPIFs into decision-making processes may help them to better influence QI.

ACK N OWLED G EM ENTS
We would like to thank all those who participated in this research.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.