Exploring the sociocultural contexts in which healthcare staff respond to and use online patient feedback in practice: In-depth case studies of three NHS Trusts

Abstract

Objectives

Patients are increasingly reporting about their healthcare experiences online and NHS Trusts are adopting different approaches to responding. However, the sociocultural contexts underpinning these organisational approaches remain unclear. Therefore, we aimed to explore the sociocultural contexts underpinning three organisations who adopted different approaches to responding to online patient feedback.

Methods

Recruitment of three NHS Trusts was theoretically guided, and determined based on their different approaches to responding to online patient feedback (a nonresponding organisation, a generic responding organisation and an organisation providing transparent, conversational responses). Ethnographic methods were used during a year of fieldwork involving staff interviews, observations of practice and documentary analysis. Three in-depth case studies are presented.

Findings

The first organisation did not respond to or use online patient feedback as staff were busy firefighting volumes of concerns received in other ways. The second organisation adopted a generic responding style due to resource constraints, fears of public engagement and focus on resolving known issues raised via more traditional feedback sources. The final organisation provided transparent, conversational responses to patients online and described a 10-year journey enabling their desired culture to be embedded.

Conclusions

We identified a range of barriers facing organisations who ignore or provide generic responses to patient feedback online. We also demonstrated the sociocultural context in which online interactions between staff and patients can be embraced to inform improvement. However, this represented a slow and difficult organisational journey. Further research is needed to better establish how organisations can recognise and overcome barriers to engaging with online patient feedback, and at pace.

Introduction

Health service policy and guidance has seen a shift, increasingly valuing patient involvement and hinting at a cultural change towards true ‘patient-centred’ care,¹ despite difficulties in defining what this means in practice. The wide ranging patient safety issues highlighted within the Francis report,² alongside other instances of poor care overlooking patient's early warning signals,^3,4 helped to make listening to patients a national priority in UK hospitals. Patient experience has also been increasingly recognised as one of the three pillars of healthcare quality, alongside clinical effectiveness and patient safety.⁵ The importance of patient experience can be seen via the lens of Martin's two key rationales for patient involvement.⁶ Namely, the democratic rationale, suggesting that patient involvement is an ethical and moral obligation of health services, and the technocratic rationale, recognising that patients provide instrumental insights into healthcare experiences.^7,8

The growing importance of patient involvement in healthcare has been reflected by a national drive to increasingly ask patients to share their experiences via a variety of tools in the UK.⁹ These can be organised into two overarching categories dependent on the extent they are initiated by the health service. Firstly, NHS-initiated feedback, which includes sources that are actively sought by, received by and interpreted by NHS services, such as formal complaints, patient advice and liaison services (PALS) and nationally distributed surveys. These feedback opportunities provide patients, their families and carers, with a point of contact, and have been heavily researched, which highlights a range of challenges for staff looking to engage with the information meaningfully.^10–14 However, a second overarching category of patient feedback tools has had far less research interest to date. Non-NHS-initiated feedback refers to unsolicited sources, which are initiated externally to the NHS by patients, carers, families, third-party organisations and/or researchers, and are sometimes, but not always, received and interpreted by NHS services. One key source of unsolicited patient feedback is Care Opinion (www.careopinion.org.uk), a national not-for-profit online platform where patients can report about their healthcare experiences anonymously using free-text narratives, and staff from provider organisations can respond. Meaningfully engaging with such sources align with the Department of Health and Social Care's vision for the future of technology in supporting an open culture where feedback is welcomed¹⁵ and supports calls from a recent themed review by the National Institute for Health Research to better engage with online tools, as their usage is expected to grow, and at an increased pace.¹⁶ Research suggests that patients are increasingly providing feedback in an unsolicited manner online, as well as reading feedback from others in this way, yet the information often remains underused by staff.¹⁷ This may be due to the unique challenges online feedback raises regarding issues such as: power imbalance, anonymity,¹⁸ selection bias and accessibility.^19–21 Additionally, research has raised concerns of gaming and misinterpretation,²² requirements of skills and resources for data analysis, a lack of responsibility felt by individuals and teams,²³ and inadequate integration with current systems and practice.²⁴

Recent research by the study authors presented a typology of responses to online patient feedback provided via Care Opinion.²⁵ Five key responses types were identified: non-responses, generic responses, appreciative responses, offline responses and transparent, conversational responses. Response types varied based on the extent they were specific and personal, how much responders embraced the transparent nature of public discussion, and whether or not responders suggested that the feedback had led to learning or impacted subsequent care delivery. Response types also appeared to have systemic influences, as variation in responding was at an organisational, rather than individual responder, level. In light of the background evidence, the objective of this study was to use the response typology²⁵ as a foundation from which to explore the sociocultural contexts in which healthcare staff use online patient feedback in practice to inform improvement via an in-depth case study approach. Specifically, the following research questions were explored within three organisations adopting different ways of responding to online patient feedback:

• - How do staff engage with online patient feedback?

• - How does this compare with how other sources of patient feedback are engaged with?

• - What are the sociocultural contexts underpinning these engagements?

Methods

The study was drafted in accordance with the consolidated criteria for reporting qualitative research (COREQ).²⁶ Authors have backgrounds in psychology, sociology, quality and safety, improvement science and applied health services research. A reflexive diary was kept to ensure interpretations were warranted by the data.

Patient and public involvement and engagement

The lay leader of the Patient Involvement in Patient Safety theme of work at the Yorkshire and Humber Patient Safety Translational Research Centre, and representatives from Care Opinion including the Chief Executive, collaborated via regular meetings to help to guide the design, development and conduct of this research.

Case selection

Ramsey et al.'s typology of staff responses to online patient feedback²⁵ was used to theoretically guide the purposeful recruitment of three local NHS Trusts, ensuring maximum variation sampling. Cases were selected based on their approaches to responding to online feedback via Care Opinion (www.careopinion.org.uk). Cases comprise a non-responding organisation (site A), an organisation that provided generic responses to all online feedback (site B) and an organisation that engaged in transparent conversation with patients online (site C). Further contextual detail of each site is provided in Appendix A. Staff within the identified sites were initially approached via email. Upon registering interest, snowballing and opportunistic sampling methods were used to recruit staff members who had, or may be expected to have, an interest in online patient feedback within each organisation.

Ethnographic methods

A focused ethnographic approach²⁷ was adopted during a year of fieldwork from March 2019 to March 2020, using an iterative process of data collection, analysis and reflection to gain insights into the wider sociocultural contexts of each site.²⁸ Healthcare has been identified as a particularly valuable environment to adopt such methods as a pragmatic and efficient way of capturing cultural perspectives, often entangled in complex relationships, organisational dynamics and multifaceted systems, which may go otherwise unnoticed via traditional research methods.^29,30

Multiple data sources were sought to provide complementary understandings of what was being done and said internally, and what was being published online.³¹ This involved in-depth semi-structured interviewing with 25 staff (7 interviews at site A, 8 interviews at site B, 10 interviews at site C). 17 were conducted face-to-face at the study site, 7 were conducted via telephone and 1 participant responded to questions via email due to limited availability. Interviewees ranged in seniority and role including volunteers, frontline staff, quality improvement, patient experience, and senior management representatives. Interview duration ranged from 23 to 74 minutes (average 41 minutes) and each were guided by an iteratively developed schedule, directing conversation towards the broad topics of (i) the role of online patient feedback within their role, service and organisation; (ii) the perceived value of online patient feedback and how that compared to other feedback sources; (iii) their approach to responding to online patient feedback; (iv) the perceived strengths, limitations and their satisfaction with the approach. Conversations were also flexible in allowing unpredicted avenues to be followed-up, based on any prior observations and interviewees’ relevant interests.³² Alongside formal interviews, ad-hoc conversations were also had with staff and accounted for within the field notes.

Based on mutual agreement between the ethnographer (LR) and participants, where suitable, observations of relevant practice were made over the year period, including routine activity, meetings, events and training sessions. Observations were often scheduled at the end of each interview where participants were asked if any practices would be appropriate to observe, or where staff had specifically mentioned practices of interest. A total of 67 hours of observations were carried out, which comprised 12 hours at site A, 23 hours at site B, and 34 hours at site C. The observer aimed to build rapport, particularly with those perceived to be gatekeepers, typical cases, significant cases, deviant cases and those with insight into organisational influences to gain a holistic view.³³ Detailed field notes were kept, including descriptive and analytical reflections, first impressions, information about significant events, pictures and diagrams to assist memory and sense-making, details of non-recorded ad-hoc discussions and copies of relevant information. Field note journals comprised >95,000 words.

Data were also collected from relevant documentary sources, including publically available and participant provided information. This comprised public responses on Care Opinion, relevant reports, training material, policies, and photographs capturing significant events.

Case study analysis

Case studies, referring to in-depth investigations of cases within their real-word contexts,^34,35 were iteratively developed via regular data sessions between all authors.²⁷ Interviews, field notes and reflexive diary notes were transcribed (by LR). Pen portrait methodology was used to synthesise the complementary data sources for the purpose of completeness.³⁶ Data were organised according to study site, making note of initial impressions based on similarities and differences within and between sources, with subsequent inductive integration where they explored shared ideas. This helped determine interesting foci, both specific to the research questions and capturing novel ideas. A basic case study structure was designed within a working document, and significant excerpts were iteratively organised to support and refine them. Inductive preliminary themes within each case study were explored, adapting pen portrait methodology which aims to focus on short summaries, but instead, maintaining the richness of the large dataset using an interpretive, narrative approach.³⁷ The representation of data sources was not necessarily equal, and all sources were not necessarily represented, but included dependent on data quality and significance to the foci identified. The refined templates were populated with agreed themes, supporting narrative and data excerpts via an iterative process of revisiting the data and research questions, and refining until a consensus was reached.

Findings

We present three in-depth case studies, each with their own themes and subthemes.

Site A (non-responder): Doing our damnedest

Site A was a non-responder, as the organisation did not publically engage with online patient feedback posted via Care Opinion.²⁵ We interpreted that this was largely because staff were busy dealing with volumes of concerns raised in other ways. We present and explore two key themes, each with their own subthemes, in relation to this case.

Doing our damnedest

Our first theme refers to idea that staff were aware that their practices were flawed, but felt they were striving to cope within the pressurised system, facing time and resource constraints and identifying unmet training needs.

Casework: logging or analysing?

The team reactively dealt with individual ‘cases’ of feedback, typically received directly via telephone or email and negative in sentiment. This initiated, what we term, a ‘casework cycle’, involving feedback being assigned to ‘case workers’. Each were tasked with logging information via their web-based incident reporting system, liaising with clinical teams for ‘answers’ and ensuring timely resolution, all before revisiting the queue of concerns. However, staff expressed worries that information was getting stuck within the system, and frontline teams were unable to piece together information regarding individual cases to develop interventions.

We’re getting the feedback in, we investigate it and go to the wards to find the right people to answer the questions, but we don’t have actions … We need to create stronger relationships … close that loop. At the minute we don’t. That’s a big issue. (PALS Officer)

Staff felt that a more proactive approach leading to meaningful change at a service and/or organisational level, was a missed opportunity, expressing frustrations of having identified continuous themes of concerns.

I feel for the fact that people’s belongings are going missing, it happens all the time. It doesn’t feel like anything is changing … there are simple solutions and I have made the suggestion. (Patient Experience Officer)

Perceived to be fulfilling a purely administrative duty, the team endured the difficulties of lacking the autonomy to deliver change, and struggling to navigate the system to create relationships with those who did. On the other hand, the Head of Patient Experience relayed how her background in nursing enabled a mutual understanding with frontline staff to make issues ‘go away’.

They are brilliant this team here, but they are admin … I do have good relationships with the Ward Managers. I can go to them and say ‘do this and do that and it will go away, the problem will go away’ and they listen. (Head of Patient Experience)

While this approach may have helped casework cycles run smoothly, we interpreted that this was not necessarily an effective strategy to innovate and improve, demonstrated in the Board report suggesting that of the 39 complaints closed in the last 3 months, 19 had been re-opened. Getting answers for patients on an individual level could be considered papering over the cracks, opposed to careful consideration of underlying issues, sharing of best practice and investment in continuous improvement.

‘Organisational firefighting’

Staff within the Patient Experience team felt that day-to-day work was often chaotic and varied, requiring complex liaising with colleagues across the organisation. We interpreted this activity as tirelessly moving from fire to fire, extinguishing the flames of concerns, which left individuals feeling overloaded with information.

You can come in on a morning and think ‘right, I’ve got so many concerns open so I’m going to go through them, try and resolve them’, and then the phone can go and that will be it. Then you’ll look at the inbox, ‘Oh good god’ … you never know on a day-to-day basis, what it’s going to be like. (PALS Officer)

Sometimes ‘organisational fires’ caused irritation, highlighted by staff wanting to make them ‘go away’. However, many recognised that like the element of fire itself, organisational fires were natural and inevitable. Without ‘organisational fire specialists’, one strategy staff suggested would alleviate pressures, was to proactively solve problem classes, rather than cases, to reduce the volume of known concerns in collaboration with clinical teams. However, urgency tended to supersede importance, and staff were too busy working to meet the overwhelming demand to plan and implement their desired proactive approach. Staff tirelessly working to resolve concerns were recognised by management as the ‘unsung heroes’, while innovators were known to have great ideas theoretically, but must not be ignorant to pressing challenges. Pressures were furthered by intensifying volumes of patients making vociferous contact with the small team, views that their work lacked priority within the wider organisational context and suboptimal staffing. Interplay of these factors were blamed for an inability to innovate, leading to there being more fires than those who were firefighting could easily manage.

We would like to be more proactive … raise the profile of the Patient Experience team with the theory that prevention is better than the cure … That’s what this service was supposed to do … That’s been lost in the ether somewhere. (PALS Officer)

Emotional labour

Staff described, what we term, the ‘emotional labour’ of being personally inundated with concerns from patients often in states of anxiety, distress and frustration. One coping strategy used to reduce the burden, was to emotionally distance from both the organisation and patients, identifying as a messenger between the two. The team self-labelled themselves as a ‘conduit’, however, being stuck in the middle contributed to both their feeling of being misunderstood by frontline staff, but also being the first port of call for distressed patients.

You are literally on the go all day and you’ll have a headache at the end … you’re just trying to help everybody … it can be quite emotive … physically draining. (Patient Experience Officer)

Contributing to the emotional labour was the unsettling nature of change. Unpredictable disruptions included moving physical location within the organisation, shifting focus for the team, high staff turnover, and the discontinuation of services.

I don’t think the Trust is seeing this properly at the moment. We’re losing a frontline service. It’s not going to be there for patients and relatives to come and speak to us and that’s sad … it does make you wonder, how much this Trust actually rates PALS. I am questioning it … I am being quite negative but I’m angry. (PALS Officer)

Some alterations were perceived positively, however, with the potential to alleviate emotional labour. For instance, the new Chief Nurse had a previous remit in patient experience which gave staff a sense of hope and opportunity to work more collaboratively with others across the organisation. Groups were also formed and in their early setup phases, dedicated to focussed discussion surrounding patient experience, and those who did monitor online feedback sporadically, recognised this offered a way to hear more positive feedback.

Somebody, anybody, nobody?

Many staff assumed the Head of Patient Experience understood organisational roles and responsibilities for monitoring and responding to online patient feedback, but felt unable to answer questions themselves. However, similarly, the Head of Patient Experience presumed that the teams themselves would know more. One Care Opinion account holder had left the organisation, and the second did not perceive, nor was recognised by the wider team, as fulfilling such role. There were also suggestions that the Quality and Governance team held responsibility, despite not self-identifying as engaging as part of their work. Engagement was inconsistent and irregular, with some sporadic responses given midway through fieldwork. Response rate to patient feedback left via Care Opinion was at 26% in December 2019, which while being one of the highest for the organisation, benchmarked 205th of 226.

[Online feedback] is not part of my remit, I have another colleague who manages that and bringing all of that together and working with the Governance team who do the incidents and you have got the serious incidents and everything else, bringing all of that together. (Senior Patient Experience Officer)

It’s just the PALS staff. The Communications might see something online and give us a heads up but we check it every day … PALS team mainly, mainly yes. Unless we think it’s something quite serious and then we would forward it on to the Complaints team … I think so. I think there is an automated thing that says ‘if you would like to get in touch’, I’m sure there is. Yes, often they don’t reply to that and then you know it’s a rant. (Head of Patient Experience)

I personally don’t see it. I am not really well qualified to answer that … That’s one of the team members who does [online feedback]. One of the team members. I don’t deal with that … My main focus is actually taking the calls or emails … What our team actually does is monitor the NHS Choices website and we always give the appropriate feedback … they are fed to the team and then we would post our responses. And again that is used to improve our services. That is dealt with by different people … people who we call the Quality team. (PALS Officer)

Hierarchy of feedback sources

Our second theme considers a perceived a hierarchy of patient feedback sources, with staff often prioritising those the organisation initiated. Working to resolve volumes of concerns raised via these sources, staff considered engaging with external sources of feedback, such as Care Opinion, supplementary activity if they managed to find the time. Many felt that patients who left online feedback were not doing so with the intention that they would be listened to, but to ‘rant’ in a public forum.

Sometimes someone will put something online and you’ll think ‘ooh heck’ but generally if they are wanting an investigation they’ll write a letter. Things are picked up online but it’s not usually severe … usually people wanting to have a rant. (Head of Patient Experience)

Most staff struggled to understand why patients would utilise online options of feedback provision via an independent provider, as the organisations services offered everything they felt that patients would need.

If you have got a complaint take it up at the point of delivery … I wouldn’t suffer in silence and write about it anonymously when I got home … there is no reason to have any difficulties with raising a complaint or a concern within the NHS. (PALS Officer)

Roadblock of anonymity

Online patient anonymity sometimes left staff without detail necessary to determine the specific nature of the concern, who to contact within the organisation to get answers and the inability to complete their usual casework cycle. The aim of ‘getting answers’ to discrete issues remained, despite not necessarily being the patients’ intended purpose of communication, causing a roadblock. Similar frustrations were felt when patients left feedback anonymously via more widely used sources.

We have to go back and ask for specific information … find out what it is that they really want answering … without the full details there is very little that we can do. (PALS Officer)

To tackle this, staff crafted their own online system which forced patients to provide identifiable information alongside their feedback, meaning they could more easily bring about ‘resolution’. This also alleviated worries that unknown patients would publicise negative or disingenuous discourse.

We said, well if someone contacts us online what information do we want? The online process is good because we got all of the information that we wanted and in the right order because we designed it. (Patient Experience Officer and PALS Lead)

Site B (generic responding organisation): Disenfranchised with online feedback

Site B adopted a generic responding style,²⁵ which we interpreted was due to resource constraints, fears of public engagement and focus on resolving know issues raised via more traditional feedback sources. We present and explore two key themes, each with their own subthemes, in relation to this case.

Siloed working

Our first theme considers how stakeholders in patient experience generally worked exclusively on specific tasks. Illustratively, a junior member of the Communications team manually provided generic responses to each patient feedback narrative, yet there was a general unawareness of online patient feedback as a phenomena across the organisation. Some recognised that this was not necessarily effective for wider organisational improvement. However, factors perceived to be outside of their control, such as systems issues, organisational processes, being in competition for resources, and time constraints, did not make collaboration easy, even for those in more strategic roles.

You can learn from different departments and hospitals … We’re all very busy, beavering away and trying to do a good job, inventing things … we could just share and save a lot of money and time … that’s my big bugbear but I’m too little of a person to solve that. (Public Health Specialist Midwife)

‘Feeding the beast’

The team invested heavily in quality administration of patient feedback received via more traditional sources using their incident reporting system. Feedback was collected, inputted and prepared to make information readily available, both routinely and on an ad-hoc basis to fulfil multiple functions. This included supporting audit and governance, producing area-specific reports and for in-depth improvement activities within services where concerns were raised. A core activity was producing detailed reports to form the basis of bi-monthly meetings attended by largely senior nursing staff. The core assumption was that clinical staff would independently action feedback outlined within the reports, with support from the central team where necessary.

We feedback to the management team and then it would be up to them how they then take that on board. (PALS Coordinator)

While it was deemed essential to ‘feed the beast’ of the incident reporting system with information, the purpose of data preparation activity was sometimes self-fulfilling. Some felt that the efforts invested by the Patient Experience team to prepare reports were often disproportionate to the useful outputs clinical teams were able to derive, with the majority of feedback disappearing into the ‘beasts’ black hole.

All of that information and data could be better used. We’re collecting stuff and then it goes into a little black hole somewhere. (Communications and Digital Manager)

Pockets of innovation

Staff felt that feedback-informed improvement often struggled to compete with wider organisational priorities, resulting in data being used to supplement predetermined decisions. However, in areas, some felt that patient feedback was genuinely valued and impactful, such as regular patient story sharing with the Board and a fresh energy for patient-centred improvement embraced by the Chief Executive and newly-appointed Head of Nursing. Maternity was also recognised as an exemplary service, as feedback sources were embraced beyond those used centrally. This included a birth thoughts clinic, digital feedback collection by midwives on the ward, a private social media group supporting bereaved parents, and a volunteer-led maternity voices group. This group brought volunteers and staff together on a bi-monthly basis and was supported by an active Facebook page used to encourage feedback from families, alleviating the burden of hospital-based appointments. Families were also offered access to coproduced website, for which the organisation was recognised with an award.

When we first put the website up a mum put a comment saying ‘I need to find out what to bring to hospital’… that mum wanted something really practical and she couldn’t find it, so we put that on. We worked with our mums, we worked with the staff, to write a list (Public Health Specialist Midwife)

Maternity staff felt that listening to patients was an intrinsic part of their caring roles, which some staff perceived was a view not necessarily shared across the organisation, as some struggled to see the value in feedback, particularly when perceived as criticism.

What does upset me is some old fashioned doctors still have entrenched attitudes towards patients who complain … they become very defensive, they don’t understand … these are human beings, they have lost somebody … doctors don’t like those kinds of things. It’s all figures and statistics. (Head of Patient Experience)

One maternity staff member expressed that they would be surprised if the organisation were responding to online patient feedback regarding their services without frontline staff engagement. However, our research suggested that this was happening without their knowledge, raising potential ethical issues.

Signposting patients elsewhere

Stakeholders disagreed about the appropriateness of their organisational approach to online feedback, involving Communications staff using their time ineffectively to provide the same generic response, regardless of feedback content, and signpost patients elsewhere in the system. Some felt that on balance, this approach streamlined efforts and reduced workloads for pressurised teams.

Instead of us having to go through all of those comments and respond on top of everything that we are doing … the system works well that we have got in place where patients are signposted to us for us to pick it up if they want us to. (Patient Advice and Complaints Manager)

Yet the majority felt that ideally, relevant stakeholders would personally respond, develop actions and publicly outline where changes had been made to close the feedback loop. There were concerns that providing generic responses would indicate an organisational culture that did not value patients or their experiences.

We should be responding personally … some kind of authentic message … if you scrolled down, you could be forgiven for thinking that it is a computer that replies. (Communications and Digital Manager)

Disenfranchised

Our second theme refers to staff at site B generally feeling disenfranchised with online patient feedback for various reasons, including resource and time constraints leading to a low-cost, in-house approach, a perception that their patient profile was better suited to traditional communication methods and apprehensions associated with public engagement.

Understanding the patient profile

While many appreciated the potential benefits that Care Opinion may offer to other patients, some felt that the culture and socioeconomic profile of their patients specifically, was one in which online communication was not best suited. This seemed to be supported by the comparatively low volume of feedback arriving digitally, however, may have reflected the perpetuating effects of lacking of internal engagement.

Here the culture is to do it almost immediately, [patients] would prefer to come and talk to you … online feedback isn’t something that has affected what we do. (PALS Coordinator)

The Head of Patient Experience was an influential decision-maker who held the view that online feedback did not necessarily suit their patients’ needs. However, at the end of fieldwork, they retired from the position and considered that when their successor takes up their post, they may adopt an alternative approach. There was recognition of discrepancies between their perception and reality in terms of the way patients wished to communicate.

In my lifetime it was predominantly letters, but now complaints say roughly 70% is coming in electronically, which I was a bit surprised at myself. I thought it was more 50:50. (Head of Patient Experience)

Judicious investment

Many expressed that staff already knew the core organisational issues, and that focus should be moved towards using existing knowledge to develop interventions, rather than investing in additional ways of hearing what they already knew. While the two barriers were in place, having the time and resources to process data and implement change outweighed that of sourcing patient feedback. Decisions to unsubscribe from Care Opinion were made during times of austerity.

It’s about giving them that space to sit down, think, plan and be imaginative … People think a lot about getting feedback but they don’t think ahead about how we are going to manage all of that … That’s why no, I’m not giving money to people like Care Opinion …. when we went into financial difficulty, I had to make those decisions. And stuff like that just went out of the window … We’ve never chucked money at external providers … always done it in-house, low-cost and low-key. (Head of Patient Experience)

Online feedback was perceived a tool available for staff to gain information about their services, but was not one they should, or would feasibly be able to, publically engage with.

You might end up spending all of your time responding and not actually getting things done. They are more to give us those trends and themes, highlighting what we need to work on … if people wanted more of a response they would come through complaints … their experiences of care when they come back are all that they need. (Patient Experience Improvement Manager)

Handle with caution

Engaging with patients in public forums removed staff from the comfort zone of surveys and discrete telephone calls, with perceived risks including the potential to breach confidentiality and information governance.

I don’t like engaging in open forums. I am not going to do that because of confidentiality. They will get a standard response … saying thank you for your feedback. It would be nice to personalise that sometimes but I am always very mindful of IG [Information Governance], GDPR [General Data Protection Regulation] and confidentiality. (Head of Patient Experience)

Other concerns included the public disclosure of clinical information, patients being more challenging in a public environment and concerns of digital exclusion.

There is a growing drive towards to electronic collection of data and it makes an assumption that everyone has the means to submit their views electronically which is absolutely not the case. (Volunteer, Governor and Patient Representative)

Site C (transparent, conversational responder): 10-year journey of challenge and perseverance

Site C provided transparent, conversational responses²⁵ to patients online as part of a wider mission for transparency, and described a 10-year journey enabling their desired culture to be embedded. We present and explore two key themes, each with their own subthemes, in relation to this case.

Embedding organisational culture and values

Our first theme refers to Care Opinion being considered integral to demonstrating organisational values, and used to inform over 90 actions in adult mental health services outlined within site C's annual report. These comprised larger transformations of practice, such as restructuring and refocussing a self-help management programme, and smaller environmental modifications, such as the provision of ramp facilities.

Sometimes our services, like every other healthcare service, gets it wrong … we don’t shy away … if you’re not hearing negative feedback, it’s because it’s being said elsewhere, that the patient is too frightened to tell you, or doesn’t have any confidence that you’ll do anything with it … we can learn so much and we can go a long way towards restoring that person’s faith in our services. (Involvement and Experience Manager)

Component of a wider ambition

While Care Opinion was the main way the organisation approached patient feedback transparently, it was an element of a wider ambition. In 2012, the organisation secured funding to build their own website and become the first in the country to publicise almost all of their patient feedback online. A Senior Analyst referred to aiming for this to become a nationally recommended system.

We’re not scared about publishing things and it having a negative effect on our organisation. I don’t think we want to hide behind anything … we’re the only Trust that does publish our feedback through our feedback website … we can only learn from any feedback that we get, whether it’s positive or negative. (Volunteering and Befriending Lead)

Staff aimed to flexibly and sensitively meet patients where they chose to share their feedback, rather than expecting patients to conform to traditional communication methods. This required non-risk-averse attitudes from three staff members who initially embraced online patient feedback on behalf of the organisations. Worries of pushback from the Board were alleviated with support for their decision. The shared vision became important in enabling buy-in from the organisation at different levels. Over time, it set a precedent for staff, increasing awareness, knowledge and acceptance that anyone who wanted to, could look at feedback patients were providing about their services in the public domain.

Have all of your feedback online and be very visible, open and honest about it. That’s really healthy for an organisation’s culture. And any organisation that doesn’t think like that needs to revisit actually what they are there to do. If you don’t want the world seeing what people are experiencing in your services, then do you really care what people are experiencing? … I would argue at that point that you care more about the reputation of your organisation. (Involvement and Experience Manager)

A journey of challenge and perseverance

Reflecting on their 10-year journey with online patient feedback, staff considered prior challenges including non-user-friendly reports getting stuck with managers and difficulties persuading staff to embrace transparency.

There were worries people could say things which could cause the organisation disrepute, pointing out things that may not be true … worries from staff that they would be personally attacked … that’s completely natural. (Head of Involvement and Experience)

The introduction of a standardised patient feedback questionnaire was also highlighted as a ‘frustrating step back’, moving focus away from the meaningful use of rich, actionable feedback, to generating volumes of data. A milestone in their Care Opinion journey was the devolution of responding responsibility, gradually increasing from three responders within the centralised team, to over 890 staff across the organisation. Internally developed guidelines supported many, including frontline staff and executive Board members, to become actively involved in monitoring, responding to and improving based upon the information. Efforts were expended to ensure attitudes were sustainably embedded via ‘super-user masterclasses’ in which Care Opinion users were asked to each invite a ‘future super-user’.

We’ve changed things so much. In forensic services where no staff had ever responded before, we get long essays from staff sometimes which are quite incredible … when people relax, they have posted a few times, they feel more able to do that. It just takes time. (Involvement and Experience Officer)

Ongoing learning journey

Our second theme considers that while the organisation was internally and externally recognised as an example of best practice, staff were keen to continue learning. For instance, the organisational topography consisting of multiple sites and a large geographical dispersion, meant that levels of engagement inevitably varied. Champions found that getting everyone on board was an ongoing challenge.

The teams and clinicians that have been actively involved really get it, but I guess it’s how we get every team member to get it? … we have the same conversations over and over and over again and that’s really tough going, to keep that motivation …. some staff are quite advanced … some are like ‘what’s that?’ (Service Improvement Facilitator)

Staff used regular meetings, events and reporting activity to widen their reach and establish consensus on shared goals staff were working towards, including synthesising widely used feedback sources such as Care Opinion, with more local feedback sources such as carer engagement groups, local engagement work, community meetings, and thank you cards. Staff also wished to improve registering changes on Care Opinion, which was considered a concept difficult to measure, particularly where changes were behavioural or cultural.

They will quite often say, ‘We took that posting to our staff handover, staff team meeting, we read it out loud, we had a conversation about it, we thought could this happen again? We compared it to other people’s experiences’ … that feedback continually has this impact on staff in opening their eyes … it does make a much bigger difference than is clear … Just this morning a manager was talking about a posting from years ago and said ‘oh this one really touched me’ … that’s the kinds of conversations with staff that I have a lot. (Involvement and Experience Manager)

Discussions focussed on developing innovative solutions to engaging particular patient groups online, as Care Opinion was not considered a one-size-fits-all approach. Dedicated staff took on responsibilities for routinely identifying teams not capturing feedback effectively, state how they would support them to improve, and liaise with others to monitor progress. One staff member was recognised with an internal ‘Care Opinion heroes’ award for engaging with a patient with intellectual disabilities online via artwork. Others developed YouTube videos to assist children using Care Opinion, used leaflets to engage with the elderly and liaised with the local Deaf Society to incorporate sign language. Additionally, a goal to emphasise the value of positive feedback derived from an interactive staff workshop. Many staff felt that concerns were informing improvements, however, the impact of praise was easily underestimated as purely recognition of staff for simply doing their jobs. Hoping to change this, staff wanted to more often share examples of excellence, boost morale and motivate staff with positivity.

If only we proudly dwelled on the compliments for as long as we shamefully dwell on the complaints. Interesting to imagine how different our culture might be from just this one small shift. (Tweet from the Clinical Director for General Health)

Some felt able to benchmark their engagement with online feedback against different organisations, however, others spoke about how they rarely got the opportunity to look outside.

I was surprised when we were looking at tendering for another service in another part of the country … we tried to find out where they had published their feedback, something about them in a public forum and there was nothing to be seen … that really surprised me … that this isn’t just normal practice everywhere else in that they publish what everyone thinks about them whether it’s positive or negative. (General Manager)

Competing organisational priorities

Staff encountered frustrations, as while they felt a culture that embraced online feedback had been successfully embedded, when it came up against competing organisational priorities it was not always able to stand up against them.

There’s a drive to hear feedback and understand what we’re getting, but I guess the word value is quite an interesting one because we still work within the NHS … Budgets are tight, we can’t wave magic wands … We can spend a lot of time doing the involvement work and hearing the feedback … sometimes we can get agreement that we’re going to deliver on it, but it can be two years down the line and you’ve still not done it. (Service Improvement Facilitator)

An example repeatedly mentioned by staff was an in-depth co-design project conducted with children over a period of 2 years to directly inform and develop a new children's unit. A key theme within feedback was that children did not want notice boards in the reception area to display formal organisational information such as health and safety or security data, but wanted a welcoming and child-friendly environment. This was agreed upon by the key stakeholders and actioned. However, during a visit from a regulatory body, an inspector questioned why they did not display their most recent report, leading to the decision being overridden and a notice board being put up. This was disheartening for those involved and highlighted a further challenge in that while patient feedback was able to directly inform change, ongoing work was required to maintain those changes and ensure that other pressures, demands and directions did not overrule them.

That commitment in the forefront is there … but I still think that the processes that we have in the organisation and some of the other things that are prioritised tend to swallow up patient experience … it doesn’t always hold the weight it should … when a service has to think about how many people have died … how it keeps people safe, risks, critical health needs, how quickly they discharge people … making sure that patient experience isn’t just thought of as a fluffy thing. (Involvement and Experience Manager)

Discussion

Our paper presents three in-depth case studies of organisations that approach responding to online patient feedback differently.²⁵ Site A did not respond to or use online patient feedback as staff were busy firefighting volumes of concerns raised in other ways. Site B adopted a generic responding style due to resource constraints, fears of public engagement and focus on resolving known issues raised via more traditional feedback sources. And site C provided transparent, conversational responses to patients online and described a 10-year journey enabling their desired culture to be embedded. The findings from each site are discussed in relation to the existing literature in detail.

Site A

The non-responding organisation involved staff firefighting volumes of feedback under pressure, leading to the prioritisation of feedback sources they initiated. The primary goal of coping under pressure can be seen in relation to a comparable emphasis on ‘putting out fires’ in complaints offices, detracting efforts from improving future care.³⁸ Paying close attention to immediate patient safety concerns, while being blind to the wider picture has also been termed ‘collective myopia’.³⁹ The overwhelming volume, variety and negativity of feedback placed a significant burden on staff considered “just admin”, illustrating a macro level issue. The findings align with the patient feedback response framework, emphasising the importance of structural legitimacy, as staff require perceptions of sufficient authority and autonomy to enact change.⁴⁰ In circumstances of emotional labour, the art of caring itself also demanded complex skills and resources,⁴¹ emphasised within an organisational context where staff were required to invest in understanding the emotions of others without reciprocity, and where staff emotions could be commodified, manipulated and shaped by wider organisational goals. Hochschild's theory of emotion work suggests that stress can arise when individuals lack confidence in reassurances they are responsible for relaying, which was openly discussed by staff who felt encouraged to act in ways that did not always fully reflect their personal beliefs and attitudes.^42,43 This has been termed cognitive dissonance,⁴⁴ and identified as a major contributing factor to reduced job satisfaction and burnout.^45,46 The importance of ensuring adequate support, supervision and training for staff dealing directly with concerned patients, while managing their own diverse workloads and the wider organisational demands, is therefore emphasised, and should be considered in both policy and practice.

Site B

The organisation provided generic responses to all online feedback for multiple reasons, including a siloed approach to working and resource constraints leading to a low-cost, in-house approach. The responsibility for improvement activity was fundamentally an expectation of those at the sharp end of care, yet often, clinical staff were distanced from the exclusivity of patient experience work, raising potential ethical issues alluded to in previous research.^20,21 Insular siloes have also been a longstanding healthcare issue, both within and between organisations,⁴⁷ and the potentially devastating implications were demonstrated by the Bristol Royal Infirmary inquiry.⁴⁸ The inquiry concluded that contributory factors to the unnecessary deaths of over 30 children included systemic fragmentation, a flawed ‘club culture’ of insiders and outsiders, lacking teamwork, and ‘too much power in too few hands’. It was also illustrated at site B that simply making data available to staff was not sufficient to drive change, which has been considered an enduring issue in patient experience.¹¹ Congruently, Pflueger⁴⁹ suggested that the process of accounting in ways such as producing patient feedback reports, may set out particular ambitions that make things knowable and accountable. However, reports may then become newly configured entities themselves, which not only represent and replace the original sentiment of patient feedback, but significantly change and transform the function of the original enterprise. Therefore, a greater focus needs to be placed on meaningfully engaging with patient experiences to improve, rather than disproportionate investment in the self-fulfilling activity of reporting, which may require buy-in at both a local and policy level.

Site C

The organisation providing transparent, conversational responses described a difficult journey over 10 years to enable their desired organisational culture to be embedded, which they were keen to reflect within their responses to patient feedback online. As a mental healthcare provider, proactively welcoming the voices of patients may, in part, be a function of cultural cohesion with the history of user activism and underpinning values of mental healthcare. However, research has contrastingly suggested that mental healthcare is one of the most challenging settings to obtain, listen to and use patient feedback, and tends to be insufficiently embedded.^19,20 Recent research found, however, that organisations that engaged with a wider variety of patient feedback mechanisms tended to be better at using the information to inform improvement, and noted the significance of building trusting relationships from a patient perspective.¹⁹ At site C, the equitable and impartial platform of Care Opinion was used to demonstrate how staff were publically responding and listening to patients as a way to nurture trust. This led to both internal and external recognition of being the gold-standard, yet learning was ongoing. The importance of constant reviewing, strengthening and improving of systems has been previously highlighted.⁵⁰ It also supports calls to recognise the importance of local ownership in patient experience, and the significance of activity being culturally embraced.^51,52 Nonetheless, the length of time and efforts to achieve this must not be underestimated. Interestingly, an arguably similar cultural transformation to site C was achieved more rapidly within an acute NHS organisation which was rated by the CQC as ‘inadequate’ in 2017.⁵³ Care Opinion was subsequently used as a strategic measure of the success of intervention implementation, with an aim to improve care and also shift the culture. Findings suggest that this was embraced positively, and relatively rapidly, however, the sustainability of this approach is yet to be fully understood.

Strengths and limitations

To the authors’ knowledge, this is the first study of its kind, having direct implications for a wide range of stakeholders in patient experience, including: policy makers, service providers, commissioners, managers, frontline healthcare staff and patients. However, a weakness of the study is that patient views and experiences were not explicitly sought, but rather outlined from the perspectives of staff, or presumed based on previous literature, and may not fully reflect a comprehensive understanding of the views of individual patients themselves. Nevertheless, patient views regarding online feedback have been previously considered in recent research.^20,21,54 Methodological limitations include that while efforts were made for the observer to be attentive to both issues relevant to the specific research questions and novel, interesting phenomena within the wider context, ethnographic methods were iterative, with new foci of attention being identified continuously throughout the research process, meaning that information may have been missed.⁵⁵ Conversely, it may have been that too much attention was paid to phenomena identified as interesting, causing novel phenomena to be missed. Finally, cases include only a subsample of eligible organisations, and were not necessarily intended to representative, but selected pragmatically and/or exemplar cases of the typology.

Future research

Study authors are undertaking a multi-case analysis to make higher-level comparisons across the three case studies, and derive recommendations for policy and practice.⁵⁶ Future research could also use the response typology²⁵ as a foundation from which to consider the patient perceptive of received the five identified response types, i.e. how do patients feel to be ignored online, provided with a generic response, thanked and/or apologised to, invited offline to continue the conversation or engaged with in a transparent conversation publically, building on previous work.⁵⁴ Finally, the wider context of the global COVID-19 pandemic may also be useful to explore in relation to the findings, potentially providing an opportunity for organisations to revaluate their approach to patient experience information and connect virtually without the added risks to infection control. Conversely, the pandemic may place further strains on resource and capacity which could potentially exacerbate issues with how online feedback is prioritised.

Conclusions

Our in-depth case studies identify a range of sociocultural barriers facing staff who do not respond to, or provide generic responses to patient feedback online. We also demonstrate the sociocultural content in which online interactions between patients and staff can be embraced and used to inform improvement. However, this represents a slow and difficult organisational journey of dedication and determination to identify and engage stakeholders over time, and embed organisational learning, which must not be underestimated. To move forward in this area, further research is needed to establish how organisations can better recognise and overcome barriers to engaging with online patient feedback more widely, and at pace, to ensure important opportunities for healthcare organisations to learn and improve are not missed.

Appendix A. Case characteristics

	Site A	Site B	Site C
Trust characteristics	Operated from two hospital sites and offered additional services based across the region, including a large range of acute services within the area and a number of specialist services beyond the area. The Trust employed around 8000 staff and cared for approximately 1 million patients annually. The most recent CQC rating was published in 2018, which rated them overall as ‘requires improvement’.	Run a 350+ bed hospital that served a population of over 250,000 people in the area, caring for over 61,000 patients in the hospital, 268,000 in clinic appointments and treating 78,000 in the emergency department. It also provided a range of district hospital services to the local community and surrounding area including emergency department services, outpatient clinics, inpatient services and maternity and children's services. A number of specialised services were also provided including cancer and surgical services, in conjunction with a local Trust The most recent CQC report was published in 2018 and awarded them overall ‘good’.	A major provider of mental health, intellectual disability and community healthcare services, with approximately 190,000 patients per year. The Trust employed approximately 9000 staff who provided a variety of services for people with mental health needs including secure mental health services. The most recent CQC rating was published in 2019 which awarded them overall ‘requires improvement’. However, within the report their activity around online feedback was highlighted as an example of outstanding practice.
Patient Experience Characteristics	The Trust did not have a formally designated patient experience team per se, but the head of patient experience managed both the small PALS and complaints teams that originally operated via the two separate hospital sites, which then merged to a single site during fieldwork.	The Trust had a formally designated patient experience team that adopted a centralised model, encompassing both the PALS and complaints teams. However, it was the communications team who were responsible for responding to feedback online including Care Opinion, but also other mechanisms such as Twitter and Facebook.	The Trust had a formally designated patient involvement and experience team that also incorporated volunteering and adopted a centralised model. This team operated separately to the PALS and complaints teams but worked closely with other teams across the trust including involvement champions within each directorate and the board of governors.
Care Opinion characteristics	At the beginning of fieldwork, the trust had 1995 stories posted on Care Opinion from patients, carers and/or their families regarding their services. This increased to 2126 by the end of fieldwork, which had been read by both staff and the public 599,406 times by April 2020. Based upon this feedback, the organisation had 0 changes planned and 0 changes made, according to those which were registered via the Care Opinion platform. Their response rate began at 0%, however, increased to 8% in Sept 2019, and reached 30% by the end of fieldwork. The trust had 2 registered staff users on the site, which was the maximum number allowed without a paid subscription. However, these individuals rarely engaged with Care Opinion, and when they did, they tended to access it via nhs.uk which Care Opinion interoperates with.	At the beginning of fieldwork the trust had 1073 stories published from patients, carers and/or their families on Care Opinion. This increased to 1193 by the end of fieldwork which had been read by both staff and the public 1,123,577 times by April 2020. Based upon this feedback, the organisation had 1 change planned and 1 change made, according to those which were registered via the Care Opinion platform. However, these were stories from 6 to 10 years ago. At the beginning of fieldwork their response rate was at 99% which decreased to 83% at the end of fieldwork. There were no active staff users registered on the Care Opinion platform as the communications team responded via nhs.uk, which the site interoperates with.	At the beginning of fieldwork there were 5883 stories published regarding their services on Care Opinion, which had received 6767 responses. This increased to 6542 stories and 7685 responses by the end of fieldwork, with both staff and the public reading those 1,787,932 times. Based on this feedback, the organisation had a total of 187 changes planned and 215 changes made, which were registered via the Care Opinion platform. Their response rate was 86% at the beginning of fieldwork, which increased to 89% at the end of fieldwork. However, response rate was calculated via Care Opinion based on the percentage of responses given to the most recent 100 stories, and due to the high volume of stories being posted about this organisation, there was a higher rate of new stories that previous activity suggests were likely to, but were yet to receive a response. The organisation had 854 active members at the beginning of fieldwork which increased to 895, meaning that many staff with diverse job roles were able to respond to online patient feedback at the Trust