Understanding how front-line staff use patient experience data for service improvement: an exploratory case study evaluation

Article history

The research reported in this issue of the journal was funded by the HS&DR programme or one of its preceding programmes as project number 14/156/06. The contractual start date was in November 2015. The final report began editorial review in May 2018 and was accepted for publication in November 2018. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HS&DR editors and production house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Louise Locock declares personal fees from The Point of Care Foundation (London, UK) outside the submitted work, and membership of the National Institute for Health Research (NIHR) Health Services and Delivery Research (HSDR) Funding Board. Jennifer Bostock declares membership of the NIHR HSDR Funding Board. Chris Graham declares a conflict of interest in financial activities outside the submitted work (he is employed by the Picker Institute). Neil Churchill declares a conflict of interest in financial activities outside the submitted work (he is employed by NHS England). John Powell is chairperson of the NIHR Health Technology Assessment (HTA) and Efficacy and Mechanism Evaluation (EME) Editorial Board and Editor-in-Chief of HTA and EME journals. He is the principal investigator on another NIHR HSDR programme-funded project that was funded under the same call [HSDR 14/04/48: Improving NHS Quality Using Internet Ratings and Experiences (INQUIRE)]. Sue Ziebland declares a conflict of interest in financial activities outside the submitted work (Programme Director of NIHR Research for Patient Benefit). She is the co-investigator on another NIHR HSDR programme-funded project that was funded under the same call (HSDR 14/04/48: INQUIRE).

Disclaimer

This report contains transcripts of interviews conducted in the course of the research and contains language that may offend some readers.

Copyright statement

© Queen’s Printer and Controller of HMSO 2020. This work was produced by Locock et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

2020 Queen’s Printer and Controller of HMSO

The importance of patient experience

Patient experience, alongside patient safety and clinical effectiveness, is widely acknowledged as a key component of quality of care. 1 Improving patient experience is a priority for the NHS, which has led the way in developing measures of patient experience, such as the NHS Adult Inpatient Survey. 2 Patients have a right to expect care that is compassionate, respectful and convenient, as well as safe and effective. Collecting data about patient experience, although important, is not enough: the data need to be used for improvement, and it is arguably unethical to ask patients to comment on their experience if these comments are not responded to and acted on. 3

Recent evidence suggests positive associations between patient experience, patient safety and clinical effectiveness for a wide range of disease areas, and between patient experience and self-reported and objectively measured health outcomes. 4–6 At a time of global recession, there is a risk that better-quality patient experience may be seen as a luxury rather than a top priority. However, the apparent conflict between maintaining tight financial control and providing good patient experience may not be as clear as is sometimes supposed.

First, we know that many of the things that matter most to patients are relational, for example paying attention to dignity, courtesy and kindness. These may not be resource-intensive in themselves (although they may involve emotional labour for staff, and may be easier to attend to if there is less pressure on staffing levels). Second, there is growing evidence linking person-centred care with decreased mortality and lower hospital-acquired infection rates, as well as a range of other organisational goals, such as reduced malpractice claims, lower operating costs, increased market share and better staff retention and morale. Hospital organisations in which care is person-centred are reported to have shorter lengths of stay and fewer medication errors and adverse events. 7–14 There is increasing evidence that good staff experience is a predictor of good patient experience. 13,15,16

Yet despite these compelling reasons for paying close attention to experience data, the quality of patient experience remains a concern, and continued media reports and inquiries have suggested that there is a long way to go in ensuring the provision of genuinely and consistently person-centred care at all levels of the organisation.

Current patient experience in the NHS

The NHS Adult Inpatient Survey in England is one of the most well-established patient-reported experience measures (PREMs); it has been providing national and trust-level data since 2002. However, the pace of change has remained slow on some of the most important questions for person-centred care. 17 The results of the 2016 NHS Adult Inpatient Survey indicate that there have been small, but statistically significant, improvements in a number of areas of care, compared with results dating back to the 2006, 2011 and 2015 surveys. 18 This includes patients’ perceptions of:

• the quality of communication between medical professionals (doctors and nurses) and patients

• the standards of hospital cleanliness

• the quality of food.

However, in some areas a decline in reported experience has been noted, which goes against a trend since 2006 of largely stable or improving scores. This includes patients’ perceptions of:

• their involvement in decisions about their care and treatment

• the information shared with them when they leave hospital

• waiting times

• the support they receive after leaving hospital.

Only 56% of patients reported feeling as involved as they wanted to be in decisions about their treatment, which is down from 59% in 2015, and the proportion feeling that doctors and nurses ‘definitely’ gave their families all the information they needed when leaving hospital was down from 66% in 2015 to 64%. Only 55% of patients felt ‘definitely’ as involved as they wanted to be in decisions about discharge, down one percentage point on 2015; and 17% of patients reported waiting longer than 5 minutes for a call bell to be answered, up from 14% in 2006.

Since 2006, the proportion of patients who say that they have been asked to give their views on the quality of their care has increased substantially, rising from 6% in 2006 to 20% in 2015, but with a decrease of one percentage point to 19% in 2016. In other words, 81% of respondents said that they were not asked to give their views on the quality of their care, which indicates a substantial missed opportunity to learn from patients.

Existing evidence on using patient feedback and patient experience data for improvement

There is already good evidence of what matters most to many patients and how they experience services;19 the gap is in how trusts can respond to the available local and national data and use them to improve care. 3,17,20,21 It should, of course, be noted that the process of gathering and analysing data on patient experiences at local level can itself be part of building momentum for the use of these data in improvement. In narrative-based approaches such as experience-based co-design (EBCD), collecting experiences (narratives) and using them for improvement are closely linked as part of the same process of cultural change. 22 Nonetheless, there has been a lack of evidence about how organisations can move beyond collecting patient experience data to using them for improvement, and how best to use different types of quantitative and qualitative data from different national and local sources.

As Gleeson et al. 23 note, this is a relatively new field that does not have a large body of published evidence, a key factor behind the decision of the National Institute for Health Research (NIHR) to commission several new studies in this field, of which this report is one.

Dr Foster Intelligent Board research20 from 2010 into how trust boards use patient experience data found substantial variation in what data were collected and also, more importantly, in the degree to which the data were effectively analysed and used to improve services. Strikingly, the research found that, over 95% of the time, hospital boards’ minuted response to patient experience reports was to note the report but take no further action. Examples of when patient experience data were used to spark debate and action were rare, as were examples of non-executive directors challenging performance based on patient experience measures.

Since the Dr Foster research, studies have continued to show that interest in using experience data at senior level is not always matched by action or the skills to work with these data. Lee et al. 24 studied how two hospital boards of directors used patient feedback. They found that although the boards used both quantitative survey feedback data and in-depth qualitative data to help them develop quality strategies and design quality improvement (QI) work, they made less use of patient feedback to help monitor their strategies and assure quality. Even though the data contributed to the development of strategy and QI, in one of the sites this ‘did not result in formal board commitment to these initiatives’. The authors suggest that there is a need for further research to investigate how boards of directors can make better use of patient feedback, from a range of sources, and combine it with wider pressures and priorities to effect change.

Martin et al. 25 explored senior leaders’ use of ‘soft intelligence’ for QI, that is, informal, localised, often narrative or observational data of a kind not easily captured in standard quantitative metrics. They noted that senior leaders were highly aware of the value of such intelligence, gleaned perhaps from conversations with staff or just from walking on to the ward, and of ‘the idiosyncratic, uncalibrated views of patients or their carers’, but that they were unsure how to turn this potentially valuable material into a useful guide to action. In leaders’ efforts to make this material useful, they often sought to apply quantitative approaches, such as aggregating multiple comments or observations, or looking for triangulation with other sources. Seeking triangulation may be misguided, given the observed mismatch between what patients say about their experience in response to survey questionnaires and what they say in interviews,26 or between survey scores and free text. 27 Formal systemisation, Martin et al. 25 argue, risks losing the spontaneous ‘untamed richness’ of insights from soft intelligence. The alternative approach, they observed, was to downgrade soft intelligence from the status of evidence to that of illustration and motivation, for example using quotations to bring dry numerical data to life, but not focusing directly on its content as a guide to improvement.

Understanding how to use qualitative evidence, including narrative and observational data, alongside numerical evidence can indeed be challenging,20,28,29 and this is undoubtedly one of the key areas in which trusts need support and training. The status of complaints in this landscape of patient experience data is contested; as unsolicited and challenging examples of patient experience, they may be actively resisted. In their analysis of how health-care professionals make sense of complaints, Adams et al. 30 note that participants tended to characterise complainants as ‘inexpert, distressed or advantage-seeking’; they rationalised their motives for complaining ‘in ways that marginalised the content of their concerns’, and rarely used the content for improving care.

Meanwhile, the NHS also faces an explosion of other unsolicited sources of patient experience data online and through social media [e.g. NHS Choices, Care Connect, Care Opinion, blogs and Twitter (Twitter, Inc., San Francisco, CA, USA)], with little guidance on how to use them. As Dudwhala et al. 31 argue, unstructured data that have not been actively sought or sanctioned by health-care organisations are unlikely to have the impact of other, more formal, data sources. Similarly, in their study of how staff raise safety concerns, Martin et al. 32 note that organisational leaders tended to prefer such concerns to be routed through formal systems in order to provide evidence for action. However, as a result, some concerns were never voiced, because of staff anxiety about using these channels and becoming drawn into bureaucratic procedures. Thus, both staff and patient experiences that come through informal routes may gain little traction.

Although there is a growing number of studies about how patient experience data are (or are not) used at board or whole-organisation level, we still know remarkably little about how front-line staff make sense of, or contest, patient experience data, what supports or hinders them in making person-centred improvements and what motivates staff – and patients and families – to get involved in improvement work.

Quality improvement programmes and techniques abound, but few are strongly evidence based and few take seriously the need to involve patients and families throughout the process. A few studies have provided some evidence of promising front-line experience-based approaches. For example, facilitated feedback of survey findings at ward team level has been shown to have the potential to improve patient experience scores. 33 Studies of EBCD have shown how locally collected narrative and observational data can lead to both specific improvements in care and changes in attitude and behaviour. 34–38 Locock et al. 39,40 reported positive findings on the use of nationally collected narrative data alongside local observational data in EBCD to generate improvements, and a randomised controlled trial of EBCD is under way in Australia. 41 A UK evaluation of the patient- and family-centred care (PFCC) approach is also in progress. 42,43 However, there remains insufficient evidence to say which types of data or QI approaches are more or less likely to be useful to front-line teams in making care more person-centred in different contexts and settings, and how well these are received.

Gleeson et al. ’s23 systematic review of approaches to using patient experience data for QI has helped consolidate the evidence base. The authors state that their focus is on PREMs, but their review includes three studies reportedly using EBCD (interviews and observations) and one using PFCC (observations), alongside seven using questionnaire survey data. This reflects a broad interpretation of the word ‘measure’ to incorporate ethnographic approaches to understanding patient experience.

The systematic review results have several points of significance for this report:

• Patient experience data were most commonly collected through quantitative PREM surveys (despite many authors acknowledging that clinical and ward staff generally find qualitative comments more insightful).

• Qualitative data were acknowledged to be more difficult to use in terms of time and expertise.

• Data were used to identify small incremental service changes that did not require a change in clinician behaviour.

• Recording of the changes that were made or the impact that they had was poor.

• None of the studies reported using ‘formal QI methods of data collection’. 23 [Arguably, EBCD and PFCC are themselves formal QI methods, but the review authors do not include EBCD and PFCC in their list of examples (total quality management, continuous quality improvement, business process reengineering, Lean and Six Sigma). As noted earlier, the process of data collection and the process of improvement are not always easily demarcated.]

• Most studies used qualitative rather than quantitative methods to measure the impact of PREMs on QI.

• Two studies which measured post-intervention PREM questionnaire results found no statistical improvement in experience.

• EBCD appeared to generate more improvement efforts than questionnaire-based data, but ‘effects of the QI interventions were not measured or reported on’. 23

• In many cases, staff reported using data not to only to identify areas for improvement but to support or provide a rationale for existing improvement projects.

Some studies have identified a mismatch between managerial expectations and how engaged and supported clinicians and other front-line team members feel to make improvements. For example, in a survey of hospital clinicians, 90.4% of respondents believed that improving patient satisfaction with their experience of hospitalisation was achievable, but only 9.2% thought that their department had a structured plan to do this. 44 Friedberg et al. 45 found that physicians’ use of patient experience reports was variable and, importantly, that little training in communication skills was provided, even though improving communication with patients was thought to be fundamental to the provision of person-centred care. These findings suggest that not enough is being done to make patient experience data available in a useful, accessible and credible way to clinical staff, and to empower them – with positive organisational support – to see improving patient experience as a priority that they can lead on.

Flott et al. 46 outline a familiar set of challenges in using feedback (in particular national survey data) at clinical level. These include:

• scepticism about the quality of the data

• lack of training in social research methods

• statistical complexity and lack of technical guidance

• aggregation of data at trust level that does not inspire local clinical ownership

• isolated data that are not linked to other relevant data sources

• contradictory results from different sources depending on how data are collected

• disparities in external support to providers to help analyse feedback and how to drive improvement

Much of the research into using patient feedback for service improvement has focused on surveys of inpatient hospital care, but Burt et al. ’s47 study of improving patient experience in primary care confirms many of the same themes. They note that general practitioners (GPs) are positive about the concept of patient feedback, but struggle to engage with it and make changes under current approaches to measurement. Within practices, and in out-of-hours settings:

staff neither believed nor trusted patient surveys. Concerns were expressed about their validity and reliability and of the likely representativeness of respondents. Staff expressed a preference for free-text comments as they provided more tangible, actionable data.

Burt et al. 47

The authors conclude that supporting primary care staff to enable them to act on patient feedback remains a key challenge, and that surveys are necessary but not sufficient to generate meaningful data for improvement.

A recent study by Sheard et al. 48 has proposed a conceptual ‘patient feedback response framework’ to understand why front-line staff may find it difficult to respond to feedback. This has three components:

• normative legitimacy (i.e. staff members express a personal belief in the importance of responding to patient feedback and a desire to act)

• structural legitimacy (i.e. staff perceive that they have sufficient ownership, autonomy and resource available to establish a coherent plan of action in response to patient feedback)

• organisational readiness (i.e. the capacity for interdepartmental working and collaboration at meso level, and senior hospital management/organisational support for staff to work on improvement).

The authors found that, although normative legitimacy was present in most (but not all) of their 17 case study wards, structural legitimacy and organisational readiness were more problematic. Even where staff expressed strong belief in the importance of listening to patient feedback, they did not always have confidence in their ability or freedom to enact change. A lack of organisational readiness and support could block change even when staff expressed high levels of structural legitimacy.

The original ‘theory of change’ underlying this study [Understanding how Front-line Staff use Patient Experience Data for Service Improvement (US-PEx)] was that high-level organisational support is necessary but not sufficient for person-centred service improvement; that many experiences that matter most to patients happen in front-line encounters; and that bottom-up engagement in person-centred improvement (as opposed to top-down, managerially driven initiatives) can be motivating for front-line staff,13 consistent with evidence that patient experience seems to be better in wards with motivated staff. 39 Sheard et al. ’s48 patient feedback response framework has provided us with an additional theoretical lens through which to view our findings.

Our study was focused on patient experience within a ward setting. However, shortfalls in patient experience often happen at the boundaries between services, for example between mental and physical health, child and adult services, adult and older people’s services, and primary and acute care. A major focus of integrated care systems is to improve care at these boundaries and even eliminate them altogether. In addition to a bottom-up/top-down perspective, therefore, we may need to add a third, lateral perspective of local but external clinical challenge. Multidisciplinary clinical teams working to improve integration of care, for example in areas such as frailty, may mutually challenge each other’s practice, culture and organisation.

Wider literature on quality improvement theory, methods and skills

There is, of course, an extensive broader literature on change in health care and what enables or frustrates change at both individual clinician and organisational levels, which in turn has fed into a more specific but still substantial literature on QI and ‘improvement science’. 49

Approaches such as normalisation process theory50 and the theoretical domains framework51 have drawn on sociological and psychological theory to develop explanations of how individuals within organisations change behaviour and adopt innovations, what motivates and deters them, what skills they require, who they listen to, and how they act collaboratively to make sense of52 and embed new practices. From the organisational change literature in health care, we know that certain common themes recur, particularly around the importance of a receptive context and an understanding of organisational history, of having both senior management and distributed leadership, of the engagement of opinion leaders and clinicians, of ‘fit’ with existing organisational goals and strategies, and of organisational support and resources. 53–56

From a human factors perspective, a daunting list of 760 challenges to the delivery of effective, high-quality and safe health care has been identified by Hignett et al. 57 The authors group these under eight headings, which, again, are remarkably consistent with other organisational studies:

• organisational culture (26.4%)

• staff numbers and competency (20.5%)

• pressure at work (19.4%)

• risk management culture (10.8%)

• communication (10.5%)

• resources (6.4%)

• finance/budget (3.6%)

• patient complexity (2.4%).

It is beyond the scope of this report to rehearse all of this wider organisational change literature, but it is clear that QI generally and patient experience improvement projects specifically are in most respects no different from other change initiatives, and that they need to pay attention to senior leadership; clinical and front-line engagement and motivation; dedicated project management support and skills; the role and attitudes of local opinion leaders/advocates for change; and the receptiveness of the context and organisational readiness to change. In a useful recent reflection on the state of the organisational change literature in health care, Fitzgerald and McDermott58 conclude that there is remarkably little evidence to support top-down, large-scale, transformational ‘big bang’ change of the kind exemplified by business process re-engineering, and that it remains ‘unproven rhetoric’. Strategies of incremental, accumulative change are, they argue, more likely to have long term impact.

Jones et al. 59 have recently constructed a measure of ‘QI governance maturity’ from working with 15 hospital boards. Consistent with wider organisational literature, they conclude that boards with higher levels of QI maturity had the following characteristics, which were particularly enabled by board-level clinical leaders:

• explicitly prioritising QI

• balancing short-term external priorities with long-term internal investment in QI

• using data for QI, not just for quality assurance

• engaging staff and patients in QI

• encouraging a culture of continuous improvement.

It is commonly argued that QI is often ineffective when isolated in organisations and that it needs to be part of wider systemic efforts. Personal and organisational membership of formal and informal networks has been cited as a factor in effective QI in a Health Foundation60 study. The mechanism(s) through which this is achieved are difficult to identify precisely, although non-hierarchical and multidisciplinary exchange of ideas may be part of the answer.

‘Improvement science’ is a contested term. Walshe61 points to the dangers of ‘pseudoinnovation’:

. . . the repeated presentation of an essentially similar set of QI ideas and methods under different names and terminologies.

Walshe61

Improving health-care services can all too easily become equated with the use of certain ‘in vogue’ tools for improving quality. Advocates for different approaches will argue strongly that their way is the best (or even the only) way. Staff may feel overwhelmed by the array of methods promising a sure recipe for success, and concerned they cannot live up to the examples set by leading organisations in the field. As Walshe61 argues, QI needs sustained investment and support, but the rapid switching from one fashionable method to another has probably damaged the chances that any of these methods will be effective.

A recent King’s Fund report62 suggests that it is important for hospitals to adopt an established method for QI, one that is ‘modern and scientifically grounded’, and to ensure that all leaders and staff are trained in it. While using one consistent method may have advantages, the evidence behind most QI approaches remains fairly weak. Cribb63 argues that it is easier to identify a gap or shortfall in current practice than to identify, with any certainty, what specific approaches are likely to improve the situation.

In a systematic narrative review for Healthcare Improvement Scotland, Powell et al. 64 argue that:

Importantly, there is no one right method or approach that emerges above the others as the most effective.

Powell et al. ,64 p. 7

Consistent with the ‘contingency theory’ of management, which suggests that there is no one best or universal way to manage businesses or hospitals,65 the authors conclude that:

The specific approach (or combination of approaches) may be less important than the thoughtful consideration of the match and ‘best fit’ (however imperfect) for the particular circumstances in the local organisations using it.

Powell et al. ,64 p. 63

Most approaches have something to offer and will work some of the time in some settings, but are also likely to fail if good conditions are not in place. Powell et al. 64 identify the following as ‘necessary but not sufficient conditions’, regardless of the approach adopted:

• provision of the practical and human resources to enable QI

• the active engagement of health professionals, especially doctors

• sustained managerial focus and attention; the use of multifaceted interventions

• co-ordinated action at all levels of the health-care system

• substantial investment in training and development

• the availability of robust and timely data through supported information technology (IT) systems.

Lucas and Nacer66 also caution against becoming too focused on tools and techniques, arguing that:

Education for improvement practices can all too easily be reduced to, for example, ‘how to use a driver diagram’ or ‘how to lead an improvement project’.

Lucas and Nacer,66 p. 12

While Lucas and Nacer66 agree that learning such techniques is helpful, they continue that:

Without a clearer picture of what improvement really looks and feels like, the ‘packaging’ of improvement can end up becoming its lived reality.

Lucas and Nacer,66 p. 12

By contrast, ‘improvers are constantly curious, wondering if there is a better way of doing something’ (p. 12). They use a kind of ‘smart common sense’66 to keep reflecting (quotations reproduced with permission from The Health Foundation66).

Finally, we note Cribb’s63 argument for greater dialogue between what might be traditionally regarded as QI research, with its instrumental and applied focus on problem-solving, and wider social science in health care, which might not be intentionally focused on improvement but might, nonetheless, bring useful insights.

Conclusion

In summary, the literature on using patient experience data for improvement is incomplete. The focus to date has been primarily on quantitative survey data and their use at board level; much less is known about the use of other types of formal and informal data, and about their use at the front line. Surveys are important as measures, but their ability to provide a nuanced guide to understanding why patient experience is as it is, and how it could be improved, is limited. Yet other forms of data and ‘soft’ intelligence may be difficult for staff to analyse and use effectively. There is little reliable evidence from the wider QI and ‘improvement science’ literature to suggest that one approach is definitively better than the others. A thoughtful, eclectic approach that is sensitive to context and engages front-line and clinical staff at the same time as offering strong senior leadership may be more important than a particular technique.

This study set out to add to this evidence by exploring how front-line hospital ward teams engage with patient experience data when encouraged to do so, what challenges they face and how they can be better supported to work on patient-centred QI.

Phase 1: secondary analysis of existing survey data and new survey of trust patient experience leads

In this section, we describe a secondary analysis of existing national patient and staff experience data, and a new survey of patient experience leads. The aim of this phase of the study was to select the six case study sites (acute trusts) to take part in this research.

Phase 2: case studies in six medical wards

Once six front-line medical wards had been selected and R&D approval had been obtained, the study moved to phase 2, a mixed-methods case study approach. This had several steps:

• a baseline patient experience survey and interviews in each site

• preparation of resources and formation of a learning community to support front-line teams to plan patient-centred QI work

• qualitative ethnographic observation of the front-line teams’ QI work

• follow-up patient experience survey and interviews.

We describe each of these steps below and, where appropriate, reflect on how our methods evolved and adapted in real-world NHS settings.

Funding of £6000 was given to each NHS trust to allow staff to take part in research-related activities (such as attending the learning community and being interviewed). The cost of time spent on QI work was expected to be borne by the trust as part of routine commitment to improving patient experience.

Phase 3: producing guidance for the NHS

Phase 3 was the preparation of an online toolkit for NHS staff on understanding and using patient experience data for QI. During the lifetime of the study, research into stakeholder perspectives on toolkits from health-care research was being undertaken as part of a Doctor of Philosophy (PhD) project at the University of Manchester,82 and early findings informed the methods adopted for phase 3. In particular, it was recommended that toolkits were more likely to have impact if they were produced and branded by a recognised source. As a result, the decision was taken not to produce the toolkit in-house, but to commission The Point of Care Foundation to produce and host it, alongside their existing and widely used toolkits on EBCD and PFCC. The core of the toolkit was the content of the resource book already produced for sites in phase 2. This was revised and edited to include findings from the US-PEx study, and illustrative films from staff at some of the participating sites. Those interviewed for the toolkit gave separate consent.

Who were the lay panel and how were they recruited?

We advertised for lay members through a range of avenues to include as diverse a mix as possible in terms of age, ethnicity, geography, health condition and type of experience of inpatient care. We particularly encouraged people who had been involved in previous service improvement work to apply, as well as those with no improvement experience. As with many patient and public involvement (PPI) groups, the panel was lacking in those aged under 30 years; however, given the subject of study, we felt that this was not of particular concern as older people tend to have more hospital experience.

The lay panel comprised 10 lay advisors (see www.journalslibrary.nihr.ac.uk/programmes/hsdr/1415606/#/; accessed 17 April 2019), all of whom had experience of NHS inpatient care (including surgical) either as a patient or as a carer. It was expected that the motivation to become part of the panel would be from having had negative experiences; however, this was not universally the case. One panel member said of his time in hospital:

My experience was a very positive one, but I was motivated to get more involved in measures aimed at improving health care and, in particular, the patient experience.

This positive attitude was reflected by others in the group. However, there were, as expected, many stories of poor experiences while in hospital. Nonetheless, the consensus was a positive one, motivated by a wish to support service improvement, rather than to complain, irrespective of whether the experience was positive or negative.

I didn’t know anything about research at the start, but I knew a lot about being a patient and this is what I tried to put forward.

Lay panel member

Who was the chairperson and what was her role?

Lay co-investigator Jennifer Bostock chaired the panel, feeding back PPI matters directly to the principal investigator and/or to the wider research team at project meetings. Jennifer Bostock was involved from the outset, contributing to the study design, drafting the funding application and reviewing the study protocol. At the project start, she reviewed all study documents and consent procedures and advised on case study data collection methods to ensure that the study was ethical and acceptable. She led the appointment of the panel and strove to have a mix of those experienced in research and those new to it in order to ensure a balanced perspective.

What was our role in the study?

The initial recruitment of lay panel members was on the basis of the following role description.

1. Advise and guide the research team and ensure the study addresses the right issues.

2. Contribute our perspective to the first learning community.

3. Attend future 1-day learning community meetings.

4. Provide ideas about what we might include in a trigger film and comment on a draft version.

5. Help provide a sense-check on emerging findings.

6. Other roles to be developed in discussion.

Relationship with the study team: lay panel view

I was surprised that so many of us were recruited and how well our input was listened to. I felt this demonstrated a real motivation of the team to listen to and enact on the patient voice.

Lay panel member

The above sentiment was generally shared by all panel members, with another remarking:

The relationship between the lay panel and the project staff was strong. Personally I felt listened to and any comments taken seriously. I was impressed by the thoughtful feedback we received from the project leader and a willingness to really engage with the lay panel.

To sum up, one panellist simply said:

It was all beautifully run . . . and it was a privilege to be involved.

Over the course of the study, some tensions emerged, particularly around distinguishing between the panel’s role in advising the research and a wish to advise the front-line teams directly (see The temptation to steer the ward teams). These tensions were addressed through discussion with the principal investigator.

Practicalities of running the lay panel

The panel met on a regular basis, together with members of the study team and the principal investigator. They also contacted each other and the principal investigator by e-mail between meetings, and were supported by lay chairperson Jennifer Bostock and the project co-ordinator. Honoraria, travel and subsistence were offered. Jennifer Bostock provided ongoing support to less experienced and/or less confident advisers to ensure that they were all able to contribute effectively.

The size of the panel

The panel may have seemed dauntingly large to the clinical teams and an agreed approach to our work and smaller panel may have been useful.

Lay panel member

The lay panel was larger than the original plan of eight members, and although this demonstrates enthusiasm from members of the public for studies of this kind, it did present some challenges. As one member said:

On reflection, perhaps the lay panel was too large . . .? A panel of perhaps 4–5 would have been adequate.

The breadth and diversity of experience and the passion for the subject meant that, at times, the size of the panel constrained its ability to reach consensus, and during the learning communities the panel size may have felt ‘daunting’ to the front-line teams. However, although the panel was a group working towards a common goal, it would be unrealistic in a study of this nature to expect total agreement, and that reflects the individual differences, experiences and attitudes within the membership. It must be recognised that a group of individuals brought together because of significant and often extreme experiences will, at some level, always remain individuals.

Advising the US-PEx study team

The panel advised the study team throughout the research; Box 1 lists some of the key impacts the principal investigator felt that the panel had had on the study. The list was discussed and added to at the last panel meeting.

The following examples are taken from the notes one panellist made:

We discussed the powerfulness of a trigger film, composed of interviews with patients discussing how they felt about their care as in-patients, and commended its use to be shown to clinical staff. We discussed the pros and cons for different timings for doing such interviews, bearing in mind the fear some patients may have of criticizing when they are still in-patients.

We discussed aspects of care that were important to us, including admission and discharge procedures and the helpfulness of information being written down as well as verbal.

We discussed ways in which clinical staff might be able to encourage ex-patients to volunteer to give feedback, and what was the best timing for this from the patients’ point of view.

Patient and public involvement at ward team level: aims and realities

Why is patient and public involvement important in a study like this and who is affected?

One of the ways in which PPI impacts on a study like US-PEx is that it affects the researchers as well as the research. 83 Research is not all about data, protocols, results and literature; it is also about the people who conduct it. What had not been anticipated at the beginning of the study was the way in which the lay panel would affect the study team, and also how involvement would affect the panellists themselves:

The involvement of patients, carers and the public is important as a resource for the project and as a sounding board within the project to focus on the things that matter to patients. There is also a motivating effect on lay panel participants to go on to join other initiatives, especially when the experience of participation is as good as it’s been on this project.

Lay panel member

This recognition, that PPI is beneficial to lay advisors as well as to research and researchers, struck a chord.

Another panel member spoke of how being involved in the study had changed her:

I can’t tell you how much I’ve changed thanks to being involved in the project. My confidence is through the roof . . . got a new motorised wheelchair which has given me my independence back. I’m doing a course in research and I’ve just got a job interview.

The same panellist spoke of how on, a personal level, the study has had an impact on how she treats clinical staff when she goes to hospital:

Before I was rather quiet, or a bit hostile towards them, after what happened to me I thought all doctors and nurses were like that, but having met and worked with them on this project I see that they are not. I’ve heard the problems they have to put up with, the pressures they deal with every day, I wouldn’t want their jobs . . . so now I am nice to the clinical staff and they are actually a lot nicer to me now. I’ve even nominated my incontinence team for an award which they think is brilliant.

Lay panel member

What now for the lay panel?

Being a part of this study has taught the lay panel many things and it is hoped that the lay panel has in turn imparted insights to the study team and the case study participants. The panel members have learned how to work in a team, how to encourage rather than dictate, how to see things from the perspective of staff as well as patients, and the benefit of exercising diplomacy when trying to effect change. Perhaps most importantly, they have learned that no matter how disappointing health-care practice may seem, it is important to understand the reasons behind this; there is always a reason and there is always hope for improvement. They have a significant role to play in achieving this through being both part of the data and part of the solution. These lessons and many others will be taken forward in work with other researchers and, most importantly, in roles as patients, carers and health-care improvement advocates.

Secondary analysis of existing national patient and staff experience data

Results from the inpatient and staff survey data analysis were used to construct a matrix for the case study selection, according to whether the trust’s average score fell in the top, middle or bottom third of the distribution for inpatient and staff survey results (Table 8).

In addition, a spreadsheet was created that showed the combined results of the analysis of existing data sources for each trust. To help with the selection of case study sites, colour coding was used to easily identify those trusts in the top, middle and bottom thirds of the distribution when sorted by overall inpatient/staff survey ranks.

NHS Friends and Family Test

The average response rate to the FFT across August, September and October 2015 was calculated for each trust and included in the spreadsheet. The average response rate ranged from 6.5% to 59.9% across trusts.

Care Opinion

The number of staff registered to use Care Opinion per trust ranged from 0 to 66. Seventeen trusts had no staff registered to use Care Opinion and only four trusts had over 17 users registered.

The proportion of Care Opinion stories per trust that were read by staff ranged from 21.6% to 100%, with only 25 trusts out of 154 having read 100% of stories. The proportion of stories responded to also ranged from 21.6% to 100%; only six trusts had responded to 100% of stories.

The percentage of stories resulting in a change (planned or made) by the trust ranged from 0% to 7% across trusts; the majority of trusts (n = 111) had not reported that they had planned or made any changes as a result of the stories.

Patient experience leads survey

The survey of patient experience leads (n = 153) was completed (or partially completed) by 60 trusts, a response rate of 39%.

Respondents were given the opportunity to opt out of the survey at three points during completion. Respondents wanting to opt out of the survey were given the choice of keeping or discarding their previous responses. Two respondents opted out of the survey: one wanted their responses to be submitted and the other wanted their responses to be discarded.

The results are based on responses from 57 trusts as three cases had to be removed (two were partial, i.e. no response data were submitted, and the third was the respondent who opted out and requested that their responses be discarded). Given that only 39% of hospital NHS trusts responded to the survey, some caution is needed when interpreting the results as it is possible that those who responded hold stronger views about patient experience data.

Of the remaining trusts, 37 (65%) expressed a willingness to take part in the second phase of the study by leaving their contact details.

Headline findings

Case study site selection

The core research team reviewed the results of the secondary analysis and key results from the patient experience leads survey and then met to discuss potential case study sites. Initial discussion established which trusts might be included or excluded based on factors such as the following:

• Trusts would be considered regardless of whether or not they were deemed ‘improvement ready’.

• Trusts would be considered regardless of whether or not they had completed the patient experience leads survey.

• Trusts in special measures would be excluded.

• Trusts without a QI team would be considered.

• The demographics and geographical location of the trust would be considered (to achieve diversity).

The team shortlisted 14 trusts based on four main factors:

1. The list comprised approximately equal numbers of trusts that were in the top, middle and bottom third based on the secondary analysis matrix.

2. The list factored in a diverse geographical spread.

3. The list included some trusts that scored more highly on the inpatient survey than on the staff survey and vice versa (i.e. mixed).

4. The list comprised trusts that had expressed willingness to take part.

From this longlist, six trusts were identified as the preferred choices. The six trusts ranged in their inpatient and staff survey ranks, and also in the extent to which they appeared to approach and respond to issues relating to patient experience (see Table 1).

In February 2016, an e-mail was sent to the six trusts inviting them to participate in the study. The invitation described what the trust would be required to do as a participating site, including giving active senior support to the project, recruiting a front-line medical ward team keen to take part, and engaging with an ethnographic researcher who would observe the team’s progress. In addition to this, the benefits from taking part were highlighted, such as receiving expert input from patient experience researchers on the research team, having ongoing improvement science and learning community support, and supporting the development of a toolkit for the rest of the NHS based on learning from the project.

Five of the original six sites contacted expressed an interest in being involved. One trust reported that it did not have the capacity to become involved in the research project. Therefore, the team selected a replacement trust with a similar profile from the secondary analysis, and this trust was very keen to participate in the research (see Table 2).

Our initial aim was to carry out the research on general medical wards in each of the chosen sites. The rationale for this was that, typically, patients on these wards have a longer length of stay than those on surgical wards, where they may be in for very short periods undergoing routine elective procedures. In addition, medical wards tend to have a higher throughput of patients than specialist wards, an important factor to consider when thinking about achieving adequate sample sizes for the baseline and post-intervention patient experience surveys.

However, from discussions with sites it became clear that there was an appetite for specialist medical units to be given the opportunity to participate. For example, one site selected operated only specialist wards and another site was keen to focus on its emergency admissions and discharge unit. The research team (including lay co-investigator JB) considered the benefits and potential issues of each ward selection in turn and in all cases decided that sites proposing a different type of medical ward for the case study should not be excluded. Indeed, it was felt that the different settings and populations would reveal different challenges to the use of patient experience data (see Chapter 3, Patient and public panel methods and reflections).

As an example, the trust operating only specialist wards demonstrated great enthusiasm for being involved and felt that it could easily find wards that would be keen to take part. To understand how the selected specialist ward compared with other wards in the study, and to be reassured that a large enough sample could be drawn for the patient experience survey, we asked the trust to provide information such as length of stay data and number of discharges per month for the ward. While the research team understood that this ward might have a very different case mix from others in the study, we felt that this could be a benefit to the research.

Case descriptions and anonymity

As noted in the results of phase 1, the six sites selected included a range of types of ward, including two general medical wards, a gastric medicine ward, two emergency medical assessment units, and a longer-term rehabilitation medical ward. The sites were geographically diverse, including both urban and more mixed ‘town and country’ catchments, and covering north, south, east, west and midlands locations. All were in England.

Appendix 1 contains an anonymised description of each case study site, including the nature of the trust and the ward concerned, what their previous experience had been of using patient feedback, who was involved and why they took part, other relevant contextual factors, what they chose to work on and how, what was achieved and what challenges were encountered (Figures 2–7).

FIGURE 2.

Case study 1 summary.

FIGURE 3.

Case study 2 summary.

FIGURE 4.

Case study 3 summary.

FIGURE 5.

Case study 4 summary.

FIGURE 6.

Case study 5 summary.

FIGURE 7.

Case study 6 summary.

‘What was achieved’ is described both in terms of findings from the baseline and follow-up survey and interviews, and from the ethnographic material gathered throughout the study. These sources of data provide different and equally important lenses through which to view each case.

The primary aims of the case studies were to understand:

• how front-line teams use different types of patient experience data

• what could make such data more credible and useful for front-line staff.

Our research therefore focused mainly on what staff understood by patient experience data for QI, how and why they used these data (or not), what they found helpful or challenging, how the wider context affected what they did, and what support they needed. Measuring the number of changes made and any impact on reported patient experience was not the main goal of the study, although it forms part of the picture.

It is undoubtedly the case that some sites went further with their QI work than others, but our intention is not to rank them in terms of ‘success’. The sites were selected to reflect a range of previous levels of experience and confidence in using patient experience data for improvement, and so it was never expected that they would all move at the same rate. Any suggestions made about what might have helped and what teams or organisations could have done differently should be seen against this background. It should also be noted that learning and cultural change may occur independently of any specific service changes.

To give staff confidence that they would not be judged publicly and could share with us difficulties as well as success, our findings are anonymised throughout. The summarised case descriptions in Appendix 1 draw on the detailed site-by-site descriptions developed by ethnographic team members as part of the analytic process. These contained a considerable amount of sensitive and identifiable information that we do not present in the report. Our findings on major themes in the data are organised thematically rather than case by case. In the more in-depth thematic chapters (see Chapters 6–8), the roles of people interviewed or observed are identified, but they are not linked back to specific sites in case this inadvertently identifies them.

Reflections on case descriptions and summarised findings

What ‘data’ did staff use and how did they use these?

Consistent with previous research (see Chapter 1), the study provides further evidence that it is not self-evident to front-line staff how to use patient experience data for improvement, or even what ‘patient experience data’ means.

Despite an emphasis on making use of existing data, teams were not always aware of sources of information they could draw on within the trust. Even if data are available and known about, they often do not provide sufficient locally relevant information. Several sites set out to collect (or use) new data as both more meaningful and more motivating.

Survey data

An expression of this was the extent to which ward teams received the baseline experience survey results as timely, relevant and actionable data. This ward-level survey was intended not for this purpose, but simply to provide a baseline measure against which to assess any measurable change in patient experience after the teams’ QI work. The findings were shared with teams as part of a formative approach. However, in some sites they were welcomed as more locally meaningful and inspiring than many existing data, such as FFT results and national survey findings. The baseline results were accompanied by an infographic (Figure 8 provides a fictionalised example) highlighting the areas of most and least positive experiences on each ward. Some patient experience officers reported that staff found this data visualisation particularly engaging and clear as a guide to action.

FIGURE 8.

Fictionalised infographic on ward experiences.

Quotations from interviews are verbatim but edited for readability by removing repeated words, ums and ers. Ellipses presented in square brackets indicate that a more substantial portion of text has been removed.

I loved the way that you presented the ‘this is what your patients are saying about you’. You did it in a very pictorial format.

Was that the Picker Infographic?

Yeah, yeah. That was really good and it was eye-catching and it made people look at it because it was bright, it was easy reading and it was eye-catching, and I think that’s what we need more and more . . . So anything like that, keep it simple and because as I say, we need to get away from this, ‘this is going to take me a lot of time away from my day job.’ . . . I just loved it. It was really, really interesting, and interesting way of showing it rather than my graphs and things. People get bored and they get number-blind as well.

I think one of the best things that we’ve had to give them is the infographic from that survey, because it’s quick to look at. It’s easy to look at. It’s easy to digest. And straight away they could see the problem areas with the red areas. I think that’s how they need to have the data presented. Just giving them graphs and data doesn’t mean anything to them. They don’t want to know all that. They just want to know, ‘where do I need to concentrate and how do I do it?’ That is what they need. The infographic does that.

One ward manager regarded the baseline survey and infographic as an authoritative guide to what was most important to patients. She felt that it cut through the confusing number of data available from the patient experience office, and was a more legitimate pointer to action than staff’s own ideas about what needed improvement:

There was so much material. It was difficult to see the wood for the trees. The two that I found the most interesting were where they were categorised into about six or seven different aspects, and then marked in red. If we’d have had that information before the 2 days away [the first learning community] that would have been key ‘cause we had three clear issues that the patients felt that we could concentrate on, and no matter how much we wanted to do other things, we had to just say, ‘Actually this is what’s important to the patients, so let’s focus on these’. Because you could drown otherwise . . . ‘Let’s not go off-piste, let’s concentrate on these three and do them really well.’

Ward manager

In another site, the patient experience officer involved in the team reported finding it hard to discern areas for improvement from existing data sources, and they returned to the baseline survey and accompanying patient experience interviews to identify new ideas:

So initially I did an analysis of all the data I could find, through our real-time feedback. Which didn’t end up being very useful, because we have FFT scores and we have comments, but I think too many of them are very positive. Which is nice, but didn’t offer any room for improvement, really. And I think some of that’s due to the fact that we know when you ask patients while they’re still in the building they give you a different response to once they’ve left and had time to reflect. So that didn’t really yield very much. Then we tried to do patient diaries. That didn’t really work very well. And then we I suppose had a closer look at the Picker survey and the interviews you’d done [baseline interviews with patients]. And that’s when we pulled out actually maybe there is something we can do about danger signals to look out for. And maybe there is something we can do to look at the discharge process. So I suppose the Picker survey and then some of the sort of comments from your interviews really ended up being the data that we actually acted on.

Patient experience officer

In other cases, the data from the survey chimed with, and confirmed, what staff had already identified as issues from other data or from their own experience. As one head of patient experience said:

We had the issue coming up about patients not always feeling informed enough about their discharge. And some of that had come through the Picker survey as well.

Comments boards

The boundary between collecting data and using them for improvement is not always clear. For example, some sites collected patient and family comments using a ‘bubble board’ or a ‘what matters to me tree’, where people could write feedback on a thought bubble or a leaf (Figures 9 and 10). Although at one level this was data collection, it was also experienced as a cultural shift towards more patient-centred care, a visible expression of commitment to the goal of improving experience. Taking action on specific suggestions was only part of the story:

We had one lady that, she’d had some really sad news, and just sitting with her for 5 minutes, chatting to her; she didn’t want to chat about what she was upset about. So, just chatting to her and saying, ‘There’s a bubble there for you if you want to write anything down,’ and she just wrote a little something. I think it was just a face with a little sad mouth, and that was enough for her. So, I think it just takes something off of someone’s shoulders . . . One lady wrote a whole bubble just on the porridge. She really enjoyed the porridge, and it was the best porridge that she had and it made her feel happy, so we put it up on the wall; we showed her it was up there and she was happy that her comment was up there . . . Since we’ve been doing this, people are recognising things I think a little bit more, and I think that’s what’s changing; not the bubbles on the board and things like that. It’s recognition that somebody needs to have a good experience.

Health-care assistant

FIGURE 9.

‘Bubble board’.

FIGURE 10.

‘What matters to me tree’.

In discussing their comments board, one ward manager explained how it seemed more tangible and meaningful to staff at all levels than, for example, traditional audit data. The physical data collection artefact became itself a motivator, an emotional and even enjoyable representation of care:

We have an awful lot of audits, and they’re important, you know. We know we have a lot of things that challenge the quality of the care we give, or that, you know, we have to write reports on, that the other staff find . . . I think it goes over their head a lot of the time, you know, they can’t relate to it. Whereas this, from whatever band you are, this you can relate to.

So, what’s different about this; what is it about this that grabbed people in the way it has?

Well, it’s more to do with how patients experience, how emotionally they feel about the ward, rather than ‘Were their intravenous antibiotics given on time? How many days length of stay? Was that extended?’ You know, it’s about how they feel about it. We get thank you cards, but to see it on a tree when you come in the ward, I just thoroughly enjoyed it.

This sense that comments boards yielded more meaningful information was echoed among staff at lower band levels. One health-care assistant commented ‘the data that we get from that is quite real’. Others noted that being encouraged to think about patient experience data had changed their thinking:

Do you think some of the improvements that have been made on the ward could have been made without using the patient experience data? How much do you think it’s depended on patient experience?

I think a few of them obviously we could have done, but would we have thought about them without the patient experience? I mean with things like the hearing aids going missing. We just became used to it. It’s just become part of our normal daily routine; complaints about this. It’s just part of the normal daily thing. So we’ve never thought about it before as to how can we change this. This Oxford project has opened us up far more, I think, to suggestions and to improvements. There’s always been the capability of doing it. But unless we actually have the thought of doing it in the first place, it’s not going to happen, is it?

So what do you think is the biggest of success so far with this project?

I think it’s bringing everybody together as a team and getting them to see that patient experience matters . . . We know that, but actually doing something about it is different . . . I think people are recognising that it’s not just about caring for somebody; it’s about making their experience good.

When you do your training it’s all very regimental – this is what you do, this is how you do it. And then when you come to work it’s the same. So, you come in, you do your washes, you do your meds, do your pills, you do beds, you do this, you do that, and then you do your notes and then you do it all over again. It almost seems like there’s no time for anything else . . . But actually there is, and it’s just finding the time . . . ‘We’ve got half an hour here, let me sit down, let me have a chat with the patients, see how they’re getting on, do they want to give us any feedback, is there anything that we could change, have they done their menu, do they like the food, like do they need to speak to somebody?’ It is just sitting down and having that extra 5 minutes that they appreciate . . . It does make you more conscious. It’s less task-focused and more patient-focused, which is what the project is about.

In two sites, staff were also invited to record ideas for change on a staff suggestions board. Some team members felt that this played an important part in generating culture change and motivating and empowering staff by giving them a sense that their experience mattered too:

The board that we had up at the beginning was ‘What ideas have you got?’ They’ve actually seen them happen. ‘What would you like to do?’ And they’re thinking, ‘Actually, well, you know, we have got some power here’. Because sometimes, particularly for the lower grades it feels sometimes that you’re a hamster on a wheel and your views aren’t always taken into account.

Ward manager

I think that came from the need for [ward manager] to try and involve all the staff in that ward. And I think she was right. They had been through a very bad time on that ward. And I think the staff really appreciated that they felt some sort of ownership. They could see that the senior staff were serious when they said, ‘We are going to make improvements on this ward’. They could see it in action. And they could see that they were being invited to take part. So I think that was a very good move on . . . And to be able to do those quick wins with that. She had done things straight away.

Patient experience officer

The staff suggestions board generated a few comments that were challenged or removed, either because they were insufficiently specific to inform change, or because the board was considered an inappropriate place to air grievances (e.g. about pay).

Observation and diaries

Staff generally did not seem to have previously considered observation as a source of patient experience data, but they found this appealing when introduced to it at the first learning community. Several experimented with observational exercises, including the pro forma in Figure 11.

FIGURE 11.

Example of pro forma.

Another site experimented with the ‘15 steps challenge’ approach:

[XXXX] explained that he had experimented with the observation model since returning from the first learning community. He explained that he had asked a volunteer (who visits the ward regularly) to do a 15 steps observation ‘from entering the hospital to sitting by a patient’s bed’. He described it as a ‘mystery shopper’ exercise in which ‘your brief is to be a patient for the day’. The volunteer was further advised (by XXXX) to ‘be as stupid as possible and ask as many questions as necessary’ . . . The volunteer provided verbal feedback that essentially comprised of the following detail/observations:

Ethnographic field note

The same site also tried patient diaries but found that these did not yield actionable improvement ideas because they were ‘too positive’. (Positive feedback, although welcomed, seemed rarely to be considered explicitly as a useful source for improvement, as we discuss further in The patient feedback silo and the role of the patient experience team.)

[T]hey trialled the diary on the ward with several patients. However, the information that they got back from these diaries was, once again, regarded as unsuitable and inappropriate in terms of quality improvement. In short, data they received here was considered ‘too positive’ and there was nothing critical on which the team could build.

Ethnographic field note

However, brief end-of-shift staff diaries were used successfully at another site (Figure 12), alongside the observations in Figure 11.

FIGURE 12.

Example of end-of-shift staff diary.

The patient feedback silo and the role of the patient experience team

Patient experience feedback (particularly unsolicited, in the form of NHS Choices or Care Opinion postings, but also solicited survey data) is often treated in a silo, managed at board level and handled separately from data about safety and other aspects of quality. Ward teams rarely seemed to access unsolicited data themselves. Most wards were routinely sent FFT data, but staff often reported having neither the time to look at them nor the skills to interpret them. Again this echoes the findings from the Phase 1 survey. In some sites, analysis was managed on wards’ behalf by the patient experience office.

Negative feedback was recognised as having valuable formative potential; staff reported that it was welcomed, and even sought in some cases, but that it was not always readily available. This may reflect a lack of connection with (for example) local teams responsible for complaints or Care Opinion data, and whether or not complaints are seen in the organisation as a resource for QI:

But our problem is we don’t really get that many bad comments. We get a lot more feedback saying about the wonderful staff and like the fantastic caring and the support that we’ve given to the families.

Ward clerk

At the same time, the potential for negative feedback to generate a defensive and resistant response was noted. One patient experience officer described how her team had had to work hard to demonstrate the value of listening to negative comments, particularly in the case of one critical narrative, which staff had felt to be unfair:

Speaking to some of the people in my team who were involved right from the start of doing the ‘real-time’ work on the wards, they would say there was a lot of resistance, a lot of suspicion. And then as we kind of demonstrated that it was kind of an open process, that it was actually helping to improve things on the wards, there was more acceptance and more wards wanted to join in the process. I found there was one ward in particular I used to hit a bit of a wall. I never felt that welcome at the start and there was always, ‘Oh, the doctors are doing their round’. There seemed to be excuses as to why I couldn’t be there and I used to get quite frustrated. But I have a really good relationship on that ward now. I think it’s about just bringing people round. I had an issue on a ward with a patient which was a really tricky one. A patient who’d had quite, not a very nice experience and I did a patient story around that which is where we go into more detail, with that patient’s consent, about what’s happened. And that story didn’t go down well at all on the ward and the staff got upset and I think they felt it wasn’t fair, that this was the patient’s perception of what had happened, and that they didn’t have the right to reply, and, ‘you’re only seeing it from the patient’s point of view’. But the point we try and make is actually, you know, your perception as a member of staff, you don’t understand the patient’s perception. That’s all we’re trying to do is to provide a mirror to that. And I was quite upset because I felt really kind of caught in the middle. But actually after a few days, the ward manager e-mailed me and said, ‘Actually, I know you’re just doing your job and you’ve given me food for thought and I’m going to try and change something as a result’.

Patient experience officer

An ethnographer working at this site observed how tensions between the patient experience officer and the ward manager played out and were resolved, partly through the intervention of a health-care assistant, which moved the debate on to a more positive footing:

[Health-care assistant] looked quite concerned and walked up and down the line reading each card and post it notes with real care and attention. He seemed genuinely concerned by the range of negative points on the wall. I listened in to what [ward manager] was saying in response to the waiting times. In short, she was not surprised by any of the comments on the wall and stated she was familiar with all the points raised. . . I could hear the frustration and possible anger in her voice [as she spoke to the patient experience team] . . . ‘We’ve tried this before, we had a dedicated person whose role was to respond only to call bells, but it didn’t work.’ Adding, ‘nothing in this project is going to combat that.’ [. . .] At this point [health-care assistant] chipped in and stated that he recognized these difficulties and that although they were negatives each point could be returned as a positive at the subsequent staff version of this event . . . [He] suggested that the comments about dignity and waiting could be fed back to staff to raise awareness of how staff talk to and treat patients. He believed that reflection on these issues will help recognize the problem.

Ethnographic field note

Positive comments were valued at ward level for boosting for staff morale:

I mean some of our cards are lovely, you know, they put some real detail in there, you know, and really personalise them, sometimes with names, certain staff’s names. Yeah it does give you a bit of a boost, it makes you feel like you’re doing something right . . . it’s a really positive experience, isn’t it, rather than a kick in the teeth [laughter] when you’re trying your best . . . It’s lovely, it’s just a small thing but it means a lot, yeah definitely.

Health-care assistant

However, their organisational potential for learning and improvement seemed under-recognised. At trust level, positive feedback on Care Opinion tends to be responded to with a ‘thank you’ to close the conversation, rather than exploring with wards how it could be used to identify what patients value and to improve care, or even as a performance indicator.

In the same way that patient experience data are siloed, we found that patient experience offices or teams also tended to be functionally separate from teams dealing with patient safety, QI, and/or complaints or patient advice and liaison services:

So you don’t have a huge amount of communication with them [QI team]?

No, no, not at all in fact, which is a bit of a shame, really. I know. I should probably do more around that really. It will be really interesting for me to know what are they actually even working on; I have no idea, no idea. Because then, we also have a service development team . . . And they have QI within their job titles as well. And they have been working with patient experience data because they’re doing a whole project on waiting times . . . purely off the back of patient experience data. So that’s been really positive. But they sit under HR [human resources]. It’s so hard to navigate.

With some notable exceptions (including one site in this study where the patient experience team were driving QI), patient experience as a function within many trusts has largely focused on data collection, ensuring that surveys are completed and FFT response rates are maximised, with a certain amount of passive feedback of findings to ward teams:

At a very kind of hands-on level, talking sort of nurses, HCAs [health-care assistants] kind of thing. How much engagement is there with patient experience do you think?

Oh that’s difficult. I don’t honestly know the answer to that. I mean I guess the sisters and charge nurses run that independently on their own ward. You get some that are far more engaged than others. But to be honest with you, because I’m one person on my own, I don’t know. I have to feed the information and hope that it’s reaching the right people basically . . . I go to things like the band 5, the band 6 development days, I go and talk at trust induction, loads of different places, but actually how much it’s filtering down – I would say that it is more, because I can see it in the FFT, as in I get so much more feedback now; the postcards are literally . . . can barely even keep up with them, we get so many in now. So I know they’re being handed out; I know that staff are engaged in that. But yeah, I don’t know honestly how much they’re involved in improvement work on the wards.

Thus, patient experience staff may not regard (or feel empowered to regard) QI on the basis of patient experience as within their remit. Meanwhile, QI teams often do not see patient experience as a particularly useful or valid source of information for QI. In some sites, this was beginning to change, and, in at least one site, this was partly because of participation in this study:

[Patient experience data] is new to the [QI] team. It was very much safety-focused before. And so everything was mainly quantitative. So, you know, it was very much about, you know, doing time trials of, you know, ‘we took Mr Potato Head apart and pulled him back together and we tried to do it quicker and quicker’. And my joke was, ‘but we didn’t ask Mr Potato Head how he felt’. So we might have got quicker, but it might not be better for the patient.

Head of patient experience

A member of the QI team was included in an initial set-up telephone call with this site but did not get closely involved, seeing it as a project primarily for the patient experience team. By the end of the study, however, the two teams were jointly reworking QI training materials to be more patient-centred:

[We’ve] looked at all the slides and we’ve rewritten their delivery and put in all the different patient experience methodologies, focus groups . . . We’ve adapted what they’re doing and everybody who’s taking a project [on QI] must have a patient experience component of it, otherwise why are they doing it? They can’t just improve safety without doing something about experience.

Head of patient experience

Even in cases where the patient experience team saw QI as definitely within their remit, front-line staff were not always aware of what support the patient experience team could offer. Realising that this was the case during the study prompted one team to change its title:

I was surprised to start off with that [ward manager] didn’t know what patient experience did and that we had those resources to actually help the wards do this improvement work, which makes me wonder, ‘wonder what everybody else thinks patient experience do’. So [head of patient experience] has changed our name to patient experience improvement team, which actually is a bit better. Perhaps it says a bit more on the tin of what we do. So that was surprising that she had no idea what we did.

Patient experience officer

Some teams found the practical support of their patient experience team invaluable in helping them make sense of data. Close involvement of the patient experience team – getting alongside ward teams and supporting them in using patient experience data – had a positive effect. Staff felt more confident and more motivated, and were more likely to take on a wider range of projects. Their work was more likely to be recognised and spread beyond the confines of the ward, and to be treated as an integral part of trust quality work.

In other sites we observed a degree of mutual misunderstanding between patient experience officers and front-line teams. One ward sister expressed an ambivalent view that greater input might have been practically helpful, but that, essentially, the front-line team could manage alone:

Yeah, I don’t think we’ve really needed it. I mean perhaps, just breaking down the feedback would be helpful maybe if [patient experience team member] could perhaps do that. But other than that, I don’t think we’ve really needed any assistance.

Ward sister

The relationship between front-line staff and the patient experience office is about more than practical data handling or project management; it reflects wider issues of power, trust or mistrust, and credibility. This relationship, and its impact on the work undertaken, is explored in more depth in Chapter 8.

Using ‘soft’ intelligence and staff experience as data

Stories, informal comments and the daily ward experiences of staff, patients and family are all important but often underutilised resources for QI.

In coding an interview with a core ward team member, one ethnographer noted:

There’s a really sad thing that [XXXX] says, which for me sums up the problem with the industry of patient experience data and quality improvement. She says ‘I hadn’t actually heard of patient experience until I went to the Learning Community’. What she means is that she hadn’t heard of patient experience as A THING; of course she knew about patients’ experiences – plural and embodied on the ward. This focus on data is so unhelpful! Instead, we should be talking about ‘listening to patients’ and ‘learning from patients’, which is probably what nurses and HCAs do anyway on a day-to-day basis.

Ethnographic coding note

Tangible, but informal, feedback from patients in the form of thank-you cards, gifts of chocolates or biscuits, or giving staff a hug before going home is virtually never treated or recorded as the expression of patient experience that it in fact represents.

There is a continued sense that data have to be countable (or formally recorded in some way) to be perceived as actionable. This may lead to a culture in which there is pressure to turn a minor comment into a complaint before it can be regarded as useful ‘data’ or to demonstrate that it is being treated seriously – and yet, as soon as a helpful, if mildly critical, comment becomes a complaint, it may raise defensive barriers.

We found that ward staff made use of multiple sources of soft intelligence but were not always aware that they were doing so. Staff could not always point to a specific source of patient experience ‘data’ that led to a particular improvement project. Sometimes they reported acting on what they felt they already knew needed changing as a result of caring for and observing people and talking to colleagues:

It [QI intervention] fits right, but I don’t think they’ve done a survey.

Sums up what I think is happening with a lot of the teams . . . the issues are known to staff and they are responding, without being able to identify any specific piece of patient feedback.

In one case, an ethnographer observed how a staff member was imaginatively channelling patient concerns through her use of language:

What strikes me about [XXXX’s] interview is how she speaks in the voice of patients and relatives (using the first person) when she’s narrating what’s needed or what’s happening on the ward. This happens at various points throughout the interview. She puts herself in their shoes. She uses direct speech, often as a dialogue between herself and the patient or relative . . . What this says to me is that her actions are based on empathising with patients and relatives rather than on data that has been extracted from them . . . Empathy and imagination should be positive resources in QI work! Suddenly data come to seem rather dry.

Ethnographic coding note

Staff gleaned insight into patient experience through their close interactions with patients and an experiential understanding of how the ward works. During the course of the study, and partly through discussion at learning community events, staff consciously expanded their interpretation of data (as did the research team and lay panel):

It depends how we look at data. I think in the ward staff before, if you said to them, ‘What’s patient experience data?’ they will say ‘Surveys’. I’m saying to them now data is any feedback at all, wherever that’s coming from and in whatever form, whether that’s coming from focus groups from the patients or anecdotal feedback from staff and patients, it’s all patient experience data . . . 9 out of 10 wards here, if you asked them what data was, they would say surveys: the National Survey, the FFT. Which I must admit when we started this project, the first thing we took along to the ward was all their survey results.

Patient Experience Officer

In this site, the ward manager introduced a suggestions board to allow all staff to contribute improvement ideas, and to show them that their views and insights were welcomed and sought:

I think that’s reinforced the feeling that, you know you need to remember that the staff are doing a job all the time. They have ideas that might improve the service.

. . .

Things like the hearing aid boxes I think came from the staff suggestion board, didn’t it, rather than from the patient experience data?

Yeah, all of that just came up because we could see the distress. Because like [ward clerk] is the person, the go-to person that they all go to when their hearing aids don’t work. She deals with a lot of the lost property, the distress of the patients that have lost an expensive hearing aid and they can’t communicate at a time when they most need to communicate clearly; the safety . . . That came from, you know, our own experience of what causes them distress.

In some cases, areas for improvement that staff identified could be ‘backed up’ by patient experience data, meeting a perceived need to justify the use of staff wisdom:

We could back those things up by other things that were being shown on the data. We could show from data that patients do better if they’ve got the same staff all the time and that the FFT positives all goes down if there is bank staff. So one of the things [ward manager] brought in was being able to offer ward staff the opportunity to fill vacant slots, rather than bringing in bank staff. So, we could, as I say, we could reinforce it with other data.

Patient experience officer

Conversely, ‘soft’ patient experience data were sometimes seen to require validation in the form of staff experience, as in the following case described by a matron:

I think the caveat with patients’ stories is, it really depends on how true you think that is for the rest of your department and that’s always difficult. But, actually, nurses on wards usually know if something has a ring of truth to it. They will hear that more, won’t they?

Matron

Where patient concerns are identified by staff rather than from patient experience data, there is a risk that staff perspectives are or continue to be privileged over patients’ perspectives when the two diverge. A matron acknowledged that staff perceptions may not be a reliable guide:

The challenge is there are several, many staff who work within the trust who will be completely unable to see the impact of their behaviour or their – let’s say – their brusqueness. No, if we have to tell people to say, ‘Hello, my name is . . .’, which we do. So that’s a real fine balance, isn’t it? ‘Cause there are people who won’t get that at all. They won’t see that that’s bad patient experience.

Matron

A head of patient experience raised concerns about not being able to demonstrate that improvement projects really did reflect patient preferences:

See, for me I’d want to say, ‘Well where’s the evidence? Where’s the evidence to substantiate that?’ and, ‘Yes, you might be right and I might agree with you’, or, ‘Yes, I agree with you and we can back this up because of this, or this, or this, or this’. But without any evidence, or something tangible to substantiate that, I think we can – not make mistakes – but we can go wrong in thinking we know what patients want.

Head of patient experience

However, another head of patient experience argued that there were limitations to both ‘hard’ and ‘soft’ data. Given that the views of some patients (e.g. people with communication difficulties or people living with dementia) can be systematically neglected by surveys, she felt that soft intelligence, held by the staff who work closely with them, can play a valuable role:

I think, you know, staff who are close to their patients can see the best. You know, this survey is very much reliant on people who can fill it in. So, actually, you’ve got lots of other patients that you’re perhaps not considering when you just take that data. And it’s also very lengthy. When you look at all the questions, you know, I am bored after 10 questions, let alone, I think there’s 87 or 89. Whereas some of the dementia patients might not be able to tell you, but you can see just by doing certain activities or things with them, that made a difference to them.

Head of patient experience

Discussion: collective sense-making and ‘mindlines’

In analysing our findings, we suggest that staff were engaged in a process of sense-making over time from a range of formal and informal sources of intelligence about patient experience. As described in the field of organisational studies,52,84 sense-making (drawing on symbolic interactionism)85 is a collective interpretive process of social interaction, whereby people arrive at a shared meaning that enables them to move to a course of action.

Ancona86 argues that sense-making:

. . . involves coming up with a plausible understanding – a map – of a shifting world; testing this map with others through data collection, action, and conversation; and then refining, or abandoning, the map depending on how credible it is.

Ancona86

For Weick,52 this perceived credibility or ‘reasonableness’ in planning a course of action is more important than seeking an objective justification. He argues that there is no single factual truth waiting to be discovered, but rather a range of perspectives that may be more or less useful in determining a way forward.

The concept of ‘clinical mindlines’ is a form of sense-making articulated by Gabbay and Le May. 87 Originally developed in the context of understanding how clinicians in general practice use research evidence to make care decisions, it can also be applied to the combined use of formal patient experience ‘data’ and soft intelligence. Gabbay and Le May87 summarise this as follows:

We found that clinicians rarely accessed, appraised, and used explicit evidence directly from research or other formal sources; rare exceptions were where they might consult such sources after dealing with a case that had particularly challenged them. Instead, they relied on what we have called ‘mindlines,’ collectively reinforced, internalised tacit guidelines, which were informed by brief reading, but mainly by their interactions with each other and with opinion leaders, patients . . . and by other sources of largely tacit knowledge that built on their early training and their own and their colleagues’ experience . . . Mindlines were therefore iteratively negotiated with a variety of key actors, often through a range of informal interactions in fluid communities of practice, interactions with and experience of patients, and practice meetings. The result was day to day practice based on socially constituted knowledge.

Gabbay and Le May. 87

Reproduced from Evidence based guidelines or collectively constructed “mindlines?” Ethnographic study of knowledge management in primary care. Gabbay J, Le May A. Vol. 329, pp. 1013–18, 2004 with permission from BMJ Publishing Group Ltd

This process resonates with the way in which ward teams in our study could not necessarily identify a linear process of specific pieces of patient experience ‘data’ prompting specific QI projects. Occasionally they could, but more often than not staff described a more informal and iterative process of drawing on clinical experience, observation, interaction and conversation – with each other and with patients and family members – and coalescing around priorities for improvement. Commonly, there is talk of how patient experience data can lead to ‘lightbulb’ moments in changing staff attitudes. Taking a ‘mindlines’ approach might suggest that there can also be a slower process of illumination, a gradual dawning rather than the flicking of a switch.

In some cases, taking part in this study provided a channel for staff to legitimise and act on these existing concerns, to make visible previously neglected problems and to create alliances for improvement. In effect, staff were seeking to create for themselves what Sheard et al. 48 describe as ‘structural legitimacy’: creating the authority and resources to take action, and making a case to higher levels of the organisation. This is an idea we develop further in Chapter 8.

At the same time, pressure at trust and national level to privilege formal, often quantitative, data remains strong. In another HSDR programme study funded under this call (HSDR 14/04/08: INQUIRE),31 an ethnographic analysis of how NHS trusts use online feedback for improvement suggests that data are more likely to be regarded as actionable for QI if they are structured, sought, solicited and/or sanctioned by the organisation. (‘Solicited’ feedback is where the trust invites feedback through channels such as Care Opinion or real-time feedback devices. This contrasts with ‘sought’ feedback, which is from sources that the trust does not actively direct people towards, such as Facebook and Twitter, but is sought out and used by the trust nonetheless.)

Equally, data that are unstructured, unsought, unsolicited and/or unsanctioned may be marginalised or disregarded, or they may simply never come to anyone’s attention. The fact that in our study Care Opinion stimulated so much early interest but failed to gain traction once staff were back in their trusts illustrates the impact of a lack of positive organisational sanction. Unintentionally, we may have promoted the use of survey data in making available at-a-glance summaries of baseline survey data.

These findings have led the research team to pose ourselves a number of questions:

• Does this mean that we define patient experience data too narrowly, and do we need to broaden the definition to encompass these informal sources?

• Do data have to come directly from patients to count as patient experience data, or can they come from the accretion of staff intelligence, multiple informal observations, discussion and embodied experience?

• Does it matter if staff harness patient experience data to legitimise action on what they already know (or believe they know)?

• How do we liberate staff creativity, empathy and action, but avoid relying on potentially mistaken assumptions without checking if these reflect real patient concerns?

Gabbay and Le May87 acknowledge that clinical mindlines ‘may seem to be a dangerous shortcut’, in that they may fail to engage with known research evidence. The same may be true of patient-centred improvement if it fails to engage with patients’ actual needs and preferences, and reinforces continued staff assumptions that they ‘know what their patients think’. The publication of numerous personal accounts by health-care professionals of revelatory experiences once they themselves become patients suggests that it may remain difficult for staff to fully appreciate the priorities and vulnerability of patients. 88

At the same time, we should not rule out the possibility that staff may indeed sometimes ‘know what their patients think’. Accumulated staff wisdom, informal observation and experience may sometimes be a more reliable guide to action than high-level survey data, or comments from a subset of patients who feel particularly strongly about something and who have the capacity and inclination to respond via sanctioned and solicited patient feedback systems.

The rhetoric of ‘putting the patient first’ is, of course, important, but it may carry with it an unfortunate implication that staff should always come second. Staff who feel valued and supported may feel more empowered and confident to engage with patient experience ‘data’. There is certainly a growing body of evidence of, at the very least, a close association between staff motivation/retention and positive patient experience, and most likely a direct causal relationship. 13–16,89 Chapter 7 reports further on improving staff experience as a strategy.

Gabbay and Le May’s87 recommendations for avoiding the ‘dangerous shortcut’ in use of research evidence can be adapted for the use of patient experience data. One key point is that individual practitioners ‘do not have the time (nor usually the skills) to rigorously review and combine all the key sources of tacit and explicit knowledge themselves’ but that key opinion leaders can be trained and supported to undertake this synthesising task – and to understand how to support professional networks and communities of practice, working with rather than against clinical mindlines.

This chimes with our findings that sites with close patient experience team support generally made more progress and were more ambitious in scope than those without. Patient experience officers are an emerging professional group whose role could develop into one of this supportive leadership, provided that they are trained not just to collect data and maximise survey response rates, but also to value and interpret a much broader range of insights. Done well, and perhaps in partnership with a more traditional QI function, this role can help liberate the curiosity of front-line staff, give them permission to use their practical wisdom, guide them and develop their skills to use patient experience insights, and strengthen their confidence to welcome feedback and work directly with patients.

We leave the final word in this chapter to one of the patient experience officers, who sent an e-mail reflecting on how her thinking about her own role had changed and what the trust will be doing differently:

I also learnt a lot from the project and I now do some things differently as I upscale to other wards – one of the main things that became clear was that it was no good just giving the data to the wards – they need support and guidance to understand what the data is telling them and how to implement changes.

What also became clear was that very valuable data could be obtained from just general conversations with patients and once the staff felt empowered to make changes, they did so quickly for these patients.

For us, the success of the project was that the staff became a team and like all new teams they need time to learn to trust each other and to work in a supportive, cohesive way which will certainly lead to improvements in the patient’s experience.

Motivated staff provide better care

Staff in several sites suggested that staff who felt well supported and valued would be better able, emotionally and psychologically, to provide good support to patients. The patient experience officer and ward manager in one site explained this perspective, arguing that good staff experience might be a necessary, but not sufficient, condition for good patient experience:

Happy staff don’t necessarily make happy patients. But unhappy staff will never make good patient experience. So I think if you’re just going to work on patient experience, you are going to make your job twice as hard. I think you’ll get there, but it’s going to be harder. If you can get the staff engaged and come in with you then you’re half way there. So I think on any ward that we take this forward, our first steps, learning from this project, is to look at the staff first. And then the patient experience. So doing it in two steps, because otherwise I think we are almost setting them up to fail just straight away by bringing patient experience without looking at the ward set-up.

Patient experience officer

The patients are very vulnerable. They’re very scared. They’re in a time in their lives where they may never have been in this environment before, and they’re extremely vulnerable. So, making them feel safe and cared for is, you know, hugely important . . . If the staff are smiling and they feel that they can give the care they want, that will come across to the patients, and the patients, in their vulnerable state, will feel more reassured.

Ward manager

The head of patient experience in another site used similar language to describe this relationship. This comment also suggests that getting good FFT scores was in itself motivating:

We’ve actually done a lot of work recently around staff experience. Because in our opinion if you look after the staff they’ll look after the patients. If you don’t have happy staff you’re not going to have happy patients. [Director of nursing] is amazing for that and she’s done so much work . . . We had our highest [FFT] recommendation rate we’ve ever had last month. It’s improved by about 3% over the past year since the director of nursing came into post. And I genuinely believe it’s because we have done so much work with staff.

Head of patient experience

This sentiment was echoed by a health-care assistant in the same site describing the positive effect of staff ideas being discussed and acted on by a local staff council:

[The staff council] is like a group of people that the staff – like we’ve got a light-bulb moment we put in the, by the staffroom. If you have a light-bulb moment you put it up. Somebody put up ‘team building’. Somebody put up ‘clock in staffroom’. Somebody put up ‘a bladder scanner’. So, they take the light-bulb moments; they take it to the council; they then discuss them, try and work on them to try and improve how the employees feel about where they’re working . . . Because I think what we found from doing this project is if the staff are happy and they want to work on the ward, that’s going to reflect onto the patients. The patients are then going to be happy and then that will reflect on the data we get. So that’s how we’re sort of interlinking it.

Health-care assistant

Staff who feel that their experience is taken seriously are more likely to be motivated and receptive to feedback

A number of sites used some form of staff suggestions board to gather views on how both staff and patient experiences could be improved. The rationale was twofold: first, that staff had a valid perspective on patient experiences and ways to improve them (‘soft’ intelligence, as described in Chapter 6); and, second, that seeking staff views and demonstrating that these were acted on would be affirming and empowering to them. This in turn could positively affect their morale and motivation:

But I think that came from the need for [ward manager] to try and involve all the staff in that ward. And I think she was right. And I think the staff really appreciated that they felt some sort of ownership. They could see that the, the senior staff were serious when they said, we are going to make improvements on this ward. They could see it in action. And they could see that they were being invited to take part. So I think that was a very good move on [ward manager]’s part, actually to get them in. And to be able to do those quick wins with that. She had done things straight away.

Patient experience officer

It was also suggested that staff who felt that their views were listened to and their expertise was respected would be more receptive to patient feedback and more likely to want to get involved in QI. The importance of creating a sense of ownership, confidence and ability to act was a common theme:

I don’t think you can do any patient experience until you’ve sorted out your staff experience. They’ve got to come first, and then look at the data. And then – or get them to look, to help them look at the data . . . And empowering them to make that decision. ‘Cause I think if they listened to their gut feeling, that’s often right. And I think what we can do with the data is to prove that what they thought was right or we can back that up. ‘Cause I think they know, don’t they, they’re on the shop floor. When I look at their Post-it notes that they did at the very beginning, some of the ideas that came up there were very much ideas that then the patients brought forward at the focus group or the data has shown to be right. But the staff knew that.

Patient experience officer

Managers argued that respecting staff’s practice-based experience and expertise meant that interventions were more likely to be implemented effectively:

The advantages are, as we know, that actually if the change comes from within, the team are more likely to do it, and also they really understand the workings and the systems on their wards, and so they know what makes it work. So, the advantages of using front-line staff definitely outweigh not using them. I’m just saying because the nursing director says do it, doesn’t really instil any confidence in anybody, let alone me. But if the staff come up with the good ideas – and they’ll also know where the waste is. They’ll also know why they do something in a certain way, and know how to change it.

Director of nursing

You need to remember that the staff are doing a job all the time. They have ideas that might improve the service . . . I’ve always said, ‘This is our team’. If we get told that we have to do something, the team will decide how they do it. Because it’s their ward, and I think that sort of strength and that feeling that the ideas board, their ideas are real. They’re not up in airy-fairy team-building days where you’re all climbing a mountain. They say, ‘We want this because this will make our job easier’.

Ward manager

This suggestion that staff-led improvement may make staff’s job easier as well as improve care for patients is explored further in Involvement in patient-centred quality improvement is motivating.

Pressures on time and the demands of a challenging work environment can impede a readiness to engage with patients and their concerns. One ward manager described the relaxed state of mind needed to engage with patient concerns. A ward clerk in the same site described the way that stressed staff tended to put up a barrier between themselves and patients and families:

It’s like being on the M25 . . . It’s very difficult to find time to sit and, you know, in a relaxed way listen . . . You’re constantly co-ordinating a ward. You really have to actually say, ‘Look, shut the door, I have the time, you tell me’. It’s a state of mind. I think you really have to be quite relaxed to be able to deal with complaints.

Ward manager

We’re just living on a knife’s edge, you know. It’s quite hard, and you sort of put yourself a bit of a barrier up – but that barrier comes through to everyone, do you know? . . . Some patients do make our lives an awful lot harder and of course it’s going to come across isn’t it, in the staff? It’s going to come across with the way the relatives view us.

Ward clerk

Involvement in patient-centred quality improvement is itself motivating

Some participants described how staff derived fulfilment directly from their involvement in patient-centred QI work. One ward manager explained how it aligned with the intrinsic motivations that drew many staff to work in the NHS:

They’re all really keen. Because I think at the heart of it, they all want to make a difference to – it sounds terribly cheesy – but they all want to make a difference to the patient. That’s what a lot of us are here for. That’s what we do extra hours for. That’s what we work for, and that’s what we work in the NHS for, you know. It’s not an easy place to work but if we have something that can make patients feel more positive about the place, and staff. I mean, go for it, you know. I think they’re all showing a real interest in it.

Ward manager

A matron explained how the trust felt that offering staff the chance to do something different by participating in the project could help re-motivate them after a difficult few months:

I think from a ward level they’ve really embraced it because they’re a ward who were struggling with vacancies, difficult kind of patients, and well-established members of staff of whom a large proportion had left in the last year . . . So the reason partly that I’d chosen them was it gave them an opportunity to be involved with something that was really beneficial, not just for patients but also for staff experience and staff engagement.

Matron

As one health-care assistant remarked, going on study leave to the learning community events as part of the project in itself felt rewarding:

Our ward very quickly, you know, we called it the Oxford project and so that’s very much still what it is, ‘Oh, the Oxford project’. And in our staff meetings we have updates on the Oxford project . . . I can’t say that there’s been any negatives. And, you know, another one of the bonuses is the fact that we’ve got to go to lovely places. So we got to go to Oxford. You know, never in my life would I have, I would never have thought I would have gone on a study session at Oxford University.

Health-care assistant

The head of patient experience in another site commented on the morale-boosting effect not just of taking part in QI through the project, but also of being recognised for it through an award within the trust. Again, the equation of ‘happy staff with happy patients’ is evident:

Yeah, we won a patient experience award – that was amazing. It was so good . . . That was about celebrating compliments and positivity, and how do we share the good that comes out, because yeah, like you say, constantly being compared to the national average and being under it and, you know, what else are we supposed to do? How do we get there? This was all about celebrating our staff basically.

And do you think that that helps them to make changes and improvements, getting positive feedback?

Yeah, it boosts morale, it makes people realise actually they’re doing a really good job, but there are a few things that you could improve on. That’s very different than ‘you’re doing rubbish’ . . . I think it makes a massive difference. Happy staff equals happy patients as well; I really believe that. If you support your staff, then they will naturally look after your patients, yeah.

Improving patient experience can directly improve staff experience

Staff identified a possible circular relationship between staff and patient experience. They suggested that poor patient experience could lead patients to behave in ways that made the working lives of employees more difficult. This could emerge from an increase in workloads (e.g. if staff are unable to respond in a timely way to call bells, the patient’s need could become more acute, requiring more staff input in the long run). It could also lead to poor patient–staff interactions, when patients were frustrated, bored or uncomfortable. This in turn could demotivate staff.

Some wards focused on introducing more activities to relieve patient boredom. One rehabilitation ward had already employed a health-care assistant as an activities co-ordinator, an unusual role in an acute hospital setting. He described the benefits of a new interactive computer system:

What’s come out of the project is kind of a computer system what helps patients engage . . . I’d like to think it’s probably gonna be the biggest thing what’s come out of the project . . . It’s amazing the difference [it] makes . . . [It] has reminiscences, has activities, we’ve got horse racing, interactive horse racing . . . We’ve had patients say, ‘I just love [it]. I’ve never had access to this in a hospital before’ . . . Patients are set for the afternoon watching an old movie or watching Dad’s Army. And it engages patients. We’d like to think that we want more interaction in terms of one-to-one stuff or group stuff. But actually sometimes, you know, patients sitting watching a movie, it’s engaging.

Health-care assistant

The ward manager in this site noted that one aim of promoting activities to avoid boredom and frustration was to reduce restlessness and avoid falls. Although they could not be certain that the interactive computer system was the cause, the number of falls on the ward had reduced since its introduction.

Another site had introduced activities boxes with cards, dominoes, colouring books and word searches. This site had also introduced textured gloves (or ‘fiddle mitts’) to occupy restless hands, particularly for people with dementia. Towards the end of the project, a health-care assistant showed these to the ethnographer and described their effects on patients and staff on the ward:

She felt the boxes and mitts were effective first as an indication that they cared about patient well-being and that, in itself, was a valuable message to send to patients. She explained that they were effective in keeping people occupied who otherwise get really bored . . . The introduction of the activity boxes and twiddle mitts has improved patients’ moods . . . That has, in turn, improved staff experience and staff morale.

Ethnographic field note

A ward manager in another site explained how interventions to improve people’s sleep and reduce pain could also directly improve staff experience:

Because even down to the eye patches, you’d say if the patient’s got more sleep their pain threshold’s going to be better, which means they’ll probably need less analgesia, which means that they’re going to be demanding less on the nursing staff . . . It sounds like one small step for mankind; do you know what I mean? But in a place like this where you’ve got a multidisciplinary team and so many different factors, every single improvement will touch someone in some way. [The doctors] think, ‘Eye patches, pah!’ you know. ‘What’s that compared to an endoscopy blah?’ you know. But in actual fact it’s simple but it’s hugely important.

Ward manager

Conclusions

A strong focus on patient experience is a priority for the NHS, but may risk implying that patient and staff experience are somehow in tension with or in opposition to each other, rather than being mutually reinforcing and compatible. Approaches such as EBCD seek to pay attention to both, seeing clinical staff as part of the solution and not part of the problem.

Our original ‘theory of change’, based on previous work,39 was that many experiences that matter most to patients happen in front-line encounters; and that bottom-up engagement in person-centred improvement (as opposed to top-down managerially driven initiatives) can be motivating for front-line staff. Conversely, QI that is perceived to be top-down and not to engage with staff values of patient care is unlikely to lead to sustainable impact and embedded cultural change. Furthermore, at a time when NHS staff have felt under extreme pressure, acknowledging the importance of their own knowledge and experience and paying attention to staff morale and motivation may in turn lead (directly or indirectly) to more patient-centred care.

Our data show that a relationship between staff and patient experience and engagement was understood by front-line staff. These findings align with Luxford et al. ’s90 conclusions that a prerequisite for patient-centred care is a sustained focus on staff experience, and with Shaller’s91 argument that:

1. to achieve the commitment of front-line staff to patient-centred care, they should be directly involved in the design and implementation of processes

2. to deliver patient-centred care, it is important to nurture an environment in which the organisation’s most important asset – its workforce – is valued and treated with the same level of dignity and respect as they are expected to show to patients. As one of his participants argued, to deliver patient-centred care an organisation needs to be human-centred, not just patient-centred.

In Luxford et al. ’s90 study, the focus on staff experience tended to be manifested in public demonstration, such as award ceremonies and newsletter articles. Although this was important to some staff interviewed in our study, small, tangible changes in staff’s working environment – even as small as a clock in the staffroom – were also experienced as a demonstration of being valued by the organisation.

Our findings suggest that a sense of control resulting from engagement in the process of change can act as a buffer to feelings of burnout and can re-motivate staff. This resonates with findings from Maben et al. ,92 who note that staff seek to alleviate the burden of a high-demand/low-control work environment, low staffing and lack of ward leadership by finding personal satisfaction caring for what they describe as ‘the poppets’: patients they like and enjoy caring for. Other patients who have a poor relationship with staff may feel more like ‘parcels’ and seek to avoid being labelled as difficult or demanding. Staff in our study were clearly aware that they found it harder to be responsive when they were stressed.

Maben et al. 13 conclude that good staff well-being is an antecedent to good patient care, echoing the intuitive observation in this study that ‘happy staff means happy patients’. There is a growing body of evidence to support this. Early work on the NHS staff survey by West et al. 93 showed clear links between staff well-being at work and patient experience and outcomes. The strongest predictor of patient mortality, for example, was the percentage of staff who felt that they were working in well-functioning teams. Significantly for this study, lower mortality also occurred in trusts where staff felt that they had opportunities to influence and contribute to improvements at work.

Dawson’s16 recent quantitative analysis of survey results from 2014 and 2015 found an association between staff experience and patients satisfaction. The following staff experience variables were found to be key predictors of patient satisfaction across a number of domains:

• work pressure felt by staff

• percentage of staff believing that the trust provides equal opportunities for career progression or promotion

• staff satisfaction with resourcing and support

• percentage of staff feeling satisfied with the quality of work and patient care they are able to deliver

• percentage of staff experiencing physical violence from colleagues in past 12 months

• percentage of staff experiencing discrimination at work in past 12 months

• effective team working

• percentage of staff agreeing that patient feedback is used to make informed decisions

• percentage of staff witnessing potentially harmful errors, near-misses or incidents in last month

• fairness and effectiveness of procedures for reporting errors, near-misses or incidents.

The three most important factors specifically for black and minority ethnic staff experience were:

• percentage of staff agreeing that their role makes a difference to patients

• percentage of staff experiencing discrimination at work in past 12 months

• percentage of staff able to contribute towards improvements at work.

However, Dawson16 concludes that good patient care:

is about more than the absence of negative experiences: it is about being able to work effectively together in effective teams, with well-designed jobs.

Dawson,16 p. 7.

Contains public sector information licensed under the Open Government Licence v3.0

Similarly, an analysis by Sizmur and Raleigh15 recently reported that patient experience is negatively associated with a range of workforce factors: higher spend on agency staff, fewer doctors and especially fewer nurses per bed, and bed occupancy. A high proportion of agency staff will make team working more difficult to achieve. This study also found that staff-reported experience and patient feedback were correlated in several areas, particularly between staff perceptions of care quality and patient-reported experience.

In North America, it has been argued that ‘care of the provider’ should be added to the ‘triple aim’ first proposed by Don Berwick of enhancing patient experience, improving population health and reducing costs, and indeed that achieving the triple aim is impossible without paying attention to the work life of health-care clinicians. 94

An evaluation of Schwartz Rounds^® as an intervention to improve staff experience through reflective peer discussion found a statistically significant improvement in staff psychological well-being. 95 Reported outcomes also included ‘increased empathy and compassion for patients and colleagues and positive changes in practice’. These reflections suggest that a binary top-down/bottom-up view of change is too limited, and that a whole-organisation focus on staff experience and well-being is important in creating a receptive context for patient-centred improvement.

As well as reflecting this growing evidence on the importance of staff experience and providing a positive working environment, we suggest that this study adds two new points:

• The act of undertaking QI activity could in itself improve staff morale, appealing to staff’s intrinsic motivation to provide good patient care. This could operate through the variables of a sense of ‘effective team working’ and increasing the sense that ‘patient feedback is used to make informed decisions’.

• Improvements to patient experience may reduce their sense of boredom and frustration, and create more positive staff–patient relationships, thereby improving staff’s own experience and making work less stressful.

Further research is needed to test these propositions.

Introduction

In this chapter, we explore the impact of the way in which front-line ward teams were constituted, the mix of staff of different grades and professions that was included, and how this interacted with each local context to affect the outcome of their QI work.

Each team sought to assemble individuals whom they felt were best placed to contribute to the project and its underlying QI ethos. Up to five core team members (including a patient or a family member) were invited to attend the learning community events, although in practice only one family carer and one other lay person were involved in any of the core team memberships. Each team could involve additional staff, either as core members or to contribute to specific aspects of a relevant improvement programme. For example, in one site, five junior doctors contributed to the project as part of their introduction to clinical practice on the ward involved. The constitution of the teams and whom they chose to involve locally was intentionally left broadly to them to determine, although it was emphasised that the first learning community was aimed at front-line staff. Director-level staff were discouraged from attending in order to give front-line teams ownership of the project.

The disciplinary and professional composition of each team varied considerably. For example, two teams consisted almost entirely of front-line ward nursing staff, whereas others included a mix of clinical and non-clinical roles, including, for example, ward manager, medical registrar, senior sister, health-care assistant, quality nurse, ward clerk, activities co-ordinator, patient experience officer, head of patient experience, pharmacist and matron. The collective core membership of the teams also reflected noticeable variation in NHS band level. As an illustration, one team consisted almost entirely of nursing staff at bands 7 and 8 (likely to be in leadership roles on the ward), in contrast to another team that comprised staff from bands 3–9.

Conjoining staff resources

The assemblage of individuals from a range of NHS band levels is important in understanding the extent to which teams were able to make progress. The multidisciplinary and mixed-seniority composition of each team brought together a range of NHS resources in newly formed patterns, resulting in collective team benefit. These NHS resources may be described as distinctive physical, material, personal, experiential and relational characteristics associated with specific roles and positions in the participating NHS trusts. For example, the range of resources available to a ward-based senior sister differs significantly from those accessed and owned by a patient experience officer; and a patient experience officer with a clinical background will wield different resources from a colleague in the same role without clinical qualifications. The formation of the six teams resulted in the conjoining of assorted NHS-specific resources to be deployed as part of QI work. Although such pooling of employment-related skills and resources is by no means unique in interdisciplinary NHS teamwork, participation in the study created new assemblages that might not otherwise have occurred.

As will be discussed throughout this chapter, the overall success (or otherwise) of the various improvement projects was observed to be strongly influenced by the range of skills and resources available to each team. In analysing these findings, therefore, we draw on the sociological concept of capital.

Capital

Capital is a familiar concept in social science. Although capital has been defined in a number of ways,96–98 it typically focuses on the way in which resources and assets interact within human relationships. It is the construct of capital as advocated by Pierre Bourdieu97,99 that has perhaps become the most influential in contemporary sociology, and that informs this analysis.

Bourdieu’s97,99 typology of capital proposes that it exists in four distinct and different forms: economic, cultural, social and symbolic capital. Each of these is summarised below. For Bourdieu, capital emerges from interactions with two further key components of his theory of social action (practice): field and habitus. Habitus can be defined as the way one unconsciously acts, interacts and behaves within the social world, according to the norms, traditions and unwritten rules of one’s particular group. Field provides context for habitus; it is the arena where one acts and that both frames conduct and provides opportunities for creating capital. For example, in this study, it is the NHS workplace. Habitus, field and capital do not exist in isolation from each other.

Capital: Bourdieu’s typology

Economic capital is perhaps the most familiar form of capital, which may exist at both individual and institutional level. Examples of economic capital include fiscal wealth, possessions, housing status and the range of physical resources and materials available to and owned by individuals and institutions in society.

Cultural capital emerges from social position (or social class) but may also be recognised within organisations and institutions (especially employment settings) as part of specific norms and values, and membership of particular institutions. Examples of cultural capital include knowledge, education, qualifications, credentials (such as work-related job titles), expertise and specific work-related skills.

Social capital relates to the range and depth of social networks available at an individual and societal level. Social capital may be viewed as purposely creating, establishing and maintaining relationships with others to acquire further social capital. Bourdieu97 considered social capital to be intrinsically linked with the perpetuation of class-based social inequality, reproducing cliques and conventions associated with elitism and power.

Symbolic capital relates to further embodied forms of socially recognised values, norms and achievements held at an individual or an institutional level. Examples of symbolic capital include an individual’s reputation, prestige and social status, which serve to reinforce and reproduce particular power relationships. Examples of symbolic capital may be found within any hierarchically organised institution such as the NHS.

The formation of teams of people from different disciplines and levels of seniority established a network of individuals with assorted levels of the four forms of capital described above. The conjoining of their respective individual capital constructed a form of new collective capital, a specific form of capital relevant to NHS teamwork that we term ‘capital in the NHS’ (i.e. capital that is specific to NHS services and service providers). We suggest that the four forms of capital combine within each of the six teams to establish a mechanism that drives and facilitates the various US-PEx QI work.

Here we consciously use the term ‘mechanism’ to allude to realist evaluation, to propose that, in each of our NHS contexts, this team capital is a mechanism that helps us explain the outcomes in each site: what teams were able to achieve, and where they came up against constraints or lack of power.

Findings: capital in NHS settings

In the following section, selected examples demonstrate the range of physical, embodied and institutionalised forms of capital available to ward teams as they developed their QI projects. We look first at positive examples of capital. Although these are discussed within Bourdieu’s97 four categories, in practice they often work in combination.

‘Negative’ capital in NHS settings

So far the emphasis has been on how teams of individual NHS staff were able to harness various forms of capital to have a positive impact on QI. We now turn to negative capital: the lack or absence of capital, the impact that this had on staff and how this constrained their ability to make progress.

The range of negative experiences articulated by front-line staff typically reflect forms of structural inequality in the NHS, and disparities connected to roles and responsibilities within the organisation. Experiences of negative capital were more typically described by team members on lower NHS band scales (bands 3–5), whereas examples of positive capital more commonly came from staff at bands 5–9. These illustrations highlight various barriers and difficulties encountered by front-line staff at lower band levels. As we have seen in the previous section, both clinical and non-clinical staff at lower band levels could and did wield capital. However, their ability to do so was greater when working in teams with access to a broad range of professional networks and forms of capital, complementing the cultural capital of front-line nursing and health-care assistant staff.

Conclusion

The formation of teams of people from different disciplines and levels of seniority automatically establishes a network of individuals with assorted levels of the four forms of capital listed above. The conjoining of their individual capital constructed a form of new collective capital, a specific form of capital relevant to NHS teamwork that we term ‘team capital in NHS settings’ (i.e. capital that is specific to NHS services and service providers). We suggest that the four forms of capital merge as one within each of the six teams to establish a mechanism that drives and facilitates the various US-PEx QI work.

Here we consciously use the term ‘mechanism’ to allude to realist evaluation in order to propose that, in each of our NHS contexts, this team capital is a mechanism that helps us explain the outcomes in each site: what teams were able to achieve, and where they came up against constraints or lack of power.

The illustrations above highlight a wide range of strengths, weaknesses, opportunities and threats that may be encountered in the formation of teams consisting of mixed band levels from a range of clinical and non-clinical disciplines. Strengths and opportunities relate to the potential for teams to apply innovative and creative methods within improvement projects, the naturally occurring conjoining of assorted capital, the value of clinical and non-clinical staff to work as one in genuinely collaborative approaches, and the possibility of maximising staff potential by involvement in such multidisciplinary teamwork. The assembling of such diverse teams has the potential to better negotiate institutional norms and practices and facilitate the QI efforts of the teams involved. In short, conjoined capital provides opportunities for access to material and embodied resources that enable good practice and facilitate improved performance.

Weaknesses and threats, however, relate to the teams’ potential to maintain the routines of the institution and to observe the hierarchical expectations of power (or lack of power) associated with team members’ band level. Teams that choose to restrict membership to a particular discipline (whether clinical or non-clinical), consisting of individuals from similar band levels, may severely limit the level of conjoined capital within their network and inadvertently establish a non-collaborative partnership. As demonstrated in the section on ‘negative capital’, such practice may serve to dis-able positive outcomes and constrain the goal of QI.

The relative lack of patient and family involvement in core team membership could be regarded as a missed opportunity that could have added to and strengthened the resources and skills available for teams to draw on.

In terms of the practical implications arising from this analysis, it is clear that the formation of the team responsible for such QI work is paramount. Purposive selection and formation of the improvement team should aim to recruit, where possible, both clinical and non-clinical expertise and, equally, vary the levels of seniority within the team. In such selection procedures, the design process establishes a pool of naturally occurring (cultural, social and symbolic) capital. In practice, the conjoined capital to emerge (with access to a wide variety of resources, networks and alliances) from this process places the QI team at an immediate advantage in developing further team-based capital.

Second, leadership for QI needs to recognise and realise opportunities from the collective capital available within the assembled team. This approach challenges normal hierarchy, and instead aims to generate a more equal and magnanimous relationship within the team and provides opportunities for individuals to promote and develop social and symbolic capital. It fosters the development of innovative working relationships and methods for QI that differ from established institutionalised norms and practices. For example, health-care assistants and consultants may work alongside non-clinical staff to design an action plan. Innovative work such as this is driven by the conjoined capital within the team, and may also significantly challenge the institutionalised norms and practices of the health-care setting concerned, but with the aim of producing a beneficial outcome for patients and staff.

These suggestions are by no means radical and are consistent with other research findings. For example, in their review of QI, Alderwick et al. 100 identify a series of lessons for NHS leaders, which include sharing responsibility for QI with leaders at all levels; developing skills and capabilities for improvement; focusing on relationships and culture; enabling front-line staff to engage in QI; involving patients, service users and carers; and not looking for ‘magic bullets or quick fixes’. The emphasis on relationships, capabilities and working as a coherent group within an established system aligns with the account of team capital given above.

Limitations

With regard to the phase 1 survey of patient experience leads, we note that, given that only 39% of hospital NHS trusts responded, it is possible that those who responded held stronger views about patient experience data.

This study was primarily an exploratory ethnographic study of how and why NHS front-line staff do or do not use patient experience data for QI. It was not a ‘what works?’ study designed to demonstrate whether particular types of patient experience data or QI approaches are more effective than others. We therefore cannot draw firm conclusions on this point, and nor can we comment on the cost-effectiveness of different approaches.

The study used a focused, team ethnography approach, which brings both challenges and advantages. The nature of ethnographic work is discussed more fully in Chapter 2, ‘Team’ and ‘focused’ ethnography.

The research team adopted a formative and supportive approach to working with the participating sites, and did not wish to create a climate in which the sites were being ranked on ‘success’, particularly given their very different starting points and amount of organisational experience with patient-centred QI methods. There is always a difficult balance to be struck between helping sites achieve the most they can, and letting events unfold and observing what choices sites make and what helps or hinders them. It is arguable that, by providing learning communities and ongoing support, the study created an artificial environment that could result in overperformance compared with normal practice.

In fact, as noted in the report (see Wider contextual factors: NHS pressures and ‘organisational distress’), the pressure on staff was such that the uptake of webinars and ongoing improvement support was very limited, and several ward teams struggled to make headway. Furthermore, most trusts have QI and patient experience support teams in-house who could provide similar input (even if, as our results show, this is not always taken up in practice).

‘Winter pressures’ during the study period affected sites’ ability to pursue their improvement projects and, indeed, to take part in study fieldwork on some occasions. It was hoped that participating sites would be able to provide some routine data (e.g. on vacancies and sickness levels) to enable us to judge whether some sites had experienced this more than others as part of understanding the local context in each site. Unfortunately, this was not possible in all sites, and the figures are too incomplete to permit meaningful comparison. However, ethnographic data would suggest that these pressures in the system were an issue for all sites.

We did not set out to compare the improvement activities of the ward selected in each site with other wards in the same trust. We are therefore unable to conclude whether working on an explicitly patient experience-focused improvement project at a time of intense pressure was more or less likely to be maintained than other kinds of QI work in other wards.

The relatively low level of patient involvement in the core ward teams was a source of disappointment. With hindsight, perhaps more could have been done to address this; the research team strongly encouraged teams to bring a patient with them to the first learning community, but this was not compulsory. A few teams whose R&D confirmation came through only just before the event would have found it challenging to find a patient team member at such short notice. The lay panel advising the research team offered to get involved with sites, although this was not strictly part of their role (see Chapter 3), and the offer was not taken up. At the same time, the low level of involvement is an interesting finding in itself.

There is some discussion of the limitations of the before-and-after patient experience survey in Appendix 4. One issue was the timing of administration. The benchmark survey covered patients discharged in January–March 2016. The sampling period for the follow-up survey was originally planned to cover the same period in 2017. However, because the unusually severe winter pressures affected teams’ ability to pursue their QI projects, the research team was concerned that using the same sampling period would mean assessing sites while it was still at too early for any changes to have had an impact. Although, ideally, the post-intervention survey would have been conducted over the same calendar months to eliminate any seasonality effect on people’s responses, the risk of potentially not capturing the impacts of the interventions was considered to outweigh this limitation.

In the event, the survey showed very few statistically significant changes in patient experience. This partly reflects the difficulty of creating a survey instrument that can be used across all contexts but is sufficiently sensitive to the great variety of local change initiatives and priorities. It may also reflect the fact that some sites chose a longer-term strategy of improving staff experience as a route to improving patient experience.

Despite these limitations, the study offers important theoretical and empirical insights into how staff engage with patient experience data for QI, and how team composition and involvement from the local patient experience team can affect what happens in practice.

Implications for practice

Building on the above summarised findings, we offer the following suggestions for NHS practice, in no particular order of priority:

• Provide specialised training and career development pathways for trust staff involved in patient experience teams. Recruitment could look to include a diverse range of staff from both clinical and non-clinical backgrounds.

• Strengthen existing moves towards greater integration and co-working between teams responsible for patient experience and QI in trusts. This includes seeing the role of patient experience teams as going beyond data collection to working on improvement. For QI teams, it includes focusing on patient experience and patient involvement, as well as targeting safety and efficiency.

• Improve the availability of more tailored, local, ward-specific surveys, and provide support and facilitation for staff to use these (e.g. infographics).

• Increase training in using less obvious forms of data, and supporting both front-line staff and boards to see these as legitimate, actionable sources of improvement ideas. The new training programme in foundations in patient experience from The Point of Care Foundation is a step in this direction.

• Promote the inclusion of a broad mix of staff in teams working on QI to maximise their collective ability to generate ideas, garner organisational support and effect change.

• Offer front-line staff more training and support in involving patients in QI, both in individual projects and in teams overseeing QI work.

• Consider and evaluate interventions to improve staff experience as a route to improving patients’ experience, as well as interventions focused more directly on patient experience.

Implications for research

We offer three key suggestions for future research:

1. There is a growing body of evidence that staff experience is one of the most important predictors of patient experience, but much of this work is based on observation of the association between measures of staff and patient experience, rather than on interventional research. We do not yet know what kinds of interventions to improve staff experience are most likely to be effective and what the costs of these would be. We therefore suggest developing and testing interventions focused specifically on staff with patient experience as the outcome, with a health economics component.

2. Studies focusing on the effect of team composition and diversity on the impact and scope of patient-centred QI.

3. Building on both this study and INQUIRE, more research into how organisations, and especially front-line staff, use and respond to unstructured feedback and soft intelligence.

Contributions of authors

Professor Louise Locock, principal investigator, led the overall design and provided academic leadership for the study, managed the ethnography team, led the development of the handbook and online resource, sat on the Study Steering Committee, led the writing of the final report and gave final approval of the manuscript.

Mr Chris Graham, co-investigator, contributed to the overall study design, led the phase 1 and the two phase 2 surveys (before and after the QI work), provided expertise to the ward teams at the learning community events, sat on the Study Steering Committee, co-authored chapters of the final report and gave final approval of the manuscript.

Ms Jenny King, co-investigator, contributed to the overall study design, co-led the phase 1 and the two phase 2 surveys, conducted and analysed patient interviews (before and after the QI work), provided expertise to the ward teams at the learning community events, was lead author of Chapter 2 on the study methods and of Chapter 4 on the phase 1 survey, co-authored other chapters of the final report and gave final approval of the manuscript.

Dr Stephen Parkin, researcher, contributed to the study design, conducted and analysed patient interviews (before and after the QI work), conducted the ethnography at three case study sites, was lead author of Chapter 8 on team-based capital in NHS settings, co-authored other chapters of the final report and gave final approval of the manuscript.

Dr Alison Chisholm, researcher, contributed to the study design, conducted and analysed patient interviews (before and after the QI work), conducted the ethnography at one case study site, was lead author of Chapter 7 on improving staff experience, co-authored other chapters of the final report and gave final approval of the manuscript.

Dr Catherine Montgomery, researcher, contributed to the study design, conducted and analysed patient interviews (before and after the QI work), conducted the ethnography at two case study sites, co-authored chapters of the final report and gave final approval of the manuscript.

Ms Elizabeth Gibbons, co-investigator, contributed to the study design, conducted and analysed patient interviews (before and after the QI work), provided expertise to the ward teams at the learning community events, co-authored chapters of the final report and gave final approval of the manuscript.

Dr Esther Ainley, researcher, contributed to the study design, conducted and analysed the phase 1 and the two phase 2 surveys, conducted and analysed patient interviews (before and after the QI work), co-authored chapters of the final report and gave final approval of the manuscript.

Ms Jennifer Bostock, co-investigator and lay representative, led the design of the PPI activities and contributed to the design of the overall study, chaired the lay panel, supported other PPI members, was lead author of Chapter 3 on patient and public involvement, co-authored another chapter and gave final approval of the manuscript.

Ms Melanie Gager, co-investigator, contributed to the overall study design, advised the research team on working with ward staff, provided expertise to the ward teams at the learning community events in her capacity as a senior intensive care unit sister and QI facilitator, co-authored chapters of the final report and gave final approval of the manuscript.

Dr Neil Churchill, co-investigator, contributed to the overall study design and was involved throughout by providing insight into the relevance of the study for NHS policy, contributed to the final report and gave final approval of the manuscript.

Professor Sue Dopson, co-investigator, contributed to the overall study design advising specifically on NHS organisational culture and change, provided expertise to the ward teams at the learning community events, contributed to the final report and gave final approval of the manuscript.

Professor Trish Greenhalgh, co-investigator, contributed to the overall study design, co-authored chapters of the final report and gave final approval of the manuscript.

Dr Angela Martin, co-ordinator, contributed to the study design and managed the conduct of the study, co-authored Chapter 3 on PPI, contributed to the final report and gave final approval of the manuscript.

Professor John Powell, co-investigator, contributed to the overall study design, provided expertise to the ward teams at the learning community events, contributed to the final report and gave final approval of the manuscript.

Dr Steve Sizmur, statistician, contributed to the overall study design, provided statistical analysis expertise, contributed to the final report and gave final approval of the manuscript.

Professor Sue Ziebland, co-investigator, contributed to the overall study design, provided expertise to the ward teams at the learning community events, contributed to the final report and gave final approval of the manuscript.

Contributions of others

The members of the lay panel (Barbara Bass, Tina Lonhgurst, Georgina McMasters, Carol Munt, Gillian Richards, Tracey Richards, Gordon Sturmey, Karen Swaffield, Ann Tomlime and Paul Whitehouse) contributed to Chapter 3 in particular.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to available anonymised data may be granted following review.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health and Social Care.

Case study 1: description summary

This case study is set within a local general district hospital in an area of high deprivation. The hospital had previously experienced a highly critical CQC inspection, with subsequent public reporting and negative discourse, which naturally had had an impact on staff morale as well as resulting in senior management resignations. Since 2012, measures have been lifted and a more positive assessment was obtained from the CQC in 2015. There are annual ‘remarkable people awards’ with categories including ward of the year, team of the year and employee of the year. In addition, feedback from online and social media is triangulated with other sources. Formal staff feedback is now more positive, based on the annual staff survey and FFT, specifically around perceptions of the care provided by the trust; the trust’s response to patient concerns; and staff recommendations of a place to work or receive care for themselves or their relatives/friends. Despite this, there are significant staff shortages, with many posts vacant, especially in challenging areas and services.

Case study 2: description summary

This case study site is a large hospital foundation trust with two district and one maternity hospital, which serves a population in an affluent county. The study covered a turbulent time in the trust; there were resignations and vacancies at senior management level and the trust had recently been placed in ‘financial special measures’.

Recent CQC inspections reported that the trust required improvement in relation to safety, effectiveness and responsiveness. However, the report highlighted that the provision of care to patients was compassionate and caring.

The survey of patient experience leads in phase 1 reported that the trust had formal arrangements to promote and co-ordinate safety and QI education for staff at all levels. However, the trust did not have a specific written plan/strategy for the collection and use of patient experience data. Furthermore, there was a lack of time to examine it owing to workforce and general system issues.

In addition, the 2015 staff survey results indicated poor performance on ‘staff engagement’ as well as poor staff ratings on the effective use of patient/service user feedback, staff motivation, staff satisfaction with quality of work and ability to deliver patient care.

Case study 3: description summary

This case study site is a hospital in a large foundation trust. It provides acute and emergency care as well as a range of other services expected in a general hospital.

The CQC Inspection Report published in 2015 gave an overall rating for the hospital as ‘outstanding’, with most services rated as ‘good’ or ‘outstanding’. This outstanding rating was premised on an inspection that concluded that staff/services were noted as ‘caring, responsive and well-led, and good in providing safe and effective care’. The report notably commented on the implementation of different models of care; strong governance structures and leadership; staff and patient engagement; proactive management of patient flow; and outstanding practice of the collection of real-time feedback at ward level, with timely action and responsiveness. This was regardless of whether the feedback was positive or negative.

There is strong leadership and active engagement with the patient experience office at a strategic level, and the trust is motivated towards improving the quality of care and patient experience.

Case study 4: description summary

This case study site provides core services and is also a research-focused teaching facility in a large urban area.

In March 2016, the CQC gave the hospital an overall rating of ‘good’, but some services were identified as requiring improvement. The report noted examples of services that had outstanding leadership; staff were dedicated, caring and supportive of each other. There was also overwhelmingly positive patient feedback. The main negative observations by the CQC related to the impact of increasing demand on staff and service delivery.

Case study 5: description summary

This case study is a large general hospital providing acute services and renal, stroke and vascular specialties on four main sites. The trust is also an accredited cancer centre providing services to a wider population.

A CQC inspection in June 2015 resulted in an overall rating of ‘requires improvement’, with some specific services assessed as ‘good’. Staff feedback via the national survey placed this trust in the worst 20% nationally, with staff feeling undervalued, facing a lack of support from senior staff, and reporting work-related stress.

In response to the 2015 CQC ratings, the trust now has a well-developed and extensive strategy, supported by the board, for improving patient experience, which is very much a task-driven approach. This includes development of a diverse range of patient feedback mechanisms and sources of data; different approaches to staff engagement, including dissemination of FFT findings and triangulation of patient feedback to staff; and specific small-scale interventions to support a good patient experience. In addition to these, a sub-board-level patient and carer experience and engagement group has been set up. There is also a dedicated QI team within the trust. When the CQC visited again in 2017, it found that significant improvements had been made and rated the trust ‘good’.

Patient experience data are actively discussed with staff in the trust and changes have been made based on the results from patient experience data collections: changes to the way staff interact with patients; changes to the way care is provided to patients; changes to the layout of the hospital, ward or department; and changes to the transactional aspects of care (e.g. cleanliness, catering).

In May 2016, the ward that had been selected as a case study changed name and location within the hospital, meaning that the staffing team and patients moved to a different physical space and the ward assumed a new identity. This occurred after the US-PEx baseline survey but before the start of the ethnographic fieldwork. Because the ward team remained the same, this was not deemed to be a problem, as the majority of content in the survey focused on either relational aspects of care (staff interactions) or experiences pre and post being on the ward, rather than on the physical features of the ward.

Case study 6: description summary

This case study site is a foundation trust providing general and specialist services to a large population. The trust has undergone significant restructuring following regional NHS changes and staff morale continues be low. In 2015, the CQC published its Quality Report, acknowledging the relative infancy of the trust (i.e. the inspection took place just 5 months after the trust’s introduction) and that a particularly challenging winter period (2014–15) had affected services. Nevertheless, despite these matters, the CQC concluded that the overall rating for the trust at that time was ‘requires improvement’. More specifically, the CQC stated that improvement was required in the areas of safety, effectiveness and responsiveness.

In addition to this, during the study, the trust was placed into ‘special measures’ with regard to finances and resources. The consequence of this is that all activity requiring funding and additional resources has to be scrutinised at numerous levels in order to justify and legitimise spending costs. This extends to any resources that may be required for the US-PEx project.

Furthermore, the hospital was operating at level 4 of the NHS Escalation Management System (i.e. extreme pressure that may affect patient safety). The hospital appears to have maintained this level of pressure for most of the winter 2016–17 period.

Sample size and detecting differences

Ward teams were asked to draw a census of all discharged patients in a given 3- or 4-month period to which a questionnaire would be sent (at two sites, a sample of 250 inpatients was selected due to the much higher number of discharges over the sampling period). Despite this generous sampling period, the before and after sample sizes for sites were small, reflecting the volume of activity in the wards. Thus, the degree of improvement needed to be identified as a statistically significant change is high. This contrasts with studies with very large sample sizes, such as the NHS inpatient survey, where the very high precision of survey estimates means that even quite small changes in scores may be identified as statistically significant. As shown in the individual case study results (see Appendix 1), the CIs around the results were wide due to the small sample sizes.

There were 36 questions in the survey for which the change in the proportion of responses could be compared between the baseline and post-intervention survey for each case study site. As we used z-tests to compare individual pairs of proportions (i.e. comparing the responses to individual response options within a question), a total of 108 individual tests were carried out for each case study site to determine if any change in the proportion of responses was statistically significant. Based on an assumption that there are no underlying differences, we would expect that 5% of these tests would be significant just by chance at a CI of 0.95 or p < 0.05 (i.e. five statistically significant results would occur just by chance in each case study site). At two of the sites (case study sites 1 and 3) only one statistically significant result was shown, while at case study sites 2 and 6, three statistically significant changes in the responses were found. At case study site 4, there were five questions in which the change in the proportion of respondents selecting one of the response options was statistically significant, so it could be argued that given the volume of testing carried out we would expect this number of tests to be significant ‘just by chance’. Similarly, in case study site 5, where eight statistically significant results were shown, five of these eight tests could have been significant simply by chance. This highlights the caution that is needed in interpreting the patient survey results. As discussed in each of the case study survey results (see Appendix 1), a number of questions showed an increase in the proportion of patients responding positively between the baseline and post-intervention surveys, which may have been a statistically significant increase if there had been a larger number of respondents.

Questionnaire design

The questionnaire used for the baseline and post-intervention surveys focused on the experience of four areas of care:

• referral to service

• inpatient care

• discharge

• support for self-management.

The same questionnaire was used for each case study site. The questionnaire was designed before knowing what the case study sites would be and what ward and types of interventions they would choose to focus on. This approach was beneficial because it meant that all sites used a common questionnaire, and the content of the survey was based on existing evidence around the elements of care likely to be important to the majority of patients. This was useful in providing sites with evidence about areas where improvements would be beneficial. However, the ability of the questionnaire to pick up changes in patient experience as a result of quite specific initiatives (such as welcome packs) merits consideration. If changes narrowly address particular elements of people’s experiences – and if changes in these do not lead to changes in other elements of care that are covered in the questionnaire – then the survey would not be likely to show differences.

Timing

The sampling months used for the baseline survey were January, February and March 2016 (with one site also including discharges in April 2016 due to a small sample size). The sampling period for the post-intervention survey was originally planned to cover the same period in 2017, but was deferred to March, April and May 2017 to allow for delays in the implementation of the interventions caused by a very difficult winter. Although ideally the post-intervention survey would have been conducted over the same calendar months to eliminate any seasonality effect on people’s responses, the risk of potentially not capturing the impacts of the interventions was considered to outweigh this limitation.

Introduction

• Background to the project and purpose of the interview: to investigate patient experiences of care.

• Emphasise confidentiality – all personal details to be removed from transcripts so no individual can be identified.

• Interview to last about 1 hour.

• Questions from participant about the interview?

• Obtain verbal consent to turn on tape recorder.

• Signing of consent form (if not already completed).

Opening narrative

• Tell me about your experience of x ward?

Referral

• If your stay in hospital was planned, did you receive any information before you were admitted?

• Can you tell me the types of things the information covered?

  Prompts:

  • Was it helpful/relevant?

  • Was any information missing/would you have liked any more information on anything?

Inpatient care

• What was the ward environment like?

  Prompts:

  • Noise, privacy, cleanliness, lighting, sleep, food, etc.

• What did you think of the care provided?

  Prompts:

  • Medication, dressings

  • Help with personal care – washing, eating, toileting

  • Dignity and respect

• How did the staff treat you? (Prompt for different staff groups.)

  Prompts:

  • Communication

  • Involvement in decisions

  • Involvement of family

  • Treated as an individual with own needs/preferences/values respected?

  • Able to discuss worries and fears?

• Did you feel you got enough information about your condition and treatment while in hospital?

• What was visiting like for you/your family members?

• Can you describe to me some examples of really good quality of care while you were there?

• Can you tell me about some things you thought could be improved?

Discharge

• Can you tell me about the few days leading up to your discharge from hospital?

• When you were being discharged did you receive any written information?

  Prompts:

  • Can you tell me the types of things the information covered?

  • Was it helpful? Would you have liked any more information?

Support for self-management

• When you got home from hospital did you feel able to look after yourself?

• Did you get any support from health or social care staff/or voluntary organisations?

• (If yes) what did this support look like?

Closing

Interviewer to wrap up with importance of gathered information for research project, and thanking participant for their involvement.

Reflections

Developing and applying a common coding framework

In the final phase of all fieldwork, a series of ethnographer data sessions aimed to harmonise the way in which data were analysed within the team. These data included documentary sources, semistructured interviews, field-based observations and individual field notes. The analysis focused on the production of a common coding framework (for use within NVivo 10) that would aim to include all aspects of the data generated from the six case study sites. Following a series of data sessions dedicated to formulating a suitable coding framework, 13 domains (or ‘parent nodes’) were identified by the three ethnographers as the key areas for analysis. These domains included ‘context’, ‘organisational culture and practices’ and ‘quality improvement’ (plus 10 others) that emerged from a shared primary analysis of each respective data set. Thematic analysis then established a wider framework that consisted of 58 separate codes (or ‘child nodes’) across all 13 domains. These 58 child nodes each represented a shared theme or topic that was discussed or observed during fieldwork (interviews and observations) and were each consistent across the various data sets generated by the three ethnographers. In addition to producing a common coding framework, the ethnographers agreed summary descriptions of each parent and child node, and collected these together in a common codebook for reference. Once the codes and codebook were validated for analytical relevance by the principal investigator, each ethnographer subsequently created an NVivo file that reflected the key domains and associated fields.

Ethnographic mapping and analytical inspiration

As noted in Chapter 2, Study methods, one ethnographer (SP) was responsible for the conduct and management of fieldwork at three case study sites, while the others worked on two sites and one site, respectively. Managing three sites created a voluminous and thematically diverse data set. To deal with this complexity, Stephen Parkin devised a system of note-taking and information management to keep the various sites separate, and during a period of reflection began sketching the ‘evolution’ of a QI project at one particular location in his field-note journal.

The need to sketch the evolution of one particular QI emerged from the development and progress of a complex, multifaceted, multicomponent project designed by the ward team concerned. More specifically, the relevant team had devised a QI project consisting of three aims and nine objectives and utilised a variety of (mainly) qualitative QI techniques to reach these various goals. To summarise the various workstreams and methods attached to this project, Stephen Parkin began charting the ward team’s activities and progress as an evolutionary and longitudinal timeline. This evolutionary timeline noted a variety of developments, including positive and negative influences, and ultimately produced a visual representation of the ward’s shared efforts. After recognising the visual impact that this chart produced (in terms of summarising the work conducted by the ward team), he repeated the exercise for two further case studies. This activity produced three very different A4 notebook sketches representing the different aims and objectives of the three case studies concerned, and also visually portrayed three different ways in which the respective ward teams had approached the task of conducting QI on the front line of NHS services. Stephen Parkin then reproduced the sketches in the form of three large-scale process maps using colour-coding and various symbolic representations to characterise the assorted process, progress, barriers and outcomes of the three case study sites’ projects. These hand-crafted maps were subsequently recreated using a software program called Visio (Microsoft, Corporation), which was developed specifically for constructing process maps. The creation of colour-coded, symbolic ethnographic process maps helped to produce an ‘at-a-glance’ description of entire improvement projects and single workstreams conducted on the three wards. The approach was subsequently developed and refined by the team in the data sessions and a dedicated ‘mapping’ meeting. Each of the ethnographers contributed to this process and all three ethnographers produced versions of the maps for their sites.

Whereas conventional process mapping seeks to improve efficiency within organisations and business settings, in this case the result was an ethnographic representation of process mapping, to portray (visually and schematically) the ethnographic object within each of the three locations. Similarly, the collapsed data from these three maps also informed and complemented the range of data sessions and other meetings held throughout the lifetime of the US-PEx study. They helped to identify positive/negative QI, effective methods of working as a ward team and the assorted negative/positive influences on project outcome (all of which are consistent with conventional process mapping). Perhaps more simply, an ‘analytical inspiration’105 unpacked the implicit and explicit narrative contained within the three maps (both collectively and individually) and helped to inform and develop the ‘team ethnography’ approach to the shared understanding and interpretation of how front-line staff engage with QI projects.

Arriving in hospital: the ward is a strange new place

• Admissions: be alert to what happened to people before they got here (delays, anxiety, bureaucratic hurdles).

• We may have been wrenched unexpectedly from caring responsibilities, tasks at home, work, or pets who need looking after.

• We need information – a guide about what to expect (preferably before admission or early on if emergency) – staff need to appreciate we don’t know what the system is, what to do, what the ropes are (e.g. where to find a towel, when to shower, when the lights can be turned out, how does the TV work).

• If we have to bring a list of medication with us, we need to know in advance; we can’t always remember a string of complex names.

• Remember that for the patient the ward is a completely alien and traumatic environment, however familiar it is to staff – we feel vulnerable.

• Whereas staff may not notice noise, lights on at night, the worry of seeing/hearing other very sick people.

• Being moved from one ward to another without notice is disorientating.

• It’s difficult often to know who is who on the ward, with unfamiliar uniforms and difficulty reading badges. Not just ‘hello my name is’ but what your role is.

• Patients who come into hospital from nursing homes are often sent with staff who do not know the details of the person and can’t tell staff much about them.

Communication between patient (and carers) and health professional

• It helps to have eye contact and really good, clear explanations.

• Handouts of what’s been said to help you remember are useful, including for those of us with some degree of memory loss.

• Online information may work for some, but one size doesn’t fit all, and some older people will not use the technology.

• If talking to medical students, please engage the person in the discussion as well.

• Don’t turn ‘hello my name is’ into a tick-box exercise.

• Having English as a second language matters whether you are a patient or a member of staff.

Care/caring/feeling like a person

• We need normal human conversation; nurses may be run ragged, but other staff on the ward may have more time.

• We are people first; the things that made us who we are in the outside world are still true in hospital.

• Sympathy and kindness may come from the most unexpected quarters and they matter a lot.

• Caring as well as nursing matters – emotional and social interaction, not just practical tasks.

• Importance of touch.

• Importance of recognising faces and remembering names.

• ‘Diamond’ nurses – nurses who stand out for their care and compassion – can transform experience.

• Privacy around personal hygiene.

• Continuity of carer, building confidence and trust.

• Timing of pain relief and other medication – when you need it, not when it’s the next drugs round.

• Don’t strip people of normal responsibility and ability to self-manage own condition(s).

• Being able to use own phone and laptop in hospital helps you feel normal.

• Loss of personal effects – valuables, practical things such as teeth and phones, personal things with sentimental value – is very important. Trivialising how much they mean to the person and saying things like ‘it’s not a hotel’ adds to distress.

• Teeth and hearing aids are important for communication; important to remember many medical ward patients will have some hearing loss, or may have difficulty speaking as a result of stroke, or be visually impaired. Staff communication training?

• Difficulty communicating because of dementia is also important – staff sensitivity to confusion and anxiety helps; patience, reassurance.

• Other conditions or disabilities may also need to be taken into account.

• It’s hard to give feedback on the spot if you feel it will affect your care.

• Small acts of care, kindness and putting things right are everyone’s responsibility, no matter how grand they think they are.

• Patients (and visiting family) can offer such small acts to each other.

• Food choices may be unhelpful for those with allergies and the options to choose from can be confusing. Staff need sensitivity about when/whether to ask people about food choices.

• Make sure food, water, buzzers, light switches and TV remote are within reach, especially for people with limited mobility.

Involving carers as appropriate

• Patients have a right to make their own decisions and may not have family, or may not want family involved at some points.

• But patients who want their family there and involved in decisions and care should be able to.

• Who are ‘protected’ mealtimes for? You may want family there to help with food and drink. Longer visiting hours?

• Patients may not want to hear some information, but carers may do.

• Carers have information and emotional needs of their own (Carers UK see it as a triangle).

Communication and coordination across team

• Conflicting messages from and need for co-ordination between different team members.

• Being on the ‘wrong’ ward; need for communications between medical team and host ward.

• Making sure staff know about other coexisting conditions (and allergies) and enable you to self-manage (e.g. diabetes and insulin).

• Getting tests and assessments done in sequence so results are available in time for meetings with the doctor.

• Having notes readily available, not in the archive – and reading them.

Discharge

• Involve the patient in decisions; don’t just tell them it’s time to go.

• Try to get discharge prescriptions sorted promptly so going home is not delayed.

• Co-ordination with primary care (GP, etc.), equipment, etc.

• Patients need a ‘what next’ in writing with them when they go home e.g. ‘see GP within 7 days’, ‘outpatient appointment letter will come by post’, ‘no follow-up needed’, etc.

• Make sure discharge information is accurate.

• Patients need a ‘what to watch for’ in writing e.g. ‘some pain is normal’, ‘if X happens speak to or see your GP’, ‘if Y happens phone 111’, etc. Nowadays, patients are often discharged when they are not 100% well, and need to know when to worry (and when not to worry).