Understanding the healthcare workplace learning culture through safety and dignity narratives: a UK qualitative study of multiple stakeholders' perspectives.

OBJECTIVES: While studies at the undergraduate level have begun to explore healthcare students' safety and dignity dilemmas, none have explored such dilemmas with multiple stakeholders at the postgraduate level. The current study therefore explores the patient and staff safety and dignity narratives of multiple stakeholders to better understand the healthcare workplace learning culture.
DESIGN: A qualitative interview study using narrative interviewing.
SETTING: Two sites in the UK ranked near the top and bottom for raising concerns according to the 2013 General Medical Council National Training Survey. PARTICIPANTS: Using maximum variation sampling, 39 participants were recruited representing four different groups (10 public representatives, 10 medical trainees, 8 medical trainers and 11 nurses and allied health professionals) across the two sites.
METHODS: We conducted 1 group and 35 individual semistructured interviews. Data collection was completed in 2015. Framework analysis was conducted to identify themes. Theme similarities and differences across the two sites and four groups were established.
RESULTS: We identified five themes in relation to our three research questions (RQs): (1) understandings of safety and dignity (RQ1); (2) experiences of safety and dignity dilemmas (RQ2); (3) resistance and/or complicity regarding dilemmas encountered (RQ2); (4) factors facilitating safety and/or dignity (RQ3); and (5) factors inhibiting safety and/or dignity (RQ3). The themes were remarkably similar across the two sites and four stakeholder groups.
CONCLUSIONS: While some of our findings are similar to previous research with undergraduate healthcare students, our findings also differ, for example, illustrating higher levels of reported resistance in the postgraduate context. We provide educational implications to uphold safety and dignity at the level of the individual (eg, stakeholder education), interaction (eg, stakeholder communication and teamwork) and organisation (eg, institutional policy). © Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

This study is the first to explore healthcare workplace learning cultures through multiple stakeholders’ patient and staff safety and dignity dilemmas.

We have incorporated previously unheard voices including fully trained doctors, other healthcare professionals and public representatives.

We have a relatively large qualitative sample and collected a large number of narratives from two UK sites, thus enhancing the transferability of our study findings.

Our four subsamples were arguably small (n=8–11), making comparisons between groups difficult.

Our sample included few participants from culturally and linguistically diverse or low socioeconomic status (SES) backgrounds, meaning that our findings better represent white and high SES stakeholders.

Introduction

The development of professionalism in healthcare students and trainees is paramount.1 While students and trainees are taught good professional practice through ethical codes, they commonly encounter safety and dignity dilemmas relating to both patients and healthcare colleagues as part of the broader workplace learning culture. 2–7 Here, we define dilemmas as: ‘day-to-day experiences in which individuals witness or participate in something that they believe to be unprofessional, unethical or immoral, which causes them some angst’ (p. 2).7

Safety and dignity in the healthcare workplace

A number of recent studies worldwide have begun to investigate healthcare students’ safety and dignity dilemmas at the undergraduate level.5–8 While patient safety has been defined as: ‘the prevention of avoidable errors and adverse effects to patients associated with healthcare’,9 dignity has been defined as: ‘how people feel, think and behave in relation to the worth or value of themselves and others’.10 Here, safety and dignity are often concerned with events involving patients as the direct target of violations.7 For example, in terms of patient safety, research has identified healthcare professionals and students making mistakes through incompetent practice, flouting regulations through poor hygiene practices and engaging in unsafe manual handling.7 With respect to patient dignity, research has found common examples of healthcare professionals calling patients derogatory names, demonstrating physical aggression towards patients and exposing their bodies for longer than is necessary.7 Interestingly, safety and dignity as broader issues have also been addressed by others exploring medical students and trainees as the targets of workplace abuse.5 11 12 This is considered a patient safety issue too, since the impact on its recipients is likely to be detrimental to their performance with patients.11 12 Indeed, we now have extensive literature on the abusive cultures of the healthcare workplace for students, trainees and staff.7 8 11 For example, healthcare students commonly report covert status-related abuse (or indignities) such as being ignored and excluded, being asked repeated questions in intimidating ways, having information withheld from them, receiving unconstructive critical feedback and being given menial tasks.6–8 13 Students also report a raft of verbal humiliations including discrimination and harassment involving protected characteristics and even physical intimidation and violence.6–8 13 Such safety and dignity dilemmas are illustrated starkly in the United Kingdom Mid Staffs Public Inquiry Report and subsequent review, highlighting grim failings involving both students and healthcare professionals (including trainees), relating to serious breaches of patient safety and dignity.14–16 It should be noted that these issues are not unique to the UK but have been documented in a variety of other countries, so could arguably be seen as a global problem.17 Rather than pointing the blame for these breaches at individuals, however, workplace culture has been identified as problematic.

Workplace cultures and raising concerns

‘When a culture is not right in an organisation, it has an impact on the professional attitudes and behaviours of the staff who work for it. Put simply, a toxic culture can pollute good people… through constant change, chronic under-staffing and unrelenting pressure, staff have kindness and compassion eroded from them’ (p. 3).15

Workplace culture comprises the structures and systems of organisations and the social facilitations and constraints that these might have on those working within the organisation.18 Quite simply, organisational culture can be described as: ‘the way things are done around here’ (p. 75).18 That said, it should be borne in mind that culture is situational and so too are definitions of it. The ways in which cultures influence learning range from the superficial to the unconscious and may involve relatively quick or slow processes. Articulating culture as a single entity in, for example, primary care can be considered ‘problematic’.19 Indeed, we should be considering distinctions between different types and levels of culture, thereby privileging both the multiplicity and complexity of workplace cultures.19–21

Workplace cultures matter: positive cultures are consistently associated with positive patient outcomes including increased patient satisfaction and reduced mortality and morbidity22 and with positive staff outcomes including improved communications and enabling high-quality care.23 24 However, a toxic culture can reinforce professionalism lapses but can also inhibit people from ‘doing the right thing’. At the undergraduate level, we often see that approximately half of medical students faced with others’ professionalism lapses do nothing in the face of those lapses, thereby complying with (or going along with) them. 6 13 25 Doing the right thing, however, includes students, trainees and qualified healthcare professionals enacting resistance, either during or after safety and dignity violations.25 26 In terms of during events, students have been shown to enact one or more of the following: direct or indirect verbal resistance, verbal or bodily role modelling, verbally demonstrating concern, disruptive or discreet bodily acts and psychological acts such as emotional withdrawal.25 With respect to after the events, students have been shown to enact one or more of the following: directly raising concerns, discreetly addressing concerns or apologising.25 Despite such resistance, it has long been argued that both patient and staff safety dilemmas are under-reported.11 26 27 Even recent findings focusing on junior doctors suggest that very few medical trainees raise concerns about safety and dignity,28–32 with 58% reporting their uncertainty around receiving the appropriate support should they raise concerns.32 Factors contributing to, for example, patient safety and workplace bullying incidents have been identified at the level of the individual (eg, patient factors such as vulnerability, healthcare professional factors such as skills, competence and workload), interpersonal (eg, communication) and organisational (eg, continuity of care, external policy context, organisational culture and physical environment).11 33 Barriers to reporting patient safety or workplace bullying incidents have been identified at the individual (eg, gender, perception bias and seniority of bully), organisational (eg, cultural censorship, hierarchy and perceptions that reporting would not change anything) and operational levels (eg, lack of training and lack of time).11 26 27 34

Gaps in the literature, study aim and research questions (RQs)

Although research has begun to examine safety and dignity dilemmas across the educational continuum, to our knowledge, none have explored the safety and dignity narratives of multiple stakeholders at the postgraduate level (eg, medical trainees, trainers, other qualified healthcare professionals and public representatives). While the views of both staff and patients have been sought in terms of dignity,26 these studies have historically emphasised the patient rather than the staff member. Given the international significance of professionalism research,35 36 an analysis of patient and staff safety and dignity narratives in the postgraduate realm should help to uncover the complexities of the workplace learning culture.36 Therefore, this study explores the safety and dignity narratives of multiple stakeholders to better understand the healthcare workplace learning culture. We address the following RQs in this paper:

What do stakeholders understand by the terms ‘safety’ and ‘dignity’ in the healthcare workplace?

What types of workplace safety and dignity dilemmas do stakeholders narrate and how do narrators act in the face of those dilemmas?

What factors are expressed in stakeholders’ dilemmas as facilitating and hindering safety and dignity cultures?

Methods

Design

A qualitative narrative interview method was employed consistent with previous undergraduate research.2 4 7 By analysing stakeholders’ narratives, we sought to better understand how they make sense of their experiences, revealing the nuances of the workplace learning culture.7 We used both group and individual narrative interviews to elicit stakeholders’ experiences of safety and dignity dilemmas. This was underpinned by social constructionist epistemology, employing interpretivism as its theoretical perspective, which suggests that there are multiple interpretations of reality and ways of knowing.37

Sampling and recruitment

Prior to recruitment, ethics approval was received from a university-based ethics committee, in addition to National Health Service Research & Development approval where necessary (Since we promised our participants that we would maintain both participant and site anonymity, we have purposely excluded the names of the ethics committees from this paper.). Informed written consent was obtained from each participant immediately before data collection, along with a short personal details questionnaire enabling the researchers to classify the sample characteristics. Maximum variation sampling was employed. Thirty-nine participants were recruited from two areas in the UK (site 1: n=25; site 2: n=14). These sites were chosen as they were ranked near the top and near the bottom (respectively) for raising concerns (ie, whistleblowing and reporting) according to the General Medical Council National Training Survey.28 Recruitment was undertaken at both sites through emails circulated by the Deanery (a Deanery is a National Health Service body in Scotland, Wales or Northern Ireland, with responsibility for postgraduate medical and dental training. In England, deanery functions are incorporated into Local Education and Training Boards.) and the researchers to members of four stakeholder groups: medical trainees, medical trainers, nurses and allied health professionals (NAHPs) and public representatives, via snowballing, word of mouth and posters. Table 1 shows the breakdown of stakeholder types and demographic information across the two sites.

Table 1

Stakeholder type and demographic information for each site

Demographics	Site 1	Site 2	Total
Age group (years)
20–39	8	5	13
40–59	10	7	17
60+	7	2	9
Gender
Male	7	7	14
Female	18	7	25
Ethnicity
White	24	13	37
Non-white	1	1	2
Social class*
1 higher managerial/admin/professional	12	10	22
2 lower managerial/admin/professional	12	3	15
3 intermediate occupations	1	0	1
4 small employers/own account workers	0	1	1
Stakeholder type
Public representative	8	2	10
Medical trainee	6	4	10
Medical trainer	4	4	8
NAHP	7	4	11

*Social classes 5–8 were not represented in the sample.

NAHP, nurse/allied health professional.

Data collection

Participants took part in individual or group interviews (conducted by either GS or SS), which were typically conducted between June 2014 and September 2015, in seminar rooms in hospital settings across the two sites (two interviews were conducted by telephone). The interviews began with participants discussing their understandings of safety and dignity, and then sharing their own experiences of safety and dignity dilemmas. Interviewers typically asked a series of prompts around these narratives (eg, what happened, who was involved, what did you do and why?). Since participants’ views were grounded in their lived experiences, their views about safety and dignity dilemmas and the workplace learning culture became apparent in personally meaningful ways. Interviews continued until participants felt that they had shared their experiences sufficiently.

The majority of participants (35/39) were interviewed individually because they were generally more comfortable discussing their experiences this way. It was also noted that the presence of one or more ‘dominant personalities’ could lead to a lack of balance in the discussion and less input from others in the group. All interviews were audio-recorded with participants’ permission. The initial set of interviews (18 individual, 2 group) were conducted by GS at site 1. To maintain consistency and develop quality, SS listened to six of these interviews before she embarked on the second group of interviews (three individual interviews at site 1 and 14 at site 2). CR listened to three interviews from each site at an early stage and gave feedback to GS and SS to further reinforce consistency and quality. All interviews were audio-recorded, transcribed and anonymised. The total amount of data collected was 27:25:54 (hours:minutes:seconds) with a mean per interview of 45:42 (range 25:20–68:08).

Data analysis

Data were analysed according to Ritchie and Spencer38 five stages of framework analysis, namely: (1) familiarisation, (2) thematic framework development, (3) indexing, (4) charting and (5) mapping and interpretation.38 The first four transcripts from site 1 (one per stakeholder group) were read and independently analysed by GS, CR and LM (familiarisation). Next, the thematic framework was developed. GS, CR and LM came together to discuss their independent analyses of the first four transcripts to develop a thematic framework for data coding (thematic framework development). Note that the themes were mostly developed inductively, but CR and LM were mindful of a previous coding framework developed in the context of undergraduate healthcare students’ professionalism dilemmas.2–7 The full coding framework for the current study can be requested from the corresponding author. Transcribed data, audios of interviews and the coding framework were entered into ATLAS.ti ready for coding (indexing): SS attended an ATLAS.ti training course and listened to all site one interviews conducted by GS and read their transcripts before coding commenced. SS listened to each interview as she coded the transcripts. SS and CR refined the coding framework as more data were analysed. After the coding was complete, SS interrogated the quotations for the different codes to make sense of the data (charting). The final mapping and interpretation stage comprised SS interrogating the coding with respect to the different stakeholder groups and data collection sites to establish similarities and differences in the data between groups. That we found few differences between the groups encouraged us to pool the results, reporting them together to provide a more synthesised and parsimonious presentation of the findings. Additionally, we explored these findings in terms of previously published research at the undergraduate level 7

Credibility was addressed via crystallisation.39 This privileges multiple researchers, perspectives, data types and modes of investigation. In this way, our exploration of the different ‘facets’ of the phenomenon from multiple angles, with four participant groups from two sites, and multiple researchers from different backgrounds, added trustworthiness to the project.

Team reflexivity

While we were an all-female research team, we had some diversity in our academic backgrounds (eg, PhDs in health psychology, cognitive psychology, forensic medicine and clinical neuroscience) and a wide range of healthcare education expertise across our team (at both undergraduate and postgraduate levels). While we had varying levels of experience with the topic of safety and dignity and qualitative research (inexperienced to experienced), we all possessed positive attitudes that qualitative methods could help to identify workplace cultures through safety and dignity narratives. We were mindful of how our previous academic backgrounds and healthcare education expertise could influence our interpretation of data.

Patient and public involvement

While public representatives (including simulated patients and lay representatives) were involved in the data collection for this study (as study participants), they were not involved in the design or data analysis. Participants were offered the opportunity to receive a copy of our findings and gave contact details expressly for that purpose if they wished to do so.

Results

We identified five themes in relation to our three RQs across the four stakeholder groups: (1) understandings of safety and dignity (RQ1); (2) experiences of safety and dignity dilemmas (RQ2); (3) resistance and/or complicity regarding dilemmas encountered (RQ2); (4) factors facilitating safety and/or dignity (RQ3); and (5) factors inhibiting safety and/or dignity (RQ3).

RQ1: what do stakeholders understand by the terms ‘safety’ and ‘dignity’ in the healthcare workplace? (theme 1)

Participants’ understandings of safety

When asked ‘what’s your understanding of safety?’, participants’ conceptualisations could be grouped into four broad categories. First, physical safety relates to the absence of physical harm or injury that could be caused by the environment, equipment or practices. This was further divided into two subthemes: sexual safety (relating to the instincts, processes and activities connected with intimate physical contact between individuals) and non-sexual safety (relating to non-intimate physical contact between individuals including threats of physical violence). Second, emotional and psychosocial safety relates to the absence of non-physical harm or injury, considering social factors and their relationship to thoughts and behaviours. For example, having a sense of psychological well-being and having a supportive team was related to emotional and psychosocial safety. Like physical safety, it was further divided into two subthemes of sexual and non-sexual emotional and psychological safety. Third, systems safety relates to sets of elements working together as part of a mechanism or network such as postoperative surgical reviews or handover processes for staff. Finally, we had a category of other types of safety, which included participants’ conceptualisations of safety that were different to physical, psychosocial and emotional or systems-related such as having sufficient knowledge to deliver a high-quality training portfolio.

It is important to flag here that our participants talked about both patient and staff safety in the healthcare workplace. However, participants’ responses to the question about their understandings of patient safety were often more detailed and complex than simple definitions, including deeper value judgements about relative importance. For example, below, a female public representative talks about patient safety from her perspective, indicating that ‘in some way’ patient psychological safety is more important, focusing particularly on the importance of trust in the patient–healthcare professional relationship (note that we indicate in brackets what type of dilemmas narrators are talking about if they do not explicitly state the type):

Well there’s physical safety… like people knowing how to get out if there’s a fire, like there not being trip hazards… like medication and medical equipment being stored safely out of children’s reach (systems safety)… there’s sexual safety as well, if somebody might feel that they’re being intimidated by somebody, either of the same sex or a different sex, and there’s the sort of psychological safety and the feeling of being able to trust people [healthcare professionals] who are looking after you… in some way it is more important than the physical aspects that people can see… as a patient with a healthcare professional you must feel safe in their hands, you must feel that they’re going to do their best to look after you, they’re going to be honest with you… and if they say they’ll do something they’re going to do it. If they don’t know they tell you how they’re going to find out, or tell you that there is no answer, and just you can then feel safe with somebody. (Public representative, female, site 1, #4)

We also see how stakeholders position psychological safety differently relative to physical safety, either explicitly (as in this narrative), or implicitly, by the dominant themes in their talk (such as in the next narrative from a trainee). Here, the male trainee talks mostly about the importance of healthcare professionals feeling safe, looked after, involved and secure within multidisciplinary teams, which he contrasts with physical safety, alluding to the physical violence that some trainees may anticipate from patients:

I suppose there’s also other kinds of safety… in the sense of being in a… multi-disciplinary team where you feel safe, being looked after by other professional colleagues around you… feeling involved and… feeling safe that your role is secure and you’re doing what you enjoy and that you’re part of a wider team (emotional and psychosocial safety), which is a very different thing to making sure you’re not going to get punched in the face (physical safety)…Working in an environment like this [a locked psychiatry ward] you have to think about safety… you also have to think about it slightly differently. (Trainee, male, site 1, #1)

As mentioned earlier in the methods, we did not identify any particular differences in understandings of safety in terms of the four stakeholder groups or between the two sites.

Participants’ understandings of dignity

When asked ‘what’s your understanding of dignity?’, participants’ conceptualisations were grouped into three categories. First, physical dignity relates to a state or quality of respect (of self or from others) and supported and/or promoted by physical elements in the environment, equipment or practices. Second, emotional and psychosocial dignity relates to a state or quality of respect (of self or from others) and supported and/or promoted by non-physical elements in the environment, equipment or practices. Finally, other types of dignity refers to participants’ conceptualisations of dignity in ways other than physical or emotional and psychosocial, such as referring to a patient by the correct name or maintaining a balance between job performance and behaviour.

Like with participants’ conceptualisations of safety, participants talked about both patient and staff dignity in the healthcare workplace. As with safety definitions, participants conceptualised dignity as similarly complex. For example, in the following narrative, a female NAHP talks about the importance of staff psychological dignity and how this can be maintained by colleagues in the workplace but is sometimes violated by disrespectful patients:

[F]rom a staff’s point of view dignity… is a slightly different… entity… [if] I have to… preserve one of my staff’s dignity in any way then I make sure that if I have to talk to them about any particularly sensitive subject it’s… not spoken about in an open forum… how I talk to them about whatever topic it is, it depends on the sensitivity of the subject… if you’re working in a group of staff that… you know, we’re polite and we use appropriate behaviour (emotional and psychological dignity)… that kind of thing… it’s difficult if you have patients and other service users that are treating you in an undignified manner (emotional and psychological dignity)…. (NAHP, female, site 1, #11)

In the following quotation, a male trainee also stresses the importance of staff emotional and psychological dignity through mentioning how staff dignity can be violated by their senior colleagues:

… I’ve been in situations where colleagues have been possibly ridiculed a bit or had their dignity taken away by senior colleagues (emotional and psychological dignity)… (Trainee, male, site 1, #1)

Another participant, a male trainer, talked about the importance of patients’ physical and emotional and psychological dignity using an example where he refrained from conducting a physical examination of a patient wearing an abaya in order to uphold her dignity:

Now the problem with cerebral palsy is it requires me to have a lot of hands on examination of muscle tone… so I had a long conversation with mum about this… mum encouraged her daughter [who was wearing an abaya] to get up on the couch for me to examine her. We had a chaperone in the room to try and minimise the dignity hit but she would not… allow me to examine her… because I was a male from outwith her culture (physical, emotional and psychological dignity). So, I had to stop, didn’t do it [the physical exam], couldn’t progress the consultation. And so in my judgment in that situation her dignity trumped clinical need because there wasn’t an urgency about that situation. (Trainer, male, site 2, #1)

What is interesting about this narrative is the complex interplay between dignity and safety, with dignity trumping safety in this particular instance. However, other participants narrated situations where they felt that safety should be privileged over dignity, for example, when patients are physically exposed in order to perform procedures safely. The same male trainer, for example, continues by explaining that his decisions (about what gets privileged when patient dignity and patient safety are at odds) depend on the context, such as the level of medical urgency:

[I]f somebody comes into resuscitation and… requires their clothing to be removed so that we can treat them, then that’s an absolute safety issue (physical safety) and their dignity I’m afraid has to wait (emotional and psychological dignity). But as soon as the time comes where we can cover [them] up with a cloth, or a robe… some bedding, we should do that (physical dignity). So we should restore dignity and respect it as soon as we can when we have to compromise it… what we shouldn’t be doing is having some kind of embarrassed awkwardness around doing a thorough job… to maintain dignity and thereby compromising safe practice. That would be a disaster, in my view. (Trainer, male, site 2, #1)

As with safety, we did not identify any differences in participants’ understandings of staff and patient dignity across our four stakeholder groups or two sites.

RQ2: what types of workplace safety and dignity dilemmas do stakeholders narrate? (themes 2 and 3)

Contextual features of the narratives

We identified 212 personal incident narratives (PINs) across our data (49 from trainers, 57 from trainees, 59 from NAHPs and 47 from public representatives). The average number (per person) of PINs given by trainers was 6.25 for site 1 and 6 for site 2; for trainees, it was 5.8 (site 1) and 5.5 (site 2); for NAHPs, it was 5 (site 1) and 6.75 (site 2) and for public representatives, it was 5.85 (site 1) and 6 (site 2). While the specific settings for the dilemmas were not always clearly identified in the narratives, safety and dignity dilemmas occurred across a range of contexts with the majority occurring in general medical hospital settings. Other environments included (in decreasing order of frequency): surgical settings, non-clinical settings, patients’ homes, mental health settings, general practice settings and residential care homes. Although the key actors within the narratives were not always clearly identified, medical trainers and consultants were most commonly identified as agents (ie, perpetrators) of safety and dignity dilemmas and patients as the objects (ie, targets) of dilemmas. Other agents included (in decreasing order of frequency): nurses, patients, trainees, AHPs, patients’ family members, carers and/or friends and medical students. Other objects included (in decreasing order of frequency): trainees, AHPs, medical trainers and consultants and nurses. Interestingly, while trainers and trainees typically identified themselves as agents, AHPs mostly identified AHPs and patients as agents. Furthermore, participants from the different groups typically told stories involving themselves and/or patients as objects of dilemmas, except trainers who commonly told stories where trainees and patients were the objects.

What are the dilemmas about?

Patient safety dilemmas

We identified six main types of patient safety dilemmas, often perpetrated by staff and impacting on both physical and emotional and psychological safety: (1) staff lack of knowledge or incompetence (n=33); (2) differences in staff clinical opinions (n=12); (3) staff requests to trainees to act beyond their capabilities without adequate supervision (n=9); (4) poor staff hygiene practices (n=4); (5) staff acting beyond their capabilities through their own volition (n=3); and (6) staff practising for their own learning needs causing patient discomfort (n=2). Fifteen further patient safety dilemmas were identified, which were isolated cases and did not fall into any of the above categories, for example, a blind patient being unable to locate their meal tray, consequently knocking their meal onto the floor.

The following narrative provides an example of the most common type of patient safety dilemma, that is, staff compromising safety through lack of knowledge or incompetence. Here, a female trainee narrated her junior doctor colleague’s experience of witnessing suboptimal care by non-medical healthcare professionals. In her narrative, the participant alleges that the suboptimal care resulted in the death of a patient, leading to an inquest with the junior doctor being signed off from work with stress:

There was a colleague I worked with last year… but… when [s/he] was really early on in [their] training… [they’d] been with a patient that ended up dying. [They’d] been looking after them but obviously [s/he] was the most junior on the team and the patient… was just really unwell, but then also had this big wound… instead of checking what… clotting was like, [other healthcare professional] then went and proceeded to sort of mess about with this wound and it just starts bleeding and the patient ended up dying… I felt awful for the F1 [Foundation Year 1 doctor] last year… because [s/he] was left in a position where [they] didn’t know how to deal with the situation cause [s/he] was too junior… [they] didn’t feel very comfortable with escalating procedures to any of the seniors at the time… and it was awful because [they] ended up having to go to… an inquest into this patient’s death… for somebody so early on in their career it’s just horrendous… [s/he] was off from stress… after um this happened (Trainee, female, site 2, #3)

Staff safety dilemmas

We identified four staff safety dilemma types, often perpetrated by patients, family or carers, colleagues and the physical environment and impacting on both physical and emotional safety: (1) physical attacks or threats by patients, family or carers (n=17); (2) physical hazards in the work environment (n=9); (3) patients, their family or carers affecting staff emotional well-being (n=4); and (4) colleagues affecting each others’ emotional well-being (n=3). Finally, seven further staff safety dilemmas were identified, which did not fall into the above categories. Examples of these included a colleague being physically attacked out of hours by another colleague and a doctor putting themselves at risk of infection in order to save a dying patient.

The most common staff safety dilemma—risk of breaches of physical safety through physical attacks from patients and their representatives—is illustrated next. Here, a female trainer reports that one of her male patients with mental health problems punched a male nurse in the back of his head. She describes the nurse as being fine, returning to work the next day, explaining that physical violence was an occupational hazard within mental health services:

[the patient] had apparently been relatively fine… made a demand for a fork… but the staff were in the process of counting forks which you’d have to do so you don’t lose any weapons and so he (the patient) was asked to wait… that precipitated an assault on the member of staff… and that led to people being a bit unsure of this patient for quite a while and wanting him to move to a higher level of security, which I didn’t allow to happen because I didn’t think it was appropriate… that’s the kind of most recent assault I can think of… Well he’s got paranoid schizophrenia, he’s also got antisocial personality disorder and I think had probably been using legal highs… so had destabilised his mental state a… bit and [he] had assaulted a member of staff in that context… it was [a] nurse who was punched to the back of the head… having approached from behind, a male nurse who didn’t suffer any significant injury… The nurse was fine, returned to work… when his next shifts were due… didn’t talk about [it], well… it’s a kind of occupational hazard ((laughs)) I mean there would have been a debrief and incident reporting and stuff but… again that wouldn’t be a very unusual thing in our service for someone to be assaulted by a patient. (Trainer, female, site 1, #3)

Patient (and family or carer) dignity dilemmas

Here we identified six key areas in which patients, their families and carers were the focus of the dignity dilemmas, with staff members being key protagonists: (1) staff compromising physical aspects of dignity (n=21); (2) staff inappropriate or inapt talk (n=8); (3) staff lack of cultural competence (n=8); (4) staff breaches of patient confidentiality (n=7); (5) staff failure to provide adequate information (n=6); and (6) staff professional practice conflicting with patient care (n=2). A further seven different, isolated patient dignity dilemmas were identified, which did not fall into the above categories. Examples of these included a GP calling a blind patient’s name in the waiting room and then walking away and a staff member becoming a patient in their own operating theatre and consequently being physically exposed to their colleagues.

One of the most common patient dignity dilemmas—risk of inappropriate or inapt talk from staff to a patient—is illustrated below. Here, a female trainee recounts seeing a patient on a stroke ward with suspected multiple sclerosis who was given information about her condition by the senior doctor in an abrupt manner and with no attempt to maintain her privacy:

I remember in a stroke ward there was a patient who, it wasn’t a stroke it was potential multiple sclerosis and a woman… had come in and had sort of pins and needles and these sorts of things and… the brain scan… showed lesions on her brain but… we couldn’t really say, ‘oh, it looks like this’… we can’t rule anything out… and it was a difficult situation really… I went up with… one of my seniors, and he went in and she [the patient] was really, really lovely, she was quite young… maybe in her forties and she… looked at us and actually offered him a chair next to her… he didn’t take the chair, he was like ‘no, no, I’ll stand’… he didn’t draw the curtains around so that meant that the… four patients were in ear shot… I mean even with curtains you still hear, but it’s just somehow it changes everything when it just feels very open… and had quite a private conversation really with these [curtains] quite open and… I ended up pulling them round… he basically gave her information in quite an abrupt way… she was fuming afterwards, she really was really upset by the whole thing… and I sort of sat with her and apologised really for the situation. (Trainee, female, site 2, #3)

Staff dignity dilemmas

Here we identified three types of staff dignity dilemmas discussed by more than one participant: (1) verbal humiliation by one staff member towards another (n=27); (2) humiliation and mistreatment situations by patients, their family or carers (n=11); and (3) staff ignoring or shunning colleagues (n=6). A further five different staff dignity dilemmas were identified, which did not fall into the above categories, for example, a trainee’s trouser seam ripping as he ran along a corridor.

We illustrate the most common staff dignity issue—verbal humiliation from other staff—below. Here, a senior female trainee recounts two separate verbal humiliation incidents. In the first, she witnessed a registrar being condescending towards a junior doctor in front of other colleagues. The second, which involved herself, comprised a consultant not accepting her clinical opinion despite her being close to becoming a consultant herself:

On a surgical ward I saw a registrar speaking with one of the juniors. One of the juniors had contacted them [the registrar] by page and asked them to come because they wanted to speak about a patient who wasn’t very well, and the registrar spoke to the junior doctor in quite a condescending, not a particularly nice way within earshot of me and at least three other healthcare professionals… and my first thought was ‘gosh, I wish if the registrar had had an issue with that junior that he’d spoken to them on their own, rather than humiliating him almost by speaking like that in front of all of us’, so yes… that’s certainly a dignity issue… I don’t think that that’s uncommon unfortunately and speaking from my point of view, something happened to me recently… with respect to a patient that I’d seen and given advice about, and the consultant phoned me back and said essentially that he didn’t really accept my opinion and wanted to speak with my consultant ((laughs)) and I said ‘I’m sorry my consultant’s not available, he’s on holiday for two weeks’… I didn’t say ‘I’m nearly at the end of my training, I could be a consultant in a month’ because I just thought… ‘that’s petty, I’ve given you my opinion, I [think] it’s a reasonable one’… I think it’s a shame that, that medicine is so hierarchical. (Trainee, female, site 1, #5)

What actions (resistance and complicity) occur in the face of dilemmas?

Our coding framework included whether the opportunity was narrated for compliance or resistance within the narratives, alongside ways of resisting. In practice, many PINs involved secondhand narratives (ie, situations where the narrator was not actually present), so it was not always possible to determine whether there were opportunities for compliance or resistance, nor the reasons given for resistance or complicity. Of our 212 narratives, only 87 (41%) narratives presented a clear opportunity for either compliance or resistance. In these 87 narratives, 27 (31%) demonstrated compliance (such as in the above quote with the trainee not saying anything to the consultant about nearly being a consultant herself: ‘I didn’t say ‘I’m nearly at the end of my training, I could be a consultant in a month”’), while 60 (69%) demonstrated resistance (such as the previous trainee closing the curtains: ‘I ended up pulling them round’, and then later sitting with the patient and apologising for the patient dignity violation: ‘I sort of sat with her and apologised really for the situation’). Across these 60 resistance narratives, we identified 148 expressions of resistance. The most common form of resistance articulated in the narratives included: directly raising concerns about the dilemma afterwards (n=30); direct verbal challenges during the event (n=27); indirectly raising concerns about the dilemma afterwards (n=25); indirect bodily challenges during the event (n=23); direct bodily challenges during the event (n=16); indirect verbal challenges during the event (n=16); and withdrawing verbally or emotionally from the experience (n=9). See table 2 for illustrative quotes.

Table 2

Illustrative quotes for resistance

Type of resistance	Context	Illustrative quote
Directly raising concerns about the dilemma afterwards	Patient dignity dilemma involving the agent (a nurse) saying something that embarrassed a patient while the narrator (a consultant) is seeing the patient. The consultant resists by speaking with both the nurse and the patient afterwards.	‘I was saying to [the patient] ‘how are your bowels doing?’… [the nurse] shouts down the ward ‘oh the bowels are ((laughs)) moving fine’… so I apologised to the patient… and then I did speak to the nurse afterwards’ (trainer, female, site 1, #5).
Direct verbal challenges during the event	Staff dignity dilemma involving the agent (an elderly patient) in general practice centre inappropriately touching an AHP (the narrator). The AHP resists by verbally challenging the patient.	‘[I] picked him up from the waiting room… he just grabbed me on the bottom… I did say ‘excuse me!’ ((laughs))… and he said ‘oh it’s okay’… there was never any recurrence’ (AHP, male, site 1, #4).
Indirectly raising concerns about the dilemma afterwards	Patient dignity dilemma where the agent (a nurse) is forced to have a sensitive conversation with a young patient with cancer in a corridor and resists by making a complaint to management afterwards.	‘It’s not something that could wait so we ended up having a conversation… in the corridor of the day unit… and I did make a formal complaint to management about that ‘cause that was just ridiculous’ (AHP, male, site 2, #5).
Indirect bodily challenges during the event	Staff safety dilemma involving the narrator (an AHP) arranging for a colleague to be in the room while they X-rayed a patient with whom they felt intimidated.	‘I just asked them to wait… and just said ‘right I’ll just go and get things organised and I’ll be back in a minute’… and got a colleague, someone else just to come in sort of… lurk around in the background just so that they were there just keeping an eye on what was happening’ (AHP, female, site 1, #9).
Direct bodily challenges during the event	Patient safety dilemma where the narrator (a patient’s relative) quietly cleaned her aunt’s overgrown and dirty nails but did not complain to the staff.	‘Her fingernails were absolutely filthy… and very overgrown… there had been an outbreak of sickness and diarrhoea on the ward… she couldn’t wash her own hands properly… [so I] cleaned [her hands and nails] properly, filed them and walked out without saying a word to any of the staff’ (public representative, female, site 1, #2).
Indirect verbal challenges during the event	Patient safety dilemma where the narrator (a trainer) witnesses a trainee making mistakes while teaching another trainee how to intubate a patient. The trainer resists by offering to show the trainee an alternative way to do the procedure.	‘I could see that [the trainee] wasn’t really teaching properly… the other trainee couldn’t do it, so I said ‘right, okay… can I show you a different way of doing that [which] might actually be easier for you?’ so I took over the airway… the other trainee recognised that I – without any verbal communication… had not thought that how he was handling the situation was going very well’ (trainer, female, site 2, #5).
Withdrawing verbally or emotionally from the experience	Staff dignity dilemma where the narrator (a trainee) witnessed the withdrawal of a colleague after she had been shouted at by a consultant in the middle of a ward round.	‘This poor junior doctor was absolutely mortified… heat radiating off them… their head was down and they were sort of shaking a bit… I kind of wish one of us had said something… I didn’t hear her speak a word for about a week… she looked haunted’ (trainee, male, site 1, #1).

AHP, allied healthcare professional.

RQ3: what factors facilitate or hinder safety and dignity cultures? (themes 4 and 5)

Factors that could facilitate or hinder safety and dignity were coded as personal, interpersonal, organisational and material. Although many of the facilitators and inhibitors were mirrors of each other (eg, communication as a facilitator and lack of communication as an inhibitor), this was not always the case, so we present them separately in this paper. As can be seen below, the most frequently cited facilitating factors (in decreasing order) were: interpersonal and organisational, with far fewer individual and material (ie, non-human) factors cited. This differed from the inhibiting factors, which were (in decreasing order of frequency): organisational, individual, interpersonal and material, with larger numbers of citations for individual and material inhibiting factors than there were for facilitating factors.

Personal facilitating and hindering factors

Here we consider how aspects relating to individuals facilitate (n=11) or hinder (n=27) safety and dignity. These aspects comprised a range of factors such as individuals’ attitudes, attributes, skills, behaviours, competence, decision making, responsibility or reputation. In terms of facilitating factors, the narrative below illustrates the light-hearted attitude of one trainee who, when he found himself physically exposed in front of staff and patients through a trouser-related accident, led him to enjoy the moment and not feel that he had lost his dignity:

I ended up just laughing… there was a bit of like squirming at first, like trying to preserve my modesty but then I was like do you know what, this is ridiculous, this is never going to happen to me again, so let’s just enjoy the moment. (Trainee, male, site 1, #1)

However, not all individual factors facilitated patient safety or dignity. For example, in the following narrative, we see how the same trainee lacked communicative competence (an individual hindering factor). Here, we see how this trainee applied the general communication rules learnt in the clinical skills environment to patients in a locked psychiatry ward, leading to a situation compromising his own safety:

They always tell us in Clinical Skills to get down to the patient’s eye level, so I squatted down next to his [the patient’s] bed and looked at him in the eye… now getting down to his level that’s a big risk for safety… all these things that I’d been taught to do as a junior doctor were a big no-no in this situation… and he suddenly leapt… I just remember thinking ‘gosh you are such an idiot… why did you get into this situation?’ (Trainee, male, site 1, #1)

Interpersonal facilitating and hindering factors

This concerns how aspects of interpersonal relationships serve to facilitate (n=27) or hinder (n=19) safety and dignity. Such interpersonal relationships included the level of support within trainee–trainer relationships, opportunities for informal debriefing, the level of professional respect demonstrated, cultural and religious norms and expectations, level of comprehension of information and managing expectations. An example of a facilitating interactional factor is in the following narrative where a paediatric audiology trainer narrates how she was very keen to be a good mentor for trainees and medical students under her care by explaining their mistakes, providing emotional support and reassurance:

It’s a lot about supporting them [learners] through doing it and then doing kind of reflective practice with them and talking about what went right, what went wrong, and… how can they learn from that and, and not put people in a bad place. (Trainer, female, site 1, #2)

Alternatively, the lack of support and understanding given to one trainee by her registrar (an interactional hindering factor) led to a patient safety issue in which the catheterisation of a child went wrong:

I was asked to catheterise a child, and I tried to catheterise the child and wasn’t initially successful. I called the registrar, said I wasn’t happy… to try again, he just told me to get on with it, so I did… I didn’t have the catheter in properly… potentially I could have done a lot of damage. (Trainer, female, site 1, #2)

Organisational facilitating and hindering factors

We identified how personal and interpersonal aspects facilitated and hindered safety and dignity and identified a number of organisational factors. For example, factors such as hierarchy, organisational systems, staffing levels, legal processes, standard operating procedures, communication between departments and operational priorities were present as facilitating (n=25) and hindering (n=31) factors. In terms of facilitating factors, faculty development was identified. For example, the following narrative highlights how a patient-related ‘aggression incident’ led to greater engagement with an aggression management course designed to teach staff how to diffuse situations involving aggressive patients or family members:

Since that issue, what happened was the audiologist and several of us actually hadn’t done our aggression management training for a long while ((laughs)) so we all updated that and did, did the course. (Trainer, female, site 1, #2)

In terms of hindering factors, the lack of NHS resources to train and have more staff available to care for patients with dementia on general medicine wards was narrated as an issue:

Everybody knows there’s cutbacks in the NHS. Our staff are getting less and less. We’re doing the bare minimum assessments and sometimes… you’re just going with your gut instinct… because we can’t do the full range of assessments we’d like to do before they go home… We’re all just kinda doing the bare minimum. (NAHP, female, site 1, #6)

Material facilitating and hindering factors

Finally, we consider how aspects of the material environment (eg, physical security, physical layout, signage, crowded or intimidating environments, implementation of safety procedures and condition and availability of medical equipment) serve to facilitate (n=1) or hinder (n=12) safety or dignity. For example, one participant narrated how the provision of physical security measures in an intensive psychiatric care unit facilitated safety:

They think about risk factors and they think about other things to keep people safe… when you walk on there ([unit], there’s two separate locked doors, one of which is a big metal thing and you’ve got cards, you’ve got keys, you’ve got someone checking you are who you say you are. (Trainer, male, site 2, #1)

However, a lack of attention to physical safety was a hindering factor. An example of this was where a substance misuse clinic moved to new premises without considering the necessary physical safety measures:

My environment isn’t as safe as it should be because we’re aware of the safety, we moved down here… and because… we’re coming down from [name of premises] which does have all those… safety features, we were a bit taken aback that there weren’t the same safety features here. (NAHP, female, site 1, #4)

Bringing together all contextual and conceptual themes (themes 2–5)

We can see how our inter-related themes play out in a couple of narrative examples. In the narrative presented in box 1, a female NAHP narrates an experience she had on a general hospital ward (setting) shortly after qualifying. She explains how a consultant (agent) pulled back the curtains of an elderly female patient (object) who was sitting on a commode during the ward round (setting). She reports that instead of leaving the patient to finish her ablutions, the consultant proceeded to talk with the patient as she sat on the commode surrounded by the healthcare team, thus breaching the patient’s physical and psychological dignity. While the NAHP describes her shock, she explains her reluctance to directly challenge the consultant during the event (compliance) but instead reports an act of resistance around indirectly raising her concerns about the consultant with the ward sister after the event (resistance). She reports that the ward sister has already directly raised her concerns with the consultant as she witnessed the patient dignity breach herself (resistance). This AHP shares a multiplicity of factors in her story that contribute to the dignity lapse, specifically alluding to individual factors relating to the patient (eg, ‘very old elderly lady’ and ‘didn’t seem to be upset or distressed’), plus organisational factors relating to time pressure and insufficient staffing (‘really pressured for time’, ‘short time’, ‘short-staffed’ and ‘under so much pressure’). Furthermore, she shares why she did not directly raise her concerns during the event, indicating that she was ‘just newly qualified’ and thus not ‘comfortable’ to raise her concerns with the consultant.

‘I’ve seen this happen a couple of times… consultants on ward rounds just pulling the curtains back and the patients are on the commode and I did see it happen… The one that I saw was in… short stay medicine so that’s really pressured for time. So I can… it’s not right but I can understand why they maybe did it’ cause they’ve got to go round all their patients… in such a short time. So the patient was sitting on the commode. They had a nightie on so they were covered but… the curtains were just pulled back and the patient was just sitting there. Then they decided to discuss the patient in front of them… it was a… female [patient] and then ask[ed] her how she was feeling and everything. It was a, a very old elderly lady. She wasn’t one of my patients but I was a bit… I was totally shocked but at the same time you’ve got to kinda keep composure if you’re with a patient and then I said to the ward sister later about what had happened and she said she’d seen it as well and she’d had a word with the doctor… She [the patient] didn’t seem to be upset or distressed from what I could hear. I couldn’t really see her… because everybody just crowded round… but as far as I’m aware like she didn’t put in a complaint or anything… I had just newly qualified when I was down in [hospital] working… although we’re all part of the same team I wouldn’t really feel comfortable going and saying to the… the doctors, like, ‘you shouldn’t be doing that’… I don’t know how they would take to me coming over and saying that and I can understand they’re so short-staffed and… under so much pressure these days’ (NAHP, female, site 1, #8).

In our second narrative (see box 2), a male medical trainer narrates an experience he had on a general hospital ward (setting). He explains how the patient (agent) follows him (object) into his office and punches him in the side of the face, thereby breaching his physical safety. This trainer shares several factors in his story that contribute to this physical safety lapse, specifically alluding to individual factors relating to the patient (eg, ‘intoxicated’ and ‘saying he had a weapon and that he was going to stab a nurse’), himself in terms of his lack of safety precautions (‘all the things you shouldn’t do in terms of ensuring your own safety’) and interpersonal factors relating to his relationship with the patient (‘he said that he wasn’t sad and hit me for writing lies in the notes’). Conversely, he shares how organisational and material factors facilitated his ultimate safety, ensuring that he was not more seriously injured (‘the kind of support systems and stuff were all activated’). While he appears quite passive in the face of this physical safety violation (note that we do not know who raised concerns by calling the police), he describes how he completed an incident form after the event.

‘I was on call [and] was called to see a man who was intoxicated, who had returned saying he had a weapon and that he was going to stab a nurse. There was a discussion about whether or not we should intervene and restrain him but I felt that the risk of that was too high because he may or may not have a weapon. He was asked to retire to bed to go to sleep… it turned out he wouldn’t, he came into the office as [I] was writing in the notes and asked me what I was writing, and I said ‘I was writing that you’re upset’. He said that he wasn’t sad and hit me for writing lies in the notes, so punched me to the side of the face. At that point the police were called, he went to leave the ward but it was locked and the police arrived. He head butted a female police officer and was arrested… I felt a bit embarrassed actually… that I managed to be attacked… all the things that you shouldn’t do in terms of ((laughs)) ensuring your own safety, so that was a bit embarrassing… the kind of support systems and stuff were all activated and… that was all fine… it was helpful to learn how you fill in the incident forms yourself and all that kind of thing as well ((laughs))’ (Trainer, male, site 1, #2).

Discussion

Summary of key findings and comparison with existing literature

In terms of our first RQ, our multiple stakeholders (including medical trainees and trainers, other healthcare professionals and public representatives) offered two key understandings regarding patient and staff safety and dignity, that is, physical and emotional and psychosocial, plus a third systems-related understanding of safety. Our stakeholders’ conceptualisations indicated an interplay between safety and dignity (with these concepts sometimes seen as contradictory), plus an interplay between, for example, the dignity of patients and staff. While previous research has distinguished between physical (eg, safety of public spaces such as falls, fire and property damage) and psychological understandings of patient safety40 and physical and emotional conceptualisations of patient and staff dignity,11 41 to our knowledge, our study is the first to elicit multiple stakeholders’ understandings of safety and dignity and their complex interplay. Given the diversity across our participant groups and settings, we thought it interesting to find an absence of differences in participants’ understandings by stakeholder group or site.

With respect to our second RQ, we identified 212 safety and/or dignity narratives, most of which took place in general hospital settings and with medical trainers and consultants as the agents and patients as the objects of the dilemmas. These contextual elements of the narratives (eg, setting and actors) are consistent with findings at the undergraduate healthcare student level.7 By far the most common types of dilemmas shared were: staff lack of knowledge or incompetence (patient safety dilemma), physical attacks or threats towards staff by patients, families or carers (staff safety dilemma), staff compromising the physical dignity of patients (patient dignity dilemma) and verbal humiliation by one staff member towards another (staff dignity dilemma). Although existing research has reported similar types of safety and dignity dilemmas,7 8 this previous research has been conducted with undergraduate healthcare students. It is therefore notable that safety and dignity dilemmas are experienced and narrated similarly by postgraduate trainees, qualified healthcare professionals and patients. Where there were clear opportunities narrated for either compliance or resistance within the narratives, narrators mostly reported resistance. Although various types of resistance were reported, the most common forms were direct actions, either during or after the events. These findings differ markedly from the undergraduate healthcare education literature. Perhaps unsurprisingly, trainees and qualified healthcare professionals narrated a higher proportion of resistance narratives (and a higher proportion of direct actions) compared with previous research with undergraduates.6 8 13 25 This no doubt speaks to the multiplicity of healthcare hierarchies in the workplace learning culture; indeed, it is easier to enact resistance (and direct forms of resistance) by stakeholders with higher status in the healthcare workplace.42 Finally, we found more similarities than differences between the sites in terms of the types of dilemmas narrated and narrators’ actions in the face of those dilemmas. This was particularly interesting to us because we purposely selected two sites at opposite ends of the spectrum in terms of rankings for raising concerns.28 That we did not find different levels of resistance in the narratives from the two sites perhaps suggests that the General Medical Council (GMC)28 rankings for raising concerns reflected differential reporting rather than different experiences of safety and dignity dilemmas at the two sites. Indeed, the site less likely to raise concerns (according to the GMC) may have been experiencing the same types and amounts of dilemmas as the site more likely to raise concerns but may have been less comfortable in raising concerns, perhaps due to differences in organisational culture.

Finally, relating to our third RQ, we found a multiplicity of factors at numerous levels (individual, interpersonal, organisational and material) thought to facilitate or hinder workplace cultures of safety and dignity. Interestingly, individual and material factors were mostly seen as hindering, whereas interpersonal and organisational factors were seen as more equally facilitating and hindering safety and dignity cultures. While our factors hindering safety and dignity cultures are consistent with those found in previous research,11 33 to our knowledge, our study is the first to explore a multiplicity of facilitating and inhibiting factors within the context of narratives of patient and staff safety and dignity dilemmas. Indeed, it is through our analysis of narratives that we have uncovered the complexities of the workplace learning culture.

Methodological strengths and challenges

Our study has a number of methodological strengths, as well as challenges, which should be taken into account when interpreting our findings. In terms of strengths, this is the first study, to our knowledge, to explore the healthcare workplace learning culture through eliciting multiple stakeholders’ narratives of patient and staff dignity and safety dilemmas. Although our study findings are broadly consistent with the professionalism literature at the undergraduate healthcare student level,7 our results have shed light on the complex interplay between patient and staff safety and dignity and its contribution to workplace culture18 and in the previously under-researched context of postgraduate medical education. While much of the previous narrative dilemmas research has been conducted with undergraduate healthcare students or trainees,7 we have included multiple stakeholders incorporating previously unheard voices such as fully trained doctors, other qualified healthcare professionals and public representatives. We have a relatively large qualitative sample overall and have collected a large number of narratives across two UK workplace contexts, thus enhancing the transferability of our study findings. Given our focused study aim, robust researcher–interviewee conversations, and our team-based analytic approach, we believe that our overall sample size was sufficient to answer our RQs.43

With respect to methodological challenges, given that we had four subsamples, our sample specificity was lower, with only 8–11 individuals in each of the four groups. This means that our exploration of patterns (ie, similarities and differences) across stakeholder groups should be treated with caution. While our participants were diverse in terms of their stakeholder group, age and gender, they were relatively homogeneous in terms of ethnicity (almost all were white) and socioeconomic class (none of the participants represented social classes 5–8). Therefore, the transferability of our findings to individuals who are culturally and linguistically diverse and from lower socioeconomic statuses may be limited. Furthermore, while we have collected stakeholders’ narratives in this study, we have analysed these using thematic analysis in order to make sense of our large volume of narrative data. While we try to present full narratives in our paper to illustrate the complex interplay between themes (eg, see boxes 1 and 2), other researchers would advocate the employment of narrative analysis. Furthermore, our data were collected between June 2014 and September 2015, so might not fully represent the current NHS. Finally, we caution the reader not to extrapolate stakeholders’ narratives, volunteered freely as part of an ethically conducted interview study, to their actual reporting of safety and dignity breaches within the workplace. Indeed, we think that participants may have been more willing to share their experiences with us than to report those experiences through formal ‘raising concerns’ channels. Interestingly, of our 27.5 hours of data collected, we were asked to redact nearly 10 min of data at the request of two participants, illustrating how sensitive some narrators were to having their experiences (although anonymously) reported.

Implications for educational policy and practice

Despite the methodological limitations of our study, our findings have several implications for educational policy and practice. Given that facilitating and hindering factors are situated at the individual, interpersonal and organisational levels primarily, we present our study implications also at these levels. First, in terms of the individual level, we would argue that all stakeholders within the healthcare workplace should be educated about patient and staff safety and dignity. Specifically, stakeholders need to understand what safety and dignity mean, how best to uphold safety and dignity, how best to enact resistance in the face of safety and dignity violations and how to maximise the multiplicity of factors contributing to safe and dignified workplace cultures. We cannot be prescriptive about how these things should be taught as educational interventions are necessarily complex and their outcomes dependent on context (eg, stakeholder and setting) but a combination of formal (eg, small group learning and patient information leaflets) and informal approaches (eg, workplace observation and feedback, mentoring and supervision) should be considered. Second, in terms of the interactional level, stakeholders need to work together to uphold safety and dignity in the workplace, enacting resistance in the face of violations and developing both their communication and conflict management skills.7 Finally, at the organisational level, healthcare leaders and managers need to develop and, perhaps most importantly, implement policy better to create positive workplace cultures mindful of work design, positive role modelling and monitoring of organisational data.44 Recent studies, for example, have illustrated the difficulties in optimising policy implementation in a workplace, which is constantly under tension between the demands of service delivery and training.45

Implications for further research

Further qualitative and quantitative research is warranted with additional subsamples of medical trainees and trainers, other healthcare professionals and patients from other sites, in order to explore the transferability of our findings beyond the UK context of the publicly funded NHS. Further studies should also strive to collect data from stakeholders who are under-represented in the current study, specifically those from culturally and linguistically diverse and low SES backgrounds. We would encourage researchers to also evaluate the outcomes of any interventions designed and implemented to improve safety and dignity cultures in the workplace. Here we would advocate the use of realist approaches to explore what safety and dignity interventions work, for whom, why and under what circumstances.45