'You're just a locum': professional identity and temporary workers in the medical profession

Internationally, there has been substantial growth in temporary working, including in the medical profession where temporary doctors are known as locums. There is little research into the implications of temporary work in health care. In this paper, we draw upon theories concerning the sociology of the medical profession to examine the implications of locum working for the medical profession, healthcare organisations and patient safety. We focus particularly on the role of organisations in professional governance and the positioning of locums as peripheral to or outside the organisation, and the in ﬂ uence of intergroup relationships (in this case between permanent and locum doctors) on professional identity. Qualitative semi-structured interviews were conducted between 2015 and 2017 in England with 79 participants including locum doctors, locum agency staff, and representatives of healthcare organisations who use locums. An abductive approach to analysis combined inductive coding with deductive, theory-driven interpretation. Our ﬁ ndings suggest that locums were perceived to be inferior to permanently employed doctors in terms of quality, competency and safety and were often stigmatised, marginalised and excluded. The treatment of locums may have negative implications for collegiality, professional identity, group relations, team functioning and the way organisations deploy and treat locums may have important consequences for patient safety.


Introduction
Internationally, there is an increasing shift towards non-standard forms of work such as temporary work (OECD 2015) and more people have 'portfolio' careers which involve them working for shorter periods or concurrently across different organisations, often without a conventional employment relationship (OECD 2014).While this increase in temporary work may yield some economic advantages for both organisations and workers such as increased labour market flexibility, it may also create or exacerbate some problems (OECD 2002).
Marginalisation and disadvantage are recurring themes in the literature relating to temporary employment (Boyce et al. 2007, Casey andAlach 2004) though studies are generally restricted to the non-professional workforce (Cheng andChan 2008, De Witte et al. 2016).Temporary work in healthcare organisations raises important issues such as team function (Wilkin et al. 2017), interactions and relationships between individuals and organisations (Ashford et al. 2007), and patient safety (Blumenthal et al. 2017).Very little research has explored temporary work in health care generally and particularly in relation to the medical profession (Gami et al. 2014).
This paper begins by providing an overview of temporary working in the medical profession and the perceived benefits and risks associated with locum working.It then sets out our theoretical framework, drawing on theories of professional identity.We seek to contribute to this body of theory in two main areas.Firstly, ideas about professional governance, hierarchy and restratification have increasingly recognised the importance of organisations and organisational context, and we explore what this means for locum doctors who have a somewhat distanced and limited organisational connection.Secondly, we examine the nature of intergroup relationships within the profession (in this case particularly between permanent and locum doctors) drawing upon social identity theory.We use qualitative data from locum doctors, locum agency staff, and representatives of healthcare organisations who use locums, to explore the experiences and perceptions of locum working (Evetts 2011, Freidson 1985, Tajfel and Turner 1979, Waring 2014) and to examine the implications for locum identity and group relations.

Background
Temporary working in the medical profession Internationally, the medical profession has seen a growing number of doctors working as 'locums' in temporary rather than permanent positions (General Medical Council 2015, 2018, Staff Care 2017, Zimlich 2014).In the United Kingdom (UK) between 2009 and 2015, the use of locums in National Health Service (NHS) hospitals almost doubled (Moberly 2016).In 2018, 8810 doctors were registered as working primarily as a locum, representing 3.6% of all registered doctors, but many more are thought to undertake some locum work alongside more conventional permanent employment (General Medical Council 2018).All doctors, other than those in their first year of the foundation programme, can work as a locum, usually standing in for a permanent member of staff when they are absent, or when a hospital or practice is short-staffed.Locums work at all levels of the profession; however, their distinctive characteristic is their temporary relationship to healthcare organisations, patients and permanent employees.
For organisations, locum working may be an important way to cope with staff shortages or vacancies, could provide greater workforce flexibility in service provision and may allow them to avoid some of the fixed costs and responsibilities of permanent employment (Fisher and Connelly 2017).For doctors, locum working may allow them greater control over their workload, offer career flexibility, increased income and improved work/life balance, and trade some occupational stability for greater autonomy (Alonzo and Simon 2008, McKevitt et al. 1999, Theodoulou et al. 2018), a phenomenon that has also been observed in other professional settings (Kirkpatrick and Hoque 2006).Work intensification, workload pressures and resource constraints in the NHS and the attendant problems of work stress and burnout may also be important factors, in the context of a wider set of healthcare reforms over the last few decades in the UK and internationally which have introduced markets, competition and managerialism (Greener andPowell 2008, Sheaff et al. 2019).
However, there are concerns about the potential negative consequences of locum working, driven in part by some high profile examples of locum failures in care over recent years (CQC 2010) and by the rising costs of locum doctor employment in the NHS (NHS Employers 2015, Rimmer 2017).Locum doctors can be perceived negatively by patients, other healthcare professionals (Morgan et al. 2000) and NHS leaders (Lind 2017).They can be regarded as less professional (Campbell et al. 2011), less ethical (Salloch et al. 2018), or untrustworthy 'outsiders' who lack commitment and have poor intentions towards the organisation (Chan et al. 2013).Locum doctors are more likely to have complaints made about them than other doctors (General Medical Council 2014) and are often perceived to present a higher risk of causing harm to patients than permanent doctors (Godlee 2010, Isles 2010, Jennison 2013).A recent review found that there was little empirical evidence to show differences in the quality and safety of locum and permanent doctors' practice, and concluded that many of the factors which may plausibly affect the quality and safety of locum practice are more concerned with the way organisations use locums and how they are deployed and supported than with the characteristics of locum doctors themselves (Ferguson and Walshe 2019).
NHS spending on locum doctors has risen rapidly in recent years -medical agency staff were estimated to have cost the NHS £1.1 billion in 2015/16 (NHS Improvement 2017), and a locum pay cap was introduced in 2015 to curb expenditure (NHS Employers 2015).Some policymakers have problematised locums, seeing them as undermining traditional and more costeffective ways of workingfor example they have been described as 'individuals who are actually available to work and are doing sobut in a way that is unfair to their permanent colleagues and is placing an unacceptable burden on the rest of the NHS' (NHS England 2017: 40).

Locum doctors and professional identity: the roles of organisations and groups
The sociology of the medical profession, the evolving nature of professional identity and the way it is constructed and maintained, the processes of professional and organisational socialisation, and the nature of group relationships within the profession have all been extensively researched (Armstrong 2002, Freidson 1985, Waring and Currie 2009) and this literature provides the theoretical context for our study.We focus here on two key themes with particular salience for the study of locum doctors and their professional identitythe place of hierarchies and organisations; and the role of groups within the profession and intergroup relations.
Freidson explored the reordering of work, identity and power within the medical profession as a strategic adaptation in response to contextual change, a process he termed restratification (Friedson 2001).He suggested that internal hierarchies within the profession, delineated by 'elites' and the 'rank and file' emerge to protect the content and context of medical work and autonomy.A number of authors have explored these professional elites and hierarchies and their growing role in governing performance (Chamberlain 2014, Waring 2014).The complexity and fluidity of these elites and the nuanced nature of restratification mean that the way they are manifested may vary in different settings (McDonald et al. 2012;Bryce et al. 2018, Waring 2014).
While hierarchical structures enable governance and allow the dissemination of codified knowledge (Freidson 2001), collegial structures within organisations permit more informal methods of socialisation and governance (Freidson 1985).Collegiality enables the transmission of tacit knowledge, the growth of trust and the preservation of shared ethical norms, such as loyalty to colleagues (Freidson 1985); while also serving to establish status closure and constrain the actions of others (Adler et al. 2008, Waters 1989).Waring and Bishop (2011: 664) have highlighted the importance of 'occupational socialisation' in the development of professional identity, as workers 'acquire not only skills and knowledge, but also shared repertoires, values and motives that contribute to a sense of identification and solidarity within the "community"'.
The importance of interdependencies between identities of professionals and the organisations within which they work has been a key focus in contemporary sociology (Evetts 2011, Waring andBishop 2011).Waring and Bishop (2011: 665) describe the organisation as an influential source of identity as it acts to 'connect and mediate the institutional and interactional aspects of occupational identity and can provide an additional source of 'ontological security'.' Evetts (2011) has contrasted traditional conceptions of professionalism as an occupational value, rooted in notions of professional autonomy through which doctors to a large extent controlled work systems and operated discretionary judgement, with 'organisational professionalism' in which extended managerial and governance arrangements have increased organisational control over professional work.For example, Evetts (2011) points to the importance of quality and safety monitoring systems, such as audits, targets, and performance reviews as means of measuring and demonstrating professionalism within organisations.Further to this, recent evidence suggests that changes in regulatory processes have strengthened interdependencies between doctors and their organisations as organisations have become intermediaries between professional regulators and doctors, enacting regulatory processes and implementing surveillance and oversight at local level on behalf of the regulator (Tazzyman et al. 2019).While Evetts (2011) acknowledges elements of continuity as well as change, her analyses, and those of others (see Muzio andKirkpatrick 2011, Noordegraff 2011) emphasise the importance of the organisation in shaping professional work and professional identities, noting the complexities of professional-organisation relationships, a dynamic which has increasingly seen professionals 're-made' as managers and aligned to organisational priorities.
We now consider the place of locum doctors in relation to concepts like occupational socialisation and organisational professionalism.Locum doctors are a growing but heterogeneous group who may have widely varying levels of prior experience, training and specialisation before working as locums, and they do not neatly fit into a particular professional stratum, nor do they have traditional organisational affiliations.For the professional elite, the growth of locum working may challenge existing models and practices for governing performance and lead to some adaptation or further restratification.For locums, because they lack some of the traditional occupational and organisational affiliations, the place of occupational socialisation in shaping professional identify may function differently than for their permanently employed peers.Locum doctors may be positioned as distanced from or peripheral to the organisation, with important implications for how they interact with the organisational systems explored by Evett (2011) and Tazzyman et al. (2019) We turn now to considering the place of groups (and the relationships between them) within the medical profession.The medical profession is not a single homogenous group, but a complex, layered and overlapping set of groupings with varying norms, values, traits and identities.Friedson observed that 'established professions come to be composed of a number of highly differentiated subcommunities, loosely held together by a common occupational title' (Freidson 2001: 144) and he noted that tensions, conflicts and contradictions between such subgroups are likely.
Reviewing sociological interpretations of professionalism and offering a critique of the overly simplistic perspective on this topic adopted by many in the medical sphere, Martimianakis et al. (2009) have argued that the development of professional identity is more than a list of characteristics or behaviours and is partially dependent on group level processes, such as professional socialisation (Martimianakis et al. 2009, Waring 2014), and the endorsement of others, particularly more experienced senior doctors (Burford 2012, Ibarra 1999, Pratt et al. 2006).Currie et al. suggest that 'professional identity is relational and legitimacy has to be actively constructed and reproduced in relation to others' (Currie et al. 2009: 944).
Evidence from discursive analysis of medical students' identity work suggests that 'becoming' a doctor is a complex transitional process that involves internalising the doctor identity (Monrouxe et al. 2011).The construction of medical professional identity has been described as a collective enterprise and the debate around definition of professional identity has been reframed to capture the social function of identity formation (Wynia et al. 2014).The organisation functions as a collective and recurring form of social action and a forum where people are generally regarded in terms of their group membership and 'seldom seen in their total human individuality' (Becker et al. 1961: 47).Studies of the socialisation of doctors during their medical education have found a complex and iterative process during which professional ideals are thought to be consolidated and outsiders become insiders (Becker et al. 1961, Brooks andBosk 2012).As Hafferty (2018) points out, those whose behaviours are regarded as misaligned with those professional ideals can be excluded or marginalised.When doctors become locums, they may be seen by others to have stepped outside the norms of professional identity (Ashforth 2001), and to be less engaged in those processes which create and maintain that collective identity (Hafferty 2018).
The extensive literature on social identity theory provides a way to understand these intragroup relations within the medical profession and the behaviours and attitudes that may emerge between individuals when they feel they belong to a group (Tajfel and Turner 1979).It posits that the social category into which an individual falls or to which they feel they belong, provides a definition of the self, according to the defining characteristics of the category (Tajfel and Turner 1979).Individuals positively identify with their group and differentiate this 'in-group' from a comparison 'out-group' on some valued dimension(s).Evidence indicates that being categorised as a group member may produce ethnocentrism and competitive intergroup behaviour (Turner et al. 1994).Research has demonstrated in-group bias and the predilection for the in-group to behave in ways that advantage their own group even at the expense of organisational goals (Haslam and Ellemers 2005).
Social identity theory suggests that individuals are cognitively represented as prototypes comprised of interrelated characteristics that represent a stereotypical member of a group (Abrams and Hogg 1990).The consequences of categorising someone as a prototype can include 'depersonalisation' as the individual is regarded as an example of the category placed upon them, rather than possessing their own unique attributes and idiosyncrasies (Hogg and Terry 2000).Contact theory hypothesises that increased interactions between groups can improve relations and reduce intergroup prejudice; resulting in the out-group member becoming 're-personalised' (Brown and Hewstone 2005).Studies of temporary workers in non-professional occupations have highlighted that permanent employees can develop negative attitudes and behaviours towards temporary workers and temporary workers can be stigmatised because of their work status (Boyce et al. 2007, Chattopadhyay and George 2001, Geary 1992).
Temporary working in the medical profession brings to the fore key questions about the organisation of transient professional work and the nature of intraprofessional group relations, particularly between locum doctors and other professional groups.The peripatetic nature of locum working means that locums often practise on the periphery of healthcare organisations and of the profession, and may consequently have a weaker connection to organisational and professional norms and values.This raises questions about how locum doctors' professional autonomy and identity is constructed and legitimised relationally, how group identities and intragroup relationships are constructed and enacted, and the nature of intraprofessional group relationships and behaviours.

The study aims and methods
The aim of this paper is to explore the experiences and perceptions of locum doctors from the perspectives of locums and those who work with them and to examine the implications for locum identity, group relations and patient safety.Our research was part of a larger study investigating reforms to the way doctors are regulated in England [Boyd et al. 2016].Ethical approval for this study was granted by the University of Manchester ethics committee (REC 15028).

Design, participants and data collection
This paper draws on data from a study conducted between 2015 and 2017 to evaluate the development of reforms to medical regulation in England which have required all doctors periodically to show they are fit to practise in order to retain their medical licence [Boyd et al. 2016].In the first part of our study, ten healthcare organisations were purposefully selected to include both NHS and non-NHS organisations varying in type, size, setting and location across England.Potential key informants who were involved in medical staffing and management were identified in each organisation (Harsh 2011) and then invited by email to take part in the study.Sixty-two participants were interviewed for this part of the study.Early data analysis highlighted that many of these participants viewed permanent and locum doctors differently, with locum doctors being regarded as problematic in comparison.This prompted the research team to design and undertake further investigations for the second part of the study with an additional five case study sites, four of which were locum agencies, and one an organisation which provided administrative services for locums.A further 17 participants from these organisations were interviewed, including eight locum doctors, three senior doctors who provided regulatory services for locum agencies, a senior doctor with responsibilities for carrying out annual appraisals for locums, and five other senior managers of locum agencies.
In total, seventy-nine participants were interviewed.Forty-nine participants were women, while 30 were men.Table 1 below summarises the primary roles of our interviewees though it is worth noting that many permanently employed doctors also have some prior experience of working as a locum.
In our first set of interviews, our interview topic guide covered a wide range of issues relating to the oversight of medical staff performance, and the implementation of changes to the way doctors are regulated, including matters such as appraisal, professional development, clinical audit, the management of complaints and systems for clinical governance [Boyd et al. 2016].Although the focus of this fieldwork was not specifically on locum doctors, the oversight of locum doctors and comparisons with permanently employed doctors were often raised by participants.In our further set of interviews with locums and locum agency staff, our interview topic guide was more directly focused on issues related to locum doctors and their experiences in terms of who locum doctors were, how locums were used and managed, how quality and performance issues were dealt with and what their relationships were to the organisations with which they were affiliated.We were also interested to know what the implications of new forms of regulation were for locum doctors and how they differed for permanent doctors.Interviews were recorded, transcribed verbatim and imported into the qualitative data management software Dedoose (Dedoose 2016).

Analysis
An abductive approach to analysis was used which integrated inductive data-driven coding with deductive theory-driven interpretation (Fereday and Muir-Cochrane 2006) and sought to position our empirical findings against a background of the existing sociological theories discussed earlier (Tavory and Timmermans 2013).This approach was chosen as it allowed themes to be inductively identified from the data by examining patterns and variation across the dataset and different locales, and relevant theories of social and professional identity to be used as a theoretical lens to interpret the findings (Timmermans and Tavory 2012).Taking a combined iterative and theoretical approach to analysis ensured that our interpretations were not restricted by theory (Gioia et al. 2013) but rather that theory was used in an exploratory way to make sense of our findings.
Firstly, all data relating to locum doctors were extracted from our first set of interviews by [XX] and combined with the data collected from our second set of interviews to examine locum working.An inductive thematic analysis was carried out to identify broad themes pertinent to locum working (King 2004) such as clinical governance, relationships between staff, team functioning and ways of working.This inductive analysis involved six phases; familiarisation with the data set relating specifically to locums, working systematically across the data set searching for meaning and patterns before initial coding, generating initial codes, refining, defining and naming themes, and lastly, producing a written account of the themes.Frequent peer debriefing provided external checks of this process and involved weekly meetings of the research team to check interpretations against the raw data, verify coding and to discuss the analytic approach (Nowell et al. 2017).Then, a deductive approach was used, in which core theoretical concepts reviewed earlier in this paper were used to explore and interpret the themes derived from inductive analysis.This was an iterative and reflexive process that involved moving backwards and forwards between phases.A reflexive process of discussing themes and interpretation with the research team at all stages of analysis ensured that a variety of perspectives were considered and analysis and interpretation was grounded in the interview data (Cassell and Symon 1994).It is important to note that findings are not intended as generalisable truths about locums or permanent doctors, but rather to serve as examples of experiences and perceptions of healthcare staff in relations to this issue, and to encourage debate about locum working and the implications for individuals, organisations and patients (Barone 2007).

Findings
Our findings are presented under three main themes.The first of these themes describes perceptions of locum working and how downward comparisons between permanent and locum doctors impacted on locum doctor identity and group relations.The second describes experiences of marginalisation and exclusion and how the differences between locums and permanent staff were manifested.The third theme examines how locum working affects organisational goals and performance, highlighting how organisational members can behave in ways which are inconsistent with the goals of the organisation.Within each theme, our findings are analysed in relation to relevant theory.

Perceptions of locums: prototypes and downward comparisons
Locum doctors perceived that they were viewed negatively and these perceptions were largely confirmed by permanently employed doctors and other healthcare staff.Locums felt they were regarded as less trustworthy, generally inferior, less qualified and capable, and more money oriented with lower levels of status and prestige than permanently employed doctors, regardless of their level of experience.
There is this little concern that you're regarded as a lesser form of doctor.Oh he's just a locum.You know, I've heard that phrase before.Oh you're just a locum then. . .The patients maybe too. . .it's not just the staff, it's the patients as well.So there's some kind of lesser qualification . ..you are treated with suspicion.You're slightly policed.I think they look at your outcomes more scrupulously.It doesn't feel good.Sometimes. . .you can see how you might feel a wee bit got at or a lesser being.It's not good for morale really, is it? . . .You don't feel quite so comfortable in your skin if they're looking at you as through you're some kind of numpty.
(Interviewee 8male, consultant locum) Locums described how they were initially treated with mistrust until other staff members got to know them.
The general attitude towards locums is, fine when they know you, but very suspicious until they know you, and so if you're going to somewhere that you've not been before, everybody's watching you and you've got to win trust, and that's not always easy, and there's a reason for that.There are a lot of locums who are really substandard and people are going to worry about them, and they can be dodgy.
(Interviewee 6male, consultant locum) Locums have stepped outside of standard ways of working and as a consequence are perhaps expected by their permanently employed colleagues to re-enter a phase that is reflective of an earlier period of professional socialisation where outsiders worked to become insiders in a limited way with limited responsibility (Wallenburg et al. 2013).Locums were again an 'unknown quantity', attempting to adapt to social norms in new environments, having to prove themselves repeatedly, despite perhaps previously earning their 'rite of passage' (Brooks andBosk 2012: 1625).Permanently employed participants perceived financial gain to be the key motivation for locum work and downward comparisons were made in terms of morality and motivation.Locums were negatively prototyped by their permanently employed peers as being less ethical and more money oriented than permanently employed NHS doctors and were perhaps seen to be diverging from traditional altruistic motivations long seen as central to medical identity and professionalism (Evetts 2011).I've got a sort of slight distaste. ..for individuals who are working as own employees, employing themselves through their own companies in order to get the tax benefit.
(Interviewee 21male, permanent senior doctor, healthcare charity) In contrast, locums cited seeking part-time work, partial retirement, avoiding workplace stress and conflict, improved work-life balance, and greater autonomy and variety as motivations for locum work.
It was making me ill actually. ..I sort of think back and think, maybe the fact that I've just had a coronary artery by-pass grafting has something to do with ten years of extreme stress at work.
(Interviewee 10male, consultant locum) Oh I haven't done [a] substantive post since 2002.I've left the NHS.It didn't suit me to be in the same building, the same place, the same people, going in year in, year out.(Interviewee 4male, Consultant locum) However, this apparent re-assertion of some of the key aspects of 'occupational professionalism' as described by Evetts (2011), such as autonomy and financial control, may be seen differently by others who have assimilated an 'organisational' form of professionalism as the basis for their own professional identity.Locum working was perhaps regarded as a form of identity violation that led to a devaluing of the doctor identity (Pratt et al. 2006).In other analyses, these discourses have been thought to demonstrate clinician-managers' alignment with and assimilation of organisation priorities (Bryce et al. 2018, McGivern et al. 2015).
It was evident from the interviews that permanently employed NHS participants perceived some doctors were motivated to work as locums because they were substandard and unable to get or maintain permanent employment.There was a perception that locum working was a route into work for doctors who were problematic.
There is an issue about the unemployable.There are people down here who are not employable in trusts for various reasons. . .There's a, sort of, vast underworld under there where you can just disappear below the radar.
(Group interview 1male, permanent Consultant doctor working in the NHS with responsibilities for the regulation of locum doctors) Locumsindividually and as a groupexperienced quite profound and sustained downward comparisons, depersonalisation and stereotyping, and their legitimacy as doctors was often implicitly or explicitly challenged.By stepping outside conventional medical career pathways and notions of service to the NHS, they were seen by some as undermining the legitimacy of the profession's collective identity (Currie et al. 2009, Hafferty 2018).Their motivations for locum working were called into question, even though in practice their espoused reasons were often more concerned with rebuilding or asserting quite traditional notions of professional freedom and autonomy (Evetts 2011) or attempting to escape deteriorating working conditions (Kirkpatrick and Hoque 2006).

Maintaining difference: marginalisation and exclusion
Locums reported that they were excluded from developmental processes and opportunities, such as continuing professional development (CPD) and multidisciplinary team meetings in the organisations in which they worked.This meant that opportunities to interact with coworkers and reflect on practice and performance with senior colleagues were reduced.
They had a whole afternoon a week where they did all their. ..they did their CPD meetings, admin, they taught the juniors, and everything, and I used to just get a list of day cases.I mean I could understand that because they were paying me to work and they had a backlog of day case surgery.It made sense to them. ..you are basically there for service, you're not there to be educated.
(Interviewee 10male, Consultant locum) Locums described how their professional needs were not necessarily considered to be the responsibility of the organisations in which they worked because they were not permanently employed by the NHS.Locums were sometimes unable to access mandated performance mechanisms, such as appraisal and audit, and were expected to source and fund their own development externally.Evetts (2011) describes organisational demands for quality control through regulatory processes, such as auditing, as attempts to measure and demonstrate professionalism.Thus, regulatory processes become reinterpreted as the promotion of professionalism.This exclusion from processes used to measure, monitor and manage medical performance could serve to exclude locums from demonstrating their professionalism and could become a mechanism for occupational constraint by means of inability to demonstrate expertise (Waters 1989).
Oh well you're not on the staff so you don't count.This occupational constraint was manifested in the types of work some locums were sanctioned to do.Locum participants felt that while they expected to complete low-risk/low skill routine procedures, work that led to the development of skills was reserved for permanent doctors.
I will do relatively few of the specific type of case, say major abdominal surgery.Because it's not what you're hired to do.You're hired to bash a waiting list of hernias, endoscopies and relatively low impact procedures.Whereas the, sort of, big broad stuff, you're not necessarily there to do, so you tend not to do it.
(Interviewee 8male, Consultant locum) Participants who worked alongside locums confirmed that locums were disenfranchised from collegial support structures and that locum working came with negative consequences that were perceived as inevitable aspects of a locum career.Positioning locums as inferior to permanently employed doctors, combined with negative judgements relating to locum pay, was used as a rationale for exclusion.
If I'm consultant in my hospital my priority is to train my own staff and my own registrar.
To train a locum -that is not my responsibility.And for example if there's only one opportunity for a junior doctor to carry out a procedure I will have to give it to my regular team rather than to a visitor.(Interviewee 1 -male, permanent consultant employed in the NHS with responsibility for carrying out locum appraisals for a locum agency) Overall, our findings suggest that once locums were positioned as a different and separate group within the profession, this allowed others to rationalise treating them differently in ways that seem likely to disadvantage them.Working as a locum meant that doctors' former professional identities were attenuated or replaced with the title of 'locum'.This placed them in a particularly limiting occupational category, with their contribution often devalued or curtailed as a consequence, and this was used in turn to rationalise their exclusion from organisational opportunities for professional development.Perhaps surprisingly, this marginalisation and exclusion extended to systems for managing performancerather than the professional elites seeking to oversee and manage locums like other groups of doctors as Freidson's restratification theory ( 2001) might suggest, we found that locums were often prevented from participating in such governance arrangements.
Organisational performance: the locum disadvantage Permanent members of staff perceived that locum working was not always conducive to patient continuity or safety; while locums felt that patient safety was sometimes jeopardised as a consequence of the ways in which locums were treated by healthcare organisations.For example, locums reported not receiving basic induction and information about local processes, which locum participants felt was associated with mistakes.
As a locum, because time is money, you actually don't get proper induction or training in the IT system, you're just, kind of, thrown in at the deep end.And, I think that, undoubtedly, has a part to play in the mistakes that are made. ... . .I think a lot of the criticisms that are laid at locum doors, is actually of the NHS's own making.They could do more to make sure that the locums are actually familiar with the working. . .The NHS has to understand that, even if it's a locum, they have to be properly trained.(Interviewee 11 -female, Consultant locum) There was a tendency for permanent NHS employees to behave in ways that served local interests even at the expense of the wider NHS or medical profession (Haslam and Ellemers 2005).For example, concerns about locum practice led to locums being 'got rid of' by senior leaders with little or no explanation, and without addressing problems, meaning they were likely to resurface elsewhere in the NHS.
One particular doctor. ..he was asked to leave after maybe a week, give or take.But this doctor, to be completely fair to him, he had great insight into his shortcomings.He felt that he was having difficulty adjusting to the NHS system. . .So he ended up in the scenario where he was busy banging on the door of the consultant, his office, the consultant was hiding in the office rather than speak to this man and explain to him where his shortcomings were.They just wanted him off the premises. ..doctors who maybe have been asked to leave, wouldn't get support from the trust. . .We wouldn't get a reference; they'll just get rid of them.That doctor would then stop taking our calls and responding to our emails, would take their job off their CV and register with another agency as if that had never happened. . .No record of them ever ballsing up a job whatsoever.That's frightening.But that's happening.
(Interviewee 12female, permanent Recruitment Director, locum agency) These findings illustrate vividly the problematic and contested nature of locum medical practice and the risks to patient safety that may arise.It seems that locum doctors were treated quite differently from permanently employed doctors in that organisations took little responsibility for these workers and did not regard them as employees, but as contractors who could be hired and fired at will and were there solely to undertake tasks as the organisation saw fit.This transactional, disengaged stance seemed likely to increase risks to safety both for the organisation and for the wider NHS, but it was perhaps the inevitable consequence of the way the locum doctor identity was constructed as separate and different from other professional groups.Overall, locum doctors might be seen as an extreme example of what has been termed the proletarianisation of medicine (McKinlay and Arches 1985).

Discussion
This study adds to the otherwise sparse empirical evidence base relating to locum working and seeks to contribute to the body of theory on professional identity and group relations in the medical profession.We now turn first to discussing that contribution, and then to considering the implications of our study for policymakers and healthcare organisations.
Freidson's notion of restratification has been widely used to explore the role of professional elites in governing the profession.This literature has tended to focus mainly on those elites and the exercise of hierarchy and authority and has perhaps paid less attention to the nature of what Freidson calls the 'rank and file' of the profession (Freidson 1985: 26).Locum doctors are part of that rank and filethe mainstream of medical practiceand we found that their experiences are often shaped by the behaviours of and interactions with other groups in that rank and file, not just (and even not mainly) by their place in the medical hierarchy, or by the behaviours of those elites.Indeed, their position in the medical hierarchy is quite ambiguousthey are heterogeneous in age, seniority, and experience and do not form a distinct 'stratum' within the profession, yet on becoming a locum this identity seems to replace their existing occupational and professional status, and often positions them at the bottom of the 'rank and file' and on the periphery of the profession and the organisation.That positioning may help to explain why instead of the professional elites adapting or extending systems of governance and performance management to take account of the emergence of locums as a group and to maintain their authority over the whole professional domain, we found they tended to exclude, implicitly or explicitly, locums from those systems.
If professionalism is understood to be constructed in organisational contexts (Evetts 2011) and organisations act as intermediaries in the relationship between professional regulators and doctors, enacting regulatory and surveillance processes on its behalf (Tazzyman et al. 2019), then locums present a challenge because of their looser organisational bonds and peripatetic relationships with both organisations and professional structures and networks.In a sense, we found that the locum identity is a kind of throwback to a past in which many or most doctors worked autonomously and outside formal organisational settings (Evetts 2011, Waring andBishop 2011), But the difference perhaps is that now, in so doing, they are a readily identifiable minority group of doctors, and their notions of professional identity run counter to a set of prevailing norms and values that are a product of modern occupational and organisational socialisation (Evetts 2011, Muzio and Kirkpatrick 2011, Noordegraaf 2011).It may be that the perception of the locum identity as 'half inside, half outside' the profession and the organisation threatens the collective identity of other medical professionals, as well as notions of organisational professionalism (Evetts 2011), with the consequence that locums were regarded by the dominant group as not 'proper' doctors, not fully committed to medicine as a vocation, or to organisational values, and perhaps therefore legitimately marginalised (Freidson 2001).
We turn now to the implications of our findings for policymakers and for healthcare organisations.The locum medical workforce is an important and necessary resource for the NHS but our findings suggest that their need for support in achieving their potential and carrying out their duties safely was sometimes neglected and locums were not always engaged with traditional support structures such as appraisal, professional development and clinical audit.Findings from this study suggest that locum doctors often do not receive equal treatment in practice and face similar difficulties to temporary workers in other sectors in relation to employment rights (Rubery and Grimshaw 2016).
These findings support earlier calls for the debate about locum doctors and quality to extend beyond the monitoring of locum performance to include the working environment (Ferguson andWalshe 2019, McKevitt et al. 1999).Our findings also suggest that poor performance by locum doctors is often poorly managed through exclusion rather than remediation.While the exclusion of a locum from a particular organisation because of performance concerns may benefit that organisation in the short-term, failure to take collective responsibility for the professional socialisation and remediation of locums does not address the risk to patient safety for the NHS as a whole in the long-term.The marginalisation of locums does not contribute towards achieving the goals of improved patient safety and collaboration (Kreindler 2015).
Further research is needed to better understand locum doctors including how locum working intersects with gender and ethnicity for example; particularly given that research has highlighted ethnic and gender discrimination at all stages of medical careers (Esmail and Roberts 2013), and in access to training and career progression (Bruce et al. 2015, Odom et al. 2007, Oikelome and Healy 2013, Rodriguez et al. 2014).Our study suggests that locum working arrangements may affect the quality and safety of care, and more research is needed to explore these aspects of locum practice.A limitation of this study was that locum perspectives came from relatively senior locums.Perspectives from locums at varying stages of their careers, other healthcare workers and indeed patients also need to be understood.

Conclusion
This study concludes that the negative effects of temporary working are present in the medical profession and that bias and prejudice exist in healthcare settings due to differences in work status between locum doctors and permanently employed doctors.Despite their relatively high occupational status as medical professionals, locum doctors individually and as a group experienced many of the difficulties seen in research on temporary workers in other sectors, such as marginalisation, stigmatisation and limited access to opportunities for training and development.Policymakers' concerns about locum doctors have been largely focused on costs and value for money, but our study suggests that quality, safety and effective team functioning may be just as important and merit greater attention.

Table 1
Summary of interviewees' organisational affiliations and roles 'Um, I do count 'cause I do work here and this is really...it's for my benefit, not for yours'."Well in that case, you can sing for it".And that's...I never got it [practice feedback].So it...that's the sort of thing you're up against sometimes. ..you don't really count, so why should we care.And that does come through...it gets under my skin certainly time to time.(Interviewee8male, consultant locum)