An Exploration of Well-being in Former Covert and Undercover Police Officers

Little is known about the stressors of working in covert and undercover policing roles and the impact these can have on the health and psychological well-being of police officers. Extant literature focuses upon the social impact of undercover and covert policing in a democratic society, especially in relation to policing political groups. Presented here are the results of an exploratory study into the lives of former police officers who have engaged in various forms of covert/undercover policing. Utilising semi-structured interviews, in a five-participant case-study design, this research investigates the impact that covert and undercover policing has on the well-being of former officers who have undertaken this role, and how they utilised coping strategies. Data were thematically analysed using Braun and Clarke’s framework (Braun and Clarke 2006). Findings were consistent in that fear of violence was a large factor that impacted the well-being and personal relationships of undercover officers. The paper concludes by outlining pertinent suggestions for future research and considers the implications for covert policing.


Introduction
Historically, the uniformed police officer has, and continues to be, the focus of much academic interest . Whether examining police culture itself, police and institutional racism (Souhami 2014) or the well-being of police officers and stress in policing (Violanti et al. 2017), it is increasingly clear that the research concentrates on the observable, public and visible aspects of policing. This has perhaps led to a distinct lack of literature concerning the more niche, specialised roles in the policing structure, such as homicide investigators who may experience different cognitive and emotional stress when investigating child homicide (Roach et al. 2016;Roach et al. 2018) and undercover police officers. Covert policing, by its very label, explicates not only the obvious, that it is practised by actors who seek to keep their activities invisible, it also implies that the structure, working culture and 'tradecraft' are all designed to remain hidden even from their parent structure. This paper argues that extant literature surrounding undercover policing is limited, and therefore our knowledge on the practices of covert and undercover policing tactics remains profoundly underdeveloped ). Certainly, one must consider Marx (1988) when beginning a discussion on covert policing; however, much of the earlier literature was a discussion of the well-being and psychological effects on undercover operatives of covert investigations (Farkas 1986;Girodo 1984Girodo , 1985. This was largely the trajectory of undercover policing research until the turn of the millennium when the interest became increasingly upon the legality and morality of undercover policing (Choo 1999;Roberts 2000). Currently, the literature on well-being is limited and sporadic (Farmer et al. 2003;Love et al. 2008;L. Miller 2006). Since 2011, the literature has invariably focused on the furore related to the undercover policing inquiry (2018). Some, however, have taken the view that academic research on undercover policing is extremely limited, and thus, we can only conclude that claims about undercover policing cannot be made in the face of limited empirical research (Kruisbergen et al. 2011;Loftus et al. 2016).
Despite the lack of research in this important area of policing practice, one can infer some of the challenges facing undercover officers. These range from isolation from loved ones for prolonged periods to being placed in dangerous lifethreatening situations. Previous research has highlighted that undercover officers may be faced with the threat of serious violence; such officers may experience losing their sense of security, as they are not able to rely on the authority which is derived from a police uniform (Arter 2005). Such responsibilities lead the present paper to argue that policing research must re-visit the impact that such roles have on the well-being of those who undertake undercover roles to protect our society. Protecting the protectors has become an imperative part of the UK Government's agenda (Home Office 2018) and we must ensure that all protectors gain equal coverage and well-being support. In a recent paper by Cartwright and Roach (2020), it has been highlighted that in the past 10 years, psychological sick leave within UK police forces has almost doubled, with 8% of police employees taking sick leave due to psychological problems. Utilising a different approach, the University of Cambridge (Police Care UK, 2019) surveyed 16,857 serving police employees and revealed estimations of up to 20% of employees suffering from post-traumatic stress symptoms. Indeed, such studies demonstrate the substantial impact that policing roles can have on police employee's well-being and psychological health. However, it is argued here that it is unlikely that those in covert roles form part of the picture outlined in the above studies; therefore, it is imperative that research thoroughly explores the impact of this niche area of policing work.
Briefly, subjective well-being refers to an individual's conception of their life satisfaction and their cognitive evaluation about their emotional state (Harrington and Loffredo 2010). How stress can affect the well-being of covert and undercover officers is of concern here, namely, how events are appraised by undercover officers as exceeding their psychological resources, thus conceptualised as stressful (Lasaruz and Folkman 1984). Kavanagh (1986) would regard the above conception of stress as imprecise, overinclusive and finds the operationalisation of the transactional model of stress as questionable. Henceforth, of equal interest is how undercover officers cope with appraised stressful situations, thus, the "cognitive and behavioural efforts to manage" the demands (Lasaruz and Folkman 1984, p. 141).
Events, such as witnessing or encountering deaths (Hartley et al. 2013), attending violent incidents (Liberman et al. 2002Violanti et al. 2017;Waters and Ussery 2007) and being involved in high-speed pursuits and arresting offenders (Anshel 2000), have been identified as activities that can cause acute stress. Others, however, argue that it is an appealing assertion to make that policing is one of the most stressful occupations (Hart et al. 1993). The media often portray the unsavoury aspects of policing, creating a stereotype that policing is unsafe and inherently dangerous (Hart et al. 1994), whilst often ignoring the mundanities. The literature in this area of policing research is indeed growing, but it is argued that the effects of doing policing on well-being should not be viewed as homogeneous (Cartwright and Roach 2020). As a result, current literature is unable to identify which operational aspects of policing work are the most susceptible to causing trauma. For example, crime scene investigators are exposed to deceased bodies, unpleasant smells, grieving relatives and many other traumatic stressors routinely. Investigators working on child abuse investigations are often exposed to traumatic imagery of children being abused, and general investigators are exposed to a range of other stressors that detrimentally affect well-being (Cartwright and Roach 2020). As such, methods for responding to and supporting these colleagues in order to minimise the detrimental impact of their role are likely to be very different and this will be no different to trauma caused by covert operations.
Covert and undercover police officers are described as those who infiltrate the criminal sphere for the purpose of intelligence gathering on criminal activities (Girodo 1985) from the inside (L. Miller 2006). The targets of undercover policing may be involved in a wide variety of criminal activity (Girodo 1991;Ilardi 2018;Schlembach 2018). Covert and undercover policing may involve covert surveillance, or a more intrusive investigation including the use of an undercover officer (UCO) deploying the use of deception (Marx 1988), and possibly engaging in criminal activities .
Undercover work has been identified as one of the most stress-inducing tasks a police officer can engage in (Kowalczyk and Sharps 2017), with undercover drug investigations being one of the most stressful assignments (L. Miller 2006). There are various demands placed on an officer engaging in undercover work, such as upon their emotional state, intellectual capacity and physical state (Macleod 1995;L. Miller 2006). They are also particularly prone to trauma (Love et al. 2008). It has been argued that various psychological symptoms are common to the undercover officer's uniformed counterpart, whereas other symptoms are present only in officers engaged in undercover policing (Kowalczyk and Sharps 2017).
The most common factor associated with stress in undercover policing is the fear of discovery, which can result in increased feelings of anxiety L. Miller 2006). Physical harm and violent retribution (Macleod 1995;L. Miller 2006) are objective dangers facing undercover police officers, should their police identity be uncovered by criminals (Girodo 1985). Critically, the failure to maintain their undercover identity could become a matter of life or death (Kowalczyk and Sharps 2017). Elsewhere, it has been noted that following a 'challenge' to the cover identity by an offender, increases in anxiety were documented, albeit minor, and left no lasting effect (Macleod 1995). In addition, it has been argued that underlying neuroticism accounts for around 5-10% of observed depressive and anxiety symptoms of undercover officers and that police leaders must question the assumption that all police personnel possess the right fortitude to carry out this role (Girodo 1985).
The existence of identity stressors within undercover policing has found less support. It is argued that the media portray a dramatic and extreme picture of undercover policing and the psychological changes it can induce (Girodo 1985). One need only point to "I Was Monty's Double" to see an extreme example, whereby MI5 concocted a plan to deceive German intelligence during the Second World War as to the location of Field Marshal Montgomery, by using a body double (James 1957). The Colonel playing the role of Montgomery described suffering heavily and commenting "As the days went by I slipped into my role so completely that to all intents and purposes I was General Montgomery" (James 1957, p. 166). Nevertheless, the argument is that reappearances of an alter-ego, when returned to the traditional police role, can occur in officers that are already predisposed to dissociative experiences, in conjunction with private rehearsals of the false personae (Girodo 1985;Girodo et al. 2002).
The literature on undercover police well-being is mixed in its findings. In some studies, there were no recorded instances of psychopathology, and termination on grounds of mental health was rare, and in fact, police officers viewed undercover work as positive, inducing personal growth as well as proving the "ultimate challenge" (Macleod 1995, p. 242). Despite observing minor escalations of anxiety, these reduced upon the termination of the operation (Macleod 1995). This contrasts somewhat with Farkas (1986), who found that amongst a sample of 82 current and former undercover officers, symptoms such as feelings of loneliness, anxiety, over suspiciousness and relationship problems were associated during undercover duties, and even persisted once the assignment had ended. To support this, Girodo (1991) found that the severity and shape of psychological symptomatology were uncannily similar to that of a psychiatric outpatient. To potentially mediate the contrasting findings, it is argued that there may be several buffers which limit the impact of stressors, for example, denial, whereby an officer simply denies or unwittingly fails to consider the possibility of physical harm should they be discovered by an offender (Love et al. 2008). This is in addition to support mechanisms provided by the police environment which may counter the impact of stressful factors (e.g. working as part of a cohesive team; supervised in a democratic way and; perceiving their work is of the highest priority) (Love et al. 2008). It is therefore important to recognise the different contexts in which an officer operates to understand the discrepancies between findings, not least to attempt to update them.
Extant academic literature is divided on whether the effects on the emotional and psychological state impact more on current or former undercover officers. Love et al. (2008), for example, argued that higher frequencies of psychopathology are reported amongst former undercover officers, over both current undercover officers and those without any undercover experience, whereas Farkas' data (1986) support the argument that reports of clinical symptomology are more frequent in current officers, rather than former. Similarly, Girodo (1985) recorded that 9% of preoperational, 26% of active and 17% of post-operational officers were at risk of psychological harm. Discrepancies may be explained by data gathering methods, as it is noted (Kowalczyk and Sharps 2017) that interviewing with a recording device can produce dramatically different responses than not using one. Questionnaires can also produce markedly different results (Kowalczyk and Sharps 2017). Over two decades lie between Farkas' study (Farkas 1986) and Love et al. (2008), thus one may surmise that support for undercover officers may have become more professional and effective, which former undercover officers may not have ready access to.
The present paper seeks to add to the extant literature to specifically explore the psychological impact of working in an undercover policing role. As police well-being research develops, it is argued here that this niche area of police work should not be left neglected based on issues with accessing this hidden cohort. This article aims to plug this gap by providing a contemporary investigation into the impact of undercover policing tactics on officers who volunteer for, often, dangerous assignments.
The main aims of the research presented here, therefore, are to (1) outline and describe the existence of stressors within undercover policing and (2) understand the use of coping strategies to mitigate negative impacts on well-being.

Ethics
Due to the nature of the project, potential ethical concerns were identified early in the study. The project was designed in accordance with the principles of The British Society of Criminology Code of Ethics ( 2015) and received ethical clearance from the School Research Ethics Panel of the University of Huddersfield.

Participants
Five former police officers were included in this study. All five were recruited due to their experience (N = 5) along the scale of covert and undercover police practices previously identified. Three of the participants were male, and two were female. The reason that former, rather than current, officers were recruited was largely due to difficulties accessing participants from this niche policing group. Participants were recruited through personal contacts and via an organisation affiliated with the police, who then acted as a snowball (Howitt and Cramer 2017), in order to contact an additional two.

Procedure
A case study research design was chosen for this area of research, seeking to explore (Yin 1981) the nature of covert and undercover policing and the effect on well-being. Cases were chosen on account of their interesting nature, seeking to produce rich in-depth descriptions of the phenomena (Ridder 2017;Yin 2013). In this study, five participants were included, and low sample sizes are typical of case study designs (Wikfeldt 2016;Yin 2013). This is an appropriate design when exploring difficult to reach populations (Yin 1981) such as the one present. Analytical generalisation, rather than statistical generalisation (Onwuegbuzie and Leech 2010), is sought here, by which it is meant "the extraction of a more abstract level of ideas from a set of case study findings − ideas that nevertheless can pertain to newer situations other than the case(s) in the original case study" (Yin 2013, p. 325). Analytical generalisations can be used to understand the scope of a theory and assess in what circumstances it does or does not apply (Firestone 1993). The case study design is utilised in order to say something about similar cases to the ones studied; if subsequent case studies support the findings, then this can be said to be constructing theory (Wikfeldt 2016). Therefore, it is envisioned that this research will be of utility to police psychologists working with this population, in order to support interventions to improve their well-being.
Semi-structured interviews were utilised as the method of data collection. The interview schedule was designed to discover stressors that could be attributed to different types of police stress such as operational and organisational (Abdollahi 2002); further questions were developed from undercover policing literature specifically to look at identity stressors (Girodo et al. 2002). The chosen method sought to unlock the meanings of officers' experiences, perceptions, beliefs and values concerning their well-being and coping as former officers (Mann 2016). A basic interview protocol was developed; subsequently, depending on the participants' responses to these questions, ad hoc follow-up questions helped to draw out and expand upon the salient aspects of each officer's experience (Hughes 2016). The interviews were conducted face to face with each participant in a location convenient to them and digitally recorded, firstly, to allow complete focus to be given in the interview and the formulation of ad hoc questions, and, secondly, to ensure the accuracy of the transcript. The use of a recording device was explained in advance and discussed with each participant. Informed consent was gained from each participant, and the full recordings were manually transcribed (Yin 2018). As Kowalczyk and Sharps caution, the use of a recording device could potentially cause variation of research data (Kowalczyk and Sharps 2017); however, post-interview discussion revealed no further information regarding their experiences, thus indicating that the presence of a recording device had little effect on the validity. Each participant was provided with a copy of their interview transcript, to ensure that participants felt that their identity was adequately protected, and to ensure a correct interpretation of their experiences. Participants are labelled in order of interview, from UCO1 to UCO5.
Participant checking of the transcripts was a means of engaging in reflective practice and is a recognised phenomenon within qualitative research (Berger 2015). Within qualitative research, it is accepted that knowledge and data are coconstructed as a result of the relationship between the researcher and the participants (Finlay 1998(Finlay , 2002a. Indeed, I recognise that I have played a central role in the construction of this knowledge (Finlay 2002b). Upon reading the transcripts, and in selecting the illustrative quotes, I felt it was important for the participants to check these also, to ensure I had correctly reflected the most important experiences that had affected them. This was also utilised to counter my own biases in interpreting the data.

Analysis
Data were analysed using Braun and Clarke's (2006) thematic analysis. Interviews were transcribed and read multiple times to identify themes (Clarke and Braun 2017). Extracts were filed into different codes, which made up the broad overarching theme. Data were continuously referred to in order to refine the codes and themes (Vaismoradi et al. 2016). NVivo 11 (QSR International 2015) was used to facilitate the analysis in order to maintain a current and up to date database of created codes. To ensure the quality of the analysis, it was ensured that the coding process was thorough and thus inclusive of the entire data, rather than a selection of a few vivid examples (Braun and Clarke 2006). Opportunities for participants to 'check' the work were extra safeguards to ensure quality. Transcripts were provided to all participants, and once data were coded and organised into broad themes, the quotes used to support and illustrate the analysis were also provided to each participant to ensure accuracy.

Results
A thematic analysis of the data gathered has allowed an understanding of the stressors that former covert police officers can face, and additionally, the coping strategies that they utilise to mitigate the negative effects of these stressors. Two thematic maps were created, with the final map resulting in three overarching themes, each with sub-themes. First, 'Risks to Well-being' as a theme was identified, incorporating 'Safety' and 'Identity Stressors' as two sub-themes. Secondly, 'Support Networks' was identified, including 'Family' and 'Support at Work' as sub-themes. Lastly, 'Coping' was the final theme, which incorporates 'Reflection', 'Alcohol use' and 'Physical Activity'.

Risks to Well-Being
Risks to well-being is split into two: Safety; and Identity Stressors. The former relates to data discussing feelings of personal safety, the risks that they may have faced, or perceived to have faced. This incorporates data from one officer who was victimised on numerous occasions in quite often a violent manner. The latter theme relates to some of the officers having to utilise another persona, how they felt about deceiving others, the shame some felt, and, finally, the possible effects of maintaining an undercover identity.
The fear of discovery permeated throughout the discussions with the former officers, both during covert surveillance work, and when engaged in undercover deception against targets. Risks, both perceived and real, included fearing violence, worrying about the presence of weapons, or in the case of UCO5, having had a car used as a weapon against them. These fears manifested psychologically into intrusive thoughts for some, who worried that they themselves were being covertly watched and followed on their way home. UCO5 felt that this was the closest to death he had ever been in his undercover career. For UCO5, this fear of discovery resulted in feelings of paranoia, anxiety and hypervigilance around the growing use of violence and weapons by offenders, evidenced by the shared knowledge of other officers being seriously harmed. For UCO1, this is demonstrated by looking in the rear-view mirror to watch for people following and UCO5 avoided locations that reminded the officer of an aggressive encounter. Fears of being sexually assaulted were highlighted by one former officer (UCO3) and argued that as a female, she was at greater risk than her male counterparts, having been involved in an operation where she ended up conducting a test purchase operation in a nightclub toilet cubicle with two men.
UCO5 was acutely aware of the risk of violent retribution: "…and only a few weeks before that a colleague had his camera found and he'd been beaten and left for dead up in [redacted]. So, I was under no doubt what risk I was at." In an account of one of his undercover operations, UCO5 attempted to protect his identity as a police officer from being revealed by challenging one of the targets in a confrontation, who noticed the camera in the buttonhole of his jacket. He utilised verbal aggression in an attempt to cement his offender persona and to reassure the target that he was legitimately illegitimate: Identity stressors were less prevalent within the study; however, where it did occur, the effects were profound. At this juncture, it is difficult to say that this increases with the level of involvement that a covert officer has with their targets. However, one can be pointed to an example where the undercover identity could be easily distinguished from the police officer identity. UCO3 describes: "I could leave it there, that's Lucy, that's not me and could switch off like that." Although UCO3 utilises the word "schizophrenia" when describing her experience of maintaining several identities, the quote above indicates that she managed to clearly distinguish between the different identities that she used as a covert police officer. She was able to 'switch' between the two. If she was able to activate the process to become someone else, then, in this case, the reverse logic is true that she can become her true identity just as easily. This role ambiguity involved suppressing the police persona and activating the false criminal persona, allowing the latter to take precedence. To avoid reappearances of the undercover persona outside of the operational context, UCO3 attributes the concern, the worry and the anxiety of the undercover operation to the undercover persona performing that role rather than allowing these feelings to be felt by her true persona. To frame the problems as belonging to 'Lucy', she effectively avoids the negative feelings as the negative feelings do not belong to her true self. This is quite different from the experience of UCO2, whose adoption of the undercover persona was problematic, in that his belief of having a strong undercover identity was fundamental to keeping himself safe from danger. Based upon someone who he had known to be a bully, he fully immersed himself into being that person and taking on their characteristics, even believing that he was a drug addict, and felt the stigma associated with this.
"And now I don't know who I am…I'm not...I'm nobody, you know what I mean?" It appears that in this case, UCO2 was unable to desist from his undercover persona, even when at home. Role ambiguity and confusion were a large source of stress for UCO2 and showed that if his role was understood by his peers and cover officer, they may have been able to cement that police identity. This would have provided a strong link with which he can latch on to prevent the profound confusion he faced. Compounding this, he subsequently faced a fear of the police, having been involved in an aggressive encounter on the street with police officers who believed he was a drug addict. Being treated like a drug addict perhaps solidified in his mind that he truly was the undercover identity he was acting.

Support Networks
Support networks as a theme were split into two, family and support at work. The former relates to the effects that undercover working had on the family of the undercover officer. The latter relates to how the UCOs were supported formally within the organisation of policing. Findings were mixed; however, the family subtheme was largely quite negative in terms of support received, interspersed with some positive factors. Formal support at work again featured barriers to receiving effective support from colleagues, and superiors, as well as negative experiences within the occupational health framework.
Regarding the family, the former UCOs generally struggled to receive support from their families. Part of this was due to the inherent nature of undercover policing, which naturally necessitates the maintenance of secrecy in all areas, both within and outside of the undercover policing context. Being unable to disclose the nature of their work was slightly disconcerting for some of the participants. Additionally, a common stressor was the uncertainty of keeping regular work hours and being unable to engage in traditional family events such as weddings and parties for example. UCO1 and UCO4 perceived this problem in their experiences, and both described the effect the antisocial hours of their police work and covert work in general had. A quote from UCO4 demonstrates this: "Yes, inevitable. Unsociable hours. It can take a toll on your marriage, in fact it did, it broke my first marriage up but that's another question. So yeah, certainly." Long hours away from home, and uncertainty around work schedules and concerns around safety are noted. UCO5, however, was in unusual circumstances, as he could pick and choose the operations he worked on, to suit his own needs. This would appear to be a rarity amongst policing unless one has a good cover officer system in place. Despite this, the issue with working long hours was central to UCO4's experience of policing, and the effect upon his personal life was extremely great. This is shared with the findings in Arter (2005), who noted that many former officers cited the long and unsociable hours as the central reason that they had left their undercover roles. UCO1 described being in a "bubble of deception", whereby she recognised that to keep her covert work secret, she had to deceive her family when they asked her about her work, describing her activities as consisting of paperwork. Having to deceive the family can also be a cause of stress and anxiety as highlighted by UCO1. This finding is in line with Farkas (1986) who argued that being unable to discuss one's work can lead to heightened anxiety. By contrast, UCO3 and UCO5 strictly managed to compartmentalise their lives so that clashes between their work and family life did not happen. Whereas UCO3 took to framing her problems within the respective compartments, i.e. undercover and home life, UCO5 saw his undercover work as an escape from an abusive relationship with his partner. Unfortunately, UCO2 suffered severely. He was unable to compartmentalise, facilitated by extreme physical appearance change, and such profound changes led to his wife saying he looked "…like a rapist…" and would not allow him to take his children to dancing classes.
Receiving formal support from work was mixed for all the participants for various reasons. Some reasons included the stigma of discussing personal problems and not wanting to appear as weak in asking for support, especially as a woman operating in a typically male-dominated area. Others simply chose not to access the support as they felt they managed to cope well. When two did attempt to access the support offered at the time, they both felt that it was low quality, and UCO2 described feeling dehumanised by the process. Humour was used by UCO5 to mitigate the emotional impact of lifethreatening situations.
"The intel guy came back in and he was laughing, and he said, 'I don't know why he didn't just shoot you, there was plenty of intel there's a gun in the car.' So, we had a laugh at that." Feedback about the occupational health system and counselling sessions was that it was of poor quality and that if one was not reminded about it often, then you would be at great risk of forgetting that it was there. UCO5 reflects on this now and finds that he lacked insight into his mental health at the time. UCO5 believes that a strong cover officer system is integral to the safety and well-being of an officer. The cover officer system in place worked when it was utilised properly, in that the cover officer acts as a liaison between the UCO and the management system. The cover officer, known as the handler, is responsible for the day to day handling of an undercover officer, whereas the controller will manage and supervise the handler, as well as have general oversight over the use of the undercover officer (College of Policing 2016).

Coping
This theme encompasses three subthemes: reflection, alcohol and physical activity. Levels of coping appeared to be quite minimal for the UCOs in this study for a variety of reasons. Reflecting appeared to be quite prevalent, and each officer utilised this practice differently. Alcohol use was utilised by some officers also. Finally, physical activity was generally linked to positive well-being and reductions in stress for most of the officers.
Generally, all participants engaged in reflection. Some employed what can be termed case-oriented reflection, where they would reflect on an operation, and constructively criticise their actions and try to learn in order to improve performance. This could also have been quite simply assessing one's actions in a case, how one may have performed better. This was used positively by the participants; however, once they considered the risks they were placed under, and the harm they perceived they had caused, then this led to rumination and feelings of guilt and shame. This largely affected UCOs 1, 2, 3 and 5, whereby negative attitudes began to creep into their processes of reflection. UCO5 illustrates this: "I sort of found myself at odds with my colleagues, but in sympathy with those I was manipulating. I think that's sort of part of the journey which has led to the state I'm in." Feelings of guilt and shame were sometimes felt by the officers, affecting some more than others, especially when the perceived vulnerability of their targets and those on the periphery were described. UCO3's reflection led her to feel sympathy for her targets in that she felt she was betraying people she had built a rapport with, and eventually, she had to "fuck them over". The guilt and shame felt were aimed towards those they perceived as particularly vulnerable, such as those who were parents and drug users, rather than those who used extreme forms of violence. Again, this is illustrated by UCO5.
"Yeah, I reflected on them, I always reflected upon the people I had manipulated and certainly reflected on that a great deal..." UCO5 experienced a transformation in his beliefs surrounding drug policy and policing, moving from a prohibitionist stance, to one that recognises how his line of work had a detrimental impact on those he used to climb the criminal ladder, often problematic and vulnerable drug users: "I suppose you could call what I did weaponising empathy because in order to be successful you had to understand the motivations of the people you were mingling with." Regarding alcohol use as a coping mechanism, it was not uncommon for the participants in this study to have engaged in it in some sense. Some, such as UCO4, saw alcohol as a way to cope through the social aspect of drinking, it is a "salient part of CID", serving as an important bonding function. UCO3's experience is similiar: "…what you'd do is after a job you'd go somewhere and have a few drinks and get pissed. That was our de-stressing…" Whereas for some, the psychopharmacological effects of alcohol were the coping strategy employed, acting as a form of self-medication to deal with the problems of life that they could not handle. Alcohol use was therefore utilised in two ways: to reduce symptoms of stress and anxiety; and due to alcohol use being inextricably linked to policing culture, which resolves stress through alcohol as a form of socialisation. The former is represented by UCO5: "Oh, this is a problem, I unwind by drinking alcohol. Which is a problem because it has become a response...a first response feeling that I'm feeling anxious…." In this study, all participants except UCO1 utilised exercise as a coping strategy. The exercise took various forms such as running, cycling, weightlifting, spinning and yoga. The effect it had upon each was quite similar, in that an important distraction was provided. Some felt that stress was reduced after doing light exercise such as cycling to the supermarket, whereas for others, the effects were more intense. For example, UCO3 saw running as a way to "stamp" out the negative thoughts and stress she endured, and explained that as her stress increased, she engaged in more running exercise: "Yeah, it was important to sort of like run it out. I still do that now, as a way to, keep things going. I can…I can just get into that state where I'm pounding the streets and not thinking about anything." Whereas UCO2 utilised exercise in both a negative and in a positive way and is quite a novel illustration of the utility it has. On the one hand, he saw running as a way to "calm his brain", whereas on the other, he saw it as a way to facilitate physical appearance manipulation in order to cement his persona as a drug addict. This arguably contributed to the identity stressors he faced due to his work as an undercover officer, while simultaneously using it as a coping strategy. In general, UCO2 chose not to cope, in order to make himself as chaotic as possible for him to blend into the culture of drugs he embedded himself into, and to better associate with other drug users. The excessive exercise served a dual purpose, in this case, to make oneself look worked out, filthy and dishevelled in order to allay suspicions of his true identity, and, he reasoned, "what better way?". After his police career ended, UCO2 engaged in other forms of exercise. At first apprehensive, UCO2 found that it was incredibly calming and finds value in coping in this manner.
"But I get in a routine, train, like [inaudible] spin on the bike. I'm addicted to that, it's my new addiction. When I go there, and I can feel it, like releases a lot of things... Exercise is a big thing." Yoga was utilised by some of the former officers, and the participants found it to help calm the mind, and to reduce negative, intrusive and distressing thoughts. UCO2, however, was initially apprehensive at practising yoga. He felt that, as a male, he would be stigmatised by using it as a coping strategy. Once he overcame this barrier, he added that it was an important and necessary part of his recovery: "It's really relaxing, and I thought no not for me, but I tried it and it's really good I enjoyed it, it kind of stills your mind a little bit..."

Discussion
The results of the present study indicate that fearing discovery is a potent stressor in undercover policing and has a detrimental impact on psychological well-being. Each participant reported feeling fear at apprehending serious violent retribution, at the prospect of being unearthed as an undercover police officer (Macleod 1995;L. Miller 2006), and in one instance, it was a matter of life or death (Kowalczyk and Sharps 2017). Participants discussed feeling anxious at being discovered and feeling like they themselves were being watched, which corresponds with the culture of "incessant paranoia" . Adequate backstopping and legend building can be an important factor in reducing the likelihood that an officer may be discovered; however, how effective this may be is uncertain due to criminal awareness of police tactics, and thus their ability to respond to undercover deployments. Backstopping is the practice of providing an undercover officer with documents that support the covert identity and can withstand intense scrutiny, such as identification documents. Legend building, however, is a process carried out by the undercover officer where they visit locations in order to make their covert identity become recognised, before engaging in any evidence gathering. Proper planning is essential to reduce the risk of harm befalling the officer, such as adequate measures where an officer can summon backup. Another consideration is that their safety should be paramount to an investigation, rather than the needs of the investigation taking precedence. One way an undercover officer can preserve their identity is through the use of verbal diversion (Jacobs 1992) in order to deflect from challenges to the undercover persona. However, others have argued that in deep cover infiltration, it is almost impossible to protect the officer in the event they are discovered as an undercover police officer (G. Miller 1987).
The existence of a good cover officer system appears to be extremely important in providing a lifeline to the undercover officer. Unfortunately, no previous literature highlights the cover officer system in use in the United Kingdom; however, the Regulation of Investigatory Powers Act Statutory Instrument 2788 sets out the responsibilities of those who have direct supervision over an officer, and those that have general oversight of an operative. This is reflected in draft authorised professional practice (College of Policing 2016) where the roles are set out, most notably the cover officer, who attends to the day to day management of the operative. This includes ensuring that their welfare is attended to (College of Policing 2015), whereas the role of the covert operations manager oversees general management and tasking of an officer. It is important that the needs of the officer are met, as it is a unique role they are performing, and they must have their concerns noted. This cover officer system, if it had worked well in the case of UCO2, may have prevented the profound identity crises he went through, and would have provided a solid base with which he could hold on to that police identity, rather than overidentifying with his role as well as encountering role strain of being both a police officer and playing the role of a drug addict. A vivid account is told in James (1957), who found himself slipping back into the role of Montgomery. The role-overidentification is observed elsewhere, as it is argued that there is a direct correlation between the sympathy felt for the target and the level of perceived distress at having betrayed them (L. Miller 2006). Some were unable to achieve the mental disassociation required to betray a close relationship without feeling sympathy or distress, experiencing what is termed undercover Stockholm syndrome (L. Miller 2006). Insights provided above may show that this may be due to the characteristics of an offender. This may be if they are vulnerable, and it would be interesting to know whether these feelings of sympathy, guilt and distress at betraying a close relationship are the same for all types of investigations. Research has shown that reappearances of the undercover persona are more frequent when combined with physical appearance changes (Girodo et al. 2002), which is concerning given that it has been identified that it is one of the most common modes of deception by undercover police officers (Jacobs 1992). UCO2's experience was the closest example of "going native" (Faupel and Watson 1988). Examples have been detailed elsewhere, where an officer returns to their undercover 'identity' once an operation has concluded (Lewis and Evans 2013).
There are similar stressful factors inherent in undercover policing as there are in conventional forms of policing which have an impact on the support they receive from home, such as working long and anti-social hours, missing family events and other social activities (Violanti et al. 2017) which are known as job-content stressors. These have been noted within previous undercover research, as Marx (1988) notes that long hours and uncertain work schedules caused stress upon the family. An added insight, however, is that the inability to discuss one's work is a necessary, yet stressful, facet of undercover policing, as found by Pogrebin and Poole (1993), whilst Farkas (1986) notes that this can cause anxiety. In concurrence with the findings here, this makes receiving support from the family qualitatively different to conventional police officers, as the relationship between partners, families and friends are marked by a bubble of deception whereby officers lie about what they do in all areas of their lives. It is also noted that there is a culture of suppressing emotions and putting up a front of toughness which can have an impact on whether an officer chooses to access support (Christopher et al. 2016;Eddy et al. 2019), especially as a female trying to prove one's worth within a generally male-dominated occupation .
Interestingly, abuse towards partners or family does not feature within the study here, apart from UCO5 being on the receiving end, whereas others have found that an officer can adopt the undercover persona so much, that they inflict their 'Neanderthal' behaviour on their family (L. Miller 2006). Although noted elsewhere that not all officers neglect their partners, there can be detrimental effects on the quality of life of the family (Hibler 1995). As in the case of UCO5, extended periods of leave should be provided for the UCO following each concluded operation. Tenure policies should also be carefully considered and operated on a case-by-case basis for each officer; however, research on length of service is inconclusive. The CoP policy on psychological support appears adequate, requiring bi-annual assessments for advanced undercover officers and both prior and following an operation, although how effectively this policy operates is unknown, and must be part of the focus of future research.
One participant described the benefits of mindfulnessbased interventions, which could perhaps be useful in improving interpersonal functioning, including positive communication with the family (Eddy et al. 2019). Defined as being conscious to one's ongoing experience (Gawrysiak et al. 2018) from moment to moment (Shapiro et al. 2006), mindfulness-based interventions may also be useful in order to reduce alcohol consumption and other poor coping strategies. It is endorsed specifically by one participant and highlighted in previous research as being useful in reducing alcohol consumption (Christopher et al. 2016). Where officers feel unable to access support if they need it, compulsory mindfulness-based training for officers to utilise on their own may reduce the stigma of an officer specifically seeking help. Perhaps a more important benefit would be by providing the officers with the skills to utilise mindfulness-based interventions when contact between handlers and the UCO is limited due to the covert nature of the work. It is argued that such action could be instrumental in providing a stable support mechanism during periods of deep cover.
The utility of exercise in alleviating some of the negative symptoms should be taken into consideration and appears to have been a vital coping mechanism employed by some of the participants. Exercise and physical activity were deemed the most beneficial coping mechanisms employed by the participants. There is the added benefit that exercise can be used when facing long periods of time without access to support. It has long been argued that exercise can alleviate depressive and anxiety symptoms (Sonstroem and Morgan 1989) and findings show that percentage change in serotonin partially mediates "the relationship between exercise and depression" (Wipfli et al. 2011, p. 474). Officers engaged in this type of police work should be pointed to and reminded of ways in which they can improve their well-being, outside of what is understood as extremely restricted access to formal support mechanisms.
There are, however, significant limitations to this study. The small number of participants interviewed is largely due to the inherently secretive nature of covert policing, which is often inaccessible to uniformed police officers , not least nearly impenetrable to academic researchers. Researchers in undercover policing have been suspected of being management spies (Mac Giollabhuí et al. 2016), or from professional standards departments. Given that researchers already struggle to study conventional policing (Cram 2018; Fox and Lundman 1974;Lundman and Fox 1978;Rigakos and Worth 2011), current academic research and access are even more so limited in the context of undercover policing (Kowalczyk and Sharps 2017). This number could, therefore, be deemed justified and is suitable for an exploratory study on the topic at hand. At present, undercover policing is undergoing political and media scrutiny and will continue to do so for a few more years (Stockdale 2018; Undercover Policing Inquiry 2018), thus presenting serious barriers to researchers engaged in this area.
Nevertheless, presented here are new insights into the realms of covert and undercover policing, shedding some light on the dangers, stressors and coping strategies used by officers in this field of policing. Covert policing is proposed here to be a distinct area of research interest, separate from traditional policing, and interested researchers would gain valuable insights if they can carry out similar research on currently serving covert and undercover police officers. This is equally important when considering the role of women in covert policing, as the experience may be markedly different. It is noted here that the risks to female officers are different than to males and thus has implications for officer safety. Loftus et al. (2016) also describe how women and minority ethnic officers are highly valued and can be seen as enabling wider operational capability, as well as offering unique insights when it comes to assessing the operational risks and hazards of an assignment. The insights here are invaluable to researchers and professionals working with covert police officers, especially providing information on help-seeking behaviour which applies to police culture generally.

Conclusion
This paper offers an insight into the experiences of former police officers who have engaged in covert police work. Although the present study is unable to provide an account of psychopathology as a result of working in an undercover role, it can outline some of the key aspects of the role that have the potential to have a detrimental effect on officers' wellbeing. The present article describes the dangers officers face, the perceived risks, and that officers can and do operate on a life and death basis. Long hours and uncertain shift schedules can be commonplace within undercover work, which can impact profoundly on the family. Extreme identity confusion featured, however, was not common. Some stressors and coping strategies are very similar to their uniformed counterparts such as using alcohol to cope; however, others are qualitatively different such as identity crises, as well as feelings of guilt and sympathy for those they target and those on the periphery such as vulnerable informants. This study makes a valuable contribution to undercover policing research, especially so neglected within policing research in the UK, and offers knowledge that is available to those working closely with UCO's. This research makes a key contribution to the understanding of a female's role in undercover policing, which is even more neglected than undercover policing research itself, as well as highlighting the necessity and utility of a wellstructured and solid cover officer system. Interested researchers would do well to investigate the role of women in covert policing, as well as looking at how an officer reintegrates into the police officer identity. An equally important recommendation would be to quantitatively measure the stress that covert and undercover policing can have, and this should be done across a sample of current police officers. This will aid selection, training and retention, as well as improving welfare strategies for this distinct and niche area of policing. What is overwhelmingly clear is that covert and undercover police officers are neglected in initiatives designed to improve officer well-being. This group of officers must not be left behind if the police institution wishes to fully support personnel in their health and psychological well-being. To neglect this group of officers in the name of secrecy would be to directly place them at the risk of serious psychological harm, and such a position cannot be countenanced if the tactic is to remain effective.

Compliance with Ethical Standards
Conflict of Interest The author declares that he has no conflict of interest.
Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed Consent Informed consent was obtained from all individual participants included in the study.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.