UK police custody officers’ psychosocial hazard exposures and burnout. Policing: an International Journal of Police Strategies & Management, 36 (3).

Purpose : Stress research in UK policing has largely neglected to account for variance in the type of psychosocial hazard officers are exposed to across policing roles, highlighting the need for role-specific research that is capable of informing similarly specific stress reduction interventions. This study aimed to develop and assess exposure to a taxonomy of psychosocial hazards specific to UK police custody work, consider the burnout profile of custody officers, explore relations between psychosocial hazard exposure and burnout, and compare the exposures of burned out and non-burned out custody officers. Design/methodology/approach : Preliminary focus groups identified a series of psychosocial hazards specific to the custody officer role. A questionnaire administered to custody officers within a UK territorial police force assessed exposure to these psychosocial hazards and burnout. Findings : Twenty-six custody-specific psychosocial hazards were identified, across nine themes. The proportion of custody officers who reported a high degree of burnout was above that found in normative data. Hierarchical regression analyses showed that exposures were positively related to emotional exhaustion and depersonalisation. Unrelated t -tests showed that respondents who reported high burnout also reported significantly higher exposures across all nine psychosocial hazard themes than those with sub-threshold burnout scores. Originality/value : This is the first study to investigate the stress-related working conditions of UK custody officers. It provides a foundation for future large-scale longitudinal studies concerned with validating the current findings and improving the health of officers engaged in this unique policing role.

In recent decades UK police forces have moved from a custody model that involved most police stations possessing a limited number of cells for the detention of prisoners to the development of dedicated custody suites within larger stations only. These typically contain a holding cell where arresting officers wait with prisoners prior to entering the suite, a prisoner processing area, cells for the detention of prisoners, interview and medical rooms, and consultation rooms for prisoners to meet with their legal representatives. The operation of a modern custody suite is the responsibility of a custody officer, who is at least of the rank of sergeant (the first supervisory rank) and has undergone specialist training. This study investigates the psychosocial hazards to which custody officers are exposed and examines relations between psychosocial hazard exposure and burnout. Prior to describing the study methodology, an overview is provided on the study's theoretical perspective and the knowledge base as it relates to custody-specific psychosocial hazards and burnout in policing.

Stress theory
Transactional stress theory conceptualises work-related stress as a process comprising three elements (Cox & Griffiths, 2010): (i) antecedent factors, namely exposure to organisational psychosocial hazards or 'stressors'; (ii) cognitive perceptual processes that give rise to the emotional experience of stress; and (iii) correlates of that experience, both individual (e.g. psychological and physical health outcomes and health-risk behaviours) and organisational (e.g. absence, impaired organisational commitment and morale, elevated intention to leave). Within this theoretical framework organizational psychosocial hazards are defined as "those aspects of work design and the organisation and management of work, and their social and organisational contexts, which have the potential for causing psychological, social or physical harm" (Cox et al., 2000, p. 14). In the policing context these might be conceptualised as "the niggling aspects of the work environment that pervades police organisations because of the structural arrangements and social life inside the organisation" (Shane, 2010, p. 815). These organisational, as opposed to operational, psychosocial hazards have been reported to be particularly problematic for officers (Alexander et al., 1993;Biggam et al., 1997;Brown and Campbell, 1990;Collins and Gibb, 2003). (For a discussion on the distinction between and impact of organizational and operational psychosocial hazards in police work see Houdmont, Kerr, and Randall, 2012).

Custody-specific stressors
UK police stress research reveals little about the psychosocial hazards to which custody officers are exposed. In recent years researchers have tended to apply generic measures of psychosocial hazard exposure that are not specific to policing (Houdmont, 2012;Houdmont et al., 2012). This approach usually reflects a preference among police forces for the generic measurement instrument developed by the UK government to assist organisations in fulfilling their legal duty to conduct psychosocial risk assessment activities. (For a discussion of this duty and approaches to its fulfilment in UK policing see Houdmont et al., 2012). Vuorensyrjä and Mälkiä (2011, p. 382) argue that police stress studies ought to apply police-specific psychosocial hazard measures given that "it is not only the intensity of stress but also the particular types of stressors that can be expected to differ from occupation to occupation." A number of UK policing studies have heeded such advice (e.g., Biggam et al., 1997;Brown et al., 1999;Collins and Gibb, 2003). However, the measures used in such studies have failed to take account of role-specific psychosocial hazards. Moreover, generic police-stressor measures typically address a host of psychosocial hazards that are of irrelevance to those engaged in the highly specialised custody officer role (such as searching for a missing person or football duty).
UK police stress studies have also consistently failed in their analyses to separate custody officers from those engaged in other policing roles. Only one study can be identified that examined officers by division and therefore addressed criminal justice (which includes custody officers) as a discrete entity (Collins and Gibb, 2003). However, the study involved a small sample of 32 officers employed in criminal justice and provided data on psychological distress only; information on psychosocial hazard exposure by division was not provided.
Taken together, these various limitations in the evidence base result in there being a paucity of evidence on the types of psychosocial hazard to which custody officers are exposed and the degree of exposure. In response, the first two aims of this study are to develop, and assess exposure to, a taxonomy of psychosocial hazards that is specific to the experience of UK police custody officers.

Burnout in Policing
The Maslach Burnout Inventory (MBI:  is the most commonly used self-report measure of burnout, and the one used in the current study. The instrument conceptualises burnout as: A psychological syndrome of emotional exhaustion, depersonalization, and reduced personal accomplishment that can occur among individuals who work with other people in some capacity. Emotional exhaustion refers to feelings of being emotionally overextended and depleted of one's emotional resources. Depersonalization refers to a negative, callous, or excessively detached response to other people, who are usually the recipients of one's services or care. Reduced personal accomplishment refers to a decline in one's feelings of competence and successful achievement in one's work. (Maslach, 1993, pp, 20-21) Numerous police stress studies have identified burnout as an important health outcome (e.g., Loo, 2004;McCarty et al., 2007;Vuorensyrjä and Mälkiä, 2011). In the current study, burnout is used as an index of custody officers' health for two reasons. Firstly, the police stress literature has often conceptualised burnout as a consequence of prolonged exposure to stressful working conditions (Vuorensyrjä and Mälkiä, 2011, p. 384). Secondly, over the course of a shift the custody officer is required to make a constant stream of decisions concerning the management of prisoners' health and safety. A poor decision can have potentially fatal consequences, highlighting the imperative for optimal cognitive functioning throughout the duration of a shift. Given that burned out employees demonstrate impaired cognitive functioning relative to non-burned out employees (Feuerhahn et al., 2013;Oosterholt et al., 2012), burnout can be identified as a construct that warrants particular attention in the examination of custody officers' working conditions and health.
In light of the evidence that points to burnout being prevalent in policing and a consequence of chronic stressor exposure, the third aim of the current study is to profile burnout in custody officers and consider this in relation to that found in normative data and previous policing studies. The fourth and fifth aims concern relationships between psychosocial hazard exposures and burnout.

Aims of the current study
Custody officers in the UK undertake a unique and challenging policing role. It is therefore surprising that no previous studies have examined the custody work environment and its health correlates. Research is required in this regard to inform the design of interventions to promote the health and wellbeing of custody officers and, by extension, operational effectiveness. In response, this exploratory study has five aims: 1. To develop a taxonomy of psychosocial hazards that is specific to UK custody officers.
2. To examine custody officers' exposure to custody-specific psychosocial hazards.
3. To profile the burnout of custody officers and to consider this in relation to normative data and previous police burnout studies. 4. To explore the strength of relations between psychosocial hazard exposures and burnout.
5. To examine the custody-specific psychosocial hazard exposures of burned out custody officers relative to that of non-burned out officers.

Setting
The study involved custody officers within a large UK territorial police force. It was commissioned by a joint branch board of the Police Federation of England and Wales in direct response to the findings of a 2011 force-wide staff survey which found that custody officers reported considerably worse working conditions and lower job satisfaction than the force average.

Study design
The study utilised a sequential mixed methods design, with an initial qualitative phase identifying the key psychosocial hazards experienced by custody officers (Phase 1). Findings from Phase 1 informed the design of a cross-sectional survey that was administered forcewide to quantify custody officers' psychosocial hazard exposure and burnout (Phase 2).

Procedure, materials, and analysis: Phase 1
Custody officers were recruited to focus groups via an email invitation issued to officers within a limited number of custody suites located across the geographical region and representing the full range of custody suites in terms of prisoner capacity. Two focus groups facilitated by the author and each involving four custody officers were conducted in September 2011. A semi-structured question schedule was used to elicit information. To help participants relax and to build rapport with the researcher they were first asked to discuss what they enjoyed about work in the custody suite. Participants were next asked to consider what they found stressful about their job. They were also asked to consider what could be done to reduce their stress. Finally, they were asked if there was anything else they'd like to discuss. The focus group discussions were audio recorded with the participants' permission and subsequently transcribed verbatim. Each lasted approximately 1.5 hours.
In accordance with the first aim of the study, thematic analysis was applied in order to identify the broad themes evident in focus group transcripts (Braun and Clarke, 2006).
Designed as "a method for identifying, analysing and reporting patterns (themes) within data" (Braun and Clarke, 2006, p. 79), the six-stage process involved (i) transcribing the data and reading (and re-reading) each transcript in detail in order to become familiar with the data, (ii) generating initial broad categories for the responses, guided by the aims of the study, (iii) collating the categories into themes and gathering the relevant data within each theme, (iv) reviewing the themes and sub-themes against the categories to ensure that the entire dataset was represented, (v) defining and naming themes and sub-themes, and (vi) production of the report involving the use of extracts from the transcripts illustrating each theme in light of the study's aims. Braun and Clarke's approach has been used extensively in occupational health psychology research where it has been shown to be an effective means by which to identify, analyse, and report themes in qualitative data (e.g., Leka et al., 2011;Payne et al., 2012).

Procedure, materials, and analysis: Phase 2
For the second phase an online questionnaire was administered to all custody officers across the force. The questionnaire was piloted among five custody officers and representatives from the Police Federation. Following refinement, an email inviting voluntary and anonymous participation in the study was sent from the researcher to the work account of each custody officer (N = 139). The email included a hyperlink to the survey that was hosted within a secure online survey facility. The questionnaire remained open for four weeks during October 2011 and weekly email reminders were sent to eligible officers.
The questionnaire examined the following variables in addition to a set of demographic constructs. Psychosocial hazards: Informed by the themes identified in Phase 1 of the study, a series of 56 items explored psychosocial hazard exposure. Respondents were asked to indicate how each aspect of their job had been in the preceding six month period.
Items were worded in the form of statements to which respondents indicated their strength of agreement on a five-point Likert-type scale of (i) strongly agree, (ii) agree, (iii) unsure, (iv) disagree, (v) strongly disagree. Example items included "staffing arrangements fail to take into account the number of cells in a block" and "I have reasonable freedom to run my custody suite as I see fit". An equal number of positively and negatively framed items were used. Burnout: This was measured via the Maslach Burnout Inventory. The Human Services version of the inventory  was used as it is designed for application with employees who are "often required to spend considerable time in intense involvement with other people" (ibid, p. 3). The inventory includes three sub-scales, which correspond to the dimensions of the burnout syndrome: emotional exhaustion (EE -feelings of being emotionally overextended and depleted of one's emotional resources), depersonalisation (DP -negative, cynical attitudes and feelings about one's clients) and reduced personal accomplishment (PA -the tendency to feel unhappy about oneself and dissatisfied with one's accomplishments on the job). Each dimension is measured by a separate subscale. EE, nine items (e.g., 'I feel emotionally drained from my work') (α = .91), DP, five items (e.g., 'I worry that this job is hardening me emotionally') (α = .73); PA, eight items (e.g., 'I have accomplished many worthwhile things in this job') (α = .68). All items were scored on a seven-point frequency rating scale ranging from 0 ('never') to 6 ('every day'). High EE and DP scores and low PA scores are indicative of burnout.
To address the second of the study aims frequency data were calculated for exposure to each potential psychosocial hazard. Following established psychosocial risk management practice (Bevan et al., 2010;Cox et al., 2002), an item was considered a psychosocial hazard if the statistical majority (≥50%) of participants responded 'strongly agree' or 'agree' with the statement (or 'strongly disagree' or 'disagree' depending on the valence of the statement).
To address the third of the study aims respondents' scores on each of the burnout dimensions were descriptively considered in relation to normative data and previous policing studies. To address the fourth aim, an overall psychosocial hazard score was calculated for each participant. This was the sum score achieved in relation to the working conditions that met the criteria for consideration as a psychosocial hazard. To investigate the relationship between psychosocial hazard exposure and burnout a series of three multiple regression analyses were performed using the burnout dimensions (EE, DP, PA) as dependent variables.
A hierarchical approach was followed, entering age, gender (coded as 0 = female, 1 = male), and tenure in custody in Step 1 in order to control for the influence of these variables, and psychosocial hazards in Step 2. To address the fifth aim, participants were dichotomised on the basis of burnout scores. A high degree of burnout is reflected in scores in the upper third of the normative distribution based on scores contributed by workers from various occupational groups, including police officers (Maslach, Jackson, and Leiter, 1996). A high degree of emotional exhaustion is identified by a score of ≥28, high depersonalisation by a score of ≥11, and low personal accomplishment by a score of ≤33. Respondents who scored above these thresholds were considered to present with a 'case' of burnout. It is important to note that the term 'case' is not used here in a diagnostic sense. Bivariate correlations were applied to identify statistically significant relationships between burnout and psychosocial hazard exposure. Subsequently, for those relationships that were statistically significant (P < 0.05), t-tests were applied to compare burnout cases to non-cases in terms of psychosocial hazard exposures.

Results: Phase 1
Analysis of focus group data revealed a series of nine psychosocial hazard themes.

Theme 1: Phone calls
The requirement to deal with a constant stream of phone calls was identified as by far the most problematic aspect of the custody officer role. The first sub-theme concerned phone calls from solicitors and prisoners' family members: Quote 1: The other day there were three phone calls from solicitors in a five minute period. From the solicitor's point of view you can understand them doing it; they're just trying to do their job. It's not unreasonable for a solicitor to say 'when's my client coming back?' They probably think we're being fussy but it's phone calls all the time.
The second sub-theme concerned switchboard putting calls through to the block that could be dealt with by others: Quote 2: It's not just solicitors phoning. The other one is 'I want to know, was Billy locked up last night?' I'm not going to tell them anyway so why are they putting these calls through to me?
Quote 3: Someone thought that when we call the doctor it would be nice to get an ETA. So what happens now is that normally the doctor turns up within 90 minutes.
We get four or five phone calls saying 'the doctor is going to be an hour', 'the doctor is 20 minutes away', 'the doctor has been delayed'. I want to say 'I'll call you if I want to know.' I only want to know if he's going to be late.

Theme 2: Physical work environment
The harsh glare produced by fluorescent lighting along with the absence of natural light was reported to be problematic: Quote 4: We're the backside of the organisation...It's dingy. X is an absolute shithole.
I hate the place with a passion. At least in Y you've a window you can look out of.
I'm surprised we don't get rickets! In addition, noise produced by prisoners (Quote 5) and staff (Quote 6) was identified as problematic.
Quote 5: The prisoners have their own tambourine set with the cell door and the hatch which is very loud. And if one or two want to play that game it resonates throughout the cell block and makes concentration very difficult. The way you get round that is to focus and narrow in your concentration, so you're missing peripheral stuff which can be important by trying to block out that sound.
Quote 6: More important than the prisoner noise sometimes is just the officer noise.
When it gets busy in there and they all want to have their own conversation; his conversation has to be a bit louder so that he can hear himself, etc, until near the end everyone's just shouting.

Theme 3: Workload
The speed at which officers were required to work was reported to be problematic along with the increased occurrence of high pressure 'peak times'. The prioritisation of tasks was made difficult by competing and constant demands (Quote 7) and it was felt that the time available to process each prisoner was inadequate (Quote 8).
Quote 7: The workload that is expected of us is horrendous. And we're getting more and more stuff that's coming in. All our custody officers are now being trained to do drug tests. When are we going to get the time to do that? Gone are the days when you'd book on a prisoner in eight minutes. And constant watches we're doing more of. Something will go wrong somewhere.
Quote 8: Our risk assessment we do when they get booked in...we do it verbally then put it on the computer. It takes 15 minutes to do a thorough one, and with a foreign language it takes an hour. I do this because the day I cut a corner on a risk assessment is going to be the day that I do something dangerous, and I've got no defence then.

Theme 4: Prisoner-related problems
Prisoners were reported to sometimes take advantage of their rights by, for example, claiming illness. This, in turn, had knock-on effects because an officer would be required to accompany an 'ill' prisoner to hospital.
Quote 9: Our vulnerable people have changed now. Because of the way we police them they get to know 'If I say I'm suicidal' they get what they see as a better service.
A bobby sitting outside the cell, someone to talk to, 'can I have a coffee?'. We ask them these questions such as 'have you ever had suicidal thoughts?' and they think 'I might as well say yes'.
Quote 10: It was a Friday night, I'd already got eight or nine in, and then I got six Romanians for burglary. I was on my own. Each of those prisoners took the nearest part of an hour to book in because they all spoke better English than me but they played the system.

Theme 5: Technology and equipment
The technology in the block, particularly computers and printers was reported as not being up to the job (Quote 11). Equipment shortages were also reported along with procurement difficulties (Quote 12).
Quote 11: The printer will say 'replace the cartridge' and the IT guy will just override the request and give it a shake...When it's turned a two minute job into a 20 minute job, when I really can't afford to spend 20 minutes on this because I've got all this happening, and then the phone goes, and then that prisoner knocks on the door, and then that cell buzzer is going. The traffic cars that the force buys are chosen to be used for 24 hours. The printers we have in custody should also be bought with a view to 24 hour use.
Quote 12: We're not as well supplied as we were before. There are gaps on our shelves. We haven't got white suits to protect officers. We haven't got enough evidence kits in the cell block. We have gaps in food. I had to go to another station the other night for coffee and milk. When you have a problem, you're not just popping upstairs to talk to someone in the station, you're talking to someone you don't even know. You're putting your faith that the request will get actioned. As a result, we're starting to over-supply in the knowledge that it generally takes two weeks to get the things you want.

Theme 6: Breaks and staff facilities
Facilities for rest breaks were reported to be inadequate and most participants reported that they took irregular breaks, if at all, due to concern about falling out of the information loop (Quote 13) and lack of appropriate facilities (Quote 14). Many felt that facilities for the storage of personal items were inadequate Quote 13: I'm not used to having rest breaks...If you're going to put me on my own then I'm not so sure I want a rest break because the problem is, who are you handing over to, how long is it going to take you to hand over to them, where are they having to come from, when you come back from your break how long is it going to take to pick the reins up again.

Theme 7: Safety risks
Frustration was expressed concerning the risk to safety produced by (i) arresting officers failing to follow the rules of the custody suite (Quote 15) and (ii) prisoners being in possession of contraband as a result of failed searches (Quote 16).
Quote 15: Someone's come in and they've already had the cuffs taken off them, they've been given a hot drink, they've not been searched, no checks have been done; you know, things that are basic and we all get trained on when we first join, for some reason it's all going out of the window.
Quote 16: The standard of arresting officers' searches is shocking. Part of it is probably the prisoner lulls them into a false sense of security. You've locked up a middle aged bank manager for shoplifting, you're not going to search him the same as your heroin addict who you know might have things on him. But that doesn't mean that the bank manager doesn't have something on him that he shouldn't...the other day someone got a tazer through!"

Theme 8: Consultation and communication
Lack of consultation was reported in relation to the introduction of new procedures and technology. For example, a buzzer system by which prisoners could call a custody officer to their cell was criticised for being inappropriate: Quote 17: ...they introduced a touch screen cell call system which obviously had no input from a practising custody team. If someone wants to speak to you inside their cell they press their buzzer and you think, 'ok, cell whatever wants to speak to me'.
To cancel the alarm you've got to physically walk to the cell. They've given us a telephone system to speak to the prisoner to ask what they want, but then you've got to walk to the cell, which might be right at the bottom of the cell block, to actually physically reset it. So by the time you've come back he's pressed it again. And it literally drives you up the wall. The force has invested so much money in that and it's not the best thought out system.

Theme 9: Staffing arrangements
Shifts were viewed as consistently understaffed. This resulted is a focusing of attention on particular types of prisoner: Quote 18: If we wanted to do a Rolls Royce job on every prisoner it would take double the staff to do it, so something has to give somewhere. As custody officers we tend to identify those prisoners that are going to bite us on the arse, and those are the ones that we pay particular attention to.
The policy of moving prisoners and staff between blocks mid-shift in order to even out demands was reported to be unrealistic and a waste of time: Quote 19: We had an instruction saying that if you're on your own you need to ring round the other custody blocks to see whose got more than one and get them to come over. To which I said 'so that one whose coming over to cover you, when does he have his break?. If that's what you want at 9pm I'll walk out of the nick. I'll catch the train into X, take another train from Y to Z to relieve that Sergeant'. It's not going to work." There was a problem with extra work caused by other custody suites refusing to take additional prisoners despite having perceived spare capacity: The view was frequently expressed that staffing arrangements failed to take account of the number of cells in a block. Finally, many officers felt that inadequate staffing arrangements resulted in officers having to undertake tasks beyond their remit, such as watching the CCTV screens fed by cameras within cells.

Participant characteristics
A total of 76 questionnaires were completed (55% response rate). Information on respondents' demographic characteristics is presented in Table 1. Comparison of the characteristics of custody officers who completed the questionnaire with those of the force's entire custody officer population showed no substantive differences.
[Insert Table 1 about here]

Psychosocial hazards in custody work
The first and second aims of the study were to develop and assess exposure to a taxonomy of psychosocial hazards specific to the custody officer role. Twenty six items met the criteria for consideration as a psychosocial hazard (Table 2).
[Insert Table 2 about here]

Burnout
The third aim of the study was to profile burnout. The mean score and standard deviation on each of the three burnout dimensions is shown in Table 3. In total, 47.8% of respondents were identified as presenting with a case of high emotional exhaustion, 47.9% high depersonalisation, and 65.7% low personal accomplishment. In the normative distribution one third of cases score above the high burnout threshold; approximately half of the respondents in the current study scored above the high burnout threshold on two of the three dimensions (EE and DP) and approximately two thirds scored above the threshold for the identification of low personal accomplishment.
Given that the MBI normative dataset is not comprised exclusively of data drawn from police officers it is useful to compare the current findings to those of previous policing studies. The MBI has been used in dozens of policing studies; however, in only four cases does the published report provide a mean sum score for each dimension. Table 3 shows that the burnout profile found in the current study was almost identical to that found in a contemporaneous study of 2,026 police officers (all job roles and ranks) drawn from a UK force (Houdmont, 2012), suggesting that custody officers have a burnout profile of equivalence to that found more generally in UK policing. Areas of contrast were found, however, when comparing current results to those of studies conducted beyond the UK. The mean EE score produced in the current study (26.7) was markedly higher than those of 19.0 and16.9 found in two Canadian studies (Loo, 1994;Stearns and Moore, 1990) and 9.8 found in a Dutch study (Kop et al., 1999).
[Insert Table 3 about here]

Relations between burnout and psychosocial hazard exposures
The fourth aim of the study was to explore the strength of relations between psychosocial hazard exposures and burnout (Table 4). When EE was regressed on psychosocial hazards (controlling for gender, age, and tenure in custody which, in combination, were not significant predictors of EE) a significant model emerged (F = 8.41, P < .001). Psychosocial hazards exposures explained 34% of the variance in EE scores. When DP was regressed on psychosocial hazards (controlling for gender, age, and tenure in custody which, in combination, were not significant predictors of DP) a significant model emerged (F = 7.55, P < .001). Psychosocial hazard exposures explained 30% of the variance in DP scores. When PA was regressed on psychosocial hazards a significant model failed to emerge.
[Insert Table 4 about here]

Differences in psychosocial hazard exposures between burnout 'cases' and 'non-cases'
The fifth aim of the study was to examine the psychosocial hazard exposures of burned out custody officers relative to non-burned out officers. Eighteen psychosocial hazards were significantly associated with at least one dimension of burnout (Table 2). Strong correlations (r = ≥.5; Field, 2009) were found between four psychosocial hazards and EE, with a further seven psychosocial hazards being moderately correlated (r = .3-.49; Field, 2009) with EE. Ten psychosocial hazards were moderately correlated with DP and five were moderately correlated with PA.
Burnout cases were compared to non-cases in respect to exposure to each of the psychosocial hazards moderately or strongly correlated with at least one dimension of burnout. A series of t-tests revealed that cases reported significantly higher psychosocial hazard exposure than non-cases.
[Insert Table 5 about here]

Discussion
This exploratory study is the first to develop a taxonomy of psychosocial hazards specific to the UK custody officer role and to consider relations between psychosocial hazard exposure and burnout among officers engaged in this highly specialised policing function. A taxonomy of 26 psychosocial hazards across nine themes was developed. Most of the psychosocial hazards were specific to the custody environment, thereby highlighting the imperative for role-specific psychosocial hazard taxonomies in work-related stress research and practice that involves custody officers. The same might apply to other specialised policing roles. The burnout profile of the sampled custody officers was similar to that found in a contemporaneous study of UK police officers (Houdmont, 2012), though considerably higher (worse) than that found in previous studies conducted in Canada and The Netherlands for which comparable data is available (Kop et al., 1999;Loo, 1994;Stearns and Moore, 1990). It is possible that general cultural differences and attitudes towards policing might explain the considerably lower burnout scores found in these studies relative to UK data (Vuorensyrjä & Mälkiä, 2011). Given the scale of the difference in burnout evident between UK custody officers and the previous Canadian and Dutch studies, it would have been desirable to examine the MBI burnout profile found in other policing studies. Unfortunately, though the literature review undertaken for the current study revealed that since 1990 the MBI has been used in at least 18 policing studies in six countries, very few of these offer a basis against which to compare the burnout scores achieved in this study due to having used a different version of the MBI, a selection of items rather than the full set of 22, or having failed to report overall subscale scores. The series of regression analyses showed that psychosocial hazard exposures were significantly associated with the EE and DP burnout dimensions, respectively accounting for 34% and 30% of the variance in burnout scores.
These findings must be viewed with caution because the sample size was at the lower limit of acceptability for the application of regression analyses. Finally, the mean score achieved on each burnout dimension was above the threshold for the identification of high burnout; approximately half of the sample was identified as presenting with a case of burnout on two dimensions (EE and DP), and approximately two thirds on the PA dimension.
Burned out custody officers reported significantly higher exposure to a set of 13 psychosocial hazards than non-burned out custody officers, highlighting these as ideal candidates for the focus of stress-reduction interventions. It is noteworthy that exposure to some of these psychosocial hazards could be reduced through relatively simple and inexpensive means. For example, the problem of arresting officers failing to follow the rules of the custody block through actions such as removing a prisoner's handcuffs before being instructed to do so by a custody officer, could be addressed through information posters placed prominently in the custody suite. Similarly, the problem of lack of consultation on the introduction of new procedures and technology could be dealt with by efforts to engage custody officers -the experts in the job -in decision making. In these respects the study offers a reminder that low cost stress management interventions often hold the potential to generate substantive health benefits. Indeed, in response to the study's findings the force in which it was conducted has since introduced a dedicated email address for solicitors as a means by which to reduce the frequency of phone calls into custody suites. Within the first month of operation this email address received more than 500 messages. In addition, the force has introduced anti-glare lighting and the regional arm of the Police Federation has issued guidance to custody officers on break taking schedules designed to maximise opportunities for the intake of food and drink without falling out of the information loop.
In terms of the wider policy agenda these findings indicate that forces should consider psychosocial issues in the design of new custody blocks. Burnout has been shown to be associated with impaired job performance among police officers (Bakker & Hauven, 2006), suggesting that work design interventions that seek to reduce burnout are likely to generate operational effectiveness gains. This is particularly relevant in the current UK policy context that has seen a move towards privately built and owned custody blocks.
Seven of the 26 psychosocial hazards identified in the current study were not significantly correlated with any of the burnout dimensions. This finding reaffirms the importance of not relying solely on exposure reports in work environment research, particularly when the working conditions under examination are psychosocial and the assumption that exposure leads to harm does not necessarily hold (Bevan, Houdmont, & Menear, 2010). In relying solely on exposure data when relations between exposures and outcomes are unknown, the risk arises that interventions might be applied to aspects of work reported by employees as problematic but which have little or no relationship with health, resulting in minimal, if any, positive impact on workers' health and wellbeing.
The psychosocial hazard taxonomy developed in the current study is centred almost entirely on organisational as opposed to operational psychosocial hazards. It appears to be the aspects of the design, management, and organisation of work rather than operational activities such as dealing with danger or violence that are primarily responsible for stress among custody officers. This finding is consistent with previous UK policing studies which have demonstrated that organisational psychosocial hazards are reported more often than operational ones by a ratio of 4 to 1 (Brown & Campbell, 1990), are perceived as more stressful than operational psychosocial hazards, and may have an equal or even greater influence on health than exposure to operational psychosocial hazards (Alexander et al., 1993;Biggam et al., 1997;Collins and Gibb, 2003).
The key value of this exploratory study lies in it being the first to explore workrelated stress among UK custody officers. As is often the case in a new avenue of scientific investigation, the findings of this initial study have successfully stimulated further research focused on this unique policing role; as a result of this study the author has been commissioned to conduct a nationwide replication and extension involving longitudinal analysis. Findings will be available in 2015. The large-scale study will provide an opportunity to validate the taxonomy of psychosocial hazards developed in the current study and address many of its limitations. These limitations include its cross-sectional design that precludes the drawing of conclusions on the direction of causation between variables, the limited sample size that prevented more sophisticated inferential analyses, and the possibility that alternative or additional psychosocial hazards might be present in custody operations that do not conform to the predominant 'police owned, police run' arrangement.