An integrative review of nursing staff experiences in high secure forensic mental health settings: Implications for recruitment and retention strategies

s (1979-2019) = 13, ProQuest Social Sciences Premium Collection (1871-2019) = 423, Web of Science (1900-2019) = 155


| INTRODUC TI ON
Nursing workforce sustainability depends on recruitment of nurses to fill vacant posts and retention of skilled and experienced staff. There is a registered nursing workforce shortage in the United Kingdom and internationally (OECD, 2016). Compared with other countries, a mismatch between supply and demand for nurses in the United Kingdom has been associated with frozen pay, reduced employment, and reduced student nurses as a consequence of economic austerity (Buchan, Charlesworth, Gershlick, & Seccombe, 2019). The UK government reported around 40,000 nursing vacancies in England in 2019 (Harding, 2019). Projected healthcare workforce shortages are a major threat to the financial viability of the National Health Service (NHS; House of Commons Committee of Public Accounts, 2019), jeopardizing the realization of the national LongTerm Plan (Harding, 2019). Shortages have been particularly felt in the mental health sector, with attrition rates of 13.6% for mental health nursing staff (Health Education England, 2017) and 12.8% of mental health nursing vacancies unfilled (NHS Improvement, 2018). This should be contextualized within an unprecedented demand for mental health services, both in the United Kingdom (Independent Mental Health Taskforce to the NHS in England, 2016) and globally (Rejm & Shield, 2019). Despite recent analyses of healthcare workforce strategy (Beech et al., 2019;Buchan et al., 2019;Durcan, Stubbs, Appleton, & Bell, 2017) and the levers affecting nursing workforce supply and demand (Gunn, Muntaner, Villeneuve, Chung, & Gea-Sanchez, 2019;Hussain, Rivers, Glover, & Fottler, 2012;Kroetzen et al., 2015), there has been limited scrutiny of factors affecting workforce sustainability within the sub-specialities in mental health nursing.
The Durcan et al. (2017) report, based on policy analysis and stakeholder consultation, identified changes to funding for nurse training, reduced employment benefits, lack of promotion of mental health nursing as an attractive career, and lack of career development for experienced nurses as specific factors affecting mental health nursing recruitment and retention. The present review was commissioned to inform English national workforce strategy in relation to the sub-speciality of forensic mental health (FMH) nursing, also described as 'nursing in secure environments'. Nursing in high secure (also known as 'maximum security') environments is a further sub-speciality within FMH. There are three high secure hospitals (HSH) in England. They house patients with mental disorders detained under the Mental Health Act 1983 who 'require treatment under conditions of high security on account of their dangerous, violent, or criminal propensities' (NHS Commissioning Board, 2013).
Concern about the impact of short staffing on patient care at two of the HSH was raised by the national healthcare regulator in 2018 (Care Quality Commission, 2018). Where short staffing and low nurse-to-patient ratios are known to affect the quality of patient care in general hospital settings, affecting the numbers of adverse events and patient lengths of stay (Aiken et al., 2014;Griffiths et al.., 2016;Rafferty et al., 2007;Sizmur & Raleigh, 2018), short staffing in the HSH was observed by the regulator to affect staff morale, monitoring of seclusion and segregation, and patients' access to appointments, ground leave, and fresh air (Care Quality Commission, 2018).
These observations make a compelling case for considering how aspects of nurses' experience may affect workforce recruitment and retention in the high secure setting.

| BACKG ROU N D
Many aspects of FMH nursing reflect the skills and competencies that are important for any mental health nursing role, such as risk assessment and developing therapeutic relationships.
There are, however, some unique features of the HSH nursing environment. Previous studies where the core competencies and features of FMH nursing have been described (e.g., Dale & Storey, 2004;Mason, King, & Dulson, 2009;Newman, Patterson, Eason, & Short, 2016;Peternelj-Taylor, 2000;Rask & Aberg, 2002;Timmons, 2010) have characterized the central challenge of FMH nursing as striking the balance between care and control, whereby nurses' primary goal 'to care' is tempered by their custodial and safety-oriented duties in the secure setting. This challenge is an ethical one, given that the nurse must provide 'an environment in which the human rights, values, customs and spiritual beliefs of the individual, family and community are respected' (International Council of Nurses, 2012, p2). Custodial duties, for example, the searching of patients and maintaining a locked environment, focus on minimizing the risk of violence and arguably restrict rather than promoting patients' rights and beliefs. Gildberg, Elverdam, and hounsgaard, L. (2010) have described this as a tension between paternalism and therapeutic engagement. For example, interactions K E Y W O R D S emotional labour, forensic, integrative review, mental health, Nursing, trauma, workforce Impact • Workforce retention strategies that incorporate education, support, and supervision should be tailored to address the specific features of nursing in high secure settings.
• Recruitment and retention strategies should take account of the characteristics of nurses who thrive in the high secure setting.
• Longitudinal research is required to measure the impact of workforce strategy on nurses' attitudes to their work and intentions to remain in the high secure setting. may have a paternalist focus because the nurse must set limits on behaviour and enforce rules while explaining those roles and supporting patients to adhere to them.
Previous systematic reviews on FMH nursing have focused on the well-being of forensic staff, finding the empirical research to be heterogeneous, site-specific, and of limited generalizability (Brown, Igoumenou, Mortlock, Gupta, & Das, 2017;Dickinson & Wright, 2008;Freestone et al., 2015). While some studies (Elliott & Daley, 2013;Jones, Janman, Payne, & L., & Rick, J. T., 1987;Kriakous, Elliott, & Owen, 2019) found more mental distress or burnout in forensic healthcare professionals versus other professionals, others found lower or comparable prevalence of stress and burnout compared with nurses working in other settings (Berry & Robertson, 2019;Chalder & Nolan, 2002;Dickinson & Wright, 2008;Elliott & Daley, 2013;Happell, Martin, & Pinikahana, 2003;Lee, Ogloff, Daffern, & Martin, 2015). Correlates for higher burnout and symptoms of stress in FMH nurses have been identified as: younger age (Berry & Robertson, 2019); whether the ward atmosphere feels safe and therapeutic (Berry & Robertson, 2019); intensity and frequency of physical aggression towards staff, when mediated by job stress (de Looff, Nijman, Didden, & Embregts, 2018); staff with low emotional intelligence and high personality traits of neuroticism and altruism (de Looff, Didden, Embregts, & Nijman, 2019); staff with low inclination to mindfulness (Kriakous et al., 2019). Most studies have been single site and cross-sectional. Also, most have not focused on the high security setting. There has been no previous study to integrate all previous research on FMH nursing experience in the high secure settings with the aim of informing workforce policy.

| Aim
To conduct an integrative review to address the research question: What are the aspects of nurses' experiences in high secure FMH hospitals that may affect recruitment and retention of registered nursing staff?

| Design
An integrative review approach was used, as described by Whittemore and Knafl (2005), after Cooper (1998). This approach enables inclusion of research from a range of methodologies and suits a review question about which there may be multiple perspectives.
Integrative reviews have five stages: problem identification; literature search; data evaluation, data analysis; and presentation. A review protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO 2019 CRD42019137259).
Following the decision flowchart developed by Flemming, Booth, Hannes, Cargo, and Noyes (2018), this review is reported according to the Enhancing transparency in reporting the synthesis of qualitative research (ENREQ) statement (Tong, Flemming, McInnes, Oliver, & Craig, 2012

| Search methods
In the problem identification stage, the team identified search criteria with reference to the research question. A preliminary search identified no studies that specifically addressed recruitment and retention in the high secure forensic setting, leading the research team to broaden their review focus to nurses' experiences in high secure FMH, from which potential factors affecting workforce sustainability could be drawn. To ensure that no papers describing studies that included high or maximum security (as opposed to low or medium security) forensic settings were missed at the search stage, we did not include search terms relating to level of security. Sifting of papers according to level of security took place at the title/abstract review stage and again at the full paper review stage if level of security was not clear from the abstract. Furthermore, references from the reviewed papers were scrutinized for potentially relevant papers. We also reviewed papers that had cited the 13 included studies.

| Search outcome
The outcomes of the literature search are detailed in the flow chart in Figure 1. Two researchers conducted the searches, comparing results and decisions around inclusion to assure consistency. Two researchers reviewed and extracted the data, with a further reviewer available to arbitrate on any differences of opinion. Data extraction and organization was undertaken by one researcher. Two researchers reviewed these papers and agreed on the final selection for inclusion. Papers were excluded at this stage if they did not present empirical research, did not contain evidence specific to high secure settings, did not focus on nursing staff experience (e.g., solely focused on patient experience). Where no specific level of security was described, or where the study described medium or low secure rather than high or maximum security, the studies were not included.

| Quality appraisal
In the data evaluation stage, the quality of studies was reviewed and agreed by two researchers, using the Mixed Methods Appraisal Tool Version 2018 (MMAT; Hong et al., 2018). The MMAT allows researchers to use comparable criteria to identify the methodological limitations of both quantitative and qualitative study designs.
Studies were appraised according to the MMAT screening questions and the five specific questions for their research design (all were either qualitative or quantitative descriptive studies; See Supporting Information).

| Data extraction and synthesis
Descriptive data from the qualitative and quantitative studies were extracted and tabulated. This included the authors and year of publication; study design; number of participants in study and comparison groups; methods of data collection and analysis; and a summary of findings. Data-based convergent qualitative synthesis (Pluye & Hong, 2014) was used to compare and describe the findings from the included studies, in relation to the study research question. A constant comparison method was used to identify and develop themes from the data. Findings from the quantitative studies were described narratively to foster comparison with the qualitative study findings.
Details of findings were tabulated in a matrix, then coded according to emergent themes and subthemes (Figure 2; Whittemore and Knafl (2005). Constant comparison enabled the identification of patterns and commonalities between studies. Two members of the team undertook the extraction and synthesis. Plausibility of analysis was discussed with the wider team on presentation of themes and subthemes. This ensured rigour in data extraction and synthesis (Whittemore & Knafl, 2005).

| Study selection
The flow diagram in Figure 1 shows the number of studies retrieved and retained at each stage of the screening process. Fifteen papers were selected for inclusion in the review, describing 13 studies. One study was described in three separate publications. Mercer, 2013, Mercer & Perkins, 2014, Mercer & Perkins, 2018.

| Study characteristics
Study characteristics are summarized in Table 1 The unique features of nursing in high secure forensic settings that may contribute to difficulties in recruiting and retaining nursing staff (N = number of studies from which the theme emerged) There were seven qualitative studies. Addo (2006)

| Quality appraisal
Study quality appraisal using the MMAT is reported in Tables S1 and S2 (Supporting Information). All studies met the screening criteria of having a clear question and presenting data to answer that question.
According to MMAT criteria, six studies (all qualitative) met all five criteria for quality, meaning that they used an appropriate approach for the study question, their data collection was adequate for their chosen methodology, the findings presented were clearly derived from the data, interpretation was substantiated and there was coherence between data source, collection, analysis, and interpretation. Dutta et al.'s (2016) study did not meet the quality standard about data collection due to focus groups being less than optimal size (having four, three, and two members respectively).
Of the quantitative descriptive studies, the common methodological limitations were sample representativeness and non-response bias. Caplan (1993) reported a response rate of 69% for his study but did not account for non-response bias in his presentation or discus- The response rates for the four sites in Reininghaus et al.'s (2007) study were between 22% and 28%. Such a low response rate must raise concerns about study quality in terms of representativeness of the sample.

| Themes
Four themes were identified during the data analysis phase, comprising 12 subthemes (Figure 2). These themes reflect what is known about the experience of working as a nurse in HSHs, offering insight into what may affect recruitment and retention in the setting.

| Engagement with the patient group
The first major theme had three subthemes. First was the challenge nurses face in developing and maintaining therapeutic relationships with patients in high secure settings, many of whom had committed serious crimes and had diagnoses of severe personality disorder.
Studies showed that HSH nurses aim to approach patients with a non-judgmental attitude to provide care. They develop therapeutic relationships in the long term and through those relationships they aim to foster hope and optimism. The second theme was how FMH nursing was characterized by 'dilemmas and balancing acts' (Beryl et al., 2018), namely the tension between care and control, therapy and risk management described earlier. Negotiating these tensions could be rewarding, with HSH nurses enjoying work between the two worlds of mental health and criminal justice if they felt adequately prepared, educated, supported, and supervised to do their job. The next subtheme was 'patients as traumatized' (Aiyegbusi & Kelly, 2015;Beryl et al., 2018), whereby having potentially committed serious and violent acts prior to admission, high secure patients may have experienced significant traumas themselves. Nurses in HSHs must work with trauma and its effects on behaviour, alongside the imperative to manage risk of violence.

| The ward social environment
The second major theme was the 'ward social environment', with the subtheme of 'differences of opinion about social climate'. Studies showed that as locked environments in locked institutions, social climates differ and fluctuate between wards and HSH nurses must be sensitive to how they may influence and respond to factors affecting social climate. Caplan (1993) used the WAS (Moos, 1974)  A further subtheme relating to the environment was 'gender and sexuality' which featured in five studies. HSH wards are single gender, but with male and female staff. Mercer (Mercer, 2013;Mercer & Perkins, 2014, 2018 directly addressed gender roles and sexuality in one male personality disorder unit in one HSH. He found that talk about pornography and sexual offences was a defining feature of the social setting, leading female staff to feel vulnerable and marginalized. Addo (2006) made a similar observation, that female staff felt vulnerable when working with sex offender patients because they had experienced being the subject of patients' sexual fantasies which could manifest as direct threats and graphic expression of these threats.

| Impact on the nurse
The third major theme: 'impact on the nurse', comprised subthemes of 'stress and distress' 'emotional hard labour', and 'survival strategies'. Reininghaus et al. (2007) and Cramer et al. (2020) used measures of stress and coping in their surveys, alongside other measures of mental health. Scores on measures of mental ill health, including burnout, were low compared with samples of other nursing populations (Cramer et al., 2020). There was a correlation between recent experience of physical assault and likelihood of psychological distress (Reininghaus et al., 2007). Higher self-esteem, higher selfconfidence, and use of a range of coping strategies were 'stress resistance resources' that could moderate the effect of physical assault on psychological distress in some nurses (Reininghaus et al., 2007). Hypothetical explanations for relatively low incidence of trauma or stress symptoms given in these studies include that those who stay working in HSH settings may be resilient individuals, who can cope with trauma and assault well, or that those nurses who experience heightened distress self-select out of the setting or had self-selected out of the research studies.
The next subtheme was 'emotional hard labour'. Alongside the discussion of 'stress' as a feature of being a nurse in HSH, the concept of' 'emotional labour' was invoked in seven studies. Emotional labour, as theorized by Hochschild (1983), is when the worker must deploy and manage their emotional resources as part of their professional duties. Beryl et al. (2018) describe nursing in HSH as 'emotional hard labour', to reflect the personal cost of HSH work, namely the likelihood of assault and the difficulty of working with a disturbed and distressed patient group. When facing 'stress and distress' and undergoing 'emotional hard labour', nurses described or exhibited several 'survival strategies' (as coined by Addo, 2006).
The third subtheme 'survival strategies' were not necessarily healthy or positive reactions to challenging and distressing situations, they could include several 'defence mechanisms': 'avoidance', 'deflection', 'detachment', 'distancing', 'disengage', 'rationalising', 'compartmentalizing', and 'not reflecting' (Addo, 2006, p. 185). Beryl et al.'s (2018) participants used black humour and compartmentalization as ways of coping in their setting, alongside accessing formal support and learning to 'understand why' their patients behaved in certain ways. Newman et al. (2019) made the correlation in their study between experience of vicarious trauma and the participants' use of avoidance, intrusion and hyper arousal, whereby, emotion avoidance' was associated with poor mental well-being.

| Implications for practice
The fourth theme was 'implications for practice', the subthemes of which were: 'educational provision', 'supervision', and 'support'.
Implications for 'educational provision' were that nurses in HSH should have access to specialized training on working with the HSH patient group, for example, on 'formulation and boundary management' (Beryl et al., 2018), psychodynamics (Aiyegbusi & Kelly, 2015), and on theoretical understanding of aspects of offending behaviour (Mercer & Perkins, 2014, 2018. The role of 'support' for nursing staff was discussed in six studies. Reininghaus et al.'s (2007) study measured 'perceived support outside work' finding it an important stress reduction resource. Regarding social support in work, the importance of 'close knit' teams (Beryl et al., 2018) and teamworking, with clear team roles and team commitment were emphasized. Team cohesion, familiarity, and team commitment were considered vital to nurses being able to work well in the HSH setting with long-term patients (Dutta et al., 2016).
In the third subtheme, study authors surmised that supervision and reflective practice should include opportunities to 'process ward dynamics' (Caplan, 1993), learn emotion-focused coping strategies (Cramer et al., 2020), address the risk of vicarious trauma (Newman et al., 2019) and experience of physical assault (Reininghaus et al., 2007).
Managers should be equipped to support staff to develop healthy coping strategies and resilient practitioners (Aiyegbusi & Kelly, 2015).
It is worth noting though that there were other 'implications for practice', specific to the individual included studies but not categorizable under broader themes. These were that differences between staff and patient perceptions of ward climate should be addressed through targeted interventions (De Vries et al., 2016), nurses' career planning should account for the effects of working full time in high-frequency violence environments, for example by negotiating time out of the setting or early retirement (Lauvrud et al., 2009) and finally, for Mercer and Perkins (2018), wider societal discourses on gender and discrimination should be recognized as affecting behaviours in the HSH setting.  (Chalder & Nolan, 2002).The review findings suggest that nurses with high self-esteem and high self-confidence are more suited to working in the HSH. These findings have implications for workforce strategies because they may guide recruitment methods, for example psychometric testing to identify nurses with certain traits and retention strategies, for example ensuring that support and supervision programs address the emotional toll of the work. Employers may also wish to ensure that HSH nurses have an adequate support network outside work and that teams are well supported because these factors appear to counter the effect of the work on the well-being of the nurse. Such recommendations accord with those of the recent reviews of forensic nurses' well-being; however, like those reviews (Brown et al., 2017;Dickinson & Wright, 2008;Freestone et al., 2015), we found that there remains a lack of empirical research testing the effectiveness of specific modes of staff support or supervision.

| Limitations
The limitations of this review are its narrow focus, solely on empirical research on nurses experiences in HSH settings. There may be insights of relevance to recruitment and retention from the wider literature on forensic nursing, for example from studies where the level of security is not specified. The lack of studies specifically analysing recruitment and retention factors in high secure settings have led us to a review that identifies likely rather than proven factors affecting nursing workforce sustainability.
Although Europe, Australia, and North America were represented, there were no studies meeting the inclusion criteria from Africa or Asia. This may be due to differences in classification of security in psychiatric hospitals between countries. The limitations of the included studies reduce the transferability of the findings. First, these were all studies of a single point in time, in specific settings.
There was no longitudinal or intervention research to review. There were some cross-site research and the mapping of themes showed commonality between studies, study sites, and types of study.
Further research is required, for example studies of changes in ward atmosphere over time and studies comparing how environments or nurses' attitudes to their work and intentions to remain in the setting change over time.

| CON CLUS ION
This review has drawn together and evaluated a body of literature not previously examined. The findings suggest that high secure settings have unique characteristics, albeit these may be heightened versions of nurses' experience in other forensic and mental health nursing settings. The study is of interest to an international audience because the challenges identified speak to the wider nursing workforce crisis.
The review findings elaborate and offer specificity on broader analyses of factors affecting nursing workforce recruitment and retention.
Findings suggest that future workforce retention strategies must incorporate tailored education, support and supervision to address the specific features of nursing in high secure settings. Furthermore, recruitment and retention strategies must take account of the characteristics of nurses who thrive in the high secure setting.

IMPAC T S TATEMENT
The review findings should inform workforce strategy in high secure forensic mental health hospitals. Recruitment and retention policy must be tailored to account for the provision of education, supervision, and support required by the ethical challenge and emotional labour of nursing in this setting.

CO N FLI C T O F I NTE R E S T
No conflict of interest has been declared by the author(s).

AUTH O R CO NTR I B UTI O N S
JO,AT,IE,EW,AMR made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; involved in drafting the manuscript or revising it critically for important intellectual content; gave final approval of the version to be published. Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content; and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.