What influences the use of seclusion? A cross-sectional study of forensic mental health nurses in a UK high secure hospital

Abstract

Purpose

Incidents of violence and aggression are a regular occurrence within adult forensic mental health inpatient settings and often lead to the use of restrictive practices such as seclusion. Such events are frequently attributed to the complexity of the patients. Research commonly focuses on patient’s characteristics and their association with seclusion use. Less attention has been centred on forensic mental health nurses’ attitudes to seclusion and the association of nursing staff characteristics.

Design/methodology/approach

A cross-sectional survey was undertaken using a standardised questionnaire, “Survey of Nurses’ Attitudes to Seclusion”. Responses were received from n = 147 nurses at a high secure forensic mental health hospital in the UK.

Findings

Key findings indicate that most participants believed seclusion should remain part of clinical practice. A correlation was identified between forensic mental health nurses’ attitudes to seclusion use and their characteristics: gender, age range, educational level and experience.

Practical implications

This paper presents novel information on seclusion reduction opportunities through modifiable workforce factors such as gender-sensitive rostering and staff training and development. Furthermore, recruitment and retention strategies should be prioritised so forensic mental health is perceived as an attractive career and a safe workplace.

Originality/value

The paucity of research in this area has prompted calls for further research to explore nursing staff characteristics and seclusion use. This is particularly important now due to the current global difficulty in the recruitment and retention of mental health nurses.

Citation

Tulloch, L., Walker, H. and Ion, R. (2024), "What influences the use of seclusion? A cross-sectional study of forensic mental health nurses in a UK high secure hospital", The Journal of Forensic Practice, Vol. 26 No. 4, pp. 201-218. https://doi.org/10.1108/JFP-05-2024-0022

Introduction

Seclusion is best defined as the separation of a patient from their peer group within a defined room or area from which they are not permitted to leave to manage behavioural disturbances (Council of Europe, 2017). Its use is only justified as a last resort when de-escalation and all other preventative strategies have been exhausted and there continues to be concern of harm to others if no action is taken [McKeown et al., 2019; World Health Organization (WHO), 2020].

Adult forensic mental health inpatient settings (AFMHIS) are volatile environments that are populated with acutely mentally unwell and high-risk offenders with a history of significant violent offences (Barr et al., 2019). Forensic mental health nurses (FMHN) who work in these settings must perform a dual role of preserving the safety of patients and others and maintaining a therapeutic nurse–patient relationship (Hui, 2017).

It is reported that almost all mental health nurses have been assaulted while on duty at some point, and this occurs at a much higher rate within forensic mental health (Renwick et al., 2019). It is, therefore, unsurprising that fear in the workplace is an issue that is encountered on a daily basis (Jalil et al., 2017). Fear of assault can impact the clinician’s ability to establish and maintain therapeutic nurse–patient relationships, as well as influence nurses’ attitudes and clinical decision-making regarding the use of restrictive practices such as seclusion (Muir-Cochrane et al., 2018).

Seclusion is primarily adopted in mental health inpatient settings, although its use is more prevalent in AFMHIS, where there is increased evidence of patient restlessness, aggression and violence (Maguire et al., 2019). It has become a controversial ethical challenge due to a lack of supporting evidence of its effectiveness, safety, utility in clinical risk management and therapeutic value (Jalil et al., 2017). It was condemned by the Ashworth Inquiry [Department of Health (DoH) 2008] in the 1990s and declared inhumane and tantamount to torture in 2013 by the United Nations “Report of the Special Rapporteur on torture and other cruel, inhuman or degrading treatment or punishment” (Mendez, 2013). For this reason, international efforts have focused on reducing the use of seclusion through the introduction of legislative guidance and models of care such as Six Core Strategies and open-door psychiatry (Bowers et al., 2015; W. Haugom et al., 2019; Lau et al., 2020; Maguire et al., 2021). Nonetheless, reports of seclusion practices being adopted across the globe perpetuate a culture of control and coercion (Ion et al., 2020).

Background

While seclusion data is regarded as an essential metric for evaluating care delivery (Barr et al., 2022a, 2022b), attempts to understand the scale of its use are hampered by inconsistencies in standardised data collection, data interpretation and data reporting across and within countries (Barr, 2023). As a result, there has been an increased call for standardised definitions of seclusion and data collection to report on the frequency and prevalence of use among mental health and AFMHIS internationally (Al-Maraira and Hayajneh, 2019). To the best of our knowledge, the four nations of the UK do not have a standardised national reporting database that captures the prevalence of seclusion use within AFMHISs. This contrasts with Australia, which can report and examine the rate of seclusion use in forensic mental health inpatient settings and published a troubling threefold increase in the incidence of seclusion use from 2008 to 2020 (Barr et al., 2023).

Research suggests that mental health nurses impact the incidence rate and persistent use of seclusion (Barr et al., 2019; Happell and Koehn, 2010). This, however, is not unexpected, considering nurses spend the most time with patients and play a significant role in determining whether or not to seclude a patient in the face of violence and aggression (Barr et al., 2022a, 2022b). Given the caring nature of the nursing profession, it seems unreasonable to suggest that nurses cause aggression. Nevertheless, there are reports of nurses’ attitudes and practices contributing to inpatient aggression and use of seclusion (Bregar et al., 2018; W. Haugom et al., 2019), particularly when patients perceive their attitudes and behaviour as being custodial or abusive (Tomlin et al., 2020).

Nursing staff characteristics, including gender, age, level of education and level of experience, have been the focus of studies within acute and AFMHIS, although the findings are inconsistent (Doedens et al., 2021). Research conducted in the acute mental health setting suggests that male nurses are more likely to adopt coercive measures (Happell and Koehn, 2010) and that seclusion use decreases with the presence of female nurses (Morrison and Lehane, 1995; Doedens et al., 2021). A further study by Barr et al. (2022a, 2022b) established no significant correlation between nurses’ gender and seclusion use. However, they did establish that registered nurses on duty contributed to higher rates of seclusion use.

A recent systematic literature review demonstrated the paucity of research on nurses’ characteristics and their influence on seclusion in the AFMHIS setting (Barr et al., 2022a, 2022b). This review identified only three studies examining nurses’ characteristics and seclusion use in AFMHIS, published between 2010 and 2020. All three studies examined differences in male and female practice and their influence on restrictive practice use and found no significant correlation (Boumans et al.,2012; De Benedictis et al., 2011; Mann-Poll et al., 2011).

The paucity of research in this area has prompted calls for further research to explore nursing staff’s characteristics and seclusion use (Boumans, 2012; Hui, 2017; Lawrence et al., 2021). This is particularly important now due to the current global difficulty in the recruitment and retention of mental health nurses (WHO, 2020), an ageing workforce (Barr et al., 2022a, 2022b) and a wave of resignations (Phiri et al., 2022). NHS Scotland Workforce Data (2023) show Mental Health Nurse vacancies have increased over the past five years reported from 712 Whole Time Equivalent (WTE) vacancies in 2018 to 1,377 WTE vacancies in 2023, nearly doubling since prior to the COVID-19 pandemic. Approximately 12% of mental health nursing positions remain vacant, raising concerns about the future workforce within Scotland’s AFMHIS (NHS et al., 2023).

An under-resourced workforce lacking experienced mental health nurses risks a workforce who do not feel safe and are insufficiently skilled to work in AFMHIS, potentially resulting in nurses resorting to the use of controlling and restrictive practices. Reports of which have led to mass media coverage and political scrutiny following the exposure of the abuse of patients by nurses at Winterbourne View, Whorlton Hall (George, 2019) and Edenfield Centre (Limb, 2024).

The researcher’s objective to provide valuable, translational knowledge that would positively aid the well-being of nurses, including the identification of modifiable workforce decisions, was an important aspect of the study design, to achieve this, a pragmatic research design to look at this problem was adopted (Goldkuhl, 2012).

The study aimed to examine FMHN’s attitudes and views on seclusion, determine whether their characteristics influence its use and answer the following questions:

Q1.

What are the nursing staff’s attitudes and views on seclusion use?

Q2.

What are the nursing staff’s motives, beliefs and feelings about seclusion use?

Q3.

What nursing staff characteristics influence or inhibit attitudes and perceptions of the use of seclusion?

Methods

Setting

The setting was a high secure adult forensic mental health inpatient setting based in the UK that provides care, assessment and treatment for up to 140 male patients, all of whom are detained under national mental health legislation.

Measures

Two instruments were used to gather demographic and attitudinal data. The first; captured nursing staff characteristics such as age range, gender, role, experience working in forensic mental health and the last occasion they cared for a patient in seclusion. The second; was the Survey of Nurses’ Attitudes to Seclusion (SNASS) (Heyman, 1987). The SNASS is a standardised questionnaire with five subsections using closed questions that are measured using a combination of three and four-point Likert scales ranging from 1 (strongly agree) and 4 (strongly disagree) to 1 (never), 3 (unsure). It seeks to explore opinions, beliefs and feelings of seclusion across different areas. Participants rated:

• motives for the use of seclusion (13 questions);

• views regarding the use of seclusion (17 questions);

• views on patients’ reaction to seclusion use (13 questions);

• views on the efficacy of seclusion (13 questions); and

• proposed changes in seclusion practice and environment (8 questions).

The SNASS has been frequently used within acute and forensic mental health settings, most recently in 2019, to measure nursing staff’s opinions, beliefs and feelings of seclusion use, including changes in seclusion practice and environment (Wynaden et al., 2001; Meehan et al., 2004; Happell and Koehn, 2010, 2010; Bregar et al., 2018, 2019). It has demonstrated good test–retest reliability of 0.62–0.79 and adequately captures FMHN’s attitudes to seclusion (Bregar et al., 2018). Cronbach’s alpha coefficient was used to assess the internal consistency of each sub-scale of the current study. The coefficient ranged from 0.65 to 0.78.

Ethical considerations

The research committee at one of the UK's high secure forensic mental health hospitals provided the study approval.

Recruitment

A potential total participant sample of N= 293 were contacted through the hospital's email system, filtered by the nursing professionals. The first author sent the email invitation, and enclosed was written information about the study, its aims, consent and confidentiality. All participants were advised that their involvement was entirely voluntary and that they were not obliged to participate should they decide not to respond. Written information on the study stipulated that participation was voluntary and anonymity would be protected throughout the data collection, storage and reporting of results and subsequent publications. No incentives were offered during the recruitment and participation process.

Potential participants were invited to take part by printing the SNASS questionnaire and returning it to the first author within 30 days. Consent was implied upon its completion and return. All returned questionnaires were securely stored and only accessible to research team members. Weekly staff bulletins were issued and served as reminders to potential participants on the study, with details enclosed on how eligible participants can engage.

Study sample

Participants were included if they met the following criteria:

• a current employee of the high secure forensic mental health hospital; and

• any grade of nursing staff (registered or unregistered).

Unregistered nursing staff are also known locally as “healthcare support workers”; registered and non-registered (healthcare support workers) nurses were invited to participate. For the purposes of this paper, the term nurse(s) includes registered and non-registered nurses. Non-nurses were excluded from the study.

A total of n = 147 members of nursing staff completed and returned the SNASS, representing a 50% completion rate. The sample had an even response by gender: 54% (n = 79) male and 46% (n = 68) female. Seventy percent (n = 103) were aged 42 years and over, and almost half, 47% (n = 70) aged 54–65. The majority of the sample, 71% (n = 105), had spent more than ten years working within forensic mental health inpatient setting, 83% (n = 122) were registered nurses, and 17% (n = 25) were unregistered. The majority of respondents, namely, 70% (n = 107), reported having cared for a patient in seclusion within the last six months, while the remainder, 18% (n = 27) between the last 1 and 2 years, 3% (n = 5) in the last 3–4 years and 6% (n = 8) over four years.

Data analysis

Data analysis was performed using IBM Statistical Package for Social Sciences version 26 (IBM Corp., 2019). Descriptive statistics were used to describe demographic profiles. Mann–Whitney U, Kruskal–Wallis and Chi-square tests were used to compare statistically significant differences between variables (Field, 2005). All results within this study were considered statistically significant at p < 0.05.

Results

Results revealed significant differences in nursing staff’s characteristics, namely, age range, gender, role and experience (in years range) of working in forensic mental health, the last occasion (in years range) they cared for a patient in seclusion compared to their opinions, beliefs and feelings of seclusion use, and motives and views of its use.

Participant’s motives for use of seclusion

Participants were invited to indicate the most likely scenario that influenced the use of seclusion. The most common responses, indicative of “sometimes” or “often” motives for the use of seclusion, were when the patient hit another patient, with the majority reporting (90%; n = 133), including instances where the patient hit a staff member (89%; n = 131). The majority of respondents rated that seclusion was never used because the patient is cursing or swearing at other patients (70%, n = 103), including not taking medication (87%, n = 128) or waking other patients at night (78%, n = 115).

There was no significant difference in opinion between gender, educational level, age or length of experience about motives for seclusion use.

Participant’s views regarding use of seclusion

The participants' views on seclusion use varied (see Table 1). Over half (55%; n = 81) reported never feeling annoyed that the patient was secluded, and 68% (n = 101) never felt angry at being made to be involved in a process that they did not agree with. However, 59% (n = 86) often felt regretful that the crisis was not resolved differently. No significant differences were found through analysis of different characteristics.

A majority, 76% (n = 112), reported feeling in control of the situation; 41% (n = 61) male and 35% (n = 51) female, and 84% (n = 124) were satisfied that the ward was running smoothly as a direct consequence of a seclusion event. The results indicated that females felt significantly less satisfied than males that the ward was running smoothly (U = 2,183.00, p = 0.03), and nursing staff with 25–30 years experience of working within forensic mental health were less satisfied (H = 47.19, p = 0.004) compared to those with only 1–5 years of experience.

More than half of the participants, 60% (n = 87), expressed the view of being fed up with the use of seclusion. Those aged (54–65) years were significantly (H = 48.441, p = 0.026) more fed up compared to the 18–29 age range. In total, 54% (n = 79) reported never feeling empowered with the use of seclusion, and the (54–65) year age group were significantly (H = 32.494, p = 0.020) less empowered by comparison to the 18–29 age range. No other differences were noted.

Participant’s perception of patient’s reaction to seclusion use

The majority viewed the patient as feeling angry (81%, n = 119), scared (67%, n = 99) or confused (75%, n = 111), also identifying the patient as feeling helpless (65%, n = 96) or depressed (57%, n = 84). Thirty-three percent (n = 48) said that they believed a patient was satisfied following seclusion; 39% (n = 57) disagreed, indicating the patient was not satisfied; 27% (n = 42) were unsure. This is a similar finding when asked if they considered the patient felt punished; 40% (n = 58) said yes, 37% (n = 54) said no and 24% (n = 35) were unsure.

Opinions differed by gender; females more often considered patients to feel depressed or sad when in seclusion (X² = 9.98, p = 0.007) and viewed the patient as being more out of control than their male counterparts (X² = 7.10, p = 0.02). When comparing the ratings of those who reported having cared for a patient in seclusion more recently, i.e. within the past six months, as opposed to a lengthier time period, i.e. over four years ago, there was a significant difference in the views on how angry the patients felt. Those with recent experience believed that seclusion makes the patients feel angry (X² = 25.968, p = 0.001) and safe (X² = 16.96, p = 0.03).

There was a difference in opinion between registered and unregistered nurses on how they viewed the emotional impact on patients. Registered nurses were likelier to report that the patient felt depressed or sad (X² = 8.20, p = 0.01) and believed that seclusion helped the patient feel calm (X² = 9.04, p = 0.01). Further differences emerged between the group that fell into the mature age range (54–65) by comparison to the 18–29 age range in relation to the feeling of disempowerment of the patient (X² = 18.368, p = 0.019) and level of helplessness (X² = 17.634, p = 0.024). The former believed the use of seclusion exacerbated both. Finally, participants with over 30 years of experience, compared to those with 6–10 years of experience, held statistically significant differences of opinion in several areas; those with over 30 years of experience considered secluded patients to feel: unhappy (X² = 25.692, p = 0.012), disgusted (X² = 21.064, p = 0.019), relieved (X² = 23.968, p = 0.021) and confused (X² = 23.009, p = 0.028).

Participant’s views on efficacy of seclusion

Nearly all (97%; n = 142) thought that seclusion helps calm the patient. Eighty-two percent (n = 120) believe it controls the patient’s behaviour, and 72% (n = 106) believe it makes the patient behave better when seclusion has ended. The majority (86%; n = 126) also believe that seclusion can often make the patient feel frustrated and 90% (n = 132) angry towards nursing staff, whereas 44% (n = 64) believe it sometimes makes the patient feel that nursing staff care about them (see Table 2).

There were statistically significant differences of opinion between participants of different genders who had experience working in forensic mental health and caring for a patient in seclusion. More females than males believed that seclusion makes the patient feel frustrated (U = 2,114.00, p = 0.005), as did those with 11–15 years of experience working within forensic mental health (H = 32.494, p = 0.020), by comparison to those with less experience, i.e. 6–10 years.

There was no further significant difference in opinion found in educational level or age range views on the efficacy of seclusion.

Participant’s proposed changes in seclusion practice and environment

The participants believed that changes were required to the seclusion practice and setting (see Table 3). Over half (52%, n = 76) agreed that changes to the seclusion room were needed, and a majority (69%, n = 101) reported that the room should be painted to have a calming effect. There was a statistically significant difference in response between registered and unregistered nurses on this matter, with registered nurses believing colour can have a beneficial impact (U = 932.00, p = < 0.000).

Sixty-six percent (n = 97) indicated the patient should be able to listen to music if they want to, and 71% (n = 104) accepted reading material could be provided. An overwhelming majority, 88% (n = 130), believed that a staff member should always remain with the secluded patient, and 91% (n = 133) were of the opinion that seclusion should remain part of clinical practice. Participants with more experience (over 30 years) working within forensic mental health also identified this as beneficial, differing significantly from those with only 1–5 years of experience (H = 44.083, p = 0.016). No further significant difference in opinion was found in gender, educational level or age range views of participants’ proposed changes in seclusion practice and environment.

Discussion

Several significant findings have emerged from this study, suggesting that differing characteristics of FMHN can play a crucial role in seclusion use. Differences emerged in gender, age range, role and experience, all of which were identified as influencing its use. Reflecting on and considering these differences could positively influence patient care delivery and outcomes.

Nursing staff’s views of seclusion

Key findings indicate that an overwhelming majority of study participants were of the opinion that seclusion should remain part of clinical practice. Most of the study participants perceived seclusion as an effective method of controlling patient behaviour, calming the patient down and improving the patient’s behaviour after seclusion. Although participants strongly denied that seclusion was used as a punishment, they did concede that seclusion should remain an option for the management of violence and aggression. This is consistent with the findings of other studies, whereby seclusion was described as an unwanted but necessary intervention to protect staff and patient safety (Green et al., 2018; Aragonés‐Calleja and Sánchez‐Martínez, 2023).

While calls reduce and eliminate seclusion has dominated the mental health care landscape, there remains agreement amongst authors across the globe that a complete ban on seclusion may not be realistic or safe at present (Barr, 2023; Gildberg et al., 2015; Mathias and Hirdes, 2015). Although current frameworks and models have effectively reduced some conflict (Maguire et al., 2018), Barr (2023) highlighted that some patients do not respond to less restrictive interactions, such as de-escalation. Occasionally, incidents increase due to inadequate nursing staff skill mix and insufficiently trained and experienced nurses (Fukasawa et al., 2018; McKeown et al., 2019).

Staff characteristics influencers and inhibitors of seclusion use

Gender.

Key findings suggest that experienced, mature females were less likely to use seclusion, unlike young, inexperienced males. This result is not consistent with the findings in acute mental health settings, where it was found that the proportion of female nurses on a ward increased the odds of a patient being secluded (Doedens et al., 2021). Within AFMHIS, reports on the influence of gender are conflicting. Urheim et al. (2020) found that an increase in the number of female nurses was significantly related to reduced violent incidents, while Barr et al. (2022a, 2022b) found no association between gender and the use of seclusion. Goodman et al. (2020) reported that patients felt safe around nurses they could trust; these were often females who possessed enhanced de-escalation skills and were seen as less dominant or threatening, which reduced the patient's fear and sense of powerlessness.

The results of this study identified that male respondents experienced more satisfaction with the use of seclusion and were of the opinion that it aided the smooth running of the ward. Male nurses also perceived secluded patients as feeling out of control, as opposed to female nurses who were of the opinion that secluded patients felt sad, depressed and frustrated, resonating with findings from Bregar et al. (2018). One possible explanation is that males are more likely to be assault victims than females, thus feeling a more significant burden and responsibility (Foster et al., 2007). As noted by Beghi et al. (2013), aggression by patients tends to be directed toward people of the same gender, and male patients are restrained the most. Combined with an over-reliance on male nurses for aggression management interventions, this can lead to patients perceiving male nurses as especially coercive, which can result in patients being less likely to respond positively to male nurses’ de-escalation efforts (Goodman et al., 2020). In addition, Forté et al. (2017) reported that following a violent incident, nursing staff would either present as hypervigilant or attain fear of violence from other patients. This may explain why nurses, in particular male nurses, who work in volatile and unpredictable environments are more likely to fear for their safety and take a more restrictive approach to violent and aggressive care management (Barr et al., 2019). The implication of a link between gender and the use of restrictive practices raises questions of whether male nursing staff assess the risk of violence and aggression through the lens of fear and for their safety? This perhaps explains why they are more often persuaded to resort to seclusion; similar findings were highlighted throughout the literature (Bregar et al., 2018; Fariña‐López et al., 2014; Gelkopf et al., 2009). It is, therefore, essential that workforce planning and the gender placement of nursing staff are carefully considered to ensure there is a balance between both female and male nurses to reduce the likelihood of patient violence and use of seclusion.

Age range.

The results highlighted differences of opinion dependent upon age range; more specifically, nursing staff with a more mature age profile: age range 54–65, thought seclusion made the patient feel disempowered and helpless, differing from those with a younger age profile: 18–29 years. These results are not consistent with findings in acute mental health settings where no significant association between nurses’ age and seclusion use were found (Doedens et al., 2021). One possible explanation for this is that more mature nurses have learned to manage their emotional reactions to violence and aggression and are more reflective in their perception of the trauma-related impact that seclusion has on the patient. Lawrence et al. (2021) note that the use of restrictive practices can have an impact on nurses’ emotional well-being, with some feeling traumatised and a sense of unease and conflict, which can lead nurses to process, reflect and understand their actions over time.

With almost half of the study participants being aged 54–65 years of age, essential workforce implications and challenges were identified, specifically regarding impending nursing turnover through retirement, resulting in a loss of forensic mental health experience and skills. This is an important issue due to the current difficulty of recruiting and retaining mental health nurses globally (WHO, 2020), compounded by an ageing workforce (Barr et al., 2022a, 2022b) and resignations (Phiri et al., 2022). It raises concerns about the future workforce lacking experienced FMHN and the potential for an increase in seclusion use as a consequence. Workforce forecasting and urgent interventions by senior managers and universities are required to influence nurse recruitment and retention, promote future nurse training programmes, provide enhanced in-service training to ensure that nurses working within AFMHIS feel safe and adequately skilled to work in AFMHIS and do not resort to controlling and restrictive practices.

Registered/unregistered nurses.

There was a difference in opinion between registered and unregistered nurses concerning how they viewed the emotional impact on patients. Registered nurses were more likely to report that the patient felt depressed or sad but believed that seclusion helped the patient feel calm. This finding may be explained by the higher level of training and education by registered nurses, leading to greater insight into the patient experience. Urheim et al. (2020) found that an increase in nurses with higher education levels was strongly associated with a decrease in violent occurrences in the AFMHIS ward.

Findings from this study suggest that unregistered nursing staff hold the presumption that seclusion has a therapeutic effect on the patient. Specifically, it makes the patient feel calmer. This is despite widespread research evidence suggesting that seclusion is harmful (Lawrence et al., 2021), with patients reporting seclusion experiences as traumatic, emotionally unsettling, disorienting and not conducive to mental health recovery (Hui, 2017).

The statistically significant finding that registered nurses believed changes in the wall colour of the seclusion room could benefit the patient's condition when compared to unregistered nurses suggests that registered nurses can see beyond the patient's mental state and recognise the negative environmental impact on patients. Research has shown that patients regard environmental conditions as reasons for aggression (Urheim et al., 2020).

Forensic experience.

Most participants had over ten years of experience working in forensic mental health inpatient settings. In line with earlier studies (Bowers et al., 2007; Happell and Koehn, 2010), nurses with extensive experience feel less in need to be in control and are less likely to use seclusion when the patient presents with, for example, inappropriate sexual behaviour when compared to those with lesser experience.

Participants with extensive work experience in this study considered the detrimental impact on secluded patients, including unhappiness, disgust and confusion. This is consistent with the findings of Krieger et al. (2020), who found that a more critical view of coercion was held by nurses with the most work experience. Our results also indicated that nurses' dissatisfaction with using seclusion as an intervention increases over time and may be due to increased exposure to seclusion, which exposes nursing staff to a negative impact on the patient. This result is not consistent with Doedens et al. (2021), who found no significant association between seclusion use and experience in mental health care.

While an overwhelming majority of participants believed that a nurse member should always remain with the secluded patient, participants with greater length of experience (over 30 years) working within forensic mental health also identified this as beneficial, more than those with only 1–5 years experience. Consistent with the findings from Holmes et al. (2015), the majority of study participants believed that providing the patient with some calming stimulus, such as music or reading material, could benefit the patient. This is an important shift in perception about the purpose of seclusion, which has traditionally been viewed as a means to de-stimulate patients. Our findings suggest that nurses are more cognisant that providing some stimulation, whether it be human interaction or some other method of occupation, is important to patients to relieve boredom and loneliness in the seclusion room.

The recency of seclusion experience.

Notably, participants who had more recent experience with seclusion believed that it impacted the patients more than those who had not been involved in the process for some time. The recency leads to greater awareness and sensitivity to negative emotions evoked through seclusion use.

Opportunities for changes in seclusion practices.

The key to preventing seclusion use is to prevent incidents of violence and aggression. However, as patient acuity rises, nurses working in AFMHIS need to be able to respond early, efficiently and effectively to prevent and manage incidents of violence and aggression while promoting an environment of patient safety and recovery. Unfortunately, empirical evidence suggests that poor organisational culture can lead to a pervasive and inappropriate use of seclusion (Paterson et al., 2013). Therefore, services must monitor and evaluate all incidents of seclusion within their service to ensure the principles of “last resort” are maintained.

The findings from this study are consistent with those from a recent study in Spain (Aragonés‐Calleja and Sánchez‐Martínez, 2023), whereby clinicians normalised restrictive practices and considered them implicit to daily practice while also acknowledging them as a “necessary evil”. While seclusion can be an opportunity to regain control in situations of escalating risk on the ward, caution is needed where seclusion becomes a default intervention for minor misdemeanours, which leads to and reinforces controlling attitudes and behaviours of nurses, which can negatively impact the social climate of the unit, leading to a cycle of violence and seclusion (Askola et al., 2018). When faced with volatile and unpredictable environments, nurses often take a more restrictive approach to managing incidents of violence and aggression due to fear for their safety and the safety of others (Barr et al., 2019). Organisations must ensure that seclusion is not used pre-emptively as a method of managing the ward environment as opposed to being a last-resort intervention in response to violence and aggression (Rabab et al., 2020).

Understanding the complex and conflicting values between staff gender differences associated with attitudes to seclusion use can also be useful when considering targeted interventions to understand in more detail the ethical challenges that staff experience and ensure a gender-balanced staffing profile (W. Haugom et al., 2019). In addition, services must prioritise training and education that sufficiently equips nursing staff working in the AFMHIS with the skills needed to meet the needs of this complex patient cohort. The emphasis is placed on development needs, communication styles and relational skills, including emotional reactions and coping skills on how to deal with patient violence, which is a fundamental influencing factor in reducing seclusion use (Gelkopf et al., 2009).

Senior managers must provide support platforms which provide opportunities for nurses to strengthen their self-awareness and reflect on practice while also having the dual benefit of optimising the resilience of nurses, which can improve the quality of care for patients in this speciality setting (Hellzén et al., 2023). Nurses should be provided opportunities to explore self-awareness through scenario-based replicated settings to develop emotional control and resilience. Thus, encouraging a recovery-focused, trauma-informed approach to the management of violence and aggression is required (Galbert et al., 2023).

Finally, forensic mental health services must prioritise recruitment and retention strategies that future-proof forensic mental health nursing as an attractive career, a safe workplace and an area where there are opportunities to develop nurses as professionals.

Limitations

This was a cross-sectional study, and responses may have varied if completed at a different time point. In addition, the results were skewed. Therefore, non-parametric tests were used. The sample was not balanced concerning age range and experience of working within forensic mental health, with the majority of the sample, 70% (n = 103) aged 42 years and over, and 71% (n = 105) had more than 10 years of experience of working in AFMHIS. It is conceivable that a younger and less experienced workforce who were willing to complete and engage in the study held differing opinions of seclusion use than those who participated; there may, therefore, have been a response bias. Finally, there was a small and varied proportion of missing data (<5%); however, due to the percentage of missing data being under 5% and at random, it would not have significantly impacted the analysis. Therefore, these cases remained in the analysis.

Conclusion

Key findings indicate that an overwhelming majority of study participants were of the opinion that seclusion should remain part of clinical practice. This finding reinforces that seclusion continues to be normalised within AFMHIS and is considered an effective tool for maintaining safety in daily practice. While findings suggest that experienced, female and mature nurses were less likely to use seclusion, unlike young, unregistered males, the results indicate an urgent need for preventative training, careful workforce planning and protection of the forensic mental health nursing speciality within nurse education should be prioritised, to influence the attitudes of future generations towards seclusion use positively.

Opportunities to improve seclusion practices and patients' experiences of seclusion were identified, including changes to the seclusion room, patient access to supportive tools such as music or reading material, and continuous support and supervision. Although it is clear, a paradigm shift in AFMHIS is required to achieve international efforts to eliminate seclusion in clinical practice.

Implications for practice

• Seclusion remains an intervention of last resort for clinicians when all the least restrictive options have been exhausted.

• Seclusion was perceived to be an effective intervention to control patient behaviour and was reported to be used when violence against staff, patients and property occurred.

• Some differences in nurses’ characteristics and attitudes were identified when evaluating views of seclusion.

• Mental health nursing staff could benefit from formal training in non-coercive practice. This would facilitate the operative implementation of effective interventions to manage complex relationships, prevent aggression and promote recovery while reducing the need for seclusion.

• Further research is required to understand the lived experiences of those exposed to violence and aggression and the subsequent use of coercive practices. This would allow for a deeper grasp of what factors and contributors, such as gender and age range, influence attitudes towards its use.

Table 1

Participant’s views regarding use of seclusion

				Mann–Whitney U				Kruskal–Wallis/ Monte Carlo		Kruskal–Wallis/ Monte Carlo		Kruskal–Wallis/ Monte Carlo
				Male/ Female		Registered/ Unregistered		Age range		Experience of FMH		Experience of seclusion
	Never n (%)	Sometimes n (%)	Often n (%)	U	(P)	U	(P)	Asymp sig.	Adj. Sig.a	Asymp sig.	Adj. sig.a	Asymp sig.	Adj. sig.a
Satisfaction with helping the patient	9 (6.1)	90 (61.2)	48 (32.7)	2,623.000	(0.775)	1,421.500	(0.533)	0.967	0.966	0.687264445	0.711	0.687264445	0.712
Annoyed that the patient was secluded	81 (55.1)	56 (38.1)	10 (6.8)	2,603.500	(0.716)	1,349.500	(0.304)	0.591	0.608	0.115702702	0.098	0.115702702	0.103
Relieved that the problem has been resolved	32 (21.8)	66 (44.9)	48 (32.7)	2,537.000	(0.533)	1,485.500	(0.826)	0.301	0.324	0.212587033	0.218	0.212587033	0.222
Satisfaction that the ward is running smoothly	23 (15.6)	75 (51.0)	49 (33.3)	2,183.000	(0.031	1,341.000	(0.296)	0.219	0.216	0.000	0.004	0.237593217	0.247
Guilt or misgivings about the necessity for secluding the patient	77 (52.4)	54 (36.7)	16 (10.9)	2,524.000	(0.483)	1,506.000	(0.913)	0.190	0.187	0.644433909	0.673	0.644433909	0.682
Regretful that the crises wasn't resolved differently	61 (41.5)	70 (47.6)	16 (10.9)	2,230.500	(0.060)	1,430.500	(0.590)	0.067	0.063	0.728421886	0.748	0.728421886	0.753
Powerful	119 (81.0)	23 (15.6)	5 (3.4)	2,500.000	(0.289)	1,364.000	(0.223)	0.071	0.062	0.938470851	0.954	0.938470851	0.952
Angry that it was a mistake to have secluded the patient	96 (65.3)	45 (30.6)	6 (4.1)	2,395.000	(0.174)	1,310.500	(0.183)	0.109	0.101	0.405925705	0.413	0.405925705	0.427
Fed up	60 (40.8)	70 (47.6)	17 (11.6)	2,589.500	(0.679)	1,420.500	(0.552)	0.003	0.025	0.558553646	0.575	0.558553646	0.586
That you have failed	90 (61.2)	48 (32.7)	9 (6.1)	2,411.500	(0.213)	1,381.000	(0.386)	0.367	0.369	0.667703188	0.695	0.667703188	0.702
Disempowered that it ended this way	80 (54.4)	54 (36.7)	13 (8.8)	2,384.500	(0.187)	1,290.500	(0.173)	0.023	0.017	0.240331687	0.239	0.240331687	0.242
In control of the situation	35 (23.8)	74 (50.3)	38 (25.9)	2,329.500	(0.131)	1,276.500	(0.162)	0.207	0.204	0.140888318	0.141	0.140888318	0.142
Angry at being made to be involved in a process that I do not agree with	101 (68.7)	38 (25.9)	8 (5.4)	2,467.000	(0.294)	1,336.000	(0.229)	0.557	0.563	0.760168478	0.815	0.760168478	0.808

Notes:

n = number; % = percentage; U = value of Mann–Whitney test; P = statistical significance; FMH = forensic mental health

Source: Authors’ own creation

Table 2.

Participants views on efficacy of seclusion

					Mann–Whitney U				Kruskal– Wallis/ Monte Carlo		Kruskal– Wallis/ Monte Carlo		Kruskal– Wallis/ Monte Carlo
					Sex		Registered/ unregistered		Experience of FMH		Experience of FMH		Experience of seclusion
	Never n (%)	Sometimes n (%)	Often n (%)	Unsure n (%)	U	(P)	U	(P)	Asymp sig.	Adj. sig.a	Asymp sig.	Adj. sig.a	Asymp sig.	Adj. sig.a
It helps them calm down	3 (2.0)	91 (61.9)	51 (34.7)	2 (1.4)	2,456.500	(0.293)	1,326.000	(0.226)	0.172	0.172	0.760	0.769	0.504	0.522
It makes them feel frustrated	12 (8.2)	103 (70.1)	23 (15.6)	9 (6.1)	2,114.000	(0.005)*	1,354.500	(0.276)	0.530	0.532	0.018	0.017	0.807	0.775
It makes them behave better when they are let out	24 (16.3)	83 (56.5)	23 (15.6)	17 (11.6)	2,398.500	(0.214)	1,343.000	(0.297)	0.914	0.915	0.104	0.106	0.045	0.044
It decreases frustrating social interactions with other people	14 (9.5)	95 (64.6)	21 14.3)	17 (11.6)	2,550.000	(0.534)	1,505.500	(0.905)	0.794	0.797	0.922	0.928	0.066	0.069
It makes them feel angry towards staff	3 (2.0)	98 (66.7)	34 (23.1)	12 (8.2)	2,453.500	(0.277)	1,407.000	(0.464)	0.057	0.053	0.824	0.834	0.519	0.536
It allows them to express angry feelings in a way that's not destructive to the rest of the ward	8 (5.4)	96 (65.3)	35 (23.8)	8 (5.4)	2,535.500	(0.487)	1,232.000	(0.072)	0.069	0.066	0.111	0.104	0.056	0.055
It makes them feel staff care about them	38 (25.9)	64 (43.5)	4 (2.7)	41 (27.9)	2,639.000	(0.845)	1,487.500	(0.836)	0.111	0.112	0.660	0.672	0.396	0.416
It changes the way they feel	17 (11.6)	93 (63.3)	9 (6.1)	28 (19.0)	2,658.000	(0.899)	1,328.500	(0.238)	0.191	0.186	0.548	0.555	0.868	0.867
It changes the way they behave	15 (10.2	104 (70.7)	19 (12.9)	9 (6.1)	2,645.500	(0.844)	1,505.500	(0.900)	0.963	0.967	0.980	0.98	0.230	0.244
It does not help patients at all	53 (36.1)	61 (41.5)	12 (8.2)	21 (14.3)	2,678.000	(0.973)	1,331.000	(0.285)	0.057	0.534	0.280	0.284	0.818	0.831
It disempowers them	34 (23.1)	70 (47.6)	27 (18.4)	16 (10.9)	2,463.000	(0.353)	1,490.000	(0.846)	0.768	0.778	0.689	0.697	0.477	0.489
It controls their behaviour	19 (12.9)	94 (63.9)	26 (17.7)	8 (5.4)	2,526.000	(0.467)	1,432.000	(0.574)	0.058	0.056	0.439	0.442	0.980	0.981
It frightens them	22 (15.0)	80 (54.4)	25 (17.0)	20 (13.6)	2,613.000	(0.755)	1,240.500	(0.106)	0.053	0.051	0.058	0.053	0.271	0.274

Notes:

n = number; % = percentage; U = value of Mann–Whitney test; P = statistical significance; FMH = forensic mental health

Source: Authors’ own creation

Table 3

Participants proposed changes in seclusion practice and environment

					Mann–Whitney U				Kruskal–Wallis/ Monte Carlo		Kruskal–Wallis/ Monte Carlo		Kruskal–Wallis/ Monte Carlo
					Sex		Registered/ unregistered		Age range		Experience of FMH		Experience of seclusion
	Strongly agree n (%)	Agree n (%)	Disagree n (%)	Strongly disagree n (%)	U	(P)	U	(P)	Asymp sig.	Adj. sig.a	Asymp sig.	Adj. sig.a	Asymp sig.	Adj. sig.a
A staff member should always remain with the patient during seclusion	75 (51.0)	55 (37.4)	15 (10.2)	2 (1.4)	2,305.500	(0.101)	1,375.000	(0.391)	0.391	0.413	0.242	0.247	0.056	0.053
Secluded patients should be able to listen to music if they want to	28 (19.0)	69 (46.9)	42 (28.6)	8 (5.4)	2,355.500	(0.167)	1,174.000	(0.051)	0.001	0.007	0.434	0.433	0.188	0.191
It should be a comfortable unlocked room available to patients who would like to use it	19 (12.9)	28 (19.0)	56 (38.1)	44 (29.9)	2,500.000	(0.448)	1,444.500	(0.663)	0.663	0.673	0.343	0.344	0.569	0.592
Seclusion should not be used	10 (6.8)	4 (2.7)	66 (44.9)	67 (45.6)	2,539.500	(0.528)	1,349.000	(0.314)	0.315	0.328	0.714	0.723	0.410	0.445
The room should be more comfortably furnished for the patient	13 (8.8)	37 (25.2)	60 (40.8)	37 (25.2)	2,638.500	(0.845)	1,437.500	(0.634)	0.634	0.644	0.569	0.575	0.764	0.784
The room should be painted to have a calming effect on the patient	27 (18.4)	74 (50.3)	36 (24.5	10 (6.8)	2,553.500	(0.576)	891.500	(<0.000)*	0.058	0.054	0.028	0.027	0.601	0.613
No changes needed to room	26 (17.7)	45 (30.6)	67 (45.6)	9 (6.1)	2,590.500	(0.690)	1,175.500	(0.053)	0.053	0.057	0.096	0.095	0.077	0.076
Reading material should be provided for the secluded patient	27 (18.4)	77 (52.4)	34 (23.1)	9 (6.1)	2,619.000	(0.776)	932.000	(<0.000)*	0.060	0.059	0.563	0.568	0.340	0.353

Notes:

n = number; % = percentage; U = value of Mann–Whitney test; P = statistical significance; FMH = forensic mental health

Source: Authors’ own creation