‘Behind This Wall’ – Experiences of Seclusion on Locked Wards for Women

Aim: To more fully understand the experiences of seclusion in a learning disability service for women. Background: This paper reports on one of the analytical themes, seclusion, from an ethnographic study exploring the lives of learning disabled women on locked wards. Method: Participant observation was used on three locked wards for women in a learning disability secure (forensic) unit in the United Kingdom. Themes from the analysis of field-notes were used to construct an interview schedule. Sixteen detained women and ten staff participated in interviews. Results: The thematic analysis produced four areas of relevance: the seclusion room environment, reasons for using seclusion, termination of seclusion and alternatives to seclusion. Conclusion: Detained women’s descriptions of seclusion portrayed a bewildering, distressing experience which violated their privacy. Alternative practices such as providing a space for anger and resolve, and time for discussion with staff were identified.


Methods and approach
This paper is drawn from an ethnographic research project with learning disabled women in locked wards. Ethnographic researchers are concerned with the meanings people attach to events in their lives (Taylor andBogdan, 1998, Goodley, 2001). By observing people in context during their everyday lives, researchers are able to develop concepts from the data rather than testing preconceived models or theories. A small number of researchers in the learning disability field have used observational methods in their research exploring the experiences of people in locked wards (Johnson, 1998, Hubert and Hollins, 2006, Owen et al., 2008. Other researchers have used ethnographic methods to enable people who are less articulate to participate in research (e.g. Booth and Booth, 1996).
I observed the daily life on three of the women's wards in the unit. Women were a minority in the service; amounting to only 20% of detainees. The wards involved were single sex and contained between 2 and 8 women at any one time, situated in a separate area from the male wards but within the same grounds. The wards that I observed were designated to me by the ward manager for reasons I was not party to. One member of staff and one detained woman declined to be involved in the observation and were not included in any field notes. Two of the wards were classified as low secure (wards are locked but clients are able to access other areas of the unit) and one was part of the medium secure unit (wards are locked and clients must stay within the two-storey enclosure at all times).
Many academics consider ethnography to be 'the only truly effective way of ascertaining the insiders' perspective, gleaned from extended association, casual conversation and everyday observation' (Angrosino, 2004:173). Doing ethnography enabled me to spend time getting to know people, and also talk about my reasons for carrying out the research. It also facilitated a trusting two-way relationship. I was concerned about inequalities in the research relationship as described by researchers such as Stacey (1988), but my aims were to make recommendations which I hoped would improve the lives of women on the unit.

Interviews
My observations produced field-notes, which I arranged into themes and used these to develop interview questions. All participants in the observation phase of the research were invited to be interviewed, and although three detained women and one member of staff declined, sixteen women and ten staff agreed to be involved (seven qualified nursing staff, two unqualified support workers and one clinical psychologist). All interviewees were white British and between the ages of 18 and 60. The questions began as open ended and general, such as, 'How do you feel about living here?' and 'How do you think services for women could be improved?' followed by further questions relevant to my knowledge of the person. Interviews lasted between 30 and 75 minutes, and were mostly single interviews, with some women requesting follow-up interviews. Staff interviews took place in the Research and Development office and interviews with the women on the wards took place in a private neutral room. At the end of the study, I provided feedback in the form of presentations as well as accessible, easy read briefs about the research for the organisation and participants.
The field-notes and interview transcripts were transcribed, and at this point they were anonymised and participants were given pseudonyms. Together, the field-notes and all interviews were analysed using NVIVO, which helped identify subthemes. I wanted to arrange all the data into themes and write about each theme in its entirety rather than looking at data from groups of participants independently. In doing this, I found that some themes were discussed by only one group or in different ways by the two groups, and these details formed part of my findings. I borrowed my analysis method from the phenomenological research tradition (Hycner, 1985) and followed the steps suggested by Hycner in order that the analysis was inductive, arising from the data. Rather than looking for descriptions of experiences, I wanted to explore the meaning that participants gave to their experiences.

Ethical Considerations
The research was given ethical approval from the NHS Local Research Ethics Committee. One of the prerequisites of approval was that I had to gain written consent from all the people involved, to allow me to write about any observation. This was a lengthy and difficult process, but it nevertheless enabled me to discuss the research in-depth with people in the unit. The detained women who participated in the research were all labelled as having mild to moderate learning disabilities and were able to give consent. Trust between myself and the women emerged over time, reinforced by conversations about my role as well as it becoming evident that I was not a member of the staff. Participants were informed that the information that they gave would be anonymous and that their name would not be written down anywhere. Pseudonyms were given during the transcription process and I have made attempts to exclude any information that could identify participants (Rogers andLudhra, 2012, Ellis andBochner, 2000).
Ethical adherence followed the guidance of Perry (2004) who advises that ethical concerns should be paramount throughout the research. Separate consent forms were completed by staff and clients, and for participation in the observation and interviews. Verbal checks were performed with participants to make sure they had understood the process of the research in accordance with the UK Mental Capacity Act (2005). Anyone who disclosed that they had been subject to abuse was referred for counselling if the issue had not been already dealt with.

Results
Identifying the results of the research involved a reflective, engaged, dynamic process that emerged from a detailed analysis of both the ethnographic field notes and the interviews. Seclusion, the most overt strategy for behaviour control, emerged as a theme that both groups discussed. The following themes emerged from a exploration of seclusion experiences: the seclusion room environment, reasons for using seclusion, termination of seclusion, and alternatives to seclusion. Given that this topic is so under-researched, it is important to provide a rich description of the findings of the study. While these issues are undoubtedly interrelated, they constituted sufficiently distinct themes to warrant detailed individual attention.

The seclusion room environment
I did not witness any incidents of physical restraint or seclusion during my fieldwork, possibly because I was a new and 'interesting' presence on the ward, and I did not spend enough time on each ward to become familiar enough to people. When I was on the wards, people often wanted to talk to me, which may have provided a distraction from contentious issues. I did, however, see women who were already in seclusion, through the window and I was shown the seclusion rooms in both areas (all have the same features but one was painted dark blue inside). There is a seclusion room in each area which services two wards: The seclusion 'suite' has a door leading from each flat, a toilet room, and a door leading into the seclusion room. There is a 2 ft by 1 ft window in the seclusion room door and a hatch at waist level which requires two keys to unlock. In each corner of the ceiling, conical mirrors at the back of the room work to show the observer areas not visible from this window. A staff member is required to observe at all times. The only furniture is a bed, which is moulded in with the floor and holds a wipe-clean mattress. Lights and sprinkler on the ceiling are covered with rounded plastic. (Field-notes) Many of the participants in my research talked about the physical aspects of the seclusion room. Here, a qualified staff member discussed the logistics of seclusion, and how difficult it is to seclude a person: From this description it is easy to imagine how difficult it would be to maintain someone's dignity when moving them to the seclusion room. The process of moving a person to this room sounds extremely distressing for everyone, and would clearly attract a lot of attention from anyone in the surrounding area. Some of the women commented on the visibility of this procedure.
Worryingly, in terms of the seclusion room environment, some of the women pointed out that the room was too cold, and others commented that the room was uncomfortable: Researcher: So what did it feel like when you were in seclusion?
Kate (SU 1): It were awful because it's bare walls. There's nothing in there. Nothing in there. They even took the mattress off me because some clients will put the mattress against the door so you can't see in. And they didn't want me doing that so they took the mattress. So I sat on a wooden bench, basically it's harder than [wood] because it's reinforced.
If the surroundings are uncomfortable and the door is locked, it suggests that the act of seclusion may be construed as punishment (Phillips, 2005); this may also be an issue when people feel as though others can see them in the seclusion room. There were many issues about the lack of privacy when secluded:

Kate (SU 1): We can see whoever's in seclusion because we're going through there, and we can see whoever's in seclusion. So it's not fair on that person who's in seclusion.
Clearly, it may have been possible for people to see a person who is in the seclusion room, and the women could not be sure that their privacy was being respected. Here, Sarah described the seclusion room experience: Sarah's experience of being secluded shows that no engagement with staff occurred. Some staff told me that they were discouraged from engaging with women in seclusion, despite this being a private opportunity where they often find the need to talk and explain the reasons for their behaviour.
From the participants' descriptions here, we can see that the process of secluding a person, as well as the uncomfortable conditions (which violate people's privacy), are experienced as unpleasant and undignified by everyone.

Reasons for using seclusion
Staff assumption on the unit was that women arrive at the unit in crisis or extreme distress, displaying behaviours that were considered to be in need of assessment and management. These behaviours may have been figured as causing the breakdown of their previous placement, or having been triggered by the move. The aim, as described by a staff member participant, was to ' contain difficult behaviours' whilst supporting people to 'learn different ways of behaving in situations' (see also Happell and Harrow, 2010). However, as Becker testifies, 'Surely, nothing distances [staff] so thoroughly from service-users as the consideration of their 'management' (Becker, 1997: 140).
The analysis from this ethnography revealed that the control and regulation of women's behaviour was central to the smooth running of the unit. The women were dependent on the staff for many aspects of life, and this dependence allowed the staff to exert control over the women's behaviours and influenced the imbalance of power on the unit; this imbalance was most salient when women were subject to behavioural interventions. Much of the discourse about women when it came to behavioural management, pointed to women as ' difficult' (Williams et al., 2001, Breeze andVRepper, 1998), a reputation which preceded them (Fish, 2015).
Staff told me that seclusion was used when people became unmanageably aggressive and physical restraint was not containing or reducing the problematic behaviour. The UK MHA code of practice advises that: 'Seclusion should be used only as a last resort and for the shortest possible time. Seclusion should not be used as a punishment or a threat, or because of a shortage of staff. It should not form part of a treatment programme.' (Department of Health, 2007: 122-123) Despite these directives, it was clear that seclusion was not being used as a last resort according to some of my interviewees, and the detained women in my study were not correctly informed about the behaviours that could trigger seclusion, bringing about fear and trepidation (e.g. Brown and Tooke, 1992  Sometimes, it seems that seclusion was used to remove a person from a situation of conflict, even when they were not the aggressor. An example of this is Andie's story: A staff member, John, acknowledged that the reasons for seclusion are not always appropriate. Here he suggested that women were sometimes secluded for the sake of others who live on the ward: John (Staff 5): A lot of the time at night they put [name] in at night, a lot, because all the other girls are in bed and she's in seclusion and she probably doesn't need to be in, but they've got to get her out of there for the sake of the other girls.
John suggested that it was not aggression that was the cause for seclusion, but ' disruption' -the seclusion room was being used for another purpose here.
A concerning issue that came out of my research was that some women could be placed in seclusion directly after admittance -possibly as a response to the distress of not being told they were staying at the unit -and staff saw this as a way of assessing the woman's needs. Here, Wendy described why seclusion is used when women arrive at first: Because of the guidance and time limits for using physical restraint, if somebody is not calming down after being restrained for a certain period of time, staff decide to use seclusion as a method to ' calm the person down', or even just to end the restraint. However, there are examples in the literature where people indicate that being placed in seclusion makes them feel more angry (e.g. Chamberlin, 1985) because of the feelings of powerlessness it evokes. Here, Sarah described feeling angrier in seclusion:  Researcher: How did you feel? Why did you cut yourself?

Sarah (SU 7): Because I was angry.
It seemed that seclusion was used to calm people down but this can be counter-therapeutic, or an aversive method of behaviour control. It also fails to sufficiently recognize that anger, aggression, self-harm, or violence may be the result of an oppressive institutional environment. Given the potential shame or stigma associated with the use of seclusion, many of the women were very keen to tell me that they had not been subjected to seclusion recently: Researcher: I have to ask about seclusion as well.

Annie (SU 16): I haven't been in for the last four years though.
Annie goes on to describe being secluded in a previous service for self-harming:

Researcher: OK then can you remember how it feels?
Annie (SU 16): Um, yes.

Researcher: What is it like? Annie (SU 16): … In [names service] one particular incident I do remember very very well, I'd been self-harming all day from the minute I'd got up and it was about half past three in the afternoon. I got put in seclusion with no clothes on in the male ward with males watching me.
Researcher: Nothing to cover yourself up with?

Annie (SU 16): It all depends on what staff it is.
It seems here that institutional context was influential: the decision to seclude was sometimes dependent on the circumstances and what staff member was making it. When the decision has been made, the person was no longer consulted or engaged with by staff and the issue causing the anger was not dealt with or resolved. Detained women not being aware of the criteria for seclusion highlights the power imbalance inherent as staff were in control of this eventuality. All of these issues demonstrate the damage that these coercive methods can do to relationships.

Termination of seclusion
There is no denying that a period of seclusion compels the person to appear calm, although what constitutes ' calm' in this environment could be simply a manifestation of defeat and immobility. Some of the women did acknowledge that it helped them to calm down, but it is impossible to deduce whether it was seclusion that prompted this or just removal from the distressing situation:

Karen (Staff 3): I suppose because you just get to know them, you know the signs and you know the triggers and
you can see in general presentation, physical presentation. They're not anxious any more, they're not red in the face, they're not shouting and screaming, they're very calm, they're talking to you, they've had some medication. There are occasions when we think they're calm and they come out and we have to put them back in again, but not very often. You just get to know the signs really.
Chamberlin concluded that when people appear to calm down after seclusion, it is because they are 'learning to play the game,' by acknowledging therapeutic benefit, and playing the game is widely recognized as the way to secure one's freedom (Chamberlin, 1985: 289). This 'playing the game' could be considered as a reflection of the care vs control dilemma within such services (Burrow, 1991): The basic ethical incongruity associated with seclusion is, on the one hand, the psychiatric ethos of maintaining/increasing personal liberty but on the other hand a dramatic suppressing of freedom. (Morrall and Muir-Cochrane, 2002: 3) Foucault comments on this duality of professed aims that psychiatry attempts to accomplish, the mutually exclusive goals of punishment and rehabilitation (Foucault, 1988). He asks how someone can be treated therapeutically in such an environment of control. Contemporary discourses of ' control', ' choice' and ' autonomy' in learning disability policy illuminate this division and pose the question, what is an ' adaptive' response to such enforced circumstances?

Alternatives to seclusion
Naturally when discussing such an aversive strategy for behaviour control, participants considered their ideas for alternative tactics. Simon, a qualified staff member, suggested that there is really no alternative to seclusion other than prolonged restraint, however he describes this as traumatising: Despite Simon's comment, many of the detained women in my research had ideas about alternative measures that could have been used before seclusion was deployed. Most of the suggestions involved having the opportunity to go to another room away from others to calm down and talk to staff about how they feel: Researcher: OK so can you think of anything that might be better for people other than putting them in the seclusion room?

Bonnie (SU 8): If there's another room away from seclusion, you know like a calm down room, I reckon that they should talk to us and say 'How do you feel, what can we do to help you?' and that.
Like Bonnie, many participants considered that some kind of space where anger in the context of an institutional regime was accepted, would reduce their need for seclusion. Other participants suggested that offering medication or extra support would help: Researcher: How could the staff help you get to a point where you wouldn't need to be in seclusion if you were very angry?

Sarah (SU 7): Give PRN [as required medication], talk to you.
Other women told me that there was nothing that could be done apart from better engagement with staff: Researcher: So if staff have to restrain you or put you in the seclusion room, what can they do to make it easier for clients?

Ruby (SU 3): Well they can't really do anything, really.
Researcher: Nothing would make it better?

Ruby (SU 3): No. Apart from keep talking to you and things like that.
Therapeutic engagement was key when people were suggesting changes to the system. If it is possible for staff to spend time with people before they get to crisis point then this may reduce the need for coercive measures. Staff members also made similar suggestions, for example: Dawn (Staff 9): I think we could do with a HDU [high dependency unit] area as well as seclusion because we have in the past used seclusion to try and take people away from the stimulus because they've been up here [gestures height with hand], but perhaps they've not needed to be secluded at that point. They might have done eventually but if they'd carried on in there. So we could do with a high dependency area where perhaps they don't need secluding yet.
This, again is pointing to a problem with the use of space; if there was another area where people could go to, then seclusion may not be needed. If it is the need for removal of stimuli that causes seclusion, there should be a better way to do this. Here, a senior member of staff talked about finding alternatives when it was decided that a woman was becoming ' dependent' on seclusion: Karen (Staff 3): Yes, there are other ways you can do it, there are other ways of keeping them in a sitting restraint and keeping them safe in the HDU, and challenging that, but it depends on the level of aggression. You know sometimes you can get to a stage where you can sit with them for the majority of the day. And that's fine, keeping them out of seclusion.

Researcher: But that's boring isn't it for everybody?
Karen (Staff 3): Well yes it is but it just stops them getting that dependence of seclusion a lot of the time. Because there's a lot that go into seclusion and then their aggression's gone [claps hands]. Well it doesn't serve a purpose then though does it. Because they're just fighting to get into seclusion and then the fight's gone. So it's about keep working through that fighting and keep -'You're safe, you're here, we're with you, let's sit in the HDU' and it's about that. And it's a long long long process, sometimes. But sometimes it works, sometimes it doesn't.
Dependence on seclusion was discussed by a number of staff, the alternative being prolonged physical restraint without progression to secluding the person. A few staff talked to me about how people can become dependent on seclusion as it is 'the end of the road' or the final method of containment. It is unclear why this may be, as none of the women spoke about this dependency and none of them told me that they wanted to be or liked being secluded. Nevertheless, it is reasonable to deduce that solitude may occasionally be preferable to communal areas at times of distress, for some people.
Within this theme, the participants show that aggression occurs relationally, and that things can be done to reduce the need for seclusion, such as talking to people and acknowledging the reasons for anger and aggression. Both groups of participants agree that more can be done to reduce the need for seclusion.

Discussion
This study shows how aversive, bewildering and distressing women with learning disabilities find the experience of seclusion, yet this method continues to be used to contain people. Further, staff members suffer from the expectation that they should use seclusion as a relevant strategy of treatment.
I did not find any evidence that the use of seclusion was fulfilling any therapeutic gain. This supports some previous research, such as that of Norton and Dolan (1995), who suggest that the institutional response to ' acting out' (such as restraint or seclusion) can perpetuate the behaviour. This is an important conclusion: framing the behaviours in a sociological and institutional context can be useful in moving beyond a focus simply on individuals towards an examination of institutional policies and practices. These institutional responses remove the potential for people to learn from their experiences, because they are serving the needs of society rather than the individuals, by simply containing the behaviour and rendering it invisible. Subsequently, there becomes a stalemate, where the behaviour carries on, producing a complementary victim/victimiser relationship. As a strategy to avoid this, Norton and Dolan (1995), suggest the use of a more democratic therapeutic model where people are involved in decisions about their care. Gentle positive engagement could be considered to effectively remove the need for seclusion. I would add to this the need for more information for everyone about definitive circumstances where institutional responses are likely to be put in place.
My findings are consistent with some previous oral history research describing the horror and the feelings of abandonment that the 'time out room' provoked for learning disabled people (Malacrida, 2005). People should be given clear and accessible information about the reasons for their seclusion, to reduce the feelings of anxiety and bewilderment so evident in my participants' descriptions. The study also demonstrates the violation of privacy associated with seclusion, an intrusion that marks out the ableist power disparities between staff and women with learning disabilities. When taking into account women's histories and the levels of abuse they have encountered, the worrying potential for retraumatisation is prominent Halstead, 2001, Fish andCulshaw, 2005).
The lack of any sort of debrief or explanation after incidents of seclusion was highlighted by the participants. It was clear from the interviews and also ethnographic observations that this was a significant problem. As highlighted above, people who are uncertain of the reasons for seclusion may inadvertently repeat the behaviour that is deemed so problematic by staff. Debriefing may assist people in coping with their own experiences of distress. Providing an opportunity to talk about concerns, responses, and conflicts may in some occasions be therapeutic, if managed appropriately.
Staff described terminating seclusion when they felt the secluded person was ' calm'. People who reported feeling calm during seclusion often simultaneously described themselves as angry and depressed. Evidently, being ' calm' does not necessarily imply a sense of well-being, it can be a manifestation of immobilization or shutting down. If the use of seclusion can be perceived as punishment or even torture (Veltkamp et al., 2008) then people are likely to act as though they have calmed down in order to be let out.
Staff were sometimes told not to engage with secluded women, this was an area where therapeutic relationships were maligned and discouraged rationalised by the suggestion that making seclusion into a positive experience would encourage women's reliance on this coercive method. I argue that coercive methods for behavior management should be investigated using concepts of power and control (Griffith et al., 2013).
This study describes the unpleasantness of coercive methods for everybody, emphasizing the damage it can do to the therapeutic relationship. There were many opportunities being missed for discussion and resolution of conflict. In Brown and Tooke's (1992) study of psychiatric patients, nurses were asked who benefits from seclusion. They replied that other patients benefitted most, then the secluded patient, hospital and lastly the staff. Patients on the other hand, claimed that the staff benefitted most, then other patients, hospital and last of all the secluded patient. This admission by both groups that the secluded person is not the primary beneficiary raises questions as to the ethics of seclusion, and the rights of one person relative to those of others. However, conflict is a real issue on the wards and sometimes the removal of one of the women is the only thing to do. In light of my participants' recommendations, I argue that this may be done via methods other than seclusion, such as providing a separate space for discussion and resolve, where anger can be articulated. Other studies indicate ways that seclusion could be reduced or eliminated, such as higher staff ratios and supporting staff to avoid burnout (Happell and Koehn, 2011) or by involving patients in seclusion reduction strategies (Gaskin et al., 2013), demonstrating that the use of seclusion is not inevitable and can depend upon staff attitudes and culture, rather than predominantly service-user characteristics.
A particular concern that this article uncovers is that occasionally, distressed women are secluded on admission. This reflects how disturbing the experience of transition from other services can be, and furthermore is at risk of violating the human rights of the service-users (Sullivan and Mullen, 2012, Fyson and Kitson, 2010, Lawson, 2004, Hendricks, 2007. On admission, women are sometimes told by staff at their previous placement that they are merely visiting the service which causes further distress when they realise they are detained and must stay (Fish, 2015). An alternative to seclusion on admission must be found, such as introducing more therapeutic and gentle admission procedures.
As detained women are often not consulted about what happens to them or informed about the reasons for the use of seclusion, it is crucial to consider how power relationships are shaped in such services as this. It is important for services to recognise people as human subjects, and to acknowledge the ways they give meaning to their lives and to specific experiences. Problematic behaviour is relational rather than individual, and although the use of seclusion (and segregation) has been rationalised as a treatment strategy throughout the past, the question for services to ask is how to imagine other ways of working.

Conclusion
Although it is important not to overestimate the degree to which learning disabled people are dissimilar from others who use services, it is equally important to observe the particularities of those services designated for this group, who have policies constructed on the basis of their impairments. This article demonstrates how learning disabled women are perceived and treated as in need of controlling, in terms of their movements, the information they are given, and the choices they can make. This controlling treatment often begins in early life, and people adapt ways of being that fulfil their needs in the short term, but may trigger further methods of coercion.
Although this research was confined to one unit, many of the women had lived in multiple inpatient services and were describing incidents of seclusion from their pasts. Despite changing regulations of seclusion practices, the potential for retraumatisation and feelings of punishment still remains. Future research should focus on ways to include service users in seclusion policies and procedure, with the goal of finding progressive ways to avoid and deal with anger within forensic services. Further research on a much wider scale is also necessary to explore the fundamental issues raised by feminist disability studies, so that disabled women's experiences are not automatically assumed to mirror those of disabled men, and are recognised as worthy of greater attention in their own right.

Funding Information
This work was supported by the Economic and Social Research Council [grant number ES/H037594/1].