‘They don't understand…you cut yourself in order to live.’ Interpretative repertoires jointly constructing interactions between adult women who self-harm and professional caregivers.

Abstract

The aim of the study was to illuminate interpretative repertoires that jointly construct the interaction between adult women who self-harm and professional caregivers in psychiatric inpatient care. Participant observations and informal interviews were conducted among six women who self-harm and their professional caregivers in two psychiatric inpatient wards, and analysed using the concept of interpretative repertoires from the discipline of discursive psychology. The analysis revealed four interpretative repertoires that jointly constructed the interaction. The professional caregivers used a “fostering repertoire” and a “supportive repertoire” and the women who self-harmed used a “victim repertoire” and an “expert repertoire.” The women and the caregivers were positioned and positioned themselves and people around them within and among these interpretative repertoires to make sense of their experiences of the interaction. It was necessary to consider each woman's own life chances and knowledge about herself and her needs. The participants made it clear that it was essential for them to be met with respect as individuals. Professional caregivers need to work in partnership with individuals who self-harm—experts by profession collaborating with experts by experience. Caregivers need to look beyond behavioural symptoms and recognise each individual's possibilities for agency.

This paper focuses on the jointly constructed interaction between adult women who self-harm and their professional caregivers based on participant observations and informal interviews at two acute psychiatric inpatient units.

Research on self-harm reports multiple meanings determined by history, culture, and tradition (McAllister, 2003), and supports the view that self-harm is a socially constructed concept.

Beliefs, attitudes, practices, and images diffuse across latitudes and longitudes and centuries. Our perceptions of self-mutilation as grotesque or beautiful, heroic or cowardly, awesome or pitiful, meaningful or senseless, derive in great part from the perceptions of those who have lived before us. (Favazza, 1996, p. 4)

Reflecting on some of the negative connotations associated with self-harm, such as “failed suicide” (a “botched” attempt), Allen (2007) discussed the term “deliberate” as a prefix to self-harm and argued that this was not only redundant, but also conveyed the self-harming act as something a person could refrain from doing through an act of will. Healthcare professionals in particular should be mindful of the words used to convey and report a client's motivations and intentions (Allen, 2007). A study by Johansson (2010) reported how meaning was produced in relation to self-harm and how this production of meaning also involved particular constructions of self-harmer identity. The self-harmer identity was associated with “normality” and “abnormality.” Self-harm was identified as “normal” predominantly within a cultural or social level (i.e., “that's how they are”) within a certain group of people, such as young people with mental or emotional problems or so-called emos. However, self-harm as “abnormal” behaviour appeared predominantly on an individual level in psychiatric discourse. The reasons for self-harm in that context are ascribed to individual emotional problems and sickness, and the recommended solution is for the persons to change themselves in some way, not for something to be changed in the individuals’ lives.

In this study we define self-harm according to Favazza's (1996) concept of self-mutilation as a repeated, impulsive behaviour where tissue damage occurs, yet it is not a suicide attempt. Instead of wishing to die, the person who self-harms wishes to be relieved from anxiety. Self-harm describes a wide range of behaviours including self-poisoning, scratching, cutting, burning, and hair-pulling (Brophy, 2006). People inflict physical harm on themselves in an effort to make their mental suffering easier to endure (Favazza, 1996).

The prevalence of self-harm is estimated to between 2 and 6% in a general population (Hawton, Rodham, Evans, & Weatherall, 2002; Klonsky, Oltmanns, & Turkheimer, 2003), and clinical signs point at self-harm as an increasing phenomenon predominantly among adolescents and young adults (Sommerfeldt & Skårderud, 2009). There are contradictory opinions whether or not there are gender differences according to sex in the prevalence of self-harm. Some authors report no differences (Klonsky et al., 2003; Marchetto, 2006), while others report that self-harm are more common among women (Landstedt & Gillander Gådin, 2011; Madge et al., 2008).

A recurring theme in recent international research is healthcare professionals’ lack of confidence in caring for persons who self-harm (Friedman et al., 2006; Gibb, Beautrais, & Surgenor, 2010; McCann, Clark, McConnachie, & Harvey, 2006). The literature reports predominantly negative attitudes towards patients who self-harm (McHale & Felton, 2010). Healthcare professionals describe them as manipulative and attention-seeking. They feel anxious about how to speak to them and how to care for them, and feelings of frustration, anger, and helplessness are frequently reported (Anderson, Standen, & Noon, 2003; Friedman et al., 2006; Hopkins, 2002). Feelings of burden (Wilstrand, Lindgren, Gilje, & Olofsson, 2007) and the need to be constantly on one's guard are also described (Thompson, Powis, & Carradice, 2008; Wilstrand et al., 2007).

Several studies illuminate experiences of care as narrated by persons who self-harm (Harris, 2000; Lindgren, Wilstrand, Gilje, & Olofsson, 2004; McAndrew & Warne, 2005; Smith, 2002; Warm, Murray, & Fox, 2002). Bywaters and Rolfe (2002) described experiences of care among persons who self-harm in a report from NCH, a voluntary organisation in the United Kingdom. In the results, persons who self-harmed felt that caregivers thought that taking care of self-harming persons was a waste of time and that they were “hopeless cases” who were going to self-harm repeatedly. These findings are echoed by Lindgren et al. (2004) and Ribe (2009). Smith (2002) interviewed people who self-harmed and noted that when an opportunity for conversation was offered by caregivers it was experienced as helpful. Nevertheless, the interviewees felt that caregivers made themselves unreachable and were more likely to provide pharmaceutical help than conversation. In a study by Lindgren et al. (2004), persons who self-harmed reported experiences of being objectified and treated as “things” or “machines” without a soul. They pointed out the importance of being viewed as human beings with assets, not only as humans with difficulties.

In summary, the literature reports predominantly negative attitudes towards patients who self-harm (McHale & Felton, 2010), and the interaction between people who self-harm and their caregivers is described as challenging and characterised by complicated feelings on both parts. There is a need to take action in relation to self-harm, focusing on the involved persons and the interaction between people who self-harm and professional caregivers, especially in psychiatric inpatient care. Details about what happens when professional caregivers and patients who self-harm meet and how this influences the interaction and the patients’ care and well-being has, to our knowledge, not earlier been described. Studying how interactions are socially constructed and framed should illuminate the beliefs and norms that influence the boundaries for possible and relevant solutions for treating people who self-harm. Therefore, the aim of this study was to explore the interpretative repertoires that jointly construct the interaction between adult women who self-harm and professional caregivers in psychiatric inpatient care.

Theoretical framework

Our theoretical framework is drawn from discursive psychology and the concepts of interpretative repertoires, negotiations (Edley, 2001; Potter & Wetherell, 2001; Taylor, 2001a, b; Wetherell, 2001a, b), and social constructionism (Burr, 2003). Language has an important role in the social construction of identities, relations, and systems of knowledge. “Identity,” as defined within social constructionism, avoids the psychological concept of personality and is an implicitly social concept. The key issue is that we, as humans, are “identifying objects” as well as human identities, and the identity we confer upon another has more to do with our own purposes than with the actual nature of the person or thing identified. The binary language we use to make up human identities, such as masculine/feminine, sane/insane, heterosexual/homosexual, and so on, is based on socially constructed categories rather than on the essences of individual people. Our knowledge and how we perceive and represent our view of the world is historical, culture-specific, and contingent. We apprehend the world and create knowledge through social interactions where we develop norms about “mutual truth” and argue about what is true or untrue. Words are important because our language constitutes our social world, constructed identities, social relationships, and our values and beliefs (Burr, 2003). Interpretative repertoires are relatively coherent ways of talking about objects and events in the world. The languages used in social conversations are usually made up of a patchwork of quotations from various interpretative repertoires and they can be both flexible and contradictory. Discursive psychologists examine how people use language to construct their lived experience of the world and their own identities through social processes in daily life, where people negotiate between available identities and use available social structures of interpretative repertoires (Edley, 2001; Potter & Wetherell, 2001; Taylor, 2001a).

Methods

Study design

We chose focused ethnography (Roper & Shapira, 2000), which deals with a distinct problem within a specific context, for this study. This method allows the researcher to observe and interview individuals or groups within their specific context (Roper & Shapira, 2000). In this study the first author conducted participant observations and informal interviews over a period of 6 months.

Context

Persons who self-harm are cared for in various settings, as both outpatients and inpatients. The present study was conducted in two locked general acute psychiatric inpatient wards with admissions 24 h a day in two different clinics in Sweden. Patients could be cared for voluntarily as well as involuntarily, in line with the Compulsory Psychiatric Care Act (SFS 1991:1128). Voluntary patients were free to leave whenever they wanted, unless caregivers deemed their mental condition too fragile for them to live without care. Adult patients with all kinds of mental illness were admitted to the wards. One of the wards had 16 beds and the other had 13 beds, with a few single rooms but mostly double rooms. When the wards were overcrowded, patients could be assigned a bed in the corridor, in a store-room, or in a meeting room. During the study period, the wards were overcrowded by two to four patients. In both wards there were both common and separate areas for patients and staff.

In general the wards were similar regarding locked doors, routines concerning times for food, access to smoking, possibilities to take a walk, and so on. However, some differences, such as the architecture were apparent. One ward (A) was built in a traditional style as a long corridor with rooms at both sides, and the dining room and common areas at one end of the corridor. The other ward (B) was built with a dining area and common areas in the centre and three annexes joined to the centre with rooms at both sides of a short corridor. Another difference concerned the food. At ward A the patients served themselves every meal in a special dining room, while at ward B the patients served themselves breakfast from a serving trolley in the dining area, and lunch and dinner were served by the professional caregivers. Furthermore there were differences regarding the patients’ access to their mobile phone, computer, and cigarettes. Ward A allowed the patients to have their own mobile phone, computer (except the wire for uploading), and cigarettes in the ward. In ward B everything was locked in and the patients had to ask caregivers for access.

The wards were staffed by registered nurses, some with and some without specialist training in psychiatric nursing, enrolled nurses in mental health, a ward manager (registered mental health nurse/registered nurse), and psychiatrists whose time was divided between the wards and other units within the psychiatric clinics. There was an informal system of “key workers” (cf. Svedberg & Lützen, 2001) with certain responsibilities for specific patients. These responsibilities could include care planning or being the patient's spokesperson during rounds or for permissions to go outside the ward, etc.

Participants with self-harming behaviour

Two registered psychiatric nurses at each ward were responsible for informing the patients, men, or women with a history of self-harm, about the purpose of the study and requesting their participation. The information was given both orally and in writing. During the observation period, no men were treated for self-harming behaviour. The participants were six women, three at each ward, who were admitted to inpatient psychiatric care because of their self-harm. The women's self-harming behaviour began several years ago and they had been admitted to different kinds of care for between 4 and 17 years (median 6 years).

The self-harming women were between 21 and 37 years of age (median 23.5 years). Three of the women were admitted to involuntarily care, in line with the Compulsory Psychiatric Care Act (SFS 1991:1128) and three were admitted to voluntary care. However, two of the women who were admitted to voluntary care had been informed that if they wanted to discharge themselves from the ward, the psychiatrist would convert them into involuntarily care. The women self-reported their diagnosis, as well as their current medication and treatment. One woman had no diagnosis and the other five women had one to three diagnoses including personality disorder, eating disorder, anxiety syndrome, psychosis, bipolar disorder, depression, and Asperger syndrome. The women reported that they were medicated with antidepressants, clozapine, benzodiazepines, hypnotics, painkillers, and acid-reducing medicines, and one of the women was treated with electroconvulsive therapy. Several nursing strategies to prevent self-harm were used in the care of these women, and the primarily goal of these interventions was to end the self-harming behaviour. Prevention strategies included the removal of objects that could be used as a ligature, distraction, and threats of negative consequences for self-harm. Another prevention strategy was the use of special observations at a different level. During the period of observation, three of the women occasionally were under special observation due to their self-harming behaviour. As inpatients, all of the women had unplanned conversations with the professional caregivers, including their key worker, at the ward. Four of the women had an ongoing contact in outpatient care (e.g., conversational therapy or physiotherapy); however, the outpatient treatment was either less intensive or suspended for the duration of inpatient treatment.

Empirical material

The empirical material in this study consists of participant observations and informal interviews. The first author, an experienced psychiatric nurse with no connection to the settings, performed all observations and informal interviews and transcribed the material verbatim.

Participant observation

Participant observations were used to describe the interaction between patients who self-harm and their professional caregivers (Patton, 2002; Roper & Shapira, 2000). Participant observations offer the opportunity to share certain experiences and have been used in studies conducted in psychiatric care (Bray, 1999; Johansson, Skärsäter, & Danielsson, 2007; Schoppman, Schröck, Schnepp, & Büscher, 2007). A researcher can take various roles depending on the focus of the observation. In this study the researcher was a participant, yet remained passive as a so-called observer-as-participant (Roper & Shapira, 2000). The observed women who self-harmed and their professional caregivers were informed about the purpose of the observations. The researcher was visible to everyone in the ward, but not involved in the care of the women who self-harmed or in a working relationship with the caregivers.

The observations took place during 3 months in 2009 and 3 months in 2010, Monday to Friday, from 6 a.m. until 10 p.m. A total of approximately 150 h of descriptive observations were performed with about 40 h of focused observations including informal interviews. The observations were performed in common and staff areas and occasionally in the woman's room or in a parlour. Initially, there was a descriptive open approach to observe actions related to the healthcare environment and to the routines of the wards. These descriptive observations were taken down as field notes and served as a context in the analysis. The focused observations were on situations where the women who self-harmed and their professional caregivers interacted with each other. Observations of a situation where the participants had a special meeting were, after approval, audio-taped.

Informal interviews

Informal interviews following some of the focused observations were not prearranged. By asking professional caregivers and women who self-harmed, individually, to reflect on their experiences of the situation observed, the researcher was able to broaden the understanding of the observations. For example, the researcher asked the participants what they experienced in the situation and what their feelings and thoughts were about their interactions in the situation (cf. Roper & Shapira, 2000).

Analysis

In this study the focused observations were analysed (with the descriptive observation as context) together with the informal interviews using the concept of interpretative repertoires from discursive psychology (Edley, 2001; Potter & Wetherell, 2001; Taylor 2001a, b; Wetherell, 2001a, b). Identities are seen as a result of constantly ongoing historical, cultural, social, and situational negotiations. They are connected to political, economical, and symbolic power and status. Some repertoires are culturally dominant/hegemonic and therefore more accessible and taken for the “truth” (Edley, 2001; Potter & Wetherell, 2001) and other repertoires are marginalised (Burr, 2003).

The formal process of analysis started with the verbatim transcriptions of the field notes and recordings from the observations and informal interviews. Then the first author read through the texts several times to get a feeling from the overall material of how interpretative repertoires were used by the women and by the professional caregivers (cf. Potter & Wetherell, 2001; Taylor, 2001a; Wetherell, 2001a, b). The next step was to read the focused observations including the connected informal interviews and mark them with codes, for example, “control,” “power struggle,” “solidarity,” and so on. After reading and coding the entire material, the first and the last author discussed the analysis so far and decided, after a discussion with the second author, to write an individual summary on each woman who self-harmed. This summary described the interpretative repertoires the women were engaged in. In order to deepen the understanding, the summaries were read through many times as new patterns were discovered (cf. Öster, Magnusson, Egberg Thyme, Lindh, & Åström, 2007; Potter & Wetherell, 2001; Taylor, 2001a; Wetherell, 2001a, b). Thereafter, the first author read through the focused observation material once again and wrote down page references from the texts where a certain repertoire was described. The other authors continued to evaluate and validate the ongoing analysis.

During the data collection, the first author's thoughts and reflections concerning observed interactions were written down in a diary. These diary notes were not used in the analysis, but they functioned as a reminder for the first author for recalling certain events that took place during an observation.

Ethical considerations

Patients in psychiatric settings are vulnerable, as they are dependent on healthcare professionals. Several ethical considerations were taken into account in this study. Whether it is ethically defendable to observe and interview people with mental illness is open for debate. Furthermore, there is a risk that they may disclose more than they really want to. Another concern is whether observations and interviews may be experienced as a violation of integrity or privacy among the involved patients and professional caregivers. The observer regards delicate situations and asks questions that are personal and emotional. On the other hand, the experience of having someone who is genuinely interested in their lives can be experienced as a relief by people with mental or emotional difficulties (Gaydos, 2005).

The participants, both the women who self-harmed and the professional caregivers, were informed about the voluntary nature of their participation. Both patients and caregivers could end their participation whenever they liked without giving any reason and without consequences for themselves, the woman's treatment, or the caregiver's work. Both verbal and written information were given to the participants, and they were asked to give their informed consent to participate. In focused observations where, for example, the participants went to a special room, the researcher again asked for consent for those specific instances. On one occasion one of the women did not want the researcher to attend the meeting. During all observations, the first author wore a badge with her name and workplace identification.

The authors are experienced in working with people who suffer from mental illness. The first author observed the participants’ reactions during the interviews and observations. Had the participants seemed to be in pain or discomfort, the data collection would have ended; however, this was never needed. The chief physicians of the psychiatric clinics and The Regional Ethical Review Board in Umeå, Sweden, approved the study (Dnr 08-034M and Dnr 2010-73-32).

Findings

The interpretative repertoires predominant when jointly constructing the interaction between adult women who self-harm and professional caregivers in psychiatric inpatient care are presented in these results. For the caregivers, the dominating repertoires were a “fostering repertoire” and a “supportive repertoire,” and for the women who self-harmed the dominating repertoires were a “victim repertoire” and an “expert repertoire.”

The women and the professional caregivers were positioned and positioned themselves and people around them within these repertoires. They took on an “equal” position, an underdog position, or an authoritative position. Although the repertoires are presented separately, they were not distinct but interwoven with each other and flexible. The repertoires will be illustrated with quotations, with pauses are marked with … and fictitious names used for the participants to protect their anonymity.

A fostering repertoire

The professional caregivers within this repertoire took on authoritative positions as fosterers. The interactions were characterised by a struggle about who has the preferential right of interpretation in various situations. Consequences of this struggle were shown in uncertainty, as well as in the restricted ability to act. Using a fostering repertoire meant positioning oneself as being the one who carries the knowledge about what is right or wrong, good or bad, acceptable or non-acceptable behaviour, normal or abnormal, etc. Within this repertoire the caregivers described trying to “raise” the women by setting demands and limits. A common situation was when a woman wanted to take a walk or go home so she could take care of mails, bills, etc. By setting rules for the women about what they had to achieve to earn the right to leave the ward, the professional caregivers communicated and gave prominence to their authoritative position. The caregivers argued that they wanted to provide help and care for the women, but they also felt they had a mandate to decide what the help should be. One of the professional caregivers described a situation when the woman and the caregiver had different opinions about care provided, and she put the blame on the woman while positioning herself as having the preferential right of interpretation:

It's hard to get into Anna's life. She has … it's difficult to … she lets no one near her. She wants help, but on her own terms. Anna is like a teenager and she finds it hard to absorb what you say. That's the way it is!

As fosterers, the professional caregivers often took on the authoritative position of experts even if their interpretations totally disagreed with what the women themselves said, as exemplified by the following interaction:

Doctor:	No one here has as a goal for you to stay here only for being here … not in any way … it isn't a goal in itself … every one of us want you to feel good enough to …
Ella interrupts:	But that doesn't work … it feels like … I have tried to explain [that they are unable to help her] since I got here, that it doesn't work … you don't get that. I don't understand … what's the problem?
Doctor:	mmhmm …
Ella:	I needed help when I got here … but not now …
Doctor:	Did you? Did you feel that yourself?
Ella:	mmhmm …
[Silence 6 seconds]
Doctor:	As I see the problem right now … it's unsteady … it's really unsteady.
Ella:	No, it's not.
Doctor:	Yes, I think so …

A supportive repertoire

The professional caregivers predominantly took on an “equal” position within this repertoire; however, sometimes they took on an underdog position together with the women. An “equal” interaction included a supportive atmosphere, positive feedback, and a concern for people around them. By providing support and giving positive feedback, these caregivers worked to empower the women they cared for. As one of the caregivers said in a conversation with one of the women:

Key-worker:

I know that what you say is the truth, Ella … You don't want to be in the ward and you don't get any better. You have found things to do that work for you … you have studied, you have had practice 2 days a week, and you have felt better in doing these things and it's worked out very well.

The professional caregivers acted predominantly as facilitators in interactions with the women in this repertoire. This was shown by being thoughtful, caring, willing to help, etc. Acting as a facilitator was also shown when the caregivers tried to demonstrate their understanding of the women's scars. The following quotation derives from a situation when a woman struggled with accepting her scars:

Fia:	But … I have realised that I … I have to … I think like this … I dress for the weather … not for wanting to hide something for another person …
Nurse:	Exactly …
Fia:	If it isn't … well … that you recently have done anything [e.g., recently cut herself] … but when it's only scars then … I know I have to live with the scars the rest of my life … and even if there is a chance to fix them, it will still be obvious that it's not unharmed skin….// [talks about how the scars look like after a plastic surgery]
Nurse:	hmm … though the scars are your war wounds … aren't they? … if you see it in that way …
Fia:	Yes …

Caregivers also acted as facilitators when they took on positions as women's spokespersons in situations when the women did not feel comfortable enough to speak for themselves. They listened and showed an interest in the women and a willingness to help them. The caregivers facilitated the women's ability to express their needs, listened, and took women's concerns under consideration. When possible, caregivers also tried to satisfy the woman's needs, sometimes even against a rule on the ward. The following conversation derives from a situation between a caregiver and a woman, Fia, when they talk about the caregiver who had acted against the ward's rules and routines:

Nurse:	I felt when I left from here … I fought a bit with myself when I left … It felt like … okay … because we had an agreement within the caring team that I should search through your things … It was me and X within the caring team who should search your things …
Fia:	hmm …
Nurse:	But I felt that … if I want someone to believe in me … and give me a chance then … I want the person to trust me … and then when I left here I thought … should I have searched through, Fia? I said to you that it felt very strange and … if it had been me … I would have found a search very difficult to accept …
Fia:	hmm …
Nurse:	And that's why … I left from here and thought … my, my, my … how will this work out … so it felt so good when X told me …
Fia:	hmm …
Nurse:	… that you, Fia, had left all your things [razors etc.] in, in the evening …

As mentioned, sometimes the professional caregivers took on the position of underdogs together with the women. Situations when this occurred were characterised by interactions negotiating consequences of rules and routines affecting the participant's ability to act. Both the women and the caregivers waited for a physician and for decisions concerning care, etc. The following conversation reflects a situation of a care provider and a woman negotiating the woman's ability to live in her own apartment supported by personnel from social services. Both the caregiver and the woman, Fia, positioned themselves as underdogs with restricted agency within the organisation:

Nurse:	Then it's this thing with your housing and the way it's turned out around that … I think … the housing [the service personnel within the housing] hasn't handled it so well … it has been a little … how should I say … It wasn't the right time to call and say that just now [the service personnel have said that Fia is not welcome back to her apartment due to her self-harming behaviour] … I think that …
Fia:	hmm …
Nurse:	They could have waited a bit …
Fia:	Yes, exactly … I think so, too … they could have … they could have waited and …
Nurse:	Actually …
Fia:	And … and …
Nurse:	And if they think so then, then … they have to manage it at an another level … not now when you struggle so much … I think so …
Fia:	Yes, because it went even harder and …
Nurse:	hmm …
Fia:	I am feeling insecure and it generates a lot of anxiety …
Nurse:	Exactly … and it has been a very peculiar … but now it's on the chief's and the doctor's table … what the actual situation looks like and how the service personnel have handled this …

For the professional caregivers, the underdog position meant being restricted in how to interact with these women. As described in the context of the study, care was organised with “key workers” who had certain responsibilities regarding specific patients. There was an implicit rule that the key workers were responsible for planning permissions and various other arrangements for their patients. A consequence of this rule was that other caregivers were restricted in both the support they could provide and the competences they could use in daily interactions with women for whom they were not key workers. The following quotation derives from a situation between a woman, Cissi, and a caregiver who was not her key worker. Cissi wanted to talk to one of the caregivers because she felt anxious, so the caregiver sat down with her. In an informal interview with the first author the caregiver said:

It's hard to talk with Cissi, because I'm not her key worker. Cissi said that she wanted to talk and so I sat down with her, and usually I am … if I am the key worker for the patient, then I am more active, I ask more questions. But I think that is hard when I am not the key worker.

Cissi described her dissatisfaction with the same conversation this way:

The conversation was useless. I want to have suggestions on things to do better. I know XX more than anyone else on duty just now, but there is a difference when I cannot talk with my key workers. It's [anxiety] like a time bomb that is ticking and in the end it explodes.

A victim repertoire

The women took on an underdog position in this repertoire, which meant being restricted, from living behind locked doors to not having access to needed support from the caregivers. A common consequence for the women was waiting—waiting for medication, waiting for the allowed time to smoke, and waiting for meals. Despite their experience of restriction, the women understood the necessity of certain rules on the ward, for example, rules about smoking. As Anna said:

I understand that the lighters have to be locked up. When I first came to the ward, I thought it was ridiculous, but now I have a certain amount of understanding for them being locked up. Because sometimes, when some patients have anxiety hysteria attacks, then they could use the lighters to set up fires, small fires with those lighters, if they were allowed to have them. So I do understand that the lighters have to be locked up.

The women described how they followed the rules and routines without questioning them and how they blamed themselves if anything went wrong. The women talked about feeling resigned, standing back, trying to please by not being in the way, and doing as the professional caregivers wanted them to do. A typical situation showing the interaction within this repertoire is the following:

Nurse	Hi, you wanted to see me, can we sit down?
Donna:	I have a real urge [to cut herself] … How was the round? Am I going to talk with the doctor?
Nurse:	If everything ends up well during the weekend, then you can talk to the doctor.
Donna:	What will end up well? If I have this big urge, should I talk about that or should I be quiet? What is it that is right?

An expert repertoire

The women took on an authoritative position within this repertoire and the interactions with the professional caregivers within this repertoire were characterised by a struggle for the preferential right of interpretation. The obvious disagreement concerned whether or not the self-harming act is an acceptable behaviour to reduce mental suffering. As Cissi expressed it:

They don't understand. You don't cut yourself because you want to die; you cut yourself in order to live.

Common attitudes in situations when the women took on an authoritative position and acted as experts, were questioning, distrust, rejection, and defiance of the professional caregivers. The interactions often concerned smoking, being allowed to go out, and what was permitted during time out of the institution. The women argued that the caregivers were unreliable and the rules in the ward were set up as important for the patients to follow but not for the caregivers, who could do what they wanted even if there were consequences for the patients. The following conversation is an example of one woman's distrust towards a physician, other staff at the ward, and the rules set up:

Doctor:	And now yesterday … you didn't come back … despite our deal.
Ella:	But nothing has happened … It's just because I don't follow your rules. You don't even follow your own rules.
Doctor:	Which one do you mean?
[Silence 3 seconds]
Ella:	How am I going to know if I can go to my practice … if I come back … I don't know that … the only thing I know is that I can go to my practice if I don't go back.
Doctor:	Our rules yesterday were pretty clear I think.
Ella:	No.
Doctor:	You met Doctor X who said …
Ella interrupts:	But I don't trust you …
Doctor:	We said 2 hours on leave … everything … you follow it and get back in time …
Ella:	But I can't know for certain anyway …
Doctor:	Because?
Ella:	Because you can change your mind whenever you want.
Doctor:	What are usually the reasons for us to change?
[Silence 5 seconds]
Ella:	I don't know …

In some situations when the women were questioned by the caregivers they reacted with rejection. The interaction was dominated by a verbal “fight” where both parts tried to argue for their preferential right of interpretation. The following describes a situation when there was a verbal agreement between the key worker and the woman Bella that she could spend the afternoon downtown with a friend:

Nurse:	What are you going to do outside this afternoon?
Bella:	Look at furniture for my new apartment with my best friend.
Nurse:	X [the key-worker] has written …
Bella interrupts:	Yes, what has X written?
Nurse:	Not so much. X has written that you're allowed to be out between meals and we have interpreted that the 2 p.m. coffee counts as a meal. This means that you have to eat lunch here and then be back here for the 2 p.m. coffee.
Bella:	But yesterday, when I was outside with XX, I was allowed to be out until 4.30 p.m.
Nurse:	Yes, but that isn't documented. Instead it's written that you have to eat your meals here and then it's written that you are going out with XX.
Bella:	No, I am going out with my best friend.
Nurse:	Yes, but you have to eat breakfast, lunch, and dinner here and be back for the 2 p.m. coffee.
Bella:	But I want to be outside until 4.30. I don't eat anything at the 2 o'clock coffee.
Nurse:	Usually that's the way it is, you eat here and then you can be outside between the meals, and your key worker isn't here today.

Sometimes, the women wanted to take matters into their own hands and provided themselves with what they needed without informing the caregivers. One of the women, Donna, described that she did not get a prescription for a salve that she knew by experience could cure her eczema. She got an ointment instead and then she described, in an informal interview with the first author, what she really wanted to do to find a solution so she could help herself:

Donna:	So, you will take this tube of ointment … [hand-salve] … bring it home … empty it out … and fill it up with the right ointment [hydrocortisone] … then you have hand-salve! But now I don't do that … but I ought to do it in order to get the right treatment. Because I know, I have had eczema since I was a kid so I know my body quite well and I know what works.
First author:	But can't you ask for the right ointment then?
Donna:	I have done that, but first they gave me another salve and it didn't work at all … then I got hand-salve … it was a little bit better but not really very good, so now I have to bring it up with them again and say [claps her hands together] now you have to do something about it!

The women struggled to take on an equal position in the interactions with the professional caregivers in this repertoire. When the interactions were more equal, there was a sense of community between the women and the professional caregivers. Commonly occurring activities such as party games, conversations about common interests, mutual expressions of concern about one another, giving compliments, and so on involved the participants. Both the women and the caregivers expressed the importance of having someone to connect to (e.g., family and friends) and having something to do.

Discussion

The aim of this study was to illuminate the interpretative repertoires that jointly construct the interaction between adult women who self-harm and professional caregivers in psychiatric inpatient care. The findings will be discussed from the basis of Table I, starting with the interaction in the upper left square.

Table. I.

Overview over interpretative repertoires jointly constructing the interaction.

		Women who self-harm
		Victim	Expert
Professional caregivers	Fostering	Fostering and victim	Fostering and expert
	Supportive	Supportive and victim	Supportive and expert

The interaction between a “fostering repertoire” and a “victim repertoire” was characterised by the authoritative position taken on by the professional caregivers and the underdog position taken on by the women. The women were unsure about almost everything and relied upon the caregivers’ preferential rights of interpretation. Caregivers’ authoritative position strengthened the women's underdog position and facilitated the women's taking cover behind rules and restrictions. According to Alexander (2006), patients’ experiences of rules in acute psychiatric wards lead to feelings of being victimised, dehumanised, powerless, humiliated, isolated, and rejected. The rules were embedded in the ward structure and therefore accepted. However, patients’ passive reaction towards the regime concealed feelings of fear, stigma, distrust, and abandonment. Deficient clarity and consistency of the rules led to strict rule enforcement and neglect of patients’ psychological needs (Alexander, 2006). Similar findings are reported in the present study and they echo aspects of Erving Goffman's work (1968) in that patients’ did not seem to be kept fully informed about the rules connected with the admission process. Goffman described that patients were not allowed to spontaneously express their own definition of the situation. Their reactions were construed as further evidence of illness because they did not act in prescribed ways. A similar line of argument can be traced back to reasoning about the efficacy of treatment (Pembroke, 2006b). Pembroke states that there is no clear evidence to support any specific clinical intervention for self-harm, be it pharmacological or psychological. Irrespective of treatment, “if it fails—or the user thinks it's a load of crap—then that's because the user is either too stupid to grasp it, untreatable, non-compliant, or being just plain difficult” (Pembroke, 2006b, p. 17).

The interaction between a “fostering repertoire” and an “expert repertoire” was characterised by a struggle about who has the preferential right of interpretation. Both the women who self-harmed and the professional caregivers took on authoritative positions and they often had contradictory interpretations and understandings of self-harm, with the women seeing it as a functional behaviour to reduce mental suffering and the caregivers viewing it as a behaviour labelled with a diagnosis. A picture of self-harm opposed to the predominant picture within the psychiatric paradigm is offered within a context of determination, strength, and courage in coping with life stressors by Ekman and Söderberg (2009). Pembroke (2006a) reports her experiences of professionals who define the progress of recovery from self-harm as total cessation of the harming. She argues that this is not realistic for most self-harmers, and if professionals demand total cessation, then they do not recognise that self-harm is in itself a coping strategy and a way to survive (Pembroke, 2006a). McAllister (2003) suggest that if healthcare professionals could think of and describe self-harm as a kind of self-soothing rather than as a symptom of illness then it might be easier for patients and caregivers to understand each other. To self-soothe is to relax, to calm, to mediate, and to nurture, which may generate creative ideas to help restore calmness.

The professional caregivers in this study adopted a “fostering repertoire.” This meant being authoritative and having the power to set limits and to enforce rules about what is and is not acceptable. This is consistent with the American sociologist Howard Becker's concept of moral entrepreneurs (Becker, 1963, p. 147), those who have the power to speak in public and to be listened to about how and why individuals, things, and phenomena should be. There are two levels of moral entrepreneurs, one that creates values and rules and the other that enforces them. When professional caregivers acted as fosterers they were to some extent creators of rules, however, they acted predominantly as guardians of existing values and rules (Becker, 1963).

The interaction between a “supportive repertoire” and a “victim repertoire” was characterised by the caregivers’ efforts to take on an equal position and the women's underdog position. Sometimes the professional caregivers supported the women's underdog position and took on an underdog position themselves. Caregivers were trapped within the organisation, especially regarding rules connected to the ward organisation. O'Donovan (2007) reported that staff within psychiatric inpatient wards felt the focus of their duty was distributing medication rather than developing therapeutic relationships with the patients. They felt hindered from engaging in therapeutic activity because of the nature of psychiatric inpatient care and the medical model of care (O'Donovan, 2007), which echoes the results of the present study. The women in this study took on an underdog position. Johansson (2010) showed that the self-harmer identity hovers between two positions, named as “victim” and “‘actor.” A position as a victim was connected to sickliness, freedom from liability, vulnerability, and self-identifying as a patient or as a victim of the society (Johansson, 2010).

The interaction between a “supportive repertoire” and an “expert repertoire” was characterised by equality between the participants and the interactions among them were filled with a sense of community. The caregivers employing the supportive repertoire were facilitators who sometimes went against existing rules and took decisions opposed to the consensus of their colleagues. The women described such caregivers as those who listened to them and took them seriously. The women in this study supported the results of several other studies (Lindgren et al., 2004; McAndrew & Warne, 2005; Pembroke, 1998, 2006a, b) by pointing out the importance of being seen, being listened to, being respected, and being treated with dignity. Furthermore, they spoke about the importance of healthcare professionals showing an interest in them as individuals with their own resources and supporting them in using their resources. Johansson's (2010) position labelled “actor” is similar to the women's “expert repertoire” in the present study. The position of actor is connected to characteristics such as self-determination, responsibility, and self-identification of oneself as a strong and self-determined actor (Johansson, 2010). Empowering strategies regard each individual as the expert on her own life seriously challenge the medical model of expert-defined assessment, diagnostics, and treatment, and redefine positions of power in healthcare settings. The focus is on the individual narrative; her own description and understanding of herself and her situation; and her own language, strengths, and abilities. The empowerment approach fosters confidence in a person's own capacity to find her own solutions and strategies (Hewitt-Taylor, 2004; Laugharne & Priebe, 2006).

Conclusions

Four interpretative repertoires were dominant in jointly constructing the interaction between the women and the professional caregivers. The women and the caregivers were positioned and positioned themselves and people around them within these repertoires to make sense of their experiences of the interaction. The interactions between a “fostering repertoire” and an “expert repertoire” and between a “fostering repertoire” and a “victim repertoire” were described as largely unsatisfying by the participants. Inflexible ward rules, disrespect for one another, and a non-listening approach contributed to the unsatisfying experiences among the participants. The interactions between a “supportive repertoire” and a “victim repertoire” strengthened the underdog position took on by the women and allowed both the women and the caregivers to “hide” behind rules and restrictions. Interactions between a “supportive repertoire” and an “expert repertoire” were described as more “equal” and included satisfying experiences. It was essential for the women to be met with respect for themselves as individuals. Furthermore, it was necessary that caregivers consider the woman's own strengths, life chances, and knowledge about herself and acknowledge her right to define her own needs.

Implications for practice and further research

The findings in this study points to professional caregivers possibilities to foster hope to the person who self-harm by offering time to meet, talk, listen, and take the individual seriously. Caregivers need to look beyond behavioural symptoms and give power back to the individual. Furthermore, they are in a position where they can provide structure and possibilities for activity to the individual. Professional caregivers need to work in partnership with the individual who self-harms—experts by profession collaborating with experts by experience.

As described earlier, the interaction between persons who self-harm and professional caregivers is filled with complicated feelings. Therefore, we believe that clinical supervision provided to professional caregivers may strengthen and improve nursing practice. There is research evidence for the effectiveness of clinical supervision on peer support and stress relief for caregivers as a means of promoting professional accountability and the development of skills and knowledge (Brunero & Stein-Parbury, 2008). Thus, it is reasonable to believe that clinical supervision provided to caregivers would improve the care for persons who self-harm as well.

Further research could focus on interventions aiming to improve care for persons who self-harm, especially within psychiatric inpatient care. Furthermore, research focusing on men who self-harm is needed.

Conflict of interest and funding

The authors declared no conflicts of interest with respect to the authorship and/or publication of this article. Financial support was provided by the Swedish Psychiatry Foundation (Psykiatrifonden), Ebba Danelius Foundation (Ebba Danelius stiftelse), and the J.C. Kempe Memorial Fund for Scholarships (Stiftelsen J.C. Kempes Minnes Stipendiefond).