Relatives' experiences of brief admission in borderline personality disorder and self‐harming behaviour

Abstract Aim The aim of this study is to describe experiences of brief admission (BA) of people with borderline personality disorder and self‐harming behaviour, from the perspective of their relatives. Design A descriptive qualitative design was chosen. Methods Twelve relatives of people with borderline personality disorder and self‐harming behaviour who had access to BA were interviewed. Data were analysed with qualitative conventional content analysis. Results One overarching category: Hope for the future and three categories occurred: Breathing space, Personal responsibility and Structure. BA created hope for the future and the relatives appreciated that BA is a freer and easily accessible form of care that enables help at an early stage, compared with usual care. When BA functions, the structure and pre‐determined days of care give relatives a breathing space, and the uncertainty diminishes for the children, as the parent can still be present during inpatient care. The lack of places was described as a disadvantage of BA.


| INTRODUC TI ON
Brief admission (BA) is a relative new nursing crisis intervention offered to patients with borderline personality disorder (BPD) and self-harming behaviour in Sweden, Denmark, Norway and the Netherlands (Helleman et al., 2014a(Helleman et al., , 2014bHelleman, Goossens, et al., 2018;Thomsen et al., 2018). BA is complementary to regular admissions and can be used in an acute crisis to promote the constructive self-regulation of emotions and reduce stress and anxiety, thereby preventing self-harm (Helleman et al., 2014a) and reducing long-term admissions (Eckerström et al., 2020). In contrast to regular psychiatric inpatient care where the physician has a gatekeeper function, BA is a patient-initiated and self-planned admission in psychiatric inpatient care limited in time. The patient plans the care content in advance. No previous studies concerning BA have been found from the perspective of relatives. Relatives of people with BPD are of great importance in the recovery process and their experiences are an important source of knowledge (Juntunen et al., 2018). Family involvement is a key to reach long-lasting positive effects on mental health and family should be recognized as important members of the healthcare team. The term "person-centred care" alone may not adequately capture the importance of the family (Feinberg, 2014). Thus, mental health services should be based on a person and family-centred care, in which it is important to listen to both the patient and the relatives' voices to gain a more comprehensive understanding (Millenson et al., 2016).

| BACKG ROU N D
The lifetime prevalence of BPD in the general population worldwide ranges from 1 to nearly 6 percent (Gunderson et al., 2018;Tyrer et al., 2010). In the mental healthcare system, BPD is one of the most common psychiatric diagnoses, and between 15%-28% of the patients are affected (Gunderson et al., 2018). BPD is a severe mental disorder characterized by chronic feelings of emptiness, difficulties controlling anger, uncertainty about one's own identity and a pattern of unstable and intense interpersonal relationships (American Psychiatric Association, 2013;Brickman et al., 2014;Gunderson et al., 2018). Nearly 65%-80% of those with BPD are also engaged in some form of self-harm, with the consequence of repeated attempts at suicide (Brickman et al., 2014;Tyrer et al., 2010). Among Swedish adolescents aged 15-17 years, 56% reported at least one episode of self-harm during the last year and nearly 7% met suggested DSM-5 criteria for a potential non-suicidal self-injury (Zetterqvist et al., 2013). Among adults, 4%-5% have a history of self-harm (Koyanagi et al., 2015).
Although self-harming behaviours tend to decrease over time, they usually leave persistent and severe social disabilities, with negative effects on life events (Gunderson et al., 2018). BPD is associated with a range of poor outcomes, including interruptions in education, and increased unemployment and family conflicts (Wertz et al., 2020). Alltogether, this leaves a high burden on both patients and their relatives (Bohus et al., 2021). Compared with relatives of people with other mental disorders, relatives of people with BPD show a higher degree of grief (Bailey & Grenyer, 2014), and many fear their close one will, due to self-harming behaviour balancing in between life and death, find it hard to take care of themselves (Ekdahl et al., 2011). Relatives often take a great amount of responsibility and are important social support in the society for people with BPD (Sharma et al., 2016). Due to the psychological suffering (Gunderson et al., 2018) and the high burden experienced by relatives (Bailey & Grenyer, 2014), there is a need for effective treatments (Gunderson et al., 2018). Options vary greatly between different countries, but first-line treatment for people with BPD includes some form of psychological intervention (Storebø et al., 2020). Dialectical behavioural therapy appears to be a valuable treatment in reducing both self-harm and suicidal ideation (Kothgassner et al., 2021). Only a few of all patients with BPD are cared for in inpatient care, and then usually for suicidal behaviour (Hawton et al., 2012). In this context, there is a lack of knowledge-based methods and there is a contradiction in admissions recommendations. BA is a nursing crisis intervention aimed at reducing long hospitalizations, preventing self-harm and encouraging the patient's autonomy, self-esteem and influence over care and treatment (Eckerström et al., 2020). A contract is designed according to the individual patient's needs in collaboration between the patient, outpatient staff and nurse in the inpatient ward. The contract contains a care and crisis plan and is shaped according to the patient's wishes. During BA, no contact with a physician is allowed (Eckerström et al., 2019). The opportunity to recover with the help of the department's routines and structure is mentioned as one of the most important parts of BA (Eckerström et al., 2020). The pre-determined time for hospitalization of a maximum of 3 days is considered helpful as it minimizes dissatisfaction in connection with discharge (Ellegaard et al., 2017). BA offers the patient an opportunity to receive help at an early stage, so that emergency care and long hospitalizations can be avoided.
BA further helps to shift the focus from the patient's difficulties to the patient's resources (Eckerström et al., 2019). Nurses experience that BA promotes the nurse-patient relationship and inspires security and continuity (Eckerström et al., 2019). Consumption of care in inpatient care for patients with severe self-harming behaviour does not seem to decrease among BA consumers, but reduces self-harm and number of days in compulsory care (Strand & von Hausswolff-Juhlin, 2015;Westling et al., 2019). Patients described BA as being primarily comforting and empowering, especially given the fact that they can use it when needed (Eckerström et al., 2020); and it provides conditions to maintain valuable daily activities and relationships (Enoksson et al., 2021). Previous research has generated results that indicate that BA is effective and helpful from the perspectives of patients (Eckerström et al., 2020;Enoksson et al., 2021;Helleman, Goossens, et al., 2018) and healthcare staff (Arnold et al., 2021;Eckerström et al., 2019). To our knowledge, no previous study explores relatives' experiences of BA. Consumers with BPD and their relatives feel excluded and experience stigma and low quality care in traditional mental health services (Barr et al., 2020). They further experience that mental health care focuses on rapid treatments, which are not helpful and, after discharge, that responsibility depends entirely on them (Ekdahl et al., 2011), which in turn leads to a mistrust in psychiatric care. Thus, considering the perspectives of relatives is of importance when implementing new interventions in mental health services (Grenyer et al., 2017). Relatives want to be involved and recognized as a resource in the care of their close one (Weimand et al., 2013). Exploring relatives' experiences of BA can contribute to a deeper knowledge useful for mental health care in order to reach a person and family-centred care (Feinberg, 2014;Millenson What does this paper contribute to the wider global clinical community?
• Due to the psychological suffering, the great amount of responsibility and high burden experienced by relatives there is a need to explore new recommendations about clinical management from the viewpoint of relatives of individuals with BPD and self-harming behaviour.
• As the structure and pre-determined days of care given during brief admission give the whole family a breathing space, and better pre-requisites for a functional everyday life, the concept could have benefits also for other psychiatric diagnoses. et al., 2016). Thus, the aim of the study was to describe experiences of (BA) for people with borderline personality disorder and self-harming behaviour, from the perspective of their relatives.

| Design
A qualitative, explorative study design was chosen, as it provides new insights and increases the researcher's understanding of a particular phenomenon (Patton, 2015). Semistructured interviews were applied, as they encourage participants to express their experiences, though in a given frame. Conventional content analysis was used for analysis (Hsieh & Shannon, 2005). Conventional content analysis is useful when the purpose of the study is to study a phenomenon in an area where other research is sparse (Hsieh & Shannon, 2005). The research process and the presentation have followed the COREQ guidelines (Tong et al., 2007).

| Sample and setting
Participants were recruited from a region containing three separate outpatient psychiatric clinics that worked in collaboration with their respective inpatient psychiatric clinics to implement BA for patients with BPD and self-harming behaviour. First, the head of the department gave the research team written permission to conduct the study. After permission was received, healthcare staff working with BA were contacted to help identify patients diagnosed with BPD and self-harming behaviour who had been assigned to BA, regardless of when and how often they have used BA, all patients have constant access even though the majority did not use BA at the moment. The inclusion criterion for being invited to participate in the study was being a relative of a patient diagnosed with BPD and self-harming behaviour, with a signed BA contract. The term relative refers to a person with whom the patient considers themself to have a close relationship (The National Board of Health and Welfare, 2021). Therefore, the patients suggested to who they had a close relationship and thus be eligible for an interview and forwarded contact information to the research team. The relatives were contacted by phone and given information about the study. Those who were willing to participate were to decide the time and place for the interview in order to create a calm and safe environment with as little risk of being interrupted as possible (Patton, 2015;Polit & Beck, 2021). Written consent was obtained from both the patients and their relatives.

| Data collection
The interviews were conducted in a secluded room at the psychiatric clinic (2), in the relatives homes (3) or by phone (7) during November 2021 to January 2022. Before the interview, there was time to establish contact and take part in informal conversations and to give the informant the opportunity to raise questions. All relatives signed a written informed consent. The data were collected through individual semistructured interviews and conducted by HR and LW. The interview guide was designed in collaboration with the research team and consisted of open-ended questions, as they tend to provide more detailed and spontaneous answers, which was considered appropriate for the purpose of the study (Patton, 2015). The questions asked were "How do you experience Brief admission as intervention for your relative?" They were also, for example, asked: "Can you tell if there is anything that you think can be developed with BA?" The question guide was supplemented with asking the relative to develop their answers, or with additional questions such as "Can you tell me more?"-so-called probing questions (Patton, 2015). The interviews were recorded as digital audio files and then transcribed verbatim using a transcription guide (McLellan et al., 2003), to facilitate subsequent analysis. The interview had a median length of 40 min and during the last interviews no new data appeared and can thus be seen as a saturation of data.
After the interviews, there was time for reflection, if required, by the informants. Two test interviews were conducted, and as no changes needed to be made, these interviews were included in the analysis. Participants were given information about setting aside approximately 60 min for the interview (Polit & Beck, 2021).

| Data analysis
The results were analysed based on a conventional content analysis (Hsieh & Shannon, 2005). The interviews were read several times with the aim of creating an overall picture of the collected data. To find and highlight exact words in the text that capture key thoughts or concepts related to the aim, the transcripts were read word by word by all authors. Then, codes were marked and labelled as close to the original transcripts as possible. Codes with similar meaning were sorted into subcategories. The content in each subcategory was sorted into a smaller number of categories. Finally, definitions for each category and subcategory were developed. All authors agreed on how to categorize subcategories and categories. The name of the categories was derived from data, not from pre-conceived categories, and thus an inductive approach was applied (Patton, 2015). The generated subcategories and categories are presented in a table (Table 1) (Hsieh & Shannon, 2005). Quotations for each category were identified from the transcripts to report the findings and strengthen the trustworthiness of the results (Polit & Beck, 2021). As a final step of analysis, one of the co-authors read excerpts of the raw data to validate the content of the categories (Patton, 2015).

| Ethics
The study has been approved by the Regional Ethical Review Board in Linköping, Sweden (Dnr: 2021-04567) and was performed in accordance with the Declaration of Helsinki (World Medical Association, 2013). By obtaining informed consent from both the patient and their relatives, the idea was to avoid the patient feeling "that someone is talking over their head". There is always a risk of provoking reactions in an interview situation. At the same time, in this case, a "silent group" is given the opportunity to express their thoughts and experiences. There was a readiness to handle emotional expressions, as the interviewers were nurses with many years of experience in psychiatric work. Both patients and relatives were informed that regardless of participation and experiences of BA, their future care would not be affected. Data are presented at a group level, which means that an individual informant should not be able to be recognized based on the descriptions.

| RE SULTS
The sample consisted of 12 relatives, of whom eight were women.
They were aged between 22-75 years (mean age of 46). Their relationship to the patient varied and consisted of five mothers (one engaged), one grandmother, three spouses, one friend and one sibling.
The result consists of an overaching category: "Hope for the future" followed by three categories "Breathing space", "Personal responsibility" and "structure" which are distinct from each other (Table 1). In each category, the varying experiences are highlighted in subcategories. In the results, the descriptions of the subcategories are illustrated by quotes from the participants, with the interview number of participants in parentheses.

| Hope for the future
Brief admission creates opportunities for relatives to have time and space to take care of themselves, while at the same time BA creates conditions for the patient to live as normal a life as possible. BA shifts focus from relatives taking responsibility and constantly watching over the patient, to the patient him/herself taking responsibility for their own well-being and care seeking. The clear structure of BA also contributes to a more easily planned and predictable everyday life with hope for an improved future for the whole family. However, the hope is conditional on BA being available when needed.

| Breathing space
The knowledge that the patient has access to BA was experienced by relatives to provide respite and reduced anxiety. BA contributed to the patient being able to get help in a simpler way and at an early stage, which was described as giving time for other things.

| Reduced anxiety
Relatives describe a general feeling that had followed them for a long time, an anxiety that the patient will be mentally impaired, as their condition had previously led to self-destructive or suicidal actions.
After the patient has gained access to BA, relatives express reduced anxiety, as BA is easily accessible in comparison with usual psychiatric inpatient care. BA requires that the patient themselfs calls the department that handles BA and expresses the desire for care. The experience of relatives is that at the times the patient has received help at an early stage of the deterioration, the downward spiral of deterioration has been able to be stopped. The reduced anxiety of relatives is also based on the knowledge that the patient is in a safe place with access to care. It appears from the statements that only the knowledge that the patient has access to BA has reduced the relatives' concerns. The abovementioned factors contribute to giving relatives the opportunity to relax in a situation that would otherwise have been stressful for both the patient and relatives: The security above all, I think, that she is under surveillance … she is in a place where she can still go out … she is not locked up. I do not think it will be as much anxiety for me … It will be a security for me that allows me to relax more. (P5)

| Time for something else
When the patient is not feeling well, relatives describe that this occupies their entire everyday life. They describe that they cannot think of anything other than the risk of the patient injuring themselfs. It also emerged that this led to their own health being left behind. With BA, relatives describe that they have been able to take care of themselves and let go of responsibility for a while: Yes, then I feel better… then he can call and say 'yes mom … now I do not feel well … I will go to bed a couple of days.' Then I know that I can kind of let go and concentrate on myself a little.
There are relatives who experience difficulties in giving the remaining family members the attention they need. They describe TA B L E 1 An overview of the categories and subcategories produced using Hsieh and Shannon (2005)  A long time of hospitalization means that so much of her skills are lost and she is forced to start all over again from the beginning … and it takes so much effort to rebuild.
It takes a lot of energy and time for both the relative and the patient to regain these skills after long care periods. The patient has many times worked on the abovementioned skills through many years of therapeutic care that have been tough for both the patient and relatives. BA is thus described as a form of care that helps the patient to maintain the skills that have been developed: It is very good that she can work two hours a day, even when she is hospitalized. She has a job she enjoys very much too, and she is educated and such. So it's good for her to come to work.
Relatives appreciated the ward where BA is offered and described a calm over knowing which ward the patient will be cared for and who works there. Together with the patient, they had positive experiences of the ward, which contributed to a calm. In previous admissions via the psychiatric emergency department, the uncertainty about where and if the patient will be admitted has been a cloud of concern, as relatives know that the patient is not feeling well in certain psychiatric wards: The care is very different, depending on which ward has a place, at that very moment, so that instead you can know, that if you apply for this, well then, it is this ward and then you know a little what kind of staff it has … to know that feels good I think. (P6) If it turns out during BA that the patient is in need of extended care, the patient will be helped to seek care at the psychiatric emer- It is difficult to get in on time if there are no places.
And I know the last few times … she has called many times before she got a place and then she had already gotten worse. Relatives in this study describe that, before the patient had access to BA, they have felt a significant concern when the patient's In this study, it emerged that relatives take great responsibility for the patient and have difficulty focusing on anything other than their mood. Many relatives of patients with BPD describe the fear that something terrible will happen that may result in the patient's suicide (Ekdahl et al., 2011). In addition, it is considered of great importance to avoid a deteriorating mood that can lead to the patient being exposed to risks. BA is also referred to as a tool for preventing suicide attempts (Arnold et al., 2021). The fear that the patient will hurt themselves or take their own life during a deterioration is something that recurs among relatives in this study, and access to BA therefore contributed to relief. Living with someone with mental illness is described by close relatives as a great responsibility (Radfar et al., 2014). Relatives estimated that BA gave the patient the opportunity to apply at an early stage, and that further deterioration could be avoided. Similar results have been described by nurses and patients (Eckerström et al., 2019;Enoksson et al., 2021).
Increased patient responsibility and care at an early stage can thus be perceived as significant and helpful factors from the perspective of both the patient (Eckerström et al., 2020;Enoksson et al., 2021), healthcare staff (Arnold et al., 2021;Eckerström et al., 2019) and, as found in this study, relatives. Relatives in this study said that they promote the patient's ability to take responsibility for, and be observant of, early signs of deterioration, in that it is the patient who has control over enrollment and discharge, which has also emerged in other studies (Ellegaard et al., 2017;Enoksson et al., 2021;Helleman et al., 2014aHelleman et al., , 2014b. This is something that could be linked to the fact that BA is a nursing intervention that strengthens the patient's autonomy, which is an important part of person-centred care (Ekman et al., 2011;Gabrielsson et al., 2015).
However, based on our result, family-centred care could be given more attention in BA. The results of this study indicate that relatives experience an improved life situation by giving time to their own interests and that the structure of the form of care provides an opportunity to plan everyday life. Eckerström et al. (2022) report in their study that the majority of patients who use BA estimate a higher quality of life in comparison with before BA (Eckerström et al., 2022). No concrete measurements about the life situation of relatives have been carried out in this study and is why it would be valuable to further investigate this in future studies. The well-being of relatives is important, as they have a supportive role for the patient (Sharma et al., 2016;Ten Have et al., 2016).
The majority of relatives appreciated the freer form of care, with a limited number of care days, because the patient's ability to, for example, take care of the children and go to work is maintained. From the patient's perspective, some experience concerns about the limited number of care days (Ekman et al., 2011). These varying perceptions can be understood from the fact that patients use BA to get a break from everyday life, especially during stress (Enoksson et al., 2021), while relatives, as mentioned above, find it positive that the patient returns home quickly to help out with daily chores. Relatives of people with BPD experience more suffering than relatives of people from other diagnostic groups (Bailey & Grenyer, 2014), and many suffer from anxiety and depression (Bailey & Grenyer, 2014;Ekdahl et al., 2011), which makes them even more exposed to mental diseases. BA may, although it is structured and restricted to a limited number of care days, seem to create space for relatives to recover. Although some patients experience a limitation in the number of care days during BA, it is perceived as helpful in reducing stress (Enoksson et al., 2021). This emerged from relatives in this study, which means that BA should not only be considered helpful from a patient perspective but also from the relatives' perspective.
In this study, relatives highlight that refusal of BA due to lack of space made some patients not so willing to use it again. This contributed to an increased concern among relatives that the patient would self-harm and led to frustration over lack of BA places. When based on person-centred care, care should be designed based on the patient's individual needs (Gabrielsson et al., 2015). From the patients' perspective, a disappointment arose if the room was occupied, making it difficult for them to call the next time (Helleman, Goossens, et al., 2018;Helleman, Lundh, et al., 2018). The knowledge of quick help is fundamental and constitutes a security in BA (Eckerström et al., 2020;Strand & von Hausswolff-Juhlin, 2015).
Relatives in this study also mention the lack of space as an obstacle to achieving this security, as a refusal can lead to the patient acting self-destructively (Helleman, Goossens, et al., 2018;Helleman, Lundh, et al., 2018). The lack of space then becomes an obstacle to working in a person-and a family-centred way, because then both the patient's and relatives' autonomy and participation are hindered by organizational factors (Ekman et al., 2011;Feinberg, 2014;Millenson et al., 2016).
The results showed that relatives can perceive BA as meaningless, and a structure with daily conversations was requested.
How the agreement is designed, and how involved the patient and relatives are, can play a role in the experience of the form of care, as the patient themself is the one who communicated wishes about what should happen during the care period (Eckerström et al., 2019;Helleman et al., 2014b). The results showed that some relatives did not know that there was an individual agreement, which could explain why the form of care is perceived as meaningless. Some relatives wanted more information and wanted to participate in BA, while others wanted to be out of care. In psychiatric care, it appears that relatives are not involved or did not receive information to the extent they desired (Cree et al., 2015).  (Juntunen et al., 2018), and involving relatives in care is also one of the basic principles in person-centred and family-centred care (Feinberg, 2014;Gabrielsson et al., 2015;Millenson et al., 2016).
Studies show that patients with BPD, without relatives present, are at risk of developing more severe symptoms related to their diagnosis (Ten Have et al., 2016). Many relatives feel excluded from care (Ekdahl et al., 2011). Excluding relatives is something that must be passed on to history. Relatives should be recognized as a partner, as it is likely to be gainful for all parties (Juntunen et al., 2018). Today, BA does not involve relatives; by asking the patient if relatives may be present when an agreement is written can be one way to include relatives.

| Limitations
When using the conventional content analysis, there is a risk that the authors fail to identify important parts of the respondents' statements (Hsieh & Shannon, 2005). The reliability of qualitative descriptive research is based on how well the authors managed to exclude irrelevant data and include and categorize relevant data according to what the respondents shared in the interviews (Polit & Beck, 2021). To increase reliability, the results have been strengthened with quotes from the transcripts, and throughout the analysis process all authors collaborated with each other so to ensure no relevant data was missing. First and last authors have solid experience of analysing qualitative data, which can also increase credibility (Patton, 2015).
Although BA is currently offered in several different countries with the common denominator "self-referral admissions to psychiatric inpatient care", the execution differs depending on how the psychiatric care is structured locally (Eckerström et al., 2022;Enoksson et al., 2021;Helleman et al., 2014a). Therefore, the transferability of the results may be limited. A clear description of the relatives, the context in which the study took place and of the research process, are given in order to make it possible for the reader to decide whether the results are transferable to another context (Patton, 2015). Most of the interviews were conducted digitally. Face-to-face interviews are considered to be the most optimal in qualitative research, as data can thereby be enriched with body language and emotional reactions among participants, which cannot be obtained through telephone interviews (Polit & Beck, 2021). In qualitative research, however, telephone interviews are described as a good alternative as they are less time-consuming and provide more flexibility at, for example, large geographical distances (Opdenakker, 2006). The possibility of the flexibility of being able to interview despite geographical distances were considered as important factors in obtaining relevant data in this study. Of the relatives, only four were men. In the selection process, the gender perspective was not taken into account as this was not considered to be relevant for the purpose of the study (Curtis et al., 2000). However, in future research, it would be interesting to shed light on whether there are differences in perceptions of BA seen from a gender perspective. It was surprisingly easy to recruit relatives to the study and they all expressed gratitude for participating. Previous studies illustrate that social stigma is the most common and challenging burden among relatives of people with BPD (Meshkinyazd et al., 2021), and many relatives feel excluded from health care (Ekdahl et al., 2011). As mentioned earlier, the relatives who participated in this study had great amounts of knowledge of their close ones' needs and were familiar with how existing health care works for them (Juntunen et al., 2018). To ensure validity, a semistructured interview guide with open-ended questions was used. This is a pre-requisite to catch the informants' thoughts, experiences and perceptions about parts of the world about which they are talking, in this case BA (Patton, 2015).

| CON CLUS ION
When BA functions and is available in an early stage to prevent selfharm, relatives said that they had a breathing space and time for recovery. Furthermore, BA contributes to relatives daring to let go of the responsibility they previously took over the patient, which promotes the patient's autonomy and relieves the burden of the relatives. Not all relatives had the information they wanted about BA. By asking the patient if relatives can be involved when an agreement is written, relatives would be given better conditions to receive information and be involved, which could also promote a person and family-centred approach in psychiatric inpatient care. Furthermore, it emerged that the children are less affected during BA than usual care, as it is clearly time-limited in its approach. This led us to the conclusion that BA is a form of care that should not only be considered helpful from a patient's perspective but also from a relative's perspective.

| RELE VAN CE TO CLINI C AL PR AC TI CE
For relatives to trust and feel secure with BA, care beds must be available when needed. As relatives have an important role in the patient's life and want to be supportive, they need to be given information and the opportunity to be included in the BA concept.
Relatives described an improved life situation for themselves and for the children, but larger quantitative studies are needed to explore this further.

ACK N OWLED G EM ENTS
Many thanks to the participants in this southern Swedish region who have participated in this study and been generous in contributing their knowledge and perspective during the interviews. Also thanks to Språkservice Sverige AB for proofreading the article.

FU N D I N G I N FO R M ATI O N
This work was supported by the Department of Psychiatry, Ryhov County Hospital, Jönköping.