A thematic analysis of care provider experiences of using self-harm abstinence agreements in psychiatric inpatient care.

AIM: Traditional methods used when managing self-harm in a psychiatric inpatient setting tend to infringe on the autonomy of the individuals receiving treatment and are often experienced as practically and emotionally challenging by care providers. Therefore, we examined care providers' experiences of an alternative method negotiating self-harm abstinence agreements, which can be viewed as a form of positive risk taking.
DESIGN: A qualitative approach using semi-structured interviews with twelve mental health professionals.
METHODS: Thematic analysis of the interviews.
RESULTS: Five themes emerged; "No-harm agreements versus constant observation and coercion," "No-harm agreements to promote independence and collaboration," "No-harm agreements' effect on ward safety," "Ambiguity surrounding the no-harm agreements" and "Ethical complexities of the no-harm agreements." These indicated perceived positive effects on the therapeutic relationship, the individuals' autonomous functioning and the ward environment, but also practical and ethical difficulties.

INTRODUCTION

Self‐harm behaviour in a psychiatric inpatient setting is a complex and challenging phenomenon for care providers. On the one hand, care providers should help individuals to manage their dysfunctional behaviours. But on the other hand, additional care and attention after an act of self‐harm risks reinforcing the self‐harming behaviour and increasing the likelihood of recurring self‐harm incidents. Developing pragmatic and efficient practices to negotiate these situations is essential for the quality of care.

BACKGROUND

Traditionally, strategies for managing self‐harm behaviour in inpatient settings have focused on risk prevention. With the aim of reducing the negative consequences of a mental health problem (by hindering self‐harm behaviour), treatment may infringe on the autonomy and integrity of individuals seeking help, for example by use of special observation or coercive measures to prevent an individual from committing suicide or inflicting severe permanent injuries upon themselves. However, excessive use of such measures can be counterproductive (James et al., 2012). The treatment period can turn into a “cat‐and‐mouse‐game” wherein the staff tries to prevent individuals from engaging in self‐harm and individuals try to bypass the measures taken by the staff (Lindkvist et al., 2019). There is no clear evidence that special observation reduces self‐harm rates (Bowers et al., 2008), and individuals under constant observation can experience it as claustrophobic and anxiety‐inducing (Breeze & Repper, 1998; Cardell & Rogers Pitula, 1999), with some having reported refraining from adequately taking care of their personal hygiene whilst under observation and even lying about their level of suicidality in hopes of escaping it (Cardell & Rogers Pitula, 1999). Coercive measures, such as mechanical restraint, are experienced as contributing to feelings of powerlessness and distrust of the ward staff (Ejneborn Looi et al., 2015). Fear of coercive measures can lead to individuals self‐harming in secret whilst on the ward and it can also deter them from seeking psychiatric care again (Ejneborn Looi et al., 2015). Thus, when care providers become overly focused on risk prevention it may end up damaging the relationship with the individual receiving care and ultimately hindering effective caregiving (Morrissey et al., 2018).

One alternative to traditional risk prevention strategies, that has been described in previous studies and can be viewed as a form of positive risk taking, is different types of agreements aimed at self‐harm abstinence, that is no‐harm contracts (O’Donovan, 2007), no‐suicide contracts (Drew, 2001) and safety agreements (Potter et al., 2005). Such agreements can be either written or verbal and are sometimes noted in the individual's treatment plan. They may involve individuals promising to reach out to staff when experiencing self‐harm impulses (O’Donovan, 2007; Potter et al., 2005) and staff asking the individual what coping strategies they find helpful (Potter et al., 2005). There is no indication that these agreements result in reduced self‐harm frequency (Drew, 2001; Potter et al., 2005) but staff have described that they can help facilitate a dialogue about safety and can result in individuals taking more responsibility (Potter et al., 2005). A more recent example of an agreement involving individuals promising to refrain from self‐harm is the Brief Admission crisis intervention, where the individual signs a contract that includes a commitment to refrain from self‐harm behaviour during the inpatient care period and to ask the staff for help and accept the help they are offered (Liljedahl et al., 2017).

In an inpatient setting such self‐harm abstinence agreements may provide an alternative to traditional risk prevention and can be viewed as a form of positive risk taking, where care providers strive to find a balance between protecting individuals from unnecessary risks whilst still providing them with opportunities for personal development and exertion of autonomy (Morgan, 2004). This type of approach involves weighing the possible short‐ and long‐term consequences of different strategies against each other and accepting that temporarily heightened risk may occasionally be necessary in order to achieve positive long‐term results (Morgan, 2004). In order to decide which risks are necessary to take, and devise a strategy for managing those risks, awareness of the individual's history and their current condition and abilities is essential (Morgan, 2004). This makes a focus on collaboration and the development of a strong therapeutic alliance an integral component of any method based in positive risk taking (Morrissey et al., 2018).

Self‐harm behaviour in the inpatient context also impacts the care providers that work with these individuals, and other patients. Clinicians generally experience it as emotionally challenging (Saunders et al., 2012). A Norwegian national screening investigation examining psychiatric care for individuals with severe and prolonged self‐harm reveals exhaustion, frustration and despair across several levels and from different vocations in the healthcare system (Holth et al., 2018). Psychiatric nurses report feeling compassion for individuals who self‐harm but also finding the behaviour disturbing and provocative (Tofthagen et al., 2014). Because of difficulty coping with strong feelings of fear and anger, staff may resort to unprofessional and unethical behaviour such as screaming at individuals or being physically rough with them (Wilstrand et al., 2007). Inability to help individuals stop harming themselves can sometimes be experienced as a personal failure (Tofthagen et al., 2014) and can cause feelings of fatigue and frustration which may have a negative influence on the care providers’ attitudes towards individuals receiving treatment (Lindkvist et al., 2019). Psychiatric nurses have also described uncertainty regarding how to care for self‐inflicted injuries, due to a fear of reinforcing self‐harm behaviour by focusing too much attention on it (Wilstrand et al., 2007).

Forms of no‐harm agreements have been used unsystematically in psychiatric wards in Sweden at least since the late 90s. However, few written accounts of these practices exist, some brief mentions by individuals speaking about their experiences of psychiatric inpatient care being exceptions (e.g. Eriksson & Åkerman, 2012), and thus, these practices need to be studied further.

The aim of this study was to explore professionals’ experiences of working with no‐harm agreements and thereby build on previous literature on how psychiatric care providers manage self‐harm behaviour and related issues in an inpatient setting.

THE STUDY

Design

The study utilized a qualitative methodological approach based on interviews with professionals that had experience of working with self‐harm in a psychiatric care context where no‐harm agreements were also used. The study was conducted at a Swedish University hospital general psychiatric inpatient ward, typically treating 12–16 inpatients, amongst which self‐harm and suicidality were common presenting problems. The ward is locked but encompasses both voluntary and coercive admissions. The practice of using no‐harm agreements typically entailed that individuals, who were treated voluntarily and had a history of self‐harm behaviour, were asked to refrain from self‐harming during inpatient care. These patients typically had previous experiences of being treated with coercive measures due to imminent and prolonged suicidality. The agreement was typically made with the resident psychiatrist when admitted to the ward. In some cases, where individuals were treated recurrently, other members of the staff (typically nursing staff) could also renew these agreements, and continuously negotiate them on a day‐to‐day basis, as part of the routine care given. As part of the agreement, individuals were asked to seek support from the nursing staff when distressed. Engaging in self‐harm at the ward would result in the individual being discharged by the resident psychiatrist and referred to seeking treatment at the emergency department anew.

A purposive sample of twelve participants was recruited through a general e‐mail invitation submitted at the units of the clinic involved in treating self‐harming patients in inpatient care. Those who agreed to participate were then contacted through e‐mail and telephone by a research assistant, who also scheduled and conducted the interviews. Ten interviews were conducted in person at the clinic and one through telephone, one was conducted at the participant's home. The participants represented the different professions involved in general psychiatric inpatient care (i.e. psychiatric nurses, orderlies and psychiatrists), and dialectical behaviour therapists involved in both inpatient and outpatient care.

Method

A semi‐structured interview guide was constructed for this study by the research team jointly, based on the theoretical framework by Morgan (2004) and the researchers’ clinical experience. The research group comprised of clinically active researchers (JB – Ph.D., clinical psychologist, SW – M.D., psychiatrist) and a research assistant (EIL), who conducted the interviews. All members of the research group had knowledge about the practices at the research site and basic familiarity with the staff at the inpatient wards through their clinical work. All interviews were audio‐recorded. The length of the interviews varied between 22 and 67 min (mean length 48 min).

The main topics included in the interview were asking participants about experiences of working with no‐harm agreements, how the no‐harm agreements were applied in practice, difficulties associated with no‐harm agreements, advantages and possibilities associated with no‐harm agreements, and care providers’ and patients’ experiences of the no‐harm agreements.

Analysis

Interviews were analysed using thematic content analysis, according to the six steps suggested by Braun and Clarke (2006). The interviews were audio‐coded and annotated within the NVivo 12 software, a method that has been suggested as an equivalent alternative to standard transcription in qualitative research (Wainwright & Russell, 2010). Two of the authors (EIL and JB) listened to the recordings, these were then reviewed several times, and themes were identified and revised in iterations throughout this process (by EIL and JB). Themes were identified based on semantic or explicit content from the data. In total five themes pertaining to the participants’ views and experiences were identified and these, and the interpretations of them, were discussed in the research group (i.e. also involving SW) to triangulate and thereby strengthen the credibility of the results (Korstjens & Moser, 2018).

Ethics

Ethical approval for the study was obtained from the regional ethics board for Lund University (Dnr: 2018/573). This approval stipulated information to participants that data was to be treated confidentially, and stored and managed in accordance with current regulations. All participants gave written informed consent to their participation, including consent for audio recording.

RESULTS

The care providers described experiences of feeling both professionally and personally challenged when trying to provide psychiatric inpatient care for some individuals with self‐harm behaviour. Below, five themes that capture central aspects of the participants’ experiences of working with the no‐harm agreements are presented.

No‐harm agreements versus constant observation and coercion

Many participants spoke of the agreements in relation to more traditional risk management methods, and it was described that individuals often appeared to be helped in refraining from self‐harming behaviour by making a no‐harm agreement at the ward.

“No‐harm agreements are an alternative method [to treatment as usual], that I feel is less abusive towards the patient, that gives more options, more… room for dialogue than coercive measures, for example, or treating someone against their will, or the special observation measures.”

However, it was also described that for a few individuals with especially severe self‐harm behaviour such agreements were not particularly effective, but they were used anyway because these individuals’ self‐harm behaviour would otherwise escalate until it became unmanageable for the care providers. The alternative to a no‐harm agreement in these cases was described as converting a voluntarily treated individual's care to involuntary treatment. This usually led to coercive measures and constant observation also being used, in many cases for up to several months because of the perceived risk of suicidal behaviour, which was considered more destructive for the individual in the long term compared to a no‐harm agreement.

“When you have tried using control, coercion and so on, for a number of years and it still doesn't help… then this is the least bad alternative. But still, it never feels good of course.”

Participants did, however, also describe that in a lot of cases the risks of relying on no‐harm agreements with certain individuals became unjustifiable after one or a few failed attempts. One participant mentioned that this could be the case with particularly young individuals with severe self‐harm behaviour who lacked alternative skills to manage self‐harm impulses. This group of individuals was described as often having recurring periods of involuntary treatment, where they could not be expected to enter into a no‐harm agreement.

No‐harm agreements to promote independence and collaboration

Several of the participants expressed that individuals who self‐harmed appeared unwilling to seek out and receive support when they were distressed, which limited the care providers’ role to primarily treating the self‐inflicted injuries after they occurred. This was described as a source of feelings of frustration, helplessness and resignation amongst the care providers. One participants described that when individuals lacked motivation to accept help it could lead to a destructive power struggle, where the individual tried to circumvent any measures aimed at reducing self‐harm behaviour. The no‐harm agreement was experienced as a positive contribution, as it was perceived as making individuals more involved in their own care, resulting in increased collaboration and acceptance of support from the care providers.

“[A no‐harm agreement] is more rewarding for me as staff to work with, it gives me more of an opportunity to motivate the patients and provide support when they are at their most distressed, instead of just bandaging them up afterwards… and, yeah, saving them when they try to kill themselves on the ward. Because sometimes, I think it feels like patients let go when they get admitted (…) that they dare to try more drastic measures here, because they know that the staff is around and can catch them.”

When care providers assumed too much responsibility for the safety and well‐being of individuals receiving inpatient care it was described to contribute to the individual becoming more passive and dependent. A majority of the participants believed that in order to improve from the psychiatric care individuals need to learn and practice strategies to regulate their emotions. A number of participants perceived the no‐harm agreement as facilitating this, by contributing to the individual taking more responsibility for their own well‐being.

“Because we know that for this type of patient, inpatient care, medication, it helps very, very little. What helps is using the staff for support and trying to take control over one's own distress (…) because everything else is just storage of the person and postponement of the problems.”

Some of the participants described that individuals at the ward generally seemed to view responsibility as something difficult or negative. One participant reasoned that care providers often assumed that everyone viewed responsibility as a fundamental aspiration and therefore seldom talked to individuals receiving care about the positive aspects of taking responsibility, or about the fact that all hospitalization, psychiatric and somatic alike, is associated with a risk of reduced autonomous functioning.

No‐harm agreements’ effect on ward safety

A number of the participants brought up that one effect of the no‐harm agreements was a reduced amount of self‐harm incidents, which was positive for both the care providers and individuals treated at the ward. It was described that the amount of self‐harm incidents prior to the use of no‐harm agreements had sometimes been unmanageable. When the work environment was characterized by frequent self‐harm incidents, the participants described it as emotionally very taxing, even traumatic and a source of continuous stress, which could lead to loss of engagement in their work.

“No‐harm agreements are a really good thing (…). It saves a lot of time, and a lot of suffering for the patient, actually. For us too, we have feelings too after all… and it's been a bloodbath on many occasions, before, it was pretty gruesome.”

The participants also described that the individuals receiving care fared better when they were not exposed to self‐harm in others as much and the ward milieu was calmer. This also meant that the care providers had more time and energy to help individuals in distress.

A number of participants brought up that on occasion care providers had been unable to keep their part of the no‐harm agreement; to help a distressed individual manage without resorting to self‐harm. This was attributed to an unmanageable workload or being understaffed. One participant mentioned that she had noticed that individuals sometimes refrained from asking for help when they perceived the ward as being understaffed or the care providers as being too busy.

“During nights and weekends, there's minimal time available, I feel. That's the sad part of this, in my opinion, we are understaffed and therefore unable to respond and catch these patients. (…) I feel that the patients are reluctant to take up too much of our time, to put too much pressure on us staff.”

It was also described that the care providers did not have time to work more systematically, for example by using situation analyses to explore what triggered an individual's need to self‐harm and that there was a perceived lack of therapeutic skills amongst the staff, making it more difficult to help individuals see patterns in their self‐harm behaviours and identifying alternative strategies for emotion regulation.

Ambiguity surrounding the no‐harm agreements

The majority of the participants emphasized that in order for no‐harm agreements to be effective it had to be very clear for the individuals receiving care what the agreement entailed, and the care providers needed to be very consistent in their application of it. At the same time, several of the participants described that consistency could be difficult to ensure due to unforeseen circumstances, for example an individual's condition drastically worsening to the point that the agreement needed to be revoked or individuals misinterpreting the conditions of the agreement. In practice, such occurrences sometimes resulted in individuals being allowed to remain at the ward after having self‐harmed.

“When you have a rule… if you are going to make exceptions, then it's not really a rule anymore. Then you have to be very clear about that in some way, and then you leave openings.”

Misunderstandings concerning the no‐harm agreements could sometimes be traced to how they were documented in the treatment plans. The care providers strove to make the agreements as individualized and specific as possible, for example by listing individuals’ usual methods of self‐harm. This sometimes resulted in individuals presuming that other forms of self‐harm, that were not explicitly stated, were allowed. When such situations arose the physician usually assessed whether the misunderstanding was reasonable, in which case the individual was not discharged. Several participants described instances when individuals had been allowed to stay on the ward despite breaking the agreement.

“The times when that has happened… where patients have told us that “no, but I misunderstood this, I thought it didn't apply when I was on temporary leave”, for example, or “I didn't think it included me trying to hang myself, I thought perhaps it only applied to cutting”, then… that can build up a bit of frustration in the staff. Because it kind of feels like they are just trying to get away with it.”

Some of the participants expressed that the formulation of the no‐harm agreement should be more focused on alternative strategies that could be helpful for the individual. It was also mentioned that whilst it was always verbally stated that the individual should ask for support instead of harming themselves, this was not always written down.

Ethical complexities of the no‐harm agreements

Several of the participants believed that the method was beneficial in the care of many individuals, for instance because fewer self‐harm episodes led to reduced feelings of guilt and shame and because it prevented individuals self‐harming in order to show that they were not ready to be discharged. At the same time, a majority of the participants brought up the risk that individuals who were discharged for breaking a no‐harm agreement could perceive it as a punishment and that this could lead to anger, guilt or shame, and possibly an increased risk of self‐harm or suicidal acts.

Many participants experienced the discharges as emotionally difficult on occasion. One participant expressed uncertainty regarding situations where individuals with a no‐harm agreement had self‐harmed and described the act as having suicidal intent. There were no experiences of individuals having committed suicide after being discharged from the ward, although self‐harm following discharges was relatively common, but some of the participants expressed concern about the opportunity.

“After all, you discharge a patient who is very upset and dysregulated. These are very, very difficult discharges. And you often feel like…. you are often worried that this particular patient will take their life at this particular time. To date, that hasn't happened”.

It was described that a few individuals had not sought out care again after a no‐harm agreement had been formulated at their first admission to the ward, which was perceived as a potential indication that they had been deterred from seeking help again after being discharged. There was said to be an opportunity for individuals who did not feel like the no‐harm agreement was helpful to discuss this with the physician, but a couple of the participants expressed doubts regarding whether all individuals would be able to do that, given their difficulties and the inherent power imbalance between physician and patient.

Some participants also brought up that individuals might find it too difficult to ask for support when they needed it. It was described that individuals sometimes questioned the fairness of different approaches being used for different patients, as it effectively meant that some individuals could self‐harm on the ward whereas others were discharged for the same behaviour. There was also considered to be a risk of individuals harming themselves in secret or developing other destructive behaviours rather than asking the care providers for support. A couple of the participants brought up that the method could be hard to understand for the patients’ families, who might get scared and upset when a patient was discharged for harming themselves when that was one of the reasons they were admitted to the ward in the first place.

“It would be good if family members were given information about the reasoning behind it, because otherwise I think it's very hard to understand and that you'd get very scared and that you'd also think the health care system isn't helping you.”

One participant described that other healthcare providers, and some individuals receiving care, could have a negative attitude towards the no‐harm agreements and consider it an inhumane practice. This critical view of the method was shared by another participant who considered the discharge to be a punishment and believed it was unreasonable to demand that individuals should be able to refrain from self‐harm. On the other hand, a couple of the participants compared the no‐harm agreement with the rules in addiction treatment, where it seemed to be less controversial to discharge individuals for using alcohol or drugs, even though it carried similar risks.

DISCUSSION

In this study, five themes emerged that capture the potentials and challenges of using no‐harm agreements with individuals who self‐harm whilst receiving psychiatric inpatient care. The first three themes emphasize the participants’ views that no‐harm agreements are useful in the treatment of the individual patient, by offering support, behavioural alternatives and encouraging the individual's agency in managing their problematic self‐harm behaviour. And further, that the practice has implications for the care providers and the healthcare system they are a part of, by offering a rationale and an alternative in managing situations where traditional approaches have historically led to problematic outcomes (e.g. escalating self‐harm behaviour whilst on the ward) and an overreliance on special observation and coercive measures to control the behaviour of individuals receiving care. The two remaining themes emphasize difficulties and challenges that this way of working presented for the care providers, both pertaining to the concrete application of the method (how it is communicated, negotiated and applied), and the implications of applying this type of “positive risk taking” (Morgan, 2004) strategy in psychiatric care (that it highlights inherent goal conflicts of care, such as that between short and long‐term goals).

More specifically, participants described that in relation to the treatment of individual patients, the no‐harm agreements had positive effects by facilitating collaboration between the individual and the care providers. The practice encouraged individuals to seek out support before they self‐harmed (rather than afterwards). This was believed to have a positive long‐term effect on the individuals’ ability to be self‐sufficient and less dependent on recurring inpatient care, as they learned to use other strategies for emotion regulation that they could then make use of outside of the ward. Encouraging individuals to actively seek out support from care providers is a quality the no‐harm agreement shares with other similar agreements (Liljedahl et al., 2017; O’Donovan, 2007; Potter et al., 2005).

Some participants, however, described complicating factors regarding the facilitation of help‐seeking behaviour. One such factor was insufficient staffing, which was described to result in individuals sometimes being reluctant to approach the care providers for fear of burdening them or that care providers were simply unavailable. Another complicating factor was a perceived lack of knowledge about self‐harm behaviour and appropriate treatment techniques amongst the staff. This is problematic as previous research has shown a correlation between the number of self‐harm accidents and the number of qualified nursing staff in psychiatric wards (Bowers et al., 2008). A method that relies on individuals seeking support from care providers depends heavily on the care providers being readily available and proficient in the therapeutic skills necessary to help the individual develop healthier coping strategies.

In addition, the no‐harm agreements were experienced as beneficial for the general ward milieu and by extension the staffs working environment. Previous research has shown that self‐harm can have a clustering or contagion effect (Beck et al., 2018; Weber, 2002), which can pose a problem when many individuals who struggle with self‐harm are treated in the same place. The participants described that the amount of self‐harm incidents prior to the no‐harm agreements could sometimes be unmanageable and that this could contribute to the working environment being experienced as stressful and traumatic. In previous studies it has been described that care providers can find self‐harming behaviour to be provocative (Tofthagen et al., 2014) and may resort to unprofessional and unethical behaviour towards individuals receiving care due to inability to cope with the negative emotions they experience (Lindkvist et al., 2019; Wilstrand et al., 2007). It is thus possible that a reduced number of self‐harm incidents on the ward could contribute to a better therapeutic relationship by contributing to increased staff well‐being.

However, the fact that the no‐harm agreements are perceived as beneficial for the care providers carries its own risks. Individuals with experience of inpatient care have described that care providers sometimes use coercive measures routinely rather than as a last resort (Ejneborn Looi et al., 2015), and it is conceivable that something similar could occur with the no‐harm agreements; that they might be used out of convenience rather than after careful consideration. Participants conveyed an impression that the no‐harm agreements had resulted in reduced self‐harm incidence at the clinic. Whilst previous studies have found no such effects (Drew, 2001; Potter et al., 2005) and this impression could potentially be explained by confirmation bias, it is also conceivable that agreements that include a negative sanction (being discharged) could reduce self‐harm incidence.

The practice of using no‐harm agreements for individuals who self‐harm described in this study is reminiscent of other types of self‐harm abstinence agreements used in psychiatric inpatient care (Drew, 2001; O’Donovan, 2007; Potter et al., 2005). A notable difference, however, is that these no‐harm agreements include discharge from the ward when individuals self‐harm instead of asking the care providers for help. It was reported that in some cases the method was used despite the care providers expecting individuals not to adhere to the agreement. In these cases, the alternative to a no‐harm agreement was perceived as likely to lead to an escalation of self‐harm behaviour, which then would be followed by coercive measures and possibly long‐term constant observation to manage the situation. Compared to this type of scenario, the no‐harm agreement and potential discharge were viewed as the lesser evil. As individuals with experience of inpatient care have described constant observation and coercive measures as having negative effects on their well‐being and attitude towards the healthcare system (Breeze & Repper, 1998; Cardell & Rogers Pitula, 1999; Ejneborn Looi et al., 2015) and staff have described that surveillance and control measures can contribute to self‐harm escalation (Lindkvist et al., 2019) it is understandable that care providers would want to avoid these types of measures. It is, however, possible that individuals who are unable to refrain from self‐harm might be impacted negatively from repeatedly being discharged from the ward.

A central question, which was implicitly touched upon by some of the participants, is whether the no‐harm agreement is an ethically viable method in the context of inpatient care. There are several risks associated with its use, for example individuals potentially hiding their self‐harm behaviour from the staff or perceiving the discharge as a punishment and attempting or completing suicide when discharged. It stands to reason that the best way to mitigate such risks would be to ensure that the individual understands and agrees with the reasoning for the no‐harm agreement. This would require an open and honest conversation with the physician, but some of the participants questioned whether individuals would feel comfortable expressing dissenting opinions due to the inherent power imbalance. This issue can be said to apply to most forms of self‐harm abstinence agreements and really any agreement between a physician and a patient who is, or perceives themselves to be, at their mercy.

Limitations

The present study only examines care providers’ experience of using no‐harm agreements and does not allow for any conclusions regarding the effects of that method. Further, the study uses a small sample size, where all participants also work within the same healthcare organization. Hence, it is difficult to know if the method would be implemented and experienced similarly in other settings. Finally, although both positive and negative experiences were mentioned by all participants, it is unknown if the experiences described are representative for other healthcare professionals, even in the studied organization. The decision to participate in the present study may be confounded with having a specific set of experiences related to no‐harm agreements or views regarding self‐harm or psychiatric care in general.

CONCLUSION

The use of no‐harm agreements can be considered a form of positive risk taking since it involves accepting short‐term risks in the hope of achieving a positive outcome, or at least avoiding a destructive one, in the long term (Coyle et al., 2018). Using this method in psychiatric inpatient care appears to have a number of positive consequences, as experienced by the healthcare providers, but also entails difficulties, including practical challenges and ethical dilemmas and potential risks. Whether the risks associated with the no‐harm agreement are worth taking needs to be explored through further study, and care providers must keep an ongoing ethical discussion regarding all aspects of caregiving for these individuals (Strand et al., 2020). Since variations of no‐harm agreements are already used in a number of psychiatric clinics it is important that the positive and negative effects of using such agreements are evaluated.

CONFLICT OF INTEREST

All authors declare no conflicts of interest.

AUTHORS’ CONTRIBUTIONS

All three authors have contributed significantly to this manuscript according to the four criteria of authorship (contribution to conception, design, data collection and analyses; drafting of manuscript; given final approval; and agreed to be accountable for all aspects of the work).

ETHICAL APPROVAL

Ethical approval for the study was obtained from the regional Ethics board for Lund University in 2018.