Engagement, Responsibility, Collaboration, and Abandonment: Nurses’ Experiences of Assessing Suicide Risk in Psychiatric Inpatient Care

Abstract

Suicide risk assessment is a complex task for nurses working in psychiatric inpatient care. This study explored psychiatric inpatient nurses’ experiences of assessing suicide risk. A qualitative design was used, and 10 interviews were subjected to qualitative content analysis. Nurses described suicide risk assessments as requiring them to create caring alliances and to take responsibility. Collaborating with colleagues was another part of nurses’ experiences, as was feeling abandoned. To make the assessment safely, nurses need a combination of caring alliances, support from colleagues, clear guidelines, training and time for collegial reflection to create a supportive working climate.

Background

According to the World Health Organization (WHO, 2019), suicide is the second leading cause of death in 15- to 29-year-olds worldwide, and over 800,000 people die of suicide every year. In Sweden, the most common external cause of death is suicide, which means a person is more likely to commit suicide than to die in a car accident. In the Swedish national cause-of-death register, 396 women and 873 men died of suicide in 2019. An additional 114 women and 205 men who died without clear intention, for example, by violence and or poisoning, were suspected victims of suicide, but this could not be confirmed (National Board of Health and Welfare, 2020). According to Reutfors et al. (2010), 23% of men and 31% of women had been cared for as psychiatric inpatients in the year before suicide and 3.2% of men and 5.1% of women committed suicide during ongoing inpatient care.

In the health care sector, suicide is considered one of the most difficult risks to assess: it can be experienced as time consuming and needs to be done repeatedly during individual conversations with the patient admitted to inpatient care. It is important that nursing staff have good knowledge of the suicidal process in order to be able to perceive suicidal communications and recognize warning signs of suicidal behaviour (National Centre for Suicide Research and Prevention [NASP], 2014). Nurses working in psychiatric inpatient care are often responsible for patients with an elevated risk of suicide (Cutcliffe & Barker, 2002). Assessing a patient’s risk of suicide can be challenging for nurses who, in an effort to make the most reliable assessment, must consider several different factors such as biological vulnerability, life stress and social support along with suicidality. Strategies include listening attentively as patients discuss their concerns and discerning whether they have a suicide plan. Prior experience, intuition and collaboration with colleagues also influence the assessment (Aflague & Ferszt, 2010).

Several studies have described suicide risk assessment as a demanding task for nurses who feel responsible for another person’s life (Jansson & Graneheim, 2018; Robertson et al., 2010; Wilstrand et al., 2007). Jansson and Graneheim (2018) reported that nurses described feeling doubtful and worried about whether or not their judgement had been correct. Newer nurses found suicide risk assessment frightening and said that it was difficult to talk with patients about suicidal thoughts. In addition, assessing patients’ risk of suicide can affect nurses’ own feelings and thoughts about life and death. Experienced nurses were considered to make more reliable assessments as they grow into their role, learn from previous experience and become more secure (Jansson & Graneheim, 2018).

Providing increased evidence-based methods and training for nursing staff caring for persons with suicidal tendencies is highlighted in the national action programme for suicide prevention approved by the Swedish Parliament in 2008 (NASP, 2014). This is in line with WHO’s (2014) plan to decrease suicide by 10% overall by 2020, in which psychiatric care plays a major and important role.

Assessment of suicide risk is a common nursing task in psychiatric inpatient care and occurs daily in meetings with patients. The assessment is considered one of the most difficult in the health care sector and a demanding task for nurses. While previous research has focused on suicide risk assessments in psychiatric open care, to our knowledge few studies have shed light on nurses’ experiences of assessing suicide risk in psychiatric inpatient care in Sweden. The aim of this study, therefore, was to explore nurses’ experiences of assessing suicide risk in psychiatric inpatient care.

Method

This qualitative study was based on semi-structured individual interviews subjected to qualitative content analysis to highlight the variations, similarities, and differences in the nurses’ experiences (Graneheim et al., 2017; Graneheim & Lundman, 2004).

Context

The study was conducted in the north of Sweden at a psychiatric clinic with both open care and inpatient units. The inpatient units for adults were two general psychiatric units and one substance-use disorder unit. Adults of either sex suffering various kinds of mental ill health were admitted to inpatient care voluntarily or involuntarily. The unit doors were locked, and hospital stays varied from days to weeks. At the time of the study, the staff on the two general psychiatric units consisted of registered nurses (RNs), some with specialist training in psychiatric and mental health nursing (RPNs), enrolled nurses (ENs) in mental health and a unit manager. Physicians and psychiatrists worked in outpatient care but participated daily in rounds and meetings for assessments of patients at the unit. Other professionals could be consulted when necessary. The number of hospital beds and staffing at the two units are presented in Table 1.

Table 1. Overview of hospital beds and staffing at the wards.

	Ward 1 16 hospital beds		Ward 2 10 hospital beds
Work shift	RN/RPN	EN	RN/RPN	EN
Day^a	2	4	2	4
Evening	2	2	2	2
Night	1	2	1	2
Day/Weekend	2	2	2	1
Evening/Weekend	2	2	1	2
Night/Weekend	1	2	1	2

^a This staffing should be able to provide care including constant observations of 1 patient and/or in the event of short term sick leave for 1 staff member.

Participants

The participants were recruited through purposive sampling and staff with experience of the subject in focus were chosen. RNs and RPNs were informed about the project verbally and in writing and invited to participate if they had at least one-year experience working in psychiatric inpatient care. Twelve nurses met the inclusion criteria, and ten gave their informed consent to participate. In total, 10 interviews were conducted with five men and five women (seven RNs and three RPNs) aged 23 to 40 years (median = 29). Years employed in psychiatric inpatient care varied from 1 to 13 (median = 3.5).

Data collection

Individual interviews were conducted using a semi-structured interview guide which could be adapted to the content that emerged during the interview. Initial questions included ‘Would you like to tell us about your experiences of performing suicide risk assessments in inpatient care?’ ‘Can you describe a situation when it was difficult for you to make an assessment?” “What can be helpful?’ and ‘How does it affect you emotionally?’ Follow-up questions were asked to encourage interviewees to clarify or develop their descriptions. The interviews lasted from 34 to 51 min (median = 42). All interviews were recorded digitally and transcribed verbatim to text.

Analysis

The transcribed data were subjected to qualitative content analysis, which involves the systematic interpretation of the overt and underlying content and can be used to analyse participants’ reflections, experiences and attitudes (Graneheim et al., 2017; Graneheim & Lundman, 2004; Lindgren et al., 2020). The text was first read several times separately by the authors, discussed to get sense of it as a whole and then divided into meaning units relevant to the aim of the study. The units were then condensed to shorten the material, while retaining their content. These condensed units were coded and sorted into groups according to their variations, similarities, and differences to create subthemes and more general themes. For example, codes such as support in decision-making, sharing the burden if something happens and consulting with more experienced colleagues were grouped together, abstracted and interpreted to form the subtheme seeking support from colleagues, which was subsumed with the subtheme feeling alone in decision-making under the theme Collaborating with colleagues.

Ethical considerations

This study was conducted according to the ethical guidelines described in the Helsinki Declaration (World Medical Association, 2013). The authors received permission to conduct the interviews by the head of the psychiatric clinic and obtained participants’ informed consent. Participant were informed about the purpose and structure of the study, possible advantages and disadvantages of participation, and the voluntary nature of their participation, which they could interrupt or withdraw from at any time (Kvale & Brinkmann, 2014). To protect participants’ confidentiality, they are distinguished in the text by number instead of name, and any data that could identify them have been changed or omitted in the report.

Results

The study resulted in 11 subthemes abstracted and interpreted into four themes, as presented in Table 2.

Table 2. Overview of sub-themes and themes in the results.

Subthemes	Themes
Relying on the relationship	Creating a caring alliance
Knowing about the patient
Trusting one’s intuition
Managing a challenging task	Taking responsibility
Being afraid of making the wrong decision
Bringing work worries home
Feeling alone in decision-making	Collaborating with colleagues
Seeking support from colleagues
Lacking knowledge and training	Feeling abandoned
Lacking routines
Feeling burdened by a high workload

Creating a caring alliance

The nurses described a number of aspects relevant to the assessment such as the importance of relying on the relationship, knowing about the patient and trusting their intuition.

Relying on the relationship

The nurses described how trustful patient–nurse relationships could help patients to speak more openly about their situation and nurses to make more reliable assessments. Such relationships were greatly aided by listening to patients, trusting their statements and taking time for longer conversations. It was considered easier to ask more directly about the patient's suicidal thoughts and to perceive if the patient withheld information during the assessment. They also worried, however, about risk of getting too close to a patient:

If there is perhaps a patient that I am very close to, then maybe my feelings for that patient will play a role – “I am afraid you will die” – and then maybe the assessment will be affected by that. (5)

The nurses reflected on the importance of letting patients retain responsibility for their own situation and to have faith in their willingness to do so. Nurses also reported that in closer relationships with certain patients, they often used oral contracts in which meant the patients promised not to self-harm or attempt suicide and to tell staff if their suicidal impulses became too strong. The nurses felt that these agreements created safety for both parties and a sense of collaboration. A broken promise, however, left nurses feeling more insecure and less able to rely on their own assessments.

Knowing about the patient

During assessments, nurses asked patients about their suicidal thoughts or plans and followed up with questions based on what was revealed. Knowledge of patients’ history and background was considered an essential part of the assessment. The nurses also tried to observe how patients interacted with each other and paid attention to the emotional content in their meetings with patient. They said that patients who often told about suicide plans or made repeated suicide attempts were not always listened to with the same attitude as other patients. Some of the nurses thought that these patients were not the ones most likely to actually take their own lives, but others perceived the risk of suicide as increasing with the number of suicide attempts.

The nurses reported that sometimes the knowledge about a patient made it more difficult to make a reliable suicide risk assessment. For example, among patients with psychosis who could have a different concept of reality or among patients with emotionally unstable personality disorders and self-harm behaviour when the intention of the self-harm act sometimes was unclear. Suicidal thoughts were often expressed by these patients and the nurses considered it challenging to know when to take it seriously. There was a fear among the nurses that self-harm would result in an accidental suicide since the acts often were considered risky and powerful.

They do these kind of small attempts, which terrify you, but at the same time it is not for the purpose of suicide… but you don’t know, and suddenly everything goes wrong, because even if they only thought of self-harming, it will really be a suicide. (9)

The nurses also highlighted challenges in assessing a patient they did not know or who had difficulties with the Swedish language. Using an interpreter during the assessment came with uncertainty about the accuracy of the interpretation. Getting to know these patients was seen as difficult, since it was challenging to have daily conversations without involving an interpreter.

Those meetings are often scheduled, it is hard to have them in the moment when the patient is not feeling well or… often it is usually a set time, like, on Tuesday at half past twelve or something like that. And it becomes different from what you usually do, just go in to the room and sit down and talk. (7)

Trusting one’s intuition

The nurses considered it important to explore patients’ suicidal thoughts and consider whether they indicated an acute danger to life or if they fulfilled any other function. To various extents nurses trusted their ‘gut feelings’ (an ineffable sense that something is not right, a lump in the stomach, an undefined feeling of unease or unpleasant ‘butterflies’) to contribute to their overall image of the patient during assessments.

You are drawn to some room simply to check on the situation, and then depends on whether you perceive something in someone’s body language or tone of voice or otherwise, I don’t know. (1)

The gut feeling could affect the type of questions the nurses asked and sometimes be the decisive factor in their assessment. Nurses felt that past experience, such as having patients who hurt themselves despite assuring the nurse that they would not, could affect their gut feelings and make them more insecure. The nurses thought their intuition or gut feeling came with experience and became more reliable over time.

I think I have developed a better gut feeling, or perhaps maybe not a better gut feeling, but you trust yourself or your feelings more. It was harder in the beginning, then you immediately ordered constant observation or called the physician… Now it is easier to take it easy or trust your intuition. (6)

Taking responsibility

Taking responsibility for assessing a patient’s risk of suicide involved managing a challenging task, being afraid of making the wrong decision and taking work worries home when it was difficult to stop thinking of them.

Managing a challenging task

The nurses described often taking responsibility for assessing suicide risk even if the physician had the formal responsibility. Assessment was considered a difficult and complex task, but also an inevitable part of the work and something one had to get used to. Nurses varied in their levels of comfort with performing these assessments. They described feeling responsible for someone else’s life as demanding and the risk of being the one who had made the wrong decision as burdensome.

If you take on a personal responsibility of someone else, you evaluate if they can go out… If something happens, it is the one that has made that assessment that, yeah, has made that decision in some way. (1)

Nurses said that the most experienced nurse on the shift, in an unspoken way, had the greatest responsibility for monitoring risks of suicide, but all staff were also individually responsible. Even those with long experience, however, sometimes felt uncomfortable taking responsibility for the assessment, especially if the patient was perceived as difficult to assess. Some nurses with longer experience and more knowledge found that assessments actually became more challenging with their greater awareness of all the factors to be considered. Others, however, over time, felt more confident, more self-reliant and more courageous about asking difficult questions and making quicker assessments. It also became easier for these nurses to wait and see how the situation progressed before imposing restraints on patients who might be at risk of suicide. Other areas of discomfort, especially at the beginning of their work in mental health care, included talking about death with patients, not being sure about how to make the assessment, and questioning their own judgement. “The first time it was… I was panicking all the time… I will let you out and you will go and kill yourself.” (5) It was also sometimes difficult to cope with the idea that some people did not want to live, which could lead to frustration and a desire to tell the patient to toughen up.

It is difficult for me to see or understand those who want to die and I have a hard time accepting that people want to die and that… it is not a natural death, that is what makes it difficult to accept. (6)

Being afraid of making the wrong decision

Feelings of insecurity and being afraid of making the wrong decision were common among the nurses. They feared misreading a patient who did not disclose their suicidal thoughts or not taking seriously a patients’ genuine suicidal communications.

I become like I have mentioned, always fearful, not fearful for the assessment but afraid of making the wrong decision… And then we are back into the topic of trusting one self and trusting the patient and such. It is often a fear that … if I have done it in the right way… (9)

They also felt uncertain and worried about assessments that were rushed due to a lack of time and the possibility that important information could be missed and only emerge later, when it might be too late. Nurses also sometimes found it difficult to rely on their own judgement and questioned whether they had performed the assessment correctly. This could lead to high levels of stress and they described being ambivalent, fearing the worst, and having a knot in their stomach and gnawing anxiety until they saw the patient again. “It gets harder to handle other situations because you are so focused on whether this assessment was right or not” (4). Sometimes, for their own protection, they ordered restraining measures (e.g. constant observation of the patient) to avoid their own feelings of unease, even when they were not entirely sure that the measures were needed. In this way, they could transfer responsibility to the physician.

Nurses said that suicide attempts or acts of self-harm at the ward, especially several over a short time, could make them more anxious and stressed, but also more accurate and reflective about their own assessments, which they believed improved their judgement.

Bringing work worries home

Nurses reports of difficulty in releasing thoughts of work at the end of their shifts varied. Some described lying sleepless at night wondering whether or not an assessment had been correct.

So, how many times have you been unable to sleep when you have worked the evening shift and there has been an incident, and you feel that you have the stress of everything at once, and then you feel when you come home, “Oh my God, what have I done now… how did it go?” (8)

Having time to reflect with colleagues before going home was thought to make it easier to let go of what had happened during the shift. It was also said to become easier with time and experience. The nurses described different ways of dealing with the concerns of work, such as calling the ward to ask how things went for a patient, discussing a situation with a colleague in their spare time or reminding themselves that the staff who had taken over would manage whatever arose in their absence. Some tried consciously to pause their thoughts and push away stressful emotions, although this was difficult when they were very worried. The nurses described being able to release their concerns completely when they revisited the patient or were assured that all was well.

Collaborating with colleagues

The nurses described feeling alone in decision-making and seeking support from colleagues.

Feeling alone in decision-making

Being the only RN, usually during evening, weekend, and night shifts, felt lonely, and working with new staff or deputies in difficult situations could increase that loneliness. Nurses felt burdened by being responsible for making difficult decisions (e.g. ordering constant observation of a patient) that were not always appreciated by other staff. They sometimes felt questioned and were concerned that colleagues’ opinions could affect their decisions. Nurses said they felt stressed and uncertain when their colleagues did not support their decisions.

If we say that I decide that a patient needs constant observation… then I just think that it is so incredibly important that you get that support from your colleagues, when you were the one who was actually responsible and made the assessment. And not that someone rolls their eyes or says “But is it really necessary?”… (10)

Seeking support from colleagues

The nurses reported that they often conferred with colleagues about their assessments. Colleagues’ opinions and thoughts were perceived as significant and important to consider even though they sometimes created uncertainty. They turned to someone they trusted, to more experienced staff or to the colleague with the closest relationship with the patient. The nurses found reflecting with others over possible actions to be supportive and comforting, and they perceived that it improved their decisions. It also allowed them to share the burden in the case of an adverse incident and made a difficult duty safer and more manageable.

The little time for reflection you have when you constantly discuss, for example, with a colleague over a cup of coffee on a coffee break, in the pharmacy room when you administer medicines, it is very much worthwhile to alleviate people’s [nurses’] anxiety, to always try to be two in the assessments, so that you do not have to be alone with it. (4)

The nurses reported that when discussions with colleagues did not feel helpful enough for them to commit to a decision, they would ask for a physician’s opinion or assessment, which could reduce their worry and relieve them of the burden of responsibility. Although their accounts varied, nurses usually said they felt supported by the physicians. Physicians’ advice and assessments, however, were often based on telephone calls with the nurses, whose information could have been obtained from other staff. Nurses considered that the possibility of a physician relying upon third-hand information could impair the assessment. Consulting over the phone was considered sufficient at times, but also uncertain, and it conferred a greater responsibility onto the nurse. The nurses described feeling insecure, especially as beginners on the psychiatric ward, when the physician did not come to the ward to assess the patient, but left the responsibility of the decision to the nurse.

Feeling abandoned

The nurses described feeling abandoned by their organizations, that they felt should provide them with knowledge and training, prescribed routines, and relief from burdensomely high workloads.

Lacking knowledge and training

The workplace had offered not one of the nurses an introduction to assessing suicide, let alone any specific education or training. “It was like 'ask if they have any suicidal thoughts and suicide plans’. Like that. Period.” (2) This lack of education fostered their uncertainty and made them less confident in their assessments. The nurses stated that they had developed their own working methods over time, but still felt uncertain about whether they were doing things the right way. They learned through reflecting with colleagues or observing others’ assessments, but were not satisfied with this learning method. As new nurses, they felt as if they were ‘thrown into’ making assessments with lacking competence or training.

It’s too important, I think, just to, “Well, you’ll probably learn sooner or later”. But it’s kind of like “Here you have a drip bag and a drip hose, you get to tinker a little, you’ll probably solve it in a couple of years”. (5)

Lack of routines

Routines for assessing suicide risk were considered desirable to increase the safety of the nursing staff and provide structure for understanding and performing assessments, but they were lacking.

I do not think there are comprehensive guidelines for suicide risk assessments for nurses within inpatient care. No, there are no routines. At least no written instructions, but rather we usually say `“this is what we do here”. But there is no protocol to refer to. (3)

Routines were thought “to tone down the seriousness” (3) and make it easier for new nursing staff to ask the patient difficult questions. Nurses’ opinions differed about documentation and the over-reliance on verbal reports about patients’ suicidal communications and perceived risk of suicide. Some nurses argued that the current practice had shortcomings, while others felt that it worked well. The nurses found it difficult to document suicide risk assessments since it was unclear what the assessments should cover and there were no routines for their documentation. They considered it difficult to gain an overall image of the patient from the available documentation and thought that nursing staff sometimes did not understand the importance of clear and adequate documentation.

Feeling burdened by a high workload

The nurses considered high pressure on the ward and lack of time as major obstacles to them making adequate assessments. Not having enough time to talk to patients to understand their situation or to read any available documentation made assessments even more challenging. The nurses described that time pressure could lead to them skipping assessments, being too hasty or basing assessments on old perceptions of the patient.

You’re under stress and a patient says. “I’m going out now, at three o'clock, can you unlock the door?” Okay, I unlock the door, and in the stress do not reflect that it may pose an imminent risk for something serious, for self-harm or suicide. (3)

Discussion

This study aimed to describe nurses’ experiences of assessing suicide risk in psychiatric inpatient care. The results show that nurses’ experiences involved creating a caring alliance, taking responsibility, collaborating with colleagues and feeling abandoned.

Nurses used their caring alliances with patients in assessing their risk of suicide. They relied on these relationships, used their knowledge about the patients and trusted their intuition. Further, nurses reported the importance of allowing patients responsibility for their own situations and trusting in the patients’ stories. These results are in line with Sun et al. (2005), who highlighted the importance of an established relationship facilitating more direct questions about suicidal thoughts. Cutcliffe et al. (2007) described the prerequisites of creating a respectful relationship as having a warm approach and listening without judging, which make it easier for the patient to form a confident attachment to the nurse. Sun et al. (2006) reported that maintaining continual contact and allowing the patient responsibility, involvement and the ability to make decisions about their own care were important factors in building a trusting relationship. According to Barker and Buchanan-Barker (2005), alliance and dialogue are essential tools in the assessment and the nursing staff should assume that patients are the experts on their own problems and needs.

Taking responsibility included feeling uncomfortable asking questions and talking about death, but this became easier with experience. Unexperienced nurses felt ‘thrown into’ making assessments without having any skills, and nursing staff need both courage and experience to discuss suicidal issues with the patient. According to Sellin et al. (2018), it is important to acknowledge each unique patient’s views about suicide, to create a nurturing and caring space for suicidal patients, and to help patients put their thoughts and feelings into words. Such a recovery-oriented caring approach has the potential to facilitate a mutual understanding of the complexities of the patient’s situation and empower patients to be involved in their own care and regain authority over their own lives.

Nurses in our study relied on their intuition in their assessments, and it could sometimes be decisive in their decisions. Intuition was considered something that emerged over time or became more dependable with experience. This is also described in a study by Jansson and Graneheim (2018), where some nurses questioned the reliability of their intuition while others described it as the tool they had the most confidence in. According to Aflague and Ferszt (2010), specialist nurses often used their intuition and could know that something was not right even if they could not specify what that ‘something’ was. This was also found in our study, with intuition described as an undefined feeling of unease or an ineffable sense that something was not right. Welsh and Lyons (2001) reported that a combination of tacit and formal knowledge as well as intuition contributed to the nurse’s skills in assessments. Tacit knowledge was gained after years of experience and based on formal knowledge augmented by variations and nuances gained through clinical experience. Relying on intuition without connecting it to knowledge was considered uncertain and unprofessional.

In this study, oral promises and no-harm agreements between patients and nurses were described as common and a way to create a sense of collaboration and safety for both parties. According to Stanley and Brown (2012), it can be risky to persuade a patient’s promise not to self-harm or commit suicide as this may obscure the patient’s actual plans. Patients can withhold information about how suicidal they are for fear of letting the nursing staff down by breaking their agreement. Clinical experience shows that these no-harm agreements can reduce nurses’ worries, but there is no evidence that such agreements do in fact prevent suicide. Aflague and Ferszt (2010) reported that no-harm agreements can be confusing for the patient and that it can be difficult to assess one’s own risk of suicide act. It is also discussed if it is right to burden the patient with the extra responsibility the agreement meant or whether it could be helpful for the patient to continue struggling (Jansson & Graneheim, 2018).

Our results showed that nurses found it more challenging to assess patients with emotionally unstable personality syndromes and self-harm behaviour since they expressed suicidal thoughts and harmed themselves more often than other patients. The nurses experienced a contradiction in their own thinking: while they thought that these patients were not taken as seriously as others, they also feared that a self-harming incident intended for other purposes could result in an accidental suicide. According to Björkenstam et al. (2016), suicide risk is significantly higher in patients with personality disorders and increases when those patients enter inpatient care. Fear that patients with self-harming behaviour might take their lives can lead nurses to focus on minimizing risks instead of supporting the patient. Ejneborn-Looi et al. (2015) found that mutual and trusting relationships with patients facilitated their personal recovery and care without the use of restrictive measures.

The results showed that assessing suicide risk was difficult and complex, and nurses varied in their confidence about taking the responsibility for such decisions. Nursing staff who felt insecure about the responsibility often ordered restraining measures to ease their own worries instead of relying on their relationship with the patient. Berg et al. (2017) highlight the importance of adopting a broader perspective on the safety of suicidal patients rather than focussing solely on physically preventing them from taking their lives. Further, that attention to the suffering of suicidal patients through creating trust and being present was considered the primary way to assess the risk of suicide and increase patient safety.

Nurses in our study often felt lonely when assessing suicide risk, especially when they were the only RNs on the unit, primarily during evening, night and weekend shifts. Notably, Bowers et al. (2011) reported that suicide attempts in psychiatric inpatient care usually occur between 6:00 PM and 9:00 PM, when staffing levels are low. Robertson et al. (2010) stated that when the working group shares decisions, the responsibility for the patient’s life does not feel as personal if something unforeseen happens. Aflague and Ferszt (2010) also highlighted the importance of reflecting with colleagues, partly to confirm one’s own assessment but also to discover whether more information is needed.

Nurses in our study considered high workloads the major obstacle to their ability to conduct adequate assessments. Berglund et al. (2016) reported that patients want more opportunities to talk about their troublesome thoughts, but are often being left alone because nurses lack the time to listen to them. Molin et al. (2016) reported that when patients felt their interactions with staff were insufficient or absent, they felt either they were invisible to staff or the staff were invisible, which led to feelings of being ignored or not involved in decisions about their own care. This is in line with Berg et al. (2017), who report that when nurses spend little time with patients, prioritize other duties, or are interrupted during conversations, patients feel lonely and insecure. Feelings of hopelessness or being worthless can then arise and lead to patients considering how they might end their lives on the ward. Graneheim et al. (2014) showed that nurses acknowledge dialogue with patients as an important task, but that the reality of psychiatric inpatient care prevents them from having these conversations, which results in an unsatisfactory work situation and feelings of insufficiency.

Finally, the results showed that a recurring issue for the nurses was their lack of knowledge and training. None had received any specific training in performing suicide risk assessment. Similar results were shown by Awenat et al. (2017), who reported that staff in psychiatric inpatient care felt inadequately equipped to deal with patients’ suicidal behaviours. Aflague and Ferszt (2010) argued that nurses should receive more training in assessing suicide risk and that employers should ensure that nursing staff have sufficient education to carry out their work according to current guidelines. Several other studies have highlighted the importance of adequate training for nurses working in psychiatric care with suicidal patients (Brunero et al., 2008; Samuelsson & Åsberg, 2002; Sun et al., 2007).

Methodological discussion

The selection of participants who were colleagues that the authors had previously worked with in psychiatric inpatient care might be seen as a limitation of this study. The advantages and disadvantages of interviewing colleagues and the issue of insider versus outsider perspective have been discussed by other researchers. McEvoy (2001) argued that shared experiences may help to develop and deepen the issue under study, and McDermid et al. (2014) warned that researchers must be aware of their dual roles and make thoughtful decisions about how to best manage them. Our experience was that our established relationships were helpful in creating a safe and comfortable environment for the participants.

Researchers’ preunderstandings of the subject are another important consideration. We reflected on our own preunderstandings from the inception to the reporting of the study, but the possibility that we were unconsciously influenced by previous experience or knowledge cannot be completely excluded. Another possible limitation is the single setting of only one psychiatric clinic, as nurses’ experiences of assessing suicide risk may vary among different contexts. However, qualitative content analysis aims to highlight variations, similarities, and differences in experiences and perceptions of the phenomenon, and the nurses in this study were rather diverse in terms of gender, age, experiences and education, which might have provided sufficient variations in their experiences. Readers may assess themselves whether the results are transferrable to their own organizations.

Conclusions

Assessing risks of suicide in psychiatric inpatient care has a significant impact on nurses’ feelings and behaviours. Nurses rely on caring alliances with patients and the support of colleagues to cope with risk assessments, but lack and need knowledge and training. The combination of caring alliances, support from colleagues, clear guidelines and structured training is a key to nurses’ ability to safely assess the risk of suicide in inpatient psychiatric care. Considering that, it is the management responsibility to offer recurrent training about the assessment of suicidal patients to all nurses and make sure it is included in the introduction of new employees. This responsibility also includes structuring time for nurses to reflect with colleagues to create a safer care for patients who are suicidal and a more supportive working climate. RPNs, with their specific competence, are in a central position to lead such specific reflections and further, to advocate suicide risk assessments that include guidelines in combination with a relational approach, in line with mental health nursing.

Author’s contributions

KD, BML, AJ and JM contributed to study design; KD, BML, AJ and JM contributed to data collection; analysis was carried out by KD, BML, AJ and JM; manuscript was prepared by KD, BML, AJ and JM. All authors are in agreement with the manuscript.

Acknowledgement

We thank the participants who generously shared their experiences.

Disclosure statement

The authors report no conflict of interest.