Self-harm is associated with a wide range of mental health problems, including borderline personality disorder, depression, bipolar disorder, schizophrenia, and drug and alcohol use issues (Department of Health, 2012). Information from the Care Quality Commission (CQC) (2015) shows that there are geographical pockets within England with a higher incidence of emergency department (ED) admissions due to self-harm. For example, the ED in which this study took place has around double the attendance rates for self-harm than the national average. This may be due to a lack of child and adolescent mental health services supporting young people, who may then use self-harm to self-regulate.
Research indicates that the negative attitudes of nursing staff towards patients who self-harm lead to poor-quality patient care and an increased likelihood of re-harming (Gibson et al, 2019). Patient perception of stigmatisation may be felt acutely by those who present to an ED for treatment and is often evidenced in psychological research (Cleaver et al, 2014).
Guidance on best practice in nursing care for patients with self-harm injuries is offered in a clinical guideline and quality statement from the National Institute for Health and Care Excellence (NICE) (2004; 2013), which includes guidance on the treatment and management of self-harm in the ED (Box 1). It states that patients should be treated with the same care, respect and privacy as any other patient, however, there is evidence that this is not always being delivered (NICE, 2004), and guidance on the delivery of care for those who self-harm is focused on registered staff (Koning et al, 2018). However, healthcare assistants (HCAs), are just as likely to encounter patients who self-harm as health professionals, such as nurses and doctors, and also require training on caring for this cohort of patients.
Box 1.Self-harm short-term management and prevention of recurrence
Source: National Institute for Health and Care Excellence, 2004:17
| 1.4 The treatment and management of self-harm in emergency departments The emergency department provides the main services for people who self-harm. Emergency department staff should assess risk and emotional, mental and physical state quickly, and try to encourage people to stay to organise psychosocial assessment | |
| 14.1 Triage | |
| 1.4.1.1 | When an individual presents in the emergency department following an episode of self-harm, emergency department staff responsible for triage should urgently establish the likely physical risk, and the person's emotional and mental state, in an atmosphere of respect and understanding |
| 1.4.1.2 | Emergency department staff responsible for triage should take account of the underlying emotional distress, which may not be outwardly exhibited, as well as the severity of injury when making decisions about priority for treatment |
| 1.4.1.3 | Consideration should be given to introducing the Australian Mental Health Triage Scale, as it is a comprehensive assessment scale that provides an effective process for rating clinical urgency so that patients are seen in a timely manner. Do not use the Australian Mental Health Triage Scale to predict future suicide or repetition of self-harm |
| 1.4.1.4 | Triage nurses working in emergency departments should be trained in the use of mental health triage systems |
| 1.4.1.5 | All people who have self-harmed should be offered a preliminary psychosocial assessment at triage (or at the initial assessment in primary or community settings) following an act of self-harm. Assessment should determine a person's mental capacity, their willingness to remain for further (psychosocial) assessment, their level of distress and the possible presence of mental illness |
Source: National Institute for Health and Care Excellence, 2004:17
ED staff have been shown to display negative attitudes towards patients who self-harm (Conlon and O' Tuathail, 2012), with ED nurses and HCAs as likely to stigmatise people with mental ill health as the general public, according to Cleaver et al (2014). Staff have reported feeling that people with mental health issues ‘take up too much time’ or those with self-harm are just ‘attention seeking’ (Giandinoto et al, 2018).
Contemporary evidence suggests that postgraduate study and in-service training are variables for changing clinicians' attitudes towards people who self-harm (McCarthy and Gijbels, 2010; Jelinek et al, 2013; O'Connor and Glover, 2017). Further evidence comes from Gibson et al (2019), who found that the use of an education intervention can lead to more person-centred intention and behaviour.
The project described in this article comprised one Plan, Do, Study and Act (PDSA) cycle (NHS England, 2018), illustrated in Figure 1: the aim was too investigate the effects of a bespoke educational intervention on the confidence and perceived knowledge levels of nurses and healthcare assistants (HCAs) in an English district general hospital ED.
Aim
The aim of this service improvement project was to assess the current knowledge and confidence of the ED nurses and HCAs in a district general hospital when caring for patients who self-harm at baseline, immediately post and 30 days following an educational intervention. The outcome measure of an increase in knowledge or confidence in the 30-day data collection set was used as a measure of success.
Methods
An educational session, consisting of five 1-hour sessions over 5 consecutive days, was delivered in person by the author. The aim was to improve participants' knowledge and skills in:
- Identifying why people may self-harm
- Understanding the link between self-harm and suicide
- Being aware of the high rate of admission with self-harm in the local area.
A short video showing an example of poor communication with a patient about their self-harm was used to introduce the group to the themes for group discussion. These were based on the NICE (2004) guideline (Box 1), which identifies how care should be: sensitive, non-judgemental, non-discriminatory and person-centred. The concept of the 4Rs (Risk, ‘Really’ listen, Reassure and ‘Real’ help) was introduced as a template that could be used as an aide-memoire when communicating with patients. Theoretically, the session was founded on constructivism, using experiential learning to facilitate making sense of the experiences of participants (Aliakbari et al, 2015).
Three separate data collection tools were devised (pre-, immediately post and 30 days post-session). See 30-day post session questionnaire (Box 2). These tools were tested for uniformity and brevity through a pilot, with feedback from urgent care setting colleagues within the same hospital trust. Both nurses and HCAs received identical data collection tools. Data were collected using a quasi-experiment design. A Likert-style scale was employed to measure perceived knowledge and confidence. Purposive sampling of available staff allowed reflection of lived experiences for those within the specific department (Healthcare Quality Improvement Partnership, 2018).
Box 2.30-day post-session questionnaire
Key: HCA=healthcare assistant; NA=nursing associate; AP=allied professional; TAP=trainee assistant practitioner
| Please note: self-harm includes self-poisoning |
| Please circle one of the following that matches your roleNurse HCA NA AP TAP |
| In your role, how often do you care for someone with self-harm injuries or deliberate self-poisoning? (please circle one)Every shift/every week/less than a week/once a month/less than once a month |
| a) Do you feel the session has increased your knowledge when caring for people who have self--harmed? (please mark on the scale below) |
| b) Has the session increased your confidence when communicating with people who have self-harmed? (please mark on the scale below) |
| c) Has the session changed your practice? (please mark on the scale below) |
| d) Have your feelings towards people who self-harm changed? (please mark on the scale below) |
| e) Do you feel the session has improved the standard of care for patients who self-harm? (please mark on the scale below) |
| How could the session be improved?… … … … … … … … … … … … … … … … … … … …. |
| What was good about the session?… … … … … … … … … … … … … … … … … … … …. |
| Please use overleaf to make any other comments about the session or the study |
Key: HCA=healthcare assistant; NA=nursing associate; AP=allied professional; TAP=trainee assistant practitioner
Pre-session questionnaires were completed before the educational intervention was delivered and included a question on frequency of exposure to patients who had self-harmed, alongside identification of role. Post-session questionnaires were completed immediately at the end of the educational session on paper. Anonymised 30-day post-session paper-based questionnaires, which could be returned maintaining confidentiality, were made available to all participants; they included free space for responses about changes in practice and session content feedback. A low response rate to the 30-day post-session questionnaire was expected, which would challenge the generalisability using traditional approaches to research methodology. However, service improvement methodology is iterative in nature and consequently does not lend itself to traditional research methodologies (Institute for Healthcare Improvement (IHI), 2022).
Risk assessment of participant and facilitator emotional safety was undertaken: the risk was mitigated by clear explanation of discussion topics in a participant information leaflet. Ethical approval was granted by the research and development team within the trust alongside Oxford Brookes University ethics committee where the author was studying for an MSc adult nursing; this study was a component of the dissertation module.
Results
Inclusion criteria for ED staff substantively employed by the trust were divided into two subsets composed of nurses (n=10), and HCAs (n=5), the latter group including one trainee assistant practitioner. A total of 15 participants took part in the training session. Demographic data indicating role and frequency of exposure to patients who had self-harmed were collected to consider whether exposure and role had any impact on the individual's perceived knowledge and confidence. Table 1 shows the number of self-reported contacts with patients who had self-harmed.
Table 1. Number of self-reported contacts with patients who had self-harmed (n=10)
| Role | Every shift | Every week | < 1 per week | 1 per month | <1 per month |
|---|---|---|---|---|---|
| Healthcare assistant | 2 | 3 | 0 | 0 | 0 |
| Nurse | 3 | 5 | 1 | 1 | 0 |
| Total | 5 | 8 | 1 | 1 | 0 |
Both qualitative and quantitative data from all participants were analysed. Measures of success in service improvement projects of this nature are determined at its conception and tracked for efficacy on completion of each PDSA cycle (IHI, 2022). This single cycle was measured as being successful because there was a reported increase in confidence and knowledge both immediately post-session and after 30 days.
Quantitative findings
Knowledge
Respondents were asked how knowledgeable they felt in caring for patients who had self-harmed before, immediately after the session and following a time lapse of 30 days.
Nurses
Initially, 2/10 nurses had felt ‘not very’ knowledgeable, with 6/10 selecting the median point and the remainder 2/10 responding that they felt ‘very knowledgeable’. Immediately after the session, 4/10 responded that their knowledge about how to care for people who have self-harmed had remained the same, with 5/10 noting a significant improvement. One nurse was called away and did not respond post-session.
At 30 days, responses from 2 of the 3 nurses who provided feedback, reported that the session had greatly improved their knowledge when caring for a person who has self-harmed.
Healthcare assistants
Among the HCA group, before the session, 2/5 had felt ‘not very knowledgeable’ and 3/5 indicated the median score (‘neither’) when caring for people who have self-harmed. Immediately after the session, 5/5 felt that the session had greatly improved their knowledge, and the 2 HCA respondents who completed the 30-day post-session questionnaire indicated an increase in knowledge.
Confidence
Nurses
Before the session, one nurse (1/10) reported feeling ‘not very’ confident in caring for someone who has self-harmed, with 7/10 selecting the median point of ‘neither’ very confident nor not very confident, with 2/10 feeling ‘very confident’. Immediately following the session, 4 responded that their confidence had remained the same and 5 reported an increase in confidence. (One nurse was called away part-way through the session and felt unable to respond post-session.)
Of the 3 nurses who responded to the 30-day post-intervention questionnaire, 2 indicated an increase in confidence.
Healthcare assistants
Before the session, 2/5 of HCAs had felt ‘not very’ confident, with 3/5 indicating ‘neither’, indicated at the median point of the questionnaire. Immediately post-session, all 5 HCAs felt that their confidence had markedly increased. In response to the post 30-day questionnaire, the 2 HCAs who provided reflective feedback reported feeling that their knowledge had increased significantly.
Perceived changes
Respondents were asked if their feelings about people who self-harm had changed because of the session, both immediately after and reflectively. Immediately post-session, 3/10 nurses reported that their feelings had changed towards people who self-harmed. At 30 days, the 2 of the 3 nurses who completed the follow-up questionnaire reported that they felt differently and also reported that they now felt that the educational intervention had improved their practice when caring for a person with self-harm injuries. All 3 nurse respondents to the 30-day questionnaire felt that the intervention had improved their practice when caring for a person with self-harm injuries.
Immediately after the session, 3/5 the HCAs reported a change in their feelings towards someone who self-harmed. After 30 days, the 2 HCAs who provided feedback reported a change in feelings and also responded affirmatively to the question: ‘Has the session improved your practice?’
Qualitative findings for both roles
Qualitative data was coded by hand using common language to identify four themes: attitude, environmental, patient factors and education, providing coherent reflections by respondents (Aveyard et al, 2016).
Attitude to caring for a person who has self-harmed
Consistent themes were identified through the qualitative responses from staff in both roles. Emotions expressed were mixed: adjectives such as ‘empathetic’ and ‘sympathetic’ were most frequently cited, however, ‘frustration’ and ‘confusion’ were also used. This mixture of antipathy and empathy are reflected in contemporary meta-analysis (Rayner et al, 2019).
Barriers to patient care
Frequent indicators of a knowledge shortfall were expressed by respondents in both roles, with responses such as ‘not equipped to help’, ‘lacking knowledge’ and ‘unsure how to engage’ noted.
Efficacy of session
Three participants particularly enjoyed the inclusive nature of the session. Respondents also reported that local information was of particular use and interest. Staff in both roles expressed that they would like further information about ‘how to talk about mental health’ and more training.
This request reflects the need for distress tolerance and managing relational issues which O'Connor and Glover (2017) considered to be more beneficial to staff than factual self-harm information. The session was well received, with constructive feedback given for future sessions and PDSA cycles.
Overall performance
Immediately post-session, half of the nurses (5/10) reported feeling that both their knowledge and confidence had improved after the intervention. After 30 days 2 of the 3 nurses who completed the follow-up questionnaire continued to feel that both their knowledge and confidence had improved. Among the HCAs, knowledge and confidence were reported as improving immediately post-session by all 5 participants and at 30 days by the 2 who provided follow-up feedback.
Just over one third of respondents across both groups (nurses and HCAs, 6/15) who completed the post-session questionnaire reported that they had felt an immediate change in their feelings about people who had self-harmed. After 30 days' reflection, all 5 respondents to the follow-up questionnaire reported that they continued to feel differently towards patients who self-harmed. Staff wrote about how their lack of knowledge and confidence affected the quality of care they gave and requested further training.
Discussion
The ED is an important access route to care for patients with self-harm injury. Patients attending the ED with self-harm injuries should expect the same care and respect as those attending for any other reason (NICE, 2013).
Currently, very few hospital trusts in England offer 24-hour mental health liaison services to the EDs (National Confidential Enquiry into Patient Outcome and Death (NCEPOD), 2017), a lack of funding alongside a shortage of registered staff have been recognised as limiting factors affecting service provision (Gibson et al, 2016). The absence of specialist liaison staff necessitates that nurses and HCAs, who do not have specialist training, holistically treat patients who have self-harmed (Cullen et al, 2019). In 2013, NICE recommended that all professionals caring and treating people who self-harm should be adequately and appropriately trained.
Guidance for nurses and HCAs on communicating with those presenting at the ED with mental illness remains vague, sparse and inconsistent. The need for such training has been explicitly identified by the Mental Health Taskforce (2016), with cross-party agreement to improve the provision of mental health services in England. Unfortunately, many nurses would not have previously received formal training in mental illness; often, trust-wide training is online and focuses on the Mental Capacity Act 2005 and legal issues around the deprivation of liberty. In 2017, NCEPOD found that 11% of ED nurses and HCAs had not received basic mental health awareness training and 39% reported that they had not received training in the management of self-harm.
Although several trusts within the UK have introduced the UK Mental Health Triage system (Sands et al, 2016) in an effort to identify risk and plan care for those in acute distress, this tool has evolved from the Australian Mental Health Triage system, which is still referred to in best practice guidance from NICE and supports specialist clinicians, such as mental health nurses, rather than ED staff.
Regrettably, contemporary research shows that ED staff often display negative attitudes towards patients who self-harm (Conlon and O'Tuathail, 2012; NICE, 2013). Postgraduate study and in-service training are variables for change in attitude about people who self-harm (McCarthy and Gijbels, 2010; Gibson et al, 2019).
This small study showed an improvement in perceived knowledge and confidence at all stages for both roles; nurses reported a higher level of knowledge and confidence pre-session compared with the HCAs. This may be attributed to formal undergraduate training or role-specific training. Nurse responses were consistently distributed in a ‘bell shape’ pre-session, indicating a normative distribution that reduces the likelihood of extremes of opinion (Polit and Beck, 2018). However, reliance on self-reporting on a sensitive subject increases the possibility of socially desirable responses (Aveyard et al, 2016), particularly contextualised in the adherence of a professional registration, as is the case with nurses. Nurse respondents are bound by the Nursing and Midwifery Council standards (2018) and it could be considered that this professional and operational sense of duty may also have an influenced their responses. Assurance of anonymity through electronic responses may mitigate this effect and should be ensured in all future versions of this project.
HCAs are rarely identified within evidence as a distinct group although their input in patient care within the ED is substantial. Importantly, the participant demographics studied accurately reflect the ratio of nurses and HCAs within this ED. Frequency of exposure to patients who had self-harmed was lower for HCAs, however the modal lowest frequency response was ‘every’ week’, indicating that this is still a significant cohort of patients. All participants repeatedly indicated that they thought it was part of their role to care for people who had self-harmed, with the modal response to frequency of care interventions ‘every week’. This mirrored the nurses' responses. Interestingly, all HCAs initially indicated their perceived knowledge as being below the median line—this level of perceived lack of knowledge reflects data collected by Jelinek et al (2013), who found that ED clinicians were concerned about gaps in their knowledge when caring for patients who had self-harmed. All HCA respondents indicated the highest level of knowledge increase post-session.
Post-session data found that all respondents had the same or more knowledge after the educational intervention. Reflective data indicated an increase in knowledge 30 days after the intervention, indicating the programme had been a success, among those who responded.
Respondents were also unanimous in answering positively when asked if the standard of care had improved. There were differences between the nurses and HCAs, with HCAs responding resoundingly affirmatively, whereas nurses thought there had been ‘some’ improvement. These findings correlate with contemporary evidence that suggests that confidence is a predictor of perceived efficacy in the care of this patient cohort (Egan et al, 2012). No other intervention or policy change relating to this cohort of patients has been recently undertaken in this environment, so an improvement in knowledge and confidence is likely to be correlated with the intervention (Moen et al, 2012).
The modal response to the question about attitudes towards patients who attend for self-injury (‘How do you feel about people who self-harm?’) was ‘sympathetic’, with ‘confusion’, ‘empathy’ and ‘pity’ also being noted. This shows a difference in findings between this service improvement project and available contemporary evidence, which indicates that the attitudes of staff towards people who self-harm are generally negative (Saunders et al, 2012; Hodgson, 2016). These sentiments did not alter throughout the study.
Some 13% of respondents (Saunders et al, 2012) indicated that they felt challenged by patients who had self-harmed, with one respondent reporting feeling frustrated by ‘frequent offenders’, demonstrating what could be a lack of knowledge about individuals who attend with self-harm injuries. Feelings of anger and frustration were identified by O'Connor and Glover (2017), who also reported that staff felt ‘that manipulative’ behaviour should not be ‘rewarded’. The choice of language used by this respondent may suggest some of these feelings towards people who self-harm. Clarke et al (2014) recognised that nurses' frustration at repeat attendance for self-harm may negatively affect attitudes, leading to antipathy and pessimism when caring for a person who had self-harmed. Furthermore, should a patient refuse treatment, they may be more likely to be denied information that might help them manage the self-harm in the future, as a result of the clinicians' frustration (Masuku, 2019).
Reflectively, respondents in this study commented that they had an increased awareness and level of understanding about self-harm and an ability to more effectively collaborate with mental health specialists and doctors following the intervention. All respondents felt that their practice had improved to some extent, with the biggest self-reported increase in practice being reported by the HCA cohort.
Notably, there are two disparities between the findings of this study and the available evidence regarding patient factors such as aggression and the care environment. Respondents to this study indicated that they were less challenged by these issues than the evidence would suggest. Only one respondent mentioned the ‘unsuitability of the environment’, which is frequently cited as a restrictive factor in care for patients who self-harm, challenging privacy and confidentiality (Conlon and O'Tuathail, 2012). No comments were made regarding a lack of personal safety or aggressive behaviour of patients, suggesting that this was not the greatest barrier for this cohort of staff when caring for patients who had self-harmed.
Limitations
Service improvement theory is concerned with meeting the markers for success that are identified when completing the ‘model for improvement’. As such, the data generated may hold limited significance in traditional research methodologies.
Project implementation
Generally, it was felt that the planning and delivery of the educational intervention was well received by frontline staff. However, it was perceived that some supervisory nurses were more reticent. ED supervisory nurses work in a very dynamic environment, where many changes have been undertaken recently, change fatigue, combined with the necessary arrangement for staff availability alongside service provision could also explain this challenge (Taylor et al, 2014). These important issues alongside resource availability to enable front line staff to attend the session need to be assured before planning any future PDSA cycles. The project was limited by resources and staff availability, this is consistently an issue in front line healthcare service improvement. However, the commitment made by senior clinicians and managers has afforded the trust a foundation for further improvement.
Managers and senior clinicians were committed to the project and were perceived to be fully supportive and prepared for improvement changes, this preparedness can significantly affect the outcome of the project (Rees, 2014). This project has demonstrated an increase in knowledge and confidence through an educational intervention. Using the measures outlined in the aim, these findings qualify this project as a success and should therefore be examined for opportunities for further development.
Conclusion
ED nurses and HCAs are often called on to care for people who have self-harmed, the care they give has been shown to affect the rate of future self-harm. The opportunity to engage with and signpost patients to mental health and third sector services can be missed if nursing staff do not feel comfortable developing a therapeutic relationship with patients who have self-harmed. Nurses and HCAs are an important component in the patient journey within the ED.
Staff confidence and knowledge greatly affect practice; training and education interventions enable staff to feel more knowledgeable and confident in their communication and practice with this patient cohort.
This service improvement project initially highlighted deficits in the knowledge and confidence of ED nurses and HCAs when caring for patients who have self-harmed.
The lack of knowledge and confidence among ED nurses and HCAs can be improved through in-house educational interventions enriching care to patients who have self-harmed. Future PDSA cycles could develop bespoke training which is relevant and achievable in all EDs in England. Data gathered from patients would further inform practice through sharing the lived experience.
KEY POINTS
- Emergency department (ED) nurses and healthcare assistants (HCAs) often lack confidence when caring for a person with self-harm injuries
- Nursing attitudes affect patient outcomes and re-harming rates
- This service improvement study identified a deficiency in knowledge and confidence within staff when caring for people who self-injure
- The educational intervention succeeded in improving staff knowledge and confidence, alongside a perception of improved patient care
- Bespoke in-house training can be delivered in a cost-effective and patient-focused manner in the ED
CPD reflective questions
- What can you do to increase awareness of the NICE guidelines on the triage of people who have a self-harm injury in your department?
- What are the particular issues within your department that could be changed to improve care of patients who have self-harm injuries?
- Reflect on any gaps in your own and the wider team's knowledge about care of a person who has self–harmed. What are the opportunities to address some of these gaps?