Levels of self-harm in the UK and Ireland are increasing, particularly in adolescents (Brunner et al, 2007; Karman et al, 2015; Griffin et al, 2018; Heyward-Chaplin et al, 2018), with more than 200 000 presentations to accident and emergency (A&E) departments for self-harm annually (Hawton et al, 2007). This has become a considerable problem for people working in healthcare settings and is exacerbated by the bed crisis.
Individuals use a range of methods to harm themselves intentionally. These can include cutting-type behaviours (including scratching and poking), hitting, banging or biting type behaviours, burning, overdosing, eating-disordered behaviour, reckless behaviour or bone-breaking (Laye-Gindhu and Schonert-Reichl, 2005; Victor and Klonsky, 2014).
As with mental illness in general, stigma around self-harm has been found in the healthcare professions (Storey et al, 2005; Friedman et al, 2006; McHale and Felton, 2010; Karman et al, 2015; Mitten et al, 2016), with research suggesting that some staff believe patients who self-harm are ‘manipulative and attention-seeking’ (Friedman et al, 2006).
Staff in both healthcare and educational settings have reported feeling that they would benefit from education to improve their knowledge of self-harm and to develop effective methods of coping with it in their job roles (Timson et al, 2012). A lack of education on self-harm has also been suggested as a reason for the stigma around it (Emerson, 2010; McHale and Felton, 2010; Karman et al, 2015).
Patients who attended A&E for care following a self-harm incident have reported experiencing negative attitudes from staff (Friedman et al, 2006; Karman et al 2015), and a notable difference in the way they are treated compared with other patients in the hospital (Horrocks et al, 2005). These patients have reported experiencing hostility, being made to feel as though they were wasting staff time and receiving ‘punitive’ treatment when presenting at A&E (Harris, 2000; Clarke et al, 2014; Owens et al, 2016).
So far, only minimal research has been carried out with the intention of reducing self-harm stigma among those in the medical and nursing professions. However, Patterson et al (2007a) delivered an educational intervention to mental health nurses, with the aim of reducing negative attitudes towards self-harm. The intervention was successful in terms of a 20% reduction in antipathy from the nurses towards patients who self-harm, which was maintained for at least an 18-month period. This kind of intervention appears to be successful in improving nurses' attitudes around self-harm, and could be applied to other disciplines and in other settings to improve attitudes and reduce the stigma around this. However, the length of this training intervention, which lasts several months, means it cannot provide a model for services where self-harm is not a regular feature. A shorter intervention is needed.
Aims
This study had two main aims. First, it intended to ascertain whether a short, one-off educational intervention into self-harm could reduce stigma and negative attitudes towards people who self-harm by increasing knowledge, and improving understanding. Its second was to research whether attitudes towards mental illness in general predicted attitudes and stigma specifically in relation to self-harm. Factors that might affect attitudes such as age and experience of self-harm were also examined.
Hypotheses
Four hypotheses were tested:
Methods
The study received ethical approval from the psychology department at a higher education institution, based on ethical guidelines from the British Psychological Society (2018). Consent was obtained from all participants after they had been given an information sheet detailing the requirements of the study.
A sample population of 80 adult nursing students from the higher education institution took part in the intervention. After taking missing data into account, the final sample who completed the full intervention was 55.
Participants were aged 18–53 years. The majority (97.5%) were female.
The students were given a copy of the Self-Harm Antipathy Scale (SHAS; Patterson et al, 2007b), and the MHAS (Time to Change, 2015) to complete to provide a baseline score for their attitudes towards self-harm and towards mental illness in general. Because of space considerations, the findings from this scale are not discussed at length, except with respect to hypothesis 2. The participants were asked demographic questions about sex and age, and if they had any friends or relatives they knew of who engaged in self-harm behaviours.
Self-Harm Antipathy Scale
The 30-item SHAS (Patterson et al, 2007b) was used to measure the attitudes of the student nurses in relation to self-harm before and after they received the educational intervention to determine any attitude change. The SHAS has six subscales: competence appraisal, care futility, client intent manipulation, acceptance and understanding, rights and responsibilities, and needs function.
Items on the inventory were rated on a seven-point Likert scale from ‘strongly agree’ to ‘strongly disagree’. The lower the overall score on this measure, the more positive the attitudes around self-harm, with strongly agree scored as a 1 and strongly disagree scored as a 7. Twenty-three items contribute to the six different factor scores. Seven items were not correlated with any of the six factors, but all 30 items are used to compute the total score on the scale.
The Cronbach's alpha scores from the present study, alongside those from the original Patterson et al (2007b), shown in parentheses, were: competence appraisal 0.78 (0.81); care futility 0.62 (0.79); client intent manipulation 0.82 (0.71); acceptance and understanding 0.39 (0.61); rights and responsibilities 0.76 (0.67); and needs function 0.52 (0.52). The total SHAS Cronbach's alpha score was 0.84 (0.89).
Mental Health Attitude Scale
The MHAS was also used before the intervention took place to assess attitudes towards mental illness in general (Time to Change, 2015). (Findings from this scale are omitted from this article with the exception of hypothesis 2.)
Educational intervention
In developing the intervention, it was important to include facts as well as personal and celebrity stories about self-harm. Both written extracts and video clips were used to reduce monotony and keep participants interested.
The intervention was kept as short as possible and unnecessary jargon was not included to minimise boredom. Facts were easily compiled, but the personal and celebrity stories were given more consideration. It was felt that a variety of celebrities, in a range of ages, who would appeal to different audiences should be included in the intervention to interest as many participants as possible, and highlight how anyone can self-harm, not just the stereotype of teenage girls.
Numerous celebrities were considered to appeal to younger students, including Johnny Depp, Russell Brand and Lindsay Lohan, but it was felt that the participants might be more sympathetic towards Demi Lovato who has less of a ‘bad girl’ image. There was also more information available regarding Lovato's self-harm to use in the intervention and for the students to look for if they wanted to carry out further research than there was for the other celebrities.
Princess Diana was chosen as the other celebrity to use in the intervention as she has almost mythical status among the British media and the general public. Many people idolised her and still do after her death. It was hoped that her story would appeal more to older participants. It was assumed that, while it was general knowledge that Princess Diana had difficulties with her marital and subsequent personal relationships, her self-harming behaviour would be less well known, and this might lead to a more sympathetic response from the participants, and open their eyes further to the different types of people who are affected by self-harm.
The personal stories and clips were chosen to outline the different methods and reasons people have for self-harming. It was important to outline a variety of reasons for self-harm, and to challenge stigma that self-harm always involves cutting. It was important that these stories demonstrated the real emotive reasons for such behaviour, and dispelled any myths about attention seeking.
The information used in the intervention was gathered from Google searches, websites such as Mind and Google Scholar, and Andrew Morton's biography of Princess Diana.
The intervention took approximately 45 minutes, and included written information and video clips. Once the intervention was over, the participants completed the SHAS (Patterson et al, 2007a) for a second time to observe if the intervention had any effect on attitudes towards self-harm.
Findings
Hypothesis 1. Student participants will have more positive attitudes towards self-harming behaviour after taking part in the educational intervention
The mean SHAS score before the intervention was 79.39 (SD=18.00, compared with 68.07 (SD=16.68) after it. This was a significant difference (Wilcoxon matched pairs statistic z=–5.303, P=0.001). Table 1 gives the SHAS factor scores before and after the training.
| Subscale | Score before | Score after | Significance |
|---|---|---|---|
| Competence appraisal | 14.50 | 12.69 | z=–2.928, P=0.003 |
| Care futility | 12.01 | 11.38 | z=–1.106, P=0.269 |
| Client intent manipulation | 10.57 | 8.90 | z=–3.209, P=0.001 |
| Acceptance and understanding | 7.82 | 6.63 | z=–3.627, P=0.001 |
| Rights and responsibilities | 9.76 | 8.73 | z=–3.635, P=0.001 |
| Needs function | 6.78 | 4.42 | z=–5.897, P=0.001 |
| Total SHAS score* | 79.39 | 68.07 | z=–5.303, P=0.001 |
This brief self-harm educational intervention led to significant improvements in the attitudes of student nurses as shown by SHAS scores. On the six subscales of the SHAS, there were significant improvements (i.e. reductions in subscale scores) on five of the six subscales. This hypothesis is confirmed.
Hypothesis 2. Participants with more positive attitudes toward mental health, as assessed by the MHAS, will be more positive in their attitudes towards self-harm
Participants were divided into positive and negative attitudes groups on the basis of a median split in their MHAS scores. The groups' SHAS scores before and after the intervention were then compared. The Mann Whitney test was used to compare the groups (Table 2).
| SHAS subscale | Before the intervention | Before the intervention | Significance | After the intervention | After the intervention | Significance |
|---|---|---|---|---|---|---|
| Positive attitudes (n=32) | Negative attitudes (n=33) | Positive attitudes (n=31) | Negative attitudes (n=32) | |||
| Competence appraisal | 12.53 | 16.28 | z=–3.177, P=0.001 | 11.00 | 14.21 | z=–2.750, P=0.006 |
| Care futility | 10.42 | 14.66 | z=–3.970, P=0.001 | 9.27 | 13.64 | z=–4.329, P=0.001 |
| Client intent manipulation | 8.62 | 12.39 | z=–3.316, P=0.001 | 7.06 | 10.53 | z=–3.173, P=0.002 |
| Acceptance and understanding | 6.79 | 9.28 | z=–3.762, P=0.001 | 5.32 | 7.89 | z=–3.856, P=0.001 |
| Rights and responsibilities | 8.94 | 11.18 | z=–2.677, P=0.001 | 8.48 | 9.48 | z=–1.088, P=0.277 |
| Needs function | 6.26 | 7.55 | z=–2.075, P=0.038 | 4.24 | 4.96 | z=–1.313, P=0.189 |
| Total SHAS score* | 69.59 | 92.25 | z=–5.121, P=0.001 | 60.08 | 77.58 | z=–3.673, P=0.001 |
Before the intervention, the two groups differed significantly on all six subscales and on total SHAS scores. Scores dropped for both groups after the intervention, showing it had changed attitudes. After the training, there were no significant differences between two of the subscales: rights and responsibilities; and needs function. Total scores were significantly different before and after the intervention between students who had positive attitudes towards people with mental illness and those who had more negative attitudes towards them. This hypothesis is confirmed.
Hypothesis 3: Older students will have more negative attitudes towards self-harm than those in the younger age category. Before the intervention, older students will score significantly higher on the SHAS than younger students
There were no significant differences between older and younger students on the SHAS. This hypothesis is rejected.
Hypothesis 4: Participants who have a friend or relative who has self-harmed will be more positive in their attitudes towards self-harm
Of the seven comparisons (six subscales and total score), six were significant. Therefore, if a participant had a friend or relative who had self-harmed, they were much more likely to have positive attitudes towards people who self-harm (Table 3). This hypothesis is accepted.
| Subscale | Previous personal exposure to self-harm (n=18) | No previous exposure to self-harm (n=41) | Significance |
|---|---|---|---|
| Competence appraisal | 11.32 | 15.52 | z=–3.186, P=0.001 |
| Care futility | 9.26 | 13.36 | z=–3.736, P=0.001 |
| Client intent manipulation | 8.05 | 11.90 | z=–3.249, P=0.001 |
| Acceptance and understanding | 6.33 | 8.52 | z=–3.141, P=0.002 |
| Rights and responsibilities | 8.30 | 10.53 | z=–2.513, P=0.012 |
| Needs function | 5.78 | 7.21 | z=–2.213, P=0.057 |
| Total SHAS score* | 65.94 | 85.32 | z=–4.019, P=0.001 |
Discussion
The results of additional questions given to the participants at the end of the intervention showed that only 51% of them felt that their attitudes towards self-harm changed as a result of the intervention, despite 79% finding the presentation informative.
Summary of main findings
There are a number of findings. The main one is that a short, educational intervention had a significant effect on changing attitudes towards people who self-harm among student nurses. After receiving the educational intervention, they demonstrated an increase in positive attitudes towards self-harming behaviour, and a reduction in negative attitudes compared with their attitudes before they took part in the intervention. This suggests that the intervention was successful in its aim to educate and change negative attitudes towards self-harm, therefore reducing stigma.
It is interesting to speculate why this intervention was as successful as it was. It could be the combination of celebrity stories, personal stories and facts that were included to provide a range of knowledge and understanding in the area. Rather than simply telling participants numerous facts, the personal and celebrity stories could have put a more emotive twist on the intervention, and led to deeper thought and more internalisation of the content. The use of video clips as well as written accounts may also have contributed to the success of the intervention in minimising boredom. The fact that the intervention was short could be a factor in its success, as the participants may not have experienced boredom as they might have in a longer intervention.
Furthermore, the two celebrities who were used as examples in this intervention were very different and were intended to appeal to students of different ages. It was thought that Princess Diana would appeal to the older students, while Demi Lovato would appeal to younger participants. It also demonstrated that people from a range of backgrounds can experience self-harm, with Princess Diana—the ‘people's princess'—engaging in self-harm as well as Demi Lovato, a contemporary pop star, who appeals to a different audience. It could be that the Princess Diana example resulted in a more sympathetic response from the students, as many did not know about her self-harm, and this could have influenced the results of the study.
Before receiving the intervention, the study found students who held more positive attitudes around mental illness in general held more positive attitudes specifically regarding self-harm, and those with more negative attitudes about mental illness in general held more negative attitudes around self-harm specifically, as was hypothesised. This is to be expected, as self-harm is a form of mental illness and those who are more positive in their attitudes around mental illness are likely to be more positive towards self-harming behaviour also.
An additional main finding of the research was that participants who knew a friend or relative who had self-harmed had more positive attitudes overall towards the behaviour than those who did not. This was expected, as it could be that such participants have already researched self-harm in an effort to figure out their relative or friend's behaviour and, as a result, understand it better. They are also more likely to be sympathetic when they know of a person they care about engaging in the behaviour.
The SHAS (Patterson et al, 2007b) has subscales that measure different aspects of attitudes towards self-harm. The biggest attitude change was found in the needs function subscale, with the participants initially having a mean score of 6.81 and after the intervention having a mean score of 4.47. The subscale with the least change was care futility, where the mean score before the intervention was 12.10, compared with a score of only 11.50 afterwards.
Comparison with previous research
Educational interventions focused on changing attitudes towards self-harm have received minimal research but no shortage of advice (e.g. Mendes, 2015). The current study was based on a previous study by Patterson et al (2007a), which demonstrated the effectiveness of an educational intervention about self-harm for qualified health professionals. However, there are numerous differences in the two pieces of research.
The Patterson et al study (2007a) differs from the current study in that the intervention took place over a period of 15 weeks, and included 12 study days, while the current study lasted only 45 minutes and was an isolated session. The present study also researched adult nursing students rather than qualified mental health professionals as in the Patterson et al (2007a) study. However, the main results of both studies were similar, in that they were both effective in improving negative attitudes towards self-harm, and resulted in participants developing more positive attitudes. The results of these pieces of research suggest that both long-and short-term interventions can have significant effects in changing negative attitudes towards self-harming. Whether this attitude change persists in the long term requires further investigation.
The results of the competence appraisal subscale differed considerably between the Patterson et al (2007a) study and the current study, with the mean score of the previous study being 25.20, and the mean score of the present study being 12.80. This difference is to be expected, however, as student nurses took part in the current research, and would have less experience in nursing patients who have self-harmed and would likely feel less competent than the registered mental health nurses who took part in Patterson et al's (2007a) study. Mental health nurses would also have had more experience in dealing with self-harming patients and would feel more able to answer those questions accurately because of their increased exposure to such patients.
The overall attitude change in the current study was greater than the overall attitude change in the Patterson et al (2007b) study. This could be because of the combination of personal and celebrity stories that were used. This combination of emotive and informative content could have resulted in the student nurses engaging more with the intervention and considering the information given to a higher degree.
The reduction in negative attitudes and therefore stigma through education, as this research demonstrates, correlates with previous educational campaigns such as Time to Change, which succeeded in educating and improving attitudes towards mental illness (Evans-Lacko et al, 2013). Other research (Ben-Zeev et al, 2010; Barber, 2012) has also suggested that increased education on mental illness would be effective in reducing stigma, and this research supports that idea.
As the intervention was based on an individual aspect of mental health (self-harm), the study also further supports research by Reavley and Jorm (2011), who proposed that interventions that focused on individual disorders would be more effective than those that concentrated on mental health in general.
Suggestions for future research
Future research could further explore the attitudes of nursing students towards self-harm, and continue to work towards changing such attitudes. Although the attitudes and attitude change of both registered and student nurses have been analysed in this study and the Patterson et al (2007a) study, a gap in the literature remains regarding the difference in the attitudes of general nursing students compared with mental health nursing students.
It would also be interesting to carry out this intervention with adult nurses who have the most contact with people who self-harm, in particular A&E staff. The bed crisis and having to deal with patients with urgent physical health problems could also mean A&E staff have different attitudes from the sample in the current study.
Another suggestion for future research would be to repeat the intervention without the celebrity stories. The use of only ordinary people could remove a potential source of bias in the research. That is, participants may have reacted positively only because of the status of the celebrities used. This might result in different findings.
A comparison of the Patterson et al (2007a) and the current research into the attitudes of the participants at different follow-up dates could also be considered to determine whether the length of the intervention influences how long the attitude change lasts. It could be that while both the short- and the longer-term interventions had significant effects immediately, the longer intervention may have longer-lasting effects on attitude change than the shorter intervention.
Implications of findings
The main implication of this research is that it is possible to address and break down stigma in relation to self-harm. The findings suggest that educational interventions aimed at changing attitudes towards self-harm are effective in making attitudes more positive.
The findings also suggest that an intervention does not have to be a lengthy one, and that even small-scale, short interventions can be effective in educating and improving attitudes towards self-harm. Therefore, it could be assumed that such interventions would be beneficial as a routine part of the theory aspect of nursing courses, to promote more positive attitudes towards self-harming patients in a general hospital setting.
Such an intervention could also benefit nurses and other qualified clinical staff to assist in making attitudes towards self-harm consistent across practitioners. In turn, it would be assumed that increased understanding and training in relation to self-harm would reduce stigma, and this would encourage those who self-harm to seek help.
Conclusions
Even a short, 45-minute educational intervention into self-harm can change the attitudes of nursing students. Students who had positive attitudes to mental health in general were also positive in their attitudes towards self-harming behaviour specifically.
This research concurs with previous work by Patterson et al (2007a) who found educational interventions effective in changing attitudes towards self-harm.
This research is novel in that it demonstrates a short intervention can achieve similar results to the 15-week intervention by Patterson et al (2007a), and it also provides data on the attitudes of adult nursing students as opposed to registered mental health nursing staff.
Longitudinal studies will help determine whether the attitude change is long-lasting. The SHAS could be used in interviews for those applying to work with patients who self-harm.
This study demonstrates that a small, simple intervention such as this can change attitudes. Small interventions can help shine a light on ‘the dark matter of stigma’ (Smith, 2013) as much as high-cost campaigns such as Time to Change.