VOL: 98, ISSUE: 17, PAGE NO: 38
Valerie Munro, BSc, RGN, is mental health coordinator, HM Prison Barlinnie, GlasgowValerie Munro, BSc, RGN, is mental health coordinator, HM Prison Barlinnie, Glasgow
Nursing is arguably the most dangerous job in the UK. A survey last year found that 75% of mental health nurses were subjected to violence at work (Mulholland, 2001); overall, 85% of nurses have experienced verbal abuse or been threatened with violence (Coombes, 1998). Yet despite these statistics, nurses' reporting of violent incidents remains inconsistent (Noble et al, 1989; Pearson et al, 1986; Nelson et al, 1997).
Employers have a legal responsibility to carry out risk assessments in working areas. Where health and safety is found to be poor, the Health and Safety Executive has the powers to serve notice. If employers fail in their duty to protect staff, they can be prosecuted (Gulland, 1997).
Proactive policies can help to reduce violence at work, but if they are to succeed there must be consistent and accurate reporting of all violent incidents as they occur. This article describes a study to investigate the factors that influence a nurse's decision to report a violent incident.
Most violent behaviour in psychiatric hospitals is directed at nurses. Fottrell (1980) cited female patients as the likeliest perpetrators, although Noble et al (1989) found that the incidents they were involved in were mainly minor, with men responsible for the most serious violence.
Violent behaviour was most frequently displayed by patients diagnosed with schizophrenia or dementia (Pearson et al, 1986; Nelson et al, 1997; Fottrell, 1980). It usually occurred in the daytime (Fottrell, 1980), and the majority of incidents detailed in all the studies were minor.
Such studies rely on accurate and consistent reporting of incidents. However, accuracy in reporting violent incidents is questionable and it has been suggested that about one-fifth of violent incidents are not reported by nurses.
This may be a consequence of staff attitudes towards such incidents: the conflict between their traditional caring role and reporting violence may cause nurses to accept that a certain degree of violence is an inevitable part of their work. Others may believe that they will be blamed or considered incompetent for allowing violent incidents to occur.
The research was carried out at the Royal Edinburgh Hospital, a psychiatric teaching hospital in central Edinburgh serving a population of around 450,000. It was designed to test three specific hypotheses:
- That patient characteristics altered the likelihood of nurses reporting violent incidents;
- That the severity of the incident was a major influence;
- That the gender of the nurse subjected to the violent behaviour influenced the likelihood that he/she reported the incident.
To test the hypotheses, 173 questionnaires were sent out to nurses and health care assistants on 10 wards across the hospital: three care of the elderly wards; three wards in adult general psychiatry; the alcohol problems clinic; a unit for brain-injured patients; the intensive psychiatric care unit; and a rehabilitation ward.
The questionnaires contained nine scenarios describing violent incidents typically encountered on the wards. Each was described as being perpetrated by one of three types of patient: a confused elderly man; a depressed middle-aged woman; and a young man with schizophrenia. Each patient was implicated in three violent incidents of differing severity, in keeping with the three-point scale for assaults devised by Fottrell (1980):
- Severity one: an assault that does not result in any detectable injury;
- Severity two: an assault resulting in minor physical injuries including bruising and abrasions;
- Severity three: an assault resulting in major physical injuries including large lacerations and fractures.
The incidents were randomly placed in the questionnaire and staff were offered five options for action: verbal report to colleagues; report to duty nurse; report to duty doctor; record incident in nursing notes; or complete an incident form. Staff were also asked to indicate their gender.
Of the 173 questionnaires sent out, 88 were returned, a response rate of 51%. They showed that the greater the level of violence relating to an incident, the less likely it was to be reported verbally. However, increased levels of violence were also associated with increased reporting to the duty nurse or duty doctor, records being entered into nursing notes more frequently, and a greater likelihood of incident forms being submitted. All results were statistically significant.
The findings supported only the second hypothesis: that the severity of an incident influenced the reporting behaviour of the nurse. Neither gender nor patient characteristics had a significant effect on reporting.
The government has emphasised its determination to eradicate violence directed at nurses. Yet its targets for reducing violence on psychiatric wards have been declared 'unachievable' because of inadequate training and support from trusts (Mulholland, 2001).
This study found that increased levels of violence were associated with decreased levels of verbal reporting to colleagues at handover. There appeared to be compensation for this through increases in other forms of reporting, for example reports to the duty nurse and completion of incident forms.
However, the results highlighted the existence of some worrying anomalies in the reporting of violent incidents. One of the scenarios (described as severity three) resulted in a fracture and a laceration requiring sutures, such that the nurses involved had to leave work to receive medical attention. Although 11% of respondents said that they would have recorded a severity three incident in the nursing notes, two nurses said they would not have submitted an incident form to their managers.
The study also found that 16% of respondents did not pass on verbal reports of violent incidents to their colleagues. While this number may not be statistically significant, it indicates that potentially a large number of nurses are walking unprepared into dangerous situations. Given that there is a greater risk of violence in the immediate aftermath of a violent incident, while the patient remains volatile, a nurse approaching the patient could inadvertently - and avoidably - reignite the situation.
Passing on clear and accurate information, without making judgements, using emotive language or dramatising the incident, is a matter of self-care, care of one's colleagues and patients as espoused by the UKCC code of conduct (1992).
Given that nurses are the most likely targets of violence in psychiatric hospitals, it is disappointing that they continue to under-report violent incidents. Under-reporting makes it difficult for employers to meet their legal obligation to protect staff from danger at work, by ensuring that risk assessments and incident investigations are carried out and any necessary remedial measures taken.
There remains some element of the old-fashioned blame culture, where a nurse's competence will be called in to question if an incident occurs on their shift. To report an incident in such a culture perhaps make nurses feel that they are drawing unnecessary attention to it.
An atmosphere of understanding and support would be more conducive to encouraging nurses to report more incidents.
Recommendations for practice
The process of carrying out the study and analysing its results led to a number of recommendations that, if implemented, could encourage nurses to report violent incidents and offer support to their colleagues.
Categorise the level and nature of violent incidents and standardise the method of recording and reporting
Although every incident is unique, a standardised framework would permit comparisons and reveal trends. It would also help to clarify which incidents should be formally reported.
Develop guidelines for reporting and handling violent incidents
Guidelines should include the information needed in each type of report, such as nursing notes, a report to the duty nurse, and so on. A further useful inclusion would be a section detailing the process the report goes through once it has been submitted.
Information could be given about specific action to be taken after each incident, including a comprehensive risk assessment in the area.
Introduce debriefing and support
Time-consuming communication, mostly involving paperwork, is involved in reporting violent incidents accurately, but this is usually at a time when stress and anxiety levels are high.
Setting aside protected time for formal/informal debriefing and for staff to complete forms would therefore be helpful.
Introduce de-escalation training
Practical training for nurses in de-escalation and managing violence should be introduced on a regular and mandatory basis. It should begin in preregistration training and be maintained throughout a nurse's career. The training should include risk assessment and management, information about the employer's responsibilities, and a forum for discussing problems and queries.
Violence against nurses in psychiatric hospitals is increasing, and nursing is now one of the most dangerous jobs in the UK, according to the government. Despite this, incidents are under-reported.
The study reported here was designed to determine the influences on the reporting behaviour of nurses. Of the three hypotheses tested, only one was supported by the study results: that it is the severity of a violent incident that influences whether it is reported. Neither gender nor patient characteristics showed significant differences.
Nurses have an obligation to protect themselves and each other from violence. Reporting violent incidents to managers can ensure that measures are enacted to prevent recurrence. Nurses should also be encouraged to offer support to colleagues when incidents take place.
Some of the above recommendations may be costly and time-consuming, but implementing them would be an investment in both staff and their safety. Some NHS trusts have already put such measures in place.
Proactive policies help to safeguard nurses who must maintain a therapeutic relationship with all patients, including those who are violent. If we could remove the blame culture surrounding violent incidents, it would help nurses to realise that they do not have to accept violence as a part of their job.