Background: International and Australian studies have highlighted workplace violence as an occupational hazard. The extent of violence in Australian psychiatric settings is uncertain because of the difficulty in determining the number and severity of assaults due to under-reporting, varying definitions of assaults, and failure to differentiate the nature of assaults. Many Australian studies have been conducted on this sensitive topic but these studies focused on problems of patients and staff. No study has been undertaken that has approached this topic from a multifaceted public health approach in Australia. The public health approach to violence is relatively new but public health strategies can help identify effective approaches for prevention of violence.
Aims: The overall aim of this study is to investigate the prevalence, nature and impacts of patient aggression, as well as the factors that are associated with the occurrence and impact of such incidents on staff in psychiatric settings.
The thesis will be structured around the Haddon Matrix. This tool has been used to develop many preventative injury programs. Applying the Haddon Matrix as a theoretical framework for the thesis will provide a better understanding of this workplace problem in psychiatric settings. It will address the different relevant aspects and not focus on one aspect.
Scope and methods: Three linked studies were conducted. The findings of each posed the research questions that led to subsequent studies. A retrospective prevalence study (August 1998) looked at patient and staff incident forms over two and half years. This was followed by a qualitative study (August 1999) involving focus groups with nursing staff. These focus groups looked at staff perceptions of assaultive incidents, their emotional impact and needs for support, and sought to establish a working definition of assaultive incidents in the workplace. A cross-sectional survey (October 2000) investigated relationships between staff-related factors (locus of control, coping, support sources, perceptions of work environment, demographics), patient assault (nature and recency) and psychological measures (GHQ 12; Impact of Event Scale).
The results from the three studies helped to design intervention programs that were introduced to the facility.
Results: The results of the prevalence study highlighted that an average of 15-20 staff injuries occurred due to patient assault each month. In comparison, an incident form designed by the author to address all patient assaults showed up to 500 incidents per week. Most of these were verbal assaults, which were not captured in standard incident forms. Nurses were involved in 93.6% of these incidents. Most injuries reported were minor and occurred during busy ward periods. Male patients aged 20 to 29 years and female patients aged 30 to 49 years accounted for the majority of aggressive incidents in the facility.
The qualitative study identified significant barriers to reporting patient assaults. These included documentation, effects on staff and their coping strategies following patient assault, workplace environment and training, education and support. The results provided a detailed perspective on a range of issues that needed to be explored in a larger, more representative sample.
The third study was a cross-sectional survey highlighting that the overall work environment played a significant part in the mental health status of staff. Approximately one-third of staff was identified as General Health Questionnaire (GHQ) cases. Staff who thought about the assaultive incidents that made an impact on them were approximately three times more likely to be GHQ cases. Gender, marital status, length of time working in a mental health area, type of work and lifestyle behaviours such as smoking and alcohol intake were not associated with mental health status as measured by the GHQ-12. Locus of Control (LOC) was significantly associated with Impact of Event Scale (IES) as those with a high IES score perceived events as outside their control and those with a low IES score perceived events as within their control. Those who had a high IES score were significantly more likely to endorse denial and self-blame as their coping strategy.
The response rate to the cross-sectional survey was 63.3% and compares favourably with previous studies conducted in Australia and overseas. Of those staff who responded to the survey, 87.2% were involved in incidents relating to patients. The most common incident was verbal assault but staff identified both verbal and physical as having the most impact on them.
Seventy-two percent of the respondents were on their guard after the incidents and 55% of them took it as part of their job. The majority of respondents used ' active' coping strategies or planned new strategies for similar circumstances and indicated that they accepted that the incident had happened and lived with it. For about three-quarters of staff, using self-distraction or thinking positively was one way of coping, and two-thirds used emotional or instrumental support as other coping strategies. Approximately half the staff received support from their colleagues.
The results from the three studies helped design the intervention programs that were implemented from January 2001 to December 2002. Incident monitoring over five years from 1999 to 2003 indicated a gradual fall in the rate of assaultive incidents due to patients.
Conclusions: The thesis confirmed that workplace violence relating to assaultive incidents against staff in psychiatric settings was widespread, grossly under-reported and capable of producing significant personal and economic costs. The results also highlighted a need for improved management support to staff victims of occupational violence. The structural and organisational changes that took place during the course of the study were significant. Changing the management style, ethos and culture of the hospital in addressing workplace violence played an important part. The monthly feedback to units on staff and patient incidents and the review of all incidents in this study also contributed to the reduction of aggressive incidents.