Abstract
International and national statistics highlight the growing significance and magnitude of the problem of workplace violence [1–3]. Workplace violence includes the following types of behaviours: homicide; physical violence; verbal abuse and threats; stalking; bullying among workers or from managers; or behaviours that induce fear, stress or workplace avoidance [4]. In the healthcare sector violence initiated by clients is increasing, with approximately one Australian health worker being a victim of homicide by a client each year [4]. It has been estimated that 13.7% of inpatients admitted to psychiatric facilities are aggressive on the ward, with 47.4% of injuries occurring to staff [5]. Victims of violence not only face immediate trauma from physical assault, but frequent exposure can lead to cumulative effects such as the development of posttraumatic stress and substance abuse disorders [3], [6], [7]. The specific emotional effects staff have reported following aggressive incidents have included anger, anxiety, helplessness, irritability, feelings of resignation, sadness, depression, shock, apathy, disbelief, self-blame, sleep disturbance, headaches, body tension and soreness, fear of patients, fear of returning to the scene of the assault and difficulty returning to work [8]. Individuals who witness violence also suffer similar emotional reactions [9]; the impact on other patients can be significant [10]. In the past, violence and aggression was often considered to be an inevitable aspect of a chosen profession [11], although now, new ‘zero tolerance’ of aggression and violence policies are being introduced in health services both in Australia [12] and overseas [13].
Responding effectively to aggression and violence requires multiple strategies, including attending to workplace design, policies and practices, staffing levels and mixture and training [1–4]. Staff training has been widely recommended, however as delivered, it is often inadequate and ad hoc in nature and there is an urgent need for published evidence-based evaluation of programs [14–18]. A survey of 839 staff from acute mental health inpatient units in the UK found that training courses varied considerably in length and content and failed to identify the specific behaviours (competencies) that are required to ensure aggression and violence are managed adequately [19]. There is also a need for comprehensive competency-based training [20] that educates staff about how they can actively participate in preventing as well as managing aggression and violence [21–23]. Several prominent cases of assault in the New South Wales (NSW), Australia healthcare setting led to the recognition that uniform aggression and violence minimization training was a high priority for the health service.
A NSW Health Department Violence Taskforce was established and allocated funding to develop a comprehensive aggression and violence minimization training program for healthcare staff. This report outlines the results of the development and pilot-testing of the program. The development of this program followed an extensive 12-month consultation process, involving the compilation of detailed feedback on three widely disseminated drafts at a project contract steering committee (see Acknowledgements). This committee took submissions from the NSW Health Learning and Development Forum and key stakeholders from the following fields: psychiatry and emergency services; nursing; medicine; occupational health and safety; security; ambulance; social work; psychology; aged care; brain injury; dental; midwifery; and early childhood. Individual feedback from senior aggression minimization trainers and consumer representatives was also collected. The basis for the first draft of the program was an existing accredited program developed on the Central Coast of NSW, with additions from the ‘INTACT’ program developed in the Illawarra region of NSW.
The final published program is now being implemented across all NSW health areas and services [24]. Key objectives of the program included the need for a comprehensive package focusing on risk assessment, prevention, management and a ‘zero tolerance’ approach to workplace violence. Further objectives were that the package be applicable to all staff at risk of violence and contain flexible delivery options that allowed practical and achievable timeframes for the competency-based training of healthcare staff. Full-day modules were designed to be broken into half-day or other flexible delivery components. Facilitator manuals were designed with the dual purpose of aiding trainer effectiveness by simplifying the trainer's task and secondarily, increasing the comprehensiveness of the content by providing greater explanations of concepts. Training materials included PowerPoint slides, detailed answers to trainerled questions, instruction in skills-based exercises and additional training and web-site resources. Comprehensive participant manuals were developed to both act as a workbook during training and as a resource for future reference. Ongoing reinforcement and development of skills and knowledge was catered for through a refresher module (module 4). Accreditation of the program was achieved through articulation with competencies from the national Health Services Training Package. Recognition of prior learning was also articulated, as previous research highlights the different needs of previously trained and untrained staff [25].
Table 1 shows the modular structure and learning objectives of the program developed [24]. The training was evaluated in a ‘train the trainer’ pilot phase with experienced aggression and violence minimization trainers sourced from across NSW and the program modules were piloted on experienced healthcare managers, clinical staff and support workers independently drawn from one representative region of NSW.
Modular structure, content and learning outcomes of the of the ‘A safer place to work – preventing and managing violent behaviour in the health workplace’ program
Method
‘Train the trainer’ workshop
Participants
Participants comprised 15 experienced healthcare trainers (9 male, 6 female; mean age = 45.1, SD = 9.33), who were independently selected by NSW Health from across the state to represent the senior aggression and violence minimization trainers. Participants had various professional health backgrounds including emergency, psychiatric and forensic medicine, aged care, dementia and brain injury, administrative, acute, community and security services. Participants had been in their current occupation for an average of 13.86 years (SD = 10.47).
Procedure and data collection instruments
The ‘train the trainer’ pilot was completed over 2 days and included an orientation to the program and a review of policies, resources, teaching strategies and assessment procedures for modules 1–4. Participants were asked to make an overall evaluation of the training and resources on completion of the workshop and made ratings on specific statements on a scale from 1 (strongly disagree) to 10 (strongly agree), regarding: (i) the ease with which the manual could be used and followed; (ii) the ease with which the content could be referred to and used; (iii) the appropriateness of the teaching strategies; and (iv) the appropriateness of the assessment and ease in administering the assessment.
Training workshops
Participants
Forty-eight experienced health care staff (33 female, 15 male; mean age = 39.15, SD = 10.74) were independently preselected by the pilot site (one representative mid-sized area health service within NSW) to represent relevant occupational backgrounds for the training evaluation and comprised the following staff: administration (3); management (11); education (17); nursing (12); medicine and allied health (2); and security (3). A consumer participated in all workshops and provided individual feedback. Participants had been in their current occupations for an average 10.36 years (SD = 10.05, range = 1–35). Educators were prominent in the pilot phase because of their learning and development expertise. Due to training space limitations and pedagogical imperatives, each module was piloted on a subsample, with modules 1–4 having 18, 20, 16, and 10 participants each, respectively, including one consumer per module. Consistent with the development of more specialized knowledge across the program, 16 staff (15 health staff and one consumer) attended more than one module. Of these, seven attended two modules, four attended three and five attended all modules.
Procedure and data collection instruments
Training outcome data was collected for each module to examine the following.
Satisfaction with training. Following training, participants indicated their response to statements on a 10-point Likert scale from 1 (strongly disagree) to 10 (strongly agree), regarding: (i) the clarity of the aims and objectives; (ii) the appropriateness of the overall content; (iii) the knowledge and skills developed in relation to aggression in the workplace, as a result of attendance; (iv) the overall satisfaction with the training; (v) whether the program would be recommended to other staff; (vi) the relevancy of the overall content covered; (vii) the clarity of the presentation; (viii) the program length; (ix) the appropriateness of the assessment; and (x) whether the assessment scenarios were easy to read and comprehend.
Knowledge and skills acquired during training. An assessment of competence was based on participants' responses to written scenarios dealing with aggressive incidents in the workplace. The assessment required that specific core competencies be satisfied in order that participants are deemed competent. Table 1 lists the desired learning outcomes for each module of the program.
Attitudes toward managing aggression. Pre- and post-measures of employee attitudes were measured with the Attitudes Toward Aggressive Behaviour Questionnaire [26]. This questionnaire consists of statements, rated on a five-point scale from strongly disagree (1) to strongly agree (5). The questionnaire was derived both from the aggression literature and more directly through the kinds of statements made by participants in training. Reliability for this instrument was assessed via the test-retest method and found to be very high (r = 0.97) [26].
Confidence in dealing with aggression. Pre- and post-measures of employee confidence were obtained using the Confidence in Coping With Patient Aggression Instrument [27]. The instrument is a 10-item self-report questionnaire that addresses areas pertaining to ability, preparation, comfort in safety, effectiveness in intervening psychologically and physically with aggressive clients for self-preservation and therapeutic intervention. Participants indicate their degree of confidence using a 10-point Likert scale with verbal descriptors as anchors. During the development of the instrument, Thackrey [27] found the it had a high degree of internal consistency with an α- coefficient of 0.92 and Allen and Tynan [28] reported an α of 0.88. Validity is supported by two studies, which showed significant differences in confidence between the trained and untrained groups between pretraining and follow-up [28], [29].
Data analysis
Overall results of the evaluation are presented as mean (standard deviation or range) and pre-post comparisons are analysed using paired t-tests. Comparisons of pre-post confidence scores as they co-vary depending on number of training modules completed, were computed as an analysis of covariance. In analyses, criterion for statistical significance was set at p < 0.05.
Results
‘Train the trainer’ workshop
The average scores (and ranges) for the evaluation statements rated by the participants, using a Likert scale (1 = strongly disagree, 10 = strongly agree) on completion of the workshop, were as follows: (i) the facilitator manual is easy to use and follow, 7.0 (range = 5–9); (ii) the content is easy to refer to, 7.0 (5–9); (iii) the teaching strategies are appropriate, 7.5 (6–9); and (iv) the assessment is appropriate and easy to administer, 7.2 (6–10).
Training workshops
Satisfaction with the training
Table 2 shows the mean (SD) satisfaction ratings for each module on the eight evaluative dimensions. Satisfaction with each module was acceptable, with a median rating of 7.70 for the program elements. With regards modules 1 and 2, two further statements were evaluated with respect to the assessment. To the statement ‘the assessment was appropriate’, participants in module 1 rated this 6.08 (2.18) and module 2 participants rated this as 7.06 (2.20). To the statement ‘the assessment scenario was easy to read and understand’, participants in module 1 rated this 6.38 (2.02) and module 2 participants rated this as 7.97 (1.58).
Mean satisfaction ratings (standard deviation) for each module of the training†
Knowledge and skills acquired during training
Participants' average percentage score (and range) for the postassessment of competency carried out for module 1 was 69% (60–84). For module 2 the average post-assessment of competency score (and range) was 66% (42–100). For module 1, 100% of participants were deemed competent, meaning that all participants could satisfactorily accomplish the learning outcomes for module 1. For module 2, 82% were deemed competent, meaning that 82% of participants could satisfactorily accomplish the learning outcomes for module 2.
Attitudes toward managing aggression
Table 3 shows the pre- and post-test means, standard deviations, paired t and significance levels for items from the Attitudes Toward Aggressive Behaviour Questionnaire [26]. The significant differences between pre- and post-measurement for four items indicates an increase in the participants' understanding of the motivations underlying aggression and improved management strategies.
Pre- and post-test mean (standard deviation) and paired t (significance) for items on the Collins Attitudes Toward Aggressive Behaviour Questionnaire†
Confidence in dealing with aggression.
For the 34 participants who completed only one module, there was a small but significant increase in confidence measured by the Confidence in Coping with Aggression Instrument [27] (pre-training mean = 62.67, SD = 19.19; post-training mean = 68.85, SD = 17.23, t = 3.26, p = 0.00). For the 15 who completed more than one module, there was a greater increase in confidence (pre-training mean = 69.85, SD = 13.99; post-training mean = 82.15, SD = 9.2, t = 4.38, p = 0.00. An analysis of covariance comparing the pre- and post-test scores for those completing more than one module was performed, controlling for the number of modules completed. This analysis found that confidence scores were significantly influenced by the number of modules completed by the participant, such that the more modules completed, the greater the confidence (ANCOVA F = 4.03, p = 0.04).
Discussion
This study describes and evaluates a new aggression and violence minimization training program developed for health care staff. An extensive consultation process on three prior drafts prior to the first training workshops helped ensure that the materials met the various needs of different areas of the health sector. Experienced aggression and violence minimization trainers evaluated the final training program and assessments. The overall evaluation of the training program from the perspective of 15 experienced aggression minimization trainers was obtained and the results indicated positive endorsement of the program. Qualitative comments obtained assisted in preparing the final published program. The program was also taught to selected experienced staff as part of the pilot assessment. At the completion of training, staff ratings indicated satisfaction with the program, an increase in knowledge and skills, a significant improvement in attitudes toward working with aggressive patients and a significant increase in perceived confidence in dealing with aggressive incidents. Additionally, the results suggest that the more modules staff completed, the greater their perceived confidence in dealing with aggressive incidents.
Some further explanation of several of the results obtained is necessary. First, with regard to participants' ratings of their satisfaction with the training, the results given in Table 2 reveal that for module 1 the item dealing with program length was rated by participants as problematic (average rating = 4.21). Written comments given by the participants suggest the main issue related to the impracticality of staff being allocated a whole day for training. This issue was addressed by offering flexible delivery options for such situations. Second, all participants did not meet the competency standard for module 2. This is because module 2 was designed for specialized clinical staff working in high-risk environments and several of the pilot participants were not familiar with this area and required additional support to meet the competency. Third, a detailed examination of the items included in the Collins Attitudes Toward Aggressive Behaviour Questionnaire [26] (Table 3) reveals that the items that did not differ significantly between pre- and post-measurement are most likely the result of the particular characteristics of this sample. The majority of participants were experienced individuals with previous training and understanding of aggressive behaviour and as such tended to respond in the predicted direction both prior to and after training. There are several limitations of the present study. First, it was outside the scope of the current study to follow up the effect of training in the workplace to assess its impact on aggression. Future research is needed to evaluate whether this training reduces violence and injury rates. Second, the pilot testing was purposely conducted on experienced staff, because they were judged to be able to give the best and fairest appraisal of the content and procedures used. It remains to be tested whether less experienced staff will find the program of benefit. A third limitation is the small sample size upon which the pilot modules were tested. A fourth limitation is that the developers conducted the training and were thus highly familiar with the materials and content. It is unclear whether independent trainers are able to deliver the training using the materials developed and achieve the same satisfaction from participants. A fifth limitation related to the difficulty in developing a broad program encompassing the needs of both clinical and non-clinical staff. For example, training in areas such as risk assessment and ways to deal with workplace bullying were important content areas for non-clinical staff, whereas more skills-based training using scenarios with a focus on communication and physical self-defence strategies were more highly valued areas for clinical staff. Achieving a balance here is not easy, and in instances where not all the needs of staff were fully met, additional specialized training outside this program may be required. A sixth limitation involved difficulties in fully implementing the ‘zero tolerance’ approach. To participants, ‘zero tolerance’ implied an attitude of withdrawal and punishment toward any individual exhibiting aggressive behaviour, when in fact the stated ‘zero tolerance’ policy was that all instances of aggression and violence should be taken seriously rather than treated simply as ‘part of the job’. It was found that this issue is complex and that additional illustrative examples and small group discussions may be required to clarify the meaning of the approach.
In conclusion, this aggression and violence minimization program was found to improve staff knowledge, skills, confidence and attitudes toward dealing with aggression and violence in the workplace. Participants were satisfied with the program and it may contribute to helping health service industry staff attain a safer workplace.
Footnotes
Acknowledgements
The authors thank the Centre for Mental Health, NSW Health Department. A violence taskforce project contract steering committee comprised: Frances Waters (chair), Kathy Baker, Trish Butrej, Maggie Christensen, Nicole Ducat, Brin Grenyer, Louise Newman, Gemma Summers and Choong-Siew Yong. We also thank Beverley Raphael and Duncan Chappell.