“A pilot study of MAP – a program for handling of aggressive behaviour in psychiatric wards. An evaluation of mediating mechanisms”

Abstract

Introduction

Staff working at inpatient psychiatric hospitals are at greater risk of being exposed to workplace aggression, with an incidence rate over 32% worldwide. Workplace aggression includes behaviours or actions that are meant to inflict harm or injury, verbally or physically, to another person. Exposure to aggressive behaviour is associated with negative work-related outcomes and higher levels of coercive measures in mental health care facilities. Several interventions aimed to prevent or reduce workplace violence have been developed, and staff training programs have shown to have some efficacy towards increasing staff knowledge and competence, in addition to reduce coercion to patients. “Management of Aggression Program (MAP)” is a Norwegian, nationally employed staff training program designed after classic triage from green to red, and is categorized as primary-, secondary- and tertiary prevention, with the aim to providing staff basic understanding of aggression, risk assessment, violence prevention, preventive communication, de-escalation, teamwork and self-regulation. This pilot study aims to explore the possible mediating mechanisms of MAP.

Methods

A qualitative exploratory study design was used. Data was collected by repeated semi-structured interviews of four participants at two different timepoints, using thematic analysis to analyse the written transcripts.

Findings and conclusions

The findings of this study suggest that themes within domain of cognitive reframing and contextualization might be the most important mediating mechanisms of the staff training program. The interviewed participants found the training program to be engaging, and the training provided the participants with knowledge and skills to be confident and conscious of their work. There is a lack of prior research or published studies about MAP, and similar programs, as of writing this report. Therefore, a need for further research on this topic is necessary. The findings of this study might be suitable for selection and operationalization of endpoints in future studies.

Introduction

Staff working at inpatient psychiatric hospitals are at greater risk of being exposed to occupational risks such as workplace aggression, with an incidence rate over 32% of healthcare workers worldwide [1]. Compared to other nations, the Nordic countries have higher reported prevalence of workplace aggression than other comparable countries [2]. There might be several reasons for the high prevalence in the Nordics, such as cultural differences or lower threshold for reporting negative occurrences. In Norway, 44% of healthcare workers reported that they had been exposed to workplace aggression the last 12 months [3], 30% of mental health care workers had been victimized more than five times the past year [4]. Verbal aggression and object aggression are most prevalent (42% and 44% respectively) over a period of 12 months [5]. Nurses and psychiatrists are at higher risks of exposure to workplace aggression [2,3,4,5].

The Norwegian Labour Inspection Authority defines work related aggressiveness as violence and threats. Threats are verbal attacks or actions that is meant to harm or frighten another person, violence is defined as any action that is intended to physically or psychological inflict damage to another person. Aggressive behaviour resulting in damage to interior or equipment is also defined as violence [6]. Long-term exposure to hostile behaviour is associated with negative work-related outcomes, such as long term absence from the workplace due to negative impact on the victims physical or psychological health [1, 6, 7], but it may also negatively influence the quality and safety of the provided healthcare [8, 9]. Staff fearing assault may in some cases be more inclined to excessive use of coercion towards patients in order to feel safe [9]. Conversely, decreased levels of conflict might contribute to reduce excessive use of force and coercion, as is suggested by the safewards model [10, 11], though this model is yet to be implemented in Norway.

Balancing complex interactions while both establishing therapeutic bonds with the patient and ensuring a safe space to interact for patient and staff, in addition to working in a context where staff may be exposed to risk of aggression, is a critical challenge of inpatient nursing. This may be complicated by the patient’s condition, such as cognition, affect or arousal, as well as the staffs previous experience with treatment of the patient. The staff’s general sense of safety at the workplace might impact the ongoing care for the patient as well.

Several interventions aimed to prevent or reduce workplace violence has been developed and are usually grouped into three categories; education and training, organisational interventions, and workplace design. Education and training of staff aim to give staff experience and skills to identify and react to aggression, and knowledge to prevent and manage incidents [7]. In Norway it is estimated that targeted training programs may reduce or prevent workplace aggression, violence or threats, and might reduce workplace absence up to 13% [6]. Existing research on the efficacy on training programs aimed at reducing violence or violence reduction in the workplace is lacking or is viewed as having low quality evidence [12, 13].

Background

Training programs can be considered to be an essential part in Evidence-based Nursing (EBN) in the sense that participants fine-tune skills that might be useful in clinical situations. It is imperative that novel training programs are developed from the best available evidence. Management of aggression program (MAP) has been developed through extensive literature review and pilot testing [14]. Equally important are evaluation to assess whether the efforts of developing and implementing programs can be justified, and whether training programs achieve their goal in being an effective intervention to reduce harm to patients and staff, or to which degree it reaches its goal.

MAP is a result of a cooperation between Centre for Forensic Psychiatry (SIFER) and regional forensic psychiatry wards with affiliation to regional hospitals in Norway [15]. The conceptualization of MAP was initiated as a response to the development of a new law, Tvangsbegrensningsloven, to reduce coercion to patients under psychiatric care in Norway [16]. MAP consist of ten chapters covering theoretical lectures, participant reflection and skill training. The theoretical chapters aim to give staff basic understanding of the phenomenon of aggression, risk assessment, violence prevention, preventive communication, de-escalation, teamwork and self-regulation (Table 1). The training program lasts over two days, where each day is half theoretical training and skill training [14, 17, 18]. The training also gives the participants opportunity to reflect upon the curriculum and cases together.

Table 1 List of chapters in MAP and corresponding learning goals [14]

MAP is a staff training program designed after classic triage from green to red, and is categorized as primary-, secondary- and tertiary prevention. Primary prevention aims to avoid violence and coercion to occur at all, secondary prevention aims to early intervene when potential violent situations escalate, and tertiary prevention is used when situations have escalated, and staff intervention is necessary to avoid damage to patients, staff or inventory (Fig. 1).

Fig. 1

Figure of modules in MAP [14]

MAP was developed in cooperation between regional hospitals Helse Vest RHF and Helse Sør-Øst RHF [15]. MAP was provided to staff in the University Hospital of North Norway (UNN HF) from 2021.

Conflict-reducing and aggression de-escalating programmes like MAP have demonstrated a reduction in the use of coercive measures in psychiatric wards [8, 19]. Use of risk assessment tools, training in use of de-escalating technics and leadership that focuses on reducing coercion have shown efficacy towards use of coercion in wards [20].

In the Norwegian “National Health and Hospital plan 2020–2023”, a deposit report from the Norwegian health department, uses MAP as an example for staff training to alter staff attitudes, competence and organization, and aims to categorize the physical techniques from MAP as an safe use of coercion [21] amidst the ongoing proposition to reduce coercion in psychiatric wards in Norway [16].

Beyond the basic course, authorized instructors provide weekly training on physical techniques for smaller groups, usually lasting one hour. This study, however, will focus mainly on the basic course.

The method of evaluating MAP was guided by Kirkpatrick’s model for training evaluation, which was designed as a method for evaluating training programs with the reasons to improve programs, maximize transfer of learning and to demonstrate the value of training [22]. Level 1, reaction, is referred to as the customer satisfaction measurement and engagement (i.e. the active participation and contribution to the learning experience), and level 2, learning, is defined as the degree to which participants acquire the intended knowledge, skills, attitude, confidence and commitment based on the participation of the training [22]. Data collection immediately after training might shed light on influence of the training on reaction and learning, e.g. if the participants find the training favourable or relevant to their jobs, if the participants believe that the training will be worthwhile to implement or if the participants think they would be able to do what they have learned during training on the job [22]. Level 3, behaviour, is defined as to which degree the participants apply what they have learned during training. Data collection 4–6 weeks after the training might be more suitable to explore the trainings influence on level three. Some skills, like communication, might require time to become implemented [22]. Level 4, results, will not be addressed by the methods of the current study.

This pilot study aims to explore the knowledge on how participants describe, reconstruct, and operationalize the input given by the training. It also aims to explore how the participants experience the training, and to examine how the staff’s self-evaluated professional behaviour related to handling workplace aggression is affected. The findings might be relevant for the selection and operationalization of endpoints, as possible mediating mechanisms of MAP in future studies.

Method

Design

A qualitative exploratory design was used to evaluate the intervention (MAP) at University Hospital of North Norway UNN HF, Åsgård during the winter of 2021 and spring of 2022. Interviews were collected immediately after the intervention and again 4 to 6 weeks later, as is suggested by levels 2 and 3 in Kirkpatrick’s model.

Sample

MAP is offered to all ward employees working at Åsgård, Norway’s northernmost psychiatric hospital located at 69 degrees north which serves 25% of mainland Norway’s specialists healthcare needs, a population of 250 000 residents. Registered Nurses (RN) (or similar with a three-year bachelor’s degree) and care-workers without formal education are the target participants of the intervention. Employment at a ward and participation on the training program were the only inclusion criteria for this study. However, one exclusion criteria had to be made. Temporarily employees and part-time workers were excluded from this study, given the uncertainty that they would be able to attend the second interview. Potential participants who were considered to have a personal relation to the interviewer, who also works as a RN, were also excluded from this study. Study participants were selected and recruited at the last day of MAP-training by the author, who approached the course participants face to face at three different training sessions.

Data collection

Data were collected week 49 2021, and weeks 7, 9, 12–14 2022 by the author. The first interviews were conducted during the first week after the course with exception of one participant where the first interview was conducted two weeks after the training. The first series of interviews had a duration of one hour. The second interviews were conducted 4–6 weeks after the first interviews and lasted 30 min. Data were collected at the workplace, in a secluded conference room or the lead authors office, after agreeing on time and date that worked best for the participants. Ahead of the interviews the author was aware of the formal training of the majority of the participants. Background variables of the participants, such as education, gender or age, although important factors to the perception of violence, was considered to be outside the scope of this study. A semi-structured approach was used. The interview guide was developed specifically for this study (Supplement 1) and piloted by conducting a single interview with one participant, and later revised in December 2021. The interview guide consisted of 13 open ended questions with possibility for follow-up questions. The questions aimed to trigger some degree of reflection in the participants. The guide was divided in three thematic sets: MAP-training, professional practice, and the patient. The questions were meant to examine how the participants experienced the training, e.g., how the participants reacted or reflected upon the training. Further questions examined how the participants attitudes or perceptions might have changed in relation to assessment of aggression-laden situations or how it affects their work. Finally, the questions asked the participants to view the staff’s attitudes and response to aggressive situations from a patient’s perspective. The second interview had similar structured interview guide as the first interview. In addition to ask the participants to view the staff or situations from a patient’s perspective, it was also meant for the participant to reflect upon their current skills and attitudes. Data were audio-recorded and stored in “Nettskjema” an online multi-step verification site owned by University of Oslo.

Participants were informed in advance about the aim of the study, their legal rights and their rights to opt-out at any time. All participants signed an informed consent to participate in the study, which includes consent to publication. This project was approved by the Norwegian centre for Research Data (NSD).

Data analysis

The data were anonymized, and transcribed using Microsoft Office 360. The analysis process followed Braun and Clarke’s six phases of thematic analysis; familiarization of the data, generating initial codes, searching for themes, reviewing themes, defining and naming themes and producing the report [23]. The first phase, familiarization, includes transcription and repeated reading of the written material. The second phase includes coding features of data that appears interesting. After the initial coding, the third phase re-focuses the analysis at the broader levels of themes, sorting and identifying potential themes. The fourth phase includes reviewing and sorting themes. The fifth phase requires the analyst to identify the essence of what the theme is about. The final phase is to generate the report [23]. Thematic analysis does not require the same level of detail in the transcript as other forms of analysis, however it requires at a minimum a verbatim account of all verbal utterances [23]. The data were coded and analysed by the lead author. The research question and Kirkpatrick’s levels provided guidance in the coding process, by allowing the author to search the dataset for potential codes. A “theoretical” thematic analysis of the material was used [23] resulting in sorting main themes beneath a fixed set of categories that corresponded to Kirkpatrick’s levels: Reactions, Learning and Behaviour. A coding hierarchy was developed, this consisted of the raw codes from the transcribed dataset, sorted under a parent code. The abstraction of meaning of the raw codes were sorted under a sub-theme. The sub-themes were then sorted under the final theme (Table 2).

Table 2 Example of structured coding scheme for two of three themes

In order to enhance trustworthiness of the study, care have been taken to piloting the topic guide. Moreover, the consolidated criteria for reporting qualitative research [24] were followed. In addition, the interview guide and raw data material from the pilot were critically reviewed by an external researcher with no affiliation to the participants workplace. The pilot data was reviewed to assess if the topic guide was specific enough to explore the participants experience of the training. Data were coded using NVivo 1.6.1 (QSR International) and necessary data compiled to useful structures to be used in the final report.

Data triangulation have been achieved by collecting data at two different timepoints (at week 1 and at week 4–6), also known as time triangulation [25].

In addition to piloting the topic guide, the COREQ-checklist have been used in order to produce the written report [24,25,26]. For instance clarifying personal characteristics of the author, reporting in detail on study design or including quotations from the participants might add trustworthiness and transparency to the findings [24].

Results

Of the eight participants recruited, four participants attended the interviews. There were several reasons for non-participation. However, the ongoing Covid-19 pandemic was the main reason. Two informants were not formally trained as healthcare professional, and the other two informants were registered nurses or similar with a bachelor’s degree in the field of healthcare. Three of the participants were male and one were female. The pilot interview has been included in the total dataset. The first interviews lasted about one hour each, and the second interviews lasted about 30 min each.

Reaction

“Intensive engagement” and “Cognitive reframing” were the extrapolated meanings connected to the two subthemes: satisfaction and engagement, within the predefined category by Kirkpatrick’s model “reaction”. Participants’ immediate response after the training was that the training program offered a great opportunity to reflect during the lectures. One participant described a feeling of tiredness, explaining: “I felt, at the end of the second day, that I wasn’t able to acquire more information (…) It was an extraordinary amount of information to go through, I wasn’t capable to process any more information at the end.”

Other participants also felt that the training was too compressed, that the curriculum was too big, or that the training program was presented within too short time to convey the curriculum.

Following the Kirkpatrick model, engagement is defined as the participants active involvement in the training. The participants seemed to agree that the skill-training was the most engaging part of the training program. One participant noted that the group discussions also generated engagement from the training-participants. One participant explained the motivation of being active in the training, saying: “(…) the way you make sure situations doesn’t escalate (…) making sure that no one gets harmed (…) I would call it caring.” A shift of mindset from force and coercion being a negative action, to be recognized as caring was coded as “cognitive reframing”.

Learning

The theme of “learning” consisted of four subthemes: Knowledge, Skills, Attitudes and Confidence, again as predefined by Kirkpatrick’s model. One participant noted a changed in awareness of own body-language when interacting with patients, explaining: “I am more conscious on my own body-language when I interact with patients (…) I haven’t thought much of this previously (.) that one sits with arms crossed and one seems less forthcoming.” The participant experienced a change in self-consciousness. This, together with a shift in focus to how one might be perceived outwards, were coded as “internalization of attitudes”.

One participant discussed the importance of being aware of the power imbalance in the therapeutic relationship. The training had reminded the participant of the patients’ experience of powerlessness and that this could affect or even damage the relationship with the patient.

The participants’ response to situations and communication skills were coded under the subtheme “skills”. One participant felt more able to engage with the patient in problematic communication, saying: “(…) by not being afraid to ask unpleasant questions, one can learn a lot if one dare to ask the patient questions.” Another participant described being more confident in participating during teamwork and knowing how to delegate tasks. The participant said: “(…) knowing how to handle the situation, especially when using force, who should talk to the patient while this is happening.” This experience of confidence to engage with patients in dialogue was coded as “interaction”.

The code “contextualization” was placed under the subtheme of “attitudes”, where participants described a new view on workplace experiences. One participant described: “After previous situations I have doubted my actions and thought whether I could do things differently. After training I have given the triangle model attention, and I think this helps me categorize my actions”.

Within the code “placement of responsibility” one participant explained: “(…) It’s easier to participate in difficult discussions with co-workers or patients with a theoretical foundation (…)”. Another participant said “One knows one’s responsibility in situations”. The participants were able to contribute to the workplace with a theoretical framework, such as MAP, as well as knowledge of correct placement of responsibility. MAP teaches the participants cooperation and management in difficult situations, including critical principles of management, communication and responsibilities towards staff, patient and workplace [14, 27]. Staff working as health personnel in Norway also has an individual responsibility making sure that patient treatment is safe. This was sorted under the subtheme of “confidence”.

Behaviour

The theme of behaviour consisted of only one code: “applying communication skills”. One participant felt a stronger commitment to take initiative to build a therapeutic relation with the patient after the training, saying: “(…) and I feel this is important (…) to avoid conflicts or unfortunate situations.”

Another participant emphasized the meaning of purposeful and direct communication, explaining: “Perhaps the teamwork in one situation was not optimal, and after training I became aware that this is not how we should do it. Now I know how to talk to co-workers and the patient during situations”.

Discussion

Main findings

The aim of this pilot study was to identify the possible mediating mechanisms of MAP using qualitative interviews as a data collection method.

Intensive engagement, cognitive reframing, internalization of attitudes, interactions, contextualization and placement of responsibility were the main findings in this study. Cognitive reframing and contextualization were considered to be the most important findings and will be further discussed. Other findings, while important, will not be discussed in this paper.

Cognitive reframing

Cognitive reframing, or cognitive reconstruction, is described as a method to consciously adjust the thinking process in order to remove forms of psychological discomfort created by behaviour or outcome [28]. Cognitive reframing may be used to help an individual by presenting alternative ways of thinking [29]. The aim of MAP is, amongst others, to help facilitate the participants to be able to face and handle aggressive situations [14]. By educating the participants in techniques for self-regulation, understanding aggression and risk assessment it is possible that the participants experience greater sense of safety when handling difficult situations. MAP offers training on self-regulating techniques such as breathing exercises, and teachings on the physiological effects of external stress on the body, i.e. sympathetic responses to external stimuli. The goal is to make the participants aware of automatic responses, and tools to handle these to avoid possible negative outcomes for the patient [14].

If the cognitive framework of the carer presupposes coercion as questionable, then use of restraint or coercion might contradict the staff’s professional ethos. Patients, for their part, might perceive the use of force as abuse or contribute to a distrust towards the participating staff. Being in a situation where one uses force against another human might be traumatic for the practitioner, and experiencing a shift from caregiver to warden might be difficult for some. However, there might be situations where use of force is needed.

One participant noted that “The way one cooperates with the patient and reduce escalation of situations and avoid harm to patient and staff (…) is care.”. The term “care” in connection with force or coercion may be frowned upon. Care is a term that has several dimensions. Care is a social relation that deals with understanding of another human. Care is also describing an attitude [30]. A Norwegian study found that some patients, in hindsight, agreed that the use of coercion or restraints have been used for their own protection [31]. The participant, however, says that avoidance of harm as result of aggressive situations or successful de-escalation in cooperation with the patient is care, an understanding that also is provided by the MAP training.

Avoiding escalation of an aggressive situation might be considered as care. The nurse understands the patient and has a common understanding of the situation with the patient, and the nurse conveys authenticity and acknowledges the patient’s situation. There might occur situations where the nurse is left with no other solutions but the “wrong” solution, such as physical restraints. Participating in the training, the participants acquire necessary knowledge and skills to provide professional care to patients. By providing a shift in standpoint towards use of force from believing that using coercion or physical restraints is bad or good, to believing this is what the patient needs right now, might be considered as a cognitive reframing.

Contextualization

MAP training and discussions are contextualized by the instructors, who also urges participants to share examples or participate in discussions. Context refers to the particularization of meanings of words and phrases, such as gaining an understanding of a phenomenon like violence or aggression. Context also refers to coherence of a larger whole that prevents particularized meanings from being isolated, for example connecting theoretical perspectives to one’s own experiences. It has a connective function and potential [32].

In other words, contextualization helps participants to connect what they learn with their work practices. It also makes it possible to view their practises in light of training and enables the participants to evaluate or rethink their previous behaviour or actions. One participant noted that “After previous situations I have doubted my actions and thought whether I could do things differently”. The contextualization seemed to trigger a process of reflection in the participant resulting in an altered point of view regarding the experienced situation. MAP might offer vertical re-contextualization, a process were new activities and insights are developed on a theoretically derived basis [32] by offering theoretical and physical education and training.

Limitations

Several aspects have been taken under consideration in the design to reduce the chance of bias. The known bias in this study is the researcher’s prior knowledge and occupation in the same field as the participants. This has been reduced by using a topic guide during the interviews. Prior knowledge of the researcher could represent a selection bias, as the interview subjects might have accepted or declined to participate based on their prior knowledge to the researcher. The sample size is small and it might therefore challenge the validity of its findings and the findings might not be representative of a larger sample size. In addition, the semi-structured approach to collecting data might skew the results as the premade questions may not tap into what the subject considers to be most important or relevant. Collecting data by different means (for example combining interviews with observation, or in this case data triangulation) might have increased the triangulation and thereby increased the trustworthiness of this study further. The fact that there was only one person conducting the interviews and analysing the data might be a weakness of the study. The researcher might also be at risk of asking leading questions, or the interviewee might show answering behaviour in terms of social desirability. On one occasion a participant asked what answer the interviewer wanted, while thinking about an answer to the question provided. The top-down approach to thematic analysis might also not have been the most ideal method for analysis, the risk of prejudices is higher than an inductive approach.

Conclusion

The findings of this pilot study suggest that cognitive reframing and contextualization might be possible mediating mechanisms of the staff training program, based on a limited sample of study participants. Cognitive reframing might give participants new perspective and help avoid automatic responses to situations that might lead to escalation. Contextualization provides possibilities to connect curriculum to clinical practices.

The findings of this pilot study might be suitable for selection and operationalization of endpoints in future studies. Further understanding of MAP is needed, and possible mixed methods studies might be suitable for further research.