Psychiatric hospital nurses' attitudes towards trauma-informed care.

WHAT IS KNOWN ON THE SUBJECT?: Research indicates many clients using mental health services have trauma histories. Consequently, mental health professionals must be aware of the impact of trauma and of how they can avoid retraumatizing service-users. Care delivered with this awareness is known as trauma-informed care (TIC). There is little research on attitudes towards TIC. To date, only one study explored these attitudes among MHNs exclusively. Additionally, a richer understanding of TIC attitudes using methods like in-depth interviews is needed. It is unclear whether knowledge of TIC results in more favourable attitudes. WHAT THIS PAPER ADDS TO EXISTING KNOWLEDGE?: MHNs in this study had little knowledge of TIC but expressed overall favourable TIC attitudes. Traumatic histories were not appreciated as causes of challenging behaviour. On rehabilitation wards, clients come to be perceived as family members and this makes it harder for MHNs to not take challenging behaviour of clients personally. MHNs face work-related traumas which interfere with their ability to provide TIC. WHAT ARE THE IMPLICATIONS FOR PRACTICE?: Findings of this study can be used to guide plans to implement TIC in psychiatric hospitals. Policymakers are called to appreciate that ensuring MHN well-being on the workplace will facilitate their delivery of TIC. TIC training initiatives for MHNs must stress the importance of acknowledging traumatic histories as causes of challenging behaviour and of maintaining professional boundaries with long-term clients. This would benefit service-users by ensuring MHNs are more trauma-informed. More research on attitudes towards TIC among MHNs is needed. ABSTRACT: Introduction Quantitative studies exploring trauma-informed care (TIC) attitudes have not used samples made up exclusively of mental health nurses (MHNs). Qualitative methods were sparingly used. Aim To examine nurses' TIC attitudes at a psychiatric hospital. Method A mixed-method design was used. One hundred and thirty-six MHNs completed the Attitudes Related to Trauma-Informed Care scale. Data were analysed using inferential statistics. A focus group interview among ten MHNs ensued. Thematic analysis was used. Results MHNs demonstrated favourable TIC attitudes. Ambivalent attitudes for the subscale "Causes" were identified. MHNs employed for less than 5 years at the hospital and those in acute settings displayed more favourable attitudes on some subscales. Three themes "Awareness," "Unhealthy boundaries" and "Inhibition" emerged from qualitative analysis. Discussion Challenges uncovered in the provision of TIC include the unacknowledged impact of trauma on challenging behaviour among MHNS, the influence of blurred professional boundaries with long-term clients on the cycle of perpetuated trauma identified by previous research and MHNs work-related traumas. Implications for practice Identified challenges to TIC integration among MHNs can facilitate the implementation of TIC in hospitals. TIC educational packages for MHNs should acknowledge traumatic histories in the aetiology of challenging behaviour and stress the importance of maintaining professional boundaries with clients.

INTRODUCTION

The high prevalence of trauma among mental health service‐users has been extensively reported (Álvarez et al., 2011; Anderson et al., 2016; Campbell et al., 2016). Consequently, the integration of a trauma‐informed ethic in mental health service delivery is being increasingly recommended (Isobel & Delgado, 2018; Wilson et al., 2017). A conceptualization of trauma‐informed care (TIC) is developed by Watson et al. (2014) as care that is cognizant of the prevalence and adverse effects of trauma, and of the potential of retraumatizing clients with trauma histories through coercive practices.

The Substance Abuse and Mental Health Services Administration (SAMHSA, 2014) describes what a trauma‐informed organization should look like in practice, listing the main features as: an appreciation of the impact of trauma, an understanding of potential paths for recovery from trauma, the recognition of signs and symptoms of trauma in clients and staff, a full integration of knowledge about trauma into policies, procedures and practices and a commitment to avoid retraumatization.

Baker et al. (2016) positioned that service providers' attitudes towards TIC constitute a particularly measurable outcome for ascertaining whether a service is trauma‐informed. The body of knowledge on mental health nurses' (MHNs) attitudes towards TIC is scant and novel. Research studies addressing this phenomenon originate mostly from the United States or Australia. It seems European research lags behind with respect to MHNs' attitudes towards TIC. Kazlauskas et al. (2016) hint at this European lag regarding the phenomenon of trauma when they outline the direct need for a detailed analysis of trauma treatment in European countries and the development of European‐level trauma‐informed healthcare policies.

A few cross‐sectional studies focusing on variables related to TIC attitudes comprise one cluster in this sparse body of knowledge. One such study is that of Jacobowitz et al. (2015), which identified factors associated with development and mediation of staff (various disciplines) PTSD in a psychiatric hospital in the United States (n = 172). It was concluded that staff with higher educational levels were more likely to have attended a recent TIC meeting. In another cross‐sectional study, Sundborg (2019) explored the relationship between knowledge and commitment to TIC (n = 118) among individuals working in health and occupying clinical, administrative and managerial roles in the United States. The author reported that the relationship between knowledge and commitment to TIC is significant and partially mediated through beliefs about trauma. Marvin and Volino Robinson (2018) provided information about knowledge and attitudes among staff members in a human service organization preparing for TIC in the United States. No statistical significance was reported between TIC knowledge and attitudes. The professional roles of staff members were not specified. None of these cross‐sectional studies have used samples made up exclusively of MHNs.

A complex variety of previously validated and non‐validated tools developed specifically for the studies in question were employed to measure a number of variables. It was noted that to date the only psychometrically evaluated tool to measure TIC attitudes is the “Attitudes Related to Trauma‐informed Care” (ARTIC) scale developed by Baker et al. (2016).

Qualitative research methodologies have been underutilized to explore this phenomenon. In one study (Stokes et al., 2017), MHNs argued whether TIC is a basic constituent of nursing practice or whether it is necessary to debate its need. The seven Canadian MHNs in this study also recognized that staff reactivity to client challenging behaviour triggers subsequent challenging behaviour from other clients. This phenomenon of repeated triggering of more challenging behaviour was described by the authors as a “cycle of perpetuated trauma.”

The third cluster of studies constituting the literature on attitudes towards TIC used pre‐test post‐test designs to evaluate the impact of TIC training on attitudes and practice. One such study is that of Hall et al. (2016), which administered an 18‐item tool before and after a single‐day TIC education package to 34 nurses from an emergency department in Australia. An increase in the ability to talk to clients about traumatic experiences was reported. However, no effect of education on comprehending how the emergency department environment can be retraumatizing was observed. In another study, Palfrey et al. (2019) matched pre‐post data available for 102 allied health professionals working in a mental health and substance abuse service in Australia in relation to a single‐day workshop on TIC. A significant increase in confidence and awareness towards TIC approaches was reported in tandem with specific examples of how practice is anticipated to improve.

International evidence on attitudes towards TIC is scant and comes mostly from the United States and Australia. With the exception of Beattie et al. (2019) and Stokes et al. (2017), existing research has seldom employed qualitative designs. Only Stokes et al. (2017) examined attitudes towards TIC using a sample made up exclusively of MHNs. Consequently, the first objective of this study is an exploration of the prevalence and distribution of TIC attitudes of all the MHNs in a psychiatric hospital in Europe. Acquiring a qualitative discernment into TIC attitudes by considering the cultural context of the hospital constitutes an additional objective. Therefore, the specific research objectives formulated in alignment with the scientific rationale for undertaking this study are;

To identify associations between psychiatric hospital nurses' attitudes towards TIC and sex, age, nursing grade, qualifications, length of time working at the hospital, work setting, participation in in‐service training or trauma‐related training.

To determine whether there appears to exist a collectively constructed reality surrounding TIC among nurses at the psychiatric hospital.

To explore the relationship between individual attitudes and the maintenance of a possible collectively constructed reality surrounding TIC among nurses at the hospital.

METHODS

Study design and theoretical underpinnings

A hybrid collection of both empiricist and interpretivist assumptions was needed to address the research objectives. Hence, a mixed‐method design was deemed most suited. Bourdieu's theory of practice was chosen as a theoretical framework. In essence, it posits that objective structures inherent in society influence subjective behaviours (agency), and the latter collectively then subsequently reproduce society (Fries, 2009). The theory is, therefore, in alignment with the multidimensionality of the research objectives. Moreover, it can be argued that attitudes are antecedent to practice. So, Bourdieu's theory is congruent with both the phenomenon under inquiry and the methods which were used to study it.

The theory is often appraised as favouring an objectivist perspective since it explains how society is reproduced and critics like Williams (1995) claim it does not allow for sufficient agency. An example of how the organization of the study was guided by its framework is how this light tipping towards objectivity in the theory of practice suggested the use of a sequential explanatory design. The study is quantitatively driven. During the qualitative phase, the participants were asked to give their interpretations of some quantitative findings. Examples include interpretations of the response rate, the ambivalent attitude scores for a particular subscale of the tool, and how their attitude scores compared to those reported by a previous study which employed the same tool.

The quantitative phase

Population

There is only one psychiatric hospital in Malta. The research population consisted of all 199 qualified MHNs working at the hospital.

Research tool

The self‐administered “Attitudes Related to Trauma‐informed Care” (ARTIC) scale (Baker et al. (2016) was used to assess attitudes. The ARTIC‐35 version was employed because it was designed for use among people who work in human services who might not be familiar with the concept of TIC. It contains 35 sets of two opposing statements grouped into five subscales designed to capture the pivotal constituents of attitudes towards TIC. Participants had to indicate, via the seven‐point bipolar Likert scale, which of the two opposing statements best represents their attitudes and to what extent. The content of the items in each subscale is as follows;

Causes: 7 items about whether respondents believe challenging behaviour is an outcome of trauma histories.

Responses: 7 items about attitudes related to how participants deal with problematic behaviour of clients.

Job behaviour: 7 items about attitudes related to whether participants think problematic behaviour of clients is perceived as a reflection of their behaviour as professionals by themselves or others.

Self‐efficacy: 7 items about attitudes pertaining to how confident participants are in their professional ability to help clients.

Reactions: 7 items about attitudes related to the perceived importance of self‐care as well as how participants react whenever they feel their work is becoming too overwhelming.

The ARTIC‐35 has satisfactory levels of validity and reliability. Baker et al. (2016) state that confirmatory factor analysis supported the five‐factor model fit of the ARTIC‐35. Internal consistency reliability using Cronbach's alpha was excellent (α = .91). Test–retest reliabilities calculated using Pearson's correlations were strong with correlations of .84 at ≤120 days, 75 at 121–150 days and .77 at 151–180 days. Validity indicators provided preliminary support for construct and criterion‐related validity.

Data collection procedures

Data were collected during August and September 2018. Envelopes containing the questionnaire and an information letter were consigned by practice development nurses to the charge nurse of every ward/unit for distribution among MHNs. MHNs were instructed to return their completed questionnaires in boxes left in each ward/unit.

Data analysis

Quantitative data were analysed using the Statistical Package for the Social Sciences (SPSS—version 26). The Shapiro–Wilk test was employed to assess the normality assumption of the five‐subscale distribution. The Spearman Rank Correlation test was applied to determine the strength and direction of associations between mean scores among different subscales. Since tests of normality revealed data were not normally distributed non‐parametric Mann–Whitney and Kruskal–Wallis tests were applied to analyse the differences between the mean ranks of each subscale of the tool by each demographic characteristic of participants. The standard α = .05 cut‐offs were utilized with the above‐mentioned tests.

The qualitative phase

Sample

Initial plans specified that the variables identified as most impinging on attitudes towards TIC during quantitative data analysis would direct the selection of participants for the focus group (theoretical sampling). MHNs were invited to participate in the focus group via a closed group on social media exclusive to nurses working at the hospital. Only ten MHNs came forward and constituted the sample. However, the purposive sample still provided a heterogeneity of the most impinging variables on attitudes towards TIC as identified by quantitative analysis.

Data collection

A focus group was chosen as the data collection tool since, as Munday (2006) purports, individual interviews would not capture a collectively constructed reality as effectively. The semi‐structured interview guide for the focus group was formulated after quantitative data analysis.

Practice development nurses invited MHNs to participate in the focus group via the group on social media exclusive to nurses working at the hospital. Upon approaching the practice development nurse, MHNs were provided with an information letter that provided details about the study and a consent form.

The focus group, which lasted 94 min, was conducted in April 2019 in a room at the training centre of the hospital which was located away from the administrative quarters. This hopefully fostered more willingness to engage in open and truthful discussions. The moderator sought to grant flexibility, grapple with conflicts and ensure involvement of all participants. A research assistant documented group dynamics and non‐verbal communication in field notes.

Thematic analysis, as outlined by Braun and Clarke (2006), was chosen to identify, analyse and report patterns within the data. This was performed manually by writing notes on the text with coloured pens. Data were approached with pre‐established questions in mind that the researcher wanted to code around (theoretical analysis). These questions were formulated after scrutinizing the quantitative data. Labels were given to features of data that appeared significant to these questions. Grouping of relevant data to each label was then performed. Codes were then sorted into potential broader themes. The choice of theoretical thematic analysis was in alignment with the sequential explanatory design of the study and the subtle supremacy of the objective paradigm in Bourdieu's theory of practice. Themes were then checked to ascertain data extracts within themes cohere and that there were clear distinctions between themes. Strategies to ensure rigour included peer reviews with colleagues to scrutinize decisions and provide evaluations of data interpretations.

Ethical considerations

The study was approved by the Faculty of Health Sciences Research Ethics Committee of the University of Malta. Information letters briefed potential participants with the study purpose and their rights. Complete anonymity throughout the quantitative phase was guaranteed, whilst the concern of maintaining confidentiality during the second phase was communicated. MHNs were also informed that in the event of any psychological distress evoked by the focus group they could contact the researcher asking to be referred for professional assistance.

FINDINGS

Separate reports for distinct phases will be presented. In alignment with the sequential explanatory design, the quantitative findings are reported first.

Quantitative findings

From a total of 199 MHNs, 136 returned the questionnaire. Demographic characteristics of the sample are shown in Table 1.

TABLE 1

Demographic characteristics of participants for the survey and focus group

Characteristic	Survey sample (n = 136)	Focus group sample (n = 10)
	N (%)	N
Gender
Male	60 (44.1)	2
Female	75 (55.2)	8
Age (years)
20–29	32 (23.5)	3
30–39	28 (20.6)	2
40–49	32 (23.5)	4
50–59	33 (24.3)	1
60+	10 (7.4)	0
Grade
Enrolled nurses	10 (7.4)	0
Staff Nurses/Senior Staff Nurses	106 (77.9)	6
Charge Nurses/Deputy Charge Nurses	19 (14.0)	4
Highest qualification
Traditional Nursing Course	29 (21.3)	0
Diploma	47 (34.6)	2
B.Sc.	51 (37.5)	6
M.Sc./PhD	8 (5.9)	2
Length of time working at the hospital (years)
<5	50 (36.8)	3
5–10	16 (11.8)	4
11+	69 (50.7)	3
Work setting
Acute	66(48.5)	8
Rehabilitation	67(49.3)	2
De‐escalation training
Yes	98 (72.1)	7
No	36 (26.5)	3
Trauma‐related training
Yes	16 (11.8)	1
No	117 (86.0)	9

As presented in Table 2, participants had contradicting opinions regarding whether client challenging behaviour is an outcome of trauma histories. This is indicated by a median score of 29 for the subscale “Causes” when the midpoint of the theoretical range of scores is 28. However, for the remaining subscales, the participants showed favourable trauma‐informed attitudes as indicated by the median scores presented for each subscale. The participants displayed more favourable TIC attitudes in relation to how they deal with client problematic behaviour, whether they think client problematic behaviour is a reflection of their behaviour as professionals, how confident they are in their professional ability to help clients and how they react when they feel their work is becoming overwhelming.

TABLE 2

Descriptives for the TIC attitude subscales by participant response

TIC attitude subscale	Theoretical range of subscale (midpoint)	Median (Interquartile range)
Causes	7–49 (28)	29 (31–26)
Responses	7–49 (28)	35 (38–30)
Job behaviour	7–49 (28)	37 (39–32)
Self‐efficacy	7–49 (28)	39 (42–33)
Reactions	7–49 (28)	38 (41–32)

Theoretical range refers to the lowest and highest score that can be obtained for the subscale. Midpoint refers to the middle point between the highest and lowest score

Table 3 presents intercorrelations obtained for the TIC attitude subscales. All subscale scores are significantly and positively correlated, that is, favourable scores in any subscale predict favourable scores in the other subscales. This indicates that the more confident MHNs are in their professional ability to help clients the more likely they are to believe that challenging behaviour is an outcome of trauma histories, to deal with problematic behaviour in trauma‐informed ways, to not perceive client problematic behaviour as a reflection of their behaviour and to attribute importance to self‐care when they are overwhelmed.

TABLE 3

Intercorrelations between subscale scores

	Causes	Responses	Job behaviour	Self‐efficacy	Reactions
Causes	1.000	.525*	.417*	.305*	.385*
Responses	.525*	1.000	.474*	.416*	.443*
Job behaviour	.417*	.474*	1.000	.429*	.435*
Self‐efficacy	.305*	.416*	.429*	1.000	.644*
Reactions	.385*	.443*	.435*	.644*	1.000

p < .001.

Non‐parametric statistical tests were applied to analyse participant responses on the TIC subscales by sociodemographic characteristics (i.e. sex, age, nursing grade, educational level, nursing speciality, years working at the hospital, work setting, participation in de‐escalation in‐service training and participation in any trauma‐related training). Statistically significant differences were identified on TIC subscales by “years working at the psychiatric hospital” and “work setting”. Kruskal–Wallis computations for the TIC subscales by years working at the hospital demonstrated a statistically significant difference in attitude scores for the subscale “Job behaviour” by length of time [χ 2 (2) = 10.741, p = .01]. Post hoc analyses indicated that MHNs working for less than 5 years at the hospital displayed more favourable attitudes on this subscale than those working between 5–10 years (p = .03) and 11 years and above (p = .02).

Mann–Whitney tests were then conducted to examine responses on the TIC subscales by work setting. Statistically significant differences between attitude scores for the subscales “Job behaviour,” “Self‐efficacy” and “Reactions” by work setting of participants (i.e. acute versus rehabilitation) were identified. MHNs working in acute wards reported more favourable attitudes on “Job behaviour” (U = 1549.50, z = −2.98, p < .001), on “Self‐efficacy” (U = 1522.00, z = −3.11, p < .001) and on “Reactions” (U = 1407.00, z = −3.62, p < .001). These comparisons between subscale mean ranks by years working at the hospital and work setting are illustrated in Table 4.

TABLE 4

Comparison between subscale mean ranks by years working at the hospital and work setting

Subscales		Years working at the hospital			Work setting
		<5 years	5–10 years	11+	Acute	Rehab
	N	50	16	69	66	67
Causes	Mean Rank	74.94	57.69	65.36	65.82	68.16
	Test statistic	χ 2 a  = 3.02, p = .22			U b  = 2133.00, p = .73
Responses	Mean Rank	72.65	62.84	65.83	64.08	69.88
	Test statistic	χ 2 = 1.20, p = .55			U = 2018.00, p = .38
Job Behaviour	Mean Rank	82.27	58.16	59.94	77.02	57.13
	Test Statistic	χ 2 = 10.64, p = .005			U = 1549.50, p < .001
Self‐Efficacy	Mean Rank	75.21	67.63	62.86	77.44	56.72
	Test statistic	χ 2 = 2.90, p = .24			U = 1522.00, p < .001
Reactions	Mean Rank	70.36	61.69	67.75	79.18	55
	Test statistic	χ 2 = 0.60, p = .74			U = 1407.00, p < .001

Kruskal–Wallis test statistic.

Mann–Whitney test statistic.

Qualitative findings

Some demographic characteristics of the ten participants of the focus group are summarized in Table 1. The three themes which emerged from the thematic analysis were Awareness, Unhealthy boundaries and Inhibition.

Theme one: Awareness

MHNs' clarifications of their awareness of some issues related to TIC constituted a good percentage of the entire data set. They expressed a profound understanding of the impact of trauma, of the need for trauma‐related training, of the gaps at the hospital which do not foster favourable attitudes and of how the context of a psychiatric facility can retraumatize clients:

Even coming into hospital is traumatizing. Even some things inherent in the routine. For example, the fact that they have to take a shower at set times.

MHNs also expressed an awareness of not being sufficiently aware of some TIC issues and attempted to “exploit” the focus group to elucidate areas which were not clear to them:

Being aware of our own traumas was the point which struck me when I first heard about it. By awareness of our own traumas do we mean resilience for example?

Traumatic histories were not acknowledged at part with mental health conditions in the aetiology of challenging behaviour. Awareness on this issue is still limited:

But your patients are not psychotic, they know exactly what they are saying, our patients, we get kicked, we get dragged by the hair, we get called names, but they are psychotic. When I know the patient is sane and knowing what they are saying, that way I take it personally yes.

Theme two: Unhealthy boundaries

MHNs demonstrated frustration over ambiguities which were experienced as work‐related traumas. These grey areas often centred around relationships with clients. They talked at length about difficulties they encounter in titrating the adequate amount of engagement with clients to avoid over‐involvement.

Nurses can become very emotionally attached to patients, you can see it's not healthy, it isn't productive, it doesn't really help the patient, and it's not fair on the patient and on the nurse. And then another issue which raises is‐how do you tackle it?

The increased difficulty of maintaining healthy boundaries with long‐term clients in rehabilitation settings was discussed Here professional boundaries more readily become blurred and consequently more emotional reactivity is experienced by clients and MHNs:

In chronic wards clients are there for a very long time, so you tend to consider them not as a client but more as family and what happens is if you get a negative response from that person your reaction won't be as professional, you respond to them as though they are family members.

Theme three: Inhibition

Participants intimated that a set of factors present in the hospital context inhibit their confidence in their professional abilities. Antecedent factors to this sense of inhibition include the blaming culture at the hospital and the lack of clinical supervision and reflective practice. These factors are inhibiting because they do not allow the nurses to turn challenging incidents into learning opportunities:

Whenever something bad happens in the ward we have an internal inquiry and the whole ward is drained. The police are there, and we get all these people coming in with their opinion, and a report is issued, and we never get the feedback of that report. There is never a debriefing after an incident.

MHNs feel they are unfairly blamed for incidents which are inevitable occurrences in a psychiatric setting. Fear of blame makes them retract their steps and refrain from giving optimal care:

If you have a patient under your care and he scratches himself or self‐harms they'll go, ‘Where was the nurse’? I mean sometimes it's impossible to stop a person from scratching themselves.

Ambiguity is another concerning feature of the context of the hospital. The MHNs expressed their anger in having to survive in an environment where policies and guidelines are lacking and procedures are not standardized:

Like for example, very simple things, level 1 supervision for some doctors the opinion is that hospital grounds are part of the hospital, for others it's not. So, one patient cannot go accompanied by a nurse and then another patient can. It's not clear, why is it not for all the patients? No clear guidelines how you should work.

These inhibiting factors within the hospital environment disempower the nursing workforce. MHNs reported ultimately behaving in ways which reinforce these inhibiting factors and further relinquish their power:

Many situations we can deal with them ourselves but telling the doctor is very convenient for us.

Combined major results

Favourable TIC attitudes and their relationship with knowledge

MHNs' total mean score on the ARTIC‐35 was 196.9 out of a maximum score of 245 indicating overall favourable TIC attitudes. During the focus group, MHNs displayed a heightened awareness of the impact of trauma, the need for trauma training and of how their work environment may be retraumatizing to clients. The participants disclosed that they were completely unaware of TIC prior to the distribution of the questionnaire. When asked to interpret their more favourable attitudes compared with those in the study of Marvin and Volino Robinson (2018), the MHNs intimated that a sample constituted exclusively of MHNs in the context of a Mediterranean culture would naturally endorse more favourable TIC attitudes.

Causes

An anomaly was observed for TIC attitudes on this subscale when compared to the other subscales. The MHNs had ambivalent perceptions concerning whether trauma histories are a cause of challenging behaviour. Qualitative findings corroborate this result since participants explained they take clients' challenging behaviour personally unless the clients are psychotic.

Job behaviour

MHNs who had been working for fewer years at the hospital and those working in acute settings were more likely to display favourable attitudes on this subscale. During the focus group, the MHNs proposed that in rehabilitation settings, where clients have been hospitalized for longer, clients eventually are perceived as family members. When professional boundaries become blurred it is more likely that professionals take client problematic behaviour personally. Emotional reactivity to problematic behaviour of clients increases and professionals might be less able to appreciate client behaviour is not a reflection of their behaviour.

Self‐efficacy

Whilst mean scores were higher for the “Self‐efficacy” subscale than they were for the other subscales, during the focus group MHNs disclosed that they were set up to feel inhibited by factors such as the pervasive blaming attitude at the hospital. MHNs working in acute settings reported higher levels of self‐efficacy.

Work setting

MHNs working in acute settings displayed significantly more favourable attitudes on the subscales “Job behaviour,” “Self‐efficacy” and “Reactions.” A shared belief at the hospital that acute wards are highly stressful to work in emerged during the focus group. Participants explained MHNs request to be transferred to calmer rehabilitation wards after a number of years.

Safety

Notwithstanding the numerous media reports addressing the poor physical safety of the hospital none of the MHNs intimated they were concerned about this. Their understanding of safety, a TIC principle advanced by Fallot and Harris (2009), is more gathered around emotional safety. Issues like the ambiguous context and not feeling supported by management were identified as their work‐related traumas.

DISCUSSION

Key findings

One of the most striking findings that emerged was the participants' unawareness of the concept of TIC, prior to involvement in the study. However, as discussed by Hall et al. (2016), health carers' unawareness of aspects related to TIC, may not be as uncommon as one may think. In fact, in their study, a sample of emergency department nurses who followed a single‐day TIC educational package still could not adequately comprehend how the emergency department environment can retraumatize clients. Despite the apparent unawareness of this concept in the current study, the respondents showed positive attitudes towards TIC and displayed a heightened awareness concerning how their work environment may retraumatize clients. This contributes to the debate in the literature concerning the link between TIC‐related knowledge and attitudes. Marvin and Volino Robinson (2018) reported that the hypothesis that more favourable attitudes towards TIC are associated with higher knowledge was not supported. Contrastingly, Palfrey et al. (2019) claimed that a single‐day TIC workshop delivered to allied health professionals working in mental health brought about significant confidence in TIC approaches, significant reduction in perceived barriers to working with trauma and specific changes to practice. One potential contributing factor to the participants' positive attitudes in this study could be related to strong cultural values focused on protection of society's most vulnerable and hospitality. To this extent, Abela et al. (2016) claim that Mediterranean nations keep being perceived as “familialistic” countries and care work is recognized as a solid moral obligation for families. This strengthens the notion that “caring” as a core value is intimately woven into Mediterranean people—a factor which may cultivate positive TIC attitudes in healthcare workers. Whilst further research is needed to explore the relationship between TIC knowledge and attitudes, this study suggests that favourable TIC attitudes are not contingent on knowledge only.

Despite the positive attitudes towards TIC in general, it transpired that trauma histories are not appreciated as understandable causes of clients' challenging behaviour. Whilst further research is required to thoroughly understand what may be causing clients to instigate violence, Beattie et al. (2019) discussed workplace violence from a trauma‐informed perspective among health professionals. In this view, traumatic fight responses were acknowledged as causes of violence and so it is concerning that professionals such as the ones in the current study may not be appreciating this potential cause of aggression.

The link between trauma and challenging behaviour was well‐explained by the MHNs who participated in a similar study by Stokes et al. (2017). The explanation focused on how trauma can spread through nurse‐patient interactions. In this way, MHNs may retraumatize clients, who in turn traumatize the MHN through their reactions and the MHN may then carry that trauma burden to other patients and perpetuate a continued trauma cycle. A new insight that emerged from this current study is the potential consequence of the blurring of boundaries, which may, in itself, be a contributing factor to certain dynamics in the nurse–patient interactions. One such example was provided by some of the participants who stated that oftentimes, those clients who remain in inpatient care for longer periods are perceived by the staff as family members. This may then explain why the participants confessed that they tend to take clients' challenging behaviour personally instead of attributing it to a possible history of trauma. This blurring of boundaries affects the “Job behaviour” component of TIC as it limits the capacity of MHNs to appreciate that challenging behaviour from clients is not intended as a personal insult to the health carer and is not a reflection of the healthcarer's behaviour.

Another important finding was the reported effect of the culture and management of the hospital. Thus, whilst scores were highest for the “Self‐efficacy” subscale, MHNs disclosed that challenging issues such as the pervasive blaming attitude, are inhibiting them. Possibly individual MHNs believe in their own personal potential but much of that is lost when they collectively come together as a professional group within this organization. This issue merits further attention since, as highlighted by the MHNs in the current study, inattention to their own work‐related traumas limits their capacity to attend to the traumas of clients. MHNs need to feel emotionally safe themselves to provide emotional safety to clients. Another take‐home message from this study, therefore, is how the collective self‐efficacy of MHNs in an organization is linked to their feelings of emotional safety.

Interestingly, MHNs working in acute settings expressed more favourable attitudes on most subscales including the “Self‐efficacy” subscale. Upon further exploration, it transpired MHNs working on acute wards eventually request to be transferred to quieter settings when they become burnt out. Thus, possibly, MHNs who still do not feel burnt out at work on acute wards view themselves more positively and hence endorse more positive TIC attitudes and higher feelings of self‐efficacy. Contrastingly, those MHNs working in rehabilitation settings had less favourable self‐efficacy scores. Whilst the blaming attitude identified by the participants may be partly responsible for these lower scores, there may be other contributing factors, such as burnout. Notably, Barbara (2019) quantitatively explored the prevalence of job‐related burnout in mental health professionals at the same psychiatric hospital and reported that burnout for MHNs is more a feature of low personal accomplishment than it is for other professionals. It is, therefore, possible that the less favourable scores on the “Self‐efficacy” subscale for MHNs in rehabilitation settings are indicating MHNs working there may feel burnt out.

Lack of emotional safety and burnout in rehabilitation settings appear to influence the self‐efficacy of the MHNs in this study. This is concerning since the link between self‐efficacy and TIC knowledge and commitment is evident, as explored by Sundborg (2019). Thus, factors that may be leading to lower reported feelings of self‐efficacy in this current study need to be explored in order to enhance a TIC culture. Reduced self‐efficacy, aggravated by factors such as the lack of clinical supervision and reflective practice that was reported in this study is a barrier to endorsement of TIC. The identification of this barrier adds to literature on the subject. For instance, MHNs in the study conducted by Stokes et al. (2017) identified staff being set in their ways, lack of time to interact with clients and altered conscious states of patients as barriers. Additionally, Bruce et al. (2018) report that trauma‐care providers rate time constraints, need for training, confusing information about TIC and worry about retraumatizing clients as other significant barriers to providing TIC.

Strengths and limitations of the study

Effective integration of quantitative and qualitative elements constitutes a strong feature of the study. The satisfactory response rate and the suitability of the ARTIC‐35 as the research tool for the quantitative phase are other assets of the study.

The major limitation was the exclusion of pool nurses from the study population for logistic reasons. The hospital employs a number of pool nurses who are not assigned to specific units but are asked to move daily across units depending on staffing needs. On the one hand, it can be argued that because they work in different settings, these nurses might have a broader appreciation of the overall culture within the organization. On the other hand, since they do not “belong” to any setting they may not have had the opportunity to deeply immerse themselves in the culture of any setting. The participants did not suggest burnt out MHNs seek to join this relieving pool. Therefore, one cannot speculate over their levels of burnout or self‐efficacy.

Moreover, the fact that only ten MHNs consented to participate in the focus group hampered the original plans of recruiting MHNs using theoretical sampling after identifying impinging variables on TIC attitudes.

Since the researcher conducting the focus group was employed as an MHN at the hospital herself, it was necessary to reflect upon and acknowledge any preconceptions which may have been brought into the research. To ensure the analysis truly represented the participants' reality the researcher asked the research assistant who attended the focus group and colleagues to evaluate interpretations. In spite of this, the experience of being an MHN in the same organization may have inevitably influenced some decisions.

What the study adds to the existing evidence

Quantitative studies exploring TIC attitudes in psychiatric hospitals or human service organizations have not to date employed samples made up exclusively of MHNs and they were conducted mostly in the United States and Australia (Bruce et al., 2018; Hall et al., 2016; Jacobowitz et al., 2015; Jordan‐Cox, 2018; Marvin & Volino Robinson, 2018; Palfrey et al., 2019; Sundborg, 2019; Williams & Smith, 2017). This study yields the cross‐sectional TIC attitudes of MHNs employed in a psychiatric hospital in Europe. It identifies the recognition of traumatic histories as causes of challenging behaviour as the domain with least favourable attitude scores among MHNs. Moreover, it reports that work setting is the most impinging characteristic on MHNs' attitudes towards TIC.

The cycle of perpetuated trauma had already been identified by Stokes et al. (2017) as the dynamic through which trauma is spread on wards through nurse‐patient interactions. The quantitative phase of the current study revealed MHNs on rehabilitation wards are less able to appreciate that client behaviour is not a reflection of their own behaviour. In the qualitative phase, MHNs suggested that this is a consequence of professional boundaries becoming more blurred with clients who have come to be regarded as family members. The influence of this enmeshment on the cycle of perpetuated trauma in a new insight afforded by this study.

Implications for mental health nursing

TIC as a framework is gaining popularity. It is likely that psychiatric hospitals all over the world will develop plans in the near future to integrate it as a guiding philosophy to care delivery. The new evidence from this study may considerably improve the outcome of such plans. The study indicates MHNs working in rehabilitation settings endorse less favourable TIC attitudes possibly because of higher levels of burnout and more difficulty with maintaining professional boundaries with patients. Consequently, successful integration of TIC on rehabilitation wards needs to follow concerted efforts to maintain MHN well‐being on the workplace. Any TIC training packages forming part of any TIC implementation strategies need to take stock of the influence of loose boundaries on the cycle of perpetuated trauma and on the importance of appreciating traumatic histories as legitimate causes of challenging behaviour.

More cross‐sectional surveys using larger samples of MHNs and the ARTIC scale are needed to corroborate or challenge the associations reported. Quasi‐experimental studies may be conducted to elicit whether a TIC training package, designed with the consideration of the new evidence of this study, would foster more favourable TIC attitudes or practice among MHNs.

CONCLUSION

This study gained an understanding of MHNs' attitudes towards TIC at the only psychiatric hospital in Malta. It contributes to previous research surrounding this phenomenon by affording a needed European inquiry employing a sample made up exclusively of MHNs. The study revealed overall favourable attitudes among the MHNs with the exception of ambivalent attitudes centred around the impact of traumatic histories in the aetiology of challenging behaviour. It also reported that TIC attitudes are less favourable on rehabilitation wards where MHNs may be more likely to be burnt out and have looser professional boundaries with clients. The study moreover suggests that for psychiatric hospitals to become trauma‐informed the work‐related traumas of employees need to be addressed. These findings should guide future TIC training and implementation initiatives undertaken by psychiatric hospitals.

RELEVANCE STATEMENT

Research has not sufficiently examined TIC attitudes of MHNs. This mixed‐method design explored TIC attitudes among MHNs employed in a psychiatric hospital. Results highlight which variables make MHNs less likely to endorse positive attitudes and which aspects of TIC require more awareness. These findings can improve the outcome of plans to integrate TIC in psychiatric hospitals by identifying challenges in the dissemination of TIC. Any TIC training packages for MHNs need to take stock of the influence of loose boundaries on the cycle of perpetuated trauma and on the importance of appreciating traumatic histories as causes of challenging behaviour.

CONFLICT OF INTEREST

No financial interest or benefit has arisen from the direct applications of this research.

AUTHOR CONTRIBUTIONS

Sarah Cilia Vincenti: Conceived and designed study; acquired, analysed and interpreted data; drafted manuscript. Paulann Grech: Designed study; analysed and interpreted data; revised manuscript critically for intellectual content. Josianne Scerri: Analysed and interpreted data; revised manuscript critically for intellectual content.

ETHICAL APPROVAL

The study was approved by the Faculty of Health Sciences Research Ethics Committee of the University of Malta.

CONSENT OF SUBJECTS

Consent has been obtained from all participants.