Journal of Psychiatric Intensive Care, 16 (1): 3–8

Journal of Psychiatric doi:10.20299/jpi.2019.002

Intensive Care Received 24 July 2017 | Accepted 16 January 2019
© NAPICU 2020

ORIGINAL RESEARCH ARTICLE

Introduction of the Brøset Violence

Checklist on a PICU

Rahna Theruvath-Chalil, John Davies, Stephen Dye

Norfolk & Suffolk NHS Foundation Trust, UK
Correspondence to: Rahna Theruvath-Chalil, Consultant Psychiatrist,
Norfolk & Suffolk NHS Foundation Trust, Woodlands Unit, Ipswich Hospital site, Heath
Road, Ipswich, Suffolk, IP4 5PD, UK; [email protected]
SD, https://orcid.org/0000-0002-0535-5332

Background and aims: This survey pragmatically appraised the utility of the
Brøset Violence Checklist (BVC) and its potential usefulness in guiding proactive
management and interventions to help reduce episodes of violence and/or aggres-
sion within a psychiatric intensive care unit (PICU).
Emphasis was placed on evaluating whether this instrument was completed prior
to an episode of violence and/or aggression, and (through development of an
appropriate action plan) the relationship it had on management plans.
Method: Use of the BVC was introduced on a PICU. Incidents of disturbed
patient behaviour were collated over a 13 week period using the DATIX incident
reporting system. BVC records completed over the 24-hour period prior to any
incident were examined. Usage of risk management plans developed after BVC
completion was identified. Plans were coded as ‘Use of medication’, ‘Environ-
mental interventions’ or ‘Restrictive practice’.
Results: 86 incidences were reported. Results suggested satisfactory completion
of BVC score sheets for all patients. Management plans were noted as being
present and robust for patients whose BVC scores were higher (≥3), as recom-
mended by the tool. It was noted that implementation of restrictive interventions
was less than use of either medication or environmental contingencies within
proposed risk management plans. However, following an episode, management
plans were not reviewed.
Conclusion: An empirically-validated measure to predict potential risk of vio-
lence within a PICU was introduced and appropriate management strategy plans
developed. Incorporating use of a structured short term risk assessment tool was
therefore deemed to be a useful addition to standard procedures.

Key words: Brøset; PICU; aggression

Financial support: This research received no specific grant from any funding agency,
commercial or not for profit sectors
Declaration of interest: None
Ethics: This project was regarded as service evaluation work and, as such, did not need
any ethics committee approval

© NAPICU 2020 3
THERUVATH-CHALIL ET AL.

Introduction term risk assessment and management (Webster et al.

2006; Douglas & Reeves 2010).
Violence or aggression refers to harm to others (that can be Patients admitted to PICU typically present with symp-
expressed both verbally and physically) or damage to the toms of severe mental illness, often with concomitant
fabric of the environment. Anderson & Busman (2002) physical health problems and an established history of
defined aggression as any behaviour directed towards violence and aggression (NAPICU 2016). Short term
another individual that is carried out with the immediate assessment of risk is therefore important, as this enables
intent to cause harm; whereby, violence is intended to staff to focus on timely identification of predictive factors
cause major harm. These behaviours are known to be and prevention of escalation to an adverse incident.
common within acute inpatient mental health settings Within a PICU, patients are predominantly admitted for
(Lozzino et al. 2015) and can often be differentiated in short-term management of behavioural disturbances that
terms of being instrumental (involving goal-directedness occur primarily, although not exclusively, within the con-
and planning) or reactive (response to provocation and text of a relapse of severe and enduring mental illness.
arousal of hostility) (Cornell et al. 1996). Although the majority of patients do not become violent
Incidents of violence and/or aggression appear to within this specialist environment, the prevention and
occur more frequently in mental health settings, as well management of aggressive behaviour needs to be care-
as in the emergency departments of acute hospitals. fully considered. To this end, national PICU guidelines
Between 2013 and 2014 there were 68 683 reported as- recommend that immediate to short term risk assessment
saults committed against NHS staff in England. Of these, should be documented shortly after admission; preferably
69% were in mental health or learning difficulty settings in a structured format (NAPICU 2014).
(NICE 2015). NICE guidance advocates conducting risk assessment
In the UK, the National Institute for Health and Care in partnership with patients. Historical factors need to be
Excellence (NICE) has produced guidance regarding the accorded special consideration as they are known to play
identification and management of risk of violence and an important role in predicting future incidents of violence
aggression (NICE 2015). Prediction of such incidents and and/or aggression (Douglas & Reeves 2010; NICE 2015).
their avoidance is nevertheless acknowledged to be diffi- Best practice suggests that patients should be asked about
cult (Lantta et al. 2016). There are known to be identifiable those factors they consider might increase their risk of
factors that can precipitate patients to enact violence and/ enacting aggression and/or violence. As part of this proc-
or aggression (Soliman & Reza 2001). These can be re- ess, it is recommended that they be encouraged to reflect
lated to personality traits, current mental health difficulties on what interventions have helped in the past to reduce this
and substance misuse (Anderson & Bushman 2002). Simi- possibility (Ockwell & Capital Members 2007). This in-
larly, the attitudes or perceptions of staff towards patients formation should subsequently be documented in a care
and vice versa, a counter-therapeutic physical environ- plan.
ment and organisational-mediated restrictions upon indi- To facilitate a comprehensive assessment of risk, NICE
vidual freedoms are also factors germane to violence and/ guidance also recommends that an empirically-validated
or aggression within mental health settings (Stuart et al. instrument, such as the Brøset Violence Checklist (BVC),
2009). be used to both monitor and reduce incidents of violence
Effective management of violent behaviour within psy- and aggression, and to aid in development of risk manage-
chiatric wards is difficult and challenging. Any episode ment and positive behavioural support plans (NICE 2015).
can create additional disturbance, or cause distress to all The BVC was developed in Norway from a study that
concerned. As a predominantly interpersonal process, vio- examined behaviours that might be germane to violence
lence and/or aggression can exert a deleterious effect on and aggression (Linaker & Busch-Iversen 1995). The six
therapeutic relationships between staff and patients (Gil- most frequently occurring behaviours that were associated
bert et al. 2008; Knowles et al. 2015). Associated with a progression to violence were identified. Logistic
interventions can sometimes involve use of physical re- regression confirmed their predictive validity (Woods &
straint, and in the severest of cases, periods of seclusion Almvik 2002).
(Stuart et al. 2009). These are likely to be traumatic Operationally, the BVC is a six-item instrument that can
experiences for both patients and staff alike, and are be used to identify patients whose behaviours might signal
known to be associated with negative mental health out- their potential to enact violence and/or aggression. Three
comes (Bonner 2007). Research has previously been patient risk characteristics and three behaviours are de-
undertaken examining factors likely to be predictive of scribed. The six variables that are evaluated by the BVC
violence and aggression (Thompson et al. 2009; Reinharth are: Confusion, Irritability, Boisterousness, Physically
et al. 2014). In addition to this, a number of assessment threatening behaviour, Verbally threatening behaviour and
tools have been developed to inform both short and long Attacking objects. Each variable present at assessment

4 © NAPICU 2020
 BRØSET VIOLENCE CHECKLIST ON A PICU

scores one point. For a total score of three or more, a plan following which consensus was reached to continue to use
or intervention should be recorded, implemented and evalu- the BVC to monitor and evaluate all patients admitted.
ated. As the instrument is short, practical and easy to This survey examined episodes of violence, incidents of
administer in routine care it is an acceptable addition to the restrictive practices, BVC completion and associated man-
MDT clinical armamentarium. agement plan implementation. Data pertaining to incidents
Since its initial development, further studies have sought of violence and/or aggression were collated over a 13-
to examine the specificity, sensitivity and inter-rater reli- week period time period.
ability of the BVC. Almvik & Woods (1999) reported ‘Incidents’ were identified from the DATIX reporting
moderate sensitivity, high specificity with an adequate system (DatixWeb Patient Safety Software: https://
level of agreement between independent raters. A further www.datixhealth.com). This is an online form designed to
retrospective study involving a PICU population (Vaaler ensure compliance with data collection and internal and
et al. 2011) demonstrated that the short-term risk of inpa- external reporting requirements. Forms are designed so
tient violence can, to a high degree be predicted by nursing that related questions are batched together and additional
staff using the BVC. sections only appear if required based on information
given.
Intervention plans were identified from the BVC check-
Aims and objectives
list scoring sheet, handover notes and care plans from the
The aim was to evaluate the introduction of the BVC on a electronic patient record system. Interventions were di-
PICU as a means to both predict violence and/or aggres- vided into:
sion and to inform proactive management of potentially
• Offered and accepted ‘as required’ pharmacotherapy
adverse incidents, thus minimising the likely distress for
(medication)
both patients and staff.
• Any form of restrictive practice (intramuscular medi-
One objective of this survey was to examine whether
cation, restraint or seclusion)
the BVC score sheet had been completed for all three
• Non-restrictive environmental interventions (contin-
nursing shifts over the previous 24 hour period for those
gency management), including one-to-one time with
patients who had been involved in violent incidents.
a primary nurse, flexible and judicious use of opportu-
A further objective was to evaluate if, when the BVC
nities for escorted leave, using gym facilities, engaging
scores were high (≥3 points) that, as recommended, there
in OT activities, and interpersonal strategies derived
was record of an associated action plan (medical, environ-
from psychological formulation.
mental, interpersonal, etc.) having been put in place to
help prevent any transition towards aggression.
Results
Within the data collection period, 86 incidents were iden-
Method tified using the DATIX reporting system. These involved
The use of the BVC risk assessment tool was implemented 20 patients.
within a 10-bed, mixed-sex PICU which admits adult There was an 82.6% completion rate (n = 71) for all
patients. The PICU is situated in the East of England and three BVC scores over the 24 hours prior to an incident
the median length of stay is 15 days. Other characteristics (one per nursing shift). Two of the three scores were
of this unit have been described elsewhere (Dye 2017). completed for 12 incidents (14.0%), whilst 3 (3.5%) had
Ahead of this, the authors (JD, SD) and ward manager only one completed. There were two incidents for which
delivered a number of training sessions. This involved the BVC score at the time of incident was not recorded.
teaching staff in small groups using a combination of For those recorded, the BVC score within the previous 24
video case presentations and ‘real-life’ clinical material to hours was observed to be highest at the time of the incident
both identify specific risk behaviours and patient charac- (see Fig. 1).
teristics, and to score each of the six items within the Of the 84 records taken at the time of incident, 60 had
measure. Formal training was undertaken to ensure ad- BVC scores of three or more. The types of incident are
equate agreement between staff members. In those shown in Table 1. For these incidents, the corresponding
instances where discrepancies in scoring were identified, a proposed intervention plan was found to be documented in
case discussion was implemented to resolve areas of disa- all cases and the incident was managed in accordance with
greement. the agreed care plan. Types of interventions are not mutu-
Incorporated into routine clinical practice, the BVC was ally exclusive, and Figure 2 displays their proportionate
rated by nurses during each of the three PICU shift cycles. usage. This shows that whilst medication is commonly
Scores were subsequently discussed and handed over at used, consideration as to the potential causal role environ-
each change of duty. This process was piloted for a month, mental factors is considered wherever possible, and inter-

© NAPICU 2020 5
THERUVATH-CHALIL ET AL.

30 16

25 14

12

Number of incidents
20
Number of incidents

10
15
8
10
6
5
4

0 2
Absent 0 1 2 3 4 5 6
0
BVC score A B C D E F G H I J K L M N O P Q R S T

Patient
Fig. 1. BVC score at the time of incident.
Fig. 3. Patients contributing to incidents.

Table 1. Types of incidents identified

recurrence of these episodes (i.e. an alteration in care plan)
Type of incident No. of incidents (BVC ≥ 3) for any patient.
Physical assault on staff 17
Criminal damage 6 Discussion
Verbal aggression 20
Boisterous behaviour 6 This survey examined the clinical effectiveness of using
Physical assault on other patients 12 the BVC to inform risk management plans for acutely
unwell and behaviourally disturbed patients on a UK-
based PICU.
Completion of the BVC scoring sheet, and subsequent
Environmental
factors/ development of care plans for those perceived to be at high
Restriction
Contingency risk for potentially enacting violence and/or aggression
management was found to be good. The PICU staff were noted to be
proactive in identifying individuals with potential to de-
ploy either reactive or instrumental violence. Whilst
patient-centred risk management plans were detailed in
their clinical record, this was not consistently documented
on the BVC score sheet.
Findings from this survey appear to indicate that epi-
sodes of behavioural disturbance are qualitatively
associated with higher scores on the BVC. Following
attainment of a score greater than or equal to 3, there is an
expectation that staff will intervene proactively in an
attempt to reduce the likelihood of a violent incident
Medication
occurring. In practice, it was encouraging to note that a
Fig. 2. Interventions noted in documentation. plurality of risk reduction strategies was considered, in-
cluding the judicious use of ‘as required’ medication,
ventions titrated to address these (e.g. time away from the interpersonal interventions and the management of envi-
ward, relocating to a low stimulus area, activities that ronmental contingencies. The finding that more
redeploy attention away from perceived provocation). non-restrictive interventions were planned may provide
A number of specific patients displayed repeated some evidence to support the assertion that staff were
incidences of disturbed behaviour (see Fig. 3). Although being more responsive to patients’ behaviours before these
care plans identified measures to be considered should escalated to the point where it became necessary to imple-
their aggressive or violent behaviour become exacerbated, ment physical intervention and associated restrictions.
there appeared no identified documentation detailing a Within this survey, the BVC was rated prior to each
consideration of a change of strategies to prevent future nursing handover; hence our prediction interval for an

6 © NAPICU 2020
 BRØSET VIOLENCE CHECKLIST ON A PICU

incident of violence and/or aggression was from one tions have utility in decreasing both the incidence of
handover to another. An empirically-derived cut off score violence and/or aggression within the PICU, and the use of
of 3 or more is predictive of a high risk for violence restrictive practices.
(Almvik et al. 2000). It was nevertheless noteworthy in
our survey that a number of incidents (28.6%) occurred
following the allocation of lower scores. This indicates a Conclusion
need to consider developing proactive risk prevention and Incorporation of the BVC into routine clinical practice
management plans even though the BVC might suggest within our PICU was successful. This was demonstrated
that a violent incident is less likely to occur. by a high completion rate of BVC checklists, documented
The BVC focuses predominantly on patient behaviours associated care plans that were followed prior to and
for predicting the likelihood of violent incidents. We following any incidents occurring. More thought needs to
recognise there might be other variables that can influence be given to amending care plans subsequently which may
the expression of overt aggression. Such behaviours, for prevent recurrence of restrictive interventions.
example, might reflect an understandable ‘reactive’ re- It is clear that no single intervention or set of interven-
sponse to preventive measures being put in place; such as tions will act as a panacea to reduce rates of conflict and
boundary setting, the denial of requests, or restrictions on containment; rather, a number of different strategies will
the availability of leave. need to be identified, deployed and evaluated specifically
to meet the individual needs of acutely unwell and dis-
Clinical implications for our PICU tressed patients. Subsequent to this survey we have
Whilst staff appeared adept at identifying incidents using developed a strategy that mirrors national guidance and
BVC, for some patients the prevention of repetition pre- examples of best practice. This includes structured risk
sented significant challenges, despite there being a care assessment leading to an MDT based psychological for-
plan. This finding was noted by the multidisciplinary mulation in order to promote least restrictive practice.
team through the process of ongoing case discussion and
weekly review meetings. On the basis of this survey, a
further risk management strategy has been introduced References
which emphasises more detailed consideration of the in- Almvik, R. and Woods, P. (1999) Predicting inpatient violence using
tra, inter and extra-personal factors likely to underpin the Brøset violence checklist (BVC). International Journal of
Psychiatric Nursing Research, 4: 498–505.
such incidents. Almvik, R., Woods, P. and Rasmussen, K. (2000) The Brøset
Augmenting the routine use of the BVC, we have Violence Checklist. Journal of Interpersonal Violence, 15(12):
introduced a structured approach to assessing those patients 1284–1296. https://doi.org/10.1177/088626000015012003
Anderson, C. A. and Bushman, B. J. (2002) Human Aggression.
identified as being at particular risk of enacting violence Annual Review of Psychology, 53: 27–42. https://doi.org/10.1146/
and/or aggression (i.e. DICES® Risk Assessment and annurev.psych.53.100901.135231
Management System; The Association for Pyschological Bonner, G. (2007) The psychological impact of restraint in acute
mental health settings: the experiences of staff and inpatients.
Therapies; https://www.apt.ac/dices-risk-assessment-in- PhD thesis, University of West London.
mental-health-training.html). Complementing this proc- Cornell, D. G., Warren, J., Hawke, G., Stafford, E., Oram, G. and
ess, a team-based integrative psychological formulation Pine, D. (1996) Psychopathology of instrumental and reactive
violence offenders. Journal of Consulting & Clinical Psychology,
(Johnstone 2014) has been developed to identify both the 64: 783–790. https://doi.org/10.1037/0022-006X.64.4.783
putative causes and precursors to adverse incidents. Where Department of Health (2014) Positive and Proactive Care: Reduc-
patients are deemed to be at significant risk of repeatedly ing the need for restrictive interventions. https://www.gov.uk/
government/publications/positive-and-proactive-care-reducing-
enacting violence and/or aggression, or where it has been restrictive-interventions/
necessary to contain this behaviour via restrictive practice, Douglas, K. S. and Reeves, K. A. (2010) Historical-Clinical-Risk-
a psychologically-informed risk formulation and manage- Management-20 (HCR-20) violence risk assessment scheme:
rationale, application, and empirical overview. In: Otto, R., Doug-
ment plan based on the principals espoused by SAFE las, K. (eds). Handbook of Violence Risk Assessment, pp.
(Shared Assessment, Formulation and Education: Meaden 147–185. Routledge.
& Hacker 2011) is developed. This facilitates positive Dye, S. (2017) Can mental health clusters be replaced by patient
typing? British Journal of Healthcare Management, 23(5): 229–
risk-taking, the social reinforcement of adaptive behav- 237. https://doi.org/10.12968/bjhc.2017.23.5.229
iours, and a timely reintegration to the communal areas. Gilbert, H., Rose, D. and Slade, M. (2008) The importance of
This is used to inform interpersonal interventions that are relationships in mental health care: a qualitative study of service
users’ experiences of psychiatric hospital admission in the UK.
directed primarily towards early identification and pre- BMC Health Services Research, 8(1): 92–104. https://doi.org/
vention of violence that negate the need for restrictive 10.1186/1472-6963-8-92
practices and is in keeping with Department of Health Johnstone, L. (2014) Using formulation in teams. In: Johnstone, L.,
Dallos, R. (eds). Formulation in Psychology and Psychotherapy:
strategy (Department of Health 2014). Further evaluation Making sense of peoples problems. 2nd edn, pp. 216–242.
is needed to provide evidence on whether these interven- Routledge. https://doi.org/10.4324/9780203380574

© NAPICU 2020 7
THERUVATH-CHALIL ET AL.

Knowles, S. F., Hearne, J. and Smith, I. (2015) Physical restraint from service users, carers and professionals, pp 48–55.
and the therapeutic relationship. Journal of Forensic Psychi- Routledge.
atry & Psychology, 24(4): 461–475. https://doi.org/10.1080/ Reinharth, J., Reynolds, G., Dill, C. and Serper, M. (2014) Cog-
14789949.2015.1034752 nitive predictors of violence in schizophrenia: a meta-analytic
Lantta, T., Anttila, M., Kontio, R., Adams, C. E. and Välimäki, M. review. Schizophrenia Research: Cognition, 1(2): 101–111.
(2016) Violent events, ward climate and ideas for violence https://doi.org/10.1016/j.scog.2014.06.001
prevention amongst nurses in psychiatric wards: a focus group Soliman, A. E. and Reza, H (2001) Risk factors and correlates of
study. International Journal of Mental Health Systems, 10(27): violence among acutely ill adult psychiatric inpatients. Psych-
1–10. https://doi.org/10.1186/s13033-016-0059-5 iatric Services, 52(1): 75–80. https://doi.org/10.1176/appi.ps.52.
Linaker, O. M. and Busch-Iversen, H. (1995) Predictors of imminent 1.75
violence in psychiatric inpatients. Acta Psychiatrica Scandin- Stuart, D., Bowers, L., Simpson, A., Ryan, C. and Tzigilli, M.
avica, 92: 250–254. https://doi.org/10.1111/j.1600-0447.1995. (2009) Manual restraint of adult psychiatric inpatients: a
tb09578.x literature review. Journal of Psychiatric & Mental Health
Lozzino, L., Ferrari, C., Large, M., Nielssen, O. and de Girolama, G. Nursing, 16(8): 749–757. https://doi.org/10.1111/j.1365-2850.
(2015) Prevalence and risk factors of violence by psychiatric 2009.01475.x
acute inpatients: a systematic review and meta-analysis. PLoS Thompson, L., Wilson, J. and Robinson, L. (2009) Predictors of
One, 10(6): e0128536. https://doi.org/10.1371/journal.pone.012 violence in mental illness: the role of substance abuse and
8536 associated factors. Journal of Forensic Psychiatry & Psychol-
Meaden, A. and Hacker, D. (2011) Problematic and Risk Behav- ogy, 20(6): 919–927. https://doi.org/10.1080/1478994090317
iours in Psychosis: A shared formulation approach. Routledge. 4097
https://doi.org/10.4324/9780203834701 Vaaler, A. E., Iversen, V. C., Morken, G., Flovig, J. C., Palmstierna,
NAPICU (2014) National Minimum Standards for Psychiatric Inten- T. and Linaker, O. M. (2011) Short-term prediction of threatening
sive Care in General Adult Services. NAPICU. https://doi.org/ and violent behaviour in an acute psychiatric intensive care unit
10.20299/napicu.2017.001 based on patient and environment characteristics. BMC Psy-
NAPICU (2016) Guidance for Commissioners of Psychiatric Inten- chiatry, 11: 44. https://doi.org/10.1186/1471-244X-11-44
sive Care Units (PICU). NAPICU. https://doi.org/10.20299/ Webster, C. D., Nicholls, T. L., Martin, M.-L., Desmarais, M. A. and
napicu.2016.001 Brink, J. (2006) Short-term assessment of risk and treatability
NICE (2015) Violence and aggression: Short term management in (START): the case for a new structured professional judgment
mental health and community settings. NICE Guideline NG10. scheme. Behavioral Sciences & the Law, 24: 747–766. https://
https://www.nice.org.uk/guidance/ng10 doi.org/10.1002/bsl.737
Ockwell, C. and Capital Members (2007) Restraint: a necessary Woods, P. and Almvik, R. (2002) The Brøset violence checklist
evil? In: Hardcastle, M., Kennard, D., Grandison, S. Fagin, L. (BVC). Acta Psychiatrica Scandinvica, 106 (suppl. 412): 103–
(eds). Experiences of Mental Health In-patient Care: Narratives 105. https://doi.org/10.1034/j.1600-0447.106.s412.22.x

8 © NAPICU 2020