bs_bs_banner

International Journal of Mental Health Nursing (2017) 26, 513–524 doi: 10.1111/inm.12389

S PECIAL I SSUE
De-escalation techniques used, and reasons for
seclusion and restraint, in a forensic psychiatric
hospital
Satu Kuivalainen,1 Katri Vehvil€ainen-Julkunen,2 Olavi Louheranta,1 Anu Putkonen,1 Eila
Repo-Tiihonen1 and Jari Tiihonen1,3,4
1
Department of Forensic Psychiatry, University of Eastern Finland, 2 Department of Nursing Science, University of
Eastern Finland, Kuopio, 3National Institute for Health and Welfare, Helsinki, Finland, and 4Department of Clinical
Neuroscience, Karolinska Institutet, Stockholm, Sweden

ABSTRACT: In Finland, the Mental Health Act determines the legal basis for seclusion and
restraint. Restrictive measures are implemented to manage challenging situations and should be
used as a last resort in psychiatric inpatient care. In the present study, we examined the reasons
for seclusion and restraint, as well as whether any de-escalation techniques were used to help
patients calm down. Seclusion and restraint files from a 4-year period (1 June 2009–31 May 2013)
were retrospectively investigated and analysed by content analysis. Descriptive statistics were
calculated. A total of 144 episodes of seclusion and restraint were included to analyse the reasons
for seclusion and restraint, and 113 episodes were analysed to examine unsuccessful de-escalation
techniques. The most commonly-used techniques were one-to-one interaction with a patient
(n = 74, 65.5% of n = 113) and administration of extra medication (n = 37, 32.7% of n = 113).
The reasons for seclusion and restraint were threatening harmful behaviour (n = 51, 35.4% of
n = 144), direct harmful behaviour (n = 43, 29.9%), indirect harmful behaviour (n = 42, 29.1%),
and other behaviours (n = 8, 5.6%). In general, the same de-escalation techniques were used with
most patients. Most episodes of seclusion or restraint were due to threats of violence or direct
violence. Individual means of self-regulation and patient guidance on these techniques are needed.
Additionally, staff should be educated on a diverse range of de-escalation techniques. Future
studies should focus on examining de-escalation techniques for the prevention of seclusion.
KEY WORDS: de-escalation, forensic psychiatry, prevention, restraint, seclusion.

(Bowers et al. 2010; Happell & Koehn 2010; Huck-

INTRODUCTION
shorn 2006; Kontio et al. 2012; MIELI plan 2009;
In recent years, there has been a growing movement Noorthoorn et al. 2016; Putkonen et al. 2013; Smith
towards reducing the use of restrictive measures in psy- et al. 2015; Wieman et al. 2014). Restrictive measures
chiatric care internationally, as well as in Finland are expensive and jeopardize the safety of, and rela-
tionships between, patients and staff (LeBel & Gold-
stein 2005; Substance Abuse and Mental Health
Correspondence: Satu Kuivalainen, Department of Forensic Psy-
chiatry, University of Eastern Finland, Niuvankuja 65, FI-70240 Services Administration 2011). The use of restrictive
Kuopio, Finland. Email: satu.kuivalainen@niuva.fi measures also presents an ethical dilemma for staff
Satu Kuivalainen, RN, MSc. (Kontio et al. 2010). In the present study, seclusion
Katri Vehvil€ainen-Julkunen, RN, PhD.
Olavi Louheranta, ThM, PhD. refers to locking a patient alone in a room that he or
Anu Putkonen, MD, PhD. she cannot leave without the assistance of staff. Res-
Eila Repo-Tiihonen, MD, PhD.
Jari Tiihonen, MD, PhD. traint refers to mechanically tying a patient to a bed
Accepted August 10 2017. using softened leather straps (Crenshaw & Francis

© 2017 Australian College of Mental Health Nurses Inc.

514
1995; Steinert & Lepping 2009). From a human rights
perspective, seclusion and restraint are justified only as
a last resort. Based on this viewpoint, these techniques
are applied only in emergencies to maintain safety in
circumstances where there is a threat of immediate or
imminent harm, and after alternative measures have
been implemented (Huckshorn 2006; Kumble &
McSherry 2010; Restraint Reduction Network 2017).
BACKGROUND
Preventive measures prohibiting seclusion and
restraint
In clinical practice, restrictive measures might partially
prevent unwanted behaviours, but do not fully elimi-
nate them, at least when used independently (Kuiv-
alainen et al. 2014). This finding has also been
reported with special observation (Stewart & Bowers
2012; Stewart et al. 2011). Therefore, therapeutic mea-
sures, as well as short-term techniques, are needed to
manage challenging situations. At least two types of
short-term measures to prevent seclusion and restraint
have been identified: first, restrictive measures, includ-
ing external measures to control patients’ behaviour,
and second, nursing measures intended to help patients
control and regulate their emotions internally without
restriction. Restrictive measures include physical
restraint, enforced intramuscular medication, and refer-
ral to a psychiatric intensive care unit (Stewart et al.
2011). One problem with using restrictive measures to
manage aggression is that they remove the power and
responsibility for the challenging situation from the
patient, who should be able to regulate his/her own
feelings in the future. Alternative measures for seclu-
sion and restraint are worth trying, because they can
occasionally be used as a substitute for seclusion (Bow-
ers et al. 2011).
De-escalation is a compilation of nursing techniques
that are used to help patients with the short-term man-
agement of violent behaviour. It has been recognized
in numerous guidelines, training programmes, and
handbooks on the management of hospital violence
(National Institute for Health and Care Excellence,
2015; Richter 2006; Roberton et al. 2012). In the pre-
sent study, de-escalation was defined similarly to the
UK National Institute for Health and Care Excellence
(2015) guidelines for the short-term management of
aggression and violence. Specifically, de-escalation
involves recognizing the early signs of agitation, anger,
and aggression, and responding to the patient’s anger
S. KUIVALAINEN ET AL.
in an appropriate way. In many cases, these techniques
include interacting with the patient to avoid confronta-
tion, and offering the patient choices in a difficult situ-
ation. The goal is to redirect the patient to reach a
calmer state, without confrontation or provocation. A
close working relationship with patients from the
beginning of care has been reported to be crucial to
de-escalation (National Institute for Health and Care
Excellence 2015). Although de-escalation is a widely-
recognized and -accepted compilation of techniques, it
has been criticized, because it mostly relies on verbal
and cognitive techniques (Sutton et al. 2013). Further-
more, there is a lack of evidence regarding its effective-
ness (Roberton et al. 2012), and the theoretical basis of
aggression has been overlooked in most studies (Rober-
ton et al. 2012). An exception of that is a study from
Australia that highlights the significance of interaction
between individuals and their environment on prevent-
ing aggression (McCann et al. 2015).
The Brøset Violence Checklist (BVC) is a risk-
assessment method that represents a successful exam-
ple of structured de-escalation techniques (Almvik
et al. 2000). The application of the BVC has led to
reduced hospital violence in acute psychiatry wards
(Abderhalden et al. 2008; Van de Sande et al. 2011). In
addition to risk assessment, this method includes a list
of preventive nursing techniques for high-risk situa-
tions. When using the method, staff implement a com-
bination of techniques, which likely explains its
effectiveness in reducing violence. The preventive tech-
niques included in the BVC are easy to implement.
Examples of techniques include general conversations
aiming to reduce aggression, one-on-one or group out-
door walks, reduction of demands, relaxation exercises,
confrontation with ward rules, discussion of risks with
the patient, and increases in medication dosage
(Abderhalden et al. 2008). In addition to reducing vio-
lence, the BVC has also been found to decrease the
use of restrictive measures in studies by Abderhalden
et al. (2008) and Van de Sande et al. (2011).
Sensory modulation is described as the ability to
automatically regulate and organize sensory experiences
in everyday life, leading to thoughtful behaviour (Te
Pou o te Whakaaro 2011). Problems with sensory mod-
ulation could lead to the overregulation or underregu-
lation of emotions, which is related to arousal levels
(Sutton et al. 2013). In psychiatric care, a sensory mod-
ulation approach might be targeted to help a patient
with overregulation or underregulation of emotions
(Champagne & Stromberg 2004). A sensory approach
is usually performed in a room where there is an
© 2017 Australian College of Mental Health Nurses Inc.
SECLUSION, RESTRAINT AND DE-ESCALATION
assortment of equipment to help individuals relax using
different senses (Chalmers et al. 2012; Novak et al.
2013; Sutton et al. 2013). The preliminary results
regarding the use of sensory modulation for managing
stressful emotional experiences are promising, espe-
cially when multiple senses are involved (Novak et al.
2013; Sutton et al. 2013). To our knowledge, there is a
lack of evidence on why sensory rooms have been per-
ceived to be helpful, although it is also known that sen-
sory rooms are not considered helpful to all patients
(Sutton et al. 2013).
Reasons for seclusion and restraint
Internationally, the use of restrictive measures and the
types of measures vary (Noorthoorn et al. 2015; Stein-
ert & Lepping 2009; Steinert et al. 2010), despite the
fact that studies have reported quite similar reasons for
seclusion and restraint. These reasons include threats
or use of violence against other people or oneself,
aggression towards property, attempts to escape from
care, inability to care for oneself (Paavola & Tiihonen
2010; Raboch et al. 2010), and agitation/disorientation
(Kaltiala-Heino et al. 2003; Keski-Valkama et al. 2009;
Larue et al. 2010).
In addition to threatening and inciting violence
(Bowers et al. 2011; Noda et al. 2013; Paavola & Tiiho-
nen 2010; Raboch et al. 2010), agitation/disorientation
has been found to be a main reason for seclusion and
restraint in some studies (Kaltiala-Heino et al. 2003;
Keski-Valkama et al. 2009; Larue et al. 2010). The
reported rate of threatening or inciting actual violence
has varied between approximately one-third and two-
thirds of incidents (Bowers et al. 2011; Paavola &
Tiihonen 2010). An even greater number of patients
have been reported to be agitated or disorientated
while secluded (92.6%, with or without restraint). In
contrast, 77.5% of secluded patients behaved aggres-
sively (Larue et al. 2010).
In addition to the imminent reasons for seclusion
and restraint, other factors have been associated with a
higher use of seclusion and restraint in psychiatric
units. Organizational factors, such as type of hospital
ward (psychiatric emergency department or intensive
care unit), have predicted a higher use of these actions
(De Benedictis et al. 2011). Greater use of seclusion
has also been connected to characteristics of the physi-
cal environment, such as a greater number of patients
in the building, presence of an outdoor space or gar-
den, and special safety measures (e.g. presence of spe-
cial communication and warning systems; Van der
© 2017 Australian College of Mental Health Nurses Inc.
515
Schaaf et al. 2013). Furthermore, the perception of the
ward atmosphere among staff members has been
related to a higher use of seclusion and restraint.
Specifically, staff perceptions of a higher level of
expression of anger and aggression among staff mem-
bers on the ward, of greater self-destructive behaviour
among patients, and of insufficient safety measures in
the hospital, have been connected to a higher use of
seclusion and restraint (De Benedictis et al. 2011).
Staff characteristics, such as having more female and
less male nurses during a shift, and less heterogeneity
in the team’s work experience, have been connected to
greater use of seclusion (Janssen et al. 2007). The vari-
ations in the use of seclusion are complicated. No sin-
gle characteristic of the ward or patients treated there
have been able to explain the differences in seclusion
use (Janssen et al. 2013).
In addition to regular care, techniques to manage
challenging situations in clinical practice are also
needed. The use of seclusion and restraint has been
successfully reduced in recent years in Finland
(Makkonen et al. 2016; Rainio & R€aty 2015). However,
in order to develop techniques to manage challenging
situations with forensic psychiatric patients, it is impor-
tant to conduct an in-depth investigation of what
occurred in clinical situations that resulted in the use
of restrictive measures. This type of in-depth, descrip-
tive information could help further reduce the use of
restrictive measures and hospital violence, and high-
light the need for education. The aim of the present
study was thus to describe the de-escalation techniques
used in psychiatric care, as well as the reasons for
seclusion and restraint.
METHODS
Study design
The present study was a cross-sectional, retrospective,
descriptive study conducted in one of two forensic psy-
chiatric state hospitals in Finland (Polit & Beck 2008).
The hospital has 284 beds for adult patients. In Fin-
land, the state hospitals admit three groups of patients
with severe mental illness and a history of serious and/
or persistent violent behaviour. The first group consists
of patients who have committed a crime, but were
found not guilty by reason of insanity (forensic psychi-
atric patients). The second group includes patients
whose treatment would be difficult or dangerous in
local psychiatric hospitals (difficult or dangerous
patients). The third group consists of persons who have
516
been referred for a forensic mental examination.
Patient care is evidence based and follows the Finnish
Current Care Guidelines for Schizophrenia. According
to these guidelines, the contents of care include a thor-
ough examination of the patient’s physical health and
mental state, pharmacotherapy, psychosocial therapy,
family intervention, patient education, and securing of
continuity in the care chain. The study hospital has
shown high adherence to these guidelines (Tuppu-
rainen et al. 2014).
The criteria for the use of seclusion in the Mental
Health Act of Finland (1116/1990) extend beyond vio-
lent behaviour or threats thereof. A patient who is
committed to involuntary psychiatric care might also be
secluded if his or her behaviour obstructs the care of
other patients or jeopardizes his or her own safety. The
disruption of other patients’ care must be severe
enough to legitimize the use of seclusion. According to
the Act, a patient might also be secluded if he or she
would likely cause significant damage to property, or if
there is another special therapeutic reason for seclu-
sion. In the Act, the criteria for mechanical restraint
are stricter than the criteria for seclusion: restraint is
only allowed in the case of self-harm or harm to others.
The restrictive measures applied must comply with the
principle of minimum restriction, which in clinical
practice means that the use of one-to-one observation
and physical restraint should be assessed before imple-
menting seclusion and mechanical restraint (Mental
Health Act (1116/1990)).
In 2012, a total of 290 adult patients received invol-
untary psychiatric treatment in the study hospital.
According to the International Classification of Dis-
eases, 10th revision, 87.74% of patients had a
schizophrenia-group diagnosis as their main diagnosis,
8.96% had a personality disorder, and 0.3% had a
mood disorder. The most common comorbid diagnoses
in 2012 were alcohol or other drug dependence (61.3%
of the patients) and personality and behaviour disor-
ders (33.7%). Intellectual disability was a comorbid
diagnosis in 5% of patients. A total of 63 forensic men-
tal examinations were performed in 2012. The individu-
als receiving forensic mental examination were most
commonly accused of homicide (46%) and other vio-
lent crimes (34%). Sexual crimes (4%), criminal mis-
chief (for example arson, 8%), and crimes targeting
property (6.4%) were also suspected of persons in
forensic mental examination. Crimes among forensic
psychiatric patients were consistently reported, as
expected. In total, 25.8% of difficult or dangerous
patients had a history of crime, even if the act was not
S. KUIVALAINEN ET AL.
connected to their current care in the hospital. In
2012, the mean length of stay for forensic psychiatric
patients was 8 years and 10 months, and 5 years and
4 months for difficult and dangerous patients, respec-
tively, and 58 days for patients receiving forensic men-
tal examination.
Ethical approval for the study was obtained from
the Research Ethics Committee of the Northern Savo
Hospital District (141//2008). The study was included
within a development project of Niuvanniemi Hospital
(Kuopio, Finland). The main objective of this project
was to reduce the use of seclusion and restraint in the
entire hospital, and the prerequisites for evidence-
based practice were thus implemented within the
research project. The studies were conducted at the
hospital level instead of the single patient level, and
the minimum amount of information for each patient
was collected as data, which remained anonymous. The
Research Ethics Committee of the Northern Savo
Hospital District stated that the study was scientifically
justified and that essential ethical considerations were
properly taken into account. The research plan fol-
lowed the principles of good clinical practice and
adhered to the legislation concerning scientific studies.
Data collection
Seclusion and restraint forms from a 4-year period (1
June 2009–31 May 2013) were investigated in the pre-
sent study. The forms were part of official patient files.
In addition, the patient files for seclusion or restraint
episodes that occurred on the same day were also
investigated to determine which de-escalation tech-
niques, if any, had been implemented to help patients
prior to the use of restrictive methods. The data were
narrative descriptions. In addition, background infor-
mation was collected on patients’ sex and legal status.
The number of cases in each category was counted.
Purposive sampling was performed by including data
from the first seclusion or restraint episode per patient
during the study period (Burns & Grove 2009). This
sampling method was used to ensure that the data
were representative and included a wide variation of
seclusion and restraint episodes from different units
and patient groups. The goal was to describe these
phenomena (Polit & Beck 2008). Descriptive statistics
and v2-test were performed using IBM SPSS Statistics
version 20 (IBM Corp., Armonk, NY, USA; Burns &
Grove 2009; Polit & Beck 2008). Data collection is rep-
resented in Figures 1 and 2. One seclusion or restraint
episode for each of the 175 total secluded or restrained
© 2017 Australian College of Mental Health Nurses Inc.
SECLUSION, RESTRAINT AND DE-ESCALATION
31 seclusion/restraint
decisions made in
another organization →
excluded
FIG. 1: Flowchart of data collection. Inclusion and exclusion criteria for the seclusion and restraint reasons analysis.
individuals was extracted from the data in the first
phase.
Before the final analysis of reasons for seclusion and
restraint, 31 seclusion and restraint periods were
excluded, because another organization had made the
decision to use seclusion or restraint. A total of 144
episodes of seclusion and restraint were included in
the analysis. In the analysis of de-escalation techniques,
only incidents in which de-escalation could have been
implemented were included. Due to the unexpected
nature of the reasons, 31 of the seclusion and restraint
episodes were excluded from the analysis of de-escala-
tion techniques.
Analysis
Qualitative content analysis was performed to investi-
gate the de-escalation techniques reported in the
nursing staff’s narrative descriptions, which were
included in the patient files. The material was
divided into smaller sections, and the de-escalation
techniques were named, coded, and grouped into cat-
egories. In the following phase, the categories were
grouped into larger concepts based on the content
that they represented (Polit & Beck 2008). Deductive
content analysis was implemented when investigating
© 2017 Australian College of Mental Health Nurses Inc.
517
1493 seclusion/restraint
episodes from 175 individual
patients from
1 June 2009 to 31 May 2013
First episode/patient
included (n = 175)
144 seclusion/restraint
decisions made in
the present organization →
included in the analysis of
reasons for seclusion/restraint
the reasons for seclusion and restraint (Elo & Kyng€as
2008). The intent was to test whether the reasons for
seclusion and restraint were the same in forensic psy-
chiatry as those identified in previous studies in gen-
eral psychiatry. In the first phase, the reasons for
seclusion and restraint were classified according to
the six categories previously used by Kaltiala-Heino
et al. (2003) and Keski-Valkama et al. (2009): actual
violence, threats of violence, property damage, agita-
tion/disorientation, undefined aggression/dangerous-
ness, and unclassified. After the initial classification
and discussion among the multiprofessional group of
authors, the analysis was furthered by re-examining
the data and reformulating the categories into the
four presented in the Results section. To evaluate
and enhance the reliability of the final coding, a pro-
portion of the seclusion and restraint episodes (20%)
were coded separately by two of the authors (SK and
OL; Burns & Grove 2009; Hallgren 2012; Polit &
Beck 2008). Agreement about the final codes was
determined by the kappa statistic. The 95% confi-
dence interval (95% CI) for the kappa was obtained
by bias-corrected, accelerated bootstrapping. Cohen’s
kappa was 0.91 (95% CI: 0.78–0.99). The level of
agreement was considered very good if kappa >0.80
(Landis & Koch 1977).
518 S. KUIVALAINEN ET AL.

1493 seclusion/restraint
episodes from 175 individual
patients from
1 June 2009 to 31 May 2013

First episode/patient
included (n = 175)

31 seclusion/restraint
decisions made in
another organization →
excluded

144 episodes included and reason

for seclusion/restraint analysed

Reason predictable in 113

Reason unexpected in
episodes → included in the
31 episodes → excluded
analysis of de-escalation
techniques

FIG. 2: Flowchart of data collection. Inclusion and exclusion criteria for the de-escalation analysis.

The first seclusion or restraint episode per patient

RESULTS
during the study period was included in the study, for
During the 4-year study period, 549 individuals were a total of 175 of such episodes. In 31 of the 175 epi-
examined or treated in the study hospital (Table 1). sodes, another organization had made the decision
There was a total of 1493 seclusion (n = 1301) and regarding seclusion or restraint, and these episodes
restraint (n = 192) episodes. were thus excluded from the final dataset. Overall, 137
(95.1%) of the 144 included episodes involved seclu-
sion, and the remaining 7 (4.9%) were episodes of
TABLE 1: Number of treated and examined patients in the study restraint. In the analysis, seclusion and restraint epi-
hospital during the study period (1 June 2009–31 May 2013) by sex sodes were considered together because of the small
(n = 549) number of restraint episodes. Nearly half (n = 68,
Male Female Total 47.2%) of the patients were those whose treatment
n n n would be difficult or dangerous in local psychiatric hos-
pitals. The second most common group consisted of
Patients who were found 203 17 220
not guilty by reason of insanity patients who had committed a crime, but were found
Difficult or dangerous patients 154 47 201 not guilty by reason of insanity (n = 43, 29.9%). The
from local hospitals remaining third group included persons who had been
Forensic mental examinations 167 26 193 referred for a forensic mental examination (n = 33,
Total 524 (59†) 90 (6†) 614 (65†)
22.9%).
†
Number of patients, who were at first on forensic mental exami- The secluded or restrained patients were male in
nation and then ordered to treatment for the reason of insanity. 105 (72.9%) cases and female in 39 (27.1%). When the

© 2017 Australian College of Mental Health Nurses Inc.

SECLUSION, RESTRAINT AND DE-ESCALATION
sex and patient groups were cross-tabulated, male sex
was related to the group of patients who had commit-
ted a crime, but were found not guilty by reason of
insanity. In contrast, female sex was associated with
patients who were considered difficult or dangerous to
treat in local hospitals. This finding was statistically sig-
nificant (v² = 14.299, df = 2, P = 0.001; Table 2).
The most common target for harmful behaviour was
another person (n = 67, 46.5%). Harmful behaviour
was targeted towards oneself only in 35 (24.3%) epi-
sodes, and towards other people and oneself in 10
(6.9%) episodes. Furthermore, 10 (6.9%) episodes were
directed towards objects. Twenty-two (15.3%) episodes
were not targeted towards individuals or objects, and
the target could not be directly described. When the
target of harmful behaviour and sex were cross-tabu-
lated, a significant connection was found (v² = 13.940,
df = 4, P = 0.007). Secluded or restrained male
patients tended to target other persons or to have no
target more often than expected. Female patients tar-
geted themselves more often than expected (Table 3).
De-escalation
In total, 113 seclusion and restraint incidents were
included in the analysis of the de-escalation techniques
(Fig. 1; Table 4). De-escalation techniques used before
seclusion and restraint were reported in 101 cases
(89.4%). In 12 (10.6%) episodes, de-escalation tech-
niques had not been implemented. The 101 cases were
categorized into two main categories: measures to help
(mentioned 150 times) and restrictions (mentioned 30
times; Table 4). Several de-escalation techniques had
been implemented prior to the restrictive episode in
some cases. The most commonly-used de-escalation
technique was one-on-one discussion with the patient
(n = 74), which included listening to what the patient
TABLE 2: Sex of secluded or restrained patients in different patient
groups (n = 144)
Patient group
Group 1 Group 2 Group 3
n n n Total
Sex
Male 26 (24.06) 40 (49.58) 39 (31.35) 105
Female 7 (8.94) 28 (18.42) 4 (11.65) 39
Total 33 68 43 144
Group 1, forensic mental examination; group 2, difficult or dan-
gerous in local hospitals; group 3, not guilty by reason of insanity;
expected values are in parenthesis; v² = 14.299, df = 2, P = 0.001.
© 2017 Australian College of Mental Health Nurses Inc.
519
wanted to say and trying to calm him/her down. Medi-
cation measures were the second most common de-
escalation technique used, and included medicating the
patient according to need (n = 37) and administering
regular medication earlier than scheduled (n = 2).
Arrangements in the facilities were the third most com-
monly-mentioned de-escalation technique. Facility
arrangement measures included escorting a patient to
his/her room (n = 15), transferring a patient to a closed
ward (n = 1), arranging a single room with one bed to
guarantee privacy (n = 1), and escorting the patient
away from a certain space (n = 1). The patients were
de-escalated by providing support with intensive obser-
vation (n = 3) and one-to-one observation (n = 9). Acti-
vating a patient was mentioned in a few cases,
including enabling him/her to smoke (n = 4) and
arranging meaningful activities for him/her to do
(n = 1). Reducing patients’ demands and giving them
time to calm down were techniques that were men-
tioned only once.
Staff also reported using techniques that could be
considered restrictions, for example, approaches that
limited patients’ privileges or movements. Nevertheless,
the aim was to avoid the use of seclusion and restraint.
In 17 cases, the interaction between the patient and
staff clearly aimed to direct or restrict a patient. Other
restrictions implemented included physical restraint
(n = 4), seclusion (before mechanical restraint, n = 2),
use of clothes to restrict movement (n = 2), limitations
on the patient’s privileges (n = 2), and urine screening
tests for drugs (n = 3).
Reasons for seclusion and restraint
As presented in Table 5, the reasons for seclusion and
restraint were classified into four categories: direct
harmful behaviour, threatening harmful behaviour,
indirect harmful behaviour, and other behaviours in
eight episodes. The most common reason for seclusion
and restraint was threatening harmful behaviour
(35.4%). Examples of this category included verbal
threats, voices urging the patient to commit violent
behaviour, and the patient’s difficulty in not obeying,
non-verbal threatening behaviour, and tension. Threat-
ening harmful behaviour was targeted towards other
people, oneself, and objects. The second reason for
seclusion and restraint was direct harmful behaviour
(29.9%). This category included cases in which there
was imminent violence before the patient was secluded
or restrained. Violent acts were targeted towards other
people, oneself, and objects. Indirect harmful
520 S. KUIVALAINEN ET AL.

TABLE 3: Targets of harmful behaviour by sex (n = 144)

Target
No target Other person Oneself Both others and self Objects
n n n n n Total

Sex
Male 19 (16.04) 55 (48.85) 18 (25.52) 7 (7.29) 6 (7.29) 105
Female 3 (5.96) 12 (18.15) 17 (9.48) 3 (2.71) 4 (2.71) 39
Total 22 67 35 10 10 144

Expected values are in parenthesis; v² = 13.940, df = 4, P = 0.007.

TABLE 4: De-escalation techniques used to avoid seclusion and restless, could not engage in interactions because of
restraint during the 4-year study period in a forensic psychiatric hos- incoherence, was fearful, or, for example, was waiting
pital (n = 113 episodes of seclusion or restraint)
for the impending end of the world. In addition, this
Measures to help n category included cases in which voluntary room obser-
Interaction 74
vation was changed to seclusion because of the
Medication 39 patient’s confused state of mind and inability to under-
Arrangement of facilities 18 stand the voluntary nature of room observation. In all
Support 12 seven restraint episodes, the reason for restraint was
Activation 5 directly harmful behaviour.
Diminished demands 1
Provision of time to calm down 1
As mentioned, the most common de-escalation mea-
Restrictions sure was one-on-one discussion, and the most common
Verbal directions 17 reason for seclusion and restraint was threatening
Other restrictions 13 harmful behaviour. When the reason for seclusion and
restraint was indirect harmful behaviour, de-escalation
had not been implemented prior to each episode.
TABLE 5: Reasons for seclusion and restraint (n = 144)

Category n %
Threats of harmful behaviour 51
Direct harmful behaviour 43
Indirect harmful behaviour 42
Other behaviour 8 5.6
Total 144
behaviours were reported nearly as often (29.1%) as
direct harmful behaviour. The category of indirect
harmful behaviour included cases in which there was
no actual violence or threat of violence, but the
patient’s behaviour was indirectly harmful to other peo-
ple or to the patient himself/herself. For example, this
category included suspected or actual alcohol or drug
use, return after absconding, actions that scared other
patients, restlessness, seclusion for a night before elec-
troconvulsive therapy to ensure that the patient did not
eat or drink before sedation, seclusion for the reason of
polydipsia to ensure that the patient did not drink a
detrimental amount of water, and mean or loud beha-
viour. The fourth category consisted of ‘other beha-
viours’ (5.6%). This category included cases with no
mentioned threat of harmful behaviour, but in which
the patient was experiencing fulminant psychosis, was
DISCUSSION
35.4
29.9 The present study was conducted by retrospectively
29.1 investigating the de-escalation techniques used before
seclusion and restraint and the reasons for seclusion
100
and restraint. The study was implemented to improve
the care of forensic psychiatric patients, as well as to
support efforts to reduce use of restrictive methods
during care.
De-escalation
The staff in the study hospital attempted to help
patients gain control of their behaviour prior to 101 of
the 113 analysed seclusion and restraint episodes; nev-
ertheless, all of the episodes resulted in seclusion or
restraint. The most frequently-implemented de-escala-
tion techniques were verbal interactions, use of medi-
cation, and arrangement of facilities. Restriction was
also mentioned in several cases. Compared to the de-
escalation techniques outlined in a successful violence
risk assessment method, the BVC, the selection of
techniques in the present study was limited (Abder-
halden et al. 2008; Van de Sande 2011). For example,

© 2017 Australian College of Mental Health Nurses Inc.

SECLUSION, RESTRAINT AND DE-ESCALATION
in the present data, there were no reports of walking
outdoors with a nurse or using relaxation exercises,
even though these techniques are easy to implement.
The study hospital is located in a beautiful park area,
and the facilities are excellent for outdoor activities.
Even the closed wards offer the opportunity to spend
time outdoors in a garden area enclosed with fences.
In the present study, constant special observation
was used more often than intermittent special observa-
tion. This finding differs from those of Stewart and
Bowers (2012), who found in an acute mental health
ward that intermittent special observation was used five
times more often than constant special observation.
This difference might be explained by the differences
between the analysed episodes. In the present study,
we analysed the use of de-escalation before seclusion
and restraint, whereas Stewart and Bowers (2012)
included other conflicts and forms of containment as
well.
Cognitive de-escalation techniques, such as one-on-
one discussion and verbal direction, were the most
commonly used. Stimulus reduction was implemented
by escorting patients to their room, providing patients
an individual room, or secluding patients (before they
ended up in restraint). It is worth noting that in one
case, staff attempted to guide the patient away from
unwanted behaviour by offering meaningful activities.
Use of the senses to redirect patients was only men-
tioned in the context of stimulus reduction (Anderson
& Bushman 2002). After the study period (in Novem-
ber 2014), the study hospital developed a care plan
form, which included an individual crisis plan.
Reasons for seclusion and restraint
In the present study, the most common reason for
seclusion and restraint was threats to commit harmful
behaviour. The staff frequently described de-escalation
techniques to help patients in the context of these
seclusion and restraint episodes. The episodes were
preceded by direct harmful behaviour with seclusion or
restraint, and were also often preceded by de-escala-
tion techniques to help the patient gain control. How-
ever, when the reason for seclusion and restraint was
indirect harmful behaviour, de-escalation techniques
were used less frequently. This decreased use might
have been a consequence of the nature of the episodes
in this category, which included returns from abscond-
ing and suspected or actual alcohol or other drug use.
However, the nature of the episodes in this category
seemed to suggest that staff might have been able to
© 2017 Australian College of Mental Health Nurses Inc.
521
react in a way that prevented the use of seclusion or
restraint.
In the present study, two of three reasons for seclu-
sion and restraint were direct or threatening harmful
behaviour. This result was in line with those of a previ-
ous study conducted in the same hospital 10 years ear-
lier (Paavola & Tiihonen 2010). That study found that
the reason for seclusion in 69.3% of the incidents was
dangerous behaviour towards other persons or towards
oneself. Actual violence or threat of violence was
reported slightly less often as a reason for seclusion in
previous studies, including one conducted in 10 Euro-
pean countries (59%, Raboch et al. 2010), one from
Japan (57.7%, Noda et al. 2013), and one from Eng-
land (49%, Bowers et al. 2011). However, this result
differs from that of a prior study from Finland (Keski-
Valkama et al. 2009). Direct or threatening harmful
behaviour was more often a reason for seclusion and
restraint in the present study than in the aforemen-
tioned study. Differences between the patient groups
likely explain the dissimilarity between our findings and
those of the study by Keski-Valkama et al. (2009).
Specifically, their data included acute psychiatric wards
in addition to forensic psychiatry, whereas the present
study was implemented in a forensic psychiatric hospi-
tal only. Acute psychiatric wards in local hospitals also
include patients with alcohol delirium, and the inclu-
sion of this population might explain the greater
reporting of agitation/disorientation as a reason for
seclusion and restraint in those wards than in forensic
psychiatry. Another explanation might be the date of
the present study. The Niuvanniemi Hospital launched
the seclusion and restraint reduction programme in
2008, and the occurrence of non-essential restrictive
measures might have reduced over time since the
launch.
Overall, there seemed to be good reasons for the
use of seclusion and restraint in the present study, with
some exceptions. In Finland, patients can be secluded
if they are likely to cause significant damage to prop-
erty. Other less-restrictive techniques must be imple-
mented before seclusion, and the use of further
techniques instead of single-seclusion episodes might
have been possible in some cases. Some of the epi-
sodes in which a patient had damaged property (e.g.
breaking a vase) appeared minor compared to the cri-
teria for significant damage of property specified in the
Mental Health Act of Finland. It is possible that seclu-
sion was not the last resort in these cases.
In the present study, seclusion and restraint were
used for fulminant psychosis in seven cases.
522
Continuous support might have been more suitable to
these patients. Other alternative measures, for example,
sensory modulation, might have been possible, espe-
cially if the patient’s condition prevented them from
calming down with cognitive techniques.
Limitations
Although the study population was nationally represen-
tative of individuals who are treated or examined in
forensic psychiatric hospitals in Finland, there are
some limitations that hinder the generalizability of the
results. Only the first seclusion or restraint episode for
each patient was included in the study material. This
purposive sampling limited the generalizability of the
numeric results, although it ensured the qualitative
diversity of the results. In future, problems with gener-
alizability can be prevented by conducting observa-
tional studies of de-escalation in clinical practice that
include all situations involving de-escalation, not only
cases that resulted in seclusion or restraint. Addition-
ally, stratifying the seclusion and restraint data at the
ward level, and randomly sampling the cases at the
ward level, would enhance the generalizability. In the
present study, we did not investigate situations in
which de-escalation techniques succeeded in avoiding
restrictions. Therefore, it is impossible to determine
whether the techniques used are more or less effective
than other techniques. Furthermore, there are certainly
other aspects affecting the path to seclusion and
restraint that were not accounted for in the present
study. Register studies are secondary analyses of data
that are not originally collected for the purposes of the
study. In these studies, it is assumed that there is a
phenomenon that can be observed via register data.
Furthermore, in a register study, it is impossible to
observe all details influencing the phenomenon, and
these unassessed variables might, in turn, bias the
results (Sund 2008). Although patient files are well reg-
ulated in Finland, in the present study, the available
data depended on how detailed the staff were in their
notes. In addition to examining seclusion and restraint
forms, the medical charts on the date of the seclusion
or restraint episode were also studied. This procedure
was followed to avoid incorrect conclusions due to the
limitations of the data. However, the medical charts
still might have been incomplete, as shown by Van
Walraven and Demers (2001). Another limitation of
the data was the sample size. The sample size was cal-
culated when planning the study, but the number of
cases was reduced during the analysis process, as some
S. KUIVALAINEN ET AL.
seclusion and restraint episodes were excluded. Despite
the original plan, relevant seclusion and restraint epi-
sodes were excluded during the study to avoid bias in
the results caused by incompatibility of choices made
by the researcher (Sund 2008). This methodology
reduced the sample size, thereby limiting the generaliz-
ability of the results.
CONCLUSION
The same de-escalation techniques were used with
most of the patients. The present study identified the
use of traditional techniques, such as verbal interac-
tions, medication changes, and arrangement of facili-
ties. Other means of supporting self-regulation among
patients in challenging situations were missing from the
data; however, self-regulation is an essential part of
patient education. Staff should be educated on a broad
range of de-escalation techniques. The staff in the pre-
sent study followed the legal criteria for seclusion and
restraint, which involved mostly violent behaviour or
threats of violence. More studies on de-escalation tech-
niques that successfully prevent restrictions are
needed, especially those that assess a larger sample size
and include both patient and staff perspectives.
RELEVANCE FOR CLINICAL PRACTICE
The present study adds to the few studies addressing
the reasons for seclusion and restraint in forensic psy-
chiatry, as well as the de-escalation techniques applied
in challenging situations, in which the use of restriction
could not be avoided. The present study highlights the
need for staff to be educated on a diverse range of de-
escalation techniques.
ACKNOWLEDGEMENT
The project was funded by the Finnish Ministry of
Social Affairs and Health through the development
fund for Niuvanniemi Hospital, Kuopio, Finland.
REFERENCES
Abderhalden, C., Needham, I., Dassen, T., Halfens, R.,
Haug, H.-J. & Fisher, J. E. (2008). Structured risk
assessment and violence in acute psychiatric wards:
Randomized controlled trial. The British Journal of
Psychiatry, 193, 44–50.
Almvik, R., Woods, P. & Rasmussen, K. (2000). The Brøset
Violence Checklist; sensitivity, specificity and interrater
reliability. Journal of Interpersonal Violence, 15, 1284–1296.
© 2017 Australian College of Mental Health Nurses Inc.
SECLUSION, RESTRAINT AND DE-ESCALATION
Anderson, G. A. & Bushman, B. J. (2002). Human
aggression. Annual Review of Psychology, 53, 27–51.
Bowers, L., Van Der Merwe, M., Nijman, H. et al. (2010).
The practice of seclusion and time-out on English acute
psychiatric wards: The City-128 Study. Archieves of
Psychiatric Nursing, 24, 275–286.
Bowers, L., Ross, J., Nijman, H., Muir-Cochrane, E.,
Noorthoorn, E. & Stewart, D. (2011). The scope for
replacing seclusion with time out in acute inpatient psychiatry
in England. Journal of Advanced Nursing, 68, 826–835.
Burns, N. & Grove, S. K. (2009). The Practice of Nursing
Research: Appraisal, Synthesis, and Generation of
Evidence, 6th edn. St Louis, MO: Saunders Elsevier.
Chalmers, A., Harrison, S., Mollison, K., Molloy, N. & Gray,
K. (2012). Establishing sensory-based approaches in
mental health inpatient care: A multidisciplinary approach.
Australasian Psychiatry, 20, 35–39.
Champagne, T. & Stromberg, N. (2004). Sensory approaches
in inpatient psychiatric settings. Innovative alternatives to
seclusion & restraint. Journal of Psychosocial Nursing, 42,
1–8.
Crenshaw, W. B. & Francis, P. S. (1995). A national survey
on seclusion and restraint in state psychiatric hospitals.
Psychiatric Services, 46, 1026–1031.
De Benedictis, L., Dumais, A., Sieu, N. et al. (2011). Staff
perceptions and organizational factors as predictors of
seclusion and restraint on psychiatric wards. Psychiatric
Services, 62, 484–491.
Elo, S. & Kyng€as, H. (2008). The qualitative content analysis
process. Journal of Advanced Nursing, 62, 107–115.
Hallgren, K. A. (2012). Computing inter-rater reliability for
observational data: An overview and tutorial. Tutorials in
Quantitative Methods for Psychology, 8, 23–34.
Happell, B. & Koehn, S. (2010). Attitudes to the use of
seclusion: Has contemporary mental health policy made a
difference? Journal of Clinical Nursing, 19, 3208–3217.
Huckshorn, K. (2006). Six core strategies for reducing
seclusion and restraint use. National Association of State
Mental Health Program Directors Research Institute.
Alexandria: NASMHPD Publications. [Cited 19 July
2017]. Available from: URL: www.nasmhpd.org/content/
six-core-strategies-reduce-seclusion-and-restraint-use
Janssen, W., Noorthoorn, E., Linge, R. & Lendemeijer, B.
(2007). The influence of staffing levels on the use of
seclusion. International Journal of Law and Psychiatry,
30, 118–126.
Janssen, W., Noorthoorn, E., Nijman, H. et al. (2013).
Differences in seclusion rates between admission wards:
Does patient compilation explain? Psychiatric Quarterly,
84, 39–52.
Kaltiala-Heino, R., Tuohim€aki, C., Korkeila, J. & Lehtinen, V.
(2003). Reasons for using seclusion and restraint in
psychiatric inpatient care. International Journal of Law
and Psychiatry, 26, 139–149.
Keski-Valkama, A., Sailas, E., Eronen, M., Koivisto, A.-M.,
L€onnqvist, J. & Kaltiala-Heino, R. (2009). The reasons for
using restraint and seclusion in psychiatric inpatient care:
© 2017 Australian College of Mental Health Nurses Inc.
523
A nationwide 15-year study. Nordic Journal of Psychiatry,
64, 136–144.
Kontio, R., V€alim€aki, M., Putkonen, H., Kuosmanen, L.,
Scott, A. & Joffe, G. (2010). Patient restrictions: Are there
ethical alternatives to seclusion and restraint? Nursing
Ethics, 17, 65–76.
Kontio, R., Joffe, G., Putkonen, H. et al. (2012). Seclusion
and restraint in psychiatry: Patients’ experiences and
practical suggestions on how to improve practices and
use alternatives. Perspectives in Psychiatric Care, 48, 16–
24.
Kuivalainen, S., Vehvil€ainen-Julkunen, K., Putkonen, A.,
Louheranta, O. & Tiihonen, J. (2014). Violent behavior in
a forensic psychiatric hospital in Finland: An analysis of
violence incident reports. Journal of Psychiatric and
Mental Health Nursing, 21, 214–218.
Kumble, S. & McSherry, B. (2010). Seclusion and restraint:
Rethinking regulation from a human rights perspective.
Psychiatry, Psychology and Law, 17, 551–561.
Landis, J. & Koch, G. (1977). The measurement of observer
agreement for categorical data. Biometrics, 33, 159–174.
Larue, C., Dumais, A., Drapeau, A., Menard, G. & Goulet,
M.-H. (2010). Nursing practices recorded in reports of
episodes of seclusion. Issues in Mental Health Nursing,
31, 785–792.
LeBel, J. & Goldstein, R. (2005). The economic cost of using
restraint and the value added by restraint reduction or
elimination. Psychiatric Services, 56, 1109–1114.
Makkonen, P., Putkonen, A., Korhonen, J., Kuosmanen, L. &
K€arkk€ainen, J. (eds.) (2016). Pakon k€ayt€on v€ahent€aminen
ja turvallisuuden lis€a€aminen psykiatrisessa hoidossa.
Discussionpaper 35/2016. Helsinki: National Institute for
Health and Welfare (THL). (In Finnish, abstract in
English) . [Cited 19 July 2017]. Available from: URL: www.
julkari.fi/bitstream/handle/10024/131697/URN_ISBN_978-
952-302-751-0.pdf?sequence=1
McCann, T. V., Baird, J. & Muir-Cochrane, E. (2015). Social
climate of acute aged psychiatry inpatient units: Staff
perceptions within the context of patient aggression and
violence. Journal of Psychiatric and Mental Health
Nursing, 22, 102–108.
Mental Health Act 1116/1990. Ministry of Social Affairs and
Health, Finland. [Cited 19 July 2017]. Available from:
URL: www.finlex.fi/fi/laki/kaannokset/1990/en19901116.pdf
MIELI Plan. (2009). National Institute of Health and Welfare,
Finland. Updated 18 May 2014. [Cited 19 July 2017].
Available from: URL: www.thl.fi/en/web/thlfi-en/research-
and-expertwork/projects-and-programmes/mieli-plan
National Institute for Health and Care Excellence (2015).
Violence and aggression: Short-term management in
mental health, health and community settings. NICE
guideline. NG10. . [Cited 19 July 2017]. Available from:
URL: www.nice.org.uk/guidance/ng10
Noda, T., Sugiyama, N., Sato, M. et al. (2013). Influence of
patient characteristics on duration of seclusion/restrain in
acute psychiatric settings in Japan. Psychiatry and Clinical
Neurosciences, 67, 405–411.
524
Noorthoorn, E., Lepping, P., Janssen, W. et al. (2015). One-
year incidence and prevalence of seclusion: Dutch
findings in an international perspective. Social Psychiatry
and Psychiatric Epidemiology, 50, 1857–1869.
Noorthoorn, E., Voskes, Y., Janssen, W. et al. (2016).
Seclusion reduction in Dutch mental health care: Did
hospitals meet goals? Psychiatric Services, 67, 1321–1327.
Novak, T., Scanlan, J., McCaul, D., MacDonald, N. & Clarke, T.
(2013). Pilot study of a sensory room in an acute inpatient
psychiatric unit. Australasian Psychiatry, 20, 401–406.
Paavola, P. & Tiihonen, J. (2010). Seasonal variation of
seclusion incidents from violent and suicidal acts in
forensic psychiatric patients. International Journal of Law
and Psychiatry, 33, 27–34.
Polit, D. F. & Beck, C. T. (2008). Nursing Research.
Generating and Assessing Evidence for Nursing Practice,
8th edn. Philadelphia, PA: Wolters Kluwer Health/
Lippincott Williams & Wilkins.
Putkonen, A., Kuivalainen, S., Louheranta, O. et al. (2013).
Cluster-randomized controlled trial of reducing seclusion
and restraint in secured care of men with schizophrenia.
Psychiatric Services, 64, 850–855.
Raboch, J., Kalisova, L., Nawka, A. et al. (2010). Use of coercive
measures during involuntary hospitalization: Findings from
ten European countries. Psychiatric Services, 61, 1012–1017.
Rainio, J. & R€aty, T. (2015). Psychiatric specialist medical
care 2013. Official Statistics of Finland. Health 2015.
Statistical Report 2/2015. Helsinki: National Institute of
Health and Welfare. [Cited 19 July 2017]. Available from:
URL: http://urn.fi/URN:NBN:fi-fe201503191963
Restraint Reduction Network. (2017). Our mission, vision,
values and principles. [Cited 17 July 2017]. Available
from: URL: http://restraintreductionnetwork.org/our-
mission-vision-value-and-principles/
Richter, D. (2006). Nonphysical conflict management and de-
escalation. In: D. Richter & R. Whittington (Eds).
Violence in Mental Health Settings: Causes, Consequences,
Management (pp. 125–144). New York, NY: Springer.
Roberton, T., Daffern, M., Thomas, S. & Martin, T. (2012).
De-escalation and limit setting in forensic mental health
units. Journal of Forensic Nursing, 8, 94–101.
Smith, G., Ashbridge, D., Davis, R. & Steinmetz, W. (2015).
Correlation between reduction of seclusion and restraint
and assaults by patients in Pennsylvania’s state hospitals.
Psychiatric Services, 66, 303–309.
Steinert, T. & Lepping, P. (2009). Legal provisions and
practice in the management of violent patients. A case
vignette study in 16 European countries. European
Psychiatry, 24, 135–141.
Steinert, T., Lepping, P., Bernhardsgr€ utter, R. et al. (2010).
Incidence of seclusion and restraint in psychiatric
S. KUIVALAINEN ET AL.
hospitals: A literature review and survey of international
trends. Social Psychiatry and Psychiatric Epidemiology,
45, 889–897.
Stewart, D. & Bowers, L. (2012). Under the gaze of staff:
Special observation as surveillance. Perspectives in
Psychiatric Care, 48, 2–9.
Stewart, D., Bowers, L. & Ross, J. (2011). Managing risk and
conflict behaviours in acute psychiatry: The dual role of
constant special observation. Journal of Advanced Nursing,
68, 1340–1348.
Substance Abuse and Mental Health Services Administration
(2011). The Business Case for Preventing and Reducing
Restraint and Seclusion Use. HHS Publication No. (SMA)
11-4632. Rockville, MD: Substance Abuse and Mental
Health Services Administration.
Sund, R. (2008). Methodological Perspectives for Register-
Based health System Performance Assessment. Developing
a Hip Fracture Monitoring System in Finland. National
Research and Development Centre for Welfare and
Health, Research Report 174. Gummerus Printing,
Vaajakoski, Finland.
Sutton, D., Wilson, M., Van Kessel, K. & Vanderpyl, J.
(2013). Optimizing arousal to manage aggression: A pilot
study of sensory modulation. International Journal of
mental Health Nursing, 22, 500–511.
Te Pou o te Whakaaro (2011). Sensory Modulation in Mental
Health Clinical Settings: A Review of the Literature. Te
Pou o te Whakaaro Nui: Auckland.
Tuppurainen, H., Kivist€ o, P., Repo-Tiihonen, E. & Tiihonen,
J. (2014). Implementation of the Finnish Current Care
guideline of schizophrenia in Niuvanniemi hospital during
2011. (In Finnish, abstract in English). Suomen
l€a€ak€arilehti, 69, 2549–2553.
Van de Sande, R., Nijman, H. L. I., Noorthoorn, E. O. et al.
(2011). Aggression and seclusion on acute psychiatric
wards: Effect of short-term risk assessment. The British
Journal of Psychiatry, 199, 473–478.
Van der Schaaf, P., Dusseldorp, E., Keuning, F., Janssen, W.
& Noorthoorn, E. (2013). Impact of the physical
environment of psychiatric wards on the use of seclusion.
British Journal of Psychiatry, 202, 142–149.
Van Walraven, C. & Demers, S. (2001). Coding diagnoses
and procedures using a high-quality clinical database
instead of a medical record review. Journal of Evaluation
of Clinical Practice, 7, 289–297.
Wieman, D. A., Camacho-Gonsalves, T., Huckshorn, K. A. &
Leff, S. (2014). Multisite study of an evidence-based
practice to reduce seclusion and restraint in psychiatric
inpatient facilities. Psychiatric Services, 65, 345–351.
© 2017 Australian College of Mental Health Nurses Inc.