bs_bs_banner

International Journal of Mental Health Nursing (2017) 26, 491–499 doi: 10.1111/inm.12383

S PECIAL I SSUE
Prolonged use of seclusion and mechanical
restraint in mental health services: A statewide
retrospective cohort study
Brian McKenna,1,2 Samantha McEvedy,3 Tessa Maguire,2,4 Jo Ryan2,4 and Trentham Furness5
1
School of Clinical Sciences, Auckland University of Technology, and Auckland Regional Forensic Psychiatry
Services,Waitemata District Health Board, Auckland, New Zealand, 2Centre for Forensic Behavioural Science,
Swinburne University of Technology, Hawthorn, 3School of Psychology and Public Health, La Trobe University,
4
Nursing Practice Development Unit, Forensicare, and 5School of Nursing, Midwifery and Para medicine, Australian
Catholic University and North Western Mental Health, Melbourne Health, Melbourne, Victoria, Australia

ABSTRACT: Seclusion and mechanical restraint are restrictive interventions that should be used
only as a last resort and for the shortest possible time, yet little is known about duration of use in the
broader context. Adult area mental health services throughout Victoria, Australia, were asked to
complete a report form for prolonged episodes of seclusion (>8 hours) and mechanical restraint
(>1 hour). The present, retrospective cohort study aimed to understand the individual (age, sex, type
of service, duration of intervention) and contextual factors associated with prolonged use of
restrictive interventions. Contextual factors describing the reasons for prolonged use of the restrictive
interventions were captured qualitatively, and then coded using content analysis. Median duration
was compared across individual factors using Mann–Whitney U-tests. During 2014, 690 episodes of
prolonged restrictive intervention involving 311 consumers were reported. Close to half (n = 320,
46%) involved mechanical restraint. Seclusion episodes (n = 370) were longer in forensic mental
health services compared to adult area mental health services (median: 24 hours and 18 min vs
16 hours and 42 min, P < 0.001). Mechanical restraint episodes (n = 320) were shorter in forensic
mental health services compared to adult area mental health services (median: 3 hours and 25 min vs
4 hours and 15 min, P = 0.008). Some consumers were subject to multiple episodes of prolonged
seclusion (55/206, 27%) and/or prolonged mechanical restraint (31/131, 24%). The most commonly
occurring contextual factor for prolonged restrictive interventions was ‘risk of harm to others’. Means
for reducing the use of prolonged restrictive interventions are discussed in light of the findings.
KEY WORDS: benchmark, mechanical restraint, restrictive intervention, seclusion.

negative emotions (Ashcraft & Anthony 2008; Bigwood

INTRODUCTION
& Crowe 2008; Larue et al. 2013; Steinert et al. 2013;
Restrictive interventions are coercive, and cause con- Stubbs et al. 2009). Restrictive interventions (seclusion,
sumers, carers, and clinicians to feel a range of physical restraint, and mechanical restraint) might be
authorized for consumers if they are considered to be
Correspondence: Brian McKenna, School of Clinical Sciences, an imminent and serious risk to themselves or another
Auckland Institute of Technology, Private Bag, Northcote, Auck- person (Wieman et al. 2014). However, the use of
land 0627, New Zealand Email: brian.mckenna@aut.ac.nz
Brian McKenna, RN, PhD. restrictive interventions can have an impact on the
Samantha McEvedy, LLB. therapeutic relationship between mental health con-
Tessa Maguire, RN, M Ment Health Sc.
Jo Ryan, RN, BEd, PGCert .
sumers and clinicians, cause serious physical harm, be
Trentham Furness, PhD. experienced as a traumatic event, and/or recapitulate
Accepted June 13 2017.

© 2017 Australian College of Mental Health Nurses Inc.

492 B. MCKENNA ET AL.

previous experiences of trauma (El-Badri & Mellsop Although the Institute has published for the first time
2008). A comprehensive review of the literature advo- this year the rates of physical and mechanical restraint,
cates that restrictive interventions should be used only duration is not reported (Australian Institute of Health
as a last resort and for the shortest possible period of and Welfare 2017).
time (McKenna et al. 2014). The aim of the present study was to understand the
The reduction, and in some cases, elimination, of individual (age, sex, type of service, duration of inter-
restrictive interventions has been a constant focus for vention) and contextual factors associated with pro-
mental health services (Bergk et al. 2011). In Australia, longed use of seclusion and mechanical restraint in the
the Commonwealth Government’s National Mental Australian State of Victoria.
Health Working Group (2005) identified reducing the
use of restrictive interventions as one of four safety pri-
METHODS
orities for mental health services. The National Mental
Health Seclusion and Restraint Project (2007–2009),
Research design
known as the ‘Beacon Project’, was developed to estab-
lish centres of excellence aimed towards reducing A retrospective cohort study design was used to analyse
seclusion and restraint in public mental health facili- data collected by the Office of the Chief Psychiatrist,
ties, including forensic mental health services (Maguire Department of Health and Human Services (DHHS) in
et al. 2012). The Beacon Project used the ‘six core Victoria for the 2014 calendar year. A retrospective
strategies’ (Huckshorn 2004) to develop a systemic cohort design is used when data have already been col-
response to drive change. The National Mental Health lected to allow the incidences or determination of a con-
Commission has also initiated a project to determine dition to be described (Mann 2003). The present study
best practice in reducing or eliminating restrictive received ethics approval from the Australian Catholic
interventions (Australian Institute of Health and Wel- University Human Research Ethics Committee (Mel-
fare 2013). bourne, Victoria, Australia; ref no. 2015-101E).
Evidence from North America suggests that efforts
to reduce the frequency of use of restrictive interven-
DHHS data collection
tions have been successful, but that prolonged duration
of seclusion remains an area that requires improvement To assist the goal to reduce the use of restrictive inter-
(Staggs 2015). Yet little is known about the duration of ventions (Department of Health 2013), the DHHS
restrictive interventions in the broader context, or the publishes the ‘frequency’ of seclusion episodes for all
factors that contribute to the prolonged use. Although public mental health services on a quarterly basis
studies have focussed on the duration of restrictive (http://performance.health.vic.gov.au/Home/Report.aspx).
interventions (Happell & Gaskin 2011; Janssen et al. However, there is no such reporting of the ‘duration’
2013; Noda et al. 2013; Tyner et al. 2012), most were of restrictive interventions. To overcome this deficit,
undertaken in specific hospital settings, with wide vari- the DHHS initiated a pilot programme to collect data
ations attributed to localized contextual factors, which on episodes of so-called ‘variance’ (i.e. prolonged use
make comparisons difficult (Janssen et al. 2011). Stud- of seclusion and mechanical restraint). Mental health
ies of the duration of restrictive interventions for entire services were asked to complete a short form captur-
jurisdictions are scarce (Steinert et al. 2010). A study ing data for every episode of seclusion that exceeded
of the frequency and duration of restrictive interven- eight continuous hours, or mechanical restraint that
tions, which mainly focussed on seclusion, reported a exceeded one continuous hour. Individual factors col-
maximum duration of 96 hours. A small number of lected for each episode included the consumer’s date
consumers, males, and high-dependency units account of birth and sex, the type of service, type of restrictive
for prolonged duration (Oster et al. 2016). The present intervention, and total duration of the episode of
state-wide study was confined to the Australian State of restrictive intervention. Contextual factors describing
South Australia. The Australian Institute of Health and the reasons for continuation of the restrictive inter-
Welfare (2017) reported national figures on the dura- vention were captured qualitatively as a written narra-
tion of seclusion. The average duration of seclusion for tive. The form was completed by a psychiatrist or
2015–2016 was reported as 5.3 hours, although this nurse and authorized by a senior clinician before
excludes forensic mental health services, where the being sent to the Office of the Chief Psychiatrist at
average duration of seclusion events is 87.9 hours. the DHHS.

© 2017 Australian College of Mental Health Nurses Inc.

PROLONGED DURATION OF RESTRICTIVE INTERVENTIONS 493

Data were analysed with the Statistical Package for

Setting
Social Scientists (SPSS version 21.0 for Windows, IBM,
Data were collected across Victoria’s 21 state-funded Armonk, NY, USA). Significance was accepted at
adult area mental health services, which provide a range P ≤ 0.05. Data regarding the duration of seclusion and
of specialist mental health services for people aged from mechanical restraint did not meet the assumptions of
16 to 64 years who have a serious mental illness (pri- normality, as Kolmogorov–Smirnov and Shapiro–Wilk
marily psychotic illnesses, bipolar disorder, and major tests were significant (P < 0.001). Therefore, the med-
depression). Each of the 21 area mental health services ian duration and interquartile ranges (IQR) are
have acute mental health inpatient services attached to reported. Mann–Whitney U-tests were used to com-
a general hospital for the management of people with pare the median duration of seclusion or mechanical
significant psychosocial disturbance and disability, until restraint by individual factor subgroups. Other individ-
they have recovered sufficiently to be managed in the ual factors (age, sex, type of service) were categorized
community (http://www.health.vic.gov.au/mentalhea and reported as numbers and percentages. Individual
lthservices/adult/index.htm). Data were also collected and contextual factors were compared between forensic
from the state’s forensic mental health secure hospital, and adult area mental health services using Pearson’s
which is a 116-bed facility operating within a recovery v2-tests.
framework, providing acute and continuing care for peo-
ple primarily accessing the service via the criminal justice
RESULTS
system (www.forensicare.vic.gov.au/our-services/thomas-
embling).
Sample description
A total of 690 reports of prolonged use of restrictive
Data analyses
intervention, relating to 311 mental health consumers,
As data were captured on hard-copy forms, they were were completed between 1 January and 31 December
first converted into electronic versions through data 2014. Of these reports, 370 (54%) were episodes of
entry of numeric (e.g. age) or categorical (e.g. sex) vari- prolonged seclusion (defined as >8 hours) and 320
ables, and via transcription of qualitative variables (i.e. (46%) were episodes of prolonged mechanical restraint
written narratives describing contextual factors). Con- (defined as greater than 1 hour).
tent analysis was used to analyse written narratives Of the 370 episodes of prolonged seclusion, 40%
describing the contextual factors surrounding each epi- (n = 146) occurred in the state’s forensic mental health
sode reported, as this approach allows coding of the hospital, and the other 60% (n = 224) in the 21 general
data and interpretation of quantitative counts of the adult area mental health services throughout the state
codes (Vaismoradi et al. 2013). Ten percent of the jurisdiction. Prolonged episodes of seclusion involved
forms were randomly selected and analysed by one 206 consumers (aged 16–64 years, mean 34  9 years,
researcher (SM) to develop a code frame. To enhance male, n = 166, 81%). Of these 206 consumers, 26 were
rigour, the code frame was checked by other members also subject to prolonged use of mechanical restraint.
of the research team, and any areas of disagreement Some consumers experienced multiple episodes of pro-
were resolved through discussion and adaptation of the longed seclusion (2–4 episodes: 21% (n = 43), ≥5 epi-
code as required (Mays & Pope 1995). sodes: 6% (n = 12), maximum episodes for one
Once the code frame had been agreed, all variance consumer over the year: n = 22).
report forms were read and systematically coded by Of the 320 episodes of prolonged mechanical
two researchers (SM and TM). Multiple codes were restraint, 16% (n = 52) occurred in the forensic mental
often required to capture the various elements health hospital, and 84% (n = 268) occurred in general
describing the contextual factors surrounding the pro- adult area mental health services. Prolonged episodes
longed restrictive intervention. Ten percent of the of mechanical restraint involved 131 consumers (aged
reporting forms were selected for quality control by 16–63 years, mean: 35  11 years, male, n = 87, 66%).
another member of the research team (BM), who Some consumers experienced multiple episodes of pro-
checked that coding was accurate and consistent. longed mechanical restraint (2–4 episodes: 21%
Coded responses were reported as numbers and per- (n = 28), ≥5 episodes: 2% (n = 3)). The maximum epi-
centages, with each code becoming a dichotomous sodes for one consumer over the year was 125 for a
‘yes/no’ response. person located in an acute inpatient mental health unit

© 2017 Australian College of Mental Health Nurses Inc.

494 B. MCKENNA ET AL.

in an adult area mental health service, who expressed a TABLE 2: Median duration of seclusion by individual factors
preference during his care planning for the use of Median duration
mechanical restraint over seclusion when experiencing of seclusion Mann–Whitney
symptoms which cause him to be verbally and physi- hours:min (IQR) U-test P-value
cally threatening or aggressive to staff or other Sex
consumers. Male 21:05 (12:50–43:40) 8540 0.002
The most commonly occurring contextual factors for Female 14:00 (11:57–23:20)
an episode of prolonged seclusion or mechanical Age (years)
restraint were ‘risk of harm to others’ and ‘symptom 18–34 18:47 (12:13–41:55) 14 499 0.953
>35 20:15 (12:14–38:05)
management’ (Table 1). Compared to seclusion,
Service type
mechanical restraint was more often planned with the Forensic 24:18 (15:25–52:15) 9285 <0.001
consumer and more likely to have involved a medical General 16:42 (11:15–25:30)
transfer offsite or to another unit, or the person Single or multiple episodes of seclusion
assessed as an absconding risk (Table 1). Single 19:00 (11:50–30:05) 15 878 0.215
Multiple 19:31 (12:45–43:00)

Medians compared using Mann–Whitney U-test. IQR, interquartile

Duration of prolonged seclusion
range.
The median duration of the 370 episodes of prolonged
seclusion was 19 hours and 30 min (IQR = 12 hours more frequent mention of symptom management in
and 14 min to 1 day, 15 hours, and 30 min). The forensic settings, and more frequent mention of police
majority of episodes (215/349 (missing values = 21), involvement, intoxication, and risk of absconding in
62%) were <24 hours, but there were 72 episodes general adult mental health settings (Table 3).
(21%) of between 24 and <48 hours, and 62 episodes
(18%) extended beyond 48 hours. The maximum dura-
Duration of prolonged mechanical restraint
tion of a prolonged episode of seclusion was just over
16 days (16 days, 1 hour, and 15 min), and related to a The median duration of the 320 episodes of prolonged
26-year-old male in the forensic mental health hospital. mechanical restraint was 4 hours (IQR = 2 hours and
The median duration of seclusion episodes was signifi- 15 min to 10 hours and 13 min). There were 202
cantly longer for males and in the forensic mental (64%) of 317 episodes of mechanical restraint (missing
health setting (Table 2). data = 3) that were between 1 hour and <6 hours’
Comparisons of contextual factors surrounding pro- duration, 49 (15%) between 6 hours and <12 hours,
longed episodes of seclusion between forensic mental and 66 (21%) that were 12 hours or more. The maxi-
health and adult area mental health settings indicated mum duration of an episode of mechanical restraint

TABLE 1: Contextual factors for prolonged duration by restrictive

intervention type TABLE 3: Contextual factors for continuation of seclusion by
service type
Seclusion Mechanical restraint
n (%) n (%) Forensic General
Contextual factors n (%) n (%) v2 P-value
Total n (%) 370 (54) 320 (46)
Risk of harm to others 355 (96) 282 (88) Total n (%) 146 (39.5) 224 (60.5)
Symptom management 196 (53) 171 (53) Risk of harm to others 142 (97) 213 (95) 1.071 0.301
Police/security involvement 56 (15) 36 (10) Symptom management 91 (62) 105 (47) 8.474 0.004
Intoxication/withdrawal 36 (10) 19 (6) Transfer to another unit 14 (10) 10 (5) 3.371 0.066
Absconding risk 34 (9) 65 (20) ECT related 13 (9) 11 (5) 2.324 0.127
Transferred to another unit 24 (6) 47 (13) Intoxication/withdrawal 4 (3) 32 (14) 14.385 <0.001
ECT related 24 (6) 43 (13) Absconding risk 2 (1) 32 (14) 18.724 <0.001
Arrived with RI in situ 23 (6) 22 (6) Arrived with RI in situ 3 (2) 20 (9) 7.714 0.005
Planned with consumer 3 (<1) 125 (39) Police/security involved 1 (1) 55 (25) 41.347 <0.001
Medical transfer off campus 1 (<1) 56 (15)
Multiple responses, therefore adds to >100%. Comparisons made
Multiple response, therefore adds to >100%. ECT, electroconvulsive using v2-tests. ECT, electroconvulsive therapy; RI, restrictive
therapy; RI, restrictive intervention. intervention.

© 2017 Australian College of Mental Health Nurses Inc.

PROLONGED DURATION OF RESTRICTIVE INTERVENTIONS 495

was close to 8 days (7 days and 20 hours), and related TABLE 5: Contextual factors for mechanical restraint by service
to a 39-year-old male in a general mental health set- type
ting. The median duration of mechanical restraint was Forensic General
significantly longer for males, those aged ≥35 years, Contextual factors n (%) n (%) v2 P-value
episodes in adult area mental health services, and for Total n (%) 52 (16) 268 (84)
those who were subject to multiple episodes of pro- Medical treatment offsite 46 (89) 10 (4) 213.64 <0.001
longed mechanical restraint (Table 4). Risk of harm to others 33 (64) 249 (93) 36.091 <0.001
Comparisons of contextual factors for the prolonged Absconding risk 33 (64) 32 (12) 70.085 <0.001
use of mechanical restraint between forensic mental Transfer to another unit 21 (40) 26 (10) 31.994 <0.001
ECT related 18 (35) 25 (9) 23.941 <0.001
health and adult area mental health settings revealed a
Symptom management 11 (21) 160 (60) 26.008 <0.001
number of differences. Perceived risk of harm to others Arrived with RI in situ 1 (2) 21 (8) 2.44 0.144
was mentioned more frequently in general services, Police/security involved 0 (0) 36 (13) 8.003 0.001
and the use of mechanical restraint was often planned Planned with consumer 0 (0) 125 (47) 40.735 <0.001
with the consumer. In the forensic mental health ser-
Multiple responses, therefore adds to >100%. Comparisons made
vice, the prolonged use of mechanical restraint was using v2-tests. ECT, electroconvulsive therapy; RI, restrictive
more frequently described as a medico-legal and proce- intervention.
dural requirement due to absconding risk when con-
sumers were transported offsite. This occurs when
forensic consumers (primarily remand or sentenced subject to multiple episodes of prolonged seclusion
prisoners) go to other hospitals to receive medical (55/206, 27%) and/or prolonged mechanical restraint
treatment, including electroconvulsive therapy (31/131, 24%) during the 12-month period.
(Table 5). In the year of the collection of these data on pro-
longed mechanical restraint, 1315 episodes of mechani-
cal restraint of any duration were reported in Victorian
DISCUSSION
mental health services (Office of the Chief Psychiatrist
The key findings of the present study relate to the 2016), compared to 14 episodes in South Australia in
extensive use of prolonged mechanical restraint in 2010 and 2011. This degree of use is concerning.
mental health services across the jurisdiction, marked Mechanical restraint should only be used in exceptional
differences in the duration of both seclusion and circumstances, due to the serious adverse effects asso-
mechanical restraint between forensic and adult area ciated with its use (Hui et al. 2013). Evidence suggests
mental health services, and some consumers being that the prevalence and duration of mechanical
restraint is particularly high among forensic mental
health inpatients (Gildberg et al. 2015). Although this
TABLE 4: Median duration of mechanical restraint by individual was supported by the findings of the present study, the
factors prevalence of the prolonged duration of mechanical
Median duration restraint was more evident in adult area mental health
of mechanical restraint Mann–Whitney services. This finding was largely influenced by one
hours:min (IQR) U-test P-value particular consumer in an adult area mental health ser-
Sex vice who was subject to 125 episodes of prolonged use
Male 4:12 (2:30–11:32) 5452.5 0.002 of mechanical restraint over the data-collection year.
Female 3:00 (1:40–6:00) Although this intervention was part of his advanced
Age (years) care planning, and his preference in situations when he
18–34 3:24 (1:46–5:28) 13 401 <0.001
>35 4:20 (2:30–13:07)
was experiencing exacerbated symptoms of mental ill-
Service type ness, there is an ethical requirement to ensure all other
Forensic 3:25 (1:52–5:00) 8365 0.008 less-restrictive interventions are continually revisited
General 4:15 (2:30–11:15) and renegotiated (Parish 2013).
Single or multiple episodes of mechanical restraint The planned use of mechanical restraints in forensic
Single 3:35 (1:44–7:40) 12 405 0.026
mental health settings is associated with behaviours
Multiple 4:05 (2:35–11:15)
linked to a complex interplay between mental illness,
Medians compared using Mann–Whitney U-test. IQR, interquartile substance abuse, medical non-adherence, and the qual-
range. ity of the therapeutic relationship (Gildberg et al.

© 2017 Australian College of Mental Health Nurses Inc.

496 B. MCKENNA ET AL.

2015). However, in the present study, the planned use interventions towards people exposed to a high number
of mechanical restraint in forensic mental health ser- of prolonged events might be an effective means of
vices often related to taking people offsite for medical reducing restrictive interventions (Hendryx et al. 2010).
reasons. In some circumstances, this was related to Therefore, a key implication arising from the results of
legal requirements, whereby it is mandatory for this research is the need for intensive reviews of
remand and sentenced prisoners (transferred to foren- repeated events, including second opinions and account-
sic mental health facilities) to be placed in mechanical ability to statutory authorities who have legislative over-
restraints when transported offsite. This is due to risk sight of restrictive interventions. Plans to prevent the
of absconding. Legally-required processes are outside use of restrictive interventions should be developed with
of clinical control. Therefore, consideration should be the person and their family (Oster et al. 2016).
given to whether it is beneficial to report prolonged In terms of the implications of the findings of the
mechanical restraints for the purpose of transfer within present study, access to whole jurisdiction datasets
the same dataset as prolonged mechanical restraint epi- potentially allows variation between services to be mon-
sodes, which can be influenced by clinical input. itored through the establishment of evidence-based
The reasons for the variation in the duration of benchmarks or standards of acceptable practice (Aus-
seclusion between adult area mental health services tralian Mental Health Outcomes and Classification
and the forensic mental health service are difficult to Network 2006). Such benchmarks allow for compar-
ascertain from the present study. Both service configu- isons across services, which can result in culture
rations cited risk of violence to others in the over- change within services and provide systemic oversight
whelming majority of cases involving prolonged (Scanlan 2010). In mental health settings, benchmark-
duration in the use of seclusion, as would be expected ing has been used to determine the effectiveness of,
(McKenna et al. 2014). Forensic mental health services and comparison with, existing treatments (Gibbons
were more likely to align this risk with clinical presen- et al. 2016; Reese et al. 2014). Benchmarking has also
tation. In adult area mental health services, the risk assisted with the development of quality indicators
was more likely to be aligned to contextual factors (Perlman et al. 2013), the use of which can assist in
unique to mental health inpatient services, such as identifying, and subsequently addressing, suboptimal
intoxication/withdrawal from illicit substances, police clinical outcomes (Fantini et al. 2016). It is heartening
involvement prior to admission, and the use of seclu- that there is evidence of moves towards a national stan-
sion to prevent absconding (leaving the premises with- dardization of data collection in Australia, which will
out permission). Intoxication/withdrawal from illicit make the development of benchmarks and future com-
substances is associated with the use of restrictive parisons across jurisdictions easier (Australian Institute
interventions in acute mental health inpatient settings, of Health and Welfare 2015). However, the marked
as is the involvement of police immediately prior to variation in the duration of both seclusion and mechan-
admission (McKenna et al. 2017). The use of seclusion ical restraint between Victoria in the present study and
as a nursing-management strategy for risk of abscond- South Australia (Oster et al. 2016) highlights the diffi-
ing in acute mental health inpatient services is counter- culties of arriving at a consensus on national bench-
intuitive, given that it might generate feelings of fear marks. Furthermore, variations between forensic and
that make absconding more likely (Muir-Cochrane area mental health services (Australian Institute of
et al. 2013). Less-restrictive alternatives, such as the Health and Welfare 2017), and in the present study,
use of intensive support and increased observation and indicate value in developing subspecialty benchmarks
engagement levels, are suggested (Grotto et al. 2014). in order to facilitate oversight by both the services
It is reasonable to assume that these contextual factors involved and statutory authorities with monitoring
within acute mental health services would be short responsibilities.
lived and quickly resolved. The findings that these con-
textual factors are associated with the prolonged dura-
LIMITATIONS
tion of restrictive interventions is not explained by the
present study, but are worthy of further investigation. Mental health services in Victoria were requested by
The findings regarding some consumers being subject the Office of the Chief Psychiatrist to report variance
to multiple episodes of prolonged restrictive interven- exceeding the benchmarks determined by the DHHS.
tions is consistent with the literature (Oster et al. 2016). It was difficult to accurately determine adherence with
Better tailoring of alternative, less-restrictive this request. Therefore, it is not possible to determine

© 2017 Australian College of Mental Health Nurses Inc.

PROLONGED DURATION OF RESTRICTIVE INTERVENTIONS 497

the extent to which the data reported are reflective of benchmarks for reporting prolonged restrictive inter-
all prolonged use of seclusion and mechanical restraint ventions might assist in reducing their use.
throughout the state of Victoria over the given time
period (i.e. the representativeness of the data RELEVANCE FOR CLINICAL PRACTICE
collected).
As with all retrospective cohort studies, the research Mental health nurses are critical contributors to reduc-
was limited to the analysis of variables collected, in this ing or eliminating, where possible, the use of restrictive
case via the form designed by the DHHS for the pur- interventions in mental health services. The ‘six core
pose of monitoring the individual and contextual factors strategies’ are a systemic, evidence-based approach to
surrounding the prolonged use of restrictive interven- reducing these controversial practices. One of the six
tions, and we note that the factors reported were not core strategies is the effective use of data to inform
fully reflective of those identified in the literature (Bul- clinical practice. The collection and reporting of
lock et al. 2014). Although the finding that males are restrictive intervention data are necessary to recognize
more likely to be involved in prolonged duration of trends in their use and measure progress towards their
restrictive interventions is consistent with the literature elimination. They also help to identify strategies to
(Oster et al. 2016), lack of more detailed information reduce the use of restrictive interventions. Mental
about demographics (e.g. diagnosis) and hospitalization health nurses are well positioned to advocate for, and
(e.g. locality, such as a high-dependency unit) to clarify implement the use of, less-restrictive alternatives.
this finding, is a limitation of the study. Space provided Although benchmarking allows system-wide monitor-
on the form to describe reasons for the prolonged use ing of practice, it also provides a means for mental
of the restrictive intervention was limited, and clini- health nurses to reflect on their own practice. Prolong-
cians completing the forms exercised choice to be brief ing the use of restrictive interventions beyond a bench-
or detailed in this regard. More research is required to mark duration signals to the nurse and the service that
understand these reasons, potentially through inter- the practice is potentially problematic, and that alterna-
viewing those directly involved, including clinicians, tive strategies need to be explored and prioritized. In
consumers, and family members. this way, the introduction of benchmarks would pro-
Finally, the present study did not involve considera- vide a means of informing clinical judgment and
tion of the use of physical restraint. In Victoria, the encouraging reflective practice to improve the quality
reporting of physical restraint is now mandatory under of mental health service delivery.
the Mental Health Act 2014. Once the quality of the
data pertaining to physical restraint is guaranteed, it REFERENCES
will be useful to describe the duration of the use of
Ashcraft, L. & Anthony, W. (2008). Eliminating seclusion and
physical restraint. The present study also excluded con- restraint in recovery-oriented crisis services. Psychiatric
sideration of specialty mental health services for chil- Services, 59, 1198–1202.
dren and adolescents, and services for older persons. Australian Institute of Health and Welfare. (2013). Mental
Health Services in Australia. [Cited 25 May 2017].
Available from: http://mhsa.aihw.gov.au/WorkArea/
CONCLUSION DownloadAsset.aspx?id=27917288164
Despite efforts to reduce or eliminate the use of Australian Institute of Health and Welfare. (2015). Key
Performance Indicators for Australian Public Mental
restrictive interventions, these data demonstrate that
Health Service. PI 15- Rate of seclusion, acute inpatient
seclusion and mechanical restraint continue to be used units, 2014-2015, Quality Statement . [Cited 25 May
for prolonged durations in forensic and adult mental 2017]. Available from: http://meteor.aihw.gov.au/content/
health settings in situations primarily where consumers index.phtml/itemld/624018
are at risk of harming others. Australian Institute of Health and Welfare. (2017). Use of
In describing episodes of prolonged seclusion and Restrictive Practices during Admitted Patient Care. [Cited
mechanical restraint in a state-wide jurisdiction, the 25 May 2017]. Available from: https://mhsa.aihw.gov.au/se
rvices/admitted-patient/restrictive-practices
present study provides the understanding required to
Australian Mental Health Outcomes and Classification
identify opportunities to reduce their use. Tailoring Network. (2006). National Mental Health Benchmarking
interventions towards consumers being subject to mul- Project Manual. Part 1: General Information for
tiple episodes of prolonged restrictive interventions and Participating Organisations. [Cited 17 November 2016].
developing specialty-specific (i.e. forensic or general) Available from: www.amhocn.org/publications

© 2017 Australian College of Mental Health Nurses Inc.

498 B. MCKENNA ET AL.

Bergk, J., Einsiedler, B., Flammer, E. & Steinert, T. (2011). Does patient compilation explain? Psychiatric Quarterly,
A randomised controlled comparison of seclusion and 84, 39–52.
mechanical restraint in inpatient settings. Psychiatric Larue, C., Dumais, A., Boyer, R., Goulet, M., Bonin, J. &
Services, 62, 1310–1317. Baba, N. (2013). The experience of seclusion and restraint
Bigwood, S. & Crowe, M. (2008). ‘It’s part of the job, but it in psychiatric settings: Perspectives of patients. Issues in
spoils the job’: A phenomenological study of physical Mental Health Nursing, 34, 317–324.
restraint. International Journal of Mental Health Nursing, Maguire, T., Young, R. & Martin, T. (2012). Seclusion
17, 215–222. reduction in a forensic mental health setting. Journal of
Bullock, R., McKenna, B., Kelly, T. et al. (2014). When Psychiatric and Mental Health Nursing, 19, 97–106.
reduction strategies are put in place and mental health Mann, C. (2003). Observational research methods. Research
consumers are still secluded: An analysis of clinical and design II: Cohort, cross sectional, and case-controlled
socio-demographic characteristics. International Journal of studies. Emergency Medicine Journal, 20, 54–60.
Mental Health Nursing, 23, 506–512. Mays, N. & Pope, C. (1995). Rigour and qualitative research.
Department of Health (2013). Providing a Safe Environment for All: British Medical Journal, 311, 109–112.
Framework for Reducing Restrictive Interventions. Melbourne, McKenna, B., Furness, T. & Maguire, T. (2014). A Literature
VIC: Mental Health, Drugs and Regions Divisions. Review and Policy Analysis on the Practice of Restrictive
El-Badri, S. & Mellsop, G. (2008). Patient and staff Interventions. Melbourne, VIC: State of Victoria
perspectives on the use of seclusion. Australasian Department of Health and Human Services.
Psychiatry, 16, 248–252. McKenna, B., McEvedy, S., Kelly, K. et al. (2017).
Fantini, G., Tibaldi, G., Rucci, P. et al. (2016). Quality of Care Association of methamphetamine use and restrictive
Indicators for Schizophrenia: Determinants of Observed interventions in an acute adult inpatient mental health
Variations among Italian Departments of Mental Health. Results unit: A retrospective cohort study. International Journal of
from the ETAS DSM Study. Epidemiology and Psychiatric Mental Health Nursing, 26, 49–55.
Sciences, https://doi.org/10.1017/S204579601600010X Muir-Cochrane, E., Oster, C., Grotto, J., Gerace, A. & Jones,
Gibbons, M., Gallop, R., Thompson, D. et al. (2016). J. (2013). The inpatient psychiatric unit as both a safe and
Comparative effectiveness of cognitive therapy and unsafe place: Implications for absconding. International
dynamic psychotherapy for major depressive disorder in a Journal of Mental Health Nursing, 22, 304–312.
community mental health setting: A randomized clinical National Mental Health Working Group. (2005). National
non-inferiority trial. JAMA Psychiatry, 73, 904–911. Safety Priorities in Mental Health: A National Plan for
Gildberg, F., Fristed, P., Makransky, G., Moeller, E., Nielsen, Reducing Harm, Health Priorities and Suicide Prevention.
L. & Bradley, S. (2015). As time goes by: Reasons and Canberra, ACT: Department of Health and Ageing,
characteristics of prolonged episodes of mechanical Commonwealth Government of Australia.
restraint in forensic psychiatry. Journal of Forensic Nursing, Noda, T., Sugiyama, N., Sato, M. et al. (2013). Influence of
11, 41–50. patient characteristics on duration of seclusion/restraint in
Grotto, J., Gerace, A., O’Kane, D., Simpson, A., Oster, C. & acute psychiatric settings in Japan. Psychiatry and Clinical
Muir-Cochrane, E. (2014). Risk assessment and absconding: Neuroscience, 67, 405–411.
Perceptions, understandings and responses of mental health Office of the Chief Psychiatrist (2016). Chief Psychiatrist’s
nurses. Journal of Clinical Nursing, 24, 855–865. Annual Report. Melbourne, VIC: Department of Health
Happell, B. & Gaskin, C. (2011). Exploring patterns of and Human Services.
seclusion use in Australian mental health services. Oster, C., Gerace, A., Thomson, D. & Muir-Cochrane, E. (2016).
Archives of Psychiatric Nursing, 25, e1–e8. Seclusion and restraint use in adult inpatient mental health
Hendryx, M., Trusevich, Y., Coyle, F., Short, R. & Roll, J. care: An Australian perspective. Collegian, 23, 183–190.
(2010). The distribution and frequency of seclusion and/or Parish, C. (2013). Change ward culture to cut violence and
restraint among psychiatric inpatients. The Journal of aggression. Mental Health Practice, 16, 6–7.
Behavioral Health Services and Research, 37, 272–281. Perlman, C., Hirdes, J., Barbaree, H. et al. (2013).
Huckshorn, K. (2004). Reducing seclusion & restraint use in Development of mental health quality indicators (MHQIs)
mental health settings core strategies for prevention. for inpatient psychiatry based on the interRAI mental
Journal of Psychosocial Nursing, 42, 22–33. health assessment. BMC Health Services Research, 13, 15.
Hui, A., Middleton, H. & Vollm, B. (2013). Coercive measures Reese, R., Duncan, B., Bohanske, R., Owen, J. & Minami, T.
in forensic settings: Findings from the literature. (2014). Benchmarking outcomes in a public behavioral
International Journal of Forensic Mental Health, 12, 53–67. health setting: Feedback as a quality improvement
Janssen, W., van de Sande, R., Noorthoorn, E. et al. (2011). strategy. Journal of Consulting and Clinical Psychology,
Methodological issues in monitoring the use of coercive 82, 731–742.
measures. International Journal of Law and Psychiatry, Scanlan, J. (2010). Interventions to reduce the use of
34, 429–438. seclusion and restraint in inpatient psychiatric settings:
Janssen, W., Noorthoorn, E., Nijman, H. et al. (2013). What we know so far and review of the literature.
Differences in seclusion rates between admission wards: International Journal of Social Psychiatry, 56, 412–423.

© 2017 Australian College of Mental Health Nurses Inc.

PROLONGED DURATION OF RESTRICTIVE INTERVENTIONS 499

Staggs, V. (2015). Trends in use of seclusion and restraint in the literature on its use, staff and patient views, and the
response to injurious assault in psychiatric units in U.S. impact of training. Journal of Psychiatric and Mental
hospitals, 2007–2013. Psychiatric Services, 66, 1369–1372. Health Nursing, 16, 99–105.
Steinert, T., Lepping, P., Bernhardsgrutter, R. et al. (2010). Tyner, S., Beckley, J., Goel, D. et al. (2012). Factors affecting
Incidence of seclusion and restraint in psychiatric the practice of seclusion in an acute mental health service
hospitals: A literature review and survey of international in Southland, New Zealand. The Psychiatrist, 36, 214–218.
trends. Social Psychiatry and Psychiatric Epidemiology, Vaismoradi, M., Turunen, H. & Bondas, T. (2013). Content
45, 889–897. analysis and thematic analysis: Implications for conducting
Steinert, T., Birk, M., Flammer, E. & Bergk, J. (2013). a qualitative descriptive study. Nursing and Health
Subjective distress after seclusion or mechanical restraint: Sciences, 15, 398–405.
One-year follow-up of a randomized controlled study. Wieman, D., Camacho-Gonsalves, T., Huckshorn, K. & Leff,
Psychiatric Services, 64, 1012–1017. S. (2014). Multisite study of an evidence-based practice to
Stubbs, B., Leadbetter, D., Paterson, B., Yorston, G., Knight, reduce seclusion and restraint in psychiatric inpatient
C. & Davis, S. (2009). Physical intervention: A review of facilities. Psychiatric Services, 65, 345–351.

© 2017 Australian College of Mental Health Nurses Inc.