Available via license: CC BY 4.0
Content may be subject to copyright.
Page 1/21
Behavioral skills training for teaching safety skills to
mental health service providers compared to
training-as-usual: a pragmatic randomized control
trial
Elizabeth Lin ( Elizabeth.lin@camh.ca )
Centre for Addiction and Mental Health
Mais Malhas
Centre for Addiction and Mental Health
Emmanuel Bratsalis
Centre for Addiction and Mental Health
Kendra Thomson
Centre for Addiction and Mental Health
Fabienne Hargreaves
Centre for Addiction and Mental Health
Kayle Donner
Centre for Addiction and Mental Health
Heba Baig
Centre for Addiction and Mental Health
Rhonda Boateng
Centre for Addiction and Mental Health
Rajlaxmi Swain
Centre for Addiction and Mental Health
Mary Benisha Benadict
Centre for Addiction and Mental Health
Louis Busch
Centre for Addiction and Mental Health
Research Article
Keywords: Workplace Violence, Violence Prevention, Behavioural Skills Training, Performance and
Competency-Based Staff Training
Posted Date: September 12th, 2023
Page 3/21
Abstract
Background:
Violence in the healthcare workplace has been a global concern for over two decades, with a high
prevalence of violence towards healthcare workers reported. Workplace violence has become a healthcare
quality indicator and embedded in quality improvement initiatives of many healthcare organizations. The
Centre for Addiction and Mental Health (CAMH), Canada’s largest mental health hospital, provides all
clinical staff with mandated staff safety training for self-protection and team-control skills. These skills
are to be used as a last resort when a patient is at imminent risk of harm to self or others. The purpose of
this study is to evaluate the ecacy by comparing two training methods of this mandated staff safety
training for workplace violence in a large psychiatric hospital setting.
Methods:
Using a pragmatic randomized control trial design, this study compares two approaches to teaching
safety skills; CAMH’s training-as-usual (TAU) using the 3D approach (description, demonstration and
doing) and behavioural skills training (BST), from the eld of applied behaviour analysis, using
instruction, modeling, practice and feedback loop. Staff were assessed on three outcome measures
(competency, mastery and condence), across three time points: before training (baseline), immediately
after training (post-training) and one month later (follow-up). This study has been registered
(ISRCTN18133140, September 6, 2023).
Results:
With a sample size of 99 new staff, results indicate that BST was signicantly better than TAU in
improving observed performance of self-protection and team-control skills. Both methods were
associated with improved skills and condence. However, there was a decrease in skill performance
levels at the one-month follow-up for both methods, with BST remaining higher than TAU scores across
all three time points. The impact of training improved staff condence in both training methods and
remained high across all three time points.
Conclusions:
The study ndings suggest that BST is more effective than TAU in improving safety skills among
healthcare workers. However, the retention of skills over time remains a concern, and therefore a single
training session without on-the-job-feedback or booster sessions based on objective assessments of skill
may not be sucient. Further research is needed to conrm and expand upon these ndings in different
settings.
Page 4/21
Introduction
Violence in the healthcare workplace has been a global concern for over two decades. In 2002, a joint
task force of the International Labour Oce (ILO), World Health Organization, Public Services
International, and the International Council of Nurses created an initiative to address this issue (1). One
result was the documentation of a high international prevalence of violence towards healthcare workers
showing that as many as half or more experienced physical or psychological violence in the previous year
(2, 3). Since then, workplace violence has become a healthcare quality indicator and been embedded in
the quality improvement initiatives of many healthcare organizations (for example, Health Quality Ontario
(4)). Conceptually, it is also reected in the expansion of the Triple Aim framework to the Quintuple Aim to
include staff work-life experience (5).
Despite these efforts, the high prevalence of workplace violence in healthcare persists (6). Two meta-
analyses, representing 393,344 healthcare workers, found a 19.3% pooled prevalence of workplace
violence in the past year among which 24.4% and 42.5% reported physical and psychological violence
experiences, respectively (7, 8). The literature also highlighted that workers in mental health settings were
at particular risk (8, 9). An Irish study found that threats and assaults were the prevalent form of violence
towards mental health workers (9). Eighty percent of mental health nurses in a Jordanian cross-sectional
correlation study reported a violent incident in the preceding 24 months (10). Workplace violence has
been associated with negative psychological, physical, emotional, nancial, and social consequences
which impact staff’s ability to provide care and function at work (11, 12). A 7-year, population-based,
follow-up study in Denmark highlighted the long-term impact of physical and psychological health issues
owing to physical workplace violence (13). Two studies, one in Italy (14) and one in Pakistan (15), have
linked workplace violence to demoralization and declining quality of healthcare delivery and job
satisfaction among healthcare workers.
Building on these efforts, the ILO published a 2020 report recommending the need for national and
organizational work environment policies and workplace training “…on the identied hazards and risks of
violence and harassment and the associated prevention and protection measures….” (16, p. 55).
Consequently, many countries (17–19) have committed to creating a safe work environment. In Ontario,
Canada, the government has provided guidelines for preventing workplace violence in healthcare (4, 20),
and our institution, the Centre for Addiction and Mental Health, launched a major initiative in 2018 to
address the physical and psychological safety of patients and staff (21). A priority component of this
initiative is mandatory training for all new clinical staff on trauma-informed crisis prevention, de-
escalation skills, and, in particular, safe physical intervention skills (22, 23).
However, the effects of such training, especially for managing aggressive behaviour, are only partially
understood. A 2015 systematic review on training for mental health staff (24) and a more recent
Cochrane review on training for healthcare staff (25) reported remarkably similar ndings. Both noted the
uneven evidence (due to methodological issues, small numbers of studies, inconsistent results) which
made denitive conclusions about the merits and ecacy of training dicult. The more consistent
Page 5/21
impacts found by Price and colleagues (24) were improved knowledge and staff condence in their
ability to manage aggression. There was some evidence of improved de-escalation skills including the
ability to deal with physical aggression (26, 27) and verbal abuse (28). However, these studies were
limited because they used unvalidated scales or simulated, rather than real-world, scenarios. For
outcomes such as assault rates, injuries, the incidence of aggressive events, and the use of physical
restraints, the ndings were mixed or dicult to generalize due to the uneven evidence.
Similarly, Geoffrion and colleagues (25) found some positive effect of skills-training on knowledge and
attitudes, at least short-term, but noted that support for longer-term effects was less sure. The evidence
for impacts on skills or the incidence of aggressive behaviour was even more uncertain. They also noted
that the literature was limited because it focused largely on nurses. They concluded, “education
combined with training may not have an effect on workplace aggression directed toward healthcare
workers, even though education and training may increase personal knowledge and positive attitudes”
(25, p. 2). Among their recommendations were the need to evaluate training in higher-risk settings such as
mental healthcare, include other healthcare professionals along with nurses, and use more robust study
designs. In addition, the literature evaluating training procedures focussed on self-reported rather than
objective measures of performance.
The present study sought to address these gaps by comparing two approaches to teaching safety skills
for managing aggressive patient/client behaviour. The setting was a large psychiatric teaching hospital;
the sample was drawn from all new clinical staff attending their mandated on-boarding training; and we
used a pragmatic randomized control trial design. Our control intervention was the current training-as-
usual (TAU) in which trainers “describe”, “demonstrate”, and “do” but without objective performance
measurement. Our test intervention was behavioural skills training (BST) (29, 30) drawn from the eld of
applied behaviour analysis (31). BST is a performance- and competency-based training model that uses
an instructional, modeling, practice, and feedback loop to teach targeted skills to a predetermined
performance level. Checklists guide the instructional sequence and the determination of whether or not
the predetermined performance threshold has been reached. Considerable evidence indicates that BST
can yield signicant improvement in skills post-training, over time, and across different settings (32–34).
It has been used to train a wide range of participants, including behavior analysts, parents, and educators,
to build safety-related skills and manage aggressive behavior (32, 35, 36).
Methods
As previously described (37), our objective was to compare the effectiveness of TAU against BST. Our null
hypotheses were that these methods would not differ signicantly in:
1. Observer assessment of self-protection and team-control physical skills.
2. Self-assessed condence in using those skills.
Study participants were recruited from all newly-hired clinical staff attending a mandatory two-week
orientation. Staff were required to register beforehand for a half-day, in-person, physical safety skills
Page 6/21
session. They were randomized to a session at the time of registration, and the sessions were then
randomized to TAU or BST. All randomization was performed by RB using GraphPad software (38). On
the day of training, informed consent for study participation was obtained privately from each attendee.
The trainers and session attendees were thus unaware of who was or was not in the study. Recruitment
began January 2021, after ethics approval, and continued until September 2021 when the target of at
least 40 study participants completing all assessments for each training condition was reached. The
target sample size was chosen to allow 80-percent power to detect a medium to large effect size (39).
Both methods taught the same 11 target skills for safely responding to patients/clients that may exhibit
harm to self or others (e.g., aggressive behaviour) during their hospital admission. These skills, dened by
the hospital as mandatory for all newly hired staff, included six self-protection and ve team-control
(physical restraint) skills (see Appendix A). Each target skill had dened components and a specic
sequence in which they were taught as outlined on performance checklists (see Appendix B for a
checklist example).
The two methods differed in how these sequences were administered. For BST, checklists guided the
training sequence (instruction, modeling, rehearsal and feedback) and indicated to the trainer when
successful performance was reached and thus when the trainee was ready to move on to the next skill
(30) (see Appendix C for BST sequence). In BST, common practice is to dene success as up to three
correct, consecutive executions (40). In our study, the trainers ensured 80% correct performance of each
skill at least once and aimed for up to 5 times in a row before moving on to the next skill. In contrast,
while TAU included elements of modeling, practice, and feedback, it did not systematically assess skill
acquisition nor impose any specic level of success before proceeding to the next skill.
Measures.
There were three outcome measures, two observer-based assessments of skill acquisition (competence
and mastery) and one self-reported condence measure. These were assessed at three time points:
immediately before training (baseline), immediately after training (post-training), and one month later
(follow-up). Descriptive information (professional role, department) was provided by the hospital for all
registrants. The research team estimated the kind of patient contact (direct, less direct, rare/low) based
on the combination of professional role and department. For condentiality reasons, the hospital did not
provide information on registrant personal characteristics (e.g., age, gender/sex), and, for the same
reasons, we did not collect this information for the study. Participants were also asked at baseline and
follow-up how many events they encountered in the previous month that required the use of these skills.
This information was collected because of our interest in testing a post-hoc hypothesis that those who
had actual experience would score higher than those who did not.
All assessments were carried out following a standardized protocol. To ensure that registrants remained
blinded to which colleagues were in the study, each registrant’s skill acquisition was assessed privately by
a research team member at baseline and post-training using the performance checklists. Only
assessments for those consenting to participate were videotaped. Study participants were then asked to
Page 7/21
return one month later for a follow-up assessment which was also videotaped. Videotapes were scored
using the BST checklists by trained observers who were blind to the participant’s training method using
the checklists developed for this study and also used in the BST session. As described previously (37),
interobserver agreement (IOA) was routinely evaluated throughout the study with the nal value being
96% across the 33% of the performance assessment videos scored for the IOA calculation.
Skill acquisition outcomes were calculated using the checklist-based observer assessments of the
videotapes. The percentage of correctly executed steps for each target skill was established. Then, these
percentages were averaged across the six self-protection target skills and across the ve team-control
target skills to create competence scores. Finally, a predened threshold of 80% was applied to the
competence scores to determine which participants met the mastery threshold (41, 42).
Statistical analysis
.
R software was used to generate descriptive statistics (frequencies, percentages) and test our hypotheses
(43). Generalized linear mixed models (GLMM) were used to test nested main and interaction effects
using likelihood-ratio chi-square statistics for the post-training and follow-up results as there were no
baseline differences. GLMM was also used to evaluate BST-TAU differences at the three study time points
(44, 45). For the BST-TAU comparisons, we used Cohen’s d as a guide for evaluating the practical
signicance of the differences for the continuous measures (competence, condence). We used Cohen’s
suggested thresholds (46) of 0.2, 0.5, and 0.8 for small, medium, and large effect sizes conservatively by
applying them to both the point estimates and 95% condence intervals. Thus, for example, a Cohen’s d
where the condence interval went below 0.2 would be interpreted as non-meaningful. For the categorical
measure of mastery, we used BST-TAU risk ratios. Condence intervals for all effect size measures were
obtained using bootstrapping. Independent-samples t-tests were used for the post-hoc analyses and,
along with chi-square tests, to compare the completers and non-completers.
Results
One hundred ninety-nine staff consented to participate in the study out of a total of 360 session
attendees (55%). Of these, 108 (54%) had been randomly assigned to a BST session and 91 (46%) to a
TAU session. Half (n=99) completed assessments at all three time points (44 % TAU; 55% BST). These 99
(hereafter ‘study completers’) constituted 28 percent of all session attendees.
Among the non-completers, 53 had been assigned to BST and 47 to TAU. Eight were classied as
incomplete because of technical software issues when video-recording one of their assessments and one
(the rst participant) because the IOA process prompted substantive changes to the assessment
checklist. The primary reason for the remaining non-completers was missing the follow-up assessment
(91 individuals: 50/53 BST, 41/47 TAU) largely due to diculties scheduling a non-mandatory event
during the pandemic (e.g., units restricting staff from leaving because of clinical staff shortages or
patient outbreaks, staff illness).
Page 8/21
Descriptive information for the expected degree of patient contact and for hospital department is shown
in Table 1 for study participants (completers, non-completers), non-participants, and the total group of
session attendees. When comparing study participants who were completers versus non-completers,
there were higher percentages of non-completers who had direct patient contact (97 vs 94%), were nurses
(48 vs 37%), or worked on inpatient units (64 vs 58%). Similarly, when compared to all session attendees,
study completers had lower proportions of direct contact with patients and of nursing staff.
Table 1:Expected Patient Contact and Department Types for Study Participants, Non-Participants, and
Total Session Attendees
Characteristic Study Participants
(n=199)
Non-
Participants
(n=161)
Total Session Attendees
(n=360)
Completers
n=99
Non-
Completers
n=100
Expected Patient Contact (%)
Direct
Nurse
Security
Other
Less Direct
Rare/None
94
37
6
51
4
2
97
48
5
44
3
0
96
47
3
47
3
1
96
44
4
47
3
1
Department Type (%)
Inpatient
Outpatient
Both
Hospital
Admin
58
9
30
2
64
9
26
1
51
14
33
1
56
11
30
1
Figure 1 depicts the self-protection and team-control competence scores for the study completers (left
and right sides, respectively). The hypothesis-testing results showed a signicant difference by training
Method (self-protection: chi-square=34.46, 1 df, p<.001; team-control: chi-square=50.42, 1 df, p<.001).
Page 9/21
There was also a signicant decline between post-training and follow-up (Time) for both skill categories
independent of Method (self-protection: chi-square=81.29, 1 df, p<.001; team-control: chi-square=56.51, 1
df, p<.001), and a signicant Method-by-Time interaction independent of Method and Time for team-
control skills (chi-square=17.41, 1 df, p<.001). BST-TAU comparisons showed no difference at baseline
for either type of skill (not shown). However, BST was signicantly better than TAU at both post-training
(self-protection: Cohen’s d=1.45 [1.02, 1.87], large ES; team-control: Cohen’s d=2.55 [2.08, 3.02]; large ES)
and follow-up (respectively – Cohen’s d=0.82 [0.40, 1.23], small ES; Cohen’s d=0.62 [0.21, 1.03], small ES).
For both methods, competence scores dropped between post-training and follow-up although not to the
original baseline levels.
Figure 1: Mean Observer-rated Competence in Self-Protection and Team-control skills across three time
points by TAU and BST
Insert Figure 1 about here
The skill mastery results for the study completers are shown in Figure 2. The mastery patterns paralleled
the competence patterns in that BST was signicantly better than TAU (self protection: chi-square=28.82,
1 df, p< .001; team-control: chi-square=72.87, 1 df, p<.001). There was also a signicant Time effect
independent of Method (self-protection: chi-square=27.54, 1 df, p<.001; team-control: chi-square=33.03, 1
df, p<.001). There were no signicant interactions for either type of skill once the effects of Method and
Time were accounted for. BST-TAU comparisons showed no difference in percent achieving Mastery at
baseline (not shown) but large risk ratios at both post-training (self-protection: 13.43 [4.01, >1000]; team-
control: 31.24 [8.45,>1000] and follow-up [self-protection: 12.30 [1.58, >1000]; team-control: 30.60 [6.75.
>1000]).
Figure 2: Percent of Participants Achieving Mastery[1] in Self-Protection and Team-control skills across
three time points by TAU and BST
Insert Figure 2 about here
Condence scores for the study completers are shown in Figure 3. The only signicant main effect was
for Time (self-protection: chi-square=57.15, 1 df, p<.001; team-control: chi-square=43.25, 1 df, p<.001).
For both skill categories, the scores increased between baseline and post-training and then dropped at
follow-up but not to the original baseline levels.
Figure 3: Mean Self-rated Condence in Self-Protection and Team-control skills across three time points
by TAU and BST
Insert Figure 3 about here
To assess what impact the high no-show rate for the follow-up could have had, we compared the
completers and the non-completers on the six post-training outcomes (self-protection and team-control:
competence, mastery, and condence). Non-completers had slightly lower scores than completers except
Page 10/21
for the two condence measures where their self-assessments were higher (not shown). However, the
only signicant difference between the two groups was for self-protection condence means (6.9 vs 6.3,
completers vs non- completers, t=2.40, 195 df, p=0.17).
In terms of past-month experience, few study completers reported events requiring self-protection (19 at
baseline, 9 at follow-up) or team-control skills (14 at baseline, 14 at follow-up). Consequently, we only
examined the presence or absence of experience without breaking it down by training method. We found
non-signicant results for both competence and mastery (not shown) but a potential impact on
condence for self-protection skills at follow-up and for team-control skills at baseline and post-training
(Figure 4).
Figure 4: Mean Self-rated Condence and 95-percent condence intervals in Self-protection and Team-
control skills across three time points by Past-month occasion to use skills
Insert Figure 4 about here
Summary and Discussion
Our strongest nding was that BST was signicantly better than TAU in improving the observed
performance of self-protection and team-control skills. While follow-up scores decreased for both
methods, BST scores remained higher than TAU scores. The impact of training on staff condence differs
from these patterns in that condence scores improved noticeably at post-training and remained
relatively high at follow-up. Further, our post-hoc analyses suggested that recent experience using safety
skills might have a greater impact on condence than on observed skill performance. We also found that
training, regardless of method, was independently associated with improved observer-scored skills and
self-reported condence.
The better performance of BST has a certain face validity since it is more structured and intensive than
TAU. However, there are at least two questions regarding whether it produced the expected results. The
BST framework requires continued rehearsal and feedback until a specied performance criterion is
reached (30). However, our mandatory safety training had practical, unmodiable constraints. The
institution required the safety-training sessions be completed in 3.5 hours which meant that BST trainers
were limited in their ability to use some of the more stringent performance criteria described in the
literature. For example, it was not practical to set the performance criterion at higher than 80 percent. In
addition, all BST completers were able to demonstrate 80-percent correct performance for each skill at
least once, but not all were able to demonstrate ve consecutive, correct executions within the allotted
time. If the requirement of ve in a row at 80% or higher had been implemented, then the post-training
scores (and potentially the 1-month follow-up scores) for the BST completers could have been higher.
Page 11/21
A second question is what level of skill retention should be expected at follow-up. The BST scores at one-
month follow-up constituted 66% and 74% of the competence scores at post-training (self-protection and
team-control, respectively) and 41% and 43% of the mastery percentages at post-training (self-protection
and team-control, respectively). Although BST and elements of performance feedback models have been
found to be effective in staff training with successful retention over time (47–51), nding appropriate
comparators for our study was challenging because there are no studies where BST has been used for
training such a large and diverse group of staff. However, the broader literature does suggest that our
results are consistent with or somewhat lower than those from other studies. Oah et al. (52) found that
45 percent of medical students retained the full set of clinical skills 18 months after completing
simulation training, and Bruno and colleagues (53) found published retention rates ranging between 75
and 85 percent across time periods between four to 24 months and across diverse disciplinary elds.
Regardless of the comparators, the loss in skill performance after one-month post-training is a concern.
Our interpretation is that reliance on a single session, even with highly structured and competency-based
methods, is not adequate particularly in the context of managing distressing events. Efforts should be
made to allow for exibility with respect to setting higher thresholds for success despite organizational
restraints for staff training. Furthermore, settings that require these skills to be performed more reliably
for both patient and staff safety (e.g., emergency departments, acute care settings, security services)
should consider on-the-job feedback or booster sessions based on objective assessments of skill rather
than on pre-set amounts of time (e.g. annual refresher). This would be more consistent with the BST
literature, as on-the-job training should occur based on an evidence-based approach.
The pattern that we found of a differential impact of training on condence versus demonstrable skills is
consistent with other research (24, 25, 54). For both of our study groups, condence ratings at 1-month
follow-up remained relatively high (83% and 86% retention for self-protection and team-control,
respectively) and did not parallel the retention rates for either competence or mastery. It is noteworthy
that our ndings repeat previously-reported patterns despite our attempts to close some of the gaps
identied: specically, studying a higher-risk setting, including non-nursing healthcare professionals, and
using a more rigorous study design (25). It is also interesting that our exploratory analyses suggest that
experience may impact condence but not skill acquisition.
The important question is how to interpret these discrepancies. It may be that condence and
competence are orthogonal constructs with different causal/relational associations with training or
experience. Barsuk, et al. (55), for example, found that the year of the medical resident and their
procedural experience did not predict skill mastery. Similarly, Choudhry and colleagues (56) in a
systematic literature review concluded that the more experienced physicians might be at risk for providing
lower-quality care. With respect to condence and performance, Magnacca and colleagues (54) found
that participants being taught to facilitate Acceptance and Commitment Training (ACT) exercises had a
high baseline condence which was inconsistent with their directly observed performance. Interestingly,
these condence ratings decreased after follow-up sessions which the authors speculated could be due
to the fact participants encountered the complexities of facilitating ACT. It may also be that the
Page 12/21
measurement method is an important factor. Self-assessment, for example, has been consistently found
to be an imperfect measure of actual performance or knowledge (57, 58). Whether or not attitudes and
condence contribute positively in ways that are independent of actual skills or negatively in that they
provide an inappropriate sense of competence merits exploration.
The major strength of our study is its design. Currently, we have identied only one other study evaluating
the impact of BST training for clinical staff using an RCT design (36). Another strength is our inclusion of
a large percentage of non-nursing, direct-care staff. Finally, we used both self-reported and observer-
assessed outcome measures. These strengths allow us to add to the evidence base already established
in the literature.
However, interpretation of our results should consider several limitations. Conducting a research study on
full-time clinical staff during a pandemic meant that a high percentage of those consenting to be in the
study did not complete their 1-month follow-up assessment. The reported reasons for missing the third
assessment (unit restrictions or short stang because of the pandemic) are consistent with the
demographic differences between completers and non-completers in that they were more likely to be
nurses or working on inpatient units. Our comparison of the post-training scores of the completers and
non-completers suggested that the no-shows had slightly lower post-training observed skill performance
(but slightly better condence ratings). If we had managed to assess the non-completers at follow-up, our
reported ndings may have been diluted although it is unlikely that this would have completely negated
the large effect sizes.
Another limitation, as identied by Price, et al. (24), is that we used articial training scenarios, though
this may be unavoidable given the low frequency of aggressive events and the ethics of deliberately
exposing staff to these events. Also, we only measured the skills directly related to handling client/patient
events. We were not able to access information on event frequency or severity, staff distress and
complaints, or institutional-level measures such as lost workdays due to sick leave, staff turnover, or
expenditures (25, 59). A further gap, which is important but dicult to assess, is whether there is any
impact of staff safety training on the clients or patients who are involved.
Given these strengths and limitations, we see our study as adding one piece of evidence that needs to be
a) conrmed or disconrmed by other researchers in both the same and different settings and b)
understood as part of a complex mix of ingredients. Future research on these fronts will hopefully
contribute to maintaining and improving workplace safety.
Declarations
This study was approved by the Research Ethics Board of the Centre for Addiction and Mental Health
(#101/2020). Informed consent was obtained from all subjects participating in the study. All
interventions were performed in accordance with the Declaration of Helsinki.
Consent for publication: not applicable
Page 13/21
Availability of data and materials: The dataset generated and analysed during the current study is
not publicly available due to the fact that it is part of a larger internal administrative data collection
but is available from the corresponding author on reasonable request.
Competing interests: none
Funding: This research was funded internally by the Centre for Addiction and Mental Health.
Authors’ contributions:All authors were involved in the study design, monitoring and implementing
the study, and review of manuscript drafts. EL was responsible for the original study design and
drafting of the full manuscript. MM, EB, and FH led the implementation of the training sessions. EB,
FH, HB, KT, and LB were involved in the reliability assessments (IOA). KD and HB were primarily
responsible for data analysis. HB and RB monitored the data collection and the ongoing study
procedures. RS and MBB assisted in the literature review.
Acknowledgements: We thank Sanjeev Sockalingam, Asha Maharaj, Katie Hodgson, Erin Ledrew,
Sophie Soklaridis, and Stephanie Sliekers for their guidance and for dedicating the human and
nancial resources needed to support this study. We also want to express our sincere gratitude to the
following individuals for facilitating physical skills sessions and for volunteering as actors in the
physical skills demonstrations: Kate Van den Borre, Steven Hughes, Paul Martin Demers, Ross Violo,
Genevieve Poulin, Stacy de Souza, Narendra Deonauth, Joanna Zygmunt, Tessa Donnelly, Lawren
Taylor, and Bobby Bonner. Finally, we are grateful to Marcos Sanchez for statistical consultation.
References
1. International Labour Organization. Joint Programme Launches New Initiative Against Workplace
Violence in the Health Sector [Internet]. 2002 [cited 2022 Aug 5]. Available from:
https://www.ilo.org/global/about-the-ilo/newsroom/news/WCMS_007817/lang--en/index.htm
2. Di Martino V. Workplace violence in the health sector. Country case studies Brazil, Bulgaria, Lebanon,
Portugal, South Africa, Thailand and an Additional Australian Study. Geneva, Switzerland; 2002.
3. Needham I, Kingma M, O’Brien-Pallas L, McKenna K, Tucker R, Oud N, editors. Workplace Violence in
the Health Sector. In: Proceedings of the First International Conference on Workplace Violence in the
Health Sector - Together, Creating a Safe Work Environment. Amsterdam, The Netherlands; 2008. p.
383.
4. Health Quality Ontario. Quality Improvement Plan Guidance: Workplace Violence Prevention. 2019.
5. Bodenheimer T, Sinsky C. From triple to Quadruple Aim: Care of the patient requires care of the
provider. Ann Fam Med. 2014;12(6):573–6.
. Somani R, Muntaner C, Hillan E, Velonis AJ, Smith P. A Systematic Review: Effectiveness of
Interventions to De-escalate Workplace Violence against Nurses in Healthcare Settings. Saf Health
Work [Internet]. 2021;12(3):289–95. Available from: https://doi.org/10.1016/j.shaw.2021.04.004
7. Li Y, Li RQ, Qiu D, Xiao SY. Prevalence of workplace physical violence against health care
professionals by patients and visitors: a systematic review and meta-analysis. Int J Environ Res
Page 14/21
Public Health. 2020;17(1):299.
. Liu J, Gan Y, Jian H, Li L, Dwyer R, Lu K, et al. Prevalence of workplace violence against healthcare
workers: a systematic review and meta-analysis. Occup Environ Med. 2019;76(12):927–35.
9. O’Rourke M, Wrigley C, Hammond S. Violence within mental health services: how to enhance risk
management. Risk Manag Healthc Policy. 2018;11:159–67.
10. Al-Azzam M, Al-Sagarat AY, Tawalbeh L, Poedel RJ. Mental health nurses’ perspective of workplace
violence in Jordanian mental health hospitals. Perspect Psychiatr Care. 2018;54(4):477–87.
11. Lim M, Jeffree M, Saupin S, Giloi N, Lukman K. Workplace violence in healthcare settings: The risk
factors, implications and collaborative preventive measures. Ann Med Surg. 2022;78:103727.
12. Lanctôt N, Guay S. The aftermath of workplace violence among healthcare workers: A systematic
literature review of the consequences. Aggress Violent Behav. 2014;19(5):492–501.
13. Friis K, Pihl-Thingvad J, Larsen FB, Christiansen J, Lasgaard M. Long-term adverse health outcomes
of physical workplace violence: a 7-year population-based follow-up study. Eur J Work Organ
Psychol. 2019;28(1):101–9.
14. Berlanda S, Pedrazza M, Fraizzoli M, de Cordova F. Addressing Risks of Violence against Healthcare
Staff in Emergency Departments: The Effects of Job Satisfaction and Attachment Style. Biomed Res
Int. 2019;May 28(5430870):1–12.
15. Baig LA, Ali SK, Shaikh S, Polkowski MM. Multiple dimensions of violence against healthcare
providers in Karachi: Results from a multicenter study from Karachi. J Pakistani Med Assoc.
2018;68(8):1157–65.
1. International Labour Organization. Safe and healthy working environments free from violence and
harassment [Internet]. International Labour Organization. Geneva, Switzerland: International Labour
Organization; 2020. Available from: https://www.ilo.org/global/topics/safety-and-health-at-
work/resources-library/publications/WCMS_751832/lang--en/index.htm
17. Wirth T, Peters C, Nienhaus A, Schablon A. Interventions for Workplace Violence Prevention in
Emergency Departments: A Systematic Review. Int J Environ Res Public Heal. 2021;18(16):8459.
1. U.S. Congress. Workplace Violence Prevention for Health Care and Social Service Workers Act
[Internet]. U.S.A.; 2020. Available from: https://www.congress.gov/bill/116th-congress/house-
bill/1309
19. Lu A, Ren S, Xu Y, Lai J, Hu J, Lu J, et al. China legislates against violence to medical workers.
TneLancet Psychiatry. 2020;7(3):E9.
20. Government of Ontario Ministry of Labour T and SD. Preventing workplace violence in the health care
sector | ontario.ca [Internet]. 2020 [cited 2020 Jul 28]. Available from:
https://www.ontario.ca/page/preventing-workplace-violence-health-care-sector
21. Safe and Well Committee. Safe & Well Newsletter. 2018; Available from:
http://insite.camh.net/les/SafeWell_Newsletter_October2018_107833.pdf
Page 15/21
22. Heckemann B, Zeller A, Hahn S, Dassen T, Schols JMGA, Halfens RJG. The effect of aggression
management training programmes for nursing staff and students working in an acute hospital
setting. A narrative review of current literature. Nurse Educ Today. 2015;35(1):212–9.
23. Patel MX, Sethi, Faisil N Sethi Barnes TR, Dix R, Dratcu L, Fox B, Garriga M, et al. Joint BAP NAPICU
evidence-based consensus guidelines for the clinical management of acute disturbance: de-
escalation and rapid tranquillisation. J Psychiatr Intensive Care. 2018;14(2):89–132.
24. Price O, Baker J, Bee P, Lovell K. Learning and performance outcomes of mental health staff training
in de-escalation techniques for the management of violence and aggression. Br J Psychiatry.
2015;206:447–55.
25. Geoffrion S, Hills DJ, Ross HM, Pich J, Hill AT, Dalsbø TK, et al. Education and training for preventing
and minimizing workplace aggression directed toward healthcare workers. Cochrane Database Syst
Rev [Internet]. 2020 Sep 8 [cited 2021 Dec 3];9(Art. No.: CD011860). Available from:
/pmc/articles/PMC8094156/
2. Paterson B, Turnbull J, Aitken I. An evaluation of a training course in the short-term management of
violence. Nurse Educ Today. 1992;12(5):368–75.
27. Rice ME, Helzel MF, Varney GW, Quinsey VL. Crisis prevention and intervention training for psychiatric
hospital staff. Am J Community Psychol. 1985;13(3):289–304.
2. Wondrak R, Dolan B. Dealing with verbal abuse: evaluation of the ecacy of a workshop for student
nurses. Nurse Educ Today. 1992;12(2):108–15.
29. Miltenberger RG, Flessner C, Gatheridge B, Johnson B, Satterlund M, Egemo K. Evaluation of
behavioral skills training to prevent gun play in children. J Appl Bahavior Anal. 2004;37:513–6.
30. Parsons MB, Rollyson JH, Reid DH. Evidence-based staff training: A guide for practitioners. Behav
Anal Pract. 2012;5(2):2–11.
31. Baer DM, Wolf MM, Risley TR. Some still‐current dimensions of applied behavior analysis. Journal of
Applied Behavior Analysis. J Appl Behav Anal. 1987;20(4):313–27.
32. Dillenburger K. Staff Training. Handbook of Treatments for Autism Spectrum Disorder. In: Matson JL,
editor. Handbook of Treatments for Autism Spectrum Disorder. Springer; 2017. p. 95–107.
33. Kirkpatrick M, Akers J, Rivera G. Use of Behavioral Skills Training with Teachers: A Systematic
Review. J Behav Educ. 2019;28(3):344–61.
34. Sun X. Behavior skills training for family caregivers of people with intellectual or developmental
disabilities: a systematic review of literature. Int J Dev Disabil. 2020;Jul 21:1–27.
35. Davis S, Thomson K, Magnacca C. Evaluation of a caregiver training program to teach help-seeking
behavior to children with autism spectrum disorder. Int J Dev Disailities. 2020;66(5):348–57.
3. Gormley L, Healy O, O’Sullivan B, O’Regan D, Grey I, Bracken M. The impact of behavioural skills
training on the knowledge, skills and well‐being of front line staff in the intellectual disability sector:
a clustered randomised control trial. J Intellect Disabil Res. 2019;63(11):1291–304.
Page 16/21
37. Lin E, Malhas M, Bratsalis E, Thomson K, Boateng R, Hargreaves F, et al. Behavioural skills training
for teaching safety skills to mental health clinicians: A protocol for a pragmatic randomized control
trial. JMIR Res Protoc. 2022;11(12):e39672.
3. GraphPad. Randomly assign subjects to treatment groups [Internet]. [cited 2023 Jun 6]. Available
from: https://www.graphpad.com/quickcalcs/randomize1.cfm
39. Van Voorhis CRW, Morgan BL. Understanding Power and Rules of Thumb for Determining Sample
Sizes. Tutor Quant Methods Psychol. 2007;3(2):43–50.
40. Erath TG, DiGennaro Reed FD, Sundermeyer HW, Brand D, Novak MD, Harbison MJ, et al. Enhancing
the training integrity of human service staff using pyramidal behavioral skills training. J Appl Behav
Anal. 2020 Jan 1;53(1):449–64.
41. Pitts L, Hoerger ML. Mastery criteria and the maintenance of skills in children with developmental
disabilities. Behav Interv [Internet]. 2021 Apr 25 [cited 2022 Oct 17];36(2):522–31. Available from:
https://onlinelibrary.wiley.com/doi/10.1002/bin.1778
42. Richling SM, Williams WL, Carr JE. The effects of different mastery criteria on the skill maintenance
of children with developmental disabilities. J Appl Behav Anal [Internet]. 2019 Jun 1 [cited 2022 Oct
14];52(3):701–17. Available from: https://pubmed.ncbi.nlm.nih.gov/31155708/
43. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R
Foundation for Statistical Computing; 2021.
44. Bosker R, Snijders TA. Multilevel analysis: An introduction to basic and advanced multilevel
modeling. In: Multilevel Analysis. London, UK: Sage Publishers; 2012. p. 1–368.
45. Pinheiro J, Bates D. Mixed-effects models in S and S-PLUS. Springer Science+Business Media; 2006.
4. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Lawrence
Erlbaum Associates; 1988.
47. Alavosius MP, Sulzer-Azaroff B. The effects of performance feedback on the safety of client lifting
and transfer. J Appl Behav Anal. 1986;19:261–7.
4. Hogan A, Knez N, Kahng S. Evaluating the use of behavioral skills training to improve school staffs’
implementation of behavior intervention plans. J Behav Educ. 2015;24(2):242–54.
49. Nabeyama B, Sturmey P. Using behavioral skills training to promote safe and correct staff guarding
and ambulation distance of students with multiple physical disabilities. J Appl Behav Anal.
2010;43(2):341–5.
50. Parsons MB, Rollyson JH, Reid DH. Teaching Practitioners to Conduct Behavioral Skills Training: A
Pyramidal Approach for Training Multiple Human Service Staff. Behav Anal Pract. 2013;6(2):4–16.
51. Sarakoff RA, Sturmey P. The effects of behavioral skills training on staff implementation of discrete-
trial teaching. J Appl Behav Anal. 2004;37(4):535–8.
52. Oah G, Ekpotu LP, Murphy S, Kane D, Gordon A, O’Sullivan M, et al. Evaluation of medical student
retention of clinical skills following simulation training. BMC Med Educ. 2019;19:263.
Page 17/21
53. Bruno P, Ongaro A, Fraser I. Long-term retention of material taught and examined in chiropractic
curricula: its relevance to education and clinical practice. J Can Chiropr Assoc. 2007;51(1):14–8.
54. Magnacca C, Thomson K, Marcinkiewicz A, Davis S, Steel L, Lunsky Y, et al. A Telecommunication
Model to Teach Facilitators to Deliver Acceptance and Commitment Training. Behav Anal Pract.
2022;15(3):752–752.
55. Barsuk JH, Cohen ER, Feinglass J, McGaghie WC, Wayne DB. Residents’ procedural experience does
not ensure competence: A research synthesis. J Grad Med Educ. 2017;9(2):201–8.
5. Choudhry NK, Fletcher RH, Soumerai SB. Systematic review: the relationship between clinical
experience and quality of health care. Ann Intern Med. 2005;142(4):260–73.
57. Lu FI, Takahashi SG, Kerr C. Myth or Reality: Self-Assessment Is Central to Effective Curriculum in
Anatomical Pathology Graduate Medical Education. Acad Pathol. 2021;8.
5. Yates N, Gough S, Brazil V. Self-assessment: With all its limitations, why are we still measuring and
teaching it? Lessons from a scoping review. Med Teach [Internet]. 2022;44(11):1296–302. Available
from: https://doi.org/10.1080/0142159X.2022.2093704
59. Leach B, Gloinson ER, Sutherland A, Whitmore M. Reviewing the Evidence Base for De-escalation
Training: A Rapid Evidence Assessment [Internet]. RAND Research Reports. RAND Corporation; 2019
Aug [cited 2021 Nov 26]. Available from: https://www.rand.org/pubs/research_reports/RR3148.html
Footnotes
1Predened as 80% or better competence as assessed by observers
Figures
Page 18/21
Figure 1
Mean Observer-rated Competence in Self-Protection and Team-control skills across three time points by
TAU and BST
Page 19/21
Figure 2
Percent of Participants Achieving Mastery[1]in Self-Protection and Team-control skills across three time
points by TAU and BST
Page 20/21
Figure 3
Mean Self-rated Condence in Self-Protection and Team-control skills across three time points by TAU
and BST
Page 21/21
Figure 4
Mean Self-rated Condence and 95-percent condence intervals in Self-protection and Team-control skills
across three time points by Past-month occasion to use skills
Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.
CONSORTchecklistsubmitted.docx
Supplementarymaterials.docx
