An integrative literature review of interventions to reduce violence against emergency department...

REVIEW

An integrative literature review of interventions to reduce violence

against emergency department nurses

Linda Anderson, Mary FitzGerald and Lauretta Luck

Aim. To critique the evidence that underpins interventions intended to minimise workplace violence directed against emergency

department nurses, to inform researchers and policy makers regarding the design, development, implementation and evaluation

of emergency nursing anti-violence and counter-violence interventions.

Background. Workplace violence perpetrated against emergency department nurses is at least continuing and at worst

increasing. Occupational violence has detrimental effects on job satisfaction, retention and recruitment, and the quality and cost

of patient care.

Design. An integrated literature review.

Method. Searches of the Cochrane Library, CINAHL, MEDLINE and the Joanna Briggs Institute between 1986–May 2007.

Included articles were appraised and then synthesised into a narrative summary.

Results. Ten primary research studies were included. Interventions were classified as environmental, practices and policies, or

skills. While each study has useful information regarding the implementation of interventions, there is no strong evidence for

their efficacy.

Conclusions. The weight of effort is still directed towards defining the phenomenon rather than addressing solutions. Studies

that assessed the efficacy of a single intervention failed to take account of context; and participatory context-driven studies

failed to provide generalisable evidence. Concerted multi-site and multi-disciplinary, action-oriented research studies are ur-

gently needed to provide an evidence base for the prevention and mitigation of violence perpetrated against emergency

department nurses.

Relevance to clinical practice. The investigation of interventions rather than repeatedly redefining the problem and directing

resources into debating semantics or differentiating ‘degrees’ of violence and aggression is recommended. This review unam-

biguously identifies the gap in research-based interventions.

Key words: accident and emergency department, aggression, emergency department, intervention, nursing, violence

Accepted for publication: 11 July 2009

Introduction

Workplace violence perpetrated against nurses is undeniably

a global issue (Stirling et al. 2001, Ferns 2005, Merecz et al.

2006) that has significant organisational and individual

sequelae (McPhaul & Lipscomb 2004, Madine & Smith

2005, Royal College of Nursing Australia 2007). Nurses, as

the largest component of the health care workforce (Foley

2004, Catlette 2005), experience the greatest burden of

workplace violence (Fernandes et al. 1999, Peek-Asa et al.

2002, Armstrong 2006); and emergency department nurses

Authors: Linda Anderson, BAppSc (MedTech), BNSc, GDipEd (Sec),

GDipEdStud (PD), GDipAppSci (Computing), MRCNA, Registered

Nurse, 52 St Albans Rd, Mount Louisa, QLD; Mary FitzGerald, RN,

DN, CertEd, MN, PhD, FRCNA, Professor of Nursing Practice

Development, Royal Hobart Hospital, School of Nursing and

Midwifery, University of Tasmania, Tasmania; Lauretta Luck, RN,

BA, MA, PhD, MRCNA, Associate Head of School and Senior

Lecturer, School of Nursing & Midwifery, College of Health &

Science, University of Western Sydney, Penrith South, DC, Australia

Correspondence: Linda Anderson, 52 St Albans Rd, Mount Louisa

QLD 4814, Australia. Telephone: 61 7 4774 8019.

E-mail: [email protected]

2520 � 2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2520–2530

doi: 10.1111/j.1365-2702.2009.03144.x

(EDNs) are one of the highest-risk nursing specialties

(Catlette 2005, Gates et al. 2006, Lau & Magarey 2006,

Ryan & Maguire 2006, International Council of Nurses

2007). A broad and inclusive definition of violence by

Perrone (1999) is used in this review: a spectrum of

behaviours ranging from passive aggression to homicide.

Considerable research effort is devoted to describing and

codifying the workplace violence phenomenon. The Univer-

sity of Iowa Injury Prevention Research Centre (UIIPRC)

developed a classification system to assist the understanding

of and to target interventions against violent events (Univer-

sity of Iowa Injury Prevention Research Centre 2001).

A summary of the classification is:

• Type I criminal intent – perpetrator has no legitimate

relationship to the workplace, often committing a crime in

conjunction with the violence.

• Type II customer/client – perpetrator becomes violent while

being served by the workplace.

• Type III worker-on-worker – perpetrator is a current or

past employee of the workplace.

• Type IV personal relationship – perpetrator does not have a

relationship with the workplace but has a personal rela-

tionship with the victim.

While nurses may experience any or all of the UIIPRC

workplace violence classes, the most common is type II,

where the perpetrators of violent acts are customers (patients,

their relatives and/or friends) who become violent while being

served by workers (nurses) (Gerberich et al. 2004, McPhaul

& Lipscomb 2004, Farrell et al. 2006). Type II violence is the

focus of this review. Additional to physical or psychological

injuries arising from type II workplace violence, suffering or

witnessing verbal or physical threats or actions erodes well-

being and job satisfaction (Fernandes et al. 1999) and

hampers nursing recruitment and retention (Armstrong

2006, Chapman & Styles 2006).

Violence and aggression reduce the quality while increasing

the cost of patient care (Arnetz & Arnetz 2001, Lau et al.

2004, Farrell et al. 2006, International Council of Nurses

2007). Despite the plethora of literature describing workplace

violence and recommending interventions to prevent or

alleviate it, research evaluating these remedies is scant (Deans

2004, McPhaul & Lipscomb 2004, Catlette 2005), the

success of interventions arbitrary (Lau et al. 2004, Luck

et al. 2007) and research programmes uncoordinated.

Four literature reviews addressing violence and aggression

in the emergency department were accessed. Stirling et al.

(2001) and Lau et al. (2004) broadly identify contemporary

research. Doughty (2005) more specifically investigates

research related to training, and Wand and Coulson (2006)

consider zero tolerance as an intervention. These reviews

confirm there is little evidence that evaluates the feasibility,

appropriateness, meaningfulness and effectiveness (FAME)

(Pearson et al. 2007) of the interventions recommended to

reduce or at least mitigate workplace violence and aggression,

despite years of effort. Moreover, the literature is dominated

by works referring to violence in psychiatric settings and

people with mental illnesses. Thus, it is prudent to undertake

an integrative review of the evidence that evaluates the

interventions specifically intended to reduce violence against

EDNs. The resulting synthesis will help transfer extant

research into the workplace and more accurately target

future research.

Aim

The aim of this integrative review is to critique the evidence

that underpins interventions intended to minimise workplace

violence perpetrated against EDNs to inform researchers and

policy makers regarding the design, development, implemen-

tation and evaluation of emergency nursing anti-violence and

counter-violence interventions.

Method

A comprehensive search strategy was developed to identify

both published and unpublished works, written in English,

between 1986–May 2007. An initial limited search of

CINAHL and MEDLINE using the following intuitive key

words was undertaken to distil suitable medical subject

headings (MeSH) vocabulary and other relevant terms:

• nurse, nurses, nursing;

• emergencydepartment, accidentandemergencydepartment;

• violent, violence;

• aggression, aggressive; and

• abuse.

To ensure that the search results were comprehensive,

multiple electronic databases were interrogated (Nicholson

2007). Identified key words and appropriate MeSH termi-

nology were selectively applied for the following:

• the Cochrane Library;

• CINAHL;

• MEDLINE; and

• the Joanna Briggs Institute (JBI).

The search strategies for the Cochrane Library, CINAHL

and MEDLINE are provided in the full report (Anderson

2007). To avoid publication bias, grey literature was sourced

via the subsequent specialised databases.

• ISI Current Contents;

• First Search; and

• Digital Dissertations.

Review Violence against emergency nurses

� 2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2520–2530 2521

Ten specialist journals were hand searched for the past

10 years and ancestry searching, appraising reference lists

from articles, was undertaken. The CINAHL and MEDLINE

searches were undertaken for a second time (September 2007)

to check for any recently published materials not available for

the first search run (May 2007).

Inclusion criteria were any research study considering:

• type II workplace violence only,

• investigating the FAME of interventions (intended to

reduce the frequency and impact of type II violence),

• set in an emergency department, with

• nurses or emergency department clientele as subjects/par-

ticipants.

Exclusion criteria for studies inappropriate to this review

were as follows:

• types I, III or IV workplace violence, set

• outside an emergency department, where

• nurses or their clientele were not the primary subjects/

participants.

Two researchers working independently appraised re-

trieved articles. The JBI appraisal tool for observational

studies supported this process (Pearson et al. 2007). The

included articles were rated according to the JBI levels of

evidence on FAME (http://www.joannabriggs.edu.au/pubs/

approach.php, retrieved 1 September 2008).

Results

One hundred and three articles were identified from the

searches of MEDLINE and CINAHL. Grey literature

searches identified one relevant retrievable thesis. Despite a

comprehensive and systematic search, there were few

primary research studies related to the evaluation of inter-

ventions intended to minimise violence and aggression for

EDNs.

After independent review by two researchers, 82 of the

articles were rejected, as their titles or abstracts described

contexts other than emergency department nursing or failed

to identify and/or evaluate interventions. Twenty-one docu-

ments were subsequently called for appraisal and retrieved as

full text. Of these, seven were later excluded as their abstracts

were misleading, and examination revealed that interventions

were not evaluated or the intent of the research was other

than the reduction in violence against EDNs (Anderson

2007).

Of the remaining 14 articles, 10 were primary research

studies and the other four were reviews. All the included

primary studies were described by their authors as quantita-

tive studies. All articles were designated level IV, according to

the JBI levels of evidence (http://www.joannabriggs.edu.au/

pubs/approach.php, retrieved 2 September 2008), equivalent

to expert opinion. Given the inclusive nature of this integra-

tive review, it was deemed important to continue the process

despite the low evidence rating and to comment in a narrative

form on appropriateness, feasibility and meaning of inter-

ventions tested in the studies.

A summary of the design, level of evidence, sample,

characteristics of the intervention and results for each of the

10 included primary studies is provided in Tables 1–3.

The articles are organised chronologically and classified into

the following three groups, based on the nature of the

intervention they evaluate (American Nurses Association

2002, Emergency Nurses Association 2006):

• workplace environment,

• workplace practices and policies and

• individual and collective skill sets.

Table 4 summarises the demographical features of the

studies, demonstrating the weight of global research effort

and reflecting the national, cultural and healthcare system

contexts that influence the transferability of the research.

Workplace environment

Three studies dealt with interventions influencing the emer-

gency department environment with the intent of changing

human behaviour (see Table 1). All involved the use of metal

detectors, as recommended by the American Medical Asso-

ciation and the American Psychiatric Association. Concerns

about metal detectors creating a poor impression with the

public and fear that metal detectors could provoke anxiety

among emergency department consumers led to mixed

feelings about their suitability (McNamara et al. 1997,

Meyer et al. 1997).

Meyer et al. (1997) and McNamara et al.’s (1997) research

investigating attitudes to metal detectors provide insights into

the applicability of perception-management strategies that

are meaningful when considered in a crisis-management

context. Some characterisations of a crisis comprise a trilogy

definition: the precipitating event, participants’ perceptions

and subjective distress (Kanel 2003). As the presence of metal

detectors appears to increase staff and clientele perceptions of

safety, this should be viewed as a positive. Neither

McNamara et al.’s (1997) nor Meyer et al.’s (1997) research

measured events, so the actual deterrent effect of the metal

detectors was not quantified.

Rankins and Hendey’s (1999) conclusion that other

interventions are required to make the emergency department

safer is not unexpected. Given the fewer weapon-based

assaults compared to verbal assaults (Presley & Robinson

2002), it is not surprising that metal detector use caused

L Anderson et al.


insignificant reductions in assault frequency, although it may

reduce assault severity (Keely 2002).

McNamara et al. (1997) and Meyer et al. (1997) address

feasibility in terms of a need to site metal detectors at all

points of entry (or restrict entry, with the commensurate

workflow consequences) and provide workable contingency

planning for non-ambulant patients brought directly into the

emergency department without security screening. Rankins

and Hendey’s (1999) conclusions that interventions addi-

tional to metal detectors were required to make the

emergency departments safer provoked insights into the

feasibility, or otherwise, of the metal detectors.

McNamara et al. (1997) recognise that equipment inter-

ventions require funding for equipment purchases and the

training liability to operate, respond to and report on the

equipment. They also identify the need to institute a system

for confiscating discovered weapons. Through-life costs are

not discussed in any detail.

McNamara et al.’s (1997) and Meyer et al.’s (1997)

research into attitudes touches on the appropriateness

of metal detectors. Although Meyer et al.’s (1997) and

McNamara et al.’s (1997) investigations are limited, ask

different questions and are undertaken in different socio-

economic contexts, they suggest that the presence of metal

detectors does not deter most people from coming to

the emergency department nor do they create a climate of

fear. Moreover, they make most staff and consumers feel

safer.

Table 1 Primary studies – environmental factors

Primary study,

country

Design, level of

evidence, sample

Characteristics of

intervention Results

Environmental factors (intended to change the environment to facilitate change in the behaviour of patients, family and friends, and nurses)

McNamara

et al. (1997)

United States of

America

Survey interview

Level IV

n = 303

142 patients (47%)

142 family (47%)

19 friends (6%)

Metal Detector (MetD). Sought

public opinion about perceptions

of safety and willingness to use

the ED if a MetD was employed.

Aimed to determine whether the

perception that MetD created bad

public relations was accurate

75% felt safe

68% satisfied with security

66% would feel better with MetD

11% worried that they may be physically harmed in ED

10% less likely to use ED with MetD

Meyer et al.

(1997)

United States of

America

‘Descriptive

observational

study’

Level IV

n = 271 (‡18 years

old)

62 patients

114 family or friends

30 doctors

35 nurses

16 security officers

14 other ED staff

Metal detector (MetD). Investi

gated attitudes towards the use of

a Metor 120 metal detector,

costing approximately $4500

USD, installed at ED entrance

under 24-hour surveillance by a

security guard

80% patrons and 85% employees liked MetD

90% patrons and 73% employees felt safer

12% patrons and 10% employees felt their privacy

was invaded

Patrons: 60% said MetD would not affect return to ED:

39% more likely and 1% less likely to return

Employees: 81% said MetD would not influence patient

decision to return: 11% more likely and 8% less likely

No significant difference in opinion was identified based

on age, sex or race

Rankins and

Hendey (1999)

United States of

America

Retrospective Chart

review

Level IV

Number of months

reviewed: Pre – 29;

Post – 25

Number of patients

treated:

Pre – 155 976;

Post – 108 994

No direct intervention applied.

Investigated weapon confiscation

(WPN) and assaults (ASLT)

before and after security system

installation from 1992–1996.

Aimed to determine the number

and frequency of WPN

confiscations and ASLT in the ED

before and after the

implementation of a security

system. Compared triage or

security (TorS) and patient care

areas (PCa)

WPN confiscation (actual &/10 000):

Pre: TorS = 2 [0Æ1]; PCa = 22 [1Æ4]; Total = 24 [1Æ5];

%before PCa = 8

Post: TorS = 23 [2Æ1]; PCa = 17 [1Æ6]; Total = 40 [3Æ7];

%before PCa = 58

7/17 WPN in PCa were from patients arriving by

ambulance who bypassed security booth and fixed MetD

ASLTs (actual & /10 000):

Pre: ED = 73 [4Æ7]; WPN = 5 [0Æ3]

Post: ED = 65 [5Æ8]; WPN = 1 [0Æ1]

WPN confiscations before entering PCa increased more

than tenfold with the security system. The total number of

ASLT was not affected by the security, indicating that

other interventions are required to make the ED a safer

place to work



Practices and policies

Two studies dealt with interventions influencing organisa-

tional practice with the intent of modifying human behaviour

(Table 2). One involves legislative change – the introduction

of AB508 – 1993 Hospital Security Act, which compels

hospitals to create systems and processes that increase

hospital physical security measures. The second concerns

practice transformation – the use of an improved reporting

form; processes that support the use of the form, data and

information it generates; and changes in systems that provide

feedback to the personnel who report the incidents of

violence and aggression.

Gray’s (2006) intervention addressed more than the

effectiveness domain, although only effectiveness was

reported. The ongoing modification of the Violence and

Aggression Incidence Report (VAIR) form throughout the

study occurred in response to nurses’ unwillingness to

undertake reporting, because the form was not sufficiently

user friendly given the time constraints of the busy emergency

department. The ongoing improvement to the form is an

example of the feasibility dimension of the intervention being

actively investigated and enhanced. Instituting policy that

mandated incident investigations and subsequent feedback to

nurses also made completing the VAIR more meaningful for

the nursing staff. The concurrent changes to the systems via

Table 2 Primary studies – practices and policy

Primary study,

country

Design, level of

evidence, sample

Characteristics of

intervention Results

Practices and policy (intended to facilitate practice that will change the behaviour of patients, family and friends, and nurses)

Peek-Asa et al.

(2002)

United States of

America

‘Quasi-experimental

non-equivalent

control group

before–after design’

Level IV

Pre (1990):

103/not specified

Post (2000):

95/180 (53Æ8%)

No direct intervention

applied. Investigated

effects via survey. Aimed

to identify changes in

security programmes and

violent events following

the passage of the 1993

Hospital Security Act

AB508

Increased reports of patients with knives (p < 0Æ001).

There were no significant changes in the nature of threats

(verbal, physical unarmed and armed) and frequency of

violent events per month between surveys. Injuries to visitors,

as consequences of violence, decreased significantly. No

statistically significant differences in the time lost to staff

injury because of violence. Insignificant increase in the staff

injuries sustained and time lost

The number of hospitals reporting violent events related to

alcohol, drugs, anger/high stress/illness, waiting times and

access control problems all decreased. Alcohol related events

decreased significantly 88Æ4–75Æ8 (p = 0Æ02) and access

control decreased from 40Æ08–11Æ6% (p < 0Æ01)

There were increases in employee training (34Æ0 – 95Æ6%,

p < 0Æ01) and policies on violence (66Æ0–77Æ5%, p = 0Æ07).

Protective measures increased overall (54Æ2–88Æ8%,

p < 0Æ01) with commensurate increases in hospital entrance

security (49Æ0 – 95Æ6%, p < 0Æ01) and surveillance cameras

(26Æ0 – 69Æ5%, p < 0Æ01). Searching (10Æ7–25Æ3%,

p < 0Æ01), secluding (27Æ2–32Æ6%) and having security on

standby (67Æ0–81Æ1, p < 0Æ05) for aggressive patients also

increased

While respondents’ perceptions of security officer adequacy

improved (23Æ8–33Æ7%), more than 66% felt them inade

quately prepared for their job

Gray (2006)

Australia

‘Non-experimental

one group pre-test

post-test design’

Level IV

Violence and Aggression

Incident Report (VAIR),

ED Critical Incident

Committee, monthly feed

back to ED staff and as

required education sessions

regarding correct usage of the

form were undertaken to

increase the report rate for

incidents of violence and

aggression in the ED

Incident reporting for Jul–Sep was compared for 2003 and

2004. When VAIR was in place, incident reporting increased

between 25–270% (per month)

Incident reporting for 2004 and 2005, comparing VAIR and

CB, showed improvement in VAIR: �30% in 2004 & �50%

in 2005. Overall improvement between 2003–2005 was

�100%

ED presentations increased from 2003–2005, �8% pa, with

a total increase in the period of 17Æ22%

VAIR per 1000 ED presentations increased from 6Æ5 in 2003–

10Æ9 in 2005

L Anderson et al.


Table

3Pri

mary

studie

s–

indiv

idual

and

collec

tive

skil

lse

ts

Pri

mary

study,

countr

yD

esig

n,

level

of

evid

ence

,sa

mple

Chara

cter

isti

csof

inte

rven

tion

Res

ult

s

Indiv

idual

and

collec

tive

skill

sets

(inte

nded

toch

ange

the

beh

avio

ur

of

eith

erper

pet

rato

rsor

vic

tim

s)

Lee

(2001)

New

Zea

land

‘Cro

ss-s

ecti

onal

mix

ed-m

odel

obse

rvat

ional

surv

ey’

Lev

elIV

A&

Edep

art

men

tnurs

es:

76/1

30

(58%

)re

sponded

12/1

8m

ale

(67%

)

64/1

12

fem

ale

(57%

)

No

dir

ect

inte

rven

tion

applied

.

Inves

tigate

def

fect

svia

surv

ey.A

imed

toex

plo

reth

eef

fect

of

aggre

ssio

n

managem

ent

train

ing

(AM

T)

and

exposu

reto

vio

lence

on

nurs

es’

self

-effi

cacy

per

cepti

ons

when

dea

ling

wit

hpati

ent

aggre

ssio

n

No

sign

ifica

nt

dif

fere

nce

inm

easu

res

of

exper

ience

:nurs

esw

ere

quali

fied

over

12

yea

rs[t

(71)

=0Æ0

1,

p>

0Æ1

]w

ith

more

than

6yea

rs’

A&

Eex

per

ience

[t(7

3)

=0Æ5

7,

p>

0Æ1

]and

over

5w

eeks’

psy

chia

tric

exper

ience

duri

ng

train

ing

[t(7

2)

=0Æ6

7,

p>

0Æ1

]

Whil

e21%

had

not

exper

ience

dphysi

cal

vio

lence

,only

4%

had

not

exper

ience

dver

bal

aggre

ssio

n

The

sam

ple

’sm

ean

self

-effi

cacy

was

‘ave

rage’

[mea

n=

20Æ3

8,

sd=

5Æ5

0,

range

=5–34]

Hig

her

self

-effi

cacy

was

ass

oci

ate

dw

ith

rece

nt

(3m

onth

s)ex

per

ience

of

ver

bal

aggre

ssio

n[t

=2Æ7

7,

df

=74,

p<

0Æ0

1]

and

bei

ng

of

managem

ent

gra

de

[t=

3Æ0

8,

df

=69,

p<

0Æ0

1]

Fer

nandes

etal

.(2

002)

Canad

a

‘Cro

ss-s

ecti

onal

pro

spec

tive

surv

ey’

Lev

elIV

Inner

-cit

yhosp

ital

wit

h55

000

ED

pati

ent

vis

its

per

annum

Six

hundre

dand

sixty

-sev

en(8

4%

)of

798

surv

eys

com

ple

ted

by

ED

hea

lth

care

work

ers

(excl

udin

gpro

tect

ive

serv

ice

staff

)

Nurs

esre

pre

sente

d56,55

and

66%

of

the

part

icip

ants

per

tim

efr

am

e

The

Pre

venti

on

and

Managem

ent

of

Aggre

ssiv

eB

ehavio

ur

Pro

gra

mm

e

(PM

AB

P)

–a

4-h

our

pro

gra

mm

e

base

don

the

Non-v

iole

nt

[sic

]C

risi

s

Inte

rven

tion

model

–in

tended

to

faci

lita

teacq

uis

itio

nof

skil

lsin

ass

essi

ng

and

pre

venti

ng

aggre

ssiv

e

beh

avio

ur

toin

crea

seca

re,sa

fety

and

secu

rity

of

clie

nts

and

staff

Rate

sof

all

types

of

vio

lent

even

ts(e

xper

ience

dand

wit

nes

sed,ver

bal

and

physi

cal)

dec

rease

dfr

om

base

line

toper

iod

2and

incr

ease

dfr

om

per

iod

2to

per

iod

3

Per

cepti

ons

of

safe

tyduri

ng

vio

lent

even

tsw

ere

report

edby

621

surv

eys.

Thes

eper

cepti

ons

wer

em

ost

lyaggre

gate

dacr

oss

all

thre

e

per

iods

Lev

els

of

physi

cal

vio

lence

dec

rease

din

per

iod

2but

incr

ease

din

per

iod

3.W

hil

eth

ein

crea

sein

physi

calvio

lence

was

not

sign

ifica

ntl

y

dif

fere

nt

than

base

line,

the

incr

ease

dver

bal

vio

lence

rem

ain

ed

sign

ifica

ntl

ylo

wer

than

base

line

level

s

Des

pit

eth

ein

crea

sed

vio

lent

even

tsfr

om

per

iods

2–3,

staff

per

ceiv

ed

the

ED

tobe

safe

rin

per

iod

3co

mpare

dto

base

line

[(O

R)

2Æ7

8;95%

CI

1Æ6

9–4Æ5

5;

p=

0Æ0

001]

than

inper

iod

2co

mpare

dto

base

line

[(O

R)

1Æ4

1;

95%

CI

0Æ8

8–2Æ2

6;

p=

0Æ1

5]

Cow

in

etal

.(2

003)

Aust

rali

a

‘Quasi

-exper

imen

tal

non-e

quiv

ale

nt

contr

ol

gro

up

bef

ore

–aft

erdes

ign’

Lev

elIV

Bef

ore

:21

MH

Unurs

es33

ED

nurs

es

Aft

er:

19

MH

Unurs

es30

ED

nurs

es

De-

esca

lati

on

Kit

:co

mpri

sing

a

post

er,

nurs

ing

staff

surv

eyto

test

pre

know

ledge

and

post

know

ledge

and

aw

are

nes

sof

de-

esca

lati

on,

a

case

-stu

dy

support

edin

-ser

vic

e

educa

tion

sess

ion

and

a

lite

ratu

re-b

ase

ddis

cuss

ion

art

icle

Only

resu

lts

for

the

ED

nurs

esare

sum

mari

sed

her

e

Anum

eric

albut

not

stati

stic

alin

crea

sew

asobta

ined

inth

eto

talm

ean

score

from

T1

(41Æ1

8)–

T2

(42Æ4

3).

The

auth

ors

state

this

sugg

ests

som

eposi

tive

effe

ctfr

om

the

inte

rven

tion

How

ever

,in

crea

ses

wer

enot

consi

sten

tth

rough

out

all

surv

eyit

ems,

sugg

esti

ng

oth

erfa

ctors

may

aff

ect

de-

esca

lati

on

know

ledge

and

aw

are

nes

s.T

he

mea

nsc

ore

was

hig

her

(by

as

much

as

0Æ7

5to

as

litt

le

as

0Æ0

4)

infive

of

the

11

item

s,but

low

er(b

yas

much

as

0Æ4

2and

as

litt

leas

0Æ0

2)

inth

ere

main

ing

six

item

s



Table

3(C

onti

nued

)

Pri

mary

study,

countr

yD

esig

n,

level

of

evid

ence

,sa

mple

Chara

cter

isti

csof

inte

rven

tion

Res

ult

s

Wes

t(2

003)

Unit

edSta

tes

of

Am

eric

a

‘Quasi

-exper

imen

tal

indep

enden

t

mea

sure

s’

Lev

elIV

30

000

ED

vis

its/

yea

r

Conven

ience

sam

ple

:

114

contr

ol

and

86

inte

rven

tion

Info

rmati

on

pam

phle

tdis

trib

ute

dto

pati

ents

and

vis

itors

at

regis

trati

on

–

aft

ertr

iage

but

bef

ore

trea

tmen

t.

Pam

phle

tcl

ari

fied

ED

purp

ose

and

pri

nci

ple

sof

tria

ge,

regis

trati

on,

wait

ing,

trea

tmen

topti

ons

and

dia

gnost

icte

sts.

Over

tA

ggre

ssio

n

Sca

le(O

AS)

use

dto

captu

repati

ents

’

and

vis

itors

’dis

pla

ysof

ver

bal

and

physi

cal

aggre

ssio

n.

OA

Ssc

ore

s

incr

ease

wit

hse

ver

ity

of

aggre

ssiv

e

dis

pla

ys,

from

0–21.

Tota

lsc

ore

reflec

tscu

mula

tive

exhib

ited

beh

avio

urs

Fre

quen

cydata

:

Contr

ol:

(114)

95

no

aggre

ssio

n;

19

som

edis

pla

ys

Inte

rven

tion:

(86)

83

no

aggre

ssio

n;

thre

eso

me

dis

pla

ys

Tre

atm

ent

gro

up

had

few

erdis

pla

ys

of

aggre

ssio

n(M

=3Æ5

)th

an

contr

ol

gro

up

(M=

16Æ7

),re

flec

ting

a79Æ1

%re

duct

ion

inaggre

ssiv

e

inci

den

ts

Sev

erit

ydata

:(O

AS

score

,n)

Contr

ol:

(n=

19)

(1Æ0

0,

7;

2Æ0

0,

5;

3Æ0

0,

6;

5Æ0

0,

1)

Inte

rven

tion:

(n=

3)

(1Æ0

0,

0;

2Æ0

0,

2;

3Æ0

0,

1;

5Æ0

0,

0)

Inte

rven

tion

gro

up

had

alo

wer

mea

nO

AS

score

(M=

0Æ0

81)

than

the

contr

ol

gro

up

(M=

0Æ3

51),

repre

senti

ng

am

ean

reduct

ion

of

81Æ3

%

F-t

est

(F=

4Æ2

10)

(p<

0Æ0

1&

crit

ical

one-

tail

F=

1Æ6

211)

sugg

ests

that

the

vari

ance

s

bet

wee

nth

egro

ups

dif

fer

sign

ifica

ntl

y

t-te

st(t

=2Æ7

81,

p<

0Æ0

05)

(p=

0Æ0

03,

one-

tail

T-v

alue

=2Æ6

045)

sugg

ests

that

the

OA

Ssc

ore

mea

ns

are

signifi

cantl

ydif

fere

nt

Larg

eH

art

ley’s

F-m

ax

(4Æ2

)su

gges

tspopula

tion

het

erogen

eity

Dea

ns

(2004)

Aust

rali

a

‘Non-e

xper

imen

tal

one

gro

up

pre

-tes

t

post

-tes

tdes

ign’

Lev

elIV

Fort

yof

60

ED

nurs

es(6

6%

)

under

wen

ttr

ain

ing

Ques

tionnai

redata

wer

eco

llec

ted

as

foll

ow

s:

Pre

:30

(75%

)

Post

:22

(55%

)

Aone-

day

train

ing

pro

gra

mm

e

inte

nded

to:

incr

ease

aw

are

nes

sof

work

envir

onm

ent,

coll

eagu

es’

stre

ngt

hs

and

wea

knes

ses,

and

fact

ors

that

influen

ceef

fect

ive

com

munic

ati

on;in

crea

se

under

standin

gof

types

of

beh

avio

ur

that

trig

ger

are

act

ion,ca

use

sand

types

of

aggre

ssio

n,appro

pri

ate

resp

onse

sand

opti

ons;

and

have

part

icip

ants

dem

onst

rate

effe

ctiv

e

avoid

ance

and

defl

ecti

on,and

secu

re

and

esco

rtte

chniq

ues

The

resu

lts

show

stati

stic

all

ysi

gnifi

cant

effe

cts

of

the

train

ing

for

ass

isti

ng

part

icip

ants

’know

ledge

and

under

standin

gof

the

code

of

pra

ctic

efo

rm

anagin

gaggre

ssiv

eE

Dsi

tuati

ons;

act

ual

know

ledge

of

the

ED

code

of

pra

ctic

e;aw

are

nes

sof

the

const

rain

tsth

at

physi

cal

lim

itati

ons

have

on

ow

nabil

ity

tore

spond

toaggre

ssio

n;and

makin

g

oth

erst

aff

aw

are

of

ow

nphysi

cal

lim

itati

ons

The

resu

lts

wer

enot

stati

stic

ally

sign

ifica

nt

for

team

resp

onse

sto

aggre

ssio

nor

duty

of

care

ques

tions

Vio

lence

inci

den

cew

as

sought

pre

inte

rven

tion

and

post

inte

rven

tion,

yet

only

pre

test

data

ispubli

shed

indet

ail

The

inci

den

cepost

-tes

tin

ciden

cem

ean

and

standard

are

publi

shed

wit

hout

any

of

the

det

ail

pro

vid

edfo

rth

epre

test

data

.R

emark

ably

,

the

auth

or

claim

sth

at

the

inci

den

ceof

aggre

ssio

nhas

bee

nhalv

edin

the

study

and

att

ribute

sth

isto

the

train

ing

inte

rven

tion

L Anderson et al.


which the data is acquired, managed and disseminated reflect

that her work is more than a quantitative study. This

consultative and iterative approach empowered the nurse

participants and shares characteristics of participatory action

research (Polit & Beck 2006).

While Gray (2006) undersells the significance of the other

elements of the modified procedure, apart from the revised

VAIR, her work to increase reporting has other implications.

Better reporting provides more comprehensive incidence

data, which informs other departmental policies, nurse

education and training, and future research objectives. Thus,

the VAIR indirectly helps to reduce workplace violence

towards the EDN.

Peek-Asa et al. (2002) recommended ongoing research to

investigate the cost effectiveness of hospital security

arrangements, where costs are indicators of feasibility.

Research that identifies the return on investment and cost

benefits analysis of changes to organisational policy and

legislative initiatives could be used to provide continued, or

additional, funding. Plus, it can shore up political support

as well as meeting occupational health and safety require-

ments while keeping employees safe and motivated,

encouraging recruitment and retention, and improving

patient care.

Individual and collective skill sets

Five studies addressed interventions intended to improve skill

sets (Table 3). Four dealt with staff skills and, while

occasionally involving teamwork, they were primarily

directed at individual training in managing aggression. The

fifth study targeted clientele – aiming to increase their

understanding of how the emergency department operates,

making consumers better informed and less inclined to

become frustrated with, or act violently because of,

functional constraints inherent in emergency department

services.

To varying degrees, the nature of the training conducted

in each of the individual skill set studies was based on

extant research that investigated training needs. In terms of

training, evaluating meaningfulness of the nature, design

and intended outcomes as perceived by the targets – nurses

– is warranted. Adults learn best when they sense that

training is immediately valuable or relevant to them

(Knowles et al. 1998) and are better motivated when they

recognise that learning addresses their needs not just their

employer’s agenda (Cross 2004). Matching nurses’ under-

standing of and preferences for learning with organisational

imperatives serves to give better return on investment from

finite training funds while meeting the professional devel-

opment and personal satisfaction requirements of the

employees (Cross 2004).

Measuring training effectiveness is always problematic;

however, much of the difficulty derives from poor project

design that fails to identify measures of effectiveness from the

outset. Research that investigates how to measure training

effectiveness in an emergency department context would be a

useful step forward.

Five of the 10 primary studies in this review dealt with

learning interventions, yet their aggregated recommendations

are weak or dubious, given the design and execution

limitations of several studies. The pre-eminence of training

as a strategy could be coincidental or it could reflect the

common perception that the simplest solution to most

problems is training. Based on this collection of research,

even if it is assumed that the gap could be remedied by

training and the delivered training was appropriate, the

effects of that training were unconvincing.

Conclusion

The local, cultural and healthcare system specificity, the

design limitations and the small scale of the 10 included

primary articles support few firm conclusions that could

Table 4 Study demographicsNature of intervention

Population*Environment

Policy and

practice

Skills and

training Total

Country of origin

Australia 1 2 3 21

Canada 1 1 33

New Zealand 1 1 4

USA 3 1 1 5 307

Total 3 2 5 10

*Population stated to the nearest million retrieved from https://www.cia.gov/library/publica

tions/the-world-factbook/index.html (accessed 28 March 2009).



confidently be used to direct clinicians. They do, however,

hold some value as expert opinion in informing other

research and planning interventions.

Because metal detectors aim to remove a hazard from the

workplace, ongoing investigation into their FAME may be

warranted. More quantitative research projects could be

undertaken to measure the effectiveness of the metal detec-

tors; and multiple, small-scale studies with improved and

homogenous designs could be synthesised as a meta-analysis.

Extending the investigations into different contexts, with

increased sample sizes, could discern whether the opinions

from these studies are generalisable to other milieus. This

situation should be investigated qualitatively to gain more

insight into people’s actual behaviours to see whether they

match the opinions gathered in these contexts.

Research into emergency department violence and aggres-

sion cites staff dissatisfaction with organisational and polit-

ical support for their plight (May & Grubbs 2002,

Emergency Nurses Association 2006). Policy and practice

interventions may mitigate the risk of violence and aggression

while concomitantly addressing staff dissatisfaction with the

status quo. Targeting qualitative research to identify the

‘appropriateness’ of interventions may clarify the nursing

workforce’s worth as an occupational resource. Investigating

‘meaningfulness’ of deliverables could shape perceptions of

nurses as human capital. Research quantifying the effects of

policy and practice change initiatives could give real

weight to recommendations. While the included studies

produced local effects, the investigations need to be replicated

in other emergency departments to consider broader

applicability.

Training is commonly seen as a panacea for all organisa-

tional problems; however, material, systems and environ-

mental or cultural difficulties are unlikely to be resolved

solely through training solutions. Before training interven-

tions are initiated, thorough needs analyses to identify the

sources of weaknesses should be undertaken. Problems

should not be assumed to be the result of training shortfalls

(Walsh 2007).

Based on the limitations of evidence revealed, the following

recommendations for practice and research are made:

• Emergency nurses should be encouraged to network and

critically review evidence regarding interventions to reduce

violence in the emergency department.

• A national/international initiative to review and coordinate

research into the reduction and mitigation of violence

against EDNs should be undertaken.

• Outcome measures that are acceptable to the profession

should be developed and applied to evaluate all interven-

tions intended to reduce violence against EDNs.

• Small-scale studies should be encouraged to use participa-

tory methods and concentrate on testing feasibility,

appropriateness and meaning for a local audience. Con-

formity with regard to outcome measures would enable

groups to benchmark and increase transferability of the

results.

• Large-scale studies aimed at proving effectiveness should

be properly funded, well-designed multi-site trials.

Relevance to clinical practice

The quality of the existing research is variable, and it

investigates effectiveness with no specific undertaking to

address the feasibility, appropriateness and meaningfulness of

interventions. The paucity of research evaluating interven-

tions demonstrates that the weight of effort, despite two

decades of investigation, is still directed towards defining the

phenomenon rather than addressing solutions (West 2003,

Lau et al. 2004). Thus, it is appropriate to transition from

definitions and descriptions of emergency department vio-

lence and aggression as a phenomenon to rigorously evalu-

ating the applicability, or otherwise, of interventions.

The way ahead lies in investigating interventions rather

than repeatedly redefining the problem and misdirecting

resources into debating semantics or differentiating ‘degrees’

of violence and aggression. This review unambiguously

identifies the gap in research products – ‘identifying the lack

of evidence’ is ‘an important first step in developing the

evidence base’ (Pearson et al. 2007). It is time to redirect the

research effort, currently scoping the problem and trying to

understand what causes emergency department violence and

aggression, to finding out what does and does not fix it

(Walsh et al. 2006).

Acknowledgements

The original honours thesis from which this review is

derived was supported by a $6000 scholarship from the

School of Nursing, Midwifery and Nutrition, James Cook

University.

Contributions

Study design: LA: 60%, MF: 20%, LL: 20%; data collection

and analysis: LA: 70%, MF: 30% and manuscript prepara-

tion: LA: 70%, MF: 20%, LL: 10%.

Conflict of interest

There is no conflict of interest for this article.

L Anderson et al.


References

American Nurses Association (2002) Preventing Workplace Vio-

lence. American Nursing Association, Silver Spring, MD. Available

at: http://www.nursingworld.org/coeh/wpviolence.htm (accessed

21 March 2007).

Anderson LM (2007) Interventions to Reduce Violence Against

Emergency Department Nurses – An Integrative Review. James

Cook University, Townsville, p. 81.

Armstrong F (2006) Keeping nurses safe: an industry perspective.

Contemporary Nurse 21, 209–211.

Arnetz JE & Arnetz BB (2001) Violence towards health care staff and

possible effects on the quality of patient care. Social Science &

Medicine 52, 417–427.

Catlette M (2005) A descriptive study of the perceptions of work-

place violence and safety strategies of nurses working in level I

trauma centers[sic]. Journal of Emergency Nursing 31, 519–525.

Chapman R & Styles I (2006) An epidemic of abuse and violence: nurse

on the front line. Accident and Emergency Nursing 14, 245–249.

Cowin L, Davies R, Estall G, Berlin T, Fitzgerald M & Hoot S (2003)

De-escalating aggression and violence in the mental health setting.

International Journal of Mental Health Nursing 12, 64–73.

Cross J (2004) Learn@ Work Day Forum. Adult Learning Australia

Inc. Available at: http://www.ala.asn.au/research/index.asp (ac-

cessed 22 October 2007).

Deans C (2004) The effectiveness of a training program for emer-

gency department nurses in managing violent situations. Australian

Journal of Advanced Nursing 21, 17–22.

Doughty C (2005) Staff training programmes for the prevention and

management of violence directed at nurses and other healthcare

workers in mental health services and emergency departments.

NZHTA Technical Brief 4, 1–59.

Emergency Nurses Association (2006). Emergency Nurses Associa-

tion. Available at: http://www.ena.org/about/position/ (accessed 21

March 2007).

Farrell GA, Bobrowski C & Bobrowski P (2006) Scoping workplace

aggression in nursing: findings from an Australian study. Journal of

Advanced Nursing 55, 778–787.

Fernandes CMB, Bouthillette F, Raboud JM, Bullock L, Moore CF,

Christenson JM, Grafstein E, Rae S, Ouellet L, Gillrie C & Way M

(1999) Violence in the emergency department: a survey of health care

workers. Canadian Medical Association Journal 161, 1245–1248.

Fernandes CMB, Raboud JM, Christenson JM, Bouthillette F, Bul-

lock L, Ouellet L & Moore C (2002) The effect of an education

program on violence in the emergency department. Annals of

Emergency Medicine 39, 47–55.

Ferns T (2005) Violence in the accident and emergency department–

an international perspective. Accident and Emergency Nursing 13,

180–185.

Foley M (2004) Caring for those who care: a tribute to nurses and

their safety. Online Journal of Issues in Nursing 9, Manuscript 1.

Available at: http://www.nursingworld.org/MainMenuCategories/

ANAMarketplace/ANAPeriodicals/OJIN/TableofContents/Volume

92004/No3Sept04/TributetoNursesSafety.aspx (accessed 21

March 2007).

Gates DM, Ross CS & McQueen L (2006) Violence against emer-

gency department workers. Journal of Emergency Medicine 31,

331–337.

Gerberich SG, Church TR, McGovern PM, Hansen HE, Nachreiner

NM, Geisser MS, Ryan AD, Mongin SJ & Watt GD (2004) An

epidemiological study of the magnitude and consequences of work

related violence: the Minnesota Nurses’ Study. Occupational and

Environmental Medicine 61, 495–503.

Gray L (2006) Reporting incidents of workplace violence in the

emergency department: how a new reporting form improved staff

compliance in documenting incidents. In School of Nursing and

Midwifery. Curtin University of Technology, Perth, Western Aus-

tralia, p. 65.

International Council of Nurses (2007) Positive Practice Environ-

ments: Quality Workplace = Quality Patient Care (Baumann A

ed.). International Council of Nurses, Geneva.

Kanel K (2003) A Guide To Crisis Intervention, 2nd edn. Brooks

Cole, California.

Keely BR (2002) Recognition and prevention of hospital violence.

Dimensions of Critical Care Nursing 21, 236–241.

Knowles MS, Holton EF & Swanson RA (1998) The Adult Learner.

Gulf Publishing, Houston.

Lau JBC & Magarey J (2006) Review of research methods used to

investigate violence in the emergency department. Accident and

Emergency Nursing 14, 111–116.

Lau JBC, Magarey J & McCutcheon H (2004) Violence in the

emergency department: a literature review. Australian Emergency

Nursing Journal 7, 27–37.

Lee F (2001) Violence in A&E: the role of training and self-efficacy.

Nursing Standard 15, 33–38.

Luck L, Jackson D & Usher K (2007) Innocent or culpable?

Meanings that emergency department nurses ascribe to indi-

vidual acts of violence. Journal of Clinical Nursing 17, 1071–

1078.

Madine K & Smith B (2005) WorkForce and your workforce.

Emergency Nurse 13, 10–11.

May DD & Grubbs LM (2002) The extent, nature and precipitating

factors of nurse assault among three groups of registered nurses in

a regional medical center. Journal of Emergency Nursing 28, 11–

17.

McNamara R, Yu DK & Kelly JJ (1997) Public perception of safety

and metal detectors in an urban emergency department. American

Journal of Emergency Medicine 15, 40–42.

McPhaul KM & Lipscomb JA (2004) Workplace violence in health

care: recognised but not regulated. Online Journal of Issues in

Nursing 9, Manuscript 6. Available at: http://www.nursingworld.

org/MainMenuCategories/ANAMarketplace/ANAPeriodicals/OJIN/

TableofContents/Volume92004/No3Sept04/ViolenceinHealthCare.

aspx (accessed 21 March 2007).

Merecz D, Rymaszewska J, Mocicka A, Kiejna A & Jarosz-Nowak J

(2006) Violence at the workplace – a questionnaire survey of

nurses. European Psychiatry: the Journal of the Association of

European Psychiatrists 21, 442–450.

Meyer T, Wrenn K, Wright SW, Glaser J & Slovis CM (1997)

Attitudes toward the use of a metal detector in an urban emergency

department. Presented at the Rocky Mountain Conference of

Emergency Medicine, Breckenridge, CO, March 1996. Annals of

Emergency Medicine 29, 621–624.

Nicholson PJ (2007) How to undertake a systematic review in an

occupational setting. Occupational & Environmental Medicine 64,

353–358.



Pearson A, Field J & Jordan Z (2007) Evidence-Based Clinical Prac-

tice in Nursing and Healthcare: Assimilating Research, Experience

and Expertise. Blackwell Publishing, Victoria, Australia.

Peek-Asa C, Cubbin L & Hubbell K (2002) Violent events and

security programs in California emergency departments before and

after the 1993 Hospital Security Act. Journal of Emergency

Nursing 28, 420–426.

Perrone S (1999) Violence in the Workplace. Research and Public

Policy Series no. 22. Australia Institute of Criminology,

Canberra.

Polit DF & Beck CT (2006) Essentials of Nursing Research: Meth-

ods, Appraisal and Utilization, 6th edn. Lippincott Williams &

Wilkins, Philadelphia.

Presley D & Robinson G (2002) Violence in the emergency depart-

ment: nurses contend with prevention in the health care arena.

Nursing Clinics of North America 37, 161–169.

Rankins RC & Hendey GW (1999) Effect of a security system on

violent incidents and hidden weapons in the emergency

department. Annals of Emergency Medicine 33, 676–679.

Royal College of Nursing Australia (2007) National Overview of

Violence in the Workplace (Rumsey M, Foley E, Harrigan R &

Dakin S eds). Royal College of Nursing, Australia.

Ryan D & Maguire J (2006) Aggression and violence – a problem in

Irish Accident and Emergency departments? Journal of Nursing

Management 14, 106–115.

Stirling G, Higgins JE & Cooke MW (2001) Violence in A&E

departments: a systematic review of the literature. Accident and

Emergency Nursing 9, 77–85.

University of Iowa Injury Prevention Research Centre (2001)

Workplace Violence: A Report to the Nation, Iowa City, IA.

Available at: http://www.public-health.uiowa.edu/iprc/ (accessed 7

May 2007).

Walsh K (2007) On equilibrium: reflections on practice development

and the philosophy of John Ralston Saul. Nursing Philosophy 8,

201–209.

Walsh K, Moss C & FitzGerald M (2006) Solution-focused

approaches and their relevance to practice development. Practice

Development in Health Care 5, 145–155.

Wand TC & Coulson K (2006) Zero tolerance: a policy in conflict

with current opinion on aggression and violence management

in health care. Australasian Emergency Nursing Journal 9, 163–170.

West LJ (2003) The effect of an intervention on the risk of eruptive

violence in the emergency department. Thesis, Southern Connect-

icut State University, New Haven, CO.

L Anderson et al.


An integrative literature review of interventions to reduce violence against emergency department...

Documents

Transcript of An integrative literature review of interventions to reduce violence against emergency department...