v healthcare 1

M A I N T A I N I N G A N D R E T A I N I N G A H E A L T H Y W O R K F O R C E

The role of aggressions suffered by healthcare workers as predictors

of burnout

Santiago Gascon, Michael P Leiter, Eva Andrés, Miguel A Santed, Joao P Pereira, Marı́a J Cunha,

Agustı́n Albesa, Jesus Montero-Marı́n, Javier Garcı́a-Campayo and Begoña Martı́nez-Jarreta

Aims and objectives. To examine the prevalence of aggression against healthcare professionals and to determine the possible

impact that violent episodes have on healthcare professionals in terms of loss of enthusiasm and involvement towards work. The

objective was to analyse the percentage of occupational assault against professionals’ aggression in different types of healthcare

services, differentiating between physical and verbal aggression as a possible variable in detecting burnout in doctors and

nursing professionals.

Background. Leiter and Maslach have explored a double process model of burnout not only based on exhaustion by overload,

but also based on personal and organisational value conflicts (community, rewards or values). Moreover, Whittington has

obtained conclusive results about the possible relationship between violence and burnout in mental health nurses.

Design. A retrospective study was performed in three hospitals and 22 primary care centres in Spain (n = 1Æ826).

Methods. Through different questionnaires, we have explored the relationship between aggression suffered by healthcare

workers and burnout.

Results. Eleven percent of respondents had been physically assaulted on at least one occasion, whilst 34Æ4% had suffered threats

and intimidation on at least one occasion and 36Æ6% had been subjected to insults. Both forms of violence, physical and non-

physical aggression, showed significant correlations with symptoms of burnout (emotional exhaustion, depersonalisation and

inefficacy).

Conclusions. The survey showed evidence of a double process: (1) by which excess workload helps predict burnout, and (2) by

which a mismatch in the congruence of values, or interpersonal conflict, contributes in a meaningful way to each of the

dimensions of burnout, adding overhead to the process of exhaustion–cynicism–lack of realisation.

Relevance to clinical practice. Studies indicate that health professionals are some of the most exposed to disorders steaming

from psychosocial risks and a high comorbidity: anxiety, depression, etc. There is a clear need for accurate instruments of

Authors: Santiago Gascon, PhD, Assistant Professor, Department of

Psychology, Zaragoza University, Teruel, Spain; Michael P Leiter,

PhD, Professor, Centre for Organisational and Development

Research, Acadia University, Wolfville, NS, Canada; Eva Andrés,

PhD, Researcher, CIBER Epidemiologı́a y Salud Pública, Unidad

Epidemiologı́a Clı́nica, Hospital 12 de Octubre, Madrid; Miguel A

Santed, PhD, Professor of Psychology and Dean of the Faculty of

Psychology, Spanish Distance University (UNED), Madrid; Joao P

Pereira, PhD, Assistant Professor, Instituto Superior de Maia, Castelo

da Maia; Marı́a J Cunha, PhD, Assistant Professor, Instituto Superior

de Maia, Castelo da Maia, Portugal; Agustı́n Albesa, PhD Student,

Psychologist and Lawyer, Department of Psychology, Zaragoza

University, Zaragoza; Jesus Montero-Marı́n, PhD, Assistant

Professor, Zaragoza University, Zaragoza; Javier Garcı́a-Campayo,

PhD, Professor, Department of Psychiatry Zaragoza University,

Zaragoza, Spain

Correspondence: Santiago Gascón, Assistant Professor, Facultad de

CC Sociales y Humanas (Psicologı́a) Ciudad Escolar s/n., 44003

Teruel, Spain. Telephone: +34 978645343.

E-mail: [email protected]

This paper is based on the study by the University of Zaragoza,

whose basic data on the incidence of violence in health care and its

distribution among centres and facilities hospitals were already

published in Gascón S, Martı́nez-Jarreta B, González-Andrade F,

Santed MA, Casalod Y & Rueda MA (2009) Aggression towards

healthcare workers in Spain: a multicenter study to evaluate the

distribution of a growing violence among professionals, health

centers, and departments. International Journal of Occupational

and Environmental Health 15, 30–36. The levels of physical and

verbal violence in Spanish hospitals were reported in Gascón et al.

(2009), and this article (Gascón et al. 2012) aims to analyse the role

that these attacks could have on burnout.

� 2012 Blackwell Publishing Ltd 3120 Journal of Clinical Nursing, 22, 3120–3129, doi: 10.1111/j.1365-2702.2012.04255.x

evaluation to detect not only the burnout but also the areas that cause it. Professional exhaustion caused by aggression or other

factors can reflect a deterioration in the healthcare relationship.

Key words: aggression, burnout, community, healthcare relationship, healthcare workers, values

Accepted for publication: 12 May 2012

[Correction added on 15 August 2013, after first online publication: the authors, ‘Eva Andrés’, ‘Miguel A Santed’ and ‘Begoña Martı́nez-

Jarreta’, together with their affiliations were added. In addition, a statement that the paper is based on a study by the University of Zaragoza,

whose basic data and findings were previously published in Journal of Occupational and Environmental Health 15, 30–36 was inserted.

Previously omitted permissions to publish Tables 1, 3, 5 and 6 were also added to the relevant tables. Finally, acknowledgements of grants by the

Spanish Ministry of Labour & INHST (2008–2010): PSIPRV and the Department of Science and Education of the Aragonese Government

(Grupo Consolidado B44 & FONDO SOCIAL EUROPEO) were also inserted in the Acknowledgements section.]

Introduction

In Spain, burnout is not recognised as an occupational

disease (Royal Decree 1299/2006, Spanish Government

2006). However, many courts pass sentences that consider

burnout to be a work accident (Martı́nez de Viergol 2005).

The Spanish Law of Work Risk Prevention (Ley 31/1995

Prevención de Riesgos Laborales, Spanish Government 1995)

requires organisations to perform an evaluation of psycho-

social risks, but these laws have limited impact, partially

because of inconsistent enforcement of the regulations

(Gascón 2006). According to the Ministry of Work, the

population with the greatest psychosocial risk is that of

hospital workers (Gestal Otero 2003). It can seem somewhat

paradoxical that those who look after our health are those

who are most vulnerable to psychological disorders (Bruce

et al. 2002).

In the 1970s, burnout was described as a disorder

characterised by physical and emotional exhaustion, which

principally affected professions that demanded intense and

direct contact with others (Leiter & Schaufeli 1996). Later,

Maslach developed the MBI questionnaire (Maslach et al.

1996), which has been widely used to measure burnout as a

three-dimensional syndrome: exhaustion, depersonalisation

and lack of professional fulfilment (Leiter 2008). In 2000,

the same author along with Leiter centred their research in

the opposite dimensions of burnout: energy, implication and

efficacy, and the work-life areas that contribute to these

dimensions: workload, control, reward, community, fairness

and values (Maslach & Leiter 1997, Leiter & Maslach

2004a,b). Workload and control were derived from the

demand-control model of Karasek and Theorell (1990)

(Leiter & Schaufeli 1996, Leiter & Maslach 2004a,b).

Reward refers to the power of reinforcement to shape

behaviour. Community reflects the support of colleagues and

superiors as interpersonal conflicts, whilst fairness comes

from research on equity and social justice. Finally, value

reflects the cognitive-emotional power of job goals and

expectations. Each one of these areas includes various

aspects of organisational life. So community includes

positive and negative elements in interpersonal relationships

between colleagues and clients or patients; rewards include

aspects that provide intrinsic work satisfaction; and values

refer to the coherence between the individual’s values and

those of the organisation (Leiter & Maslach 2009, Leiter

et al. 2010).

The authors agree that overload and the control related to

exhaustion were not the only variable in burnout; the

aforementioned areas can erode the other dimensions and

generate cynicism and exhaustion (Leiter et al. 2008). As well

as exhaustion, the second aspect of burnout, implication,

includes ‘depersonalisation or cynicism’. With this dimension

reaching beyond the physical and emotional well-being of the

individual, the concept also expands its ability to concectarse

with the world. The third dimension includes the self-

perception that the worker has on their effectiveness and

performance at work.

This experience, maintained and becomes chronic over

time, erodes the vision of the worker on his ability to

transform and improve their working environment. These

three experiences, acting in an interdependent manner, form

what Leiter and Maslach considered a syndrome of three

dimensions (Laschinger & Finegan 2005, Leiter & Maslach

2009, Leiter et al. 2008).

As has been noted in several studies, a high level of

inconsistency in values, such as interpersonal conflicts, may

predict a significant proportion of the three dimensions of the

syndrome (Leiter & Maslach 2004a,b, Leiter et al. 2008). On

the other hand, in some professions such as health or

education, violent attacks on professionals from patients or

students have increased over the last years. Physical aggres-

sion, insults and threats can affect important areas such as

community, work rewards or a value conflict (Winstanley &

Whittington 2004, Gascón et al. 2008).

Maintaining and retaining a healthy workforce The role of aggressions suffered

� 2012 Blackwell Publishing Ltd Journal of Clinical Nursing, 22, 3120–3129 3121

In recent years, some countries have developed zero

tolerance laws against these types of behaviours, but the prob-

lem continues to be considered a health issue of the first order

(Whittington 2002). In health, most of the published studies

have been centred on psychiatry because it is a source of

violence for many reasons (McKinnon & Cross 2008),

distinct from other areas in health, especially in the nursing

profession (Dunn et al. 2007). However, recent studies have

revealed that violence is not exclusive to only one area in

hospitals (Martı́nez-Jarreta et al. 2008). Psychiatry is not the

only place in which violence takes place; in large hospitals, it

is superseded in the emergency area (Gascón et al. 2009b).

The nursing profession suffers continuous physical aggression

because of direct contact with the patient. Doctors and other

professionals are equally exposed, particularly to nonphysical

aggression (Martı́nez-Jarreta et al. 2008).

Aims and hypothesis

The aims of the study were to ascertain the weight that

aggression (physical, threats, verbal abuse, etc.) could have

when it comes to explaining burnout levels in the healthcare

sector and if it is produced in a similar way in doctors and

nursing professionals.

Health professionals are exposed to a wide variety of stress

sources: daily contact with sickness and death, attending

continuous urgent demands, uncertainty, the need to update

skills and knowledge, etc. (Gascón 2006). We consider that

violent episodes could contribute to exhaustion that manifests in

the variables of exhaustion, depersonalisation and inefficacy,

and the opposing dimensions: energy, involvement and efficacy.

Method

This was a retrospective study, using self-reporting, for the

purpose of researching levels of burnout and the experi-

ences of aggression and violence over the preceding twelve

months. The study was performed in three hospitals and 22

rural and urban primary care centres in Spain.

Participants

The survey included only permanent staff working continu-

ously in the same post for at least one year. A stratified

sample, proportional to the number of healthcare workers in

each centre, was studied. The questionnaires were distributed

in informative sessions carried out with groups of between 15

and 20 people, who were provided with information about

the study and how to answer the study questions. Partici-

pants returned the survey to boxes located in each of the

healthcare facilities. Of the 3000 surveys delivered, 2279

were returned and 1826 were used according to the criteria

of proportionality. The proportion, broken down by profes-

sion, was 603 doctors (33Æ02%), 878 nursing professionals

(48Æ08%), 144 administration staff (7Æ9%), 23 managers

(1Æ26%), 49 patient attendants (2Æ7%), 129 technical staff

and other professions (7Æ06%). The average age of the

participants was 41Æ84 years (SD 8Æ427); 64Æ2% were women

and 35Æ8% were men Table 1.

Instruments

The health workers completed a booklet containing the

following questionnaires:

• A demographic data record asked for personal, family and workplace information.

• Maslach Burnout Inventory-General Survey (MBI, Mas- lach et al. 1996). The 22 items are framed as statements of

job-related feelings (e.g. ‘I feel burned out from my work’;

‘I feel confident that I am effective at getting things done’)

and are rated on a 6-point frequency scale (ranging from

0 = ‘never’ to 6 = ‘daily’). Burnout is reflected by higher

Table 1 Number of participants and percent of workforce represented, by facility type and post (n = 1826)

Nurses

(%)

Doctors

(%)

Managers

(%)

Administration

(%)

Patient attendants

(%)

Others

(%)

Total participants

(%)

Large hospital 361 (20Æ0) 256 (21Æ5) 4 (13Æ0) 68 (27Æ4) 27 (0Æ6) 87 (25Æ7) 803 (43Æ9 Medium-sized hospital 169 (19Æ9) 91 (19Æ4) 14 (53Æ8) 22 (12Æ5) 16 (10Æ4) 31 (6Æ4) 343 (18Æ8) Small hospital 157 (52Æ7) 50 (35Æ7) 2 (11Æ7) 14 (14Æ6) 6 (3Æ3) 11 (37Æ9) 240 (13Æ14) Rural primary care 41 (7Æ6) 56 (11Æ7) 3 (33Æ3) 10 (22Æ8) 0 (0Æ0) 0 (0Æ0) 110 (6Æ02) Urban primary care 150 (80Æ2) 150 (67Æ3) 0 (0Æ0) 30 (75Æ0) 0 (0Æ0) 0 (0Æ0) 330 (18Æ07) Total 878 603 23 144 49 129 1826

Reproduced with permission from Maney Publishing from: Gascón S, Martı́nez-Jarreta B, González-Andrade F, Santed MA, Casalod Y &

Rueda MA (2009) Aggression towards healthcare workers in Spain: a multicenter study to evaluate the distribution of a growing violence among

professionals, health centers, and departments. International Journal of Occupational and Environmental Health 15, 30–36. Available at: http://

www.maneypublishing.com/journals/oeh.

S Gascon et al.

� 2012 Blackwell Publishing Ltd 3122 Journal of Clinical Nursing, 22, 3120–3129

scores on exhaustion and depersonalisation/cynicism and

lower scores on efficacy.

• Areas of Work-life Scale (AWS, Leiter & Maslach 2004a) comprises 45 items: 16 offered information about the

opposing dimensions to burnout (energy, involvement and

efficacy), and 29 items produce distinct scores for six areas

of work-life from a positive dimension – manageable

workload, control, reward, community, fairness and val-

ues (Leiter et al. 2010). The item measuring the perceived

congruence employed in different areas of their work is,

for example, ‘I have enough time to do what’s important

in my job’ (workload), or ‘Working here forces me to

compromise my values’ (values). Respondents indicate

their level of agreement on a five-point Likert-type scale

ranging from 1 (strongly disagree) to 5 (strongly agree).

The score is subsequently reversed negative items (Leiter

& Maslach 2004a). The AWS items were developed from

a series of staff surveys conducted by the Centre for

Organisational Research & Development (Leiter & Har-

vie 1998, Leiter et al. 2008) as a means of assessing the

constructs underlying their analysis of the six areas of

work-life. The scale has yielded a consistent factor struc-

ture across samples (Leiter et al. 2010).

• Aggression Questionnaire (Gascón et al. 2009b) contained descriptions of various forms of aggression, in line with

what the Occupational Safety and Health Administration

(Cal. OSHA 1995) classifies as type II (Table 2), that is,

exercised by customers, users or patients, excluding

aggression not related to the workplace and aggression

by work colleagues or superiors. For each type a definition

was provided. The definitions, adopted from various

international organisations (Di Martino 2002), had previ-

ously been used by other authors (Winstanley & Whit-

tington 2004), included definition for physical aggression,

verbal threats, threatening behaviour and verbal abuse

(Table 3). Participants were asked whether they had any of

the listed types of violence and how often in the previous

12 months (regardless of the type or the resulting lesion),

using a Likert-type scale (0 = never; 1 = never, but has

been witnessed to it happening to others; 2 = on one

occasion; 3 = on two or more occasions; 4 = on more than

five occasions).

Ethical considerations

Authorisation from the Ethics Committee (CEICA, Aragon

Institute of Health Sciences – IACS) was obtained. Participa-

tion was voluntary and the professionals studied authorised

the inclusion of their data in the survey, by ticking a box. All

the answers were treated anonymously and the question-

naires were destroyed, after the data were collected.

Methodology

A series of correlation and regression analysis examined the

relationship between aggression on burnout dimensions, and

its relative contribution to the two process model proposed

by Leiter. Excessive workload predicted fatigue that, in turn,

predicted depersonalisation and low levels of efficacy.

Moreover, a mismatch in values contributed significantly to

a regression analysis predicting each of the dimensions of

burnout syndrome (Leiter et al. 2008).

Results

Survey shows high levels of physical and verbal abuse

against health professionals during the previous year

(Gascón et al. 2009a). Eleven per cent of respondents

Table 2 Types of violence investigated

Type of violence

in the workplace Definition

Type I Without relation to

work. Delinquency

Type II Carried out by clients,

users, patients and pupils

Type III Carried out by members of the

workforce: bosses, colleagues,

subordinates

Type IV Domestic violence, or personal

problems without relation to work

California Occupational Safety and Health Administration (OSHA

1995). Guidelines for workplace security.

Table 3 Definition of violent behaviour investigated

Aggression Definition

Physical

aggression

Intentional behaviour with the

use of physical force, producing

physical, sexual or psychological

damage: kicking, slapping, stabbing,

pushing and pulling, biting and pinching

Verbal threats or

threatening behaviour

The promise of using physical strength

or power, which produces the fear

of physical, sexual, psychological

damage or other negative consequences

Insults or slander Verbal behaviour which humiliates,

degrades or shows lack of respect

Reproduced with permission from John Wiley & Sons Ltd from:

Winstanley S & Whittington R (2004) Aggression towards health

care staff in a UK general hospital: variation among professions and

departments. Journal of Clinical Nursing 13, 3–10. Doi: 10.1111/

j.1365-2702.2004.00807.x.

Maintaining and retaining a healthy workforce The role of aggressions suffered

� 2012 Blackwell Publishing Ltd Journal of Clinical Nursing, 22, 3120–3129 3123

(201) had been physically assaulted on at least one occasion,

5% (92) on more than one occasion, whilst 34Æ4% (628)

had suffered threats and intimidation on at least one

occasion and 23Æ8% (435) repeatedly, and 36Æ6% (668)

had been subjected to insults on at least one occasion and

24Æ3% (444) repeatedly.

The incidence was higher in large hospitals (Table 4), with

very high levels in services such as Accidents & Emergencies

and Psychiatry, with the incidences, respectively, of 48 and

26Æ9% for aggression, 82Æ1 and 64Æ1% for insults and 87Æ2

and 58Æ6% for threats (Table 5).

No statistically significant association was observed between

physical aggression and the victim¢s gender (v2 = 3Æ498; p = 0Æ610), nor in relation to professional category

(v2 = 8Æ295; p = 0Æ141) (Table 6). This form of violence was

associated more with a specific type of centre and services. Both

forms of violence, physical and nonphysical aggression,

showed significant correlations with symptoms of burnout,

as much in the MBI dimensions (emotional exhaustion,

depersonalisation and inefficacy), as in the dimensions of

AWS (energy, involvement and efficacy) (Table 7).

By multiple regression analysis were analysed the resulting

patterns of the CFA further. The outcome variables were the

positive dimensions which Leiter & Maslach proposed for

burnout: energy, involvement and efficacy. For involvement/

cynicism, energy/exhaustion was entered as a predictor in

the first step; for efficacy, cynicism was entered as a

predictor in the first step, following a process model of

burnout. In the second step of these analyses, both values

and workload were permitted to enter in a stepwise fashion:

Table 4 Percentage of aggression by centre

Centre

Large hospital

(%)

Medium-sized

hospital (%)

Small hospital

(%)

Urban primary

health centres (%)

Rural primary health

centres (%) v2 Meaning

Physical

aggression

21Æ9 21Æ7 9Æ5 17Æ4 11Æ3 26Æ435 p < 0Æ001

Insults 58Æ2 56Æ3 49Æ7 59Æ5 58Æ5 0Æ087 p = 0Æ100 Threats 55Æ7 56Æ3 43Æ2 57Æ0 58Æ5 0Æ120 p = 0Æ003

Table 5 Percentage of aggression by area

Service Surgical (%) Central (%) Medical (%) A&E (%) Psychiatry (%) Others (%) v2 Meaning

Physical

aggression

6Æ3 17Æ1 9Æ2 48Æ0 26Æ9 20Æ0 45Æ903 p < 0Æ001

Insults 62Æ3 25Æ0 47Æ1 82Æ1 64Æ6 25Æ0 19Æ995 p < 0Æ001 Threats 62Æ3 24Æ8 44Æ6 87Æ2 58Æ6 29Æ4 25Æ825 p < 0Æ001

Reproduced with permission from Maney Publishing from: Gascón S, Martı́nez-Jarreta B, González-Andrade F, Santed MA, Casalod Y &

Rueda MA (2009) Aggression towards healthcare workers in Spain: a multicenter study to evaluate the distribution of a growing violence among

professionals, health centers, and departments. International Journal of Occupational and Environmental Health 15, 30–36. Available at: http://

www.maneypublishing.com/journals/oeh.

Table 6 Percentage of aggression by post

POST

Admin

(%)

Patient

attendant (%)

Manager

(%)

Nurse

(%)

Doctor

(%)

Others

(%) v2 Meaning

Physical

aggression

7Æ5 18Æ2 10Æ0 17Æ0 19Æ4 11Æ1 8Æ295 p = 0Æ141

Insults 55Æ0 50Æ0 40Æ0 54Æ6 61Æ6 39Æ3 11Æ614 p = 0Æ040 Threats 42Æ5 39Æ1 65Æ0 49Æ9 60Æ7 35Æ7 22Æ678 p < 0Æ001

Reproduced with permission from Maney Publishing from: Gascón S, Martı́nez-Jarreta B, González-Andrade F, Santed MA, Casalod Y &

Rueda MA (2009) Aggression towards healthcare workers in Spain: a multicenter study to evaluate the distribution of a growing violence among

professionals, health centers, and departments. International Journal of Occupational and Environmental Health 15, 30–36. Available at: http://

www.maneypublishing.com/journals/oeh.

S Gascon et al.

� 2012 Blackwell Publishing Ltd 3124 Journal of Clinical Nursing, 22, 3120–3129

either one or both of the predictors could enter the equation

if their coefficient was significant at the 0Æ05 level. For

energy, values and workload were entered in this fashion in

step 1 (Leiter et al. 2008).

The analysis confirms the relationships of six areas with

the three positive dimensions of burnout. The samples from

the doctors and nurses were analysed separately as they

were the two biggest groups and the most affected by

violent episodes.

In the doctor sample, the six areas and aggression episodes

(except physical assault) explained the 67% variance in

energy: workload (t = 11Æ892, p £ 0Æ001; B = 0Æ336), control (t = 4Æ581, p £ 0Æ001; B = 0Æ149), reward (t = 5Æ615, p £ 0Æ001; B = 0Æ184), community (t = 4Æ359, p £ 0Æ001; B = 0Æ134), values (t = 4Æ407, p £ 0Æ001; B = 0Æ149), having been insulted (t = 3Æ926, p = 0Æ02; B = 0Æ162) and having

suffered threats (t = 4Æ008, p £ 0Æ001; B = 0Æ107). Although physical aggression was encountered in the doctor sample,

the results were not statistically significant. The energy

dimension in nurses was explained by workload (t = 9Æ322,

p £ 0Æ001; B = 0Æ349), reward (t = 4Æ135, p £ 0Æ001; B = 0Æ153), values (t = 4Æ018, p £ 0Æ001; B = 0Æ140), having been insulted (t = 3Æ826, p £ 0Æ001; B = 0Æ188) and having suffered threats (t = 3Æ919, p = 0Æ03; B = 0Æ101).

In involvement, AWS and aggression explained 53% vari-

ance in the doctor sample: workload (t = 3Æ046, p = 0Æ002;

B = 3Æ046), control (t = 2Æ204, p = 0Æ02; B = 0Æ081), reward

(t = 5Æ925, p £ 0Æ001; B = 0Æ221), values (t = 4.912, p £ 0Æ001; B = 0Æ188), physical aggression (t = 3.151, p = 0Æ05; B =

0Æ236), insults (t = 5Æ231, p = 0Æ001; B = 0Æ117) and threats

(t = 4Æ211, p £ 0Æ001; B = 0Æ193); in nurses: workload (t = 3Æ155, p £ 0Æ001; B = 0Æ077), reward (t = 4.672, p £ 0Æ001; B = 0Æ333), values (t = 5.202, p = 0Æ001; B = 0Æ201), physical

aggression (t = 3Æ285, p £ 0Æ001; B = 0Æ138), insults (t = 5Æ862, p = 0Æ002; B = 0Æ206) and threats (t = 2Æ383,

p £ 0Æ001; B = 0Æ200). The 37% variance in efficacy by doctors was explained by

control (t = 3Æ147, p = 0Æ02; B = 0Æ125), reward (t = 4Æ275,

p £ 0Æ001; B = 0Æ172), community (t = 4Æ280, p £ 0Æ001; B = 0Æ162), fairness (t = 2Æ548, p = 0Æ01; B = 0Æ104), values

(t = 1Æ463, p £ 0Æ001; B = 0Æ060), insults (t = 2Æ772, p £ 0Æ001; B = 0Æ100) and threats (t = 1Æ907, p £ 0Æ001; B = 0Æ095); for nurses: control (t = 3Æ478, p = 0Æ03;

B = 0Æ136), reward (t = 5Æ016, p £ 0Æ001; B = 0Æ228), values (t = 2Æ367, p £ 0Æ001; B = 0Æ053), physical aggression (t = 3Æ682, p £ 0Æ001; B = 0Æ049, insults (t = 3Æ160, p £ 0Æ001; B = 0Æ098) and threats (t = 1Æ891, p = 0Æ002; B = 0Æ086) Table 8.

The study confirmed the double process model of burnout

(Leiter et al. 2008). First, an excess charge weakens the

professional and prevents them regain their energy. On the

other hand, mismatches in the values and interpersonal

conflict, which include aggression, showed their ability to

predict the dimensions that conform to burnout syndrome.

The healthcare professionals reported moderately low

levels of workload and reported a negative evaluation of

the other five areas of work-life: rewards, community and

values that can be measured by the number, gravity and type

of aggression.

To ascertain the role of aggression in burnout, a

structural equation model (SEM) was conducted, using

the AWS and aggression levels, including the ten freed

error correlations. For the AWS, the analysis included three

indicators for each of the three subscales as defined in

Leiter et al. (2010), permitting the analysis to focus

primarily on the structural relationships among the con-

structs and to deemphasise the factor structure of the AWS

that has been established elsewhere (Leiter & Maslach

2004a,b). The mediated model produced an adequate fit

regarding the v2/df and RMSEA ( v2ð642Þ=1683Æ27,

p < 0Æ001; v2/df = 2Æ62; CFI = 0Æ895, RMSEA = 0Æ044).

The modification indices identified three areas of work-life

for which values did not fully mediate their relationships

with the burnout aspects: reward with exhaustion, com-

munity with cynicism and reward with efficacy. Adding

these three paths to a partially mediated model provided a

Table 7 Correlations (Rho Spearman) between burnout dimensions and types of aggression

MBI dimensions AWS dimensions

Emotional

exhaustion

Depersonalisation/

cynicism

Personal

accomplishment Energy Involvement Efficacy

Physical aggression 0Æ109** 0Æ145** �0Æ015 �0Æ189** �0Æ172** �0Æ117** Insults 0Æ167** 0Æ215** �0Æ137** �0Æ211** �0Æ273** �0Æ229** Threats 0Æ198** 0Æ237** �0Æ178** �0Æ248** �0Æ307** �0Æ325**

n = 1Æ826. **Correlations are significant at the 0Æ01 level (2 tailed).

Maintaining and retaining a healthy workforce The role of aggressions suffered

� 2012 Blackwell Publishing Ltd Journal of Clinical Nursing, 22, 3120–3129 3125

better fit (v2ð639Þ=1588Æ90, p < 0Æ001; v 2/df=2Æ49; CFI =

0Æ904, RMSEA=0Æ042) that was a significant improvement

over the mediated model (v2ð3Þ = 94Æ37, p < 0Æ001) and

bringing the CFI to an adequate level. In the partially

mediated model, aggression demonstrated relationships

with the areas such as reward, community and fairness

and with values which demonstrates relationships with the

three dimensions of burnout (Fig. 1).

Table 8 Regression analysis: positive dimension of burnout. Criterion variables: dimensions opposed to burnout. Predictor variables: areas of

Work Life (Leiter & Maslach 2008)

Unstandardised

coefficients Standardised Coefficients

B SE b t Sig.

Dependent variable: energy

doctors n = 603

Constant 2,509 1,484 1,690 0,091

Workload 0,788 0,056 0,336 11,892 0,000

Control 0,624 0,119 0,149 4,581 0,000

Reward 0,690 0,100 0,184 5,615 0,000

Community 0,329 0,075 0,134 4,359 0,000

Values 0,851 0,096 0,149 4,407 0,000

Insults 0,462 0,058 0,162 3,926 0,020

Threats 0,395 0,112 0,107 4,008 0,000

Dependent variable: energy nurses

n = 878

Constant 3,491 1,422 1,571 0,077

Workload 0,908 0,056 0,349 9,322 0,000

Reward 0,890 0,100 0,153 4,135 0,000

Values 0,626 0,096 0,140 4,078 0,000

Insults 0,441 0,058 0,188 3,826 0,000

Threats 0,288 0,112 0,101 3,919 0,030

Dependent variable: involvement

doctors n = 603

Constant 6,276 0,746 8,412 0,000

Workload 0,186 0,028 0,098 3,046 0,002

Control 0,232 0,060 0,081 2,204 0,028

Reward 0,298 0,050 0,221 5,925 0,000

Values 0,436 0,048 0,188 4,912 0,000

Physical aggression 0,177 0,050 0,236 3,151 0,050

Insults 0,244 0,058 0,117 5,231 0,001

Threats 0,217 0,062 0,193 4,211 0,000

Dependent variable: involvement

nurses n = 878

Constant 5,387 0,648 7,329 0,020

Workload 0,197 0,180 0,077 3,155 0,000

Reward 0,301 0,043 0,333 4,672 0,000

Values 0,232 0,049 0,201 5,202 0,001

Community 0,130 0,054 0,194 4,117 0,000

Physical aggression 0,198 0,049 0,138 3,285 0,000

Insults 0,316 0,047 0,206 5,862 0,002

Threats 0,170 0,056 0,200 2,383 0,000

Dependent variable: efficacy

doctors n = 603

Constant 5,588 0,613 9,119 0,000

Control 0,255 0,049 0,125 3,147 0,002

Reward 0,198 0,041 0,172 4,275 0,000

Community 0,133 0,031 0,162 4,280 0,000

Fair 0,081 0,032 0,104 2,548 0,011

Values 0,258 0,040 0,060 1,463 0,000

Insults 0,166 0,033 0,100 2,272 0,001

Threats 0,290 0,051 0,095 1,907 0,000

Dependent variable: efficacy

nurses n = 878

Constant 4,779 0,707 8,924 0,053

Control 0,366 0,053 0,136 3,478 0,003

Reward 0,252 0,049 0,228 5,016 0,000

Values 0,199 0,046 0,053 2. 367 0,000

Physical aggression 0,201 0,050 0,049 3,682 0,001

Insults 0,206 0,041 0,098 3,160 0,000

Threats 0,290 0,049 0,086 1,891 0,002

S Gascon et al.

� 2012 Blackwell Publishing Ltd 3126 Journal of Clinical Nursing, 22, 3120–3129

It is very important to note that only 19Æ8% of healthcare

workers felt they were supported by management (Gascón

et al. 2009a). In multiple regression analyses, this variable

was seen to be a modulating factor of the psychological effect

of aggression (p = 0Æ001; odd ratio = �0Æ771) by emotional exhaustion and (p = 0Æ001; odd ratio = �0Æ496) by deper- sonalisation.

Discussion

As shown in other studies, health workers show a significant

risk of verbal and physical abuse by patients or their compan-

ions (Arimatsu et al. 2008). This is shown by the results in our

study, especially in very complex hospitals and in areas like

emergencies and psychiatry (Gascón et al. 2009b). Every

professional in contact with the public is exposed to physical

and verbal violence, which particularly affects nursing and

medical staff (Whittington 2002, Winstanley & Whittington

2004, McKinnon & Cross 2008). The latter, along with the

direction staff, is additionally exposed, because they must

make important decisions and threats become a way to

influence them (Martı́nez-Jarreta et al. 2008).

Violent episodes are one of the many risks health profes-

sionals have to face, and, as it has been shown in many studies

(Dunn et al. 2007, Arimatsu et al. 2008), they have clear

effects on physical and psychological symptomatology. This

could perhaps help explain why doctors and nurses are the two

professions with the highest psychiatric comorbidity (Bruce

et al. 2002, Löwe et al. 2003). The association between non-

physical violence and anxiety and symptoms of post-traumatic

stress disorder was seen to be statistically significant in several

studies (Whittington 2002, Gascón et al. 2009a), and others

have shown that the burnout syndrome, like depression,

anxiety and other disorders, affects health professionals (Löwe

et al. 2003, Findorff et al. 2004). In this survey, both forms of

violence showed a similar impact on burnout, as much in the

MBI dimensions, as in the dimensions of AWS.

The syndrome is not an ‘either/or’ question, but a slow and

insidious process. Thus, it can and must be observed in health

institutions to prevent it. The need is clear for accurate

instruments of evaluation to detect not only the burnout but

also the areas that cause it.

In addition to providing a good instrument for measuring

the burnout syndrome (MBI), in 2000 Leiter and Maslach

created a questionnaire measuring the opposing dimensions

of the syndrome, as well as the causes that may contribute to

its development (Leiter & Maslach 2004a,b). The authors

developed the theory of the double process of burnout, which

explains that a worker can get burnt out not only because of

an excess of workload, but also because of conflicting values,

interpersonal problems in the work place, absence of reward

and lack of equity (Leiter et al. 2008).

This study confirms the model of two processes of

burnout: a evident process of workload–exhaustion and

another that shows the relationship that a mismatch in

values and interpersonal conflict has on the three dimensions

of burnout syndrome. Aggression has shown an important

contribution in the value congruence and the energy and

involvement dimensions. So that confirms that burnout is

more than an exhaustion syndrome (Leiter & Schaufeli

1996).

As various authors have pointed out, it is necessary to

develop strategies focused on minimising the occurrence of

violent episodes in health care and to minimise their possible

impact, when they occur, by deploying both a programme of

prevention and a programme of medical, psychological and

legal assistance (McKinnon & Cross 2008). Managing

violence with safe and effective protocols should be a goal

for hospital¢s managers and a necessary skill for profession- als, particularly for mental health nurses (Dunn et al. 2007),

emergency specialists (Martı́nez-Jarreta et al. 2008) and

other healthcare professionals.

On the basis of the results of this study, our team have

proposed prevention protocol, which includes medical and

Manageable workload

InvolvementAggression

Reward

Fairnes

Community

Efficacy

Outcomes

Energy

Values

–.05

–.59

–.47

.42–.75

.17

.16

–.62

.29

.14

.16

.38 .53

Figure 1 Casual model coefficients:

aggression – areas work-life (n = 1826).

Maintaining and retaining a healthy workforce The role of aggressions suffered

� 2012 Blackwell Publishing Ltd Journal of Clinical Nursing, 22, 3120–3129 3127

legal counsel as well as psychological support for the victims.

We are currently working on intervention models in which

the professionals themselves have proposed measures of

change to be implemented in hospitals with the support of

management.

In view of the results, we cannot establish a cause–effect

relationship between aggression and burnout. The variables

studied could contribute to a maladaptive cycle in which the

violence contributes to exhaustion and cynicism, but we can

also consider that these events can promote aggressive behav-

iour of a patient who does not feel well served. Further studies

are needed to establish the contribution of intervention

programmes: both violence prevention and in providing

professional skills to manage potentially conflicting situations.

Acknowledgements

This study was conducted with support from the Ministry of

Health of Spain and the IIISASO (International Research of

Social, Environmental and Occupational Health Institute).

In addition, this work was supported by grants from the

Spanish Ministry of Labour & INHST (2008–2010): PSIPRV

and the Department of Science and Education of the

Aragonese Government (Grupo Consolidado B44 &

FONDO SOCIAL EUROPEO).

Conflict of interest

None.

References

Arimatsu M, Wada K, Yoshirakawa T, Oda

S, Taniguchi H, Aizawa Y & Higashi T

(2008) An epidemiological study of

work-related violence experienced by

physicians who graduated from a

Medical School in Japan. Journal of

Occupational Health 50, 357–361.

Bruce M, McAvoyr G, Raue P, Brown EL,

Meyers BS, Keohane DJ & Jagoda D

(2002) Major depression in elderly

home health care patients. American

Journal of Psychiatry 159, 8.

California Occupational Safety and Health

Administration (1995) Guidelines for

Workplace Security. Department of

Industrial Relations, Division of Occu-

pational Safety and Health, San Fran-

cisco, CA.

Di Martino V (2002) Workplace Violence in

the Health Sector-Country Case Studies

[Brazil, Bulgaria, Lebanon, Portugal,

South Africa, Thailand, and an

Additional Australian Study]; Synthesis

Report. OIT/OMS/CIE/ISP, Ginebra

(SWZ).

Dunn K, Elsom S & Cross W (2007) Self-

efficacy and locus of control affect

management of aggression by mental

health nurses. Issues in Mental Health

Nursing 28, 201–217.

Findorff MJ, McGovern PM, Wall M,

Gerberich SG & Alexander B (2004)

Risk factors for work related violence

in a health care organization. Injury

Prevention 10, 296–302.

Gascón S (2006) Intervención Sobre Riesgos

Psicosociales: Una Asignatura Pendi-

ente. Available at: https://www.preven

cionintegral.com (accessed 28 August

2012) [in Spanish].

Gascón S, Casalod Y, Jarreta BM, Abecia E,

Luna A, Cárceles MD, Santed MA,

González-Andrade F & Bolea M

(2009a) Aggression against healthcare

workers: an approach to the situation in

Spain and the victims psychological ef-

fects. Legal Medicine (Tokyo) 11(Sup-

pl. 1), S366–S367.

Gascón S, Martı́nez-Jarreta B, González-

Andrade F, Santed MA, Casalod Y &

Rueda MA (2009b) Aggression towards

healthcare workers in Spain: a multi-

center study to evaluate the distribution

of a growing violence among profes-

sionals, health centers, and depart-

ments. International Journal of

Occupational and Environmental

Health 15, 30–36.

Gestal Otero J (1993) Riesgos del Trabajo

del Personal Sanitario, 2nd edn. Inter-

americana-McGraw-Hill, Madrid [in

Spanish].

Karasek R & Theorell T (1990) Halthy

Work: Stress, Productivity, and the

Reconstruction of Working Life. Basic

Books, New York, NY.

Laschinger HK & Finegan J (2005) Empow-

ering nurses for work engagement and

health in hospital settings. Journal of

Nursing Administration 35, 439–449.

Leiter MP (2008) A two process model of

Burnout and work engagement: distinct

implications of demands and values.

Giornale Italiano di Medicina del

Lavoro ed Ergonomia Supplemento A,

Psicologia 30, A52–A58.

Leiter MP & Schaufeli WB (1996) Consis-

tency of the burnout construct across

occupations. Anxiety, Stress, and Cop-

ing 9, 229–243.

Leiter MP & Harvie P (1998) Conditions for

staff acceptance of organizational

change: burnout as a mediating con-

struct. Anxiety, Stress, and Coping 11,

1–25.

Leiter MP & Maslach C (2004a) The Areas

of Worklife Scale Manual. Centre For

Organizational Research & Develop-

ment, Wolfville, NS.

Leiter MP & Maslach C (2004b) Areas of

worklife: a structured approach to

organizational predictors of job

burnout. In Research in Occupational

Stress and Well Being: Vol. (3)

Emotional and Physiological Processes

and Positive Intervention Strategies

(Perrewé P & Ganster DC eds). JAI

Press/Elsevier, Oxford, pp. 91–134.

Leiter MP & Maslach C (2009) Nurse

turnover: the mediating role of burnout.

Journal of Nursing Management 17,

331–339.

Leiter MP, Gascón S & Martı́nez-Jarreta B

(2008) A two process model of Burn-

out: their relevance to Spanish and

Canadian nurses. Psychology in Spain

12, 37–45.

Leiter MP, Gascón S & Martı́nez-Jarreta B

(2010) Making sense of worklife:

a structural model of burnout. Journal

of Applied Social Psychology 40, 57–

75.

Löwe B, Gräfe K, Kroenke K, Zipfel S,

Quenter A, Wild B, Fiehn C &

S Gascon et al.

� 2012 Blackwell Publishing Ltd 3128 Journal of Clinical Nursing, 22, 3120–3129

Herzog W (2003) Predictors of psy-

chiatric comorbidity in medical

outpatients. Psychosomatic Medicine

65, 764–770.

Martı́nez de Viergol A (2005) La consider-

ación del sı́ndrome de burnout como

constitutivo de la contingencia profe-

sional de Accidente de Trabajo. Revista

del Ministerio de Trabajo y Asuntos

Sociales 59, 213–224 [in Spanish].

Martı́nez-Jarreta B, Gascón S, Santed MA,

González-Andrade J & Goicoechea J

(2008) Medical- Legal analysis of

aggression towards health care profes-

sionals. An approach to a silent reality

and its consequences on the health.

Medicina Clı́nica Barcelona 128, 307–

310.

Maslach C & Leiter MP (1997) The Truth

About Burnout. Jossey Bass, San Fran-

cisco, CA.

Maslach C, Jackson SE & Leiter MP (1996)

Maslach Burnout Inventory Manual,

3rd edn. Consulting Psychologists Press,

Palo Alto, CA.

McKinnon B & Cross W (2008) Occupa-

tional violence and assault in mental

health nursing: a scooping project for a

Victorian Mental Health Service.

International Journal of Mental Health

Nursing 7, 9–17.

Spanish Government (1995) Ley de

Prevención de Riesgos Laborales (Law

on Prevention of Occupational Risks).

Ley 31/1995 de 8 de noviembre

(Law 31/1995 of 8 November). Gazette

No. 269 (11/10/1995), Madrid [in

Spanish].

Spanish Government (2006) Real Decreto

por el que se aprueba el cuadro de en-

fermedades profesionales en el sistema

de la Seguridad Social y se establecen

criterios para su notificación y registro

(Royal Decree Approving the Schedule

of Occupational Diseases in the Social

Security System and Establishes Criteria

for Notification and Registration). Real

Decreto 1299/2006 de 10 de noviembre

(Royal Decree 1299/2006 of 10

November). BOE (Official State Gaz-

ette) No. 302 (19/12/2006), Madrid [in

Spanish].

Whittington R (2002) Attitudes toward pa-

tient aggression amongst mental health

nurses in the ‘zero tolerance’ era: asso-

ciations with burnout and length of

experience. Journal of Clinical Nursing

11, 819–825.

Winstanley S & Whittington R (2004)

Aggression towards health care staff in

a UK general hospital: variation among

professions. Journal of Clinical Nursing

13, 3–10.

The Journal of Clinical Nursing (JCN) is an international, peer reviewed journal that aims to promote a high standard of

clinically related scholarship which supports the practice and discipline of nursing.

For further information and full author guidelines, please visit JCN on the Wiley Online Library website: http://

wileyonlinelibrary.com/journal/jocn

Reasons to submit your paper to JCN: High-impact forum: one of the world’s most cited nursing journals, with an impact factor of 1Æ316 – ranked 21/101 (Nursing

(Social Science)) and 25/103 Nursing (Science) in the 2012 Journal Citation Reports� (Thomson Reuters, 2012).

One of the most read nursing journals in the world: over 1Æ9 million full text accesses in 2011 and accessible in over 8000

libraries worldwide (including over 3500 in developing countries with free or low cost access).

Early View: fully citable online publication ahead of inclusion in an issue.

Fast and easy online submission: online submission at http://mc.manuscriptcentral.com/jcnur.

Positive publishing experience: rapid double-blind peer review with constructive feedback.

Online Open: the option to make your article freely and openly accessible to non-subscribers upon publication in Wiley

Online Library, as well as the option to deposit the article in your preferred archive.

Maintaining and retaining a healthy workforce The role of aggressions suffered

� 2012 Blackwell Publishing Ltd Journal of Clinical Nursing, 22, 3120–3129 3129

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material.