PICOT Question and Literature Search

The connection between illness representations of Alzheimer’s disease and burnout among social

workers and nurses in nursing homes and hospitals: a mixed-methods investigation

Shiri Shinan-Altmana*, Perla Wernerb and Miri Cohenb

aSchool of Social Work, Bar-Ilan University, Ramat Gan, Israel; bDepartment of Gerontology, University of Haifa, Mt. Carmel Haifa, Israel

(Received 31 May 2014; accepted 5 January 2015)

Objective: To examine the relationship between Alzheimer’s disease (AD) illness representations and burnout among social workers and nurses, based on the self-regulatory model. Method: A mixed-methods study was conducted. First, 327 social workers and nurses completed measures of cognitive and emotional representations, burnout (emotional exhaustion, depersonalization, lack of personal accomplishment), role variables, knowledge about AD, emotion-focused coping, problem-focused coping, and demographic and occupational characteristics. Second, interviews were conducted with eight social workers and nurses to uncover their perceptions about AD and their burnout experience. Results: Using structural equation modeling and controlling background variables, findings indicated that emotional representations were associated with burnout while only some of the cognitive illness representations were associated with burnout. While cognitive illness representations were associated directly to burnout, the association between emotional representations and burnout was mediated by emotion-focused coping. The trimmed model showed a good fit of the data and explained 32.2% of the variance in emotion-focused coping, 51% of the variance in emotional exhaustion, 37.7% of the variance in depersonalization, and 22.6% of the variance in lack of personal accomplishment. Interviews demonstrated that AD characteristics were perceived as affecting participants on both personal and professional levels; the participants expressed negative feelings towards AD and stated that these perceptions and feelings had led them to burnout. Conclusions: AD illness representations may be a risk factor for developing burnout. New directions for intervention programs, aiming to reduce burnout, should be examined.

Keywords: illness perceptions; professionals; work stressors

Introduction

Each year, 20% to 40% of patients with Alzheimer’s dis-

ease (AD) are hospitalized for an average of 3.7 days per

person-year and statistical forecast predict that these per-

centages will continue to grow even more in the upcoming

years (Fillenbaum, Heyman, Peterson, Pieper, & Weiman,

2000; Fong et al., 2012). Social workers and nurses are

the two primary groups who provide care for AD patients

(Kwok, Lam, Yip, & Ho, 2011) in general hospitals and

in nursing homes. These two groups are required to cope

with the burden of care associated with AD patients

(Schmidt, Dichter, Bartholomeyczik, & Hasselhorn,

2014; Shinan-Altman, Werner, & Cohen, 2014).

Burnout is defined as a syndrome experienced in

response to chronic on-the-job stressors (Maslach, Schau-

feli, & Leiter, 2001), and it has both consequences of

organizational factors and a driver of suboptimal well-

being and productivity (Cimiotti, Aiken, Sloane, & Wu,

2012; Rossler, 2012;Van Bogaertp, Clarke, Willems, &

Mondelaers, 2013). Based on empirical data, Pines and

Maslach (1978) clustered burnout symptoms across a

three-dimensional scale: emotional exhaustion, refers to

feelings of being emotionally exhausted by one’s work;

depersonalization, refers to impersonal responses and lack

of feelings towards clients; lack of personal

accomplishment, refers to feelings of incompetence and

unsuccessful achievement in one’s work. Various factors

may create burnout among professionals (Garrosa,

Rainho, Moreno-Jim�enez, & Monteiro, 2010; Jourdain & Chenevert, 2010). Thus, the severity and complexity of

clients’ problems have already proved to be a major

source of burnout among professionals, in general (Cohen

& Gagin, 2005; Shinan-Altman & Cohen, 2009), and

among professionals who care for AD patients, in particu-

lar (Todd & Watts, 2005). AD patient caregivers are espe-

cially vulnerable to burnout, due to the patients’ high

dependency on professionals for daily functioning (Pek-

karinen et al., 2006) and to AD patients’ challenging

behavior (Mackenzie & Peragine, 2003). Indeed, burnout

among professionals who take care of AD patients was

associated with low motivation to respond to the patients’

physical and emotional needs (Todd & Watts, 2005).

In order to reduce professional burnout, it is crucial to

understand the factors associated with it. It was shown

that work stressors such as role conflict, role ambiguity,

and work overload are major causes of burnout in work

settings (Maslach et al., 2001). Furthermore, it was found

that professional burnout is related, among other factors,

to professionals’ perceptions of their patients’ illness, as

well as to their emotional reactions to patients (Edberg &

*Corresponding author. Email: [email protected]

� 2015 Taylor & Francis

Aging & Mental Health, 2016

Vol. 20, No. 4, 352�361, http://dx.doi.org/10.1080/13607863.2015.1008983

Hallberg, 2001; Todd & Watts, 2005). However, only a

few studies have examined the relationships between ill-

ness perceptions and burnout (Edberg & Hallberg, 2001;

Todd & Watts, 2005). According to these studies, profes-

sionals may experience feelings of despair and frustration,

which may lead to burnout, as a result of misconceptions

about patients, negative beliefs towards the behavior of

patients (Edberg & Hallberg, 2001), and negative beliefs

and emotional reactions to AD (Todd & Watts, 2005).

In order to assess professionals’ beliefs and emotional

reactions towards their patients’ AD, our study was based

on the self-regulatory model (SRM). The SRM provides

an integrated and empirically validated model for dealing

with beliefs and coping related to illness (Leventhal,

Meyer, & Nerenz, 1980). According to the model, when

an individual is confronted with an illness, he or she will

attempt to assign meaning to this illness by accessing per-

ceptions about the illness. The SRM describes the ways in

which individuals perceive illness (defined as ‘cognitive

illness representations’) and react to it emotionally

(defined as ‘emotional representations’).

Based on Moss-Morris et al.’s (2002) conceptualization

of the SRM model, the cognitive component consists of

seven categories of subjective beliefs about the illness:

identity � symptoms that one attributes to the disease; causes � the perceived causes of the symptoms; timeline � beliefs about the course of the illness; consequences – the perceived impact of the illness; control – the perceived

ability to control the illness; illness coherence � the extent to which patients perceive that they understand the illness

and its implications, and cyclical timeline � beliefs about the temporal changeability of the illness (Moss-Morris et

al., 2002). The emotional component refers to the produc-

tion of emotional responses to the disease such as worry,

anger, and anxiety (Leventhal et al., 1980). According to

Leventhal et al. (1980), illness representations affect coping

strategies, defined as the cognitive and behavioral actions a

person takes (or does not take) to enhance health and treat-

ment (i.e., cure or control), and rehabilitate from the illness.

In summary, the model suggests that perceptions about an

illness and coping strategies can have an impact on psycho-

logical well-being.

The SRM (Leventhal et al., 1980) was developed to

explain illness representations among sick people, but

recently it has also been used to examine illness represen-

tations of healthy people and professionals (Insel, Meek,

& Leventhal, 2005; Shinan-Altman et al., 2014). Never-

theless, the literature discussing illness representations

among professionals, in general, and in the area of AD, in

particular, is still limited. Thus, the aim of the present

study was to build and test a comprehensive model to

examine the relationship between illness representations

and burnout. In addition, knowledge about AD, work

stressors, and personal and professional variables were

assessed in relation to illness representations and burnout.

Methodology overview

To address the research aim, we used a mixed-methods

approach combining both quantitative and qualitative

methodologies. Mixed-methods research offers an oppor-

tunity for a deeper examination of the research questions

and provides an opportunity to increase the breadth and

depth of understanding, and enhance the quality of data

interpretation (Johnson, Onwuegbuzie, & Turner, 2007).

In the current study, we sought to enhance and illustrate

the results from the quantitative method with results from

the qualitative method by integration between the two

methods (Meissner, Creswell, Klassen, Plano Clark, &

Smith, 2011).

Study 1: quantitative method

Methods

Participants

Overall, 360 social workers and nurses were asked to par-

ticipate in the study, and 327 consented, rendering a

response rate of 90.8%, a total of 205 nurses and 122

social workers. Reasons for non-participation were mainly

lack of interest and time constraints. Inclusion criteria

were: being a social worker or nurse who works with AD

patients (we asked participants whether they are working

or used to work with patients who were diagnosed by psy-

chogeriatric as having AD) and having tenure of over 3

months in one’s current place of work. The latter condi-

tion was based on the assumption that there is a need for a

minimum time of exposure and familiarity with AD

patients, in order to provide basic answers to the study’s

questionnaire. In addition, working in wards with no AD

patients was considered as an exclusion criterion.

As shown in Table 1, participants’ mean age was 40;

their mean years of education was 16. Most participants

were women; more than half were married, born in Israel,

Table 1. Participants’ demographic characteristics (n D 327). Variables

Gender, N (%)

Male 31 (9.9)

Female 283 (90.1)

Age,M (SD), range 40.83 (10.68), 22�27 Education,M (SD), range 15.94 (2.13), 10�25 Number of children,M (SD), range 2.29 (1.10), 0�6 Marital status, N (%)

Single 50 (15.5)

Married 230 (70.3)

Divorced 28 (8.7)

Other 15 (4.6)

Place of birth, N (%)

Israel 194 (60.6)

Former Soviet Union 99 (30.9)

Europe�America 14 (4.4) Asia�Africa 13 (4.1)

Income�, N (%) Above average 78 (25.8)

Average 61 (20.1)

Below average 164 (54.2)

�Average monthly Israeli’s wage is 9330 NIS (about 2600$) according to the Israel Central Bureau of Statistics (2012).

Aging & Mental Health 353

with an average of two children. Finally, more than three-

quarters of participants reported that their gross household

income was below average, according to Israeli standards.

Measures

Illness representations were assessed using the revised ill-

ness perception questionnaire and included eight dimen-

sions: identity, causes, timeline, consequences, control,

coherence, cyclical timeline, and emotional representa-

tions (IPQ-R; Moss-Morris et al., 2002). For the illness

identity dimension, participants were asked whether they

perceived a list of 13 symptoms as being related to

advanced stages of AD (0 D no, 1 D yes). Symptoms of AD were adapted from the American Alzheimer’s Associ-

ation (2009) and from other studies (Hoyer & Roodin,

2009; Orengo et al., 2008). Symptoms of AD as shown by

the American Alzheimer’s Association website are the

most common symptoms of AD as shown in other updated

studies (Casanove, Starkstein, & Jellinger, 2011; Moss et

al., 2012).

An overall index was calculated by summing the num-

ber of symptoms rated as being related to AD. Total

scores ranged from 0 to 13, with higher scores indicating

more perceptions that the symptoms are related to AD.

For the other dimensions (causes, timeline, conse-

quences, control, coherence, cyclical timeline, and emo-

tional representations), participants were asked to indicate

their degree of agreement on a five-point scale (1 D strongly disagree, 5D strongly agree) with five statements concerning acute/chronic timeline (e.g., ‘AD lasts a short

time’), four items about the cyclical nature of AD (e.g.,

‘AD is very unpredictable’), six items about severe conse-

quences of AD (e.g., ‘AD is a serious condition’), five

items about the coherence of AD (e.g., ‘I don’t understand

AD’), and six items dealing with emotions caused by AD

(e.g., ‘AD makes me feel angry’). The control domain

consisted of 11 items which, according to Moss-Morris et

al. (2002), loaded onto two factors: six items representing

beliefs about personal control (e.g., ‘The course of AD

depends on the patient’) and five items representing

beliefs about treatment (e.g., ‘There is very little that can

be done to improve AD’). The causal domain consisted of

18 attribution items which, according to Moss-Morris et

al. (2002), were divided into four sub-dimensions: six

items related to psychological attributes (e.g., ‘stress or

worry’), seven items dealt with risk factors (e.g., heredity,

diet), three items related to immunity (e.g., ‘a germ or

virus’), and two items reflected accident or chance (e.g.,

‘chance or bad luck’). Overall indices for all dimensions

were calculated as the mean of the items per scale, with a

higher score indicating a stronger perception of the attri-

butions assessed. The internal consistency of these sub-

scales was modest to excellent (Cronbach’s a ranged from .68 to .90).

The Maslach Burnout Inventory was developed by

Maslach and Jackson (1981) to assess burnout in human

service workers. The inventory consists of three sub-

scales: emotional exhaustion, depersonalization, and lack

of personal accomplishment (Maslach & Jackson, 1996).

The participants were asked to rate their degree of agree-

ment with each item on a seven-point scale from 1 D completely disagree to 7 D completely agree. The mean score was calculated; a high score indicated high levels of

burnout. The questionnaire was translated into Hebrew

with internal consistency (a) that ranged from .89 to .90 for the emotional exhaustion subscale, .79�.89 for the depersonalization subscale, and .71 for the lack of per-

sonal accomplishment subscale (Cohen & Gagin, 2005;

Shinan-Altman & Cohen, 2009). In the present study, the

respective figures for these subscales were .86, .77, and

.79.

The work stressors questionnaire was developed by

Gonzalez-Roma and Lloret (1998) and included role con-

flict and ambiguity scales and role overload scale. The

role overload subscale was measured by one item. All

items were scored on a five-point scale ranging from 1 D completely disagree to 5 D completely agree. The items’ means were calculated; a high score in each subscale indi-

cated high levels of role conflict, role ambiguity, or role

overload. The role conflict subscale consisted of five

items, with internal consistency levels (a) of .85 and .61 in previous studies (Gonzalez-Roma & Lloret, 1998;

Shinan-Altman & Cohen, 2009), and .73 in the present

study. The role ambiguity subscale consisted of four items

with internal consistency levels (a) of .78 and .83 in pre- vious studies (Gonzalez-Roma & Lloret, 1998; Shinan-

Altman & Cohen, 2009), and .77 in the present study.

The coping strategies scale was developed by Carver,

Scheier, and Weintraub (1989) to assess coping strategies.

In the current study, we used the COPE, a 30-item Hebrew

version questionnaire, based on the scale developed by

Carver et al. (1989), with the items divided into 15 coping

strategies � two items per strategy. The participants were asked to rate the extent to which they used each coping

option in stressful situations. A rating scale ranging from

0 to 3 was used; 0 D not at all and 3 D a great deal. A fac- tor analysis with Varimax rotation showed a two-factor

structure, as reported by Carver et al. (1989): Problem-

focused scale, included six coping strategies (a D .66); emotion-focused scale included four coping strategies

(a D .67). Knowledge about AD was assessed using a 30-item

true/false AD knowledge test assessing the nature, symp-

toms, course, etiology, diagnosis, and treatment of AD

(Carpenter, Balsis, Otilingam, Hanson, & Gatz, 2009). An

overall index of knowledge was calculated by summing

the correct answers. Total scores ranged from 0 to 30,

with a higher score indicating a greater knowledge of AD.

The test has adequate reliability (test�retest and internal consistency) and validity (content, predictive, concurrent,

and convergent) (Carpenter et al., 2009).

The questionnaire was translated from English to

Hebrew and then retranslated into English. The original

and the translated versions were compared to ensure accu-

racy of the content. The internal reliability of the ques-

tionnaire in the current study was modest (Cronbach’s a D .60).

Sociodemographic and professional details included

gender (female/male), age, marital status (single/married/

354 S. Shinan-Altman et al.

widowed/divorced/separated), number of children, years

of education, place of birth (Israel, Asia/Africa, Europe/

America, other), place of work (medical center, nursing

home), profession (social worker, nurse), years of tenure

in profession and income (above average, average, or

below average income in Israel). Average monthly Israel-

i’s wages is 9330 NIS (about 2600$) as published by the

Israel Central Bureau of Statistics (2012).

Procedure

A pre-test was conducted with 10 social workers and 10

nurses to test the clarity of the items, and to identify diffi-

culties in understanding them. Next, managers of five

medical centers and 40 managers of nursing homes were

asked for permission to include the institute’s workers in

the study; five nursing home managers refused to partici-

pate, stating that the questionnaire was too long. The

study was approved by the institutional review board in

each of the five medical centers and by the University of

Haifa’s Ethics Committee for the nursing homes. Partici-

pants were included in the study after signing a consent

form. Next, in order to protect the participants’ privacy,

interviews were coded anonymously to a password-

protected file.

Statistical analysis

Descriptive statistics were used to describe participants’

demographic, personal, and professional characteristics,

as well as the research variables. Pearson correlations

were used to assess the associations between the main

research variables (identity, psychological attributes, risk

factors, immunity, accident or chance, timeline, conse-

quences, self-control, treatment-control, cyclical timeline,

illness coherence, emotional illness representations, prob-

lem-focused coping strategies, role conflict, role ambigu-

ity, role overload, knowledge about AD, emotional

exhaustion, depersonalization, and lack of personal

accomplishment). These analyses were performed using

SPSS software (17th version). Structural equation model-

ing (SEM) using AMOS 4 (Arbuckle, 2005, 2009) was

used to assess the fit to a model with a single latent

variable.

SEM was chosen over measured variable path analysis

because it corrects for measurement errors, allows the

investigation of complex models, and enables tests of

direct and indirect effects.

According to SEM, the fit of the data to the model is

assessed by varying indices of fit, including an overall

chi-square to assess the degree of fit between the esti-

mated and the observed variables, for which a lower value

indicates a better fitting model, as well as the normed fit

index (NFI), Tucker-Lewis index (TLI), and comparative

fit index (CFI), for which a score of .90 is the lowest

acceptable value (Bentler, 1990). Finally, the root mean

square error of approximation (RMSEA), for which a

value equal to or less than .05 indicates a good fit of the

model, assesses the magnitude and the direction of the

parameter estimates (Arbuckle, 2011).

Results

Table 2 shows the means (SDs) and ranges of the study

variables. The mean score of burnout was relatively high

for the dimensions of emotional exhaustion and deperson-

alization, while the mean score of lack of personal accom-

plishment was relatively low. Participants believed that

the main causes of AD stem from risk factors followed by

psychological factors. Accident or chance and immunity

were reported at lower levels as causes of AD. Most of

the participants perceived AD as being a chronic, cyclical

disease, with severe consequences; they also believed that

AD could be controlled through treatment, rather than by

personal control, and perceived their understanding of the

disease as moderate. The mean score of emotional repre-

sentations was relatively high. Regarding work stressors

(role ambiguity, role conflict, and role overload), mean

scores indicated a medium level of experienced stressors.

Finally, participants used more emotion-focused coping

strategies than problem-focused coping strategies.

According to Table 3, a few moderate statistical sig-

nificant associations were found between AD illness rep-

resentations and burnout. Emotional exhaustion and

depersonalization were positively associated with percep-

tions of AD as being cycle (cyclical timeline) and as

caused by psychological factors, risk factors, accident or

chance with believing that AD could be controlled

through personal control and with emotional representa-

tions. Lack of personal accomplishment was positively

associated with cyclical timeline and with emotional rep-

resentations. Finally, AD emotional representations were

positively associated with emotional exhaustion, deper-

sonalization, and lack of personal accomplishment,

namely, the higher the participants expressed negative

Table 2. Means, SDs, and ranges of study variables.

Variables Mean SD Range

AD representations

Psychological attributes 2.60 .50 1.00�5.00 Risk factors 2.90 .60 1.00�4.70 Immunity 1.80 .61 1.00�4.50 Accident or chance 2.00 .51 1.00�5.00 Timeline 4.25 .67 2.40�5.00 Consequences 4.15 .65 1.67�5.00 Self-control 2.50 .52 1.50�4.00 Treatment control 3.60 .60 1.50�5.00 Cyclical timeline 2.56 .51 1.00�4.00 Illness coherence 2.28 .60 1.00�4.75

Emotional illness representations 2.67 .64 1.00�5.00 Emotion-focused coping strategies 1.73 .39 1.00�2.88 Problem-focused coping strategies 3.12 .50 1.50�4.00 Role conflict 2.52 .64 1.20�4.60 Role ambiguity 1.96 .61 1.00�4.00 Role overload 2.74 .76 1.00�5.00 Knowledge about AD 11.22 2.71 12.00�29.00 Burnout

Emotional exhaustion 3.04 1.12 1.00�6.56 Depersonalization 1.80 .96 1.00�6.60 Lack of personal accomplishment 2.91 1.23 1.00�6.12

Aging & Mental Health 355

emotional representations about AD, the higher the level

of burnout they experienced.

Examining the associations between coping strategies

and burnout dimensions revealed that stronger associa-

tions were found between emotion-focused coping strate-

gies and burnout, compared to problem-focused coping

strategies. This indicates that the higher the use of emo-

tion-focused coping strategies, the higher the experience

of burnout was. In addition, statistically significant and

positive associations were found between role overload,

role conflict, and role ambiguity for the three dimensions

of burnout. In other words, the more participants reported

on role conflict, role ambiguity, and role overload, the

more they reported on depersonalization, emotional

exhaustion, and a lack of personal accomplishment.

Finally, moderate negative associations were found

between knowledge about AD and emotional exhaustion,

meaning that the more the participants knew about AD,

the less emotional exhaustion they expressed.

Following these analyses, SEM was conducted to

identify the variables associated with burnout. Only varia-

bles found to be significantly associated with at least one

of the burnout dimensions were included in the structural

equation analysis (knowledge about AD, identity, casual

attributions of psychological factors, risk factors, immu-

nity, accident or chance, timeline, consequences, personal

control, treatment control, illness coherence, cyclical

timeline, emotional illness representations, problem-

focused coping, emotion-focused coping, emotional

exhaustion, depersonalization, and personal

accomplishment).

The trimmed model showed a good fit of the data

[x2 (n D 591) D 866.281, p D .000; CMIN/DF D 1.47;

NFI D .80; TLID .91, RMSEAD .04]. The general model explained 51% of the emotional exhaustion variance,

37.7% of the depersonalization variance, 22.6% of the

lack of personal accomplishment variance, and 32.2% of

the emotion-focused coping strategies variance. Figure 1

shows the relations found in the final model (standardized

b) with boxes for observed variables and oval boxes for

latent variables. According to the figure, positive associa-

tions were found between cognitive illness representations

(risk factors, cyclical timeline, and illness coherence) and

various dimensions of burnout. This means that illness

representations were associated in a different manner with

each of the burnout dimensions. Moreover, cognitive ill-

ness representations were associated directly with burn-

out, rather than through coping strategies; meaning, there

is a direct connection between cognitive illness represen-

tations and burnout. Regarding emotion-focused coping

strategies, according to Figure 1, these strategies are a

major factor linking emotional illness representations and

emotional exhaustion, depersonalization, and a lack of

personal accomplishment.

Study 2: qualitative method

Method

Participants

Qualitative methodology is based on small purposeful

samples that provide an in-depth understanding of the

phenomenon under investigation (Dekel, Goldblatt, Kei-

dar, Solomon, & Polliack, 2005). Thus, participants were

chosen according to theoretical sampling principles;

meaning, we selected subjects with demonstrated

Table 3. Correlations between AD representations, coping strategies, role variables, and knowledge about AD with burnout subscales (emotional exhaustion, depersonalization, lack of personal accomplishment).

Emotional exhaustion Depersonalization Lack of personal accomplishment

AD representations

Identity ¡.05 ¡.01 .03 Causes

Psychological attributes .17�� .16� .04 Risk factors .17�� .16� .01 Immunity .10 .09 .02

Accident or chance .16�� .16� .08 Timeline .03 ¡.06 ¡.04 Consequences .12 ¡.11 ¡.06 Self-control .17�� .17�� .09 Treatment control ¡.01 ¡.09 ¡.03 Cyclical timeline .15� .12� .11�

Illness coherence .01 .08 ¡.03 Emotional illness representations .32�� .21�� .18��

Emotion-focused coping strategies .38�� .38�� .25��

Problem-focused coping strategies .04 ¡.07 ¡.01 Role conflict .25�� .25�� .14�

Role ambiguity .16�� .18�� .32��

Role overload .33�� .18�� .16�

Knowledge about AD ¡.16�� .08 .11 �p < .05; ��p < .01.

356 S. Shinan-Altman et al.

theoretical relevance to the phenomenon under study

(Shkedi, 2003). The participants were eight professionals

who took part in the quantitative research. The average age

was 41 (SD D 10.5; range D 27�59) and the mean profes- sional tenure was 14 years (SD D 9.4, range D 1�35).

Measures

Data were gathered by means of semi-structured in-depth

interviews. This tool is a flexible framework for examin-

ing the meanings of behaviors, feelings, beliefs, and per-

ceptions (Shkedi, 2003). The purpose of the interview was

to explore participants’ experience and their perceptions

regarding AD, as well as their experience of burnout. An

interview guide was constructed including key questions

to simulate primary descriptions and concepts that partici-

pants might use to describe their reality regarding the phe-

nomenon under study (Shkedi, 2003). The interview

guide domains included perceptions of AD characteristics

and their impact on participants as professionals, and the

impact of AD perceptions on burnout.

Procedures

The research protocol was approved by the institutional

review board of the medical centers participating in the

study and by the University of Haifa’s Ethics Committee

for nursing home professionals. The sample size was

guided by the principle of ‘theoretical saturation’; mean-

ing, the interviews continued until no new information

came up. All professionals agreed to take part in the in-

depth interviews. Before starting the interview, partici-

pants received an explanation regarding the study’s aim

and signed a consent form, stating their agreement to par-

ticipate. Each interview was tape-recorded and later tran-

scribed and content analyzed.

Data analysis

The content analysis process was based on interview tran-

scripts according to three steps. In the first stage, a re-

reading of all interviews was performed, in order to thor-

oughly understand the thoughts, perceptions, and emo-

tions of participants regarding their work with AD

patients. In the second stage, ‘units of meaning’ were

identified in each interview. Units of meaning refers to a

sentence or several sentences that can stand alone, and yet

are structured and meaningful within the context of the

research topic (Shkedi, 2003). In the third stage, concepts

and units of meaning with common denominators were

identified and formulated into ‘central themes.’ These

themes included the abstraction and unification of all find-

ings (Evans, Patterson, & O’Malley, 2001). Data were

carefully coded by the three researchers of the current

study and verified by an external coder. Differences

between the coders were resolved through discussion until

Figure 1. The study’s model.

Aging & Mental Health 357

consensus was reached (e.g., whether quotations reflected

the theme title). All interviews were translated into

English.

Results

An analysis of the interviews showed that AD charac-

teristics were perceived by participants as affecting

them on both personal and professional levels, and as

leading to burnout. Participants expressed negative feel-

ings towards AD and feelings of self-blame regarding

the care of these patients, and explicitly stated that such

feelings had led to personal burnout. Three main themes

were identified:

Theme 1: ‘Decrease in emotional energy’

Participants described a lack of energy, feelings of

overload, and the deterioration of emotional resources,

leading to emotional exhaustion. This depletion was

explained mainly as a consequence of participants’ con-

stant need to give of themselves, as an integral part of their

role. In addition, participants also felt that the emotional

energy flowed in one direction only – from the professio-

nals to their patients � and that they (the professionals) did not receive any feedback from their patients. Thus, partici-

pants described a direct effect of AD characteristics on

their physical and emotional functioning.

. . . Sometimes even one AD patient can depress me. Since I experience so much sadness and frustration, I have no motivation. . . lack of communication with them frustrates me and angers me. . . I am more impatient at home. . ., I have no strength, I don’t feel like talking, everything annoys me; I just want to go to bed and sleep until tomorrow. . .[K.]

. . . In situations where you internalize the pain and sorrow of others you also feel burnout. . .You show your patients strength and optimism, a kind of hope. . .but we are human beings. . .it is also difficult for me. . .It hurts me to see them (the Alzheimer’s patients)..a heavy emotional bur- den. . . [F.]

Theme 2: ‘Depersonalization of the patient’

Some participants expressed a feeling of cognitive dis-

tance from their patients. However, it seems that they felt

guilty about this distance on both professional and per-

sonal levels. On the professional level, the essence of the

professionals’ role is to treat the problems of their various

patients, regardless of the characteristics of the patients’

disease. On the personal level, people help each other.

Feeling a sense of distance from their patients’ problems

led to professional and personal dilemmas, which caused

some guilt, on the one hand, but also allowed them to

express these difficult emotions.

. . . I accept the definition that they [Alzheimer’s patients] do not have faces. I mean, you can’t know what they’re thinking and feeling, what he thinks . . . this is difficult for me because I connect to people through their stor- ies. . .because of this, it’s easier for me to treat them dif- ferently, without emotion, without involvement. . .I can’t believe I’m saying this. . . [U.]

. . .They lose their humanity, they are like a shadow of a human being, can’t decide what they want, can’t express what they feel, like an animal who can’t say what hurts. . . it’s not nice to say . . . there is a person in there, but there is no normal communication. . .that’s how I see them and it helps me cope . . . . I have also become less tolerant, less sensitive . . . [A.]

Theme 3: ‘Sadness at perceiving loss/deficits in

patients’

Participants were aware that they experienced a vari-

ety of emotions � positive and negative feelings � towards AD patients, such as sadness and compassion

along with anger and frustration. However, in complicated

situations, such as when they witnessed a grandfather who

did not recognize his young grandson, they broke down

emotionally. Participants found these situations as diffi-

cult to accept and expressed feelings such as fear, anger,

and anxiety.

. . . I can handle everything . . . give medications, clean, feed. . .but to hear . . . Avner (a pseudonym) cursing his little grandson � that, I couldn’t handle . . .Yesterday, I was angry with him. . . I told him � Avner, its Dan, your grandchild. . . .but he yelled at him and told him to go away. . . afterwards, I felt sorry . . . .first for Dan, and then for Avner. . . [S.]

Discussion

Professionals’ illness representations regarding their

patients’ disease are one of the factors that may enhance a

deep and broader understanding of professional burnout;

however, these factors have not yet been studied

sufficiently.

The current study’s aim was to assess the connection

between AD perceptions and burnout, using a sample of

nurses and social workers caring for patients with AD.

Our results showed that the most positive, strong, and con-

sistent associations were found between emotional repre-

sentations and burnout. Similar findings were reported by

Todd and Watts (2005), who, in a study conducted among

nurses and psychologists, demonstrated that negative

emotional responses towards patients who have memory

decline were related to a higher experience of burnout

among professionals.

Our findings regarding the positive associations

between emotional representations and burnout strengthen

the theoretical structure of the SRM (Leventhal et al.,

1980) by stressing the simultaneous occurrence of two

separate processes, as demonstrated in the model: the pro-

duction of cognitive illness representations and the pro-

duction of emotional representations and their differential

impact on the outcome variable. Emotional representa-

tions were consistently associated with the three dimen-

sions of burnout, while only some of the cognitive illness

representations (cyclical timeline, risk factors, and illness

coherence) were inconsistently associated with the three

dimensions of burnout.

Furthermore, the finding that the most positive, strong,

and consistent associations were found between emotional

illness representations and burnout has theoretical and

358 S. Shinan-Altman et al.

practical implications. Theoretically, many previous stud-

ies that examined the SRM (Leventhal et al., 1980) did

not include emotional illness representations. Thus, our

study, which examined the SRM using all of its compo-

nents, supports the argument that the model should be

examined as a whole. Practically, professional burnout

programs should examine professionals’ emotional reac-

tions to their patients’ illnesses by exploring whether they

experience anger, fear or anxiety towards AD patients and

how are these emotions influence the quality of care they

provide. It is interesting to see that only a few dimensions

of cognitive illness representations were associated with

burnout: perceiving AD as cyclical was associated with

the experience of emotional exhaustion, depersonaliza-

tion, and a lack of personal accomplishment. Risk factors

(e.g., heredity, diet) as causal attributes were associated

with emotional exhaustion, and illness coherence was

associated with a lack of personal accomplishment.

According to this finding, professionals’ perceptions of

their patients’ disease are an integral part of the daily care

they provide, and a significant factor in the formation of

professional burnout. Furthermore, according to our

model, cognitive illness representations were associated

directly with burnout, without the mediation of coping

strategies. This finding is unique because, to the best of

our knowledge, the existing theories of professional burn-

out do not demonstrate a direct association between ill-

ness perceptions and burnout (Dierendonck, Wilmar,

Schaufeli, & Buunk, 2001; Maslach & Leiter, 2005).

These theories (for example, those of Dierendonck et al.,

2001; Maslach & Leiter, 2005) focus mainly on the per-

sonal and organizational factors that influence the forma-

tion of professional burnout.

Despite the interesting and important findings of the

quantitative part of our research, these findings alone did

not provide us with a full understanding of the associa-

tions between cognitive illness representations and burn-

out. The qualitative part, however, provided us with a

more thorough appreciation of the phenomenon under

study. According to the qualitative interviews, partici-

pants invested emotional energy in the care of their AD

patients, but since they felt that they did not receive any

positive feedback from their patients, it led them to feel

emotionally drained. Thus, participants described desper-

ation, a lack of energy, and feelings of emotional over-

load, which are manifestations of emotional exhaustion

(Maslach & Jackson, 1996). Furthermore, participants’

negative perceptions of care led to feelings of a lack of

personal achievement and developing a sense of failure,

manifested in their expressions of low self-esteem and a

low sense of capability as regards their choice of profes-

sion and the performance of their duties, clear expressions

of a lack of personal accomplishment (Maslach & Jack-

son, 1996).

However, participants also reported positive experien-

ces associated with the care of AD patients, since the care

they provided gave them a sense of professional meaning.

Interestingly, the perception that care had meaning for

them also led to depersonalization, another dimension of

burnout (Maslach & Jackson, 1996). Perhaps, out of the

fear of getting too close to patients, both physically and

emotionally, participants tended to form a cognitive dis-

tance between themselves and their AD patients. This

cognitive and emotional distance allowed them to protect

themselves from involvement and reduce the emotional

consequences, which often arises as a result of the nature

of their duties.

According to the in-depth interviews, representations

of emotional illness included, on the one hand, the feeling

that caring for AD patients is rewarding. On the other

hand, representations of negative emotional illness

towards AD were also experienced, which led caregivers

to emotional exhaustion and a sense of a lack of personal

accomplishment. Participants referred primarily to the

fact that, unlike other diseases, in which verbal interaction

with patients exists and feedback can be given regarding

the care they provide, caring for AD patients, especially

in advanced stages, lacks this interaction and feedback,

especially due to the behavioral challenges and memory

problems of AD patients. This lack of possibility to value

the care they provide often leads professionals to experi-

ence feelings of despair and frustration.

Overall, the qualitative research findings demonstrated

that the experience of taking care of AD patients involved

cognitive and emotional perceptions, which led to burn-

out. Two main explanations can illustrate why the quanti-

tative research findings only partially supported the

perception that there is an association between profes-

sionals’ illness representations and burnout. First, the

qualitative research findings revealed that participants

nurtured additional perceptions and feelings regarding the

care of AD patients that they could not express through

the structured IPQ-R (Moss-Morris et al., 2002). Second,

the qualitative research participants referred to specific

and complicated situations related to AD characteristics.

These situations exhausted participants, both cognitive

and emotionally, and led them to experience burnout.

Moreover, in the quantitative research, there were no case

studies of care situations, but only descriptions of clinical

characteristics of AD. In conclusion, the combination of

quantitative and qualitative methodologies has given us a

more thorough understanding of the relationship between

cognitive and emotional illness representations and burn-

out, since the clinical characteristics of the disease

together with the perceptions of the experience of care of

AD patients provided a more exhaustive explanation for

the creation of burnout. Therefore, it is recommended to

expand the IPQ-R (Moss-Morris et al., 2002) by adding

some open questions in order to receive more information

about the experience of care among professionals, which

cannot be expressed by the closed questionnaire.

Four main limitations regarding our study should be

noted. First, the use of a convenience sample does not

allow us to generalize the results, nor does it provide an

accurate representation of all social workers and nurses.

Second, the study was confined to workers in one country,

which limits the generalizability of the findings. Further

studies in other countries are called to address this limita-

tion. Third, the answers given could have been affected

by social desirability, while those who declined to

Aging & Mental Health 359

participate might be suffering more burnout. Fourth,

symptoms of AD were adapted from the American

Alzheimer’s Association website although the American

Alzheimer’s Association has been criticized for its over-

medicalization of brain aging and its focus on fundraising

for an absolute cure. However, the symptoms of AD as

shown by the American Alzheimer’s Association are the

most common symptoms of dementia as demonstrated in

other studies (Casanove et al., 2011; Moss et al., 2012).

Despite these limitations, our study is the first to

examine the connection between illness representations

and burnout. Its findings expand the limited body of exist-

ing knowledge regarding professionals’ illness representa-

tions. Practically, the findings present a mechanism for

recognizing professionals who are at risk of developing

burnout, namely, those who have AD illness representa-

tions (such as perceiving AD as being contagious) and

expressing negative emotions (such as fear and depres-

sion) towards the care of AD patients. They also suggest

new directions to improve the design of programs that

aim to reduce professional burnout: intervention programs

should address not only the organizational and instrumen-

tal factors contributing to burnout, but also focus on pro-

fessionals’ cognitive and emotional reactions towards

their patients’ illness, with reference to the specific and

unique characteristics of social workers and nurses. These

intervention programs should include learning and enrich-

ment of intervention skills and knowledge about AD. Fur-

thermore, it is also suggested that these programs would

give professionals the choice to see AD not as a loss of

self but as a natural stage of life (Whitehouse & George,

2008) with an approach to care that looks far more to

human than to medical solutions (Kitwood, 1997). Fol-

lowing these programs, an improvement in the quality of

care would defiantly improve the quality of life of AD

patients, who often experience directly the implications of

their providers’ burnout

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Alzheimer’s Association. (2009). Ten warning signs of Alzheimer’s disease. Retrieved February 19, 2009, from http://www.alz.org/AboutAD/10signs.htm

Arbuckle, J.L. (2005). AMOS 6 user’s guide. Meadville, PA: AMOS Development Corporation.

Arbuckle, J.L. (2009). AMOS 18 user’s guide. Chicago, IL: SPSS.

Arbuckle, J.L. (2011). AMOS 19 user’s guide. Meadville, PA: AMOS Development Corporation.

Bentler, P.M. (1990). Comparative fit indexes in structural mod- els. Psychological Bulletin, 107, 238�246.

Carpenter, B.D., Balsis, S., Otilingam, P.G., Hanson, P.K., & Gatz, M. (2009). The Alzheimer’s disease knowledge scale: Development and psychometric properties. The Gerontolo- gist, 25, 1�12.

Carver, C.S., Scheier, M.F., & Weintraub, J.K. (1989). Assessing coping strategies: A theoretically based approach. Journal of Personality and Social Psychology, 56, 267�283.

Casanove, M.F., Starkstein, S.E., & Jellinger, K.A. (2011). Clinico- pathological correlates of behavioral and psychological symp- toms of dementia. Acta Neuropathologica, 122, 117�135.

Cimiotti, J.P., Aiken, L.H., Sloane, D.M., & Wu, E.S. (2012). Nurse staffing, burnout, and health care-associated infection. American Journal of Infection Control, 40, 486�490.

Cohen, M., & Gagin, R. (2005). Can skill-development training alleviate burnout in hospital social workers? Social Work in Health Care, 40(4), 83�89.

Dekel, R., Goldblatt, H., Keidar, M., Solomon, Z., & Polliack, M. (2005). Being a wife of a veteran with posttraumatic stress disorder. Family Relations, 54(1), 24�36.

Dierendonck, D.V., Wilmar, B., Schaufeli, A., & Buunk, B.P. (2001). Burnout and inequity among human service profes- sionals: A longitudinal study. Journal of Occupational Health Psychology, 6(1), 43�52.

Edberg, A.K., & Hallberg, I.R. (2001). Actions seen as demand- ing in patients with severe dementia during one year of inter- vention. Comparison with controls. International Journal of Nursing Studies, 38(3), 271�285.

Evans, M.J., Patterson, M., & O’Malley, L. (2001). Bridging the direct marketing�direct consumer gap: Some solutions from qualitative research. Qualitative Marketing Research: An International Journal, 4(1), 4�17.

Fillenbaum, G., Heyman, A., Peterson, B., Pieper, C., & Wei- man, A.L. (2000). Frequency and duration of hospitalization of patients with AD based on Medicare data: CERAD XX. Neurology, 54, 740�743.

Fong, T.G., Jones, R.N., Marcantonio, E.R., Tommet, D., Gross, A.L., Habtemariam, D., . . . Inouye, S.K. (2012). Adverse outcomes after hospitalization and delirium in persons with Alzheimer disease. Annals of Internal Medicine, 156(12), 848�856.

Garrosa, E., Rainho, C., Moreno-Jim�enez, B., & Monteiro, M.J. (2010). The relationship between job stressors, hardy per- sonality, coping resources and burnout in a sample of nurses: A correlational study at two time points. International Jour- nal of Nursing Studies, 47, 205�215.

Gonzales-Roma, V., & Lloret, S. (1998). Construct validity of Rizzo et al. (1970). Role conflict and ambiguity scales: A multisample study. International Association of Applied Psychology, 47(4), 535�545.

Hoyer, W.J., & Roodin, P.A. (2009). Adult development and aging (6th ed.). Boston, MA: McGraw-Hill.

Insel, K.C., Meek, P.M., & Leventhal, H. (2005). Differences in illness representation among pulmonary patients and their providers. Journal of Health Psychology, 10(1), 147�162.

Johnson, R.B., Onwuegbuzie, A.J., & Turner, L.A. (2007). Toward a definition of mixed methods research. Journal of Mixed Methods Research, 2, 112�133.

Jourdain, G., & Chenevert, D. (2010). Job demands � resources, burnout and intention to leave the nursing profession: A questionnaire survey. International Journal of Nursing Stud- ies, 47(6), 709�722.

Kitwood, T. (1997). Dementia reconsidered: The person comes first. Philadelphia, PA: Open University Press.

Kwok, T., Lam, K.C., Yip, A., & Ho, F. (2011). Knowledge of dementia among undergraduates in the health and social care professions in Hong Kong. Social Work in Mental Health, 9, 287�301.

Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common-sense representations of illness and danger. In S. Rachman (Ed.), Contributions to medical psychology (pp. 7�30). New York: Pergamon.

Mackenzie, C.S., & Peragine, G. (2003). Measuring and enhanc- ing self-efficacy among professional caregivers of individu- als with dementia. American Journal of Alzheimer’s Disease and Other Dementias, 18(5), 291�299.

Maslach, C.S., & Jackson, S.E. (1981). The measurement of experienced burnout. Journal of Occupational Behavior, 2, 99�113.

360 S. Shinan-Altman et al.

Maslach, C.S., & Jackson, S.E. (1996). Maslach’s burnout inven- tory � human services survey (MBI-HSS). In C. Maslach & S.E. Jackson (Eds.), MBI manual (3rd ed.) (pp. 3�18). Palo Alto, CA: Consulting Psychologists.

Maslach, C.S., & Leiter, M.P. (2005). Banishing burnout: Six strategies for improving your relationship with work. San Francisco, CA: Jossey Bass.

Maslach, C.S., Schaufeli, W.B., & Leiter, M.P. (2001). Job burn- out. Annual Review of Psychology, 52, 397�422.

Meissner, H.I., Creswell, J.W., Klassen, A.C., Plano Clark, V.L., & Smith, K.C. (2011). Best practices for mixed methods research in the health sciences. Washington, DC: National Institutes of Health, Office of Behavioral and Social Scien- ces Research.

Moss, A.S., Wintering, N., Roggenkamp, H., Khalsa, D.S., Waldman, M.R., Monti, D., . . . Newberg, A.B. (2012). Effects of 8-weekmeditation program on mood and anxiety in patients with memory loss. The Journal of Alternative and Complementary Medicine, 18(1), 48�53.

Moss-Morris, R., Weinman, J., Petrie, K.J., Horne, R., Cameron, L., & Buick, D. (2002). The revised illness perception questionnaire (IPQ-R). Psychology and Health, 17(1), 1�16.

Orengo, C.A., Khan, J., Kunik, M.E., Snow, A.L., Morgan, R., Steele, A., . . . Graham, D.P. (2008). Aggression in individu- als newly diagnosed with dementia. American Journal of Alzheimer’s Disease and Other Dementias, 23(3), 227�232.

Pekkarinen, L., Sinervo, T., Elovainio, M., Noro, A., Finne-Sov- eri, H., & Leskinen, E. (2006). Resident care needs and work stressors in special care units versus non-specialized long-term care units. Research in Nursing and Health, 29 (5), 465�476.

Pines, A., & Maslach, C. (1978). Characteristics of staff burnout in mental health settings. Hospital and Community Psychia- try, 29(4), 233�237.

Rossler, U. (2012). Stress, burnout, and job dissatisfaction in mental health workers. European Archives of Psychiatry and Clinical, 262(2), 65�69.

Schmidt, S.G., Dichter, M.N., Bartholomeyczik, S., & Hassel- horn, H.M. (2014). The satisfaction with the quality of dementia care and the health, burnout and work ability of nurses: A longitudinal analysis of 50 German nursing homes. Geriatric Nursing, 35, 42�46.

Shinan-Altman, S., & Cohen, M. (2009). Nursing aides’ attitudes to elder abuse in nursing homes: The effect of work stressors and burnout. The Gerontologist, 49(5), 674�684.

Shinan-Altman, S., Werner, P., & Cohen, M. (2014). Social workers’ and nurses’ illness representations about Alz- heimer disease: An exploratory study. Alzheimer Disease and Associated Disorders, 28, 73�78.

Shkedi, A. (2003). Words that try to touch: Qualitative research � theory and practice. Tel Aviv: Ramot-Tel Aviv University (Hebrew).

Todd, S.J., & Watts, S.C. (2005). Staff responses to challenging behaviour shown by people with dementia: An application of an attributional-emotional model of helping behaviour. Aging and Mental Health, 9(1), 71�81.

Van Bogaertp, P., Clarke, S., Willems, R., & Mondelaers, M. (2013). Nurse practice environment, workload, burnout, job outcomes, and quality of care in psychiatric hospitals: A structural equation model approach. Journal of Advanced Nursing, 69(7), 1515�1524.

Whitehouse, P.J., & George, D. (2008). The myth of Alzheimer’s. New York, NY: St. Martin’s Griffin.

Aging & Mental Health 361

Copyright of Aging & Mental Health is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.