burnout and health behaviors in health professionals from

1

Burnout and health behaviors in health professionals from seven European countries

Anna Alexandrova-Karamanova1,2, Irina Todorova1, Anthony Montgomery3,Efharis

Panagopoulou4,Patricia Costa

5, Adriana Baban

6,Asli Davas

7, Milan Milosevic

8, Dragan

Mijakoski9

1Health Psychology Research Center, Po Box 238, 1113Sofia, Bulgaria

2Department of Psychology, Institute for Population and Human Studies – Bulgarian Academy

of Sciences,Acad. G. Bonchev str., bl. 6, fl. 5/6, 1113 Sofia, Bulgaria

3Department of Education and Social Policy,University of Macedonia, 156 Egnatia Street, GR-

540 06 Thessaloniki, Greece

4Medical School, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece

5BusinessResearchUnit - InstitutoUniversitáriodeLisboa, Av. DasForc_asArmadas, Edificio

ISCTE-IUL, 1649-026Lisbon, Portugal

6Department of Psychology, Babes Bolyai University, 37, RepubliciiStreet, 400015Cluj-Napoca,

Romania

7Department of Public Health,Ege University, HalkSagligi AD 35100 Bornova,Izmir, Turkey

8School of Medicine, Andrija Stampar School of Public Health,University of Zagreb,

Rockefellerova 4, 10000Zagreb, Croatia

9 Institute of Occupational Health of RM, WHO Collaborating Center, II MakedonskaBrigada

43,Skopje, Republic of Macedonia

The research leading to these results has received funding from the European Union’s Seventh

Framework Programme [FP7-HEALTH-2009-single-stage] under grant agreement no. [242084].

Corresponding author: Anna Alexandrova-Karamanova, Department of Psychology, Institute for

Population and Human Studies – Bulgarian Academy of Sciences, Acad. G. Bonchev str., bl. 6,

fl. 5/6, 1113 Sofia, Bulgaria. E-mail: [email protected];tel. +359 2 979 30 29; fax:

+359 2 870 32 17.

2

Abstract

ObjectivesWithin an underlying health impairing process work stressors exhaust employees’

mental and physical resources and lead to exhaustion/burnout and to health problems, with

health-impairing behaviors being one of the potential mechanisms, linking burnout to ill

health.The study aims to explore the associations between burnout and fast food consumption,

exercise, alcohol consumption and painkiller use in a multinational sample of 2623 doctors,

nurses and residents from Greece, Portugal, Bulgaria, Romania, Turkey, Croatia and Macedonia,

adopting a cross-national approach.

MethodsData are part of theinternational cross-sectional quantitative ORCAB survey.The

measures included the Maslach Burnout Inventory and the Health Behaviors Questionnaire.

ResultsBurnout was significantly positively associated with higher fast food consumption,

infrequent exercise, higher alcohol consumption and more frequent painkiller use in the full

sample, and these associations remained significant after the inclusion of individual differences

factors like age, gender, staff position and unit tenure.Cross-nationalcomparisons showed

significant differences in burnout and health behaviors, and some differences in the statistical

significance and magnitude (but not the direction) of the associations between them.Health

professionals from Turkey, Greece and Bulgaria reported the most unfavorable experiences.

ConclusionsBurnout and risk-health behaviors among health professionals are important both in

the context of health professionals’ health and well-being and as factors contributing to medical

errors and inadequate patient safety. Organizational interventions should incorporate early

identification of such behaviors together with programs promoting health and aimed at the

reduction of burnout and work-related stress.

Keywordsburnout;health behaviors; health professionals; cross-national

3

Introduction

The burnout phenomenon is most often viewed as a syndrome, resulting from prolonged

job stress, including the dimensions of emotional exhaustion (feelings of being overextended and

depleted of one’s emotional and physical resources), depersonalization or cynicism (a negative,

callous, or excessively detached response to various aspects of the job), and reduced personal

accomplishment (feelings of incompetence and a lack of achievement and productivity at work)

(Maslach et al. 2001).Burnout is considered an important issue for a diverse range of

occupations.Since its initial conceptualization it is deemed particularly relevanttocare-giving and

human service occupations, which are interpersonally stressful and emotionally demanding

(Maslach et al. 1996).Burnout has been identified as a serious issue for nurses, physicians and

residents within a number of cultures and countries, among them the USA (Shanafelt, Boone et

al.2012), Canada (Boudreau et al. 2006), the UK (Sharma et al. 2008), the Netherlands (Prins et

al. 2010), Japan (Kanai-Pak et al. 2008), Brazil (Tironi et al. 2010), Germany (Schulz et al.

2009), France(Dreano-Hartz et al. 2015), as well as within countries in the South and South

Eastern Europe, e.g. Italy (Grassiand Magnani2000),Spain (Castelo-Branco et al. 2007),

Portugal(Maroco et al. 2016), Serbia (Putnik and Houkes2011), Romania (Popa et al. 2010),

Greece (Zis et al. 2014), Turkey (Demir et al. 2003).Burnout in healthcare professionals is a

globalconcern both due to its detrimental effect on their health and well-being and because of its

potential impact on quality and safety of patient care.

Comparative analyses of burnout and its determinants and outcomes have been conducted

among medical professionals from multiple nations.Within the International Psychiatry

Residents/ Trainee Burnout Syndrome studyburnoutis studied in respect to working conditions,

mental health and personality within European residents in psychiatry from 24 countries

(Jovanovic et al. 2009; Beezhold et al. 2010). The European General Practice Research Network

explores working conditions, health behaviors and others as factors for burnout among family

doctors from 12 European countries (Soler et al. 2008).Within the European NEXT study

violence from patients, different aspects of teamwork, full-time work,fixed night shifts and time

pressure were determinants of burnout in nurses from 10 European countries(Estryn-Behar et al.

2008), and high prevalence of burnout was a major risk factor for nurses’ intention to leave

(Estryn-Behar et al. 2007).The International Hospital Outcomes Study found that higher levels of

4

burnout were associated with lower quality of care among nurses from six countries (Poghosyan

et al. 2010). Other studies among nurseslink burnout to working conditions, social work

environments, subjective work stressors, personality characteristics, job satisfaction, and

intention to quit (Armstrong-Stassen et al. 1994; Turnipseedand Turnipseed 1997; Schaufeli and

Janczur 1994). All these international studies revealed significant cross-national differences in

burnout rates and burnout relationships with factors from the working environment, health and

personality variables, and quality of care.

One influential theoretical explanation of the development of burnout is provided by the

Job Demands-Resources model(Bakker and Demerouti 2007; Demerouti and Bakker 2011;

Demerouti et al. 2001). Within this theoretical framework all factors associated with job-related

stress are classified in two general categories: job demands (physical, psychological, social or

organizational aspects of the jobrequiring sustained physical/psychological effort or skills and

therefore associated with certain physiological/psychological costs) and job resources (functional

in achieving work goals, reducing job demands and the associatedcosts, stimulating personal

growth, learning, and development).It has been assumedthat an underlying health impairing

process stands in the development of burnout in which high job demands exhaust employees’

mental and physical resources and therefore lead to the depletion of energy (a state of

exhaustion) and to health problems. The Job Demands-Resources modelhas been evidenced to

have high predictive value for employee health and well-being(Bakker and Demerouti 2007;

Demerouti and Bakker 2011).

Reviews analyzing the relationship between burnout and the individual health of

employees indicate that burnout is associated with poor mental health (anxiety, depression,

somatic complaints), poor self-rated health, poor health behaviors, sleep disturbances, obesity,

infectious diseases, reproductive functions, musculoskeletal disorders, and with an increased risk

of diabetes, cardiovascular disease and cardiovascular-related events (Shirom et al. 2005;

Melamed et al. 2006; Shirom 2010a; Schaufeli 2003).It has been proposed that one of the

potential mechanisms, linking burnout to cardiovascular disease and additional physical

morbidities, might be through health-impairing behaviors (Melamedet al. 2006).Researchers

often assume that health-impairing behaviorsfunction as a denial or avoidance coping strategy

and are used for relieving distress in the short term (Shirom 2010a). Though a number of studies

havefound links between chronic stress and different poor health behaviors, theassumption that

5

health behaviors serve as coping is rarelyactually and explicitlyexplored(Park and Iacocca 2014).

While the studies on the relationships between chronic stress/burnout and health behaviors are

correlational in nature and conclusions about causality are limited, there is usually an assumption

that stress is the cause of poor health behaviors. But the reverse position could also be

considered: viewed as wellness practices, behaviors like exercise and healthy eating are studied

as resources withthe potential to prevent burnout in physicians (Shanafelt, Oreskovichel al. 2012;

Swetz et al. 2009; Weight et al. 2012).

The current study explores burnout in relation to several health-impairing behaviors

which have been associated with burnout in different populations, including health professionals

(e.g. Ahola et al. 2012; Goldberg et al. 1996; Gorter et al. 2000; Kristensenet al. 2005; Moustou

et al. 2010; Peterson et al. 2008; Soler et al. 2008): alcohol consumption, lack of exercise

(sedentary lifestyle), unhealthy eating behavior (fast food consumption) and self-medication with

painkillers (analgesics).

The aim of the study is to investigate the relationship between two dimensions of burnout

(emotional exhaustion and depersonalization) and alcohol consumption, physical exercise, fast

food consumption and self-medication through painkillers within a sample of health

professionals from seven South and South Eastern European countries, adopting a cross-national

approach.

Methods

The ORCAB study

The present analysis utilizes data from the international survey “Improving quality and

safety in the hospital: The link between organizational culture, burnout and quality of care

(ORCAB)”, funded by the European Commission within the 7th Framework Programme. The

ORCAB project includes cross-sectional quantitative and qualitative studies among physicians,

nurses and residents in university hospitals from seven South and South Eastern European

countries: Greece, Turkey, Portugal, Romania, Bulgaria, Croatia, and Macedonia. Its aim is to

benchmark the organizational and individual factors that impact quality of care and patient

safety, and design bottom-up interventions that both increase quality of care and health

6

professionals’ well-being. The theoretical model of ORCAB is based on the assumptions of the

Job Demands-Resources model and studies quality of care, patient safety and health

professionals’ well-being as outcomes, organizational culture and work demands as independent

variables and burnout and work engagement as potential mediators (Montgomery, Panagopoulou

et al. 2011).

Materials and procedure

Burnout

Burnout was measured with the Maslach Burnout Inventory – Human Services Survey

(MBI-HSS, Maslach et al. 1996). Validated versions of the MBI were used, where available, and

if such didn’t exist, the national research teams translated the inventory (and other measures used

within the study) following Harkness’ (2003) instrument translation procedure.

The present study was restricted to the Emotional Exhaustion (EE) and Depersonalization

(DP) dimensionsof burnout, building on the theoretical assumptions Job Demands-Resources

model. The authors of the model do not consider reduced personal accomplishment as a core and

separate dimension of burnout pointing to evidence that reduced personal accomplishment

correlates less strongly with EE and DP than they correlate with each other, has weaker

relationships with other variables, may be interpreted as a possible consequence of burnout and

is suggested to reflect a personality characteristic similar to self-efficacy (Demerouti et al. 2001;

Demerouti and Bakker 2008).This view is reflected in the Oldenburg Burnout Inventory

(Demerouti et al. 2003). The idea was reconfirmed in an analysis of the convergent validity of

four most widely used burnout measures, showing that burnout is best represented by two

underlying, strongly related factors, namely exhaustion and withdrawal (Qiao and Scaufeli

2011). It is concluded that burnout is best assessed through the EE and DP subscales of the

Maslach Burnout Inventory or through the OldenburgBurnout Inventory. The EE scale

comprised 9 items, and the DP scale – 5 items. Example items are: “I feel emotionally drained

from my work” (EE) and “I feel I treat some patients as if they were impersonal objects”(DP).

Participants responded in a 7-point Likert scale, ranging from 0 to 6, indicating the frequency in

which they had certain types of feelings in relation to their job from “never”to “every day”. The

7

subscales total scores range from 0 to 45 for EE and from 0 to 30 for DP. Within the present

study both subscales showed high reliability, with Cronbach’s alphas for the full sample being

0.91 for Emotional Exhaustion and 0.83 for Depersonalization, showing excellent and good

internal consistency, respectively. Within the national samples internal consistency coefficients

for the Emotional Exhaustion subscale ranged from 0.84 to 0.92, and for the Depersonalization

subscale Cronbach’s alphas ranged from 0.68 to 0.87(alpha fell below the 0.7 threshold only in

Croatia).

Health Behaviors

The Health Behaviors Questionnaire is a validated questionnaire that has been used

previously with healthcare professionals and civil servants (Moustou et al 2010; Nella et al.

2015;Tsiga et al. 2015). It measures protective health behaviors (e.g. exercise, healthy eating)

and risk health behaviors (e.g. drinking alcohol, smoking, unhealthy eating, self-medication).

The four behaviors, included in this analysis, were evaluated through a single item, regarding

frequency of fast food consumption per week (How many times in a week do you eat fast food?),

frequency of exercise per week (How many times do you exercise per week?), frequency of

alcohol consumption per week (On average, on how many days of the week do you drink

alcohol? (Regardless of the quantity), and frequency of painkiller use per week (On average, how

often per week do you use painkillers?). Participants responded by writing the number,

indicating the frequency of the behavior. The percentage of answers in the range 0-7 is as

follows: 97.9% for fast food, 99.8% for exercise, 100.0% for alcohol consumption and 99.6% for

painkiller use.

Procedure

After approval from the Administrative Boards and Ethical Committees of the

participating hospitals in each country, health professionals from the 7 countries filled in the

structured ORCAB questionnaire. Researchers and research assistants from the national teams

distributed the questionnaire in a hardcopy format. Participants were asked to complete and

return it sealed in an anonymous envelope.

8

Statistical analyses

All statistical analysis were done by means of IBM SPSS Statistics 22.As a first step the

associations between the burnout dimensions and health behaviors were assessed through

bivariate correlations, and items, measuring eating and substance (ab)use behaviors, which had

the strongest relationships with the burnout dimensions,were chosen for further exploration.

Tertiles were created for the two burnout dimensions and participants were classified as having

high, average or low levels of EE or DP in accordance with the identified cut-offs.Descriptive

statistics was used to define the means of emotional exhaustion, depersonalization and health

behaviors. One-way ANOVA was used to determine if there were significant differences

between the means of the burnout dimensions and the health behaviors in the groups of health

professionals from different countries. Post-hoc tests (Tukey HSD test) were conducted in order

to make pair-wise comparisons and determine which means were significantly different from

each other.To test the relationship between burnout dimensions and health behaviors, multiple

linear regressions were performed for every behavior as an outcome variable. For the whole

multinational sample, a total of eight hierarchical linear regressions were run (4 with Emotional

Exhaustion as an independent predictor and 4 with Depersonalization as an independent

predictor) with fast food consumption, exercise, alcohol drinking and painkiller use being the

dependent variables. The first block of independent predictorsincluded either EE or DP. In the

second block of variables the individual factors age,gender, staff position (employing two almost

evenly distributed groups of doctors/residents vs. nurses), and unit tenure were added. In the

third block of predictorsthe variable country was added. As a categorical variable with 7 levels

‘country’ was dummy coded with Greece being the reference category.As a final step, the

models including EE or DP in the first block and the individual factors in the second block were

tested separately in every one of the participating countries.

Results

Participants

9

Health professionals from university hospitals in Greece, Turkey, Portugal, Romania,

Bulgaria, Croatia, and Macedonia were recruited through convenience sampling. The total

number of participants was 2623, of whom 627 were doctors, 1431 were nurses and 565 were

residents (Table 1). The largest sample was collected in Greece (N=594), and the smallest in

Croatia (N=193). While nurses prevail in all national samples, some of the samples include very

small numbers (Portugal) or no residents at all (Croatia). At the same time the proportion of

residents is higher than the proportion of doctors in several other countries (Greece, Turkey and

Romania). Given the high proportion of nurses (93.6% of whom are women), women

predominate in the full sample (75.5% women vs. 24.5% men), with the two genders most

evenly represented in Greece and most disproportionately represented in Portugal and Romania.

The mean age of participants is 38.7 years (SD = 10.18 years), being the highest in Croatia and

Bulgaria (about 43-44 years), and the lowest in Portugal and Turkey (about 35 years). The

average reported tenure was 8 years and 10 months (SD = 9 years), with unit tenure ranging from

1 month to 45 years and 9 months. Higher average unit tenure compared to the other samples

were observed in Croatia (17 years and 11 months), and the shortest average unit tenure is

observed in Greece (6 years and 4 months).

-----------------------------------------------------------------------------------------------------------------

Insert Table 1 about here

-----------------------------------------------------------------------------------------------------------------

Burnout rates and frequencies of health behaviors

Tertiles for each of the two burnout scales (EE and DP) were created to allow

comparisons across countries for different prevalence categories. The participants were classified

as having high, average or low levels of emotional exhaustion or depersonalization in accordance

with the following cut-offs: EE: high ≥ 26, low ≤ 14; DP: high ≥ 7, low ≤ 2. In our European

sample high EE and especially high DP are defined at lower absolute scores than high EE and

high DP in Maslach’s normative sample of American medical professionals (EE: high ≥ 27, DP:

high ≥ 10) (Maslach, Jackson & Leiter, 1996).

10

Results showed that 31.9% of the health care professionals in the overall sample

experienced high EE and 33.2% experienced high DP. High levels of both burnout dimensions

were observed in 20.9% of the overall sample (Table 2).High emotional exhaustion amounted to

more than 50% in Turkey and to more than 35% in Greece and Bulgaria; in the group of the

remaining countries it ranged from 15% to 21%. High depersonalization was found in almost

60% of Turkish health professionals, in Greekhealth professionals it was above 40%, and in the

remaining countries it ranged from 16% to 23%. High levels of both burnout dimensions at the

same time were found in 45% of Turkish, 26% of Greek and 15% of Bulgarian health

professionals. In Portugal, Romania, Croatia and Macedonia proportions of health care

professionals with high levels of both burnout dimensions were about 9%.

-----------------------------------------------------------------------------------------------------------------


-----------------------------------------------------------------------------------------------------------------

The means for the overall sample are 20.68 (SD=12.52) for EE and 5.71 (SD=6.16) for

DP. In the separate countries the mean values for EE showed large differences ranging from

theaverage EE of15.61 in Romania to the high EE of27.88 in Turkey, and for DP ranging from

the average DP of2.94 in Macedonia to the high DP of9.99 in Turkey. Results from one-way

ANOVA and Tukey HSD post-hoc multiple comparisons acknowledged the statistical

significance of the differences (F=50.62 for EE, p<0.001; and F=64.11 for DP, p<0.001). Tukey

HSD post-hoc multiple comparisons tests showed significance of differences between Turkey

and all other countries, with the observed mean in Turkey being well above the EE values in the

other countries. The second highest mean of EE,positioned in the upper end of the average EE

scores,was observed in Greece and Bulgaria (Greece and Bulgaria differed from the other

countries, but not between themselves). Portugal, Romania, Croatia and Macedonia were the

final group of countries, that all shared a similar level of EEin the lower end of the average EE

scores. Cross-national analysis of depersonalization revealed a somewhat dissimilar picture. The

highest level of DPwas again observed in Turkey, and the second highest DP(positioned at the

threshold of high DP) was found in Greece. In third place came Bulgaria, Romania, Portugal and

11

Croatia, with almost equal means, corresponding to average DP. The lowest DP (at the lower end

of the average DP scores)was observed in Macedonia.

Regarding health behaviors, results showed that medical professionals (full sample)

consumed fast food on average about 2.5 times per week, exercised on average about 1.5 times

per week, consumed alcohol and used painkillers on average about once a week (Table 2). One-

way ANOVA revealed significant differences in health behaviors between countries, being most

pronounced in respect to fast food consumption (F=106.34, p<0.001). Post-hoc multiple

comparisons differentiated three groups of countries that differed significantly between each

other. Most often fast food was consumed in Bulgaria and Greece – nearly 4 times per week. The

next group of countries included Turkey and Macedonia, where fast food consumption equaled

twice per week, and the third group comprised of Romania, Croatia and Portugal, where fast

food was consumed once a week. Exercise levels in all countries were similar, amounting to

slightly more than once a week to about twice a week (about 1.5 times per week onaverage). Yet

some differences were observed (F=5.36, p<0.001). The highest physical activity frequency was

observed in Romanian health professionals, and it differed significantly from physical activity

levels in Greece, Portugal, Turkey and Croatia. Macedonian health professionals came after their

Romanian colleagues and their physical activity level differed significantly from that of their

colleagues from Greece. Alcohol and painkillers were both used on average once a week by

health professionals (full sample). Alcohol consumption was most often reported in Bulgaria,

who differed significantly from the group of all other countries except Greece (F=19.07,

p<0.001). Second in alcohol consumption cameGreece (differing significantly from all countries

except Portugal), and Portugal health professionals were third in alcohol consumption, differing

significantly from their Romanian and Croatian colleagues. Painkillers were most often used by

Turkish health professionals – twice weekly – which was double than the use of such

medicaments among their colleagues from the other countries (F=28.43, p<0.001).

Associations between burnout and health behaviors

The results from the 8 hierarchicalregressions ran with the full sample with fast food

consumption, exercise, alcohol drinking and painkiller use being the dependent variablesare

presented in Tables 3 and 4 (Table 3 includes results with EE being an independent variable, and

12

Table 4 includes results with DP being an independent variable). Model 1 regressions included

EE or DP as the independent predictors,in Model 2 regressions age, gender, staff position

(doctor/resident vs. nurse), and unit tenure were addedin a second block of independent

predictors, and in Model 3 regressions the 7-level categorical variable ‘country’ was added in a

third block with Greece being the reference category.The results regarding EE in Model 1

regressions demonstrated that it was significantly positively associated with fast food

consumption, alcohol drinking and painkiller use, and negatively associated with frequency of

weekly exercise (Table 3). The values of the regression coefficient indicated the strongest

relationshipin regard to self-medication through painkillers and the weakest relationshipin regard

to alcohol consumption.

-----------------------------------------------------------------------------------------------------------------


-----------------------------------------------------------------------------------------------------------------

The proportion of the variance in health behaviors which was explained in Model 1

ranged from close to 0% in regard to alcohol consumption, to 7% in regard to painkiller use.

When adding the individual differences variables in Model 2, all the above described

associations remained significant. The R2 change values showed that adding the individual

variablesin Model 2 led to a significant increase in the total variance explained by 1% regarding

exercise, by2% regardingpainkiller use, by5% regarding fast food consumption and by 10%

regarding alcohol consumption. Thus alcohol consumption was the behavior most strongly

(positively) associated with individual characteristics.It was positively related tobeing a male,

being a doctor/resident, and with older age. More frequent fast food consumption was associated

with being a male, being a doctor/resident,and to a shorter unit tenure; more engagement in

exercise was associated with being a male; and more frequent self-medication through

painkillerswas related to being a female. To summarize the significance of the individual

variables, gender was significantly associated with all of the studied behaviors, staff position was

significantly related to fast food consumption and alcohol drinking, unit tenure was significantly

associated only with fast food consumption, and age – only with alcohol consumption.

13

Adding the country from which the participating health professionals came in Model 3

led to a significant increase in the total variance explained by 1% regarding exercise, by 3%

regarding alcohol consumption, by 4% regarding painkiller use, and by 18% regarding fast food

consumption, with associations between EE and health behaviors remaining significant. In

regard to fast food consumption, participants’ country turned out to be its most important

determinant. The inclusion of ‘country’ changed the associations of health behaviors with some

of the individual factors, i.e. more frequent fast food consumption now being associated with

younger age, but not associated with unit tenure anymore; and more frequent alcohol

consumption not associated with age anymore.

The results from the hierarchical regression with DP as independent variable proved to be

quite similar to the above described results. In Model 1 regressions DP was significantly

positively associated with fast food consumption, alcohol drinking and painkiller use, and

negatively associated with frequency of weekly exercise (Table 4). The values of the regression

coefficient indicated the strongest relationship in regard to painkiller use and the weakest

relationships in regard to alcohol consumption and exercise.Comparing the relationships of EE

with health behaviors and of DP with health behaviors we noticed that EE had stronger

relationships with exercise and painkiller use that DP had, and DP had stronger relationship with

alcohol consumption than EE had.

-----------------------------------------------------------------------------------------------------------------


-----------------------------------------------------------------------------------------------------------------

The proportion of the variance in health behaviors which was explained in Model 1

regressions ranged from 1% in regard to exercise and alcohol drinking to 3% in regard to fast

food consumption and 4% in regard to self-medication through painkillers. When adding the

individual differences variables in Model 2, all the above described associations remained

significant. The R2 change values showed that adding the individual variables in Model 2 led to

(as in the case with EE) a significant increase in the total variance explained by 1% to 4%

(regarding exercise, fast food consumption and painkiller use) and to 9% regarding alcohol

drinking. The results regarding the individual variables in Model 2 regressions with DP as an

14

independent variable were analogous to the above described results for Model 2regressions with

EE as an independent variable.

Adding the country from which the participating health professionals came in Model 3

led to a significant increase in the total variance explained by 1% regarding exercise, by 3%

regarding alcohol consumption, by 5% regarding painkiller use, and by 19% regarding fast food

consumption, with associations between DP and health behaviors remaining significant.

Participants’ countrywasthe most important determinant of fast food consumption. The inclusion

of ‘country’ changed the associations of health behaviors with some of the individual factors, i.e.

more frequent fast food consumption now being associated with younger age, but not associated

with unit tenure anymore; more frequent alcohol consumption not associated with age anymore;

and more frequent painkiller use associated with older age now.

Cross-national differences

In line with our cross-national approach associations between burnout and health

behaviors, adjusted for individual difference factors, were analyzed separately in all of the

participating countries as well.Table 5 presents the results from the linear regression models

including emotional exhaustion or depersonalization as the single predictor variable (Model 1),

so that the R2values for the two burnout dimensions can stand out. Adding the individual

differences variables in the second block of independent predictors within Model 2 didn’t change

the associations between burnout and health behaviors – they all remained significant.

-----------------------------------------------------------------------------------------------------------------


-----------------------------------------------------------------------------------------------------------------

Cross-national comparisons showed that statistically significant associations displayed

the same relationships as described above: higher levels of EE or DP were related to more

frequent fast food consumption, less frequent exercise, more frequent alcohol drinking and self-

medication through painkillers. However, while in the full sample burnout dimensions were

15

significantly related to all health behaviors, in many cases the associations between them did not

reach statistical significance in the separate countries(Table 5).

For example, in Portugal and in Croatia only one significant relationship was observed in

regard to EE (in Portugal EE was significantly associated with exercise, and in Croatia – with

painkiller use), and in Macedonia the statistically significant associations were two (one between

EE and alcohol drinking, and one between DP and painkiller use). In Turkey burnout dimensions

were associated with all behaviors (with only one exception – the association between EE and

alcohol consumption), and in the remaining countries the significant relationships add up to

fourin Greece and Bulgaria and to five in Romania (in all three countries EE was associated with

exercise and painkiller use; in Greece and Bulgaria DP was associated with fast food

consumption; in Greece and Romania DP was associated with exercise; in Bulgaria and Romania

DP was associated with alcohol consumption; and in Romania EE was associated with alcohol

consumption).

Some of the observed associations seem to be more universal than others, e.g. EE was

positively related to self-medication through painkillers in all countries except Portugal and to

infrequent exercise in all countries except Croatia and Macedonia, while EE was associated with

fast food consumption, and DP with painkiller use only in Turkey; similarly EE was associated

with alcohol consumption only in Romania. Interestingly, exercise frequency and painkiller use

were associated more often with EE than with DP, while DP was more commonly associated

with fast food and alcohol consumption.

Discussion

Burnout rates

According to our findings, one in every five health professionals in the overall sample

had high scores on both burnout dimensions, and the overall sample means purported to an

average level of both emotional exhaustion and depersonalization. Both emotional exhaustion

and depersonalization reached striking proportions in Turkish health professionals. Emotional

exhaustion was also quite high in Greek and Bulgarian health professionals, and

depersonalization – in Greek health professionals. The means for burnout in the separate

16

countries correspond to high EE and high DP in Turkey, upper end of the average EE scores in

Greece and Bulgaria, and the threshold of high DP in Greece. In Portugal, Romania, Croatia and

Macedonia the means corresponded to average levels of EE and DP. In summary, considerable

cross-national differences in regard to EE and DP were observed in our sample of European

health professionals, in agreement with similar studies (e.g. Soler et al. 2008; Beezhold et al.

2010; Poghosyan et al. 2010; Turnipseed and Turnipseed 1997; Schaufeli and Janczur 1994). For

example, within the EGPRN study, comprising European family doctors from Greece, Bulgaria,

Turkey and Croatia, Turkey was a country with lower EE and DP, Greece had higher proportions

of doctors with high DP, and Bulgaria had higher proportions of doctors with high EE (Soler et

al. 2008).

It has been proposed that the differences owing to country of origin may represent

differences due to the health care systems of the participating countries(Soler et al. 2008). Both

Turkey and Bulgaria shared the implementation of major healthcare reforms in the beginning of

the new century, and both Greece and Bulgaria shared the consequences of severe socio-

economic crisis in the past years. Thus we might think of additional stressors encountered by the

health professionals in these countries such as financial and professional insecurity and role

ambiguity in the face of changing health system’s regulations, and working in the situation of

shortage of resources (e.g. lack of materials, equipment and personnel, underpayment), which

contribute to higher levels of burnout and its consequences.Within our study and the framework

of the Job Demands-Resources model we found that physical, organizational and emotional work

demands were highest in Turkey and Greece, and quite high in Bulgaria.

Frequencies of health behaviors

It was found that medical professionals exhibited a largely sedentary lifestyle, exercising

on average only 1.5 times per week, obviously failing to meet recommended amounts of physical

activity needed for health (WHO 2010). They were also prone to unhealthy eating: fast food was

consumed on average 2.5 times per week, which could be considered regular consumption

(Anderson et al. 2011; Odegaard et al. 2012). Alcohol consumption seemed to be infrequent

(about once a week), while painkiller use amounted to once a week on average (in Turkey

17

amounting to twice a week), which could be considered frequent use (Dale et al. 2015;

Samuelsen et al. 2015).

Based on these findings we might assume that there is a risk for medical professionals’

health and well-being due to their unhealthy lifestyle, characterized by physical inactivity,

frequent fast food consumption and frequent painkiller use.The risks associated with physical

inactivity are well established – it has been identified as the fourth leading risk factor for global

mortality; for cardiovascular disease, diabetes and cancer and their risk factors as raised blood

pressure, raised blood sugar and overweight (WHO 2010).Regarding regular fast food intake, it

has been found that adults consuming Western-style fast food ≥2 times per week had an

increased risk of developing type 2 diabetes mellitus and dying of coronary heart disease

(Odegaard et al. 2012). The potential inappropriate use of painkillers is associated with increased

risk of cardiovascular disease, gastrointestinal damage, renal disease, renal failure and

interactions with other drugs (Dale et al. 2015; Samuelsen et al. 2015).

The cross-national data point to higher health risks associated with fast food consumption

for Bulgarian and Greek health professionals, who ate fast food on average about four times per

week; the risk is high also for their Turkish and Macedonian colleagues, whose average fast food

consumption was 2 times per week. Turkish health professionals might be at higher risk

associated with regular painkiller use, as they take such medications on average twice weekly

(double than the use of their colleagues from the other countries).

Associations between burnout and health behaviors

It was found that higher emotional exhaustion and higher depersonalization were

significantly associated with more frequent fast food consumption, less frequent exercise, more

frequent alcohol consumption and more frequent painkiller use when analyses were performed

with the full sample, and these weak to moderate associations remained significant after the

individual differences variables had been added to the model.Significance of the burnout-health

bahaviors relationship was retained also after adding participants’ country in the regression.Such

relationships between burnout and the studied health behaviors have been found in other studies

withhealth professionals’ samples, exploring health behaviors as outcomes of burnout. For

example, burnout was associated with an unhealthy diet in dentists (Gorter et al. 2000); it was

18

related to fast food consumption in ambulance workers (Moustou et al. 2010); and stress was

related to fast food consumption in nurses (Steptoeetal. 1998). Lower participation in physical

exercise was associated with burnout in emergency physicians (Goldberg et al. 1996), dentists

(Gorter et al. 2000) and ambulance workers (Moustou et al. 2010). Conversely, engaging in sport

has been related to lower burnout in nurses (Chayu and Kreitler 2011).Burnout was found to be

linked to increased alcohol consumption in dentists (Gorter et al. 2000; Winwood et al. 2003),

emergency physicians (Goldberg et al. 1996) and European family physicians from 12

countries(Soler et al. 2008), to alcohol abuse or alcohol dependence in surgeons (Oreskovich et

al. 2012) and to patterns of social drinking in ambulance workers (Moustou et al. 2010). Higher

levels of burnout predicted weekly painkiller use in employees from the human service sector,

including health professionals (Kristensen et al. 2005).

The burnout-health behaviors relationship can be viewed as a part of an underlying health

impairing process, in which work stressors (job demands) exhaust employees’ mental and

physical resources and lead to burnout/exhaustion and to health problems (Bakker and

Demerouti 2007; Demerouti and Bakker 2011; Demerouti et al. 2001), with health-impairing

behaviors being one of the potential mechanisms, linking burnout to ill health (Melamed et al.

2006). It is proposed that health-impairing behaviors might function as a coping strategy for

relieving distress in the short term (Shirom 2010a).The exact mechanisms through which burnout

links with health behavior needs further investigation and have to be addressed in future

research.For example, painkiller use was the behavior most strongly associated with the burnout

dimensions in our study. It might be expected that frequent painkiller useis tied to the need for

coping with pain and discomfort in cases of existing somatic complaints, established as

consequences or concomitants of burnout (Soares et al. 2007; Bauer et al. 2006; Schaufeli and

Buunk 2003; Schaufeli 2003; Shirom et al. 2005; Melamed et al. 2006). It has been found,

however, that the association between stress and painkiller use remains significant and graded

after adjusting for stress related pain and discomfort (Koushede et al. 2009).

For the specific behaviors, the country from which the participants came from was the

most important determinant of fast food consumption, implying that this behavior is most

strongly associated withcultural appropriateness and to the extent cultures are influenced by and

espouse other lifestyles that are not traditional for that culture (i.e. Western eating practices).

Individual factors were the most important determinant of alcohol consumption.After controlling

19

for country, more frequent alcohol consumption was associated with being a male, and being a

doctor/resident. This is in line with the well-known fact that the frequency of alcohol

consumption and related health risks are higher in men compared to women (WHO 2014). The

same explanation may apply to the staff position of doctor/resident which correlated with male

gender, while virtually all nurses were female.More frequent fast food consumption was

associated with being a male, being a doctor/resident,and with younger age. Regular fast food

consumption has been higher among men and younger adults in other studies as well (Anderson

et al. 2011).More frequent consumption of fast food in males and doctors/residents may be

related to the less proneness of men to prepare (healthy) food, especially when they have such a

busy schedule – the first reason for fast food has been ‘quick and convenient’ (Anderson et al.

2011). Younger age may be associated with the more dynamic lifestyle of younger people with

less attention to the preparation of healthy food, especially if living on their own.More

engagement in exercise was associated with being a male, with males being more physically

active than female being a well-established gender difference. More frequent self-medication

through painkillerswas related to being a female and to older age. More frequent self-prescribed

painkiller use by females is consistent with other studies (Dale et al. 2015;Golar 2011; Naveed et

al. 2014; Samuelsen et al. 2015), and the use of some types of analgesics increases with age

(Dale et al. 2015).

In some of the countries burnout dimensions were not significantly associated with

certain health behaviors, but when the relations were significant, the same direction of

associations was observed. Most universal across countries was the fact that emotional

exhaustion was associated with infrequent exercise and self-medication through painkillers,

which may be due to a common underlying psychological/physiological mechanism. Turkey was

the country, in which virtually all associations were significant and stronger than in the other

countries. Significant associations between burnout dimensions and several health behaviors

were found also in Bulgaria, Greece and Romania.

Strengths and limitations of the study

The main strength of the study is that it assesses the underinvestigatedrelationships

between burnout and health behaviors among health professionals in a number of European

20

countries simultaneously, adopting a cross-national approach. The study comprises health

professionals from seven South and South Eastern European countries thus providing data on

burnout and health behaviors for a region of Europe that is rarely explored.

Alimitation of this study is its cross-sectional design, which does not allow exploration of

underlying mechanisms linking burnout to health behaviors ormaking conclusions regarding the

direction of causality between burnout and health behaviors. Another limitation is that health

behaviors are self-rated and not measured by objective indicators.Among the limitationsof the

studyis that it does not investigate the role of other potential confounders in the burnout - health

behaviors association, such as personality, general philosophical outlooks to life, family situation

and relationships, etc. Also, our convenience sampling approach does not allow us to generalize

our conclusions to the whole population of interest – health professionals working in university

hospitals in the seven South and South Eastern European countries. Thus the observed cross-

national differences might be due to specifics of the samples.The differences among the national

samples in respect to proportions of different staff within the samples might also be a factor to

generalizability of conclusions, as some research infers that significant differences in burnout

and its correlates between doctors, nurses and residents are present (e.g. Embriaco, Papazian et

al. 2007; Panagopoulou et al. 2006).

Conclusions

The study found that burnout dimensions emotional exhaustion and depersonalization

were significantly positively related toseveral risk health behaviorsin a multinational sample of

healthcare professionals from South and South Eastern Europe. Burnout dimensions wereweakly

to moderatelyassociated with higher fast food consumption, infrequent exercise, higher alcohol

consumption and more frequent painkiller use.Moreover, theseassociations remained significant

after the inclusion of individual differences factors like age, gender, staff position and unit

tenure. Within its cross-national approach the studyrevealedsignificant differences between the

participating countries in the means of the burnout dimensions and health behaviors, as well as

some differences in the statistical significance and magnitude (but not the direction) of the

associations between burnout and health behaviors.The country from which the participants

21

came from was the most important determinant of fast food consumption and individual factors

were the most important determinant of alcohol consumption.

Moreover, the study of high risk behaviors among health professionals is important not

only in the context of health behaviors having a negative impact on health and well-being, but

also in the context of these professionals’ roles as care providers. Risky health behaviors (e.g.

psychoactive substance use) have been implicated as factors contributing to medical errors and

inadequate patient safety, as well as affecting medical professionals’ health promotion activities

directed toward patients (Bakhshi and While 2014; Fie et al. 2013; Zhu et al. 2011; Pipe et al.

2009). Organizational interventions should incorporate early identification of such behaviors

together with programs promoting health and aimed at the reduction of burnout and work-related

stress, including the promotion of work engagement, which has been found to foster protective

health behaviors (Alexandrova-Karamanova et al. 2013; Shirom 2010b).

Funding This study was funded bythe European Union’s Seventh Framework ProgrammeGrant

Number 242084.

Conflict of interest The authors declare that they have no conflict of interest.

Ethical approvalAll procedures performed in studies involving human participants were in

accordance with the ethical standards of the institutional and/or national research committee and

with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

References

AholaK,HonkonenT, PirkolaS, IsometsaE, KalimoR, NykyriE et al(2006)Alcohol dependence in

relation to burnout among the Finnish working population.Addict 101:1438-

1443.doi:10.1111/j.1360-0443.2006.01539.x

AholaK, Pulkki-RabackL,KouvonenA, RossiH, AromaaA, LonnqvistJ(2012) Burnout and

behavior-related health risk factors: results from the population-based Finnish Health

2000 study. J Occup Environ Med54:17-22.doi:10.1097/JOM.0b013e31823ea9d9

22

Akvardar Y,Demiral Y, Ergor G, Ergor

A(2004)Substanceuseamongmedicalstudentsandphysiciansin a

medicalschoolinTurkey.Soc Psych PsychEpid 39:502-506. doi:10.1007/s00127-004-

0765-1

Alexandrova-KaramanovaA, NaydenovaV, TodorovaI(2013)[Work engagement, health

behaviours and subjective well-being in medical professionals from university hospitals].

Angazhiranost v rabotata,

zdravnipovedeniaisubektivnoblagopoluchieprimedicinskispetsialistiotuniversitetskibolnit

si. Bulg J Psychol 1-4:67-84. Bulgarian

AndersonB, RaffertyAP, Lyon-CalloS, FussmanC, ImesG(2011)Fast-food consumption and

obesity among Michigan adults. Prev Chronic Dis 8:A71.

http://www.cdc.gov/pcd/issues/2011/jul/10_0186.htmAccessed 13 October 2014

Armstrong-StassenM, Al-Ma ‘AitahR, CameronS, HorsburghM(1994)Determinants and

consequences of burnout: A cross-cultural comparison of Canadian and Jordanian nurses.

Health Care Women Int 15:413-421.doi:10.1080/07399339409516133

BakhshiS, WhileAE(2014)Health professionals’ alcohol-related professional practices and the

relationship between their personal alcohol attitudes and behavior and professional

practices: A systematic review.Int J Environ Res Public Health 11:218-

248.doi:10.3390/ijerph110100218

Bakker AB, Demerouti E (2007) The Job Demands-Resources model: State of the art. Journal of

Managerial Psychology, 22:309-328.

BauerJ, StammA, VirnichK,WissingK, MullerU, WirschingM,

SchaarschmidtU(2006)Correlation between burnout syndrome and psychological and

psychosomatic symptoms among teachers. Int Arch Occup Environ Health 79:199-

204.doi:10.1007/s00420-005-0050-y

BeezholdJ, JovanovicN, AndlauerO,PodlesekA, PappS, FerrariS, MihaiA(2010) YPS01-01 - The

international resident/trainee burnout study: boss international. EurPsychiat 25 Suppl

1:1698.doi:10.1016/S0924-9338(10)71677-1

BoudreauRA, GriecoRL, CahoonSL,RobertsonRC, WedelRJ(2006)The pandemic from within:

Two surveys of physician burnout in Canada. Can J Community Ment Health 25:71-88.

doi: 10.7870/cjcmh-2006-0014

23

Cartwright M, Wardle J, Steggles N, Simon AE, Croker H, Jarvis MJ

(2003)Stressanddietarypracticesinadolescents. Health Psychol 22:362-369.

doi:10.1037/0278-6133.22.4.362

Castelo-Branco C, Figueras F, Eixarch E, Quereda F, Cancelo MJ, Gonzales S, Balsch, J (2007)

Stress symptoms and burnout in obstetric and gynaecology residents. BJOG: An

International Journal of Obstertrics&Gynaecology, 114:94-98. doi: 10.1111/j.1471-

0528.2006.01155.x

ChayuT, KreitlerS(2011)Burnout in nephrology nurses in Israel.NephrolNurs J 38:65-77.

CunradiCB, Chen M-J, LiptonR(2009)Association of occupational and substance use factors

with burnout among urban transit operators.J Urban Health 86:562-

570.doi:10.1186/1747-597X-2-36

Demerouti E, Bakker AB (2008) TheOldenburgBurnoutInventory: A

goodalternativetomeasureburnoutandengagement. In J. Halbesleben (Ed.),

Stressandburnoutinhealthcare. NovaSciences.

Demerouti E, Bakker AB (2011) The Job Demands– Resources model: Challenges for future

research. South African Journal of Industrial Psychology,37:1-9.

Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB (2001)The job demands-resourcesmodel

of burnout. J ApplPsychol 86:499-512. doi:10.1037/0021-9010.86.3.499

Demerouti, E., Bakker, A. B., Vardakou, I., &Kantas, A.(2003).

Theconvergentvalidityoftwoburnoutinstruments: A multitrait-

multimethodanalysis.EuropeanJournalofPsychologicalAssessment, 19,12–23.

Demir A, Ulusoy M, Ulusoy MF (2003) Investigation of factors influencing burnout levels in the

professional and private lives of nurses. International Journal of Nursing Studies, 40,

807-827.

Dreano-Hartz S, Rhondali W, Ledoux M, Ruer M, Berthiller J, Schott AM, Monsarrat L, Filbet

M (2015) Burnout among physicians in palliative care: Impact of clinical settings. Palliat

Support Care, 14:1-9.

EmbriacoN, PapazianL, Kentish-BarnesN, PochardF, AzoulayE(2007)Burnout syndrome among

critical care healthcare workers. CurrOpinCrit Care 13:482-488.

Estryn-Behar M, vanderHeijdenB, Ogińska H,Camerino D, LeNézet O, Conway PM, Fry

C,Hasselhorn HM (2007) Theimpactofsocialworkenvironment, teamworkcharacteristics,

24

burnout, andpersonalfactorsuponintenttoleaveamongEuropeannurses. Medical Care,

45:939-950.

Estryn-Behar M, vanderHeijdenB, Camerino D, Fry C,LeNézet O, Conway PM, Hasselhorn HM,

the NEXT study group (2008)Violence risks in nursing – results from the European

‘NEXT’ study. Occup Med 58:107-114. doi:10.1093/occmed/kqm142

FieS, NormanIJ, WhileAE (2013)The relationship between physicians’ and nurses’ personal

physical activity habits and their health-promotion practice: A systematic review. Health

Educ J 72:102-119.doi:10.1177/0017896911430763

Golar SK (2011)Useandunderstandingofanalgesics (painkillers)byAstonuniversitystudents.

Bioscience Horizons 4:71-78. doi:10.1093/biohorizons/hzr009

GoldbergR, BossRW, ChanL, GoldbergJ, MallonWK, MoradzadehD et al (1996)Burnout and its

correlates in emergency physicians: Four years' experience with a wellness booth.

AcadEmerg Med 3:1156-1164.doi:10.1111/j.1553-2712.1996.tb03379.x

GorterRC, EijkmanMAJ, HoogstratenJ(2000)Burnout and health among Dutch dentists. Eur J

Oral Sci 108:261-267.doi:10.1034/j.1600-0722.2000

GrassiL, MagnaniK(2000)Psychiatric morbidity and burnout in the medical profession: An

Italian study of general practitioners and hospital physicians. PsychoterPsychosom

69:329-334.doi:10.1159/000012416

Green DE, Walkey FH, Taylor AJ (1991)The three-factor structure of the MaslachBurnout

Inventory: A multicultural, multinational confirmatory study.J SocBehavPers 6:453-472.

Harkness JA (2003) Questionnaire translation. In: Harkness JA, van de Vijver F,Mohler, P

(eds)Cross-cultural survey methods. John Wiley and Sons,New York,pp 35–56

JovanovicN, BeezholdJ, AndlauerO, KuzmanMR, PodlesekA, HanonC, SchulzeB(2009)Burnout

among psychiatry residents: The International Psychiatry Resident/Trainee Burnout

Syndrome Study (BoSS). Die Psychiat 6:75-79.

Kanai-PakM, AikenLH, SloaneDM, PoghosyanL(2008)Poor work environments and nurse

inexperience are associated with burnout, job dissatisfaction and quality deficits in

Japanese hospitals. J ClinNurs 17:3324-3332.doi:10.1111/j.1365-2702.2008.02639.x

Kandiah J,YakeM, JonesJ, MeyerM(2006)Stress influences appetite and comfort food

preferences in college women. Nutr Res 26:118-123.doi:10.1016/j.nutres.2005.11.010

25

Kristensen TS, Borritz M, Villadsen E, et al. (2005) The Copenhagen Burnout Inventory: a new

tool for the assessment of burnout. Work & Stress;19(3):192-207.doi:

10.1080/02678370500297720

Maroco J, MarocoAL, Leite E, Bastos C, Vazao MJ, Campos J (2016) Burnout in Portuguese

healthcare professionals: An analysis at the national level. Acta Med Port, 29:24-30.

MaslachC, JacksonSE, LeiterMP(1996)MBI: The Maslach Burnout Inventory:

Manual.Consulting Psychologists Press, Palo Alto

MaslachC, SchaufeliWB, LeiterMP(2001)Job Burnout.Annu Rev Psychol 52:397-

422.doi:10.1146/annurev.psych.52.1.397

McDonaldT, SiegallM(2004) Burnout and expectancies about alcohol use: Drinking behavior in

a sample of university professors.

B>Questhttp://www.westga.edu/~bquest/2004/burnout.htmAccessed on 20 October 2014

Melamed, S., Kushnir, T., Shirom, A. (1992) Burnout and risk factors for cardiovascular disease.

Behav Med 18:53-60.doi:10.1080/08964289.1992.9935172

MelamedS, ShiromA, TokerS, BerlinerS, ShapiraI(2006) Burnout and risk of cardiovascular

disease: Evidence, possible causal paths, and promising research directions. PsycholB

132:327-353.doi:10.1037/0033-2909.132.3.327

Mikolajczyk, R. T., El Ansari, W., Maxwell, A.E. (2009)Food consumption frequency and

perceived stress and depressive symptoms among students in three European countries.

Nutr J 8:31.doi:10.1186/1475-2891-8-31

MontgomeryAJ, BradleyC, RochfortA, PanagopoulouE(2011)A review of self-medication in

physicians and medical students. Occup Med 61:490-497.doi:10.1093/occmed/kqr098

MontgomeryAJ, PanagopoulouE, KehoeI, ValkanosE(2011)Connecting organisational culture

and quality of care in the hospital: is job burnout the missing link?J Health Organ

Manag25:108-123.doi:10.1108/14777261111116851

MoustouI, PanagopoulouE, MontgomeryAJ, BenosA(2010)Burnout predicts health behaviors in

ambulance workers. Open Occup Health & Safety J 2:16-18.

Naveed S, Ghayas S, Jawed A, Kanwal F, Mohiuddin B, Sabeen E, Sabuhi B

(2014)Unnecessaryuseofpainkillerindifferentagegroups.DHRInt J Med Sci 5:89-95.

26

Nella D, Panagopoulou E, Galanis N, Montgomery A, Benos A (2015) Consequences of job

insecurity on the psychological and physical health of Greek civil servants.BioMed Res

Int,Article ID 673623. doi: 10.1155/2015/673623

NevanperaNJ, HopsuL, KuosmaE, UkkolaO, UittiJ, LaitinenJH(2012) Occupational burnout,

eating behavior, and weight among working women. Am J ClinNutr95:934-

943.doi:10.3945/ajcn.111.014191

O’Connor DB, Jones F, Conner M, McMillan B, Ferguson E

(2008)Effectsofdailyhasslesandeatingstyleoneatingbehavior. Health Psychol 27Suppl

1:S20–S31.doi:10.1037/0278-6133.27.1.S20

OdegaardAO, Koh WP, Yuan JM, Gross MD, Pereira MA (2012)Western-

stylefastfoodintakeandcardiometabolicriskinanEasterncountry. Circulation, 126:182-188.

doi: 10.1161/CIRCULATIONAHA.111.084004

OreskovichMR, KaupsKL, BalchCM, HanksJB, SateleD, SloanJ et al(2012)Prevalence of

alcohol use disorders among American surgeons. Arch Surg 147:168-174.

doi:10.1001/archsurg.2011.1481

PanagopoulouE, MontgomeryAJ, BenosA(2006)Burnout in internal medicine physicians:

Differences between residents and specialists.Eur J Int Med 17:195-

200.doi:10.1016/j.ejim.2005.11.013

ParkCL, IacoccaMO(2014)A stress and coping perspective on health behaviors: theoretical and

methodological considerations. Anxiety Stress Copin 27:123-

137.doi:10.1080/10615806.2013.860969

PetersonU, DemeroutiE, BergstromG, SamuelssonM, AsbergM, NygrenA(2008)Burnout and

physical and mental health among Swedish healthcare workers. J AdvNurs 62:84-

95.doi:10.1111/j.1365-2648.2007.04580.x

Pines AM, Aronson E (1988)Careerburnout: Causesandcures. FreePress, New York.

PipeA, SorensenM, ReidR(2009)Physician smoking status, attitudes toward smoking, and

cessation advice to patients: An international survey. Patient EducCouns74:118-

123.doi:10.1016/j.pec.2008.07.042

PoghosyanL, ClarkeSP, FinlaysonM, AikenLH(2010)Nurse burnout and quality of care: Cross-

national investigation in six countries. Res Nurs Health 33:288–

298.doi:10.1002/nur.20383

27

PopaF, RaedA, PurcareaVL, LalaA, BobirnacG(2010)Occupational burnout levels in emergency

medicine–a nationwide study and analysis. J Med Life 3:207-215.

PrinsJT, Hoekstra-WeebersJ, Gazendam-DonofrioSM,DillinghGS, BakkerAB, HuismanMet

al(2010)Burnout and engagement among resident doctors in the Netherlands: a national

study. Med Educ 44:236-247.doi:10.1111/j.1365-2923.2009.03590.x

PutnikK, HoukesI(2011)Work related characteristics, work-home and home-work interference

and burnout among primary healthcare physicians: A gender perspective in a Serbian

context. BMC Public Health 11:716.doi:10.1186/1471-2458-11-716

Qiao H, Scaufeli, WB (2011) Theconvergentvalidityoffourburnoutmeasuresin a Chinesesample:

A confirmatoryfactor-analyticapproach. Applied Psychology: An International Review,

60 (1), 87-111.

Ramsted M (2001) Alcohol and suicide in 14 European countries.Addiction 96Suppl 1:S59-

75.doi:10.1046/j.1360-0443.96.1s1.6.x

SchaufeliWB(2003)Past performance and future perspectives on burnout research. SA J

IndPsychol 29:1-15.doi:10.4102/sajip.v29i4.127

SchaufeliWB, BuunkBP(2003)Burnout: An overview of 25 years of research and theorizing.

In:Schabracq MJ, Winnubst JAM, Cooper CL (eds)The handbook of work and health

psychology. Wiley,Chichester, pp 383–425

Schaufeli WB, JanczurB(1994)Burnout among nurses: A Polish-Dutch comparison.J Cross-Cult

Psychol25:95-113.doi:10.1177/0022022194251006

SchaufeliWB, LeiterMP, MaslachC(2009)Burnout: 35 years of research and practice.Career Dev

Int 14:204-220.doi:10.1108/13620430910966406

Schulz M, Damkroger A, Heins C, Wehlitz L, Lohr M, Driessen M, Behrens J, Wingenfeld K

(2009)Effort-reward imbalance and burnout among German nurses in medical compared

with psychiatric hospital settings.J PsychiatrMent Health Nurs, 16:225-233. doi:

10.1111/j.1365-2850.2008.01355.x.

ShanafeltTD, BooneS, TanL, DyrbyeLN, SotileW, SateleD et al (2012)Burnout and satisfaction

with work-life balance among US physicians relative to the general US population.Arch

Intern Med 172:1377-1385.doi:10.1001/archinternmed.2012.3199

28

ShanafeltTD, Oreskovich MR, Dyrbye LN, Satele DV, Hanks JB, Sloan JA, Balch CM (2012)

Avoiding burnout: the personal health habits and wellness practices of US surgeons. Ann

Surg 255:625-633. doi:10.1097/SLA.0b013e31824b2fa0

SharmaA, SharpDM, WalkerLG, MonsonJRT(2008)Stress and burnout in colorectal and

vascular surgical consultants working in the UK National Health Service. Psycho-Oncol

17:570-576.

Shirom A (2010a) Employee burnout and health: Current knowledge and future research paths.

In:Houdmont J,LekaS (eds)Contemporary occupational health psychology, vol 1.Wiley,

Chichester, pp 59–77

ShiromA(2010b)Feeling energetic at work: On vigor’s antecedents. In: Bakker AB, Leiter MP

(eds)Work engagement: A handbook of essential theory and research.Psychology Press,

Hove, pp 69–84

Shirom, A., &Melamed, S. (2006). A

comparisonoftheconstructvalidityoftwoburnoutmeasuresamongtwogroupsofprofessionals.

International JournalofStressManagement,13, 176–200.

ShiromA, MelamedS, TokerS, BerlinerS, ShapiraI(2005)Burnout, mental and physical health: A

review of the evidence and a proposed explanatory model. Int Rev Ind Organ

Psychol20:269-309.

SoaresJJF, GrossiG, SundinO(2007)Burnout among women: Associations with

demographic/socio-economic, work, life-style and health factors. Arch Women’s Ment

Health 10:61-71.doi:10.1007/s00737-007-0170-3

SolerJK, YamanH, EstevaM, DoddsF, SpiridonovaAsenovaR, KaticM et al; European General

Practice Research Network Burnout Study Group(2008)Burnout in European family

doctors: the EGPRN study. Fam Pract 25:245-265.doi:10.1093/fampra/cmn038

Steptoe A, Lipsey Z, Wardle J (1998)Stress,

hasslesandvariationsinalcoholconsumption,foodchoiceandphysicalexercise: A diarystudy.

BritJ Health Psych 3:51-63.doi:10.1111/j.2044-8287.1998.tb00555.x

Swetz KM, Harrington SE, Matsuyama RK, ShanafeltTD, Lyckholm LJ (2009)Strategies for

avoiding burnout in hospice and palliative medicine: Peer advice for physicians on

achieving longevity and fulfillment. J Palliat Med 12:773-

777.doi:10.1089/jpm.2009.0050

29

TironiMOS, SobrinhoCLN, de Souza BarrosD, ReisEJFB, FilhoESM, AlmeidaA et al

(2010)Professional burnout syndrome among intensive care physicians in Salvador,

Brazil. Brazil Rev Assoc Med Bras56:656-662.doi:10.1590/S0104-42302009000600009

Tsiga E, Panagopoulou E, Niakas D (in press) Health promotionacross occupational groups: One

size doesn’t fit all. Occup Med-c

TurnipseedDL, TurnipseedPH(1997)A bi-cultural analysis of the costs of caring: nursing burnout

in the United States and the Philippines.Career Dev Int 2:180-

188.doi:10.1108/13620439710173670

Weight CJ, Sellon JL, Lessard-Anderson CR, Shanafelt TD, Olsen KD, Laskowski ER

(2013)Physical activity, quality of life, and burnout among physician trainees: The effect

of a team-based, incentivized exercise program. Mayo Clin Proc 88:1435–

1442.doi:10.1016/j.mayocp.2013.09.010

WinwoodPC, WinefieldAH, LushingtonK(2003)The role of occupational stress in the

maladaptive use of alcohol by dentists: A study of South Australian general dental

practitioners. Aust Dent J 48:102-109.doi:10.1111/j.1834-7819.2003.tb00017.x

WHO (2014)Global status report on alcohol and health.WHO Press, Geneva

WHO (2010)Global recommendations on physical activity for health.WHO Press, Geneva

WHO (2009)Alcohol: Country profiles: Turkey.

http://ec.europa.eu/health/alcohol/policy/country_profiles/turkey_country_profile.pdfAcc

essed on 20 October 2014

ZhuDQ, NormanIJ, WhileAE (2011)The relationship between doctors’ and nurses’ own weight

status and their weight management practices: A systematic review. Obes Rev 12:459-

469. doi:10.1111/j.1467-789X.2010.00821.x

Zis P, Anagnostopoulos F, Sykioti, P (2014) Burnout in medical residents: A study based on the

Job Demands-Resources model. The Scientific World Journal, article ID 673279, 10

pages. doi: 10.1155/2014/673279

30

Table 1. Sample description - health professionals from sevenSouth and South Eastern European countries

Greece Portugal Bulgaria Romania Turkey Croatia Macedonia All

N (%) 594 (22.6) 249 (9.5) 394 (15.0) 411 (15.7) 496 (18.9) 193 (7.4) 286 (10.9) 2623 (100)

Age range Mean SD

19-66

39.6

8.89

23-62

35.2

10.64

22-68

43.3

10.82

23-64

36.8

9.15

22-66

35.3

8.76

20-65

43.9

11.28

20-63

38.1

10.85

19-68

38.7

10.18

Gender Male – N (%) Female – N (%)

198 (33.4)

395 (66.6)

37 (14.9)

212(85.1)

98 (25.9)

280(74.1)

65 (15.9)

343(84.1)

118 (23.8)

378(76.2)

40 (20.7)

153(79.3)

80 (28.2)

204(71.8)

636 (24.5)

1965 (75.5)

Staff position Doctor – N (%) Nurse – N (%) Resident - N(%)

123 (20.7)

280 (47.1)

191 (32.2)

69 (27.7)

167(67.1)

13 (5.2)

140 (35.5)

197(50.0)

57 (14.5)

87 (21.2)

205(49.9)

119 (29.0)

77 (15.5)

301(60.7)

118 (23.8)

60 (31.1)

133(68.9)

0 (0)

71 (24.8)

148(51.7)

67 (23.4)

627 (23.9)

1431 (54.6)

565 (21.5)

Unit tenure range Mean SD

1m – 35y 6y 4m 7y 2m

1y – 36y 8y 7m 8y 3m

1m – 40y 9y 9m

10y 3m

1m – 37y 7y 11m 7y 11m

1y – 42y 8y 5m 8y 1m

6m - 45y 9m 17y 11m 11y 4m

1m – 35y 8y 9m 9y 0m

1m - 45y 9m 8y 10m 9y 0m

31

Table 2. Descriptive statistics for emotional exhaustion, depersonalization and health behaviors in

health professionals from 7 South and South Eastern European countries

Greece Portugal Bulgaria Romania Turkey Croatia Macedonia All

EE Mean (SD)

22.71

(11.55)

16.82(9.53)

21.33

(12.97)

15.61

(10.80)

27.88

(12.88)

17.49

(10.80)

15.89

(11.79)

20.68

(12.52)

Low Average High

26.3%

34.7%

39.1%

48.2%

36.9%

14.9%

35.8%

28.7%

35.5%

54.0%

29.2%

16.8%

17.9%

28.2%

53.8%

44.0%

34.7%

21.2%

50.3%

31.8%

17.8%

36.5%

31.6%

31.9%

DP Mean (SD)

6.95

(5.73)

3.99

(4.44)

4.22

(5.39)

3.98

(4.71)

9.99

(7.71)

3.88

(4.33)

2.94 (4.49)

5.71

(6.16)

Low Average High

25.8%

31.0%

43.3%

47.8%

30.1%

22.1%

53.8%

23.9%

22.3%

51.3%

25.1%

23.6%

20.2%

21.0%

58.9%

51.8%

29.5%

18.7%

65.7%

17.8%

16.4%

41.3%

25.5%

33.2%

EE & DP High

26.4%

9.2%

15.2%

9.7%

45.2%

8.8%

9.1%

20.9%

Fast food Mean (SD)

3.57

(2.65)

0.95

(1.50)

3.80

(2.52)

1.02

(1.63)

2.44

(2.40)

1.01

(1.40)

2.28

(1.63)

2.41

(2.43)

Exercise Mean (SD)

1.26

(1.72)

1.20

(1.59)

1.47

(1.88)

1.83

(2.08)

1.38

(1.89)

1.30

(2.01)

1.69

(1.97)

1.45

(1.89)

Alcohol Mean (SD)

1.22

(1.35)

1.04

(1.79)

1.47

(1.70)

0.67

(1.07)

0.87

(1.23)

0.56

(1.13)

0.70

(1.32)

0.98

(1.41)

Painkillers Mean (SD)

0.97

(1.40)

0.81

(1.70)

0.82

(1.53)

0.78

(1.29)

2.03

(1.75)

0.87

(1.31)

0.91

(1.50)

1.03

(1.54)

Tertiles of burnout dimensions: EE: high ≥ 26, low ≤ 14; DP: high ≥ 7, low ≤ 2.

32

Table 3. Hierarchicalregression of the relation betweenemotional exhaustion and health behaviours in health professionals – overall sample

b SE β t F R2 ∆ R2

Fa

st

foo

d

Model 1 EE

Model 2 EE Age Gender Staff position Unit tenure

Model 3 EE Age Gender Staff position Unit tenure Portugal Bulgaria Romania Turkey Croatia Macedonia

0.03

0.03 - 0.00 - 0.54 - 0.42 - 0.00

0.01 - 0.04 - 0.24 - 0.40 0.00 - 2.64 0.52 - 2.53 - 1.41 - 2.28 - 1.22

0.00

0.00 0.01 0.13 0.12 0.00

0.00 0.01 0.12 0.11 0.00 0.23 0.15 0.15 0.14 0.19 0.16

0.14***

0.14***

- 0.01 - 0.10*** - 0.09*** - 0.12***

0.07**

- 0.19*** - 0.04*

- 0.08*** 0.03

- 0.24*** 0.07**

- 0.39*** - 0.23*** - 0.26*** - 0.16***

6.65

6.92 - 0.40 - 4.06 - 3.60 - 4.32

3.36 - 6.90 - 1.99 - 3.77 1.22

- 11.52 3.35

- 16.72 - 9.79

- 11.76 - 7.45

44.18***

30.43***

65.88***

0.02

0.07

0.25

0.02***

0.05***

0.18***

Exe

rcis

e

Model 1 EE



- 0.03

- 0.03 0.00 - 0.42 0.01 0.00

- 0.03 0.00 - 0.47 0.06 0.00 - 0.15 0.23 0.44 0.28 - 0.09 0.28

0.00

0.00 0.01 0.10 0.09 0.00

0.00 0.01 0.11 0.09 0.00 0.20 0.13 0.13 0.13 0.17 0.14

- 0.17***

- 0.17***

- 0.01 - 0.10***

0.00 0.03

- 0.17**

0.00 - 0.11***

0.02 0.02 - 0.02 0.04

0.09** 0.06* - 0.01 0.05*

- 8.16

- 8.16 - 0.50 - 3.99 0.13 0.25

- 7.46 - 0.01 - 4.43 0.62 0.69 - 0.76 1.70 3.36 2.18 - 0.53 2.00

70.64***

18.26***

10.30***

0.03

0.04

0.05

0.03***

0.01***

0.01**

Alc

oh

ol

Model 1 EE



0.01

0.01 0.02 - 0.64 - 0.36 - 0.00

0.01 0.01 - 0.59 - 0.36 0.00 - 0.19 0.27 - 0.40 - 0.26 - 0.54 - 0.42

0.00

0.00 0.00 0.07 0.06 0.00

0.00 0.00 0.07 0.06 0.00 0.14 0.09 0.09 0.09 0.12 0.10

0.06**

0.08*** 0.11*** - 0.20*** - 0.13***

- 0.04

0.05* 0.04

- 0.19*** - 0.13***

0.03 - 0.03

0.07** - 0.11*** - 0.07**

- 0.11*** - 0.10***

2.90

3.73 4.03 - 8.86 - 5.73 - 1.45

2.23 1.45 - 8.28 - 5.77 1.01 - 1.43 3.04 - 4.44 - 2.97 - 4.71 - 4.39

8.40**

48.49***

29.93***

0.00

0.10

0.13

0.00**

0.10***

0.03***

33

Pa

inkill

ers

Model 1 EE



0.03

0.03 0.00 0.38 0.08 0.00

0.02 0.01 0.41 0.05 0.00 - 0.13 - 0.21 - 0.10 0.81 - 0.12 0.07

0.00

0.00 0.00 0.08 0.07 0.00

0.00 0.00 0.08 0.07 0.00 0.16 0.11 0.10 0.11 0.13 0.11

0.26***

0.25***

0.01 0.11***

0.03 0.04

0.19***

0.05 0.12***

0.02 0.02 - 0.02

- 0.05* - 0.02

0.19*** - 0.02 0.02

12.21

12.03 0.19 4.59 1.14 1.44

8.68 1.84 4.97 0.73 0.54 - 0.81 - 2.00 - 0.95 7.51 - 0.93 0.60

148.99***

38.46***

26.60***

0.07

0.09

0.12

0.07***

0.02***

0.04***

EE, age and unit tenure are included as continuous variables. Dichotomous categorical variables are coded as: gender (man = 0, woman = 1), staff position (doctor/resident = 0, nurse = 1). Categorical variable with 7 levels country is dummy coded with Greece being the reference category. Constant is included. b – unstandardized regression coefficient; SE – standard error of the regression coefficient; β – standardized regression coefficient; t –test of each predictor’s significance; F –test of overall model’s significance; R2– coefficient of determination (proportion of explained variance), ∆ R2 - R Square Change. Levels of statistical significance:***p<0.001; **p<0.01; *p<0.05.

34

Table 4. Hierarchicalregression of the relation betweendepersonalization and health behaviours in

health professionals – overall sample

b SE β t F R2 ∆ R2

Fa

st

foo

d

Model 1 DP

Model 2 DP Age Gender Staff position Unit tenure

Model 3 DP Age Gender Staff position Unit tenure Portugal Bulgaria Romania Turkey Croatia Macedonia

0.07

0.06 0.00 - 0.49 - 0.38 - 0.00

0.05 - 0.04 - 0.22 - 0.40 0.00 - 2.67 0.60 - 2.49 - 1.48 - 2.37 - 1.12

0.01

0.01 0.01 0.13 0.12 0.00

0.01 0.01 0.12 0.11 0.00 0.23 0.15 0.15 0.14 0.19 0.16

0.16***

0.14***

0.00 - 0.09*** - 0.08**

- 0.12***

0.12*** - 0.18*** - 0.07**

- 0.08*** 0.04

- 0.24*** 0.09*** - 0.38*** - 0.24*** - 0.26*** - 0.15***

7.80

6.83 0.14 - 3.71 - 3.22 - 4.32

5.92 - 6.62 - 3.34 - 3.77 1.31

- 11.80 3.91

- 16.65 - 10.27 - 11.62 - 6.88

60.83***

30.17***

68.74***

0.03

0.06

0.26

0.03***

0.04***

0.19***

Exe

rcis

e

Model 1 DP



- 0.04

- 0.04 - 0.01 - 0.46 - 0.02 0.00

- 0.03 - 0.00 - 0.52 0.03 0.00 - 0.02 0.19 0.53 0.26 - 0.04 0.33

0.01

0.01 0.01 0.11 0.09 0.00

0.01 0.01 0.11 0.09 0.00 0.20 0.14 0.13 0.13 0.17 0.14

- 0.12***

- 0.12***

- 0.03 - 0.11***

- 0.01 0.03

- 0.11**

- 0.01 - 0.12***

0.01 0.02 - 0.00 0.04

0.10*** 0.05* - 0.01 0.06*

- 5.54

- 5.73 - 0.96 - 4.41 - 0.21 0.94

- 4.83 - 0.23 - 4.93 0.32 0.64 - 0.09 1.42 4.02 2.03 - 0.21 2.28

30.66***

11.45***

7.30***

0.01

0.03

0.04

0.01***

0.01***

0.01**

Alc

oh

ol

Model 1 DP



0.03

0.02 0.02 - 0.62 - 0.35 0.00

0.02 0.01 - 0.59 - 0.34 0.00 - 0.20 0.32 - 0.37 - 0.30 - 0.51 - 0.37

0.01

0.01 0.00 0.07 0.06 0.00

0.01 0.00 0.07 0.06 0.00 0.13 0.09 0.09 0.09 0.11 0.10

0.12***

0.11*** 0.12*** - 0.20*** - 0.13***

- 0.04

0.10*** 0.05

- 0.19*** - 0.13***

0.03 - 0.03

0.08*** - 0.10*** - 0.08**

- 0.11*** - 0.09***

5.45

5.29 4.41 - 8.65 - 5.46 - 1.39

4.67 1.67 - 8.24 - 5.42 1.09 - 1.46 3.54 - 4.13 - 3.38 - 4.47 - 3.80

29.66***

51.60***

31.69***

0.01

0.11

0.14

0.01***

0.09***

0.03***

35

Pa

inkill

ers

Model 1 DP



0.05

0.05 0.00 0.44 0.12 0.00

0.03 0.01 0.46 0.08 0.00 - 0.25 - 0.17 - 0.18 0.85 - 0.17 0.03

0.01

0.01 0.00 0.09 0.07 0.00

0.01 0.00 0.08 0.07 0.00 0.16 0.11 0.10 0.11 0.13 0.11

0.19***

0.20***

0.03 0.12***

0.04 0.04

0.13*** 0.06*

0.13*** 0.03 0.02 - 0.04 - 0.04 - 0.04

0.20*** - 0.03 0.01

8.66

9.35 0.86 5.20 1.64 1.41

5.58 2.05 5.55 1.07 0.64 - 1.60 - 1.61 - 1.71 7.78 - 1.31 0.26

74.91***

26.78***

22.17***

0.04

0.06

0.11

0.04***

0.03***

0.05***

DP, age and unit tenure are included as continuous variables. Dichotomous categorical variables are coded as: gender (man = 0, woman = 1), staff position (doctor/resident = 0, nurse = 1). Categorical variable with 7 levels country is dummy coded with Greece being the reference category. Constant is included. b – unstandardized regression coefficient; SE – standard error of the regression coefficient; β – standardized regression coefficient; t – test of each predictor’s significance; F –test of overall model’s significance; R2 – coefficient of determination (proportion of explained variance), ∆ R2 - R Square Change. Levels of statistical significance:***p<0.001; **p<0.01; *p<0.05.

36

Table 5. Regression models of the association betweenemotional exhaustion and depersonalization and health behaviours in health professionals from sevenSouth and South Eastern European

countries

Greece Portugal Bulgaria Romania Turkey Croatia Macedonia

EE

β 0.05 - 0.09 0.07 - 0.01 0.18*** 0.02 0.01

R2 0.00 0.01 0.01 0.00 0.03 0.00 0.00

F 1.24 0.81 1.40 0.01 13.69*** 0.09 0.02

DP

β 0.16*** 0.07 0.12* 0.08 0.25** - 0.11 0.06

R2 0.03 0.01 0.01 0.01 0.06 0.01 0.00

Fa

st f

oo

d

F 13.90*** 0.49 4.23* 2.19 29.46*** 2.18 0.91

EE

β - 0.17*** - 0.26** - 0.19** - 0.13** - 0.25*** - 0.07 - 0.09

R2 0.03 0.07 0.04 0.02 0.06 0.01 0.01

F 17.57*** 8.18** 11.12** 5.96** 26.52*** 0.91 2.39

DP

β - 0.09* - 0.09 - 0.06 - 0.14** - 0.17*** - 0.01 - 0.05

R2 0.01 0.01 0.00 0.02 0.03 0.00 0.00

Exe

rcis

e

F 4.63* 0.87 1.16 6.79** 12.36*** 0.04 0.61

EE

β - 0.04 0.04 0.04 0.14** 0.03 - 0.01 0.04

R2 0.00 0.00 0.00 0.02 0.00 0.00 0.00

F 1.03 0.18 0.48 7.49** 0.32 0.01 0.49

DP

β 0.08 - 0.06 0.15** 0.21*** 0.12* - 0.04 0.15*

R2 0.01 0.00 0.02 0.04 0.02 0.00 0.02

Alc

oh

ol

F 3.20 0.43 6.75** 17.12*** 5.91* 0.33 5.87*

EE

β 0.24*** - 0.05 0.15** 0.23*** 0.26*** 0.17* 0.14*

R2 0.06 0.00 0.02 0.05 0.07 0.03 0.02

F 32.52*** 0.26 7.00** 21.19*** 21.50*** 5.76* 5.28*

DP

β 0.06 - 0.07 0.10 0.07 0.26*** 0.14 0.04

R2 0.00 0.00 0.01 0.01 0.07 0.02 0.00

Pa

ink

ille

rs

F 1.77 0.45 3.13 1.70 21.50*** 3.67 0.46

Only results from the linear regression models including emotional exhaustion or depersonalization as thesingle predictor variable (Model 1) are presented here. Constant is included.

Levels of statistical significance:***p<0.001; **p<0.01; *p<0.05.

burnout and health behaviors in health professionals from

Documents