CME AVAILABLE FOR THIS ARTICLE AT ACOEM.
ORG
The Mediating Role of Psychophysic Strain in the Relationship
Between Workaholism, Job Performance, and Sickness Absence
A Longitudinal Study
Alessandra Falco, PhD, Damiano Girardi, PhD, Luca Kravina, PhD, Elena Trifiletti, PhD,
Giovanni Battista Bartolucci, MD, Dora Capozza, Prof, and Nicola A. De Carlo, Prof
Objective: To test a theoretical model in which workaholism predicts both
directly and indirectly, via psychophysic strain, job performance and sick-
ness absences. Methods: A multimethod study was performed examining
a sample of 322 workers in a private company. The study was articulated
into two phases, over a time period of 15 months. Workaholism was assessed
using a self-report measure (time 1). Psychophysic strain was measured by
the occupational physician, performance by the supervisor, and data on sick-
ness absences were collected from the companys database (time 2). Results:
Results highlighted a positive relationship between workaholism and psy-
chophysic strain. Psychophysic strain was negatively associated with job
performance and positively associated with sickness absences. In addition,
workaholism predicted sickness absences. Conclusion: Workaholism nega-
tively affects the health of workers. This is associated with lower working
performance and greater sickness absences.
W orkaholism, or the tendency to work excessively hard in
a compulsive way,1 is a construct that is widespread and
observed all over the world. The prevalence of workaholism varies
between 8% and 17.5% in different samples of graduate workers with
peaks of 25% in samples of professionals such as doctors, lawyers,
and psychologists.2 Other sociodemographic variables (eg, sex, age)
do not seem to have a stable link with workaholism but are affected
by the composition of the sample, the different conceptualizations
of workaholism, and the different instruments used.3
According to McMillan and ODriscoll,3 workaholism has
a multifactor origin. From an etiological point of view, in fact, it
is difficult to find a univocal explanation to the onset of worka-
holism: personal dispositions, sociocultural experiences, behavioral
structures, and vicarious learning can all be factors that favor the
onset of workaholism.2 For this reason, according to the literature,
it is necessary to analyze workaholism considering its different as-
pects (eg, cognitive, affective, behavioral)4 using a multidisciplinary
approach.3
In accordance with the aforementioned definition by Schaufeli
et al,1 it is possible to identify the two central nuclei of workaholism:
working excessively (WE) and working compulsively (WC). It is
From the FISPPA Section of Applied Psychology (Drs Falco, Girardi, and Krav-
ina, Prof Capozza, and Prof De Carlo), University of Padova, Padova, Italy;
Department of Philosophy, Education and Psychology (Dr Trifiletti), Univer-
sity of Verona, Verona, Italy; and Department of Cardiologic, Thoracic and
Vascular Sciences (Prof Bartolucci), University of Padova, Padova, Italy.
The authors declare no conflict of interest. No funding is declared.
Authors Falco, Girardi, Kravina, Trifiletti, Bartolucci, Capozza, and De Carloand
have no relationships/conditions/circumstances that present potential conflict
of interest.
The JOEM Editorial Board and planners have no financial information related to
this research.
Address correspondence to: Alessandra Falco, PhD, FISPPA Section of Ap-
plied Psychology, University of Padova, Via Venezia, 8-35131 Padova, Italy
(alessandra.falco@unipd.it).
Copyright  C 2013 by American College of Occupational and Environmental
Medicine
DOI: 10.1097/JOM.0000000000000007
JOEM r Volume 55, Number 11, November 2013
Copyright 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Learning Objectives
r Discuss the definition and multifactorial origin of worka-
holism, along with its "two central nuclei."
r Summarize the new findings on the relationship between
workaholism and psycho-physic strain, including the effects
on sickness absence and job performance.
r Discuss the study implications for the occupational physi-
cians role in primary, secondary, and tertiary prevention of
workaholism.
important to underline that workaholism is characterized by a si-
multaneous presence of high levels of the two dimensions.1 As un-
derlined by Schaufeli et al,1 both WE and WC lead workers to be
particularly involved in their work. On the one hand, WE leads work-
ers to take on a considerable amount of work (eg, overwork), whereas
WC leads to an obsessive commitment to work, which particularly
involves mental and emotional demands. Shimazu and Schaufeli5
identify WE and WC, respectively, as the behavioral and cognitive
dimensions of workaholism.
According to McMillan et al,6 workaholism can be considered
as a stable characteristic of the individual. Workaholism affects styles
of coping,5 cognitive aspects (eg, obsession, rumination),4,7 and
emotiveaffective aspects of the individual (eg, obsessive passion),8
as well as typical individual behaviors (eg, working a great num-
ber of hours). Workaholism is moreover associated with numerous
negative consequences at an individual (eg, psychophysic strain,
emotional breakdown),1 familial (eg, workfamily conflict),1 and
organizational levels (eg, poor social relationships).4
This study refers to the theoretical framework proposed by
Hart and Cooper,9 according to which both organizational character-
istics and individual characteristics contribute to the well-being of
the worker. This, in turn, determines outcomes both at an individual
level (eg, job performance [JP]; sickness absences [SAs]) and at an
organizational level (eg, customer satisfaction, productivity loss).
As well as this indirect effect, organizational and individual charac-
teristics also have a direct link with these outcomes. According to
the authors, organizational characteristics are the objective aspects
of the work environment or the workers subjective perceptions re-
garding this environment. Individual characteristics are intended as
factors associated with individual differences among workers, such
as personality dimensions, styles of coping, individual attitudes, and
behaviors.
It is claimed, therefore, that workaholism can be considered
an individual characteristic,6 capable of influencing the health and
performance of the worker. The aim of the present work consists,
therefore, in testing a theoretical model in which workaholism pre-
dicts psychophysic strain, which, in turn, is related to JP and SAs (ie,
psychophysic strain mediates the relation between workaholism and
JP/SAs). In addition, workaholism is expected to exercise a direct
effect on JP and SAs.
1255
Falco et al
WORKAHOLISM AND PSYCHOPHYSIC STRAIN
In literature, numerous studies have highlighted a positive re-
lation between workaholism and psychophysic strain.5,1012 By defi-
nition, workaholics (high WE and high WC) dedicate a considerable
amount of time to their work (eg, work overtime, work at home in the
evening).10 Prolonged exposure to job demands means insufficient
possibility for a complete recovery during after-work hours.7 This
determines a breakdown at an emotive or cognitive level and, in the
long run, the appearance of psychophysic symptoms.13,14
This can be explained on the basis of effort recovery model,15
according to which the performance of ones work requires efforts
that must be followed by an adequate psychophysiological unwinding
during after-work hours (ie, recovery). If this does not occur, the
worker finds himself or herself in a suboptimal situation (ie, sustained
activation) and must invest compensatory effort to achieve adequate
performances.16,17 This affects the necessity for a greater recovery
time. Thus, a vicious circle is established, which can, with time,
cause health problems.7,15,16,18
In addition, workaholics think obsessively about their work
also during after-work hours (perseverative thinking).4 This can pro-
voke emotional distress,13 feelings of guilt, and withdrawal symp-
toms (eg, anxiety symptoms) in the moment in which it is impossi-
ble for workaholics to dedicate themselves to their work.4 To avoid
these negative emotions, workaholics think about work also during
after-work hours, which determines less possibility for recovery ex-
periences, in terms of both psychological detachment from work
and relaxation.18 Psychological detachment from work refers to the
ability of individuals to mentally switch off from work. Psycho-
logical detachment implies not only being absent from work, but also
neither working at home (eg, checking e-mails) nor thinking about
job-related issues (eg, future tasks) during after work hours.1821
Relaxation is a process characterized by low psychological, phys-
ical, and physiological activation that is positively associated with
positive affect. Relaxation can be achieved by practicing relaxation
techniques or by engaging in activities that require little physical or
intellectual effort (eg, a light walk, listening to music).18,2123
Lack of psychological detachment and relaxation may impair
the recovery process. Indeed, a negative association has been ob-
served between recovery experiences (ie, psychological detachment
from work and relaxation) and fatigue, health complaints, emotional
exhaustion, and sleep problems.20,21,24
Therefore, we hypothesize that workaholism will be positively
related to psychophysic strain (Hypothesis 1).
PSYCHOPHYSIC STRAIN, JOB PERFORMANCE, AND
SICKNESS ABSENCES
Literature has evidenced a negative relation between psy-
chophysic strain and JP.25,26 From a theoretical point of view, this
can be explained by the fact that people invest resources to carry
out their work, in terms of skills, general cognitive abilities, and
task-relevant knowledge.27 These resources are limited, so the con-
cept of resource allocation is central. If an individual invests his
or her own resources entirely in work, he or she will obtain high
levels of performance. Vice versa, if the resources are destined else-
where, JP will suffer.27 In this perspective, negative affective states
connected to anxiety and depression (ie, psychological strain)25 can
induce the individual to divert his or her resources from perform-
ing the task toward the management of negative emotions (ie, affect
regulation),27,28 or they can generate perseverative thinking about
his or her state of health (ie, rumination).7 This leads to the use
of cognitive resources for activities not linked to performing ones
work (ie, task-irrelevant thoughts) and, consequently, lower JP.27 In
literature, moreover, other mechanisms are described through which
the negative affective states connected to psychological strain can
influence performance. For example, negative affective states can
1256 
C 2013 American College of Occupational and Environmental Medicine
Copyright 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
JOEM r Volume 55, Number 11, November 2013
negatively influence the self-efficacy of the worker, motivation, and,
therefore, JP.29
An analogous situation can be revealed in relation to phys-
ical strain, in which the pain or discomfort generated by physical
symptoms can interfere with the individuals cognitive resources,
generating rumination or shifting the attention of the individual to-
ward the specific symptom.25
In accordance with the theoretical framework proposed by
Hart and Cooper,9 SAs are considered a consequence of psy-
chophysic strain. Sickness absences mean the nonattendance of an
employee at work because of health complaints when this atten-
dance is instead expected by the employer.30 Sickness absences are
considered a valid indicator of the health of the worker.31 Some stud-
ies have, in fact, evidenced an association between SAs and future
long-term self-rated health status,32 future disability pension,33 and
mortality.34
From an empirical point of view, numerous studies have shown
a positive relation between psychophysic strain and SAs.30,31,35 Ac-
cording to Darr and Johns,35 work stress can be seen as a pro-
cess in which the individual gradually develops psychological (ie,
psychological strain), physical (ie, physical strain), and behavioral
responses (ie, SAs) when faced with characteristics of the work
environment perceived as threatening or potentially harmful. In
their meta-analysis, Darr and Johns35 demonstrated how the per-
ception of stressors determines psychophysic strain and, conse-
quently, SAs. Overall, we therefore hypothesize that psychophysic
strain will be negatively related to JP and positively related to SAs
(Hypothesis 2).
WORKAHOLISM, JOB PERFORMANCE, AND
SICKNESS ABSENCES
In literature, some studies have considered the relation be-
tween workaholism and JP. According to Schaufeli et al,11 worka-
holics work hard rather than smart. They create difficulties for them-
selves and their colleagues, are rigid, inflexible, and perfectionist,
as well as not inclined to delegate. This, in the long run, can create
conflict and friction in the workplace, which determines low social
support and therefore low JP.4,11,36,37
In addition, as previously mentioned, JP depends on the re-
sources that an individual dedicates to his or her work (ie, resource
allocation).27 According to Gorgievski and Bakker,36 workaholism is
characterized by an obsessive passion for ones job and is associated
with negative emotions (eg, guilt or feeling hassled) while or after
carrying out ones job. As a consequence, workaholics may use a
part of their cognitive resources to deal with these negative emotions
(ie, affect regulation),27,28 determining lower JP.
Apart from this cognitive strategy, workaholics can also resort
to a behavioral strategy. According to van Wijhe et al,18 workaholics
interpret their own negative emotions as dissatisfaction with their
performance (ie, sensation of not having done enough) and try to
handle these emotions by remaining mentally or physically involved
in work during after-work hours (eg, by creating extra work). This
precludes the possibility of an optimal recovery during the evening
and determines more negative emotions the day after which, in turn,
results in lower JP.
In addition, some authors proposed that psychophysic strain
could mediate the relationship between workaholism and JP. Accord-
ing to Ng et al,4 indeed, workaholics work with energy and dedicate
a lot of time to work over a short-term period. Nevertheless, in the
long run, workaholics can undergo a decline in the state of their
psychophysic health, which determines lower JP.4
Overall, we expect that workaholism exerts both a direct and
an indirect effect (through psychophysic strain) on JP. In particu-
lar, we expect workaholism to positively predict in the long run
psychophysic strain, which, in turn, is negatively related to JP (ie,
JOEM r Volume 55, Number 11, November 2013
negative indirect effect). In addition, workaholics, because of their
perfectionism and obsessive passion for work,36,38 may use part of
their cognitive resources to deal with negative emotions (such as
guilt or feeling hassled) and receive in the long run low social sup-
port from colleagues and supervisors, which negatively affects JP (ie,
negative direct effect). Therefore, we hypothesize that psychophysic
strain partially mediates the relationship between workaholism
and JP.
H3a: Workaholism exerts a negative indirect effect on job perfor-
mance through psychophysic strain.
H3b: Workaholism exerts a negative direct (residual) effect on job
performance.
To date, only a few studies have considered the relation be-
tween workaholism and SAs.11,39 Nevertheless, some characteristics
of workaholics lead to believe that a direct negative relation exists
between workaholism and SAs.1 Workaholics, because of their ob-
sessive stimulus to work, are reluctant to disengage from work,11
inclined to disregard the eventual occurrence of symptoms of psy-
chophysical strain to dedicate themselves to work.6 Workaholics, as
a result, are reluctant to be absent even when they are ill, putting into
practice behaviors representative of presenteeism (ie, going to work
even when ill).1,40,41
Overall, we expect that workaholism exerts both a direct ef-
fect and an indirect effect (through psychophysic strain) on SAs. In
particular, we expect workaholism to positively predict psychophysic
strain, which, in turn, is positively related to SAs (ie, positive indirect
effect). In addition, because workaholics are reluctant to be absent
from work even when they are ill, we hypothesize that workaholism
negatively predicts SAs (ie, negative direct effect). Therefore, we hy-
pothesize that psychophysic strain partially mediates the relationship
between workaholism and SAs.
H4a: Workaholism exerts a positive indirect effect on sickness ab-
sences through psychophysic strain.
H4b: Workaholism exerts a negative direct (residual) effect on sick-
ness absences.
METHODS
Participants and procedure
We conducted a longitudinal study in a mechanical engi-
neering company located in North-East Italy. A sample of workers
(N = 322) was obtained, through probability sampling, from the
firms workforce of 500 workers. The sample consisted of 86 women
and 234 men (2 missing data). Most respondents were aged between
40 and 50 years (45.5%), 38.5% were younger than 40 years, and
15.2% were older than 50 years (3 missing data). For the work po-
sition, 52.8% were blue collars and 45.9% were white collars or
managers (4 missing data). The majority (85.4%) had been with the
company for more than 5 years (13.7% less than 5 years; 3 missing
data).
Participants took part in the study on a voluntary basis. They
were informed in advance by the management that the research con-
cerned work-related stress and was articulated in two phases, over a
period of 15 months. In December 2010 (time 1), workers completed
a baseline questionnaire including a self-report measure of worka-
holism and possible confounders (ie, age, sex, job position, length
of service). The questionnaire was administered in group sessions of
approximately 40 workers, under the supervision of an occupational
physician and an occupational psychologist. Each worker was given
an identification code to connect the self-report measure to subse-
quent measures (psychophysic strain, SAs, and JP). This procedure
took place after having obtained the written consensus of the worker,
whose anonymity was guaranteed by the occupational physician and
the occupational psychologist.

C 2013 American College of Occupational and Environmental Medicine
Copyright 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Workaholism and Performance: The Role of Strain
In February 2012 (time 2), workers were convened by the
occupational physician who assessed psychophysical strain using a
structured interview.42,43 The evaluation of workers performance
was supplied by the supervisor, in the context of the evaluation
of annual performances. In particular, the supervisor expressed a
judgment for each worker, using a 10-point scale. The evaluation
concerned the quality and quantity of the results obtained by the
worker in the course of 2011. Finally, data concerning SAs for 2011
were obtained directly from the firms database.
Measures
Workaholism was measured with the Italian version12,44 of
the Dutch Workaholism Scale.1 The scale includes two dimen-
sions: working excessively (six items) and working compulsively
(four items). Sample items are as follows: I seem to be in a hurry
and racing against the clock, I spend more time working than on
socializing with friends, on hobbies, or on leisure activities (WE);
I feel that theres something inside me that drives me to work hard,
I feel obliged to work hard, even when it is not enjoyable (WC).
Participants were asked to indicate how often they had these feelings
on a six-point frequency scale, ranging from 1 (never) to 6 (always).
Cronbach was 0.77 for WE, 0.84 for WC, and 0.84 for the overall
scale of workaholism.
Job strain was measured using 17 physical and psychological
symptoms taken from a scale developed in the Italian context.42,45,46
An occupational physician rated how often, during the previous 6
months, the worker felt each symptom, as a consequence of situations
related to work. Examples of items are nausea, slow and difficult
digestion, memory problems, and sleep problems. The occupational
physician answered on a six-point frequency scale, ranging from 1
(never) to 6 (always). Cronbach was 0.87.
Workers performance was measured by asking their super-
visor to answer the following question: In your opinion, in what
percentage the employee X has accomplished his/her goals during
the last year? The 10-point response scale ranged from 1 (10%) to
10 (100%).
The number of days of sickness absence for 2011 was ob-
tained from the companys records. The measure was split into two
indicators, corresponding to the absence days for the first and second
2011 semesters.
Data Analysis
Descriptive statistics and correlations between variables were
obtained with SPSS 17. Hypotheses were tested using regression
with latent variables (LISREL 8.7).47 In the regression model, worka-
holism predicts psychophysic strain, which, in turn, is related to JP
and absenteeism. In addition, also the direct paths from workaholism
to the outcome variables were estimated (see Fig. 1). Two indicators
were created for workaholism, corresponding to the WE and WC
dimensions. As to job strain, three indicators were obtained after
the procedure of item-to-construct balance.48 We tested a one-factor
model to obtain the item loadings. Using the three items with the
highest loadings as anchors, we added to them, in an inverted order,
the three items with the next highest loadings. This procedure was
continued by allocating items with lower loadings to higher loaded
parcels. For SAs, the two indicators were the number of absence
days in the first and second semesters of 2011. The error term of
performance (single-item measure) was constrained to zero.
The goodness of fit of the model was evaluated using the
chi-squared test. A model fits the data well when chi-squared test is
nonsignificant. Three additional goodness-of-fit indices were used:
the comparative fit index (CFI), the standardized root mean squared
residual (SRMR), and the root-mean-square error of approximation
(RMSEA). Hu and Bentler49 suggested that the fit is satisfactory
when CFI is equal to or greater than 0.95, and SRMR and RMSEA
are equal to or less than 0.08. The reliability of mediation was tested
1257
Falco et al JOEM r Volume 55, Number 11, November 2013

0.18** Performance y4

0.39***
Psychophysic
Workaholism FIGURE 1. Regression model of the
strain
effects of workaholism on workers
0.33***
Sickness y5
performance and sickness absences
x1 x2 y1 y2 y3 absences through psychophysic strain (n = 322).
y6 Goodness-of-fit indices for the model
are 2 (15) = 14.12, P = 0.52; CFI =
1.00; SRMR = 0.02; RMSEA = 0.00. Only
significant parameters (completely
standardized) are reported. **P< 0.01.
0.40*** ***P< 0.001.

using the Sobel test.50 Additional regression models were performed
with SPSS 17.0 to control for the effects of sex, age, job position, and
length of service. These control variables were dummy coded. For sex
and job position, a single dummy variable was created, assigning 0
to women and blue collars, and 1 to men and white collars/managers,
respectively. For age and length of service, two dummies were cre-
ated. For age, in the first dummy, 0 was assigned to between 40 and
50 and over 50 and 1 was assigned to below 40; in the second
dummy, 0 was assigned to below 40 and over 50 and 1 was
assigned to between 40 and 50. For length of service, in the first
dummy, 0 was assigned to below 5 years and between 5 and 19
years and 1 was assigned to over 19 years; in the second dummy,
0 was assigned to below 5 years and over 19 years and 1 was
assigned to between 5 and 19 years. We performed four regres-
sion models. In the first, workaholism was the dependent variable;
sex, age, job position, and length of service were the predictors. In
the second model, psychophysic strain was the dependent variable;
workaholism and the control variables were entered as predictors. In
the third and fourth models, JP and SAs were the dependent vari-
ables; the predictors were workaholism, psychophysic strain, and the
control variables.
All analyses were performed with LISREL 8.747 and SPSS
17.0. The null hypothesis was rejected when P < 0.05.
RESULTS
Descriptive statistics and correlations between variables are
reported in Table 1. Participants reported moderate levels of worka-
holism. The physician evaluated the psychophysic symptoms as in-
frequent. Moreover, employees were absent from work because of
sickness for approximately 1 week and they were evaluated by their
supervisor as having accomplished most of their goals. Correla-
tions between variables were in the expected direction, except for
the association between workaholism and JP, which was nonsignifi-
cant. Nevertheless, the absence of a significant relationship between
workaholism and JP does not preclude the testing of the media-
tion model.51 The possibility that mediation can exist even if the
independent and dependent variables are not significantly related is
TABLE 1. Correlations Between Study Variables (n = 322)
Variable Mean SD 1 2
acknowledged by several authors.5154 In fact, it is possible that a
suppression effect has occurred. Within a mediation model, a sup-
pression effect is present when the direct (residual) and indirect
effects have opposite signs and are similar in magnitude.5153 Given
that the total effect (not mediated) is the sum of the indirect and
residual direct effects, when a suppression process occurs, the direct
link between the independent variable and the dependent variable
will be close to zero.
The hypothesized mediation model showed an excellent fit:
2 (15) = 14.12, P = 0.52; CFI = 1.00; SRMR = 0.02; RMSEA =
0.00. As shown in Figure 1, in which only significant parameters
are reported, workaholism was positively related to psychophysic
strain (Hypothesis 1), which, in turn, is related to lower levels of
performance (Hypothesis 2); the indirect effect was 0.10 (95% CI
0.18 to 0.02). The application of Sobel test showed that mediation
was significant (z = 2.25, P = 0.024), thus confirming Hypothesis 3a.
Contrary to Hypothesis 3b, the direct link between workaholism and
JP was not significant. In addition, results showed that the indirect
and residual direct effects were similar in magnitude (0.10 and
0.05, respectively), but opposite in sign, suggesting the presence
of a suppression effect. This explains why the bivariate correlation
between workaholism and JP was not significant (see Table 1).
As to SAs, psychophysic strain was positively associated with
absence days (Hypothesis 2); the indirect effect was 0.91 (95% CI,
0.29 to 1.23). The Sobel test confirmed that mediation was signifi-
cant (z = 3.13, P = 0.002); Hypothesis 4a was, therefore, confirmed.
Moreover, consistent with Hypothesis 4b, a direct negative effect of
workaholism on SAs was found. Therefore, as expected, the direct re-
lationship between workaholism and absences was negative, whereas
the indirect relationship via psychophysic strain was positive.
We also controlled for the effects of age, sex, job position,
and length of service. Results showed that including these variables
as predictors in the regression did not alter the relationships between
workaholism, psychophysic strain, JP, and SAs. Nevertheless, job
position was significantly related to workaholism, with white collars
and managers reporting higher workaholism than blue collars (b =
0.29, t = 5.11, P < 0.001); SAs were, in contrast, higher for blue
collars than for white collars and managers (b = 0.23, t = 3.98,
P < 0.001). A significant effect of sex on psychophysic strain also
emerged: strain symptoms were higher among women than among
men (b = 0.24, t = 4.64, P < 0.001). No significant effects of
demographic variables on performance were found.
3 4

1. Workaholism 3.34 0.93 ...

2. Psychophysic strain 1.57
3. Job performance 6.97
4. Sickness absences 7.32 11.95 0.22***
**P 0.01. ***P < 0.001.
DISCUSSION
In this study, as expected, a positive relationship between
0.42 0.32*** ...
workaholism and psychophysic strain has been revealed (Hypothe-
1.26 0.02 0.16** . . .
sis 1). Workaholics dedicate a lot of time to their work and think
0.14** 0.02 . . . obsessively about it during after-work hours. These inclinations de-
termine less opportunity for recovery and, in the long run, health
problems.16 We also found a direct and negative relation between

1258 
C 2013 American College of Occupational and Environmental Medicine

Copyright 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
JOEM r Volume 55, Number 11, November 2013
psychophysic strain and JP, as well as a direct and positive relation
between psychophysic strain and SA (Hypothesis 2).
Hypothesis 3, which stated that psychophysic strain partially
mediates the relationship between workaholism and JP, was par-
tially supported. Indeed, the indirect effect of workaholism on JP
through psychophysic strain was significant (Hypothesis 3a sup-
ported), whereas the direct effect was not (Hypothesis 3b not sup-
ported). Therefore, psychophysic strain fully mediates the relation-
ship between workaholism and JP. This study highlights the central
role of psychophysic strain in the relation between workaholism and
JP. Indeed, in the long run, workaholics can undergo a decline in the
state of their psychophysic health, which determines lower JP.4 This
is a strong point of our work, because previous studies, which ana-
lyzed the relationship between workaholism and JP, but did not take
into account impaired psychophysic health as a mediator, showed
inconsistent findings. For example, in two cross-sectional studies,
Shimazu and Schaufeli5 found a weak negative association between
workaholism and JP whereas Schaufeli et al11 found a weak positive
association between workaholism and extra-role performance (ie,
activities that promote organizational effectiveness but are not part
of employees formal role requirements),55 while they found no asso-
ciation between workaholism and in-role performance (ie, activities
that are related to employees formal role requirements).55 In addi-
tion, in a longitudinal study, Shimazu et al10 found no association
between workaholism and subsequent change in JP.
Finally, a direct and negative relation emerged between worka-
holism and SAs, as well as an indirect positive relation through
psychophysic strain. Therefore, Hypothesis 4 was fully confirmed.
Psychophysic strain partially mediates the relationship between
workaholism and SAs. This study highlights the importance of psy-
chophysic strain also in the relation between workaholism and SAs.
Indeed, when we consider psychophysic strain as a mediator, two dis-
tinct mechanisms emerge: a positive indirect effect of workaholism
on SAs through strain, and a negative direct effect that, at least par-
tially, offsets each other. This could provide a possible explanation
for why previous studies, which did not consider the mediating role
of psychophysic strain, did not find any association between worka-
holism and SAs.11,39 Nevertheless, it is worth noting that in this
study, the direct negative effect was stronger than the indirect posi-
tive effect and, in fact, the overall correlation between workaholism
and SA (latent variable, Fig. 1) was negative, corresponding to 0.27
(P = 0.003). Therefore, in this study, workaholism restrains SAs.
Overall, this study can be referred to the theoretical frame-
work proposed by Hart and Cooper.9 Workaholism, considered as an
individual characteristic, influences workers well-being, with con-
sequences at an individual and organizational level, especially in
terms of lower levels of JP and more SAs.
CONCLUSION
The present work supplies several new contributions to the
literature. First, few studies exist, which take into account the re-
lation between workaholism and two outcomes: JP and SAs. Also,
the hypothesis of the meditational role of psychophysic strain is
new. Moreover, this study uses a longitudinal design, and variables
were revealed using different information sources. These are self-
reports for workaholism, the evaluation of occupational physician
for psychophysic strain, the judgment of the supervisor for JP, and
the company database for SAs. Our research design overcomes the
limits of previous studies, which are often cross-sectional, and based
on self-report measures (with the exception of the study by Shimazu
et al).10 The longitudinal design allows us to obtain information
about the direction of causality, and the use of different types of
measures allows us to contain the common method bias.56
Nevertheless, this study presents some limitations. First, the
sample used is a convenience sample, which limits generalization.

C 2013 American College of Occupational and Environmental Medicine
Copyright 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Workaholism and Performance: The Role of Strain
Second, a single indicator of overall JP was employed, which does not
allow the distinction between in-role performance (ie, activities that
are related to employees formal role requirements) and extra-role
performance (ie, activities that promote organizational effectiveness
but are not part of employees formal role requirements).55 Some pre-
vious studies, in fact, evidenced a positive relation between worka-
holism and extra-role performance.11 Finally, psychophysic strain,
JP, and SAs were analyzed simultaneously at time 2. This does not
allow us to test the direction of causality for the relationship between
strain and JP and between strain and SAs.
These limitations lead to singling out potential future devel-
opments. First of all, a full three-wave panel design would help
clarify the direction of causality between psychophysic strain and
JP/SAs.56 Moreover, the relations examined here should be analyzed
considering a wider time perspective. It can be hypothesized that
with an increase in the interval between time 1 and time 2, the psy-
chophysic symptoms can get even worse, with even more serious
consequences for the individual and the organization. It could be
interesting to take into account other objective indicators of malaise
(eg, actual turnover, disability pension). Finally, the assessment of
some physiological indicators (eg, cortisol, cytokines) could help
in understanding the mechanisms that regulate the relation between
workaholism and psychophysic strain, according to the allostatic
load theory.16,57
Possible interventions could concern both the workers and
the organization. In suggesting potential interventions, three lev-
els of prevention can be distinguished: primary (ie, reducing the
risk of workaholism among workers), secondary (ie, identifica-
tion and training of the workers at risk of workaholism), and ter-
tiary (ie, minimizing the negative consequences of workaholism on
health).58
At the primary level, the occupational physician could sug-
gest to the management interventions aimed at modifying the work
environment; in particular, it is advisable to construct practices that
permit a better balance between working life and private life, avoid-
ing the allocation of rewards (career, salary rises) purely on the
basis of the quantity of work realized.4 As underlined by van Wijhe
et al,58 therefore, a change is necessary at the level of organizational
culture, which should be primarily aimed at balancing the work and
private spheres, and should use reward systems based on working
smart rather than hard.
Among the secondary-level interventions is highlighted the
identification of potential workaholics on the basis of individual
characteristics (eg, self-efficacy, neuroticism),4,59 the number of
hours worked and the position held (eg, manager, supervisor).
In addition, potential workaholics could be encouraged to detach
themselves from work and to recover the invested resources. For
example, the workers at risk could be encouraged to reduce persever-
ative cognitions during after-work hours (eg, rumination, worry)18
or to carry out regenerative or relaxing activities during work
breaks (eg, physical activity, meditation).16 In addition, training
programs aimed at increasing individual psychological resources
can be useful to prevent workaholism and favor the development
of work engagement (eg, self-esteem, resilience, and active coping
style).58,6063
At a tertiary level, the occupational physician can imple-
ment, either personally or through psychologists or psychothera-
pists, cognitivebehavioral interventions. For example, one type of
intervention could be aimed at helping workaholics to substitute ir-
rational beliefs (eg, I must respect the deadline at all costs) with
more realistic beliefs (eg, I will do everything possible to respect the
deadline or, otherwise, I will inform my supervisor in good time)
and, consequently, to reduce worries connected to work (cognitive
restructuring).58 Another type of intervention could include mus-
cle relaxation techniques to reduce the tension attributable to the
negative affective states linked to workaholism.58
1259
Falco et al
REFERENCES
1. Schaufeli WB, Taris TW, Bakker AB. It takes two to tango: workaholism is
working excessively and working compulsively. In: Burke RJ, Cooper CL, eds.
The Long Work Hours Culture: Causes, Consequences and Choices. Bingley,
UK: Emerald; 2008:203225.
2. Sussman S. Workaholism: a review. J Addict Res Ther. 2012;6:110.
doi:10.4172/2155-6105.S6-001.
3. McMillan LHW, ODriscoll MP. The wellsprings of workaholism: a compar-
ative analysis of the explanatory theories. In: Burke RJ, Cooper CL, eds. The
Long Work Hours Culture: Causes, Consequences and Choices. Bingley, UK:
Emerald; 2008:85111.
4. Ng TWH, Sorensen KL, Feldman DC. Dimensions, antecedents, and con-
sequences of workaholism: a conceptual integration and extension. J Organ
Behav. 2007;28:111136. doi:10.1002/job.424.
5. Shimazu A, Schaufeli WB. Is workaholism good or bad for em-
ployee well-being? The distinctiveness of workaholism and work en-
gagement among Japanese employees. Ind Health. 2009;47:495502.
doi:10.2486/indhealth.47.495.
6. McMillan LHW, ODriscoll MP, Marsh NV, Brady EC. Understand-
ing workaholism: data synthesis, theoretical critique, and future design
strategies. Int J Stress Manag. 2001;8:6991. doi:10.1023/A:100957312
9142.
7. Bakker AB, Demerouti E, Oerlemans W, Sonnentag S. Workaholism and
daily recovery: a day reconstruction study of leisure activities. J Organ Behav.
2012;34:87107. doi:10.1002/job.1796.
8. Ho VT, Wong SS, Lee CH. A tale of passion: linking job passion and cognitive
engagement to employee work performance. J Manag Stud. 2011;48:2647.
doi:10.1111/j.1467-6486.2009.00878.x.
9. Hart PM, Cooper CL. Occupational stress: toward a more integrated frame-
work. In: Anderson N, Ones DS, Sinangil HK, Viswesvaran C, eds. Handbook
of Industrial, Work, and Organizational Psychology. Thousand Oaks, CA:
Sage; 2001:93114.
10. Shimazu A, Schaufeli WB, Kubota K, Kawakami N. Do workaholism and
work engagement predict employee well-being and performance in opposite
directions? Ind Health. 2012;50:316321. doi:10.2486/indhealth.MS1355.
11. Schaufeli WB, Taris TW, Bakker A. Dr. Jekyll and Mr. Hyde: on the dif-
ferences between work engagement and workaholism. In: Burke R, ed. Work
Hours and Work Addiction. Northhampton, MA: Edward Elgar; 2006:193
252.
12. Kravina L, Falco A, Girardi D, De Carlo NA. Workaholism among manage-
ment and workers in an Italian cooperative enterprise. TPM. 2010;17:201
216.
13. Taris TW, Schaufeli WB, Verhoeven LC. Workaholism in the Netherlands:
measurement and implications for job strain and worknonwork conflict. Appl
Psychol. 2005;54:3760. doi:10.1111/j.1464-0597.2005.00195.x.
14. Shimazu A, Demerouti E, Bakker AB, Shimada K, Kawakami
N. Workaholism and well-being among Japanese dual-earner cou-
ples: a spillover-crossover perspective. Soc Sci Med. 2011;73:399409.
doi:10.1016/j.socscimed.2011.05.049.
15. Meijman TF, Mulder G. Psychological aspects of workload. In: Drenth PJ,
Thierry H, de Wolff CJ, eds. Handbook of Work and Organizational Psychol-
ogy. Hove, UK: Psychology Press Ltd; 1998:533.
16. Geurts SAE, Sonnentag S. Recovery as an explanatory mechanism in the
relation between acute stress reactions and chronic health impairment. Scand
J Work Environ Health. 2006;32:482492. doi:10.5271/sjweh.1053.
17. Binnewies C, Sonnentag S, Mojza EJ. Daily performance at work: feeling
recovered in the morning as a predictor of day-level job performance. J Organ
Behav. 2009;30:6793. doi:10.1002/job.541.
18. van Wijhe CI, Peeters M, Schaufeli WB, Ouweneel E. Rise and shine: recovery
experiences of workaholic and nonworkaholic employees. Eur J Work Organ
Psychol. 2012;21:114. doi:10.1080/1359432X.2012.663527.
19. Sonnentag S. Psychological detachment from work during leisure time:
the benefits of mentally disengaging from work. Curr Dir Psychol Sci.
2012;21:114118. doi:10.1177/0963721411434979.
20. Sonnentag S, Binnewies C, Mojza EJ. Staying well and engaged when
demands are high: the role of psychological detachment. J Appl Psychol.
2010;95:965976. doi:10.1037/a0020032.
21. Sonnentag S, Fritz C. The Recovery Experience Questionnaire: develop-
ment and validation of a measure for assessing recuperation and unwinding
from work. J Occup Health Psychol. 2007;12:204221. doi:10.1037/1076-
8998.12.3.204.
22. Sonnentag S, Binnewies C, Mojza EJ. Did you have a nice evening? A
day-level study on recovery experiences, sleep, and affect. J Appl Psychol.
2008;93:674684. doi:10.1037/0021-9010.93.3.674.
1260
Copyright 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
JOEM r Volume 55, Number 11, November 2013
23. van der Klink JJ, Blonk RW, Schene AH, van Dijk FJ. The benefits of inter-
ventions for work-related stress. Am J Public Health. 2001;91:270276. doi:
10.2105/AJPH.91.2.270.
24. Sonnentag S, Bayer UV. Switching off mentally: predictors and consequences
of psychological detachment from work during off-job time. J Occup Health
Psychol. 2005;10:393414. doi:10.1037/1076-8998.10.4.393.
25. Ford MT, Cerasoli CP, Higgins JA, Decesare AL. Relationships be-
tween psychological, physical, and behavioural health and work per-
formance: a review and meta-analysis. Work Stress. 2011;25:185204.
doi:10.1080/02678373.2011.609035.
26. Chang CH, Rosen CC, Levy PE. The relationship between percep-
tions of organizational politics and employee attitudes, strain, and be-
havior: a meta-analytic examination. Acad Manage J. 2009;52:779801.
doi:10.5465/AMJ.2009.43670894.
27. Beal DJ, Weiss HM, Barros E, MacDermid SM. An episodic process model
of affective influences on performance. J Appl Psychol. 2005;90:10541068.
doi:10.1037/0021-9010.90.6.1054.
28. Beal DJ, Trougakos JP, Weiss HM, Green SG. Episodic processes in
emotional labor: perceptions of affective delivery and regulation strate-
gies. J Appl Psychol. 2006;91:10531065. doi:10.1037/0021-9010.91.5
.1053.
29. Seo M, Ilies R. The role of self-efficacy, goal, and affect in dynamic motiva-
tional self-regulation. Organ Behav Hum Decis Process. 2009;109:120133.
doi:10.1016/j.obhdp.2009.03.001.
30. Taimela S, Laa ra E, Malmivaara A, et al. Self-reported health problems and
sickness absence in different age groups predominantly engaged in phys-
ical work. Occup Environ Med. 2007;64:739746. doi:10.1136/oem.2006
.027789.
31. Vaa nanen A, Toppinen-Tanner S, Kalimo R, Mutanen P, Vahtera J, Peiro
JM. Job characteristics, physical and psychological symptoms, and social
support as antecedents of sickness absence among men and women in the
private industrial sector. Soc Sci Med. 2003;57:807824. doi:10.1016/S0277-
9536(02)00450-1.
32. Vahtera J, Westerlund H, Ferrie JE, et al. All-cause and diagnosis-specific sick-
ness absence as a predictor of sustained suboptimal health: a 14-year follow-
up in the GAZEL cohort. J Epidemiol Community Health. 2010;64:311317.
doi:10.1136/jech.2008.083923.
33. Kivimaki M, Forma P, Wikstrom J, et al. Sickness absence as a risk marker of
future disability pension: the 10-town study. J Epidemiol Community Health.
2004;58:710711. doi:10.1136/jech.2003.015842.
34. Vahtera J, Pentti J, Kivimaki M. Sickness absence as a predictor of mor-
tality among male and female employees. J Epidemiol Community Health.
2004;58:321326. doi:10.1136/jech.2003.011817.
35. Darr W, Johns G. Work strain, health, and absenteeism: a meta-analysis. J
Occup Health Psychol. 2008;13:293318. doi:10.1037/a0012639.
36. Gorgievski MJ, Bakker AB. Passion for work: work engagement versus
workaholism. In: Albrecht SL, ed. Handbook of Employee Engagement: Per-
spectives, Issues, Research and Practice. Northhampton, MA: Edward Elgar;
2010:264271.
37. Shimazu A, Schaufeli WB, Taris TW. How does workaholism affect worker
health and performance? The mediating role of coping. Int J Behav Med.
2010;17:154160. doi:10.1007/s12529-010-9077-x.
38. Clark MA, Lelchook AM, Taylor ML. Beyond the Big Five: how narcissism,
perfectionism, and dispositional affect relate to workaholism. Pers Indiv Diff.
2010; 48:786791. doi:10.1016/j.paid.2010.01.013.
39. Burke RJ, Matthiesen S. Workaholism among Norwegian journal-
ists: antecedents and consequences. Stress Health. 2004;20:301308.
doi:10.1002/smi.1025.
40. Taris TW, Geurts SAE, Schaufeli WB, Blonk RWB, Lagerveld SE. All day
and all of the night: the relative contribution of two dimensions of worka-
holism to well-being in self-employed workers. Work Stress. 2008;22:153
165. doi:10.1080/02678370701758074.
41. Falco A, Girardi D, Parmiani G, et al. Presenteismo e salute dei lavoratori:
effetti di mediazione sullo strain psico-fisico in unindagine longitudinale.
G Ital Med Lav Erg. 2013;35:138150.
42. Falco A, Girardi D, Sarto F, et al. A new scale for measuring the psycho-
physical effects of work-related stress in a perspective of methods integration.
Med Lav. 2012;103:288308.
43. Falco A, Girardi D, Marcuzzo G, De Carlo A, Bartolucci GB. Work stress and
negative affectivity: a multi-method study. Occup Med (Lond). 2013;63:341
347. doi:10.1093/occmed/kqt054.
44. Falco A, Kravina L, Girardi D, Dal Corso L, Di Sipio A, De Carlo NA. The
convergence between self and observer ratings of workaholism: a comparison
between couples. TPM. 2012;19:311324. doi:10.4473/TPM19.4.5.

C 2013 American College of Occupational and Environmental Medicine
JOEM r Volume 55, Number 11, November 2013
45. Trifiletti E, Vianello M, Capozza D. Validazione delle scale del Test Qu-
Bo con modelli di analisi fattoriale confermativa. In: De Carlo NA, Falco
A, Capozza D, eds. Benessere Organizzativo e Stress Lavoro-Correlato.
Metodi di Valutazione e dIntervento. Milano: FrancoAngeli; 2013:464
479.
46. Trifiletti E, Di Bernardo GA, Falvo R, Capozza D. Patients are not fully
human: a nurses coping response to stress. J Appl Soc Psychol. In press.
47. Joreskog KG, Sorbom D. LISREL 8.70 for Windows [Computer Software].
Lincolnwood, IL: Scientific Software International; 2004.
48. Little TD, Cunningham WA, Shahar G, Widaman KF. To parcel or not
to parcel: exploring the question, weighing the merits. Struct Equ Model.
2002;9:151173. doi:10.1207/S15328007SEM0902_1.
49. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure
analysis: conventional criteria versus new alternatives. Struct Equ Model.
1999;6:155. doi:10.1080/10705519909540118.
50. Baron RM, Kenny DA. The moderatormediator variable distinction in social
psychological research: conceptual, strategic, and statistical considerations. J
Pers Soc Psychol. 1986;51:11731182. doi:10.1037/0022-3514.51.6.1173.
51. MacKinnon DP. Introduction to Statistical Mediation Analysis. Mahwah, NJ:
Erlbaum; 2008.
52. Shrout PE, Bolger N. Mediation in experimental and nonexperimental studies:
new procedures and recommendations. Psychol Methods. 2002;7:422445.
doi:10.1037/1082-989X.7.4.422.
53. MacKinnon DP, Krull JL, Lockwood CM. Equivalence of the medi-
ation, confounding and suppression effect. Prev Sci. 2000;1:173181.
doi:10.1023/A:1026595011371.
54. Judd CM, Kenny DA. Estimating the Effects of Social Interventions. New
York: Cambridge University Press; 1981.

C 2013 American College of Occupational and Environmental Medicine
Copyright 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Workaholism and Performance: The Role of Strain
55. Borman WC, Motowidlo SJ. Task performance and contextual performance:
the meaning for personnel selection research. Hum Perform. 1997;10:99109.
doi:10.1207/s15327043hup1002_3.
56. Zapf D, Dormann C, Frese M. Longitudinal studies in organizational stress
research: a review of the literature with reference to methodological issues. J
Occup Health Psychol. 1996;1:145169. doi:10.1037/1076-8998.1.2.145.
57. Hansel A, Hong S, Camara RJ, von Kanel R. Inflammation as a psychophys-
iological biomarker in chronic psychosocial stress. Neurosci Biobehav Rev.
2010;35:115121. doi:10.1016/j.neubiorev.2009.12.012.
58. van Wijhe CI, Schaufeli WB, Peeters MCW. Understanding and treating
workaholism: setting the stage for successful interventions. In: Burke RJ,
Cooper CL, eds. Risky Business: Psychological, Physical and Financial Costs
of High Risk Behavior in Organizations. Farnham, UK: Gower; 2010:107
134.
59. Burke RJ, Matthiesen SB, Pallesen S. Personality correlates of workaholism.
Pers Indiv Differ. 2006;40:12231233. doi:10.1016/j.paid.2005.10.017
60. Bakker A, Schaufeli WB, Leiter M, Taris TW. Work engagement: an emerging
concept in occupational health psychology. Work Stress. 2008;22:187200.
doi:10.1080/02678370802393649.
61. Di Sipio A, Falco A, Kravina L, De Carlo NA. Positive personal resources and
organizational well-being: resilience, hope, optimism, and self-efficacy in an
Italian health care setting. TPM. 2012;19:8195. doi:10.4473/TPM19.2.2.
62. Dal Corso L, Floretta P, Falco A, Benevene P, De Carlo A. The reper-
tory grid technique in a researchintervention on work-related stress. TPM.
2013;20:155168. doi:10.4473/TPM20.2.4
63. Girardi D, Falco A, Dal Corso L, Kravina L, De Carlo A. Interpersonal conflict
and perceived work stress: the role of negative affectivity. TPM. 2011;18:257
273.
1261