ORIGINAL ARTICLE

Association Between Organization Culture, Health Status, and

Presenteeism
Yao-Tsung Chang, MPH, Chien-Tien Su, MD, PhD, Ruey-yu Chen, PhD, Ching-Ying Yeh, PhD,
Pai-Tsang Huang, MD, Chiou-Jong Chen, PhD, and Ming Chu, MPH

organizational culture as “the identity of an organization and its peo-

Objective: The aim of this study was to examine relationships of organiza-
ple that is created, maintained, and transformed by its employees and
tional culture and health behaviors with presenteeism. Method: Data of a
leaders.”22 Schein23,24 defined it as “a deeply rooted level of basic
self-reported questionnaire were collected from 816 employees, who joined
assumptions and beliefs that is ingrained within an organization.”
the study on a voluntary basis, in seven enterprises in northern Taiwan.
According to The Health Communication Unit guidelines in 2009,25
Results: Organizational culture and health behaviors were found to be signifi-
organizational culture is one worksite health dimension which, along
cantly associated with presenteeism. After adjusting for confounding factors,
with occupational safety and health as well as individual health be-
the number of health complaints seemed to be more suitable than chronic
haviors, contributes to a fully healthy worksite.
diseases in predicting presenteeism. Conclusions: This study result implied
Three measurement methods exist for organizational cul-
that advantage could be taken of organizational culture and employees’ health
ture. The first is Quinn’s Organizational Culture Assessment Instru-
behaviors to reduce presenteeism in the workplace.
ment questionnaire and competing values framework26 ; the second,
Schein’s three levels of culture model works,23,24 ; and the third,
Shain’s Stress Satisfaction Offset Score (SSOS) index.27 The SSOS

P resenteeism was defined as ever getting ill or feeling uncomfort-

able but still going to work1 or working even when ill, which
causes loss of productivity.2 Presenteeism can lead to productiv-
index contains four questions about cultural conditions: control, de-
mand, reward, and effort. After summing and balancing these condi-
tions, the SSOS index is commonly used to predict health risks and
ity losses and wasted medical expenses. Ricci and Chee3 showed is suitable for measuring organizational culture as an aspect of the
that obesity and chronic fatigue cost more than US$42.29 billion psychosocial work environment.
and US$31 billion, respectively, annually in the United States.4 In Karlsson et al28 conducted a study that explored relationships
2007, Schultz and Edington5 published his work on health-related of organizational culture and health status with presenteeism. Their
productivity loss in the Harvard Business Review, indicating that pre- results showed that an individual’s health status was a mediator be-
senteeism accounted for 62% of overall productivity loss, whereas tween the psychosocial work environment and presenteeism. As the
medical costs only accounted for 24%. It is clear that presenteeism only longitudinal study available on the relationship between orga-
has been a long-term problem in the workplace and will likely cause nizational culture and presenteeism, their study was thus suitable for
even more losses in the future. Therefore, ever more research was use to verify findings in Taiwan. If a worker’s individual health status
conducted in the past decade, with the aim of elucidating relation- is indeed a mediator of organizational culture and presenteeism, then
ships between presenteeism and various risk factors, and estimating chronic diseases or subjective health complaints (SHCs) might be a
the loss of productivity to enterprises.3,4,6–14 more appropriate surrogate of health status than the 12-item short
According to World Health Organization (WHO)’s form health survey indicator used by Malin et al, which indirectly
guidelines,15 a healthy worksite encompasses four dimensions: the measures health status.
physical work environment, psychosocial work environment, per- Subjective health complaints refer to symptoms experienced
sonal health resources, and enterprise community involvement. In by an individual with or without a defined diagnosis, and no study
addition to those studies which addressed relationships between per- was found which examined the relationship between SHCs and pre-
sonal health risks and presenteeism, some others looked at the role senteeism. Subjective health complaints constitute both everyday
played by psychosocial worksite environments.16,17 For example, experiences and health problems, which are common reasons for
low job satisfaction and dismissal pressure caused increased presen- temporary disability and sick leave in adults.29–34 Compared with
teeism, whereas lower social support led to elevated psychological chronic diseases, we consider SHCs a better indicator of a worker’s
pressure, which in turn caused more presenteeism.18 Researchers health status because presenteeism may occur before overt diseases
believe that mental health is more important to presenteeism than manifest themselves. Recent research also demonstrated that SHCs
physical health.19–21 Within the dimension of the psychosocial work are an important factor in presenteeism.35 In addition, health be-
environment, however, we found only a few studies that dealt with haviors are related to both one’s health status and presenteeism, and
the relationship between organizational culture and presenteeism. they cannot be ignored as potential covariates or confounders in such
Organizational culture has several different definitions. In- studies.
dustrial College of the Armed Forces guidelines in 1999 defined The null hypotheses of this study were as follows: (1) organi-
zational culture is related to presenteeism; (2) one’s health status is
a mediator between them, (3) SHCs are more suitable than chronic
From the Taipei Medical University (Mr Chang, Dr Su, Dr Chen, Dr Yeh, Dr Chen, diseases, as an indicator of one’s health status, in predicting presen-
and Mr Chu); Taipei Medical University Hospital (Dr Su); Taipei Wanfang teeism; and (4) health behaviors are confounders of the relationship
Hospital (Dr Huang); Occupational Safety and Health Administration
(Dr Chen); and Taiwan Health and Productivity Management Association between organizational culture and presenteeism.
(Mr Chang, Dr Su, Dr Chen, Dr Yeh, and Mr Chu), Taipei.
The authors declare no conflicts of interest. METHODS AND PROCEDURES
Address correspondence to: Ruey-yu Chen, PhD, School of Public Health,
Taipei Medical University, No. 250, Wu-Hsing St, Taipei 110, Taiwan Study Population
(rueyyu@tmu.edu.tw, rueyyu535@gmail.com).
Copyright C 2015 by American College of Occupational and Environmental
We collected data from employees in seven enterprises in
Medicine northern Taiwan, including one communication company, three high-
DOI: 10.1097/JOM.0000000000000439 tech manufacturers, one financial corporation, one environmental

JOEM r Volume 57, Number 7, July 2015 765

Copyright © 2015 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited.
Chang et al JOEM r Volume 57, Number 7, July 2015

clean-up enterprise, and one government agency. All workers were SHCs, numbers of chronic diseases, and health behaviors. After con-
aged 18 to 55 years and received a letter inviting them to volunteer ducting univariate and bivariate analyses, variables were selected to
for the study. In total, 816 workers joined the study by returning a construct multivariate models with a logistic regression. Each model
self-reported questionnaire, resulting in a response rate of 46% to was adjusted for demographic variables. Multivariate models were
63% among seven enterprises, and the average response rate was analyzed at a 95% significance level (P < 0.05). The analyses were
55.2%. The study was approved by the Joint Institutional Review conducted using PASW 18.0 software for Windows (SPSS, Chicago,
Board of Taipei Medical University. IL).

DATA COLLECTION RESULTS

Questionnaire In total, 816 workers completed the questionnaires (62.5%
All participants completed a self-reported questionnaire about male vs 27.5 female employees), with a mean age of 36 years, 85.9%
organizational culture, individual and sociodemographic variables, with an educational level above university, 45.8% were married, and
health behaviors, and presenteeism. with a mean weekly work time of 45 hours. All variables showed
The measure of presenteeism was based on a single question significant differences between “without presenteeism” and “with
that assessed how often employees had come to work during the presenteeism” (Table 1).
last month despite being ill. Four response categories consisted of We found that 38.6% of participants had presenteeism at least
“never,” “one to three times,” “four to six times,” and “more than 1 day, 31.4% 1 to 3 days, 4.0% 4 to 6 days, and 3.1% more than 7
seven times.” The question was modified from the work of Aronsson days. The mean SSOS index was 0.52. Those without presenteeism
et al.1 had an SSOS index of 0.72, whereas those with presenteeism had an
Organizational culture was measured by the SSOS index of 0.22. The difference was statistically significant (Table 2).
questionnaire,27 which contains four questions. Every question is Differences between numbers of SHCs were significant with
rated on a Likert five-point scale, ranging from “strongly agree” to” regard to presenteeism. All participants had a mean number of 1.63
strongly disagree.” Two questions about pressure were scored −1 SHCs, and 32.7% of participants had no SHCs, whereas those with
point for agree and strongly agree, whereas the other two about work presenteeism had an average of 2.88 SHCs.
character were scored 1 point for agree and strongly agree. All ques- Most participants did not meet the health behavior standard;
tions were scored 0 point for the answers “I’m not sure,” “disagree,” 30% to 40% of them consumed enough vegetables and fruits, and
and “strongly disagree.” After summing up the scores, the SSOS 6% had enough physical activity, although there were only 8.5%
index reflexes organizational culture and includes four dimensions smokers. There was a significant difference between health behavior
of reward and effort balance, work participation, time pressure, and scores, as those with presenteeism had a mean of 8.86, which was
mental fatigue. lower than those without presenteeism.
Fourteen common SHCs were inquired about in our question- Because sex, age, marital status, and weekly work hours are
naire, including a cold, difficulties concentrating, memory problems, associated with presenteeism, and most health behaviors had signif-
allergic rhinitis, limb pain, neck and lower back pain, insomnia, icant differences between subjects with or those without presen-
lethargy, stomach pain, chest tightness and palpitations, menstrual teeism, these demographic variables and overall health behavior
pain, anxiety, irritability, and irritable bowel syndrome. points were used in a regression analysis. The logistic regression
Chronic diseases were also inquired about with a single analysis demonstrated that the number of SHCs had a mediating
multiple-choice question. Fourteen kinds of chronic diseases were effect (Table 3). Model 1 showed that SHCs were the major pre-
listed: hypertension, heart diseases, allergies, cancers, gastrointesti- dictive variable of presenteeism, with a predictive ability of 42%.
nal problems, diabetes, arthritis, depression, chronic liver diseases, Being male, being older, and having longer weekly work hours were
asthma, anxiety, thyroid problems, gum diseases, and dry eyes. associated with higher presenteeism odds ratios. After adjusting for
There were four questions about health behaviors (ie, veg- organizational culture, healthy behavior final points, and the number
etable consumption, fruit consumption, physical activity, and tobacco of chronic diseases, model 2 showed that organizational culture sig-
use). For vegetable consumption, fruit consumption, and physical nificantly predicted presenteeism. Model 3 showed that those with
activity, we asked about the frequency of such behaviors during the a higher SSOS index, thus enjoying healthier organizational cul-
week, and whether or not tobacco was used. This behavior score was ture, had a lower presenteeism risk. The most important factor in
summed up with the three questions not pertaining to smoking as organizational culture was mental fatigue.
follows: one point for 0 day, two points for 1 day, three points for 2
to 3 days, four points for 4 to 5 days, and five points for 6 to 7 days. DISCUSSION
Final scores ranged 5 from 15 points. In this study, we examined the relationships of organizational
Sociodemographic variables included the age, sex, educa- culture and health behaviors with presenteeism among 816 employ-
tional level, and marital status. Educational level was categorized ees in various enterprises. A lower SSOS index was found to be as-
as follows: less than senior high school, university, and higher than a sociated with greater presenteeism, and work pressure was the only
master’s degree. Marital status was classified as not currently married significant component of the SSOS index. Those with presenteeism
and married at the time of completing the questionnaire. had more SHCs and lower health behavior scores than the group
averages and those without presenteeism, respectively. Moreover,
Statistical Analysis the model analyses showed that both SHCs and organizational cul-
Health behaviors were rated as “meeting the standard” and ture significantly predicted presenteeism. After adjusting for health
“not meeting the standard.” According to World Health Organization status using SHCs and chronic diseases, the SSOS index became
suggestions as to health behaviors, “5 days a week” was rated as insignificant. Therefore, SHCs seemed to function as a mediator
“healthy” and scored one point, and then a final score was calculated. of organizational culture and health behaviors with presenteeism, a
Presenteeism was dichotomized as “without presenteeism” and “with finding shared by Malin et al.28 Subjective health complaints seemed
presenteeism.” The numbers of SHCs and chronic diseases were to be a more suitable indicator of health status than chronic diseases
counted to obtain a score. because people with overt diseases might still work and make fewer
Chi-square tests or t tests were used to assess presenteeism SHCs to keep their job. Thus, chronic diseases ceased to be a valid
against sociodemographic variables, SSOC scores, numbers of indicator of presenteeism under such a condition.

766 
C 2015 American College of Occupational and Environmental Medicine

Copyright © 2015 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited.
JOEM r Volume 57, Number 7, July 2015 Association Between Organization Culture, Health Status, and Presenteeism

TABLE 1. Description of Sociodemographics

With Without
Total Presenteeism Presenteeism
Variables n (%) n (%) n (%) χ2

Sex 18.39**
Male 510 (62.5) 168 (53.3) 342 (68.3)
Female 306 (37.5) 147 (46.7) 159 (31.7)
Age, yrs 9.34**
18–29 214 (26.2) 86 (28.0) 113 (23.5)
30–39 374 (45.8) 156 (50.8) 218 (45.4)
≥40 199 (24.4) 65 (21.2) 149 (31.0)
Mean 36.07 34.40 37.14 4.055**a
SD 9.33 7.66 10.124
Education 5.12#
Senior high school or less 115 (14.1) 34 (10.9) 81 (16.3)
University 388 (47.5) 160 (51.4) 228 (45.8)
A master’s degree or more 306 (37.5) 117 (37.6) 189 (38.0)
Marital status 6.71*
Not currently married 432 (52.9) 184 (59.4) 248 (50.0)
Married 374 (45.8) 126 (40.6) 248 (50.0)
Weekly work, hrs 26.61**
0–40 67 (8.2) 100 (32.3) 244 (49.8)
41–60 389 (47.7) 173 (55.8) 216 (44.1)
≥61 344 (42.2) 37 (11.9) 30 (6.1)
Mean 44.77 45.57 44.26 − 2.908**a
SD 6.25 6.83 5.805
Total 816 315 501

#0.1 < P < 0.05; *P < 0.05; **P < 0.01.

a
Analysis with t test.
SD, standard deviation.

As far as we are aware, the SSOS index as a potential in- in turn improved employees’ mental health. The interrelationship
dicator of presenteeism has never been studied. Only its compo- among them thus seems bidirectional or circular, instead of a simple
nents (ie, control, demand, reward, and effort) were investigated, causal one.
and “work control” was found to be related to presenteeism. This In recent years, “healthy culture” has gradually gained its im-
connection has something to do with work participation, cowork- portance. Nevertheless, it seems that the aspect of “healthy work
ers’ and managers’ support, and the degree to which one’s work culture” and “healthy culture of worksite” is quite different. With
is under one’s control. Leineweber et al36 showed that if someone worksite health promotion increasingly focusing on the support-
lacked coworkers’ or managers’ support and approval for his/her ive environment (ie, elements which sustain healthy behaviors), the
decisions, the situation would lead to increased pressure and greater social–ecological model in the design of health promotion inter-
presenteeism. Ashby and Mahdon37 showed that a feeling of tak- vention has been adopted more frequently.42 By constructing an
ing group responsibility was also a factor that affected how work- environment that supports healthy behaviors, this healthy culture or
ers choose absenteeism or presenteeism when they get ill. This healthy climate created is viewed as a key point to reducing health
study result did not demonstrate that work control was a signif- risks, such as obesity, imbalanced diet, and low physical activity, and
icant factor in the SSOS index. Perhaps this indicates that there hence comes the assessment tool “Lifegain Healthy Culture Audit,”
is a causal relationship between work control and pressure. As which evaluates whether the worksite supports employees’ health or
to “reward and effort balance,” Siegrist’s effort–reward imbalance not.43,44 This seems to differ in aspects of the SSOS, job content
model38 was most popular. If an employee felt that he/she was not questionnaire, and effort–reward imbalance models we discussed.
getting sufficient rewards at work, psychological pressure tended “Working culture” may not influence healthy behaviors directly, but
to increase. Low payment and long work times lead to increased it poses a determinant factor in absenteeism or presenteeism when
presenteeism.1,39 someone feels sick. For this reason, to further explore the impact
Work pressure can take many forms and the best-known of culture on health, it is imperative to look into these separate
is time pressure (ie, completing a mission on time, without dimensions, the “healthy working culture,” and the “healthy cul-
delays).36,40,41 This factor, however, was not significant in this study. ture about worksite environment.” Furthermore, our data show that
Psychological pressure was more predictive of presenteeism than SSOS scores and healthy behaviors were both significant on model
was physical health.35 Certain studies inquiring about presenteeism 3 and model 4, and we may say both of them had some influence
among patients with chronic diseases took mental health problems, on presenteeism, but unsure of whether the working culture has a
such as depression and anxiety, into account.11–14 Interestingly, some direct influence on behaviors. It should be researched deeper in the
studies showed that work effectiveness reduced work pressure, which future.


C 2015 American College of Occupational and Environmental Medicine 767

Copyright © 2015 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited.
Chang et al JOEM r Volume 57, Number 7, July 2015

TABLE 2. Description of SSOS Index, Healthy Behavior, Subjective Health Complains, and Chronic Diseases

With Without
Total Presenteeism Presenteeism
n (%) n (%) n (%) χ2

SSOS index 38.03**

+2 225 (27.6) 56 (17.8) 169 (33.7)
+1 194 (23.8) 78 (24.8) 116 (23.2)
0 237 (29.0) 92 (29.2) 145 (28.9)
−1 104 (12.7) 56 (17.8) 48 (9.6)
−2 56 (6.9) 33 (10.5) 23 (4.6)
Mean 0.52 0.22 0.72 5.885**a
SD 1.21 1.23 1.16
SHCs 272.93**
0 267 (32.7) 10 (3.2) 257 (51.3)
1 212 (26.0) 77 (24.4) 135 (26.9)
2 141 (17.3) 78 (24.8) 63 (12.6)
≥3 196 (24.0) 150 (47.6) 46 (9.2)
Mean 1.63 2.88 0.85 − 17.924**a
SD 1.86 2.09 1.15
Chronic diseases 22.27**
0 523 (64.1) 176 (55.9) 347 (69.3)
1 205 (25.1) 87 (27.6) 118 (23.6)
≥2 88 (10.8) 52 (16.5) 36 (7.2)
Mean 0.50 0.69 0.39 − 5.403**a
SD 0.80 0.98 0.64
Vegetables consumed/day 15.58**
<5 492 (60.3) 217 (68.9) 275 (55.0)
≥5 323 (39.6) 98 (31.1) 225 (45.0)
Fruits consumed/day 7.76**
<5 576 (70.6) 240 (76.2) 336 (67.1)
≥5 240 (29.4) 75 (23.8) 165 (32.9)
Physical activity, d 0.73
<5 717 (87.9) 278 (94.6) 439 (93.0)
≥5 49 (6.0) 16 (5.4) 33 (7.0)
Mean 2.52 2.31 2.65 2.953**a
SD 1.62 1.52 1.67
Smoke 1.33
Yes 69 (8.5) 31 (10.0) 38 (7.7)
No 737 (90.3) 279 (90.0) 458 (92.3)
Healthy behaviors
Mean 9.30 8.86 9.58 4.391**a
SD 2.30 2.12 2.37
Total 816 315 501

#0.1 < P < 0.05; *P < 0.05; ** P < 0.01.

a
Analysis with t test.
SD, standard deviation; SHC, subjective health complaint; SSOS, Stress Satisfaction Offset Score.

We did not include alcohol consumption or drug use in the clarify the relationships among physical activity, work pressure, and
health behaviors as other studies did because both of these behavior SHCs.
are uncommon in Taiwan.45 Rather, we inquired about vegetable and Although SHCs can be very culturally/geographically spe-
fruit consumption, but little research has explored the associations of cific, few articles have been published in Taiwan. In 2012, one re-
such health behaviors with presenteeism until now.17,46,47 Our result search was conducted in Taiwan on the association with SHC and
showed that greater vegetable and fruit consumption was healthy, different occupation category,49 and the result showed that muscu-
and was significantly associated with decreased presenteeism. There loskeletal pain has the highest prevalence, followed by dizziness,
is no study with a similar design with which we could compare eye irritation, and abdomen pain. Although SHCs may vary from
our results. Physical activity is associated with SHCs and chronic job to job, musculoskeletal pain remains the dominant problem.
diseases. More physical activity can reduce psychological pressure This happened to have a similar result as one research done in the
and indirectly decrease SHC problems.48 More studies are needed to United States,50 which nonetheless indicated that improving work

768 
C 2015 American College of Occupational and Environmental Medicine

Copyright © 2015 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited.
JOEM r Volume 57, Number 7, July 2015 Association Between Organization Culture, Health Status, and Presenteeism

TABLE 3. Association Between Organizational Culture, Healthy Behavior, Subjective Health Complains, and
Presenteeism—Using Logistic Regression Analysis (n = 816)

Model 1 Model 2 Model 3 Model 4

Sex, OR (CI)
Male 1.00 1.00 1.00 1.00
Female 1.475* (1.007–2.159) 1.452# (0.967–2.179) 2.281** (1.608–3.235) 2.108** (1.486–2.990)
Marital status, OR (CI)
Not currently married 1.106 (0.721–1.698) 1.134 (0.724–1.777) 0.950 (0.649–1.390) 0.984 (0.670–1.443)
Married 1.00 1.00 1.00 1.00
Age, OR (CI) 0.976* (0.953–1.000) 0.974# (0.948–1.000) 0.978# (0.957–1.00) 0.974** (0.953–0.995)
Weekly work hours, OR (CI) 1.035* (1.006–1.065) 1.035* (1.004–1.066) 1.033* (1.006–1.062) 1.034* (1.006–1.062)
SSOS index, OR (CI) 0.864# (0.739–1.012)
−2 4.081** (2.173–8.431)
−1 3.283** (1.902–5.664)
0 1.619* (1.040–2.520)
+1 1.663* (1.006–1.062)
+2 1.00
Reward, OR (CI) 1.017 (0.828–1.248)
Effort, OR (CI) 0.860 (0.682–1.085)
Time pressure, OR (CI) 1.098 (0.869–1.389)
Mental fatigue, OR (CI) 1.450** (1.151–1.827)
Healthy behavior, OR (CI) 0.954 (0.779–1.167) 0.791** (0.664–0.941) 0.804* (0.674–0.958)
SHCs, OR (CI) 2.553** (2.190–2.976) 2.391** (2.028–2.818)
Chronic diseases, OR (CI) 1.205 (0.926–1.568)
Adjusted R2 0.421 0.422 0.138 0.148

#0.1 < P < 0.05; *P < 0.05; **0.01.

CI, confidence interval; OR, odds ratio; SHC, subjective health complaint; SSOS, Stress Satisfaction Offset Score.

satisfaction did not help reduce SHCs. Even in an ideal worksite cul- bility cost for employers is relatively cheap compared with the United
ture environment, SHCs can still exist, and most employees suffering States and most European countries. Besides, most companies have
from SHCs do not seek medical assistance.51 Owing to this, rather no short paid sick leaves, except for workers’ compensation; we
than a good indicator of good worksite culture, SHCs can serve as might assume that we have higher presenteeism and SHCs than
good indicators of presenteeism behavior. Hence, we may predict countries mentioned, and we need to take greater heed to the long-
that if we intend to reduce SHCs, there are two ways to approach term health effects on our employees and the policies introduced
this: one is improving employees’ health risk and healthy behavior in the future. So even our research may lack adequate evidence to
to reduce “true SHCs”, and the other is improving their work satis- properly recommend for improving worksite culture to reduce SHCs
faction to reduce the “pseudo SHCs”—just to complain about their and presenteeism, SHCs may be regarded as a potential sign of
work. In short, we need more research into SHCs in the workplace. illnesses.
On the basis of our result, some recommendations are pro- A major limitation of this study is the cross-sectional design
posed. More research effort should be placed on the association that did not allow us to draw causal relationships of organizational
between SHCs, presenteeism, healthy behaviors, and healthy work- culture and health behaviors with presenteeism. Second, this study
ing culture because SHCs might be a predictor to long-term health might have suffered from the problem of selection bias. Because em-
status. For worksite health promotion intervention, the effectiveness ployees were recruited from a national health promotion program,
of lowering SHCs when reducing health risks will also be a research those with greater interest in health issues may have been more will-
area. For policies and management recommendations, on the basis of ing to join the study, thus creating a selection bias. The response
the relationship between absenteeism and working culture, a further rate of this questionnaire is not significantly different, from 46% to
probe into presenteeism and hence forming suitable policies to im- 63%, with government agency the highest, partly due to the obedient
prove employees’ health and productivity will be advisable. A 2014 nature of public servants. The lowest response rate of the financial
research, which compared the company policy and health status, cooperation is due to the high proportion of field personnel. A prob-
indicated that the cost of modifiable health status was much lower able reason for this 55.2% average response rate may be because
than that of modifiable policies on medical care and absenteeism.52 participating organizations are all involved in the worksite health
Nevertheless, considering the long-term effect on worksite culture, promotion program conducted by the government.
presenteeism, and health status, the cost of health status to compa- One of our research limitations is the scarcity of relevant
nies is underestimated. In fact, health risks often induce a number of literature on this subject, which makes it difficult for us to com-
chronic diseases and increase health cost; conversely, better work- pare the SHCs with other countries, and to effectively explore how
site cultures and policies can inhibit health risks such as obesity and cultural and geographical specifics can have the impact on the dis-
unhealthy diet and habits. At the same time, with the support of crepancy and the impact on SHC and presenteeism. The strength and
well-implemented medical care and policies, it not only promotes uniqueness of this study lie in the fact that for the first time in the
employees’ health but also reduces the company’s health cost in a literature, the SSOS index as an indicator of organizational culture
long term. In Taiwan, we have national health insurance, and the lia- was used to predict presenteeism. Because we used fewer questions,


C 2015 American College of Occupational and Environmental Medicine 769

Copyright © 2015 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited.
Chang et al JOEM r Volume 57, Number 7, July 2015

more dimensional questions are left for us to learn about in the fu- absenteeism and presenteeism. J Occup Environ Med. 2008;50:1228–
ture. In addition, our study population encompassed white-collar and 1243.
blue-collar workers, as well as private sector and public sector em- 21. Munir F, Yarker J, Haslam C, et al. Work factors related to psychological
ployees, resulting in a wider and more diversified representativeness. and health-related distress among employees with chronic illnesses. J Occup
Rehabil. 2007;17:259–277.
22. Industrial College of the Armed Forces. Organizational culture. Strategic
CONCLUSIONS Leadership and Decision Making. National Defence University; 1999. Avail-
This is the first study to use the SSOS index as an indicator able at: http://www.ndu.edu/inss/books/Books-201999/StrategicLeadership
of organizational culture, and to explore relationships of organiza- andDecision-making-Feb99/pt4ch16.html.
tional culture, health behaviors, and health status with presenteeism. 23. Schein E. Organizational Culture and Leadership. San Francisco: Jossey-
Both organizational culture and health behaviors were affected by Bass; 1988.
mediation of the health status, whereas the latter was represented 24. Schein E. The Corporate Culture Survival Guide. John Wiley & Sons;
1999.
more by SHCs than by chronic diseases. It would be advantageous
to enterprises to make good use of organizational culture and health 25. The Health Communication Unit. Organizational Culture: From Assessment
to Action. University of Toronto; 2009.
behaviors to influence presenteeism in a positive direction.
26. Quinn C. Diagnosing and Changing Organizational Culture. John Wiley &
Sons; 2006.
REFERENCES 27. Shain M, Suurvali H. Investing in Comprehensive Workplace Health Pro-
1. Aronsson G, Gustafsson K, Dallner M. Sick but yet at work: an empirical study motion. Centre for Addiction and Mental Health (CAMH) and the National
of sickness presenteeism. J Epidemiol Community Health. 2000;54:502–509. Quality Institute (NQI); 2001.
2. Koopman C, Pelletier KR, Murray JF, et al. Stanford presenteeism scale: 28. Karlsson ML, Björklund C, Jensen I. The effects of psychosocial work factors
health status and employee productivity. J Occup Environ Med. 2002;44: on production loss, and the mediating effect of employee health. J Occup
14–20. Environ Med. 2010;52:310–317.
3. Ricci JA, Chee E. Lost productive time associated with excess weight in the 29. Haugland S, Wold B. Subjective health complaints in adolescence—reliability
U.S. workforce. J Occup Environ Med. 2005;47:1227–1234. and validity of survey methods. J Adolesc. 2001;24:611–624.
4. Ricci JA, Chee E, Lorandeau AL, Berger J. Fatigue in the U.S. workforce: 30. Kawada T, Ooya M. Workload and health complaints in overtime workers: a
prevalence and implications for lost productive work time. Occup Environ survey. Arch Med Res. 2005;36:594–597.
Med. 2007;49:1–10. 31. Hildingh C, Luepker RV, Baigi A, Lidel E. Stress, health complaints and
5. Schultz AB, Edington DW. Employee health and presenteeism: a systematic self-confidence: a comparison between young adult women in Sweden and
review. J Occup Rehabil. 2007;17:547–579. USA. Scand J Caring Sci. 2006;20:202–208.
6. Sanderson K, Tilse E, Nicholson J, Oldenburg B, Graves N. Which presen- 32. Roelen CAM, Schreuder1 KJ, Koopmans PC, Groothoff JW. Perceived job
teeism measures are more sensitive to depression and anxiety? J Affect Dis. demands relate to self-reported health complaints. Occup Med. 2008;58:
2007;101:65–74. 58–63.
7. Gates DM, Succop P, Brehm BJ, Gillespie GL, Sommers BD. Obesity and 33. Ihlebaek C, Brage S, Eriksen HR. Health complaints and sickness absence in
presenteeism: the impact of body mass index on workplace productivity. Norway, 1996–2003. Occup Med. 2007;57:43–49.
J Occup Environ Med. 2008;50:39–45. 34. Grøvle L, Haugen AJ, Ihlebaek CM, et al. Comparing working condi-
8. Gisbert JP, Cooper A, Karagiannis D, et al. Impact of gastroesophageal reflux tions and physical and psychological health complaints in four occupational
disease on work absenteeism, presenteeism and productivity in daily life: a groups working in female-dominated workplaces. J Psychosom Res. 2011;70:
European observational study. Health Qual Life Outcomes. 2009;7:90. 548–556.
9. Campo M, Darragh AR. Work-related musculoskeletal disorders are associ- 35. Taloyan M, Aronsson G, Leineweber C, et al. Sickness presenteeism predicts
ated with impaired presenteeism in allied health care professionals. J Occup suboptimal self-rated health and sickness absence: a nationally representative
Environ Med. 2012;54:64–70. study of the Swedish working population. PLOS One. 2012;7:1–8.
10. Mannion AF, Horisberger B, Eisenring C, et al. The association between 36. Leineweber C, Westerlund H, Hagberg J, et al. Sickness presenteeism among
beliefs about low back pain and work presenteeism. J Occup Environ Med. Swedish police officers. J Occup Rehabil. 2011;21:17–22.
2009;51:1256–1265. 37. Ashby K, Mahdon M. Why Do Employees Come to Work When Ill? AXAPPP
11. Burton WN, Pransky G, Conti DJ, Chen CY, Edington DW. The association Health Care Report: 2010:22–23.
of medical conditions and presenteeism. J Occup Environ Med. 2004;46:838– 38. Peter R, Siegrist J. Psychosocial work environment and the risk of coronary
845. heart disease. Int Arch Occup Environ Health. 2000;73:S41–S45.
12. Loeppke R, Taitel M, Haufle V, et al. Health and productivity as a business 39. Löve J, Grimby-Ekman A, Eklöf M, Hagberg M, Dellve L. “Pushing oneself
strategy: a multiemployer study. J Occup Environ Med. 2009;51:411–428. too hard”: performance-based self-esteem as a predictor of sickness presen-
13. Iverson D, Lewis KL, Caputi P, Knosp S. The cumulative impact and asso- teeism among young adult women and men—a cohort study. J Occup Environ
ciated costs of multiple health conditions on employee productivity. J Occup Med. 2010;52:603–609.
Environ Med. 2010;52:1206–1211. 40. Aronsson G, Gustafsson K. Attendance presenteeism: prevalence, attendance-
14. Kowlessar NM, Goetzel RZ, Carls GS, Tabrizi MJ, Guindon A. The relation- pressure factors, and an outline of a model for research. J Occup Environ Med.
ship between 11 health risks and medical and productivity costs for a large 2005;47:958–966.
employer. J Occup Environ Med. 2011;53:468–447. 41. Hansen CD, Andersen JH. Going ill to work—what personal circumstances,
15. Burton J. WHO Healthy Workplace Framework and Model: Background and attitudes and work-related factors are associated with sickness presenteeism?
Supporting Literature and Practices. WHO; 2010. Soc Sci Med. 2008;67:956–964.
16. Virtanen M, Kivimäki M, Elovainio M, Vahtera J, Ferrie JE. From insecure to 42. Green LW, Richard L, Potvin L. Ecological foundations of health promotion.
secure employment: changes in work, health, health related behaviours, and Am J Health Promot. 1996;10(4):270–281.
sickness absence. Occup Environ Med. 2003;60:948–953. 43. Allen J. Building supportive cultural environments. In:O’Donnell MP, ed.
17. Heponiemi T, Elovainio M, Pentti J, et al. Association of contractual and Health Promotion in the Workplace. NY: Delmar Publishers, Inc, Albany;
subjective job insecurity with sickness presenteeism among public sector 2002:202–217.
employees. J Occup Environ Med. 2010;52:830–835. 44. Golaszewski T, Hoebbel C, Crossley J, Foley G, Dorn J, The reliability
18. Wang JL, Schmitz N, Smailes E, Sareen J, Patten S. Workplace characteristics, and validity of an organizational health culture audit. Am J Health Stud.
depression, and health-related presenteeism in a general population sample. J 2008;23(3):116–123.
Occup Environ Med. 2010;52:836–842. 45. 21st Century Foundation. Taiwan Alcohol Use Survey. Available at: http:
19. Huang DT. A New Tool to Measure the Relationship Between Health-Related //www.21stcentury.org.tw/uploads/m file20110112171429.pdf, 2010, Table
Quality of Life and Workforce. School of Industrial and Systems Engineering; D.7, p.65. Accessed March 25, 2015.
2008. 46. Terry PE, Xi M. An examination of presenteeism measures: the association of
20. Hilton MF, Scuffham PA, Sheridan J, Cleary CM, Whiteford HA. three scoring methods with health, work life, and consumer activation. Popul
Mental ill-health and the differential effect of employee type on Health Manag. 2010;13:297–307.

770 
C 2015 American College of Occupational and Environmental Medicine

Copyright © 2015 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited.
JOEM r Volume 57, Number 7, July 2015 Association Between Organization Culture, Health Status, and Presenteeism

47. Burton WN, Chen CY, Conti DJ, Schultz AB, Edington DW. The association 50. Svensen E, Arnetz BB, Ursin H, Eriksen HR. Health complaints and satis-
between health risk change and presenteeism change. J Occup Environ Med. fied with the job? A cross-sectional study on work environment, job satis-
2006;48:252–263. faction, and subjective health complaints. J Occup Environ Med. 2007;49:
48. Gerber M, Pühse U. Do physical activity and fitness protect against stress- 568–573.
induced health complaints? A review of the literature. Scand J Public Health. 51. Ihlebaek C, Eriksen HR, Ursin H. Prevalence of subjective health complaints
2009;37:801–819. (SHC) in Norway. Scand J Public Health. 2002;30:20–29.
49. Huang SL, Lee HS, Li RH, Lai YM, Chen ALC, Tang FC. Differences in health 52. Lynch WD, Sherman BW. Missing variables: how exclusion of human re-
complaints among taiwanese workers in different occupational categories. sources policy information confounds research connecting health and business
J Occup Health. 2012;54:241–249. outcomes. J Occup Environ Med. 2014;56:28–34.


C 2015 American College of Occupational and Environmental Medicine 771

Copyright © 2015 American College of Occupational and Environmental Medicine. Unauthorized reproduction of this article is prohibited.