This article was published in an Elsevier journal. The attached copy
is furnished to the author for non-commercial research and
education use, including for instruction at the author’s institution,
sharing with colleagues and providing to institution administration.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
Author's personal copy
Sedentary Behavior in Dutch Workers
Differences Between Occupations and Business Sectors
Marielle P. Jans, PhD, Karin I. Proper, PhD, Vincent H. Hildebrandt, MD, PhD
Sedentary behavior is an independent risk factor for excess body weight and other health
problems. There are no published data on sitting time at work, or how this is related to
occupation and sector (branch of business). No published study has shown whether
extended sitting at work is compensated for by sitting less during leisure time.
This study used data from a continuous cross-sectional survey, from 2000 to 2005
(N?7720). Workers were asked how many minutes they spent sitting during the preceding
day, both at work and in their leisure time. To test differences in sitting times among
occupational groups and sectors, descriptive analyses and analyses of variance were carried
out in 2006.
On average, the Dutch working population reported sitting for 7 hours each day, one third
of which was at work. Occupational groups and sectors differed significantly in sedentary
behavior, mainly involving sitting periods at work. Workers spending long periods sitting at
work did not compensate by sitting less during their leisure time.
Workers spend a substantial part of their waking and working time seated. Those who sat
for long periods at work did not compensate for this lack of activity by adopting
less-sedentary behaviors during leisure time. To prevent health problems, the best
approach may be to reduce sedentary behavior at work, when traveling to and from work,
and during leisure time.
(Am J Prev Med 2007;33(6):450–454) © 2007 American Journal of Preventive Medicine
syndromes, and diabetes.1–7Sedentary behavior in-
volves activities with a very low energy expenditure
(1.0–1.8 metabolic equivalents [MET]), performed
mainly in a sitting or supine position. To date, studies
have used hours spent watching TV as the main indi-
cator of sedentary behavior.2–4,6,7However, watching
TV is only one of many sedentary activities. Today,
many jobs involve sitting at work; more than one
quarter of the Dutch working population has only
seated work.8Very few published studies have investi-
gated how long workers spend doing seated work, or
how the sitting time at work relates to total sitting time
There are major differences in sedentary behavior
between occupations and sectors (branches of business).
arious studies have shown that, besides too little
physical activity, sedentary behavior is an inde-
pendent risk factor for overweight, metabolic
This is borne out by previous studies on the amount of
physical activity9and obesity10in different occupations.
These studies also have shown that occupations involv-
ing a greater degree of activity have a relatively low
incidence of obesity. The most physically inactive occu-
pations have relatively high prevalence of obesity. Nev-
ertheless, there are some occupations and sectors in
which this link is less clear. Factors such as socioeco-
nomic status (SES) and differences in sedentary behav-
ior also may confound the association.
Sitting time at work is determined largely by the
nature of the work in question. It may be expected that
workers in occupations and sectors with relatively long
periods of sitting at work compensate for this lack of
activity by sitting less during their leisure time. Con-
versely, workers who are seldom seated while working
may spend more time sitting down during their leisure
time. To date, no published studies have explored the
relationship between occupation and sedentary behav-
ior, both at work and during leisure time.
This study describes how many hours per day the
Dutch working population spends sitting down, and
what proportion of this is work-related. The differences
among various occupational groups and sectors in total
sitting time, sitting time at work, and sitting time during
leisure time were examined in a representative sample
of the Dutch working population. It was hypothesized
From the TNO Quality of Life (Jans, Hildebrandt), Leiden;
Body@Work Research Center Physical Activity, Work and Health,
TNO-VUmc (Proper, Hildebrandt); and Department of Public and
Occupational Health, EMGO Institute, VU University Medical Center
(Proper), Amsterdam, The Netherlands
Address correspondence and reprint requests to: Marielle P. Jans,
PhD, TNO Quality of Life, P.O. Box 2215, 2301 CE Leiden, The
Netherlands. E-mail: Marielle.Jans@tno.nl.
Am J Prev Med 2007;33(6)
© 2007 American Journal of Preventive Medicine • Published by Elsevier Inc.
0749-3797/07/$–see front matter
Author's personal copy
that occupations and sectors with relatively long sitting
times at work might compensate during their leisure
time by spending less time sitting.
Study Population and Design
The data were derived from a continuous cross-
sectional survey known as Injuries and Physical
Activity in the Netherlands from 2000 to 2005.
Approximately 10,000 individuals per year were
interviewed by telephone, using a computer-based
survey system. Sampling via random-digit dialing was used, as
it provided a representative sample of Dutch households.
Once a list had been compiled of all the individuals in a given
family, the computer selected one of them to act as the
respondent. One quarter of those interviewed were given
more-detailed questions about the time spent sitting or
supine. These questions referred to the previous day or to 2
days before, to minimize recall bias. No interviews were
conducted on Sundays, so the questions posed on Monday
alternately dealt with “yesterday” and “the day before yester-
day.” Distributing the observations evenly throughout the
days of the week and using a large number of respondents
made it possible to derive a complete picture of the sedentary
behavior of the Dutch population. Only data on workers were
used. Statistics Netherlands (CBS) defines a worker as some-
one aged between 15 and 64 who works at least 12 hours per
Sitting time at work on the previous day was determined by
the following question: Could you indicate approximately
how many minutes you spent sitting at work yesterday? The
same questions were asked with respect to sitting during
traveling to and from work, while performing domestic tasks,
and during leisure time—both throughout the day and
during the evening, separately. Individuals also were asked
how many hours they had spent in bed the previous night.
Total daily sitting time was calculated by totaling the individ-
ual sitting periods. Total sedentary time included the sum of
total sitting time plus the time spent lying in bed. The
psychometric properties of the questions used to evaluate
sitting time are unknown.
Occupational Group and Sector
All respondents were asked about their occupa-
tion as well as the sector in which they worked.
From the occupation, sector, main activities, and
education and income levels of the respondents,
the occupations were assigned a code. The seven
main categories of occupations and 28 sectors
defined by CBS were used for the analyses (Table
1). Data also were collected on the respondent’s age, gender,
highest level of schooling (on a 7-point scale), and number of
working hours. Following the definition of CBS, full-time
workers worked at least 36 hours per week. A further distinc-
tion was made between workers with a small part-time job
(12–25 hours per week) and workers with a large part-time
job (26–35 hours per week).
Data were weighted to family size, age, and gender, to match
the structure of the entire population of the Netherlands,
applying a weighting factor to the entire data file. After this
weighting, only data of workers were used for further analy-
ses. To calculate mean sitting times per occupational group
and sector, descriptive analyses were carried out without any
correction for certain variables because corrected means
would not be representative of the specific occupational
group or sector, as characterized by a given gender-based,
age-based, and education-based distribution. Furthermore,
one-way analyses of variance (ANOVA) were carried out to
test differences in sitting times among occupational groups
and sectors. To determine which occupational groups and
sectors had extremely high or low sitting times, the 95%
confidence intervals (CIs) of the mean time spent sitting per
occupational group and sector were calculated. Since
weighted data cannot be used in the analysis of variance and
in the calculation of CIs, nonweighted data were used. By
using weighted and nonweighted data, the weighted mean
could fall outside the nonweighted 95% CI of the mean. Nine
on page 502.
Table 1. Sedentary behavior by occupational group (weighted means [95% CI for nonweighted means])
Subdivisions of total sitting timeb
Traveling to and
Legislators and senior managers
Scientific and artistic professions
Commercial workers (n?635)
Trade, industrial, or transportation
Service workers (n?796)
Agricultural occupations (n?228)
930 (914–956)495 (481–521)181 (168–197) 38 (33–43)3 (1–3) 103 (95–116) 171 (166–182)
aDifferences between the occupational groups were statistically significant (p?0.01).
bAs a result of rounding, the sum of the periods spent sitting may deviate slightly from the row total.
CI, confidence interval.
December 2007 Am J Prev Med 2007;33(6)
Author's personal copy
sectors with fewer than 100 respondents were excluded from
the analyses, which were carried out in 2006.
General Characteristics of the Study Population
Of the 60,628 respondents, 25,954 met the definition
of worker. Of these, 7720 were given detailed ques-
tions about sedentary behavior. The mean age of the
study population (N?7720) was 39 (SD?11), 60%
were men, and 38% had a higher vocational educa-
tion or university. Twenty percent had a small part-
time job, 16% had a large part-time job, and 64%
On average, Dutch workers spent 862 minutes (well over
14 hours) per day either sitting or supine (Table 1). Of
this, an average of 423 minutes (7 hours) per day was
spent sitting. The evenings were spent mainly sitting
(almost 3 hours per day). Work and traveling to and
from work accounted for one third of the total sitting
time (well over 2 hours per day). This average also
covered nonworking days (weekends and holidays),
during which no time was spent seated at work. For
those who had worked on the previous day, the
average time spent seated at work and while traveling
to and from work was 237 minutes (almost 4 hours)
Full-time workers spent 444 minutes per day sitting.
Of this total, 163 minutes (almost 3 hours) per day were
spent sitting during work and during traveling to and
from work. The total sitting time and the sitting time
during work and during traveling to and from work was
significantly higher for full-time workers than for part-
time workers (p?0.001).
Table 1 shows the differences in sitting time among
occupational groups. There were significant (p?0.01)
differences among the occupational groups in total
sedentary time, total sitting time, sitting time during
work, during traveling to and from work, during house-
work, and during the day (exclusive of work). The
“legislators and senior managers,” the “clerks,” and the
“scientific and artistic professions” sat significantly
longer than the average worker. The “agricultural
workers,” the “service workers,” the “trade, industrial,
or transportation occupations,” and the “commercial
workers” had a significantly lower total sitting time. The
difference between these extremes in occupations was
162 minutes. Full-time workers showed the same overall
There were significant (p?0.05) differences among the
sectors in total sedentary time, total sitting time, sitting
time during work, during traveling to and from work,
during the day (exclusive of work), and during the
evening (Table 2). Those working in computerization,
commercial services, transportation, banking and insur-
ance, and government and judicial organizations sat
significantly longer than the average worker. Workers
in the catering industry; other service functions; agri-
culture, horticulture, fishing; health care; and retail
and wholesale trade had significantly lower total sitting
times. The difference between these extremes in sec-
tors was 218 minutes.
Sedentary Behavior at Work and During
In terms of total sitting time, occupational groups and
sectors differed mainly in sitting time at work. Evening
sitting times differed only slightly (Table 1 and 2).
Where the sitting time at work was above average, the
average number of minutes during leisure time spent
“sitting for other purposes” and “sitting during the
evening” were actually slightly higher than those for
workers who spent less time than average sitting at work
This study of the Dutch working population showed
that, on average, workers were sitting for 7 hours per
day. Well over 2 hours of this total involved time spent
at work and traveling to and from work. While the
number of hours spent sitting at work was not particu-
larly excessive, this average also included nonworking
days (such as weekends and holidays), during which no
time was spent sitting at work. As was shown in the
results, for those who had worked on the previous day,
the average time spent sitting at work and while travel-
ing to and from work was almost 4 hours per day. It can
be concluded that work is a major source of sedentary
behavior as it accounts for one third of the total sitting
These figures represent average values. In some
occupations and sectors, the work-related share of the
total sitting time was much higher. Computerization
had the highest work-related share (45%), while the
lowest was found among service workers (19%).
Different occupations and sectors differed only mar-
ginally in sitting time during leisure periods. Therefore,
there was no indication for the existence of compensa-
tory behavior. Workers in occupations and sectors with
relatively long sitting times at work were not sitting less
during their leisure time than those with short sitting
times at work. This supports recent findings among the
American Journal of Preventive Medicine, Volume 33, Number 6 www.ajpm-online.net
Author's personal copy
full-time working population in Australia.11No rela-
tionship was found between the amount of sitting time
at work and levels of physical activity during leisure
There is some evidence to suggest that heredity is a
factor related to physical activity and sedentary behav-
ior. Genes have been shown to account for more than
half of the individual differences in sport behavior.12,13
While the analyses used a large and representative
database, the fact that these data were self-reported may
mean that the results are biased. As yet no instrument
has been found to measure sedentary activities in a
valid and reliable manner within an epidemiologic
setting. This should be a topic for future research. It is
difficult to estimate periods spent sitting at various
times of the previous day. Individuals might overesti-
mate the length of time. There is no evidence to
suggest that this tendency to overestimate is associated
with any particular occupation or sector.
The low response also might have caused bias. Over
the 6 years, about 50% of the individuals who had been
called did not respond, mainly because they refused to
participate (35%). A small part of the nonresponse was
because they were not reachable by telephone.14As the
focus of the survey was mainly on all kinds of injuries
instead of on sedentary behavior only, bias due to
selective response is assumed to be small. In spite of
this, selection bias cannot be excluded as individuals
who were not reachable might show less-sedentary
behavior than the average worker. This could have
resulted in a small overestimation of the sitting times
presented in this study. Another point of discussion is
that only land lines were used in this study. At the
beginning of the study, 98% of Dutch households were
reachable by land lines, but this percentage is declin-
ing. In the future, this might cause a higher selection
Bias also might have been caused by the weighting
that was applied to match the structure of the entire
population of the Netherlands. When selecting workers
from the entire database, the question arose as to
whether this weighting was actually correct. There are
two reasons for assuming that the weighting produced
very little bias. First, after applying the weighting, the
age and gender structure of the workers showed rea-
sonable correspondence with CBS data for the Dutch
Table 2. Sedentary behavior by sector (weighted means [95% CI for nonweighted means])
Subdivisions of total sitting timeb
Traveling to and
Banking and insurance
Government and judicial
Paper, printing, publishing
Culture, sport, and
Petroleum, rubber, chemical
Food and stimulants
Metal industry (n?348)
Post and telecommunications
Welfare work (n?259)
Retail and wholesale trade
Health care (n?1051)
Other service functions
Catering industry (n?225)
917 (902–944)476 (462–501) 174 (155–185) 28 (23–32)4 (2–6)102 (95–117)169 (166–183)
905 (865–956)461 (423–509)150 (123–186)18 (12–24)7 (0–14) 100 (79–119) 186 (171–206)
859 (826–901)425 (388–458) 123 (98–145)23 (16–29) 5 (2–7)110 (87–125) 164 (153–184)
834 (795–896)412 (381–464)92 (71–122) 18 (10–27)6 (0–11)118 (97–142)178 (164–200)
818 (779–875) 400 (358–445)82 (57–107)19 (12–24) 6 (2–12)111 (91–134)182 (160–203)
765 (723–808)326 (296–363) 62 (42–86)12 (7–16)5 (2–8) 86 (68–106)161 (145–181)
aDifferences between the sectors were statistically significant (p?0.05).
bAs a result of rounding off, the sum of the periods spent sitting may deviate slightly from the row total.
CI, confidence interval.
December 2007Am J Prev Med 2007;33(6)
Author's personal copy Download full-text
working population as a whole. Second, the differences
between the weighted and nonweighted data were
small, and they did not point to any other conclusions.
To improve comparability with the two previous studies
of workers from the same database,9,10the same weight-
ing method was adopted.
These findings demonstrate the need to encourage
physical activity in the workplace. First, this study shows
that work accounts for one third of the total sitting
time. Since sedentary behavior is an independent risk
factor for health problems, policies to encourage phys-
ical activity in the workplace should prompt workers to
adopt a more-active lifestyle and help workers to reduce
sedentary behavior. While largely interrelated, these
aims nevertheless require slightly different approaches.
For instance, company fitness programs help workers to
adopt a more-active lifestyle, but have very little effect
on sedentary behavior. However, such behavior can be
affected by encouraging workers to take active breaks at
regular intervals.15Examples of active breaks include
lunchtime strolls; retrieving supplies, coffee, or printer
output from the other end of the corridor; using
break-reminder software; performing some stretches
and strengthening exercises; yoga; or visiting col-
leagues instead of e-mailing them. Second, since work-
ers do not compensate for long periods of sitting at
work, policies are needed to discourage sedentary
behavior and to encourage physical activities. Those
whose work involves long sitting periods should be
alerted to the need to compensate this by engaging in
physical activity at (before or after) work, as well as by
reducing their time spent sitting.
Finally, the findings show that attempts to encourage
physical activity in the workplace should be tailored to
the occupation and sector in question, given the often
enormous differences between one working environ-
ment and another. In some sectors, workers are forced
to adopt unhealthy, sedentary behavior for most of the
day. To date, studies evaluating the interventions aimed
at a reduction of the time spent sitting are lacking.
From both a public and an occupational health per-
spective, such research is needed.
No financial disclosures were reported by the authors of this
1. Martinez-Gonzalez MA, Martinez JA, Hu FB, Gibney MJ, Kearney J. Physical
inactivity, sedentary lifestyle and obesity in the European Union. Int J Obes
Relat Metab Disord 1999;23:1192–201.
2. Salmon J, Bauman A, Crawford D, Timperio A, Owen N. The association
between television viewing and overweight among Australian adults partic-
ipating in varying levels of leisure-time physical activity. Int J Obes
3. Kronenberg F, Pereira MA, Schmitz KH, et al. Influence of leisure time
physical activity and television watching on atherosclerosis risk factors in
the NHLBI Family Heart Study. Atherosclerosis 2000;153:433–43.
4. Hu FB, Leitzmann MF, Stampfer MJ, Colditz GA, Willett WC, Rimm EB.
Physical activity and television watching in relation to risk for type 2
diabetes mellitus in men. Arch Intern Med 2001;161:1542–8.
5. Hu FB, Li TY, Colditz GA, Willett WC, Manson JE. Television watching and
other sedentary behaviours in relation to risk of obesity and type 2 diabetes
mellitus in women. JAMA 2003;289:1785–91.
6. Jakes RW, Day NE, Khaw K-T, et al. Television viewing and low participation
in vigorous recreation are independently associated with obesity and
markers of cardiovascular disease risk: EPIC-Norfolk population-based
study. Eur J Clin Nutr 2003;57:1089–96.
7. Bertrais S, Beyeme-Ondoua J-P, Czernichow S, Galan P, Hercberg S,
Oppert J-M. Sedentary behaviours, physical activity, and metabolic syn-
drome in middle-aged French subjects. Obes Res 2005;13:936–44.
8. Smulders PGW, Andries F, Otten FWJ. How do Dutch people think about
their work? Design, quality and first results of the TNO Worksituation
Survey (in Dutch). Hoofddorp: TNO Work and Employment, 2001.
9. Proper KI, Hildebrandt VH. Physical activity among Dutch workers. Differ-
ences between occupations. Prev Med 2006;43:42–5.
10. Proper K, Hildebrandt V. Overweight and obesity among Dutch workers.
Differences between occupations (in Dutch). In: Ooijendijk WTM, Hilde-
brandt VH, Hopman-Rock M (eds). Report Physical Activity Measured
2000–2004 (in Dutch).Leiden: TNO Quality of Life, 2006.
11. Mummery WK, Schofield GM, Steele R, Eakin EG, Brown WJ. Occupational
sitting time and overweight and obesity in Australian workers. Am J Prev
12. Carlsson S, Andersson T, Lichtenstein P, Michaëlsson K, Ahlbom A.
Genetic effects on physical activity: results from the Swedish Twin Registry.
Med Sci Sports Exerc 2006;38:1396–401.
13. Stubbe J. The genetics of exercise behavior and psychological well-being
(Thesis). Amsterdam: Vrije Universiteit, 2006.
14. Ooijendijk WTM, Hildebrandt VH, Hopman-Rock M. Physical activity in
the Netherlands 2000–2005 (in Dutch). In: Hildebrandt VH, Ooijendijk
WTM, Hopman-Rock M (Eds). Report Physical Activity and Health 2004/
2005 (in Dutch). Leiden: TNO Quality of Life, 2007.
15. Taylor WC. Transforming work breaks to promote health. Am J Prev Med
American Journal of Preventive Medicine, Volume 33, Number 6 www.ajpm-online.net