PREVENTIVE MEDICINE 26, 138–143 (1997)

ARTICLE NO. PM960117

Smoking Behavior and Attitude Toward Smoking Regulations and

Passive Smoking in the Workplace1
A Study among 974 Employees in the German Metal Industry

HERMANN BRENNER, M.D., M.P.H.,*,2 JÜRGEN BORN, DIPL. SOZ.,† PETER NOVAK, M.D., PH.D.,† AND
VOLKER WANEK, PH.D.†

*Department of Epidemiology, University of Ulm, D-89069 Ulm, Germany; and †Department of Medical Sociology, University of Ulm,
D-89069 Ulm, Germany

Key Words: environmental tobacco smoke; passive

Background. Smoking regulations at the workplace
have been found to be acceptable and effective in
many studies conducted in the United States. There is
limited knowledge, however, on acceptance and ef-
fects of smoking regulations in European countries,
particularly among blue collar employees.
Methods. We conducted a survey on smoking behav-
iour and attitude toward smoking regulations and
passive smoking in a South German metal company. A
self-administered questionnaire was mailed to 1,500
predominantly blue collar employees of whom 974 par-
ticipated in the study (response rate 64.9%).
Results. About 30% of the employees were not al-
lowed to smoke at their immediate work area. Among
them, about 95% of both smokers and nonsmokers
agreed with this smoking policy. More than 60% of
nonsmoking blue collar workers were bothered by
passive smoking at work whether or not smoking was
allowed at their immediate work area. In contrast, the
proportion of nonsmoking white collar employees who
were bothered by passive smoking varied from 52% if
smoking was allowed at their immediate work area to
18% if smoking was not allowed. Prevalence of active
smoking and average amount of smoking among ac-
tive smokers were considerably lower among employ-
ees who were not allowed to smoke at work than
among other employees. These differences were partly
due to confounding by occupation, however, which
was strongly related to both smoking habits and smok-
ing policy.
Conclusions. Our results, which confirm and extend
previous findings, give further support to the accept-
ability and potential effectiveness of smoking regula-
tions at the workplace. Particular efforts should be
devoted to limit both active and passive smoking
among blue collar employees. © 1997 Academic Press
1
This work was supported by Bundeszentrale für gesundheitliche
Aufklärung, Köln, Germany.
2
To whom reprint requests should be addressed. Fax: ++49-731-
5026453.
138
smoking; regulations; workplace.
INTRODUCTION
Concern about the negative health effects of passive
smoking has led to widespread introduction of smoking
regulations in the workplace in some countries, such as
the United States [1,2] and Australia [3]. In contrast,
there are only very limited legal regulations of smoking
in the workplace in the European Community [4,5].
This particularly applies to the Federal Republic of
Germany where smoking remains common. In 1992,
35.1% of men and 20.6% of women were current smok-
ers [6]. Lack of smoking regulations is surprising,
since, although the health risks of passive smoking are
considerably less than those of active smoking, the
number of persons injured by involuntary smoking is
much larger than the number injured by any other
environmental agent already under regulation to limit
exposure [7,8]. Also, it has been demonstrated that for
many nonsmokers exposure to environmental tobacco
smoke (ETS) at the workplace exceeds exposure to ETS
from any other source [9]. As a result, exposure to ETS
at the workplace accounts for a large proportion of mor-
bidity and mortality from passive smoking among non-
smokers, including lung cancer and other respiratory
diseases [10].
Reluctance in adopting restrictive smoking policies
in the workplace may result from concern about their
social acceptance although many studies from the
United States and Australia have confirmed that bans
on workplace smoking are acceptable and effective
public health measures [see Ref. 8 for a review of the
pre-1991 literature, 11,12]. Most previous studies,
however, have been conducted among predominantly
white collar working populations. In particular, a large
proportion of previous studies have been carried out
among employees in the health care sector [13], and
only very few studies have been reported from Europe

0091-7435/97 $25.00
Copyright © 1997 by Academic Press
All rights of reproduction in any form reserved.
 SMOKING REGULATIONS AT THE WORKPLACE 139
[14–16]. In the present study, we assessed smoking be- Table 1 provides basic descriptive statistics on sociode-
havior and attitudes toward smoking regulations and mographic characteristics of participants and smoking
passive smoking in the workplace in a predominantly regulations at the immediate work area. The distribu-
blue collar working population in the German metal tion of basic sociodemographic characteristics among
industry. respondents was similar to that of the entire popula-
tion of eligible employees, although male blue collar
MATERIAL AND METHODS employees were slightly underrepresented. The vast
This analysis is based on a survey on attitudes to- majority of employees were men (86%). The age range
ward health promotion at the workplace carried out in was from 19 to 65. Female employees were on the av-
a company from the metal industry located in the re- erage somewhat younger than their male colleagues.
gion of Ulm in the south of Germany. The company is The majority of both male (74%) and female (63%) em-
mainly engaged in processing of copper. A standard- ployees were married. Levels of school education
ized, self-administered questionnaire (75 items) was among male employees were relatively low, and more
mailed to 1,500 employees of German nationality in than two-thirds of male employees were blue collar
Spring 1995 by the company’s health insurance plan. workers. Most of the latter were working in large pro-
Repeated mailings were employed to enhance the re- duction halls, where they were affected by a variety of
sponse rate. Questionnaires were sent back and ana- occupational exposures, including noise and heat. In
lyzed anonymously. Questions addressed working con- contrast to their male colleagues, the majority of
ditions, perceived health status, health behavior and women were working as clerical employees in private
lifestyle, occupational health protection and health or shared offices. Overall, a majority of 62% of study
promotion, and sociodemographic factors. participants were blue collar workers. About one-
The present analysis focuses on smoking behavior fourth of study participants were current smokers.
and attitudes toward smoking regulations and passive Smoking prevalence was higher among male (27%)
smoking in the workplace. Participants were asked, than among female (17%) employees. Smoking preva-
among other things, whether smoking was or was not lence in this study was well below the national average
allowed at their immediate work area. Attitudes to- which is consistent with previously observed regional
ward smoking regulations were ascertained by agree- variations of smoking habits in Germany [6]. Smoking
ment or disagreement with the smoking regulation at was banned in some sections of the company for vari-
their immediate work area. In addition, participants ous reasons, including safety regulations and compa-
were asked whether (and, if so, how strongly) they ny’s internal arrangements. Overall, smoking was pro-
were bothered by other people’s smoking at the work-
place and in a variety of other settings (such as public
buildings or transportation systems). Attitudes toward TABLE 1
smoking regulations and passive smoking were strati- Sociodemographic Characteristics, Smoking Status, and
fied by current smoking status (smoker versus non- Smoking Regulations at the Immediate Work Area of Study
smoker), smoking regulation at the immediate work Participants
area, and occupation (blue collar versus white collar).
In addition, prevalence of smoking and average total Males Females Total
(n 4 838) (n 4 134) (n 4 974)
numbers of cigarettes smoked per day by current
smokers were compared between employees who were Age
allowed and employees who were not allowed to smoke 19–35 years 35% 50% 37%
36–50 years 38% 27% 37%
at their immediate work area. Multiple linear regres-
51–65 years 27% 23% 26%
sion with the number of cigarettes smoked per day as Marital status
the dependent variable was employed to control the Single 23% 31% 24%
impact of smoking regulations on the amount of smok- Married 74% 63% 72%
ing for potential confounding by other factors. The fol- Widowed/divorced 4% 7% 4%
School education
lowing potential confounders were included as inde- ø9 years 72% 47% 69%
pendent variables in the regression model along with >9 years 28% 53% 31%
the regulation of smoking at the immediate work area: Occupation
age (three categories), sex, marital status (categories: Blue collar 69% 13% 62%
married or other), level of school education (categories: White collar 31% 87% 38%
Smoking status
ø9 years, > 9 years), and occupation (categories: blue Never smoker 41% 59% 44%
collar or white collar). Ex-smoker 32% 24% 31%
Current smoker 27% 17% 25%
RESULTS Smoking regulation
The questionnaire was completed and returned by Smoking allowed 74% 38% 69%
Smoking not allowed 26% 62% 31%
974 of 1,500 eligible employees (response rate 64.9%).
140
hibited at the immediate work area of about 30% of
employees.
Table 2 shows the results on attitudes toward smok-
ing regulations and passive smoking at the workplace.
An overwhelming majority of smokers agreed with the
smoking regulation at their immediate work area. This
applied, whether smoking was or was not allowed.
There were no major differences between blue and
white collar employees in this respect. In contrast,
agreement with the existing smoking regulation varied
strongly between nonsmokers who were or were not
allowed to smoke at their immediate work area: it was
much higher among employees working at smokefree
workplaces (regardless of the reason of smoking prohi-
bition) than among other employees. This difference
was particularly pronounced among white collar em-
ployees. In this group, large differences were also found
in the proportion of nonsmokers who were bothered by
involuntary smoking at work, which ranged from
52.6% if smoking was allowed at the immediate work
area to 18.2% if smoking was not allowed. In contrast,
more than 60% of nonsmoking blue collar workers were
bothered by involuntary smoking, whether or not
smoking was allowed at their immediate work area.
Interestingly, there was also a minority of active smok-
ers who were bothered by other employees’ smoking.
The proportion of nonsmokers who were bothered by
involuntary smoking at the workplace was also com-
pared with the proportion of nonsmokers who were
bothered by involuntary smoking in other settings (see
Table 3). Among blue collar nonsmokers and among
those white collar nonsmokers who worked at areas
where smoking was allowed, about three times as
many employees were bothered by passive smoking at
the workplace than by involuntary smoking in the fam-
ily, the other main source of involuntary smoking [10].
The proportion of white collar nonsmokers who were
bothered by involuntary smoking did not differ be-
tween employees who worked in areas where smoking
TABLE 2
Proportions of Employees Who Agreed with the Smoking
Policy Currently in Effect at Their Immediate Work Area and
Who Were Bothered by Involuntary Smoking at Work, Cat-
egorized by Own Smoking Status, Occupation, and Smoking
Policy
Smoking policy n Agree P Bothered
Smokers
Allowed 185 96.2% 13.6%
Not allowed 57 94.7% 0.63 14.6%
Nonsmokers, blue collar
Allowed 317 70.7% 61.1%
Not allowed 69 85.5% <0.01 65.6%
Nonsmokers, white collar
Allowed 139 70.5% 52.6%
Not allowed 158 97.5% <0.001 18.2%
BRENNER ET AL.
was allowed and employees who worked in areas where
smoking was not allowed for settings other than the
workplace. This suggests that the large difference
found in the proportion bothered by involuntary smok-
ing at the workplace reflects a true exposure effect
rather than structural differences between both
groups. There were also only minor differences be-
tween the proportions of blue collar nonsmokers and
white collar nonsmokers who were bothered by invol-
untary smoking in settings other than the workplace.
Tables 4 and 5 provide data on the smoking habits of
employees by type of smoking regulation at their im-
mediate work area and occupation. Overall, smoking
prevalence was 19.7% among employees who were not
allowed to smoke at their immediate work area com-
pared with 28.5% among employees who were allowed
to smoke (P 4 0.005, see Table 4). This large difference
was partly explained, however, by the overrepresenta-
tion of white collar employees (who had a considerably
lower smoking prevalence) among employees who were
not allowed to smoke at their immediate work area.
Nevertheless, some (though statistically not signifi-
cant) difference persisted even after stratification by
occupation.
Among active smokers, the average number of ciga-
rettes smoked per day was 14.0 for smokers who were
not allowed to smoke at their immediate work area
compared with 18.3 cigarettes per day among other
employees (crude difference −4.3, 95% confidence inter-
val −6.9,−1.6) (see Table 5). The amount of smoking
was also lower among female smokers than among
male smokers, among smokers with higher levels of
school education than among other smokers, and
among white collar employees than among blue collar
employees. A difference of about two cigarettes per day
by type of smoking regulation at the immediate work
area persisted after simultaneous adjustment for all
other variables listed in Table 5 using multiple linear
regression, but this difference was no longer signifi-
cantly different from 0 (95% confidence interval −5.0,
+0.8).
DISCUSSION
An inherent limitation of our study is its cross-
sectional nature. This implies that, in theory, differ-
ences in smoking habits between employees working
P
under various types of smoking regulations might be
due to the fact that smoking habits of employees influ-
enced the type of smoking regulation at their immedi-
0.85
ate work area rather than vice versa, or that selection
of the work area by employees was influenced by smok-
ing habits. From a practical perspective, both possibili-
0.50
ties appear to be extremely unlikely, although they
cannot be ruled out with absolute certainty on the ba-
sis of our data.
<0.001
In our study, an important factor contributing to the
 SMOKING REGULATIONS AT THE WORKPLACE 141
TABLE 3
Proportion of Nonsmokers Who Were Bothered by Involuntary Smoking in Various Settings, Categorized by Occupation
and Smoking Policy at the Immediate Work Area

Occupation
Blue collar White collar
Setting Smoking allowed Smoking not allowed Smoking allowed Smoking not allowed
At the workplace 61.1% 65.6% 52.6% 18.2%
In the canteen 14.3% 17.2% 15.9% 15.9%
In the family 24.1% 17.7% 18.5% 17.5%
In restaurants 83.6% 95.3% 91.9% 92.3%
Among friends 63.6% 64.5% 59.7% 61.5%
In public buildings 53.2% 65.6% 63.2% 60.0%
In transportation systems 51.9% 60.4% 60.8% 63.0%

major differences in smoking habits by type of smoking
regulation at the immediate work area was confound-
ing by other factors, in particular, type of occupation,
which was strongly related to both smoking policy and
smoking habits. Nevertheless, some difference in the
average total number of cigarettes smoked per day be-
tween smokers who were and who were not allowed to
smoke at their immediate work area persisted after
adjustment for potential confounders in multiple re-
gression analyses. The adjusted difference was not sig-
nificantly different from 0, however, and it was smaller
than in most previous studies, the majority of which
were conducted among predominantly white collar
working populations [12,14,15,17–19]. This may be
partly explained by limited enforcement of smoking re-
strictions in this company which is suggested by the
considerable proportion of employees who were both-
ered by involuntary smoking at the workplace even if
smoking was not allowed at their immediate work
area. Protection from involuntary smoking seems to
have been particularly ineffective for blue collar work-
ers, who predominantly worked in large production
halls. A total smoking ban in the entire work area
rather than selective restrictions in specific sections
appears to be needed to achieve meaningful protection
from involuntary smoking under such conditions.
Our findings regarding attitudes toward restriction
of smoking at the workplace confirm results of previous
studies which showed that such restrictions are ac-
TABLE 4
Prevalence of Current Smoking by Occupation and
Smoking Policy at the Immediate Work Area
Smoking
Occupation policy n Prevalence
Blue collar Allowed 476 32.1%
Not allowed 96 25.0% P40.17
White collar Allowed 172 18.0%
Not allowed 192 17.2% P40.83
Total Allowed 663 28.5%
Not allowed 293 19.7% P40.005
cepted by both nonsmokers and smokers and that ac-
ceptance tends to increase after their implementation
[3,8,16]. In our study, almost 95% of smokers who were
not allowed to smoke at their immediate work area
agreed with this restrictive policy. These high levels of
agreement are consistent with the observation that
many smokers consider restrictive smoking policies as
an aid to reduce or quit smoking rather than a dis-
criminating measure [16].
The high levels of agreement of nonsmokers with
either type of smoking policy are not surprising given
that smoking policies do not interfere with their own
habits. Nevertheless, the much lower agreement with
the existing policy in work areas where smoking was
allowed points to the dissatisfaction with exposure to
their colleagues’ smoking at their work area under
such conditions. This is consistent with the finding that
more than 50% of white collar nonsmokers who worked
in areas where smoking was allowed and more than
60% of blue collar workers (who seem to have been
exposed to involuntary smoking even if smoking was
not allowed at their immediate work area) were both-
ered by involuntary exposure to tobacco smoke at the
workplace. Although a majority of nonsmokers also in-
dicated being bothered by involuntary smoking in res-
taurants, among friends, and in public buildings, the
large share of time spent at the workplace underlines
the importance of this particular source of exposure
[20]. Although the time spent in the family exceeds the
time spent at the workplace, a considerably higher pro-
portion of nonsmokers were bothered by exposure to
involuntary smoking at the workplace than by expo-
sure to involuntary smoking in the family. This par-
ticularly applied to blue collar employees and to those
white collar employees who were working in areas
where smoking was allowed. In fact, it has been dem-
onstrated that exposures to ETS in workplaces that
allow smoking are comparable with, and often greater
than, exposures in the family of nonsmokers married to
smokers. The latter have been shown to be at increased
risk for lung cancer and heart disease [9,10,21]. Invol-
142 BRENNER ET AL.

TABLE 5
Average Number of Cigarettes Smoked per Day by Current Smokers, Categorized by Smoking Policy at the Immediate
Work Area and Sociodemographic Variables

Crude Adjusted
Average differencea differencea,b
n number (95% CI) (95% CI)
Smoking policy
Allowed 185 18.3 Ref Ref
Not allowed 57 14.0 −4.3 (−6.9,−1.6) −2.1 (−5.0,+0.8)
Sex
Male 220 17.9 Ref Ref
Female 22 11.5 −6.3 (−10.2,−2.4) −2.4 (−6.7,+2.0)
Age
19–35 years 114 17.9 Ref Ref
36–50 years 79 16.1 −1.9 (−4.4,+0.7) −2.9 (−5.6,−0.2)
51–65 years 50 17.6 −0.3 (−3.3,+2.7) −0.4 (−3.5,+2.8)
Marital status
Married 154 17.5 Ref Ref
Other 89 16.8 −0.7 (−3.1,+1.7) −0.6 (−3.1,+1.9)
School education
ø9 years 177 18.3 Ref Ref
ù10 years 63 14.5 −3.7 (−6.3,−1.1) −2.2 (−5.0,+0.5)
Occupation
Blue collar 176 18.7 Ref Ref
White collar 63 13.1 −5.7 (−8.1,−3.2) −3.4 (−6.4,−0.4)
a
Ref, reference category.
b
Adjusted for all other variables listed in the table by multiple linear regression.

untary smoking at the workplace thus represents a
major health hazard [22].
Given the increasing evidence on the adverse health
effects of exposure to involuntary smoking [10,21] and
on the potential effectiveness of smoking regulations in
the workplace in reducing both exposure of nonsmok-
ers to involuntary smoking and the amount of smoking
among active smokers [9,14,15], implementation of
smoking regulations should be enforced. This particu-
larly applies to countries like Germany, where smok-
ing prevalence continues to be high. Our data suggest
that particular efforts should be made to limit both
active and passive smoking among blue collar working
populations. Restrictive smoking policies should be
embedded in comprehensive health promotion pro-
grams in the workplace [23–25] and introduced with-
out discriminating against smokers. Our study among
a predominantly blue collar working population adds
to the evidence that smoking regulations implemented
in that way may have the additional beneficial effect of
reducing rather than increasing potential tensions be-
tween smokers and nonsmokers.
REFERENCES
1. Rigotti NA, Pashos CL. No-smoking laws in the United States:
an analysis of state and city actions to limit smoking in public
places and workplaces. JAMA 1991;266:3162–7.
2. Novotny TE, Romano RA, Davis RM, Mills SL. The public health
practice of tobacco control: lessons learned and directions for the
states in the 1990s. Ann Rev Public Health 1992;13:287–318.
3. Borland R. Changes in prevalence of and attitudes to restrictions
on smoking in the workplace among indoor workers in the state
of Victoria, Australia, 1988–90. Tobacco Control 1992;1:19–24.
4. Sasco AJ, Dalla-Vorgia P, van der Elst P. Comparative study of
anti-smoking legislation in countries of the European Economic
Community. Lyon: International Agency for Research on Can-
cer, 1992; IARC Technical Rep. No. 8.
5. Goldman L. The effect of implementation of non-smoking poli-
cies in the workplace: an update on the law. Addiction 1993;88:
1321–5.
6. Statistisches Bundesamt. Gesundheitswesen, Fachserie 12,
Reihe S.3. Fragen zur Gesundheit [in German]. Stuttgart:
Metzler–Poeschel, 1994.
7. U.S. Department of Health and Human Services. The health
consequences of involuntary smoking: a report of the Surgeon
General. Washington: U.S. Department of Health and Human
Services, Public Health Service, Centers for Disease Control,
1986; DHHS Publication No. (CDC) 87-8398.
8. Fielding JE. Smoking control at the workplace. Ann Rev Public
Health 1991;12:209–34.
9. Hammond SK, Sorensen G, Youngstrom R, Ockene JK. Occupa-
tional exposure to environmental tobacco smoke. JAMA 1995;
274:956–60.
10. U.S. Environmental Protection Agency. Respiratory health ef-
fects of passive smoking: lung cancer and other disorders. Wash-
ington: U.S. Environmental Protection Agency, 1992; EPA Pub-
lication No. 600/6-90/006F.
11. Hocking B, Borland R, Owen N, Kemp G. A total ban on work-
place smoking is acceptable and effective. J Occup Med 1991;33:
163–7.
12. Wakefield MA, Wilson D, Owen N, Esterman A, Roberts L.
Workplace smoking restrictions, occupational status, and re-
duced cigarette consumption. J Occup Med 1992;34:693–7.
 SMOKING REGULATIONS AT THE WORKPLACE
13. Brenner H, Mielck A. Restrictions of smoking at the workplace
and smoking behavior: a review of the literature [in German].
Soz Praeventivmed 1992;37:162–7.
14. Brenner H, Mielck A. Smoking prohibition in the workplace and
smoking cessation in the Federal Republic of Germany. Prev
Med 1992;21:252–61.
15. Brenner H, Fleischle BMM. Smoking regulations at the work-
place and smoking behavior: a study from Southern Germany.
Prev Med 1994;23:230–4.
16. Brenner H, Fleischle BMM. Social acceptance of smoking regu-
lations in the workplace. Eur J Public Health 1994;4:17–21.
17. Sorensen G, Rigotti N, Rosen A, Pinney J. The effects of a work-
site nonsmoking policy: evidence for increased cessation. Am J
Public Health 1991;81:202–4.
18. Jeffery RW, Kelder SH, Forster JL, French SA, Lando HA, Bax-
ter JE. Restrictive smoking policies in the workplace: effects on
smoking prevalence and cigarette consumption. Prev Med 1994;
23:78–82.
19. Brigham J, Gross J, Stitzer ML, Felch LJ. Effects of a restricted
work-site smoking policy on employees who smoke. Am J Public
Health 1994;84:773–8.
143
20. Emmons KM, Abrams DB, Marshall RJ, Etzel RA, Novotny TE,
Marcus BH, et al. Exposure to environmental tobacco smoke in
naturalistic settings. Am J Public Health 1992;82:24–8.
21. Fontham ETH, Correa P, Reynolds P, Wu-Williams A, Buffler
PA, Greenberg RS, et al. Environmental tobacco smoke and lung
cancer in nonsmoking women: a multicenter study. JAMA 1994;
271:1752–9.
22. Repace JL, Lowrey AH. An enforceable indoor air quality stan-
dard for environmental tobacco smoke in the workplace. Risk
Anal 1993;13:463–75.
23. Fielding J. Evaluations, results and problems of worksite health
promotion programs. In: Cataldo M, Coates T, editors. Health
and industry: a behavioral medicine perspective. New York:
Wiley, 1986.
24. Pelletier K. A review and analysis of the health and cost-effective
outcome studies of comprehensive health promotion and disease
prevention programs. Am J Health Promot 1991;5:311–5.
25. Sorensen G, Lando H, Pechacek TF. Promoting smoking cessa-
tion at the workplace: results of a randomized intervention
study. J Occup Med 1993;35:121–6.