occup med (lond) 1996 lam 351 5

7/24/2019 Occup Med (Lond) 1996 Lam 351 5

1/5

Occup. Med. Vol. 46, No. S, pp. 351-355, 1996

Copyright @ 1996 Rapid Science Publishers for SOM

Printed in Grea t Britain. All rights reserved

0962-7480/96

Smoking and exposure to

occupational hazards in

8^304

workers in Guangzhou, China

T. H. Lam ,* C. Q . Jiang,*

W. W.

Liu,

1

W.

S . Zhang,*

J. M . He* and C. Q. Zhu

f

*Department of Comm unity Medicine, The University of Hong Kong,

Patrick Manson Building South Wing, 7 Sassoon Road Pokfulam,

Hong Kong; ^Guangzhou Occupational Diseases Prevention and

Treatment Centre, Huang Po Dong, Guangzhou 510420, China

This cross-sectional study aimed to describe the pattern of smoking in relation to

occupational

hazard

exposure in a working population in Guangzhou, China. In

1994,

data on smoking and occupational hazard exposure from occupational health

records of 8,304 subjects aged 35 years or older from 47 randomly selected factories

were studied. About 49 of the men and 55 of the wome n were exposed to dust,

chemicals or other hazards. The prevalence of smoking was 56.1 in men and 2.0

in women. The prevalence of sm oking in men was higher in those w ho were younger,

with primary education or who were workers. In women, those who were older, with

primary education or in management jobs had higher smoking prevalence. In men

and women, subjects who were exposed to occupational hazards had higher smoking

prevalence: the highest (71.6 ) was found in male workers expose d to dust. Urgent

tobacco control measures are needed to prevent the epidemic of smoking-related

and occupation-related diseases in the workplace in China.

Key

words:

China; occupational hazards; smoking.

Occup. Med. Vol. 46, 351-355, 1996

Received 20 M arch 1996; ccepted in final form 18 June 1996

INTRODU TION

The high prevalence of smoking and the increasing

incidence of mortality attributable to smoking in China

is a major public health problem locally and globally.

1

While most studies on smoking are usually published

in the Chinese literature, a few have recently appeared

in international journals and have aroused much

concern and interest.

2

4

In the field of occupational health in China, the m ain

concern is with exposure to occupational hazards;

smoking is often not a priority for health promotion

or protection in the workplace. There are few reports

which focus specifically on smoking prevalence in

relation to exposure to occupational hazards in the

working population. A study in a large petrochemical

complex in Shanghai showed that smoking was posi-

tively associated with sick leave after adjustment for

age,

alcohol consumption and exposure to chemicals.

Correspondence and reprint requests to:Prof. T. H. Lam, Department

of Community Med icine, The University of Hong Kong, Patrick Manson

Building South Wing, 7 Sassoon Road, Pokfulam, Hong Kong. Tel:

(+852) 2819 9287 or (+852) 2819 9280; Fax: (+852) 2855 9528.

However, the prevalence of smoking in the men

exposed to chemicals was similar to those who were

not exposed (80% compared with 79%).

5

A study on

byssinosis in Guangzhou showed that 67.6% of the

men in two cotton factories smoked, compared with

61.9% in control workers who had no history of

exposure to dust or other toxic substances.

6

Because

of the small number of woman workers who smoked,

no further analysis on female smoking patterns was

possible in both studies.

This study aimed to describe the pattern of smoking

in a working population aged 35 years or older in

relation to their exposure to occupational hazards and

other demographic variables.

METHODS

In 1 988, the 'Guang dong Province Lab our Safety and

Hygiene Regulations' were enacted by the Guangdong

Province People's Congress. From 1990-1992, under

the direction of the Guangzhou Occupational Diseases

Prevention and Treatment Centre (GODPTC), an

byguestonNovember4,2015

http://occmed.oxfordjournals

.org/

D

ownloadedfrom
http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/http://occmed.oxfordjournals.org/


2/5

352 Occup. Med. Vol. 46. 1996

Occupational Health Surveillance Record System was

established basedon theregulations.T he aimwasto

establish two main types of records: (1) Individual

Worker's Health Surveillance Record(a20 page record

book) and (2) Industrial Hygiene Records (13 data

forms),

forall factories ofcounty levelorabove. The se

records were designed by the GODPTC. Occupational

health surveillance teams were formed

by

medical

and

hygiene personnel

who had

undergone

a

training

programme specifically organized

for

establishing

the

record system. Allworkers were medically examined

and were classified into exposed and unexposed

categories. Their demographic characteristics, work

history, medical history and smoking history were also

recorded in their individual records. Those with

abnormalities detected were followed-up andreceived

advice

or

treatment

as

approp riate. Workers classified

as exposed

to

hazardous substances were

to be

exam-

ined annually.

Industrial hygiene measurements were recorded

for

each factory and were evaluated to identify areas which

needed improvements. Advice onim provementswas

given, followed

by

further evaluations.

By the end of

1992 when

the

record system

was

completed,

860

factories

and

about 400,0 00 workers were included.

This covered

84 of all

factories

and 81 of all

workers eligible. Because informationinthe individual

worker's records was not computerized and there were

no resources to do so,detailed analysis of workers'

data, such as the prevalence of smoking, werenot

available.In 1994,the present study wasinitiatedas

a pilot study

to

plan

for the

setting

up of a

cohort

to

study the effect of smoking and occupational ex posures

on mortality. Factory doctors from

47

randomly

selected factories were requested

to

retrieve

28

items

of information, including demographic characteristics,

smoking history (amount smoked per day andyears

of smoking), occupational exposure category (e.g.

dust, chemicals, physical and other hazards, without

specifying thenature of the substances) andcommon

diseases detected, from therecordsofall workers who

were aged

35

years

or

older when the record was first

established, onto standard data retrieval forms.

T he

lower

age

limit

of 35

years

was

chosen because

the

cohort

had to

include older workers

to

increase

the

yield of mortality in future follow-up studies (there

were not enough resources to include youngersub-

jects).

The response rate of thefactory doctorswas

100%.

The data were entered

and

checked

by

using

EPI-

I N F O

and

a n a l y z e d

by

S P S S - P C . S t a t i st i c a l

procedures used included

2t e s t

> Fisher exact test,

-test, analysis ofvariance,and the Coxproportional

hazard model. The latter wasused to estimate the

prevalence rate ratio (P RR ) for smoking after adjusting

for other factors.

7

Two factories (with

a

total

of 324

workers) were

randomly selected from

the 47

factories

for

double

data retrieval

and

entry

and the

agreement

was

over

98%.

Smokers were defined as those who smoked atleast

one cigaretteperday. As there werefewoccasionalor

ex-smokers (only 30 subjects), they were classifiedas

non-smokers.

RESULTS

A totalof8,304 persons aged 35orabove, 4,637 men

(55.8%)and 3,667 women (44.2%), from 47 factories

were included. The distribution of the factories by

nature of industry was: light industry, 17; mechanical

and electrical, eight; chemical, six; pharmaceutical,

three; packaging, three; textile, three; construction

materials, two; electronics, one; fuel, one; metallurgy,

one;

others, two. Twelve

of the

factories were under

the GODPTC and the rest were from the eight district

anti-epidemic stations of Guangzhou. Each district

contributed threeto seven factories.

Table

1

shows

the

demographic characteristics

of

the sample. About 49%

of the men and

55 %

of the

women were classified

as

subjects exposed

to one or

more occupational hazards.

The

commonest occupa-

tional exposure wastochemicals, followed by physical

factors (e.g., noise) anddust (mainly silica, cemen t,

welding and organic dusts).

Table 1

.

Demographic character ist ics

of the

sample

Age (yrs)

3 5 - 3 9

4 0 - 4 4

4 5 - 4 9

50 -54

5 5 - 5 9

60 -64

65 +

Education (missing

Primary

Secondary

Post-secondary

Male

(n= 4,637)

No .

1,127

1,179

91 8

89 5

44 1

50

27

I 166

1,149

2,989

415

Marital status (missing

18

Single

Married

135

4,489

Occupation (missing40

Management

Workers

1,369

3,246

Occupational exposures (missing

Dust

Chemicals

Physical factors

Mixed exposure*

Other factors*

No exposure

390

63 7

477

121

750

2,260

24.3

25.4

19.8

19.3

9.5

1.1

0.6

25.2

65.6

9.1

2.9

97.1

29.7

70.3

2)

8.4

13.7

10.3

2.6

16.2

48.8

C

No .

1,529

1,284

63 2

159

28

17

18

1,269

2,195

121

106

3,556

68 7

2,962

23 8

51 7

19 0

45

65 2

2,025

Female

n

=

3,667)

41.7

35.0

17.2

4.3

0.8

0.5

0.5

35.4

61.2

3.4

2.9

97.1

18.8

81.2

6.5

14.1

5.2

1.2

17.8

55.2

'Mixed exposures were exposures tomore thanone of thefactors abo ve;

other factors included biological, radioactive

and

other less w ell-defined

factors.



.org/

D

ownloadedfrom


3/5

T. H. Lam

et

a/.: Smoking and occupational hazards in China 353

Table 2. Prevalence of smoking in men and women

Age (yrs)

3 5 - 3 9

4 0 - 4 4

4 5 - 4 9

5 0 - 5 4

5 5 - 5 9

60 +

Education

Primary (< 6 yrs)

Secondary (6-12 yrs)

Post-secondary (12+ yrs)

Marital status

Single

Married

Occupation

Management

Workers

Occupational exposures

Oust

Chemicals

Physical factors

Mixed exposure

Other factors

No exposure

All subjects

Male

Wo.of smokers Prevalence

64 2

67 2

531

52 4

21 5

16

x

2

70 5

1,679

184

x

2

71

2,524

. 3

69 0

1,897

x

2

276

35 7

29 3

83

35 2

1,238

*

2

2,600

57.0

57.0

57.8

58.5

48.8

20.8

= 52.7, df = 5, p< 0.001

61.4

56.2

44.3

=

36.1,

df =2 , p< 0.001

52.6

56.2

t

2

= 0.7, df = 5, p

0.4

50.4

58.4

= 25.3, d f= 1 . p < 0 .001

70.8

56.0

61.4

68.6

46.9

54.8

= 74.4, df = 5, p < 0.001

56.1

X

2

= 2737.0,

Female

No. of

smok rs

Prevalence

33

20

10

10

1

1

x

=

32

41

1

x

2

2

73

x

2

20

54

x

2

10

34

7

1

1

22

x

=

75

df

1, p < 0.001

2.2

1.6

1.6

6.3

3.6

2.9

17.0,df = 5, p = 0.004

2.5

1.9

0.8

= 2.6, df =2 , p = 0.27

1.9

2. 1

=

0.01,

df

= 1 ,

p = 0.9

2.9

1.8

= 2.8, df =

1,

p = 0.9

4. 2

6.6

3.7

2. 2

0.2

1.1

82.0,

df = 5, p < 0.001

2.0

Th e prevalence of smoking was 56.1% in men and

2.0%

in women (Table 2). The prevalence of smoking

in men was the highest in those who were younger,

i.e. aged 35-54 years (57-59%), with primary educa-

tion (61%) and who were workers (58%). Although

the prevalence of smoking in women was low, some

notable differences were obvious. The highest preva-

lence of smoking was seen in older women. Although

not statistically significant due to small sample size,

the management personnel had higher smoking preva-

lence (2.9%) than the workers (1.8%) but there was

a similar decreasing trend of smoking with increasing

educational level.

In relation to occupational exposures, the highest

prevalence of smoking in m en was found in those who

were exposed to dust (71%). In women, it was found

in those exposed to chemicals

(6.6%),

followed by dust

(4.2%).

In Table 3, the increase in risk of smoking was first

estimated by the crude PRR which showed that, in

men, increased risk of smoking was associated with

younger age, lower education, being a worker and

being exposed to dust. After adjusting for each other,

the results were similar. After stepwise p roc edu res,

only age, education and occupation al exposures

remained in the final model. The adjusted PRR's in

the stepwise model were similar to those in the model

which included all five variables and are not reported

here.

In women, re-grouping was necessary for age and

education because of the small number of smokers.

The crude PRR showed that increased risk of smoking

was associated with older age, lower education and

exposure to chemicals and dust. After adjustment, the

results were similar. After stepwise proc edur es, the two

factors remaining in the model were age and occupa-

tional exposures.

In Table 4, after breakdown by occupation and ge nder,

the group with the highest prevalence of smoking was

identified: 71.6% of workers exposed to dust were

smokers. With the exception of female workers, the

subjects exposed to dust had the highest prevalence



.org/

D

ownloadedfrom


4/5

354 Occup.Med. Vol.46, 1996

Table 3. Crude and adjusted prevalence rate ratio (PRR) for smoking in men and women

Age (yrs)

Male: 60+

55-59

35-54

Female: 35-49

50+

Education

Post-secondary

Secondary

Primary

Marital status

Single

Married

Occupation

Management

Workers

Occupational exposures

No exposure

Dust

Chemicals

Physical factors

Mixed exposure

Other factors

Crude PRR

1.00

2.35

2.77

1.00

1.27

1.38

1.00

1.07

1.00

1.16

1.00

1.29

1.02

1.12

1.25

0.86

(95 Cl

(1.41-3.90)

(1.69-4.53)

(1.09-1.48)

(1.18-1.63)

(0.84-1.35)

(1.06-1.27)

(1.13-1.47)

(0.91-1.15)

(0.99-1.27)

(1.00-1.56)

(0.76-0.96)

Male

Adjusted PRR (95 Cl)

1.00

2.35

2.77

1.00

1.19

1.32

1.00

1.10

1.00

1.03

1.00

1.23

1.01

1.07

1.21

0.86

(1.41-3.90)*

(1.69-4.54)*

(1.01-1.41)*

(1.10-1.59)*

(0.86-1.39)

(0.93-1.14)

(1.07-1.40)*

(0.90-1.14)

(0.94-1.22)

(0.96-1.51)

(0.76-0.96)*

Crude PRR

1.00

2.96

1.00

1.39

1.00

1 .09

1.00

0.63

1.00

3.87

6.05

3.39

2.05

0.14

Female

(95 Cl) Adjusted PRR (95 Cl)

(1.59-5.48)

(0.88-2.20)

(0.27-4.43)

(0.37-1.05)

(1.83-8.17)

(3.54-10.35)

(1.45-7.94)

(0.28-15.17)

(0.02-1.05)

1.00

2.92

1.00

1.59

1.00

0.95

1.00

0.53

1.00

3.90

6.81

4.23

2.46

0.15

(1.50-5.65)*

(0.97-2.61)

(0.23-3.86)

(0.30-0.94)

(1.79-8.50)*

(3.96-11.71)*

(1.79-9.99)*

(0.33-18.29)

(0.02-1.09)

These remained statistically significant after stepwise procedures. In the final stepwise model: in men, marital status and occupation were dropped and in

women,

education, marital status and occupation were dropped.

Table 4. Prevalence of smoking in management and workers by occupational exposure and gender

Male

Female

No. of

smokers

30

63

18

16

16 3

400

Z

2

= 10.6

245

294

274

67

18 9

827

X

2

=

58.1 ,

Prevalence

65.2

58.3

43.9

61.5

46.8

50.0

, df = 5, p=0.06

71.6

55.7

63.3

70.5

47.1

57.2

df =5, p < 0.001

No. of

smokers

Prevalence

4

5

1

0

1

9

2 tailed:

6

28

6

1

0

13

2 tailed:

16.7

6.8

6. 3

0

0. 7

2.1

Fisher exact test,

dust vs. no exposure, p =

0.003

2.8

6.4

3.4

2.4

0

0.8

Fisher exact test,

dust vs. no exposure, p - 0.02

Management

Dust

Chemicals

Physical factors

Mixed exposure

Other factors

No exposure

Workers

Dust

Chemicals

Physical factors

Mixed exposure

Other factors

No exposure

of smoking.The adjusted PRR's (after adjusting for

age,educationandmarital status)for subjects exposed

to dust vs. unexp osed subjects w ere: (1) male ,

management: 1.29 (0.89-1.87);(2)male, workers:1.21

(1.05-1.40);(3)female, management: 4.95 (1 .30-18.89)

and

(4)

female, workers:

3.70

(1.40-9.78).

On average,

the men

smoked

17.9

cigarettes

per day

(95%

confidence interval

[CFJ =

17.6-18.2)

and

they

had smoked

for 20.4

years

(Cl =

20.0- 20.8 ) whereas

the women smoked

9.1 (Cl =

6.6-11.6) cigarettes

per

day

for 8.1

years

(Cl =

5.8-0.4). Analysis

of

variance

showed

no

increase

in

amount smoked

per day in

dust-exposed subjects compared with

the

unexposed

subjects (data

not

shown).



.org/

D

ownloadedfrom


5/5

T. H. Lam

et al :

Smoking and occupational hazards in China 355

DIS USSION

In the first national smoking prevalence survey in

subjects aged 15 or above in China in 1984,

65.71%

of male workers and 7.32% of female workers were

smokers (regular and occasional).

8

Male workers '

smoking prevalence was the highest among all occu-

pational groups, followed by peasants (male:

63.73%;

female: 6.63%) and governm ent functionaries or cadres

(male: 59.20%; female: 6.51%). In Guangd ong, of which

Guangzhou is the provincial capital, the prevalence

was 66.92% in male workers and 69.77% in male

cadres; the corresponding figures in females were

7.07% and 1.68 . Excluding occasional smokers

(those who smoked less than one cigarette per day),

the prevalence of regular smokers (those who smoked

one or more per day), in men for workers and cadres

was 62.68% and 64.69% respectively and the corre-

sponding figures for women were 5.87% and 1.68 .

Since 1984, although a second national survey was

condu cted in 199 1, prevalence of smoking in workers

is not yet available. It should be noted that the above

data were obtain ed by interviews in comm unity surveys

and are not comparable with our data which were

based on medical examinations by factory doctors in

the workplace. Nevertheless, these data show clearly

that smoking in workers is certainly a major public

and occupational health problem in China.

In ou r review of the literature, we did not find rep orts

on smoking in relation to various occupational expo-

sures in China. In the present study, as a descriptive

cross-sectional study, we have described the prevalence

of exposures to smoking and occupational hazards in

a reasonably representative working population in a

major city in China and we have identified the groups

most at risk. The latter should be the most important

target for preventive measures.

Smoking and exposure to occupational hazards

especially dust and chemicals are the two most

important health hazards in the working population.

In addition to the independent effects of each factor,

the two factors can have synergistic (additive or

multiplicative) effects when smoking workers are

exposed to the work hazards. One important example

is lung cancer.

9

The high prevalence of smoking in

workers exposed to dust is a major concern and the

comb ined exposu res, if not prevented, will lead to very

high incidence of lung cancer and other respiratory

disorders in the next decade. Although we did not

measure exposure to environmental tobacco smoke,

the high prevalence of smokers suggests that many

non-sm okers are exposed to passive smoking. A recent

study in China showed that passive smoking in the

workplace was associated with coronary heart

disease.

10

While the present emphasis in occupational health

in China is on prevention of exposure to occupational

hazar ds, we urge that the preven tion of smoking should

also be a top priority for health promotion in the

workplace. This is particularly relevant for the factory

doctors as they are in the best position to help their

patient workers to quit smoking. Advising patients to

quit smoking should be more straight-forward for the

doctors than to advise management to improve the

work environment because the latter is constrained by

resources and technology. It is useful for occupational

health doctors to remember that when two risk factors

interact (e.g. smoking and dust), addictively or mul-

tiplicatively, removal of one factor (e.g. smoking) will

take away the effect of that single factor and the

interaction effect of both factors (smoking plus dust)

and the health benefit will be substantial. Smoking

cessation is the most cost-effective health promotion

measure in the workplace and this is particularly the

case in China and other developing countries in which

occupational exposures are usually high and smoking

prevalence is increasing. Urgent tobacco control

measures are needed to prevent the epidemic of smoking-

and occupation-related diseases and mortality in the

workplace in China.

KNOWLEDGEMENTS

We would like to thank the Clinical Trial Service U nit,

University of Oxford for funding this study, Dr Sun

Yat Sen Foundation Fund of the Faculty of Medicine,

The University of Hong Kong for supporting Dr C. Q.

Jiang's visitorship to Hong Kong, Professor R. Peto

for advice, all the factory doctors for their participation

and Ms M. Chi for secretarial assistance.

REFEREN ES

1. Peto R, Lopez AD, Boreham J, Th un M , Heath Jr C.

Mortality

from smoking indeveloped countries1950-2000. Oxford, UK:

Oxford University Press, 1994: 101-103.

2. Yu JJ, Mattson ME, Boyd GM ,etal.A comp arison of smoking

patterns in the People's Republic of China with the United

States.

JAmMedAssoc

1990; 264: 1575-1 579.

3. Vartiainen E, Du D, Marks JS,etal.Mortality, cardiovascular

risk factors, and diet in China, Finland, and the U nited States.

Public HealthRe p 1991; 106: 41^6 .

4.

Wang W Q , Dob son AJ. Cigarette smoking and sick leave in an

industrial population in Shanghai, China.IntJ Epidemiol1992;

21: 293-297.

5.

Gong YL, Koplan

JP,

Feng W,et

al.

C igarette smoking in China,

prevalence, characteristics and attitudes in Minghang District.

JAmMedAssoc 1995; 274: 1232-1234 .

6. Jiang CQ , Lam TH , Kong C,etal.Byssinosis in Guan gzhou ,

China.OccupEnviron Med 1995; 52: 268-27 2.

7. Lee J. Odds ratio or relative risk for cross-sectional data? Int

JEpidemiol

1994; 23: 201-20 3.

8. Weng XZ, Hong ZG, Chen DY, Zhang M, Chen BZ, Tin BC.

Acollection ofdataofthe1984 national smokingsample survey.

Beijing: People's Health Publishers, 1987. (In Chinese)

9. Harrington JM, Saracci R. Occupational cancer: clinical and

epidemiological aspects. In: Raffle PAB, Adams RH, Baxter

BJ, Lee WR, eds.

Hunter's iseases of Occupations, 8th Edn.

London: Edward Arnold, 1994: 654-688.

10. He Y, Lam TH , Li LS,

et

al.Passive smoking at work as a risk

factor for coronary heart disease in Chinese women who have

never smoked.

Br MedJ

1994; 308: 380-3 84.



.org/

D

ownloadedfrom

occup med (lond) 1996 lam 351 5

Documents