Upsala J Med Sci 95: 45-52 

Association between Exposure to Asbestos and Pleural 
Effusion. Results from a Questionnaire 

Study of 31000 Persons 
Per Malmberg and Gunnar Hillerdal 

National Institute of Occupational Health, S-17184 Solna, Sweden and Institute of Pulmonary Medicine, 
University Hospital, S-75185 Uppsala, Sweden 

ABSTRACT 

All visitors to a general health survey in 1979, 17,140 men and 14,371 women, completed a 
questionnaire on smoking habits, exposure to asbestos, silica and welding fumes, and diseases such 
as pleural effusion, pneumonia, cough, asthma and diabetes. Seven per cent of the men reported 
exposure to asbestos, 10% to welding fumes, and 6% to silica. Among those who reported work 
related dust exposure there was a higher proportion of smokers, and smokers exposed to dust 
smoked more tobacco per day than non exposed smokers. In the group of men 30-59 years of age, 
who did not indicate exposure to occupational pollutants 2.7% reported previous pleural effusions. 
However, among asbestos exposed men of the same ages, the prevalence was more than doubled 
(5.7%, p<O.Ol). This finding was highly significant in a logistic regression model where age and 
smoking habits were included. The data indicate that 10% or more of diagnosed cases of pleurisy 
could be associated with previous asbestos exposure. 

INTRODUCTION 

Beginning in 1963 and ending in 1985, all inhabitants in the county of Uppsala aged 15 years or 
more were invited to a general health survey every second or third year. During the visit blood 
pressure, hemoglobin, and urine tests were performed and a small size chest roentgenogram was 
taken. Between 50% and 70% of the population participated each time. 

In 1979, all visitors were asked to complete a questionnaire regarding exposure to asbestos, 
silica, welding fumes, and occurrence of certain pulmonary and other symptoms and diseases. The 
aim was to investigate the relation between smoking habits and exposure to some common 
occupational pollutants occurring before 1979 and future pulmonary diseases in a prospective study. 

During the study of the questionnaire responses at follow-up time, observations were made 
which were considered to be of general interest. Thus twice as many men who reported exposure to 
asbestos also reported a history of pleural effusion, compared to men who did not report asbestos 
exposure. Since the survey was performed in 1979, when the relation between pleural effusion and 

45 



asbestos exposure were practically unknown among practicing physicians, the data should be less 
influenced by recall bias than would be the case in similar studies performed at later dates. 

Men Exposed to 
all Asbestos Welding Silica 

Never- 
smokers 40.4 30.2 32.2 29.4 
Ex-smokers 26.2 30.2 28.9 28.4 
Smokers 33.4 39.6 38.9 42.3 

MATERIAL AND METHODS 

Women 
a l l  

54.5 
14.8 
30.7 

All visitors at the health survey were asked to complete a questionnaire. In the questionnaire a 
tobacco smoker was defined as a person who had smoked at least one cigarette per day for one year 
or more (or about 400 cigarettes, or 5 packs of tobacco (75 gram packs) or 200 cigars). An ex- 
smoker was defmed as a smoker who had stopped smoking since at least six months. The answers 
were coded as "Yes", "No" or "Missing". Smokers were also asked to estimate the average tobacco 
consumption per day, the year they started to smoke and the year they stopped smoking, if ex- 
smokers. In the calculation of total tobacco consumption one cigarette was assumed to contain one 
gram of tobacco, a cigar two grams and a parcel of tobacco 75 grams 

The participants were also asked to report if they had been exposed to asbestos dusts. Similarly 
they were asked to report exposure to silica dust and welding fumes. In a third section of the 
questionnaire they were asked to indicate if they presently suffered from or had previously suffered 
from diseases or conditions as listed in Table 4. No distinction was made between "No" replies and 
missing information in the questions regarding dust exposure or diseases. 

The responses were coded twice for improved accuracy and fed into a computer (MicroVAX 11) 
with the database program MIMERB. The statistical analysis was performed using the SAS 
statistical package@, using the Chi2-methcd and logistic analysis of variance (cat-mod procedure). 

RESULTS 

all 

46 

Questionnaires from 17,140 men and 14,371 women contained identification numbers which 
could be used for calculation of age and were thus used in the investigation. Another 100 
questionnaires from men and 70 from women did not contain identification numbers and were 
excluded. In 70 questionnaires from men and 180 from women information on the smoking habits 
were missing. The age dismbution is shown in Fig. 1. 

There were more tobacco smokers among men aged 40 or more than among women and male 
smokers smoked more than femaIe smokers (Tables 1 and 2, Fig. 2); but at younger ages the 
opposite was found. The average age when smoking was commenced was practically constant for 

age groups in men 30 years of age or older. 



Men current-smokers Ex-smokers 
Non- Non- 
exposed Exposed exposed Exposed 

dday 13.6 14.5 14.1 15.9 I :i:M) 10.4 
(S.D.) (8.2) (8.2)l (9.1) (11.6)1 (6.4) 

Women 

smokers smokers 
Current EX- 

1 p<O.OOl for difference between non-exposed and exposed groups. 

Fig. 1. 

Age distribution. Three- 
point moving averages of 
each one year age group. 

2'0 4 0  6 0  Y E A R S  

Fig. 2. 

Age distribution of current 
smokers and ex-smokers 
(plus current smokers) 
among male and female 
participants. Percent of all 
participants in each sex 
and age group. 
See legend to Fig. 1 for 
further detail. 

Exposure to asbestos, welding fumes and silica are shown in Table 3. Among men who 
reported exposure to occupational pollutants there were more smokers, and dust exposed smokers 
smoked more tobacco per day (Table 2) than among age-matched referents. Exposure to 
occupational dusts, particularly asbestos, was reported more often among young men than among 
older age groups (Fig. 3). Of the men aged 30 to 59 years, 8.1% reported exposure to asbestos. 

The number of persons who reported diseases or symptoms are shown in Table 4. The 
occurrences of diseases and symptoms (asthma, pneumonia, pleurisy, rib fractures, diabetes and 
chronic cough) in different age groups are illustrated in Fig. 4-7. As could be expected, the 
prevalence of pneumonia, pleurisy and diabetes increased with age, but asthma was more common 
in the younger age groups. 

47 



Table 3. Exposure to occupational agents and smoking habits (per cent) 

Never- Women 
Smokers smokers smokers 

Asbestos 
Welding 10.4 12.1a 11.5 8.3 0.2 
Silica 5.7 6.2 4.1 0.2 
~ 4 . 0 0 1  for differences between smoking categories. 
bp<0.01 for differences between smoking categories. 

Men 
all CSa XSb NSC 

Pneumonia 13.6 14.6l 15.8 11.4 
Pleuraleffusion 2.9 2.9l 4.1 2.3 
Rib fracture 9.9 11.3l 11.1 7.9 
Diabetes 1.6 1.3l 2.4 1.3 
Asthma 3.6 2.g2 3.8 3.0 
Chroniccough 2.0 3.7l 1.5 0.9 

Of men aged 30 to 59 years who did not report exposure to occupational dusts 2.7% reported 
previous pleuritic disease. Similar prevalences of reported pleural disease were found in men who 
reported exposed to silica dust or welding fumes. However, among men aged 30-59 years who 
reported exposure to asbestos, more than twice as many (5.7%) reported previous pleural effusion 
(Table 5). This difference was not found in men less than 30 years of age. The association between 
exposure to asbestos and pleurisy was significant ( ~ ~ 0 . 0 1 )  in a logistic regression model which 
included age and smoking habits. 

Women 
all CS XS NS 
12.1 11.g2 14.3 11.8 
2.2 1.9ns 2.4 2.4 
3.2 2S2 2.7 3.6 
1.3 0.7l 0.8 1.8 
3.2 3.lnS 3.8 3.0 
1.7 3.01 1.5 1.1 

Table 5. Prevalence of positive answers to question: "Have you or have you ever had pleuritis", 
according to age category and dust-exposure in men. Percent of group and group size. 

15-29 years 
30-44 years 
45-59 years 
60+ years 
All(n=17140) 

aExcluding case S 

No exposure silica asbestosD 
or weldin$ 
0.9% (3807) 0.8% (396) 0.9% (234) 
1.7% (4482) 1.9% (644) 4.4% (498) 
4,1% (3072) 3.7% (376) 8.2% (268) 
5.4% (2942) 7.1% (240) 9.7% (175) 
2.8% (14303) 2.8% (1656) 5.4% (1 175) 

who also reported exposure to asbestos. bIncluding mixed exposure. 

48 



-.-.- s i l i c a  
8 - a s b e s t o s  --- w e l d i n g  ported exposure to occu- 15 

See legends to Fig. 1,2 

Fig. 3. 

Age distribution of re- 

pational dusts among male 
participants. 

for further detail. 
10 

5 

A G E  

20 3 0  4 0  5 0  6 0  Y E A R S  

?=* p l e u r i t i s  
--'d r i b  f r a c t u r e  9---- 

Fig. 4. 

Reported frequency of 20"6 
earlier or present pleuritis 
and rib fractures among 1 5  
male and female partici- 
pants. 
See legends to Fig. 1,2 
for further detail. 

A G E  

2 0  30 4 0  5 0  6 0  Y E A R S  

5 

p n e u m o n i a  

and pneumonia among 1 0  

See legends to Fig. 1,2 5 

Fig. 5. 

Reported frequency of 
earlier or present asthma 

male and female partici- 
pants. 

for further detail. 
A G E  

2 0  30 4 0  5 0  6 0  Y E A R S  

?='diabetes 20%l 

among male and female 5 

Fig. 6. 

Reported frequency of 
earlier or present diabetes 

participants. 
See legends to Fig. 1,2 
for further detail. 

A G E  

l o  1- - 2 0  30 4 0  5 0  6 0  Y E A R S  
4-908571 49 



Fig. 7. 20961 

?--’ c h r o n i c  cough 

1 0  

5 

Reported frequency of 
earlier or present chronic 
cough among male and 
female participants. 
S e e  legends to Fig. 1,2 
for further detail. 

2 0  30 4 0  50 6 0  Y E A R S  

DISCUSSION 

The present data confirm some commonly known observations, such as the circumstance that 
older men smoke more than women whereas the opposite is found among young subjects. 

Among those who reported exposure to silica, asbestos or welding fumes there were more 
smokers and dust exposed smokers smoked more than age matched controls. These findings agree 
with other studies (4, 8) and point to an association between occupation and lifestyle characteristics, 
which should be considered when interpreting data from different occupational settings. 

Young subjects reported present or earlier diagnosis of asthma more often than middle aged 
subjects. This finding was unexpected. If the incidence of asthma is constant, an increasing 
proportion of subjects who have or have ever had the diagnosis would be expected with increasing 
age, rather than the opposite. Several explanations are possible, such as a recent change in 
diagnostic criteria or improved diagnostic capacity. With increasing age, episodes of childhood 
asthma may have been forgotten or middle aged subjects with a history of asthma may have chosen 
not to attend the health survey. The finding can therefore only be regarded as circumstantial evidence 
in the present debate regarding a possible rise in the true incidence of asthma i n  recent years. 
However, the results of the survey do not refute this notion. 

Men who had indicated exposure to asbestos also reported having suffered from pleurisy more 
than twice as often as men who did not report asbestos exposure. This finding needs to be 
interpreted with consideration of the representativity of the material and of possible bias in the 
answers. 

The health survey was in operation from 1963 to 1985, and the participation was fairly constant 
through the years, as was the distribution of participation among different age groups. Thus, for the 
age groups 35 to 65 years, the participation figures have been from 60 to 70%. Those who are 
invited for the first time, i.e. at age 15 to 17, have the highest participation rate (70% or more), b u t  
the second and third time - at age 18 to 25 - the interest seems to be low. Above age 70, again fewer 
persons will attend. The exact figures from the year 1976 have been published previously (8) and 
were as follows: 15-44 years of age,65 %; 45 to 64 years, 72%; above age 65, 48%; and total, 
64%. According to unpublished studies the main reasons for non-padcipation was either that the 

50 



subject was sick and already under medical supervision, or especially among young subjects, a 
feeling of good health, and thus no need for a free medical checkup. 

Between 6-10% of the men indicated exposure to asbestos, silica or welding fumes and about 
3% of all men indicated previous or present episode(s) of pleurisy. The observed frequency of 
reported pleurisy may be influenced by positive and negative selection, but selection can hardly 
explain the findings. Even if a higher proportion of asbestos exposed subjects who had suffered 
from pleurisy would attend the survey, than in the general population, such a selection bias would 
not be of sufficient magnitude to explain the figures obtairied. Furthermore, men exposed to silica or 
welding fumes should have similar bias, but these groups did not have an increased prevalence of 
pleurisy compared to men who did not report exposure to occupational dusts. 

Another possible confounding factor is recall bias. Thus if a relationship between asbestos 
exposure and pleurisy were known to the medical profession or to the lay public, a careful search of 
previous exposure to asbestos may have been performed by the subjects at the time of the diagnosis. 
This would increase the likelihood of remembering such exposure when the questionnaire was 
completed. Benign asbestos pleural effusion is now a well-recognized entity (1, 6, 7). This 
association between asbestos exposure and pleurisy was, however, f i s t  reported in the 1970’s (5, 
10, 12) and was not commonly known among the medical profession at the time of the study 
(1979). The association was virtually unknown in the general public at the time. Therefore recall 
bias is an unlikely explanation. 

There are few studies regarding how many in the general population who have been 
occupationally exposed to asbestos. In the USA it has been estimated that 14 million living workers 
have had significant exposure, which would be more than 10 per cent of the adult male population 
(11). In Norway, in a study of 21,453 men above age 40, a mean of 18% had been occupationally 
exposed to asbestos. In industrialized municipalities, the figure was 9 to 26%, and i n  non- 
industrialized 7 to 8% (8). Uppsala County is not very heavily industrialized and the figures agree 
well with the non-industrialized municipalities in Norway. For the whole of Sweden, the percentage 
figures would probably be more like those of the USA. 

If about 3% of men aged 30 - 60 years report having had pleurisy and twice as many among the 
ten percent of men who also indicate exposure to asbestos, then about 10% of all reported pleurisy 
could be related to asbestos exposure. Studies where subjects with pleural disease have been 
subjected to thoracoscopy have indicated similar proportions (2, 3, 9). The figure 10% may 
underestimate the true proportion of asbestos related pleurisy since this type of pleurisy is often 
associated with few symptoms. Furthermore, a significant number of those who reported pleurisy i n  
the present study who had not indicated asbestos exposure, may have been exposed to asbestos 
unknowingly or may have forgotten this exposure at the time of the inquiry. 

ACKNOWLEDGEMENTS 

This work was supported by the Swedish National Work environment fund. We wish to thank 
mr Henrik Modin, mrs Gun Rosin and mrs Britt-Marie Sundblad for excellent assistance in 
organizing and editing of data. We also would like to express our thanks to mr Per Nasman at the 
institute for statistical analysis, University of Stockholm for statistical advice. 

5 1  



REFERENCES 

1. 
2 .  

3. 

4. 

5. 
6. 

7. 

8. 

9. 

10. 

11. 

12. 

Anonymus: Benign asbestos pleural effusions. Lancet 1:1145-1146, 1988. 
Aubert, M., Parent, B., Brambilla, C. & Bwa, A.: Our experience with thoracoscopy 
(author's transl). Poumon Coeur 37:75-77, 1981. 
Boutin, C., Cargnino, P. & Viallat, J.: Thoracoscopy in chronic pleural effusion (author's 
transl). Poumon coeur 3523-29, 1979. 
Covey, L. S .  & Wynder, E. L.: Smoking habits and occupational status. J Occup Med 

Eisenstadt, H.: Pleural effusion in asbestosis (Letter). N Engl J Med 290:18, 1974. 
Epler, G. R., McLoud, T. C. & Gaensler, E. A.: Prevalence and incidence of benign asbestos 
pleural effusion jn a working population. JAMA 247:617-622, 1982. 
Hillerdal, G. & Ozesmi, M.: Benign asbestos pleural effusion: 83 exudates in 60 patients and 
a review of the literature. Eur J Respir Dis 71:113-121, 1987. 
Hilt, B., LangArd, S . ,  Lund-Larsen, P. G. & Lien, J. T.: Previous asbestos exposure and 
smoking habits in the county of Telemark, Norway - a cross-sectional population study. 
Scand J Work Environ Health 12561-566. 1986. 
Luther, R., Wetzer, K., Kiihe, W., Spormann, H. & Bottger, W.: Asbestos-induced pleural 
diseases-thoracoscopic aspects. Endoscopy 20:4-6, 1988. 
Mamson, S . :  Monosymptomatic exudative pleurisy in persons exposed to asbestos dust. 
Scand J Respir Dis 56:263-272, 1975. 
Nicholson, W. J., Perkel, G. & Selikoff, I. J.: Occupational exposure to asbestos: population 
at risk and projected mortality 1980-2030. Am J Ind Med 3:259-311,1982. 
Sluis, C. G. & Webster, I.: Acute pleurisy in asbestos exposed persons. Environ Res 5:380- 
392, 1972. 

23537-542, 1981. 

Address for reprints: 

Per Malmberg 
National Institute of Occupational Health 
S-171 84 Solna Sweden 

52