American Journal of Respiratory and Critical Care Medicine

Rationale: Occupational exposures to vapors, gas, dust, or fumes have been shown to be a risk factor of airway obstruction in cross-sectional studies in the general population.

Objectives: Our aim was to study the relationships between specific occupations and occupational exposures during a 9-yr follow-up period and changes in lung function and symptoms of chronic bronchitis.

Methods: Subjects from the general population aged 20 to 45 yr were randomly selected in 1991–1993 within the European Community Respiratory Health Survey. Follow-up took place from 1998 to 2002 among 4,079 males and 4,461 females in 27 study centers. A total of 3,202 men and 3,279 women twice completed lung function measurements. Job history during follow-up was linked to a job exposure matrix and consequently translated into cumulative exposure estimates.

Main Results: Individuals exposed to dusts, gases, and fumes during the period of follow-up did not have a steeper decline of FEV1 than did individuals with consistently white-collar occupations without occupational exposures (relative change among men and women, + 1.4 and −3.1 ml/yr, respectively; p > 0.2), nor an increase of prevalence or incidence of airway obstruction defined as an FEV1/FVC ratio of less than 0.7. The incidence of chronic phlegm increased in men exposed to mineral dust (relative risk, 1.94 [1.29–2.91]) and gases and fumes (relative risk, 1.53 [0.99–2.36]), which was not modified by smoking.

Conclusion: Occupational exposures to dusts, gases, and fumes occurring during the 1990s are associated with incidence of chronic bronchitis, although these did not impair lung function in a population of relatively young age.

Occupational exposures to vapors, gases, dusts, or fumes have been shown to be a risk factor of airway obstruction in studies in representative samples of the general population (110) and recently in a comprehensive review (11). These studies have the advantage of including all types and durations of exposures, but at the cost of reducing validity in the measurement of exposure in comparison to workforce-based studies. The use of expert consensus and a job-exposure matrix (JEM) has been reported to increase the validity of exposure assessment in the general population relative to self-reported exposure (1214). In a cross-sectional study among 13,255 young subjects from the general population in 30 centers in 14 countries in the European Community Respiratory Health Survey (ECRHS-I), chronic bronchitis symptoms were associated with self-reported occupational exposures and among smokers with occupational exposures assessed by using a JEM (1). In general, although FEV1 tended to be lower in certain occupational categories, lung function was not associated with self-reported or JEM exposures in ECRHS-I. Nevertheless, results were heterogeneous across countries, and significant relationships between lung function and occupational exposures were found in Spain and New Zealand (8, 9). Cross-sectional studies, however, are limited by the self-selection of the populations, particularly regarding the choice of job. In the follow-up of the ECRHS, a structured, detailed job history was recorded and linked to a general population JEM. The follow-up of this young population for approximately 9 years allowed a relative ageing of these subjects, which is more relevant for evaluating lung function defects and risk factors. Our aim was to study the relationships between specific occupations and occupational exposures during the follow-up period and changes in lung function and symptoms of chronic bronchitis.

Subjects from the random sample selected for ECRHS-I, performed in 1991–1993 (15), who completed a lung function test were included in 27 centers (n = 12,417). Follow-up took place from 1998 to 2002 among 4,079 males and 4,461 females. From these subjects, 76% (3,202 and 3,279, respectively) completed a lung function measurement at baseline and at follow-up. Responders were slightly older and had somewhat better baseline lung function, but there were no marked differences in occupational exposure at baseline as compared with nonresponders.

Of the 27 centers that measured lung function in both surveys and collected the job-history data, 21 used the same spirometer (16). The maximum FEV1 and maximum FVC of up to five technically acceptable maneuvers were determined (17). Change in FEV1 (ΔFEV1) and FVC (ΔFVC) was expressed per year of follow-up (i.e., a negative value represented decline). Airway obstruction was defined as an FEV1/FVC ratio lower than 0.7.

Chronic cough and chronic phlegm (presence of the symptom on most days for at least 3 mo/yr) reported in the second survey were used to define chronic bronchitis as the presence of both symptoms.

All occupations and industries from jobs held during follow-up were recorded using free text entry, and subsequently coded using the International Standard Classification of Occupations-88 classification (18). Codes from all centers of a country were systematically checked by local experts who were trained in the coordinating center. Occupations were grouped into 15 wider categories. Exposures to biological dust, mineral dust, and gases and fumes were assigned using a general population JEM. This JEM classifies each job of an individual's job history into none, low, or high exposure to each of the three types of occupational exposures. Duration of assigned exposure in months was weighted by intensity (with high exposure giving a weight of 4 and low exposure a weight of 1). Occupational exposures at baseline did not differ between participants and nonparticipants (p > 0.1). At baseline, 13% of both participants and nonparticipants were exposed to high levels of any type of occupational exposures and 22% of participants and 23% of nonparticipants were exposed to low levels.

The analyses were performed for men and women separately, and secondarily stratified by smoking status and type of exposure. Associations between occupational exposures during follow-up and lung function change were evaluated using multiple linear regression models, adjusting for center, age, height, and body mass index. “Ever having had a high exposure” and “ever having had a low but never a high exposure” were compared with those classified as consistently “nonexposed.” The distribution of ΔFEV1 was tested for heterogeneity between geographic areas (English-speaking, and northern, central, and southern Europe) (1). The associations between both airway obstruction and chronic phlegm and occupational exposures were examined estimating incidence rates using Poisson regression, with the absolute net change in prevalence between baseline and follow-up determined using generalized estimating equation models for binomial outcome and identity link.

Table 1

TABLE 1. Respiratory health characteristics of the study population by sex and survey



Males (n = 3,951)

Females (n = 4,312)

ECRHS-I
ECRHS-II
ECRHS-I
ECRHS-II
Smoking status
 Never-smokers, %41.438.149.146.1
 Ex-smokers, %20.930.218.825.6
 Current smokers, %37.731.732.028.3
 No. cigarettes 4.68 3.61
Atopy, %*28.726.620.418.8
Symptoms, asthma, % 6.7 9.2 8.210.3
Chronic phlegm, % 7.3 6.9 5.4 5.8
Chronic cough, % 7.1 6.9 7.4 7.1
Chronic cough with phlegm, % 3.0 2.8 2.9 2.7
FEV1/FVC < 0.7, % 5.0 6.9 2.9 4.5
FEV1, L 4.35 4.07 3.22 3.00
Height, m 1.77 1.64
Age, yr34.0142.9833.9442.88
BMI, kg/m2
24.51
26.12
23.23
24.97

*Atopy refers to a specific IgE > 0.35 kU/ml to any of cat, grass, or house mites.

Symptoms refers to shortness of breath, attacks of asthma or taking asthma medication the last 12 months.

3,202 males, 3,279 females with lung function data in both surveys.

Definition of abbreviations: BMI = body mass index; ECRHS = European Community Respiratory Health Survey.

shows the description of variables of interest in each survey in men and women. The average follow-up time between ECRHS-I and ECRHS-II was 8.9 yr (range, 5.8–11.7 yr). During this time, active smoking decreased, particularly in men, as well as slightly did the prevalence of respiratory symptoms, such as phlegm and cough. Symptoms of asthma, airway obstruction, and body mass index increased in both sexes. As expected, FEV1 declined with increasing age.

During the follow-up period, 52% of men and 37% of the women were exposed to dust, gases, or fumes, particularly to gases and fumes (Table 2)

TABLE 2. Descriptive statistics of occupational exposures




Males (n = 3,951)

Females (n = 4,312)
Biological dust, %
 None75.668.5
 Low18.329.9
 High 6.1 1.6
  Cumulative exposure, mo*
   Median (25–75%)103 (58–123)96 (49–109)
Mineral dust, %
 None66.587.1
 Low21.311.0
 High12.3 1.9
  Cumulative exposure, mo*
   Median (25–75%)108 (84–278)85 (36–108)
Gas and fumes, %
 None51.868.2
 Low33.329.5
 High14.9 2.3
  Cumulative exposure, mo*
   Median (25–75%)108 (82–201)94 (45–109)
Any of above, %
 None47.962.9
 Low27.632.6
 High24.5 4.5
  Cumulative exposure, mo*
   Median (25–75%)
224 (110–510)
158 (90–230)
Occupational Categories
n
Males (%)
n
Females (%)
 Consistently white-collar1,92948.821,87443.46
 At least once in following occupation
  Health care 159 4.02 82719.18
  Cleaning 111 2.81 313 7.26
  Transport 288 7.29  31 0.72
  Other manual services 175 4.43 154 3.57
  Construction 231 5.85   6 0.14
  Painting 52 1.32   3 0.07
  Metal industry 332 8.40  31 0.72
  Chemical and related  69 1.75  21 0.49
  Electrical 168 4.25  15 0.35
  Wood, paper, and textile 142 3.59  68 1.58
  Food processing  84 2.13 104 2.41
  Other industry and mining  69 1.75  26 0.60
  Agriculture, fishery, forestry 123 3.11  47 1.09
 Not classifiable
 543
13.74
1,359
31.52

*Among the exposed, months weighted by type of exposure: 1 = low, 4 = high.

. Almost half of the subjects consistently worked in white-collar jobs. Jobs in the metal industry and transport industry among men, and health care and cleaning (in addition to homemaking) among women, were the most prevalent occupations. Homemakers were included in the nonclassifiable (with regard to occupational exposure) group.

Table 3

TABLE 3. Average change in fev1 (ml/yr) in nonexposed participants during follow-up, and additional relative change (se)* among participants with low and high levels of exposure



Males

Females



Nonsmokers
Ex-smokers
Smokers
Nonsmokers
Ex-smokers
Smokers
Males§
Females§
Biological dust
 None (ref)−23.30−17.30−29.11−25.51−23.21−27.37−23.72−25.02
 Low−0.15 (2.56)3.81 (2.78)0.33 (3.07)0.80 (1.54) −7.37 (2.02) −1.70 (2.04)1.26 (1.62)−1.88 (1.05)
 High4.27 (4.26)−4.98 (4.47)−0.34 (5.04)−4.27 (5.92)−11.24 (7.84)−13.57 (7.10)−0.61 (2.67)−8.78 (3.93)
Mineral dust
 None (ref)−23.45−16.72−29.24−25.30−24.96−27.76−23.73−25.51
 Low−1.55 (2.50)2.37 (2.58)0.76 (2.97)1.63 (2.42) −3.36 (3.08) −2.56 (2.77)0.62 (1.56)−0.74 (1.56)
 High6.75 (3.32)−4.65 (3.51)−0.06 (3.31)−6.89 (5.42)−12.92 (6.52) −3.90 (6.65)0.81 (1.95)−7.38 (3.53)
Gas and fumes
 None (ref)−24.19−18.36−28.05−25.37−22.79−27.30−24.21−24.93
 Low3.61 (2.19)1.46 (2.33)−1.02 (2.71)0.03 (1.60) −8.26 (1.99) −3.10 (2.02)1.75 (1.39)−3.00 (1.06)
 High1.02 (2.98)5.73 (3.32)−2.97 (3.27)2.07 (5.16)−11.26 (6.92) 6.48 (6.33)1.13 (1.84)0.37 (3.46)
Any exposure
 None (ref)−23.96−18.49−27.88−25.83−22.80−27.11−24.14−25.04
 Low0.93 (2.34)3.37 (2.46)−2.41 (2.98)1.60 (1.52) −6.88 (1.96) −2.57 (2.01)1.05 (1.49)−1.67 (1.03)
 High
3.10 (2.50)
2.15 (2.77)
−1.53 (2.86)
0.64 (3.68)
−12.29 (4.75)
 −2.33 (4.45)
1.45 (1.57)
−3.14 (2.43)

*Adjusted for age, age2, height, body mass index, change in body mass index, and length of follow-up.

Smoking defined according to both ECRHS-I and ECRHS-II. Subjects inconsistently classified were excluded (n = 277).

p < 0.05.

§Models also adjusted for smoking and number of cigarettes in ECRHS-II for smokers.

Definition of abbreviation: ref = reference category.

shows the lung function decline in ml/yr for each category of occupational exposure defined as “ever high-exposed” and “ever low-exposed” but not high-exposed stratified by smoking and sex. A negative sign means a yearly loss in milliliters in comparison to the reference category, whereas a positive sign means a yearly gain in milliliters. Lung function decline was not associated with any occupational exposure in either men or women. However, after stratifying by smoking, there was a larger decline among women ex-smokers to both low and high levels of any exposure and, after stratifying by type of exposure, there was a larger decline among women exposed to high levels of both biological and mineral dust and to low levels of gases and fumes. There were no differences between the four geographic areas (p for heterogeneity > 0.5). Adjustment for baseline FEV1 did not change the results shown in Table 3, nor did adjusting for asthma or atopy at baseline or replacing “ever” in a high- or low-exposure job by cumulative exposure estimates at any of the different categorizations of duration (results not shown). Stratification by age groups did not modify any of the coefficients shown in Table 3, nor did restriction to subjects older than 25 years at baseline. Finally, assessment of lung function decline as a dichotomous variable (defining as a substantial decline subjects with more than 2 SDs of the average FEV1) did not give any additional information. Analysis by specific occupations showed a lack of association with lung function decline in men and women as did the analysis by occupational exposures at baseline (results available on request). Stratification by time windows of exposure during the follow-up did not provide any association with FEV1 decline in either sex.

The prevalence of airway obstruction increased significantly between the two lung function tests among both males and females (Table 1), but this increase was similar regardless of the occupational exposure during the follow-up period (i.e., there were no significant differences between the increase among those ever exposed to low or high levels and those not exposed; Table 4)

TABLE 4. Absolute net change* and incidence of airway obstruction (fev1/fvc < 0.7) according to occupational exposures by sex



Males

Females
Incidence
Incidence

Baseline
 Prevalence (%)
Absolute net change/yr
 (95% CI)
%
RR (95% CI)
Baseline
 Prevalence (%)
Absolute net change/yr
 (95% CI)
%
RR (95% CI)
Biological dust
 None2.190.17 (0.08 to 0.27)0.7112.200.14 (0.07 to 0.22)0.781
 Low2.250.27 (0.06 to 0.47)0.540.39 (0.05 to 2.99)2.35Model did not converge0.410.55 (0.16 to 1.91)
 High4.060.16 (−0.24 to 0.57)0.841.34 (0.17 to 10.30)1.98Model did not converge2.702.91 (0.37 to 22.86)
Mineral dust
 None3.700.16 (0.05 to 0.27)0.5912.600.11 (0.04 to 0.19)0.711
 Low1.83*0.34§ (0.17 to 0.52)1.161.71 (0.52 to 5.62)2.710.31 (0.04 to 0.57)0.380.51 (0.07 to 3.86)
 High5.440.17 (−0.12 to 0.46)0.390.66 (0.08 to 5.31)0.040.71 (0.03 to 1.39)2.272.86 (0.38 to 21.55)
Gas and fumes
 None2.390.14 (0.03 to 0.25)0.7512.540.13 (0.04 to 0.22)0.721
 Low3.240.26 (0.10 to 0.43)0.610.83 (0.25 to 2.77)2.410.19 (0.04 to 0.33)0.711.00 (0.35 to 2.83)
 High1.960.16 (−0.04 to 0.37)0.620.44 (0.05 to 3.50)2.91−0.27 (−0.70 to 0.16)0.00
Any exposure
 None2.500.13 (0.02 to 0.25)0.6012.110.14 (0.05 to 0.24)0.781
 Low2.980.31 (0.13 to 0.49)0.741.26 (0.36 to 4.43)2.250.12 (−0.01 to 0.26)0.500.66 (0.21 to 2.03)
 High
2.38
0.15 (−0.01 to 0.32)
0.77
1.01 (0.25 to 4.09)
1.02
0.36 (−0.02 to 0.74)
0.99
1.13 (0.15 to 8.78)

*Adjusted for geographic area, smoking status, time of follow-up.

Adjusted for age, smoking status, number of cigarettes in ECRHS-II for smokers, and length of follow-up among those with FEV1/FVC > 0.8 at baseline.

p < 0.05.

§p for interaction < 0.05 (i.e., the net change in this group is different than among the nonexposed).

Definition of abbreviation: CI = confidence interval.

. Only females ever highly exposed to biological dust (n = 69) and to mineral dust (n = 78) showed an increase in the incidence of airway obstruction (relative risk [RR], 2.91 and 2.86, respectively) compared with nonexposed females, although this change was not statistically significant (at p > 0.1).

The prevalence of chronic phlegm at baseline increased with increasing exposures to mineral dust and gas and fumes in both sexes (Table 5)

TABLE 5. Absolute net change* and incidence of chronic phlegm according to occupational exposures by sex



Males

Females
Incidence
Incidence

Baseline
 Prevalence (%)
Absolute net change/yr
 (95% CI)
%
RR (95% CI)
Baseline
 Prevalence (%)
Absolute net change/yr
 (95% CI)
%
RR (95% CI)
Biological dust
 None4.93−0.04 (−0.15 to 0.07)4.3313.850.05 (−0.05 to 0.16)3.881
 Low5.11−0.01 (−0.24 to 0.22)5.341.19 (0.82 to 1.73)5.64−0.16 (−0.34 to 0.02)3.820.96 (0.67 to 1.36)
 High4.42−0.15 (−0.52 to 0.23)4.741.00 (0.52 to 1.92)1.610.11 (−0.44 to 0.66)3.570.92 (0.24 to 3.55)
Mineral dust
 None4.10−0.05 (−0.16 to 0.06)3.4114.120.00 (−0.10 to 0.09)3.631
 Low6.170.00 (−0.25 to 0.24)6.551.78 (1.25 to 2.54)5.790.16 (−0.18 to 0.49)6.041.49 (0.98 to 2.28)
 High6.300.03 (−0.30 to 3.70)7.461.94 (1.29 to 2.91)9.08−0.80§ (−1.53 to −0.06)1.590.46 (0.07 to 3.14)
Gas and fumes
 None4.48−0.04 (−0.17 to 0.08)3.5214.540.06 (−0.04 to 0.16)3.771
 Low5.13−0.08 (−0.25 to 0.09)5.491.49 (1.05 to 2.10)6.59−0.17 (−0.35 to 0.01)3.991.00 (0.71 to 1.43)
 High5.750.07 (−0.23 to 0.36)6.061.53 (0.99 to 2.36)6.92−0.13 (−0.80 to 0.54)4.881.22 (0.46 to 3.22)
Any exposure
None4.51−0.06 (−0.19 to 0.07)3.3914.640.06 (−0.05 to 0.17)3.881
Low4.48−0.01 (−0.19 to 0.17)4.871.44 (0.98 to 2.10)6.06−0.12 (−0.28 to 0.04)3.820.96 (0.68 to 1.36)
High
5.88
−0.01 (−0.23 to 0.22)
6.50
1.71 (1.18 to 2.49)
6.38
−0.25 (−0.68 to 0.17)
3.82
0.96 (0.43 to 2.14)

*Adjusted for geographic area, smoking status, exact time of follow-up in years.

Adjusted for age, smoking status, number of cigarettes in ECRHS-II for smokers, and length of follow-up among those without any phlegm at baseline.

p < 0.05.

§p for interaction < 0.05 (i.e., the net change in this group is different than among the nonexposed).

For definition of abbreviation, see Table 4.

. In contrast, the change in prevalence from baseline to the end of the study was not statistically significantly different for any of the three categories of exposure. There was, however, a tendency toward a decrease in the prevalence among women ever exposed to low levels of biological dust and gas and fumes and in those exposed to high levels of mineral dust during follow-up, in contrast to a small increase among those never exposed. This paradoxical change contrasts with an increase of the incidence of chronic phlegm during the follow-up in men, particularly in those ever highly exposed to mineral dust and/or gas and fumes, which was significant in comparison to those never exposed. This increase was homogeneous in all areas except Belgium, Germany, and Switzerland, where RRs were below 1, although the test for heterogeneity was not statistically significant (p > 0.1). Stratification by smoking did not modify these results, which exhibited a similar strength of association among never-smokers and current smokers. Analysis by specific occupations showed a significant relative increase in the incidence of chronic phlegm in men working in the following categories: construction (RR adjusted for smoking and area, 1.64; 1.00–2.80), metal (RR, 2.02; 1.29–3.19), electrical (RR, 2.47; 1.46–4.18), and not classifiable (RR, 1.92; 1.25–2.87) in comparison to men consistently employed in white-collar professions. No significant increase in incidence by occupation was observed among females. The same results were obtained for chronic bronchitis (results available on request).

During the follow-up of ECRHS, no greater decline of FEV1 was observed among individuals occupationally exposed to dusts, gases, or fumes (except for female ex-smokers) than among those not exposed, nor was there any increase of airway obstruction related to occupational exposures. The incidence of chronic phlegm showed a relative increase among men with exposures to mineral dust and gas and fumes.

A major limitation in occupational follow-up studies in the general population is the bias related to selection of the subjects (19). Three types of selection events might contribute: (1) selection of subjects at baseline, before the study started (i.e., most susceptible individuals avoid high exposures); (2) selection of subjects during follow-up (e.g., subjects with symptoms at baseline reduce their exposure during follow-up to a higher degree than subjects without symptoms or impaired lung function [20]); and (3) loss to follow-up (subjects who are more susceptible tend to participate more during the follow-up). Selection at baseline might be avoided if cohort studies started early in life, such as in school-age subjects. Nevertheless, this problem has a smaller impact when outcomes of interest are changes in symptoms, lung function, or incidence, such as in the present study, than it does in cross-sectional studies. Changes in exposure during follow-up in relation to symptoms might bias our analysis because subjects in our cohort with chronic cough and phlegm at baseline were somewhat more likely to decrease their exposure to mineral dust, biological dust, and gases and fumes during follow-up in comparison with subjects without symptoms at baseline (data not shown). However, exclusion of subjects with chronic bronchitis or airway obstruction at baseline did not result in any change in the lack of association with lung function. In addition, we measured any occupational exposure during the follow-up or cumulative exposure during follow-up and not just the current occupation at the time of the second examination, which also reduces the potential self-selection during the follow-up. Finally, loss to follow-up did not seem to be related to occupational exposures at baseline because there were no differences in occupational exposure between participants and nonparticipants, and adjustment for participation rates at the center level, another technique reducing the potential of such a bias, did not change the coefficients presented (data not shown).

In general, occupational exposures during follow-up were unrelated to lung function impairment (i.e., FEV1 decline or increase of prevalence or incidence of airway obstruction), which is in contrast with previous general population studies (46) but in agreement with the ECRHS-I study (1). Again, an exception should be made for female ex-smokers (25% of the female population). Possible explanations of the overall lack of effect on lung function include the young age of our cohort (mean age at baseline, 32 yr) and its short cumulative exposure period, and improvement of working conditions with aging related to seniority in the job (21, 22). Regarding age, decline of lung function (23), and prevalence of airway obstruction (7), increases are less pronounced during youth and early adulthood. Despite their relatively young age, a significant association with smoking was found among the present men and women (a decline of 4.8 and 5.1 ml/yr, respectively, compared with never-smokers) (16), similar to that found in other studies at the corresponding age ranges (24). An alternative explanation for the lack of significant findings on lung function refers to the misclassification of the occupational exposures. However, this is not coherent with the lack of effect of high-risk exposures, which are the ones assigned with a lower error rate and a higher specificity when using JEMs (12). In addition, the fact that similar results were obtained with estimates of cumulative exposure, with baseline occupation or with specific occupations, suggests that the misclassification bias is unlikely to explain the lack of significant results. A different explanation refers to the self-selection process of individuals, but as discussed above, this potential bias probably does not explain the lack of significant findings with decline of FEV1.

The prevalence of chronic phlegm and chronic bronchitis at baseline increased with occupational exposure, which coincides with the ECRHS-I cross-sectional study (1). This effect could be explained by residual confounding by socioeconomic status (SES), because variables other than occupational exposures related to SES, such as diet, early life exposures, or area of residence, may play a role in the baseline respiratory outcomes (25, 26). The fact that the incidence during the follow-up in men was higher in those exposed to mineral dust or gas and fumes indicates that the potential interference with SES might not only occur at baseline (i.e., that subjects of low SES enter in high-exposure jobs) but also at follow-up (i.e., that workers of low SES either had higher exposures or were more susceptible or more likely to increase their exposure). In fact, in our study, SES based on occupational history was significantly associated with both prevalence of chronic bronchitis at baseline and incidence of chronic bronchitis (data not shown). To try to separate the effect of occupational exposures from the effect of SES, we adjusted for a surrogate of the latter—namely, education (i.e., age of completion of formal studies). After adjusting for education, results of the RR of incidence did not change, rejecting in part the confounding role of social class.

The increase of incidence of chronic phlegm and chronic bronchitis was not accompanied by an increase of prevalence, because remission counterbalances incidence, or due to the perceived or actual fluctuations of the disease. However, the fact that we excluded all individuals who reported any phlegm at baseline, and not only chronic phlegm, increases the sensitivity of our criteria used to define a free-of-disease population to measure incidence and therefore reduces the potential bias for a baseline association. Changes in exposure during follow-up in relation to symptoms might interfere with these findings because subjects in our cohort with chronic cough and phlegm at baseline decreased their exposures to mineral and biological dust during follow-up in a higher proportion than subjects without symptoms (data not shown). However, if present, this effect would underestimate the association, suggesting that the association between occupational exposures and incidence of chronic bronchitis could be true. It is unlikely that this association was explained by residual confounding of asthma (27), because adjustment for baseline asthma or atopy did not substantially modify the results.

In the pooled analysis, similar results were observed in men and women in relation to lung function. However, in the subgroup analysis, female ex-smokers and women exposed to high levels of both biological and mineral dust had a larger decline and an increase in the prevalence of airway obstruction than women never exposed, a pattern not observed among men. What explains this association among female ex-smokers is uncertain, given the lack of consistency with the males and the lack of differences in baseline atopy (i.e., a possible surrogate of susceptibility) by smoking status in women. A possible explanation could be the role of chance, given the multiple comparisons in the stratified analysis and the small number of women included in the analysis (1.5% exposed to biological dust, 1.8% to mineral dust). In relation to chronic phlegm and chronic bronchitis, the exposure-related increase in baseline prevalence was observed in both sexes, whereas the increase in incidence was only observed in men. Again, the low prevalence of high exposure to biological dust mineral dust and gas and fumes (2.5%) in women may explain this lack of effect on incidence among women. Nevertheless, the failure to identify an adverse effect on ventilatory function arose from very carefully obtained objective data, whereas the identification of chronic bronchitis in men arose subjectively from self-reported symptoms alone. It could be that women in a community setting (not women seeking compensation for occupational disease) were more reluctant to admit to regular phlegm than men. Finally, the discrepancies between symptoms and lung function agree with the cross-sectional analysis of ECRHS-I on a lack of association between both chronic bronchitis and chronic phlegm and lung function (28) and suggest that lung function impairment and chronic bronchitis symptoms are two different entities, at least at early ages.

Overall, in the ECRHS follow-up study conducted during the 1990s, in a young general population sample from centers in industrialized countries, occupational exposures were not related to changes in lung function except in the few highly exposed women and ex-smoking women. Improvement of working conditions through job change, seniority within a job, or overall reduction of exposures over time with ageing, together with the expected relatively small decline at these ages might explain the present findings. Similar to the cross-sectional analysis, we found an increased incidence of chronic bronchitis symptoms in relation to occupational exposures, suggesting inflammation and mucus hypersecretion related to airborne pollutants in the current workplace.

Investigators from the Occupational Working Group in ECRHS-II:

Belgium: South Antwerp and Antwerp City (M. Van Sprundel); France: Paris (B. Leynaert); Germany: Munich (K. Radon); Italy: Verona (A. d'Errico, M. Olivieri), Pavia (S. Villani); Spain: Barcelona (J. M. Antó, J. Sunyer, M. Kogevinas, J. P. Zock, R. Garcia-Esteban, Estel Plana), Galdakao (N. Muniozguren, I. Urritia), Oviedo (F. Payo); Sweden: Uppsala (D. Norback), Goteborg (K. Toren, L. Lillienberg), Umea (B. Lundback); Switzerland: Basel (N. Künzli); The Netherlands: Utrecht (H. Kromhout); United Kingdom: Norwich (D. Jarvis), Ipswich (D. Jarvis).

The following organizations funded the local studies in ECRHS-II included in this article:

Albacete: Fondo de Investigaciones Santarias (grant code: 97/0035-01, 99/0034-01, and 99/0034-02), Hospital Universitario de Albacete, Consejeria de Sanidad; Antwerp: FWO (Fund for Scientific Research)–Flanders Belgium (grant code: G.0402.00), University of Antwerp, Flemish Health Ministry; Barcelona: SEPAR, Public Health Service (grant code: R01 HL62633-01), Fondo de Investigaciones Santarias (grant code: 97/0035-01, 99/0034-01, and 99/0034-02) CIRIT (grant code: 1999SGR 00,241) “Instituto de Salud Carlos III” Red de Centros RCESP, C03/09, Red RESPIRA, C03/011 and Red INMA G03/176; Basel: Swiss National Science Foundation, Swiss Federal Office for Education and Science, Swiss National Accident Insurance Fund (SUVA); Bergen: Norwegian Research Council, Norwegian Asthma and Allergy Association (NAAF), GlaxoWellcome AS, Norway Research Fund; Bordeaux: Institut Pneumologique d'Aquitaine; Erfurt: GSF-National Research Centre for Environment and Health, Deutsche Forschungsgemeinschaft (grant code: FR 1526/1-1); Galdakao: Basque Health Department; Goteborg: Swedish Heart Lung Foundation, Swedish Foundation for Health Care Sciences and Allergy Research, Swedish Asthma and Allergy Foundation, Swedish Cancer and Allergy Foundation; Grenoble: Program Hospitalier de Recherche Clinique–DRC de Grenoble 2000 no. 2610, Ministry of Health, Direction de la Recherche Clinique, Ministere de l'Emploi et de la Solidarite, Direction Generale de la Sante, CHU de Grenoble, Comite des Maladies Respiratoires de l'Isere; Hamburg: GSF–National Research Centre for Environment and Health, Deutsche Forschungsgemeinschaft (grant code: MA 711/4-1); Ipswich and Norwich: National Asthma Campaign (UK); Huelva: Fondo de Investigaciones Santarias (grant code: 97/0035-01, 99/0034-01, and 99/0034-02); Montpellier: Program Hospitalier de Recherche Clinique–DRC de Grenoble 2000 no. 2610, Ministry of Health, Direction de la Recherche Clinique, CHU de Grenoble, Ministere de l'Emploi et de la Solidarite, Direction Generale de la Sante, Aventis (France), Direction Régionale des Affaires Sanitaires et Sociales Languedoc–Roussillon; Oviedo: Fondo de Investigaciones Santarias (grant code: 97/0035-01, 99/0034-01, and 99/0034-02); Paris: Ministere de l'Emploi et de la Solidarite, Direction Generale de la Sante, UCB-Pharma (France), Aventis (France), Glaxo France, Program Hospitalier de Recherche Clinique–DRC de Grenoble 2000 no. 2610, Ministry of Health, Direction de la Recherche Clinique, CHU de Grenoble; Pavia: GlaxoSmithKline Italy, Italian Ministry of University and Scientific and Technological Research (MURST), local university funding for research 1998 and 1999 (Pavia, Italy); Portland: American Lung Association of Oregon, Northwest Health Foundation, Collins Foundation, Merck Pharmaceutical; Reykjavik: Icelandic Research Council, Icelandic University Hospital Fund; Tartu: Estonian Science Foundation; Turin: ASL 4 Regione Piemonte (Italy), AO CTO/ICORMA Regione Piemonte (Italy), Ministero dell'Università e della Ricerca Scientifica (Italy), GlaxoWellcome spa (Verona, Italy); Umeå: Swedish Heart Lung Foundation, Swedish Foundation for Health Care Sciences and Allergy Research, Swedish Asthma and Allergy Foundation, Swedish Cancer and Allergy Foundation; Uppsala: Swedish Heart Lung Foundation, Swedish Foundation for Health Care Sciences and Allergy Research, Swedish Asthma and Allergy Foundation, Swedish Cancer and Allergy Foundation; Verona: University of Verona, Italian Ministry of University and Scientific and Technological Research (MURST), GlaxoSmithKline Italy.

1. Zock JP, Sunyer J, Kogevinas M, Kromhout H, Burney P, Antó JM, and the ECRHS Study Group. Occupation, chronic bronchitis, and lung function in young adults. Am J Respir Crit Care Med 2001;163:1572–1577.
2. Melbostad E, Eduard W, Magnus P. Chronic bronchitis in farmers. Scand J Work Environ Health 1997;23:271–280.
3. Heederik D, Kromhout H, Burrema J. Biersteker k, Kromhout D. Occupational exposures and 25-year incidence rate of non-specific lung disease: the Zutphen study. Int J Epidemiol 1990;19:945–952.
4. Korn RJ, Dockery DW, Speizer FE, Ware JH, Ferris BG. Occupational exposures and chronic respiratory symptoms: a population-based study. Am Rev Respir Dis 1987;136:298–304.
5. Kryzanowski M, Kauffman F. The relation of respiratory symptoms and ventilatory function to moderate occupational exposure in the general population. Int J Epidemiol 1988;17:397–406.
6. Vieggi G, Prediletto R, Paoletti P, Carrozzi L, di Pede F, Vellutini M, di Pede C, Giuntini C, Lebowitz MK. Respiratory effects of occupational exposure in a general population sample in north Italy. Am Rev Respir Dis 1991;143:510–515.
7. Bakke PS, Bste V, Hanoa R, Gulsvik A. Prevalence of obstructive lung disease in a general population: relation to occupational title and exposure to some airborne agents. Thorax 1991;46:863–870.
8. Sunyer J, Kogevinas M, Kromhout H, Antó JM, Roca J, Tobias A, Vermeulen R, Payo F, Maldonado JA, Martinez-Moratalla J, et al. Pulmonary ventilatory defects and occupational exposures in a population-based study in Spain. Am J Respir Crit Care Med 1998;157:512–517.
9. Fishwick D, Bradshaw LM, D'Souza W, Town I, Armstrong R, Pearce N, Crane J. Chronic bronchitis, shortness of breath, and airway obstruction by occupation in New Zealand. Am J Respir Crit Care Med 1997;156:1440–1446.
10. Suadicani P, Hein HO, Meyer HW, Gyntelberg F. Exposure to cold and draught, alcohol consumption, and the NS-phenotype are associated with chronic bronchitis: an epidemiological investigation of 3387 men aged 53–75 years: the Copenhagen Male Study. Occup Environ Med 2001;58:160–164.
11. Balmes J, Becklake M, Blanc P, Henneberger P, Kreiss K, Mapp C, Milton D, Schwartz D, Toren K, Viegi G. American Thoracic Society statement: occupational contribution to the burden of airway disease. Am J Respir Crit Care Med 2003;167:787–797.
12. Kromhout H, Heederik D, Dalderup LM, Kromhout D. Performance of two general job-exposure matrices in a study of lung morbidity in the Zuphten cohort. Am J Epidemiol 1992;136:698–711.
13. Kauppinen T, Mutanen PO, Seitsamo JT. Magnitude of misclassification bias when using a job-exposure matrix. Scan J Work Environ Health 1992;18:105–122.
14. de Vocht F, Zock JP, Kromhout H, Sunyer J, Anto JM, Burney P, Kogevinas M. Comparison of self-reported occupational exposure with a job exposure matrix in an international community-based study on asthma. Am J Ind Med 2005;47:434–442.
15. Burney PG, Luczynska C, Chinn S, Jarvis D. The European Community Respiratory Health Survey. Eur Respir J 1994;7:954–960.
16. Chinn S, Jarvis D, Melotti R, Luczinska C, Ackerman-Liebrich U, Antó JM, Cerveri I, De Marco R, Gislason T, Heinrich J, et al. Smoking cessation: maximum benefit for lung function requires control of weight gain. Lancet 2005;365:1629–1635.
17. American Thoracic Society. Standardization of spirometry: 1994 update. Am J Respir Crit Care Med 1995;152:1107–1136.
18. International Labor Office. International Standard Classification of Occupations: ISCO 88. Geneva, Switzerland: International Labor Office; 1991.
19. Radon K, Goldberg M, Becklake M. Healthy worker effect in cohort studies on chronic bronchitis. Scand J Work Environ Health 2002;28:328–332.
20. Kristman V, Manno M, Cote P. Loss to follow-up in cohort studies: how much is too much? Eur J Epidemiol 2004;19:751–760.
21. Vermeulen R, de Hartog J, Swuste P, Kromhout H. Trends in exposure to inhalable particulate and dermal contamination in the rubber manufacturing industry: effectiveness of control measures implemented over a nine-year period. Ann Occup Hyg 2000;44(5):343–354.
22. Kromhout K, Vermeulen R. Long-term trends in occupational exposure: Are they real? What causes them? What shall we do with them? Ann Occup Hyg 2000;44:325–327.
23. Robbins DR, Enright PL, Sherrill DL. Lung function development in young adults: is there a plateau phase? Eur Respir J 1995;8:768–772.
24. James AL, Palmer LJ, Kicic E, Maxwell PS, Lagan SE, Ryan GF, Musk AW. Decline in lung function in the Busselton Health Study: the effects of asthma and cigarette smoking. Am J Respir Crit Care Med 2005;171:109–114.
25. Shohaimi S, Welch A, Bingham S, Luben R, Day N, Wareham N, Khaw KT. Area deprivation predicts lung function independently of education and social class. Eur Respir J 2004;24:157–161.
26. Prescott E, Lange P, Vestbo J. Socioeconomic status, lung function and admission to hospital for COPD: results from the Copenhagen City Heart Study. Eur Respir J 1999;13:1109–1114.
27. Hendrick DJ. Occupation and chronic obstructive pulmonary disease. Thorax 1996;51:947–955.
28. Sunyer J, Basagaña X, Roca J, Urrutia I, Jaen A, Antó JM, Burney P. Relations between respiratory symptoms and spirometric values in young adults: the European community respiratory health study. Respir Med 2004;98:1025–1033.
Correspondence and requests for reprints should be addressed to Dr. Jordi Sunyer, Institut Municipal d'Investigació Mèdica–IMIM, Dr. Aiguader 80, 08003 Barcelona, Spain. E-mail:

Related

No related items
American Journal of Respiratory and Critical Care Medicine
172
9

Click to see any corrections or updates and to confirm this is the authentic version of record