ASTHMA AND OCCUPATION: A POPULATION-BASED STUDY AMONG YOUNG

DANISH ADULTS

Jesper Rasmussen, MD1, Jesper Bælum, MD 2, Lars R Skadhauge, MD 3, Maria C Mirabelli,

PhD 4, 4a, 4b, Jan-Paul Zock, PhD 4, 4a, 4b, Søren Dahl, MD 3, Øyvind Omland, MD 5, David

Sherson, MD 6, Hans C. Siersted, MD 7, Gert Thomsen, MD 8, Torben Sigsgaard, MD 9

1Department of Occupational and Environmental Medicine, Hospital of Southern Jutland,

Haderslev, Denmark; 2 Department of Occupational and Environmental Medicine, Odense

University Hospital, Odense, Denmark; 3 Department of Occupational and Environmental

Medicine, Hospital of Southern Jutland, Haderslev, Denmark; 4 Centre for Research in

Environmental Epidemiology (CREAL), Barcelona, Spain; 4a Municipal Institute of Medical

Research (IMIM-Hospital del Mar), Barcelona, Spain; 4b CIBER Epidemiología y Salud

Pública (CIBERESP), Spain; 5 Department of Occupational Medicine, Hospital of Aalborg,

Denmark; 6 Department of Occupational and Environmental Medicine, Vejle Hospital;

Denmark; 7 Department of Respiratory Medicine, Odense University Hospital, Denmark; 8

Department of Occupational Medicine, Hospital of South West Jutland, Denmark;

9Department of Environmental and Occupational Medicine, Aarhus University, Aarhus,

Denmark.

Corresponding author and requests for reprints

Jesper Rasmussen, Department of Occupational and Environmental Medicine, Hospital of

Southern Jutland, Simmerstedvej 1-3, DK-6100 Haderslev, Denmark, Telephone: 0045

73530222, Fax: 0045 74520373, E-mail: jesper.rasmussen@shs.regionsyddanmark.dk

Running Title: Asthma and Occupation Word count: 4,237

Number of tables: 7 tables Number of figures: 2 figures

ABSTRACT

Objectives: To describe occupations and occupational exposures associated with the

prevalence of asthma among men and women in Denmark.

Methods: Information about occupation and asthma symptoms was collected from 7,271 men

and women, aged 20 to 44 years, who participated in a population-based, cross-sectional

survey of respiratory health in Denmark. Three respiratory health outcome definitions were

used: current wheeze, current asthma and doctor-diagnosed adult-onset asthma. The current

or last held job was coded according to ISCO-88 and linked to an asthma-specific job

exposure matrix. Associations between occupational exposures and the respiratory health

outcome definitions were evaluated using log-binomial regression analysis, stratified by

gender and adjusted for county, age and smoking status.

Results: Occupation as a cleaner or caretaker was associated with an increased prevalence of

current asthma compared to non-manual workers. These associations were most pronounced

among women (prevalence ratio (PR) 2.17; 95% confidence interval (CI) 1.47-3.21 and PR

1.50; 95% CI: 1.02-2.18) and were not modified by nasal allergy. Among men, exposure to

high-molecular-weight agents was associated with an increased prevalence of current asthma.

A greater prevalence of current asthma was found among women exposed to industrial

cleaning agents than in respondents unlikely to be exposed to any asthmagenic compounds.

Conclusions: Our data indicate a consistently increased asthma risk in female cleaners and

caretakers and in men with workplace exposure to high-molecular-weight agents.

Key words: work-related asthma; job exposure matrix; prevalence

2

INTRODUCTION

Asthma is the most common occupational respiratory disorder in Western industrialised

populations1, and 15-20% of the population burden of asthma may be attributable to

occupational exposures.2;3 The list of occupations and exposures reported to be associated

with asthma is expanding, and over 250 workplace agents have been identified as specific

causes of occupational asthma.4;5

Population-based studies have reported an increased risk of asthma among

agricultural workers, bakers and other food processors, cleaners, chemical processors,

farmers, laboratory technicians, nurses, painters, personal care workers, plastic and rubber

workers, spray painters, waiters and welders.2;6-9 Work as a domestic or professional cleaner

and frequent use of cleaning sprays have been reported to increase the risk of asthma.13-18

Healthcare workers and hospital technicians also experience increased risk of asthma,

possibly due to the use of specific products at work,10;11 and an increased risk of new-onset

asthma has been reported among nursing professionals with occupational exposure to

cleaning products and disinfectants.12

Estimates in population-based studies mainly report work-related asthma without

further characterisation as occupational asthma. Because work-related asthma includes

asthma caused by the workplace (i.e., occupational asthma) as well as pre-existing asthma

that is exacerbated by occupational exposures (i.e., work-exacerbated asthma), work-

exacerbated asthma is also an important area of research interest.19 Epidemiological research

indicates that work-exacerbated asthma is common.20 Particularly high percentages have

been reported among garment workers, janitors, restaurant workers and security guards,

possibly due to irritant exposure in the workplace.21 Participants exposed to either high-

3

molecular-weight (HMW) or low-molecular-weight (LMW) agents, high peak irritant

inhalation accidents, biological enzymes, flour dust, latex, mite and other animal-derived

proteins, textiles, industrial cleaning agents and highly reactive chemicals have shown

increased asthma risk in epidemiologic studies conducted using an asthma- specific job

exposure matrix (JEM).2;9;22

Respiratory symptoms among men and women have been attributed not only to

occupation, but also to exposures that are likely to be relevant to the exposures encountered

in Denmark. A study of Dutch farmers and agricultural industry workers reported an

association between endotoxin exposure and respiratory effects, (such as wheezing and

wheezing with shortness of breath), among those with a farm childhood.24 This finding

suggests that growing up on a farm may increase, rather than decrease, the risk of adult

asthma symptoms when accounting for farming-related endotoxin exposure. Also, a higher

risk of occupational asthma has been reported among individuals having a parental history of

asthma.2 A positive association between rhinitis and asthma has been reported in both adult

men and women, but without accounting for occupational exposure.25 Finally, a study

reported that individuals who grew up in large families were also likely to experience adult

asthma, although occupational exposures were not accounted for.26 Given the potential for

the exposure-outcome relationships to vary, not only between men and women, but also

between individuals having these a priori suspected or known risk factors, it is important to

include these risk factors in the study of associations between occupational exposures and

asthma.

In Denmark, the prevalence of self-reported asthma among adults has increased from

2.9% in 1987 to 6.4% in 2005.27 In the Danish Risk Factor for Adult Asthma study (RAV),

4

we found an asthma prevalence of 7.0% in 2004. 28 The aetiology of this increase is unknown

and may partially be explained by occupational exposure, although associations between

occupational exposure and asthma have rarely been studied in Denmark.

In 2005, Denmark was the European country with the highest percentage of female

workers, with women accounting for 47.5% of the labour force.29 Few studies have

investigated gender differences in the associations between occupational exposure and

asthma. One study reported that women were more exposed to asthmagenic agents than men

and that associations between asthmagenic agents and severe asthma were stronger among

women than men.9 Another study reported an increased incidence of respiratory symptoms

among only female woodworkers exposed to dry wood.23 In a study of trends in the Danish

work environment in 1990-2000, the prevalence of cleaning, clerk, military and textile jobs

decreased, while academic, computer, and managerial jobs increased. The number of part-

time cleaners has declined considerably, resulting in more hours of cleaning work per

individual. In addition, skin contact with cleaning agents among nurses has increased.30 With

the unusually high percentage of women in the Danish workforce overall, and in several

occupations having a high risk of asthma in particular, the prevalence of work-related asthma

may plausibly be expected to be higher than those reported in other European countries.

The purpose of this study was to investigate how gender influences the relationship

between occupation and asthma in a large population-based cohort of adults in Denmark. In

addition, we assessed the effects of growing up in a rural area or on a farm, parental asthma

history, nasal allergy and the number of siblings in modifying the relationship between

occupational exposures and asthma.

5

METHODS

STUDY DESIGN AND POPULATION

The RAV study used a cross-sectional study design following the protocol used for the

European Community Respiratory Health Survey (ECRHS).31 Between December 2002 and

January 2004, an extended ECRHS screening questionnaire was mailed to a gender- and age-

stratified random sample of 10,000 adults aged 20 to 44 years. Participants were drawn from

a general population-based sample of residents in five counties in western Denmark, a

geographic region having a total population of 1,800,000. The study was approved by local

ethical committees.

QUESTIONNAIRE

The questionnaire included items about asthma symptoms, asthma and suspected asthma risk

factors, including the current or last held job, smoking, parental asthma, growing up in a rural

environment or on a farm, the number of siblings26 and nasal allergy.

OCCUPATIONAL EXPOSURE ASSESSMENT

Employment information was obtained using the following question that defined the current

or last held job: ‘What is your current or last held job?’ The current or last held job was

coded according to the Danish version of the International Labour Office system, ISCO-88

(International Standard Classification of Occupations).32 Two estimates of exposure were

used. First, jobs were grouped into 26 categories of potentially high-risk jobs and a reference

group consisting of participants who worked in professional, clerical, or administrative jobs,

as previously defined by the ECRHS.2 Second, an asthma-specific JEM33 was used to

estimate exposure to high- and low-risk asthmagenic agents. For each ISCO-88 job code, the

asthma-specific JEM classified the job as exposed to or not exposed to 18 high-risk

asthmagenic agents or settings, which we further subdivided into four groups: HMW agents,

6

LMW agents, mixed environments or high irritant peaks. The four low-risk asthmagenic

agents or settings were engine exhaust, environmental tobacco smoke, possible irritants and

low- risk antigens. Furthermore, we grouped the JEM exposure categories into two groups:

exposure to any high- risk agents or any low- risk agents. The low-risk group was not

exposed to any high-risk agent, but in each of the two separate groups, participants may have

had more than one specific high-risk or low-risk exposure. Overall, 754 participants could

not be classified into any exposure category using the asthma-specific JEM. To avoid

excluding these participants, we considered this group more likely to be unexposed than

exposed, so in our analysis, they are included in the reference group. The expert step,

recommended when using the asthma-specific JEM, was not applied in this study.

ASTHMA DEFINITION

We used three definitions of asthma based on diagnosis or symptoms that are highly

suggestive of asthma, ranging from a relatively high sensitivity to a relatively high

specificity. Current wheeze was defined by an affirmative response to the question ‘Have

you had wheezing or whistling in the chest at any time during the last 12 months when you

did not have a cold?’ Current asthma was defined by an affirmative response to any of the

following questions: ‘Have you had an attack of asthma in the last 12 months?’, ‘Have you

been woken by an attack of shortness of breath at any time in the last 12 months?’ and ‘Are

you currently taking any medicine for asthma?’ The third group, doctor-diagnosed adult-

onset asthma, comprised participants who had their first asthma attack after the age of 16

and answered affirmatively to both of the following questions, ‘Have you ever had asthma?’

and ‘Was it confirmed by a doctor?’ These definitions are consistent with those used in

previous ECRHS studies.2;13

7

POTENTIAL EFFECT MODIFIERS

Gender is considered the main potential effect modifier. Smoking status was classified as

follows: non-smokers answered ‘Have you ever smoked for as long as a year?’ in the

negative. Current smokers were defined by an affirmative response to both of the following

questions: ‘Have you ever smoked for as long as a year?’ and ‘Do you smoke now?’ Ex-

smokers were defined by an affirmative response to the question ‘Have you ever smoked for

as long as a year?’ and a negative response to the question ‘Do you smoke now?’ Parental

asthma was defined by an affirmative response to either of the questions: ‘Did your mother

ever have asthma?’ or ‘Did your father ever have asthma?’ Growing up in a rural area was

defined by an affirmative response to the question ‘Did you grow up in a rural area?’

Growing up on a farm was defined by an additional affirmative response to the question ‘Did

you live on a farm?’ Growing up in a rural area and on a farm with animals was defined by

an affirmative response to all three questions: ‘Did you grow up in a rural area?’, ‘Did you

live on a farm?’ and ‘Were there animals on the farm?’ The number of siblings was

determined by asking the question ‘How many siblings do you have or did you have?’ Nasal

allergy was defined by an affirmative response to the question ‘Do you have nasal allergies

(e.g. hay fever)?’

STATISTICAL ANALYSIS

Differences in the distributions of gender and other characteristics between the study

population and the excluded population were assessed using the chi-squared or Student’s t-

test. We used log-binomial regression models to assess associations between 26 separate

categories of potential high-risk jobs, 22 separate JEM high-risk asthmagenic agent

exposures or 5 separate JEM low-risk asthmagenic agent exposures and current wheeze,

current asthma and doctor-diagnosed adult-onset asthma. Associations for each of the

exposure-outcome relationships stratified by gender were estimated using separate models.

8

Referent categories for the job and JEM exposure categories were the job reference group

and the unexposed JEM group, respectively. Confounders were selected a priori. Therefore,

all models were adjusted for age (in 5-year strata), smoking status (non-smoker, ex-smoker,

current smoker), and county (five counties). All associations are presented as prevalence

ratios (PRs)34 with 95% confidence intervals (CIs).

We evaluated potential modification of effects using a stratified analysis to assess the

associations of any JEM high-risk asthmagenic or any JEM low-risk asthmagenic exposure

with current asthma and compared these associations with those of unexposed individuals;

these associations were conducted among the entire study population and for the strata of

gender, smoking status, nasal allergy, parental asthma, growing up in a rural area or on a

farm and the number of siblings. Each separate model was adjusted for age, gender and

county (in analyses other than those grouped by gender), and for smoking status (in analyses

other than those grouped by smoking status). We assessed differences in stratum-specific

associations using a Wald test for the statistical significance of an interaction term introduced

into a Poisson regression model with robust error estimation. We chose the Poisson

regression model in this portion of the analyses to resolve problems of non-convergence with

the log-binomial regression models. Analyses using the JEM were repeated after the

exclusion of jobs with uncertain exposure estimates (nexcluded=1,771). For all tests, p≤0.05

was considered statistically significant. All analyses were performed using Stata version SE

10.1 (Stata Corporation, College Station, Texas, USA).

9

RESULTS

A total of 7,271 participants (response rate, 73%) returned the questionnaire. Overall, the

participation rates were higher among women (p<0.001) and among the older age groups

(p<0.001). We excluded 642 respondents who were classified as occupationally inactive,

including homemakers (n=27), unemployed (n=43), individuals not working because of poor

health (n=19), full-time students (n=302), retirees (n=58) and respondents with missing

occupational history information (n=193). We also excluded 302 respondents with missing

information regarding asthma outcomes or any of the potential confounders and effect

modifiers considered in our analysis. There were differences between those who were

excluded because of occupational inactivity and missing data (n=944) and those participants

eligible for analysis in the study. The excluded population was younger (p<0.0001) and had a

higher percentage of females (p=0.0006) than males. The prevalences of current wheeze and

current asthma were higher among the excluded persons than in the study population

(p<0.0001 for both), but there was no difference in the prevalence of doctor-diagnosed, adult-

onset asthma between the two populations. There were fewer non-smokers (p=0.0390) and a

higher percentage of ex-smokers (p=0.0263) in the excluded population.

After these exclusions, 6,327 participants remained eligible for inclusion in our

occupational analysis. Table 1 summarises the descriptive statistics for the participants

stratified by sex. The participants averaged 33 years old, and the response rates were higher

in the older age groups (30-44 years) than in the younger age groups (20-29 years). The

prevalences of current asthma and doctor-diagnosed, adult-onset asthma were higher in

women than in men (p=0.043 and p=0.0001, respectively), and wheeze was more common in

men, although not significantly so (p=0.094). There were no differences by county.

10

Table 2 shows the gender-specific associations between occupation and current

asthma. These analyses generated significantly increased prevalences of current asthma for

female cleaners and caretakers. The increased PR of current asthma for cleaners and

caretakers was similar in those with and without nasal allergy (data not shown). A small but

borderline-significant increase in the PR of current asthma was observed for female medical

and pharmacy workers other than nurses. Fully adjusted PRs for occupational categories of

potential interest such as hairdressers, welders, plastics and rubber workers, and paper

workers could not be generated because of the small number of current asthma cases in each

occupational group. The results of all analyses for current wheeze and doctor-diagnosed

adult-onset asthma are presented in Supplementary Tables 1 and 2.

Table 3 shows the associations between occupational exposures categorised using the

asthma-specific JEM and current asthma. In these analyses, exposure to HMW agents was

associated with current asthma in men, whereas the PR for women was borderline-

significant. Among men, the elevated PR was driven by elevated prevalences of several

specific HMW agents, namely, flour, plants and mites (data not shown). Exposure to LMW

agents was less clearly associated with current asthma. However, the nested category of

cleaning agents was related to current asthma in women but not in men. A significantly

increased current asthma PR was associated with exposure to possible irritants only among

women. The results of all analyses for current wheeze and doctor-diagnosed adult-onset

asthma are presented in Supplementary Tables 3 and 4.

The main jobs identified with possible irritant exposure in women were cleaners and

caretakers (71%), food and tobacco processing (9%), and construction and mining (16%).

Figures 1 and 2 show the PRs for current asthma due to any JEM high- or low-risk asthma

11

agent exposure in separate groups as defined by potential effect modifiers, including gender.

Figure 1 shows that the PRs of current asthma related to any high-risk asthma agent exposure

were significantly lower for non-smokers (p=0.048 for interaction) compared to ex-smokers

or current smokers. No significant differences were observed for gender, nasal allergy,

parental asthma or having grown up in a rural or farm environments. While dichotomising

the number of siblings (0-2 siblings vs. 3 or more), a significant interaction (p=0.04) was

found between having three or more siblings and exposure to high-risk asthma agents

regarding the presence or absence of current asthma.

Figure 2 shows that the PRs of current asthma that were related to low-risk asthma

agent exposure were significantly higher for those growing up on a farm and those growing

up on a farm with animals (p=0.027 and p=0.041, respectively, for interaction) compared

with those who did not grow up on a farm or in a rural area. No significant differences were

observed for gender, nasal allergy, smoking status or parental asthma. While dichotomising

the number of siblings (0-2 siblings vs. 3 or more), a significant interaction (p=0.0023) was

found between having three or more siblings and exposure to low-risk asthma agents

regarding the presence or absence of current asthma.

After excluding jobs with uncertain JEM (n=1,771) (data not shown) the PRs among

men were slightly increased for current asthma and high- and low-risk asthma agents, but the

PR for HMW was no longer significant (PR=1.67, 95% CI: 0.95-2.92). Among women,

slightly decreased PRs for current asthma and HMW agents, mixed environments and low-

risk asthma agents were observed, and the PR for current asthma and LMW agents was

modestly elevated. Cleaning agent exposure remained significantly elevated (PR=1.38, 95%

12

CI: 1.02-1.86). Possible irritant exposure was no longer significantly associated with current

asthma (PR=0.87, 95% CI: 0.23-3.28).

After exclusion of jobs with uncertain JEM, the stratified analysis no longer showed

significant interactions for non-smokers between current asthma and high-risk asthma agents

or between growing up on a farm with or without animals and current asthma and low-risk

asthma agents. When dichotomising the number of siblings (0-2 siblings vs. 3 or more), a

significant interaction remained between having three or more siblings and exposure to both

high- and low-risk asthma agents in relation to current asthma (p=0.0137 and p=0.0278,

respectively).

13

DISCUSSION

This population-based study among young adults in five Danish counties showed a

consistently increased risk of current asthma among female cleaners and caretakers and

among males exposed to high-molecular-weight agents. Workplace exposure to irritants was

associated with an increased risk of current asthma in women. Those who grew up on a farm

or had three or more siblings showed increased current asthma risk associated with

occupational exposures. The increased PR of current asthma for cleaners and caretakers was

similar in those with and without nasal allergy.

A high risk of occupational asthma in cleaners has been reported in other

populations.2;6;8 Among cleaners, chemicals such as bleach, ammonia-containing compounds

and disinfectants including glutaraldehyde and formaldehyde have been identified as specific

causes of respiratory disorders, including asthma. Increased risk of asthma has also been

related to specific job tasks, such as cleaning floors, windows, mirrors, ovens and dishes.35

Some studies have identified specific professional cleaning products associated with asthma,

including bleach and sprays.14;18 Still, it remains unclear how much of cleaning-related

asthma is related to specific sensitisation and how much is due to airway inflammation

induced by a mixture of irritants.35

The positive association between HMW agent exposure and the prevalence of asthma

is consistent with results from other population-based studies of asthma prevalence and

incidence conducted using the asthma-specific JEM.2;9 We found that exposure to LMW

agents was less clearly associated with asthma, although the nested category of cleaning

agents was related to asthma in women but not in men. These results are partly consistent

with those of the ECRHS study and the French Pollution Atmosphérique et Affections

14

Respiratoires Chroniques (PAARC) survey, where increased asthma risk was found in

subjects exposed to industrial cleaning agents and reactive chemicals.2;9 Industrial cleaning

agents could be considered as LMW according to the JEM, but some of them are also

irritants.2

Workplace exposure to irritants (not high peak-exposure) showed an increased PR of

asthma in women. Evidence is growing for the importance of repeated moderate irritant

exposures in the development of asthma.21;36 Evidence is also accumulating about the effects

of irritant exposure on the occurrence of asthma in specific occupations, such as cleaning or

the pulp and paper industries.2;37 Recent population-based studies suggest an increased

relative asthma risk in occupations with low-to-moderate respiratory irritant exposure, and

that these exposures are common.37 Additional studies are needed to determine the airway

effects of such exposures.

Women were more exposed to asthmagens in the present survey than men, but less

exposed to other agents. The associations appear to be different between men (a significant

association with HMW in general) and women (a significant association with industrial

cleaning agents). This is probably explained by gender differences in job exposures, job

choice and suscepbtibility.9;23;38;39 However, further research is needed to elucidate whether

these findings reflect differences in gender susceptibility or in gender-specific occupational

exposures.

The observed increase in asthma risk associated with occupational exposures in

participants who grew up on a farm is consistent with a study of Dutch farmers and

agricultural industrial workers, which showed slightly higher associations between endotoxin

15

exposure and respiratory effects, such as wheezing and wheezing with shortness of breath.24

The increased prevalence of asthma associated with occupational exposures in those with

more siblings is partly supported by the literature. In an analysis of ECRHS data, it was

found that participants who grew up in large families experienced more asthma in

adulthood.26 However, in our study we showed a possible protective effect against asthma in

participants with more siblings and a farm childhood, without accounting for occupational

exposures.40 Whether the increased prevalence of asthma among participants growing up on

a farm or with more siblings and occupational exposures reflects different susceptibility,

occupational exposures or job choice in these two groups needs to be studied further.

In contrast to results from previous studies, we were not able to confirm elevated PRs

for established high-risk occupations and exposures, such as farmers, bakery workers,

painters, laboratory technicians, plastic and rubber workers, welders, latex, highly reactive

chemicals and textiles.6-8 This may be explained by a lack of statistical power for some

groups and occupational exposures. Alternatively, a lower-than-expected prevalence of

asthma in these risk occupations may reflect differences in occupational risk factors or job

selection tendencies in Denmark compared with other European countries.29;30 We examined

the effect of the current job or last held job, not the job occupied at the time of the

exacerbation of respiratory symptoms. This might have lead to a healthy-worker effect in our

results that is, the over-representation of participants who are more resistant to occupational

exposures and an underestimation of associations between asthma and occupational

exposures.6;7 Differing exposures and awareness of symptoms in different social classes may

also be a factor leading to over-or under-estimation of possible associations between

symptoms and exposures. We believe that the participants’ general knowledge about

occupational asthma and its causes are limited, and therefore it is unlikely that participants in

16

certain jobs were more likely to report symptoms or become diagnosed with asthma than

those in other jobs. The application of JEM may reduce reporting or recall bias, thus

minimising potential overestimation of the prevalence of asthma in relation to occupational

exposure. However, the JEM has the limitation that exposure variations within job titles are

not taken into account, and it does not identify asthma risks associated with unknown

agents.22;33 Another limitation of our study was that we applied the JEM without the

recommended expert reviewing step. This might have resulted in our insufficiently

incorporating country-specific exposure risks and in reduced specificity in estimates for jobs

with uncertain exposures. In our main results, we saw only minor changes after excluding

jobs with imprecise exposure estimates. In an ECRHS study, the associations did not change

considerably after the expert review step,7 so we believe that our use of the JEM without the

expert reviewing step was likely to have had minimal impact on our findings.

Discrepancies with previous findings could also have arisen from asthma

misclassification, as our diagnoses were based on self-reported respiratory symptoms and/or

medication. This study did not include validation of asthma diagnosis by bronchial hyper-

responsiveness or atopy status using the skin-prick test. This could lead to an under

estimation of the associations between differing occupational exposures and asthma.

Nevertheless, our results were fairly consistent using the three asthma definitions (ranging

from relatively sensitive to relatively specific). The ECRHS definition of asthma symptoms

or medication has been validated against bronchial responsiveness,8 indicating that such

underestimation, if present, would be minimal. We did not validate the reported association

of nasal allergy with atopy status using the skin-prick test, which could lead have led to the

underestimation of the associations of asthma with nasal allergy and occupation.

17

The response rate in this study was fair (73%), but it could have biased the results

away from the null hypothesis. Nevertheless, we believe that potential bias was non-

differential according to the primary aim of the survey, namely, identifying risk factors for

adult asthma in general, and not occupational risk factors in particular. Therefore, it is

unlikely that participants in certain jobs were more likely to participate in the study.

In 2007, 102 cases of occupational asthma were identified in the Danish National

Board of Industrial Injuries. A crude estimation with our data showed a mean asthma

incidence rate of 3.3 per 1000 person-years in 2003.28 Using a conservative population-

attributable risk of 15% of asthma due to occupation, we estimate 1,341 new cases per year

of occupational asthma among the employed population in Denmark, suggesting that

occupational asthma is considerably under-reported in Denmark.

Occupational asthma is the most frequently reported occupational respiratory disorder

in western industrialised populations.1As the true prevalence of occupational asthma is

underestimated, there remains a potential problem for many individuals who are exposed to

known asthmagenic agents in Denmark and other industrialised countries.

Main conclusion: Our data demonstrate a consistent increased prevalence of asthma in

female cleaners and caretakers and in men with workplace exposure to HMW agents.

18

ACKNOWLEDGEMENTS

The authors thank Estel Plana (CREAL, IMIM, Barcelona, Spain) for her counselling on data

analysis. The authors appreciate the RAV cohort members’ contributions to this work.

The study has been founded by Hospital of Southern Jutland, Department of Research,

Sønderborg, Denmark, and Department of Occupational and Environmental Medicine,

Hospital of Southern Jutland, Haderslev, Denmark. Institute of Clinical Research, University

of Southern Denmark, Department of Occupational and Environmental Medicine, Odense

University Hospital, Odense, Denmark. Faculty of Health Sciences, University of Southern

Denmark, Odense, Denmark. Institute of Regional Health Services Research, University of

Southern Denmark, Odense, Denmark. Global Allergy and Asthma European Network (GA2

LEN), Work Package: WP 2.4 Occupation and Adolescents, EU Framework programme for

research contract n0 FOOD-CT-2004-506378, University of Ghent, Belgium

19

REFERENCES

Reference List

1. Nicholson, P. J., P. Cullinan, A. J. Newman Taylor, P. S. Burge, and C. Boyle. 2005.

Evidence based guidelines for the prevention, identification, and management of

occupational asthma. Occup Environ Med 62:290-299.

2. Kogevinas, M., J. P. Zock, D. Jarvis, H. Kromhout, L. Lillienberg, E. Plana, K. Radon,

K. Toren, A. Alliksoo, G. Benke, P. D. Blanc, A. hlman-Hoglund, A. D'Errico, M.

Hery, S. Kennedy, N. Kunzli, B. Leynaert, M. C. Mirabelli, N. Muniozguren, D.

Norback, M. Olivieri, F. Payo, S. Villani, S. M. van, I. Urrutia, G. Wieslander, J.

Sunyer, and J. M. Anto. 2007. Exposure to substances in the workplace and new-onset

asthma: an international prospective population-based study (ECRHS-II). Lancet

370:336-341.

3. Toren, K. and P. D. Blanc. 2009. Asthma caused by occupational exposures is common

- a systematic analysis of estimates of the population-attributable fraction.

BMC.Pulm.Med 9:7.

4. Malo, J. L. and M. Chan-Yeung. 2009. Agents causing occupational asthma. J Allergy

Clin.Immunol. 123:545-550.

5. van Kampen, V., R. Merget, and X. Baur. 2000. Occupational airway sensitizers: an

overview on the respective literature. Am.J Ind.Med 38:164-218.

6. Kogevinas, M., J. M. Anto, J. B. Soriano, A. Tobias, and P. Burney. 1996. The risk of

asthma attributable to occupational exposures. A population-based study in Spain.

Spanish Group of the European Asthma Study. Am.J Respir Crit Care Med 154:137-

143.

20

7. Fishwick, D., N. Pearce, W. D'Souza, S. Lewis, I. Town, R. Armstrong, M. Kogevinas,

and J. Crane. 1997. Occupational asthma in New Zealanders: a population based study.

Occup Environ Med 54:301-306.

8. Kogevinas, M., J. M. Anto, J. Sunyer, A. Tobias, H. Kromhout, and P. Burney. 1999.

Occupational asthma in Europe and other industrialised areas: a population-based

study. European Community Respiratory Health Survey Study Group. Lancet

353:1750-1754.

9. le Moual, N., S. M. Kennedy, and F. Kauffmann. 2004. Occupational exposures and

asthma in 14,000 adults from the general population. Am.J Epidemiol. 160:1108-1116.

10. Mirabelli, M. C., J. P. Zock, E. Plana, J. M. Anto, G. Benke, P. D. Blanc, A. hlman-

Hoglund, D. L. Jarvis, H. Kromhout, L. Lillienberg, D. Norback, M. Olivieri, K.

Radon, J. Sunyer, K. Toren, S. M. van, S. Villani, and M. Kogevinas. 2007.

Occupational risk factors for asthma among nurses and related healthcare professionals

in an international study. Occup Environ Med 64:474-479.

11. Delclos, G. L., D. Gimeno, A. A. Arif, K. D. Burau, A. Carson, C. Lusk, T. Stock, E.

Symanski, L. W. Whitehead, J. P. Zock, F. G. Benavides, and J. M. Anto. 2007.

Occupational risk factors and asthma among health care professionals. Am.J Respir Crit

Care Med 175:667-675.

12. Arif, A. A., G. L. Delclos, and C. Serra. 2009. Occupational Exposures and Asthma

among Nursing Professionals. Occup Environ Medoem.

13. Zock, J. P., E. Plana, D. Jarvis, J. M. Anto, H. Kromhout, S. M. Kennedy, N. Kunzli, S.

Villani, M. Olivieri, K. Toren, K. Radon, J. Sunyer, A. hlman-Hoglund, D. Norback,

21

and M. Kogevinas. 2007. The use of household cleaning sprays and adult asthma: an

international longitudinal study. Am.J.Respir.Crit Care Med 176:735-741.

14. Medina-Ramon, M., J. P. Zock, M. Kogevinas, J. Sunyer, Y. Torralba, A. Borrell, F.

Burgos, and J. M. Anto. 2005. Asthma, chronic bronchitis, and exposure to irritant

agents in occupational domestic cleaning: a nested case-control study. Occup Environ

Med 62:598-606.

15. Medina-Ramon, M., J. P. Zock, M. Kogevinas, J. Sunyer, and J. M. Anto. 2003.

Asthma symptoms in women employed in domestic cleaning: a community based

study. Thorax 58:950-954.

16. Rosenman, K. D., M. J. Reilly, D. P. Schill, D. Valiante, J. Flattery, R. Harrison, F.

Reinisch, E. Pechter, L. Davis, C. M. Tumpowsky, and M. Filios. 2003. Cleaning

products and work-related asthma. J Occup Environ Med 45:556-563.

17. Karjalainen, A., R. Martikainen, J. Karjalainen, T. Klaukka, and K. Kurppa. 2002.

Excess incidence of asthma among Finnish cleaners employed in different industries.

Eur Respir J 19:90-95.

18. Nielsen, J. and E. Bach. 1999. Work-related eye symptoms and respiratory symptoms

in female cleaners. Occup Med (Lond) 49:291-297.

19. Henneberger, P. K. 2007. Work-exacerbated asthma. Curr.Opin.Allergy Clin.Immunol.

7:146-151.

20. Caldeira, R. D., H. Bettiol, M. A. Barbieri, J. Terra-Filho, C. A. Garcia, and E. O.

Vianna. 2006. Prevalence and risk factors for work related asthma in young adults.

Occup Environ Med 63:694-699.

22

21. Berger, Z., W. N. Rom, J. Reibman, M. Kim, S. Zhang, L. Luo, and G. Friedman-

Jimenez. 2006. Prevalence of workplace exacerbation of asthma symptoms in an urban

working population of asthmatics. J Occup Environ Med 48:833-839.

22. Zock, J. P., N. Cavalle, H. Kromhout, S. M. Kennedy, J. Sunyer, A. Jaen, N.

Muniozguren, F. Payo, E. Almar, J. L. Sanchez, J. M. Anto, and M. Kogevinas. 2004.

Evaluation of specific occupational asthma risks in a community-based study with

special reference to single and multiple exposures. J.Expo.Anal.Environ Epidemiol.

14:397-403.

23. Jacobsen, G., V. Schlunssen, I. Schaumburg, and T. Sigsgaard. 2009. Increased

incidence of respiratory symptoms among female woodworkers exposed to dry wood.

Eur Respir J 33:1268-1276.

24. Smit, L. A., D. Heederik, G. Doekes, C. Blom, Z. van, I, and I. M. Wouters. 2008.

Exposure-response analysis of allergy and respiratory symptoms in endotoxin-exposed

adults. Eur Respir J 31:1241-1248.

25. Shaaban, R., M. Zureik, D. Soussan, C. Neukirch, J. Heinrich, J. Sunyer, M. Wjst, I.

Cerveri, I. Pin, J. Bousquet, D. Jarvis, P. G. Burney, F. Neukirch, and B. Leynaert.

2008. Rhinitis and onset of asthma: a longitudinal population-based study. Lancet

372:1049-1057.

26. Svanes, C., D. Jarvis, S. Chinn, E. Omenaas, A. Gulsvik, and P. Burney. 2002. Early

exposure to children in family and day care as related to adult asthma and hay fever:

results from the European Community Respiratory Health Survey. Thorax 57:945-950.

27. Ekholm O., Kjøller M., Davidsen M., Hesse U., Eriksen L., Christensen A.L., and

Grønbæk M. Public Health Report, Denmark 2005 & development since 1987. 2007.

23

National Institute of Public Health, Copenhagen, Denmark.

Ref Type: Report

28. Skadhauge, L. R., J. Baelum, H. C. Siersted, D. L. Sherson, S. Dahl, G. F. Thomsen, O.

Omland, E. Taudorf, and T. I. Sigsgaard. 2005. [The occurrence of asthma among

young adults. A population-based study in five west Danish counties]. Ugeskr.Laeger

167:648-651.

29. Kines, P., H. Hannerz, K. L. Mikkelsen, and F. Tuchsen. 2007. Industrial sectors with

high risk of women's hospital-treated injuries. Am.J Ind.Med 50:13-21.

30. Burr, H., J. B. Bjorner, T. S. Kristensen, F. Tuchsen, and E. Bach. 2003. Trends in the

Danish work environment in 1990-2000 and their associations with labor-force

changes. Scand.J Work Environ Health 29:270-279.

31. The European Community Respiratory Health Survey II Steering Committee. 2002.

The European Community Respiratory Health Survey II. Eur Respir J 20:1071-1079.

32. Statistics Denmark. Discoeloen 2004. 2003.

Ref Type: Catalog

33. Kennedy, S. M., M. N. Le, D. Choudat, and F. Kauffmann. 2000. Development of an

asthma specific job exposure matrix and its application in the epidemiological study of

genetics and environment in asthma (EGEA). Occup Environ Med 57:635-641.

34. Skov, T., J. Deddens, M. R. Petersen, and L. Endahl. 1998. Prevalence proportion

ratios: estimation and hypothesis testing. Int.J Epidemiol. 27:91-95.

35. Jaakkola, J. J. and M. S. Jaakkola. 2006. Professional cleaning and asthma.

Curr.Opin.Allergy Clin.Immunol. 6:85-90.

24

36. Balmes, J. R. 2002. Occupational airways diseases from chronic low-level exposures to

irritants. Clin.Chest Med 23:727-35, vi.

37. Tarlo, S. M. 2003. Workplace irritant exposures: do they produce true occupational

asthma? Ann Allergy Asthma Immunol. 90:19-23.

38. mich-Ward, H., P. G. Camp, and S. M. Kennedy. 2006. Gender differences in

respiratory symptoms-does occupation matter? Environ Res. 101:175-183.

39. Schachter, E. N., E. Zuskin, E. L. Moshier, J. Godbold, J. Mustajbegovic, J. Pucarin-

Cvetkovic, and A. Chiarelli. 2009. Gender and respiratory findings in workers

occupationally exposed to organic aerosols: a meta analysis of 12 cross-sectional

studies. Environ Health 8:1.

40. Hansen, C. L., J. Baelum, D. L. Sherson, L. R. Skadhauge, H. C. Siersted, O. Omland,

and T. I. Sigsgaard. 2005. [The significance of conditions when growing up on the

prevalence of atopy in adult Danes]. Ugeskr.Laeger 167:4570-4574.

25

FIGURE LEGENDS

Figure 1. Prevalence ratios of current asthma in participants exposed to JEM high-risk

asthma agents (n=4,981). Participants with high-risk asthma agent exposures (n=1,309)

compared to unexposed individuals (n=3,672), both overall and grouped by: gender; nasal

allergy (29 missing values); smoking status; parental asthma history; growing up in rural

area, on a farm or on a farm with animals; and the number of siblings. Prevalence ratios with

95% CIs were calculated for each group and were adjusted for age, county, gender (in

analyses other than those grouped by gender) and smoking status (in analyses other than

those grouped by smoking status). Each association was derived from a separate regression

model.

Figure 2. Prevalence ratios of current asthma in participants exposed to JEM low-risk asthma

agents (n=5,018). Participants with low-risk asthma agent exposures (n=1,346) compared to

unexposed individuals (n=3,672), both overall and grouped by: gender; nasal allergy (32

missing values); smoking status; parental asthma history; growing up in rural area, on a farm

or on a farm with animals; and the number of siblings. Prevalence ratios with 95% CIs were

calculated for each group and were adjusted for age, county, gender (in analyses other than

those grouped by gender) and smoking status (in analyses other than those grouped by

smoking status). Each association was derived from a separate regression model.

26

AUTHORS CONTRIBUTIONS

All authors have contributed to: formulation in the concept phase of the basic scientific

problem; planning of analyses and formulation of methodology; analysis of the concrete

investigation; presentation, interpretation and discussion of the results. All authors approved

the final version of the submitted report

Conflict of Interest Statement: All authors state no conflict of interest

27

TABLE 1. CHARACTERISTICS OF THE STUDY POPULATION BY GENDER

Men (n= 2,937) Women (n= 3,390) Age, years (mean (SD*)) 33.1 (6.9) 32.8 (7.0)

Age group, years (%) 20-24 25-29 30-34 35-39 40-44

14.9 18.4 21.2 22.4 23.1

16.2 19.3 21.0 21.8 21.7

County (%)

Funen Vejle Ribe Southern Jutland Northern Jutland

20.7 20.6 18.7 19.8 20.2

20.7 20.1 18.7 20.5 20.0

Smoking (%)

Non-smokers Ex-smokers

50.9 20.1 29.0

49.8 22.1 28.1

Current smokers Definitions of asthma (%)

Current wheeze Current asthma Doctor-diagnosed adult-onset asthma

14.4 8.0 2.9

12.9 9.4 4.8

Nasal allergy** (%) 22.5 23.2 Parental asthma (%) 13.1 17.9

Number of siblings (%)

0 1 2 3 4 or more Don’t know/missing

4.5 35.4 28.7 15.2 9.9 6.3

4.8 36.6 29.9 15.2 10.3 3.2

38.5 19.3

Growing up environment In a rural area (%) In a rural area on farm (%)

On farm with animals (%) 18.3

36.7 18.4 17.7

*SD, standard deviation **Data missing for 39 participants

28

Table 2. Prevalence ratios (PR) with 95% Confidence Interval (95% CI) for associations between occupational groups and current asthma by gender adjusted for county, age, and smoking status

Men Women Occupational group Number

Number current asthma

Number no current asthma

PR (95% CI) Number Number current asthma

Number no current asthma

PR (95% CI)

Legislators, managers, administrators (reference) 1,204 103 1,101 1.00 1,924 166 1,758 1.00 Cleaners and caretakers 45 5 40 1.20 (0.51-2.82) 124 24 100 2.17 (1.47-3.21) Hairdressers, barbers, beauticians 2 0 2 NE 39 0 39 NE Nurses 4 0 4 NE 158 13 145 0.97 (0.56-1.66) Other medical and pharmacy 56 4 52 NE 577 63 514 1.25 (0.95-1.65) Agriculture and forestry 166 12 154 0.80 (0.45-1.43) 63 7 56 1.24 (0.61-2.52) Wood workers 142 13 129 1.07 (0.61-1.86) 6 1 5 NE Bakery workers 16 3 13 NE 4 0 4 NE Food and tobacco processing 70 7 63 1.22 (0.59-2.53) 83 10 73 1.41 (0.78-2.56) Chemical and physical science technicians 3 0 3 NE 15 1 14 NE Plastic and rubber workers 13 0 13 NE 6 0 6 NE Chemical processors 0 0 0 NE 0 0 0 NE Welders and flame cutters 11 0 11 NE 0 0 0 NE Metal making and treating 99 11 88 1.26 (0.70-2.28) 6 0 6 NE Other metal workers 221 16 205 0.85 (0.51-1.42) 16 2 14 NE Electrical processors 106 7 99 0.74 (0.35 -1.57) 6 1 5 NE Painters 22 3 19 NE 28 1 27 NE Spray painters 4 1 3 NE 0 0 0 NE Textile, leather and fur workers 4 0 4 NE 13 1 12 NE Paper workers 5 0 5 NE 3 0 3 NE Printing workers 16 4 12 NE 3 0 3 NE Glass and ceramics workers 7 0 7 NE 5 1 4 NE Construction and mining 192 13 179 0.79 (0.45-1.39) 24 1 23 NE Drivers 128 8 120 0.74 (0.37-1.48) 12 2 10 NE Remainder transport and storage 70 4 66 NE 27 2 25 NE Remainder blue-collar 203 11 192 0.63 (0.34-1.16) 132 11 121 0.99 (0.55-1.78) Not classifiable 128 9 119 0.79 (0.41-1.52) 116 12 104 1.17 (0.67-2.05)

NE, not estimated due to less than 5 asthma cases.

Table 3. Prevalence ratios (PR) with 95% Confidence Interval (95% CI) for associations between occupational exposure groups and current asthma by gender adjusted for county, age, and smoking status Men Women Exposures grouped according to Asthma-specific Job Exposure Matrix*

Number total

Number current asthma

Number no current asthma

PR (95% CI) Number total

Number current asthma

Number no current asthma

PR (95% CI)

Not exposed (reference) 1,498 118 1,380 1.00 2,174 188 1,986 1.00 High risk asthma agents 482 44 438 1.14 (0.82-1.59) 827 87 740 1.21 (0.95-1.54) HMW agents 153 19 134 1.59 (1.01-2.51) 709 77 632 1.25 (0.97-1.61) Animals 33 5 28 1.91 (0.85-4.32) 28 1 27 NE Fish 11 1 10 NE 5 0 5 NE Flour 17 4 13 NE 4 0 4 NE Plants 1 1 0 NE 1 0 1 NE Mites 1 1 0 NE 20 6 14 3.55 (1.76-7.14) Enzymes 16 3 12 NE 4 0 4 NE Latex 42 2 40 NE 630 67 563 1.23 (0.94-1.60) Bioaerosols 71 8 63 1.44 (0.73-2.86) 23 3 20 NE Drugs 3 0 3 NE 161 14 148 0.95 (0.56-1.64) LMW agents 272 24 248 1.08 (0.71-1.65) 553 59 494 1.22 (0.92-1.61) Reactive chemicals 82 5 77 0.76 (0.32-1.81) 525 57 468 1.25 (0.94-1.65) Isocyanate 25 1 24 NE 8 0 8 NE Cleaning agents 40 2 38 NE 438 53 385 1.40 (1.04-1.87) Wood dust 0 0 0 NE 0 0 0 NE Metals 194 19 175 1.21 (0.76-1.91) 60 5 55 0.97 (0.42-2.27) Mixed environments 235 20 215 1.09 (0.69-1.72) 77 8 69 1.18 (0.60-2.31) Metal working fluids 81 8 73 1.24 (0.62-2.47) 8 0 8 NE Textile 4 0 4 NE 11 1 10 NE Agricultural antigens 150 12 138 1.05 (0.60-1.86) 58 7 51 1.36 (0.67-2.77) High irritant peaks 45 5 40 1.37 (0.59-3.18) 5 0 5 NE Low risk asthma agents 957 72 885 0.94 (0.71-1.25) 389 44 345 1.30 (0.95-1.77) Exhaust 212 11 201 0.66 (0.36-1.20) 16 2 14 NE Environmental tobacco smoke 11 2 9 NE 35 1 34 NE Possible irritants 418 31 387 0.91 (0.62-1.33) 146 21 125 1.68 (1.10-2.55) Low risk antigens 450 37 413 1.04 (0.73-1.48) 347 40 307 1.31 (0.95-1.81)

NE, not estimated due to less than 5 asthma cases. *According to published matrix33. Participants can be categorised in more than one exposure category in high risk asthma agents and low risk asthma agents, therefore, number exceed 100% in these groups.

FIGURE 1

FIGURE 2

32

Supplementary table 1. Prevalence ratios (PR) with 95% Confidence Interval (95% CI) for associations between occupational groups and current wheeze by gender adjusted for county, age, and smoking status

Men Women Occupational group Number

Number current wheeze

Number no current wheeze

PR (95% CI) Number Number current wheeze

Number no current wheeze

PR (95%CI)

Legislators, managers, administrators (reference) 1,204 184 1020 1.00 1,924 237 1,687 1.00 Cleaners and caretakers 45 10 35 1.42 (0.83-2.42) 124 24 100 1.50 (1.02-2.18) Hairdressers, barbers, beauticians 2 0 2 NE 39 3 36 0.60 (0.20-1.80) Nurses 4 1 3 1.99 (0.34-11.60) 158 15 143 0.80 (0.49-1.32) Other medical and pharmacy 56 9 47 1.13 (0.62-2.05) 577 87 490 1.19 (0.95-1.50 Agriculture and forestry 166 19 147 0.81 (0.52-1.26) 63 8 55 0.97 (0.50-1.88) Wood workers 142 16 126 0.69 (0.42-1.11) 6 1 5 1.12 (0.19-6.70) Bakery workers 16 1 15 0.49 (0.07-3.33) 4 0 4 NE Food and tobacco processing 70 12 58 1.01 (0.59-1.73) 83 13 70 1.31(0.79-2.18) Chemical and physical science technicians 3 1 2 2.87 (0.84-9.81) 15 0 15 NE Plastic and rubber workers 13 1 12 0.52 (0.09-3.16) 6 2 4 2.83 (0.95-8.43) Chemical processors 0 0 0 NE 0 0 0 NE Welders and flame cutters 11 2 9 1.19 (0.37-3.80) 0 0 0 0.83 (0.58-1.17) Metal making and treating 99 16 83 0.99 (0.62-1.57) 6 0 6 NE Other metal workers 221 32 189 0.91 (0.65-1.29) 16 4 12 1.98 (0.83-4.71) Electrical processors 106 14 92 0.89 (0.54-1.48) 6 2 4 2.66 (0.93-7.67) Painters 22 3 19 0.84 (0.28-2.47) 28 1 27 0.28 (0.04-1.92) Spray painters 4 1 3 1.35 (0.22-8.35) 0 0 0 0.83 (0.59-1.17) Textile, leather and fur workers 4 0 4 NE 13 1 12 0.64 (0.10-4.11) Paper workers 5 1 4 1.05 (0.19-5.98) 3 0 3 NE Printing workers 16 4 12 1.63 (0.71-3.78) 3 0 3 NE Glass and ceramics workers 7 1 6 1.09 (0.19-6.31) 5 4 1 4.87 (2.64-9.01) Construction and mining 192 24 168 0.76 (0.51-1.13) 24 4 20 1.43 (0.59-3.47) Drivers 128 15 113 0.74 (0.45-1.20) 12 2 10 1.39 (0.38-5.09) Remainder transport and storage 70 8 62 0.71 (0.37-1.36) 27 1 26 0.29 (0.04-1.99) Remainder blue-collar 203 28 175 0.88 (0.61-1.27) 132 14 118 0.86 (0.52-1.44) Not classifiable 128 19 109 0.99 (0.65-1.50) 116 15 101 1.02 (0.63-1.66)

NE, not estimable due to small numbers of cases

SUPPLEMENTARY TABLE 2. PREVALENCE RATIOS (PR) WITH 95% CONFIDENCE INTERVAL (95% CI) FOR ASSOCIATIONS BETWEEN OCCUPATIONAL GROUPS AND DOCTOR DIAGNOSED ADULT-ONSET ASTHMA BY GENDER ADJUSTED FOR COUNTY, AGE, AND SMOKING STATUS

Men Women Occupational group Number

Number doctor diagnosed adult-onset asthma

Number no doctor diagnosed adult-onset asthma

PR (95% CI) Number Number doctor diagnosed adult-onset asthma

Number no doctor diagnosed adult-onset asthma

PR (95% CI)

Legislators, managers, administrators (reference) 1,204 38 1,166 1.00 1,924 93 1,831 1.00 Cleaners and caretakers 45 3 42 1.96 (0.62-6.17) 124 9 115 1.59 (0.82-3.11) Hairdressers, barbers, beauticians 2 0 2 NE 39 2 37 1.17 (0.30-4.58) Nurses 4 0 4 NE 158 3 155 0.40 (0.13-1.23) Other medical and pharmacy 56 0 56 NE 577 31 546 1.14 (0.76-1.69) Agriculture and forestry 166 6 160 1.27 (0.54-3.00) 63 6 57 1.94 (0.88-4.29) Wood workers 142 2 140 0.49 (0.12-2.01) 6 0 6 NE Bakery workers 16 1 15 2.36 (0.33-16.74) 4 0 4 NE Food and tobacco processing 70 2 68 0.93 (0.23-3.80) 83 3 80 0.75 (0.24-2.32) Chemical and physical science technicians 3 0 3 NE 15 0 15 NE Plastic and rubber workers 13 0 13 NE 6 0 6 NE Chemical processors 0 0 0 NE 0 0 0 NE Welders and flame cutters 11 0 11 NE 0 0 0 NE Metal making and treating 99 2 97 0.61 (0.15-2.50) 6 0 6 NE Other metal workers 221 4 217 0.58 (0.21-1.61) 16 3 13 4.29 (1.51-12.21) Electrical processors 106 3 103 0.93 (0.29-2.99) 6 1 5 3.77 (0.63-22.51) Painters 22 1 21 1.61 (0.23-11.46) 28 2 26 1.39 (0.36-5.32) Spray painters 4 0 4 NE 0 0 0 NE Textile, leather and fur workers 4 0 4 NE 13 0 13 NE Paper workers 5 0 5 NE 3 0 3 NE Printing workers 16 2 14 5.24 (1.33-20.62) 3 0 3 NE Glass and ceramics workers 7 0 7 NE 5 0 5 NE Construction and mining 192 3 189 0.56 (0.17-1.80) 24 2 22 2.11 (0.54-8.21) Drivers 128 7 121 1.78 (0.82-3.89) 12 0 12 NE Remainder transport and storage 70 0 70 NE 27 0 27 NE Remainder blue-collar 203 8 195 1.27 (0.60-2.70) 132 3 129 0.48 (0.15-1.50) Not classifiable 128 3 125 0.71 (0.22-2.28) 116 4 112 0.75 (0.28-2.00)

NE, not estimable due to small numbers of cases

34

Supplementary table 3. Prevalence ratios (PR) with 95% Confidence Interval (95% CI) for associations between occupational exposure groups and current wheeze by gender adjusted for county, age, and smoking status Men Women Exposures grouped according to Asthma-specific Job Exposure Matrix*

Number total

Number current wheeze

Number no current wheeze

PR (95% CI) Number Total

Number current wheeze

Number no current wheeze

PR (95% CI)

Not exposed (reference) 1,498 219 1,279 1.00 2,174 261 1913 1.00 High risk asthma agents 482 71 411 0.98 (0.77-1.25) 827 113 714 1.12 (0.91-1.38) HMW agents 153 30 123 1.29 (0.93-1.80) 709 97 612 1.13 (0.91-1.40) Animals 33 5 28 1.01 (0.46-2.23) 28 4 24 1.19 (0.48-2.95) Fish 11 2 9 1.04 (0.29-3.73) 5 1 4 1.98 (0.36-10.79) Flour 17 1 16 0.49 (0.07-3.38) 4 0 4 NE Plants 1 1 0 5.87 (4.03-8.56) 1 0 1 NE Mites 1 0 1 NE 20 3 17 1.36 (0.48-3.85) Enzymes 16 1 15 0.58 (0.08-3.99) 4 0 4 NE Latex 42 9 33 1.47 (0.85-2.55) 630 86 544 1.13 (0.90-1.41) Bioaerosols 71 15 56 1.34 (0.86-2.10) 23 2 21 0.74 (0.19-2.76) Drugs 3 0 3 NE 161 15 146 0.82 (0.50-1.34) LMW agents 272 43 229 0.98 (0.73-1.31) 553 84 469 1.22 (0.97-1.53) Reactive chemicals 82 14 68 1.17 (0.74-1.87) 525 81 444 1.23 (0.98-1.56) Isocyanate 25 5 20 1.36 (0.65-2.85) 8 2 6 1.88 (0.57-6.29) Cleaning agents 40 7 33 1.14 (0.60-2.17) 438 68 370 1.24 (0.97-1.59) Wood dust 0 0 0 NE 0 0 0 NE Metals 194 30 164 0.93 (0.66-1.32) 60 9 51 1.26 (0.68-2.34) Mixed environments 235 29 206 0.84 (0.59-1.20) 77 8 69 0.85 (0.44-1.66) Metal working fluids 81 11 70 0.80 (0.46-1.40) 8 0 8 NE Textile 4 0 4 NE 11 0 11 NE Agricultural antigens 150 18 132 0.92 (0.60-1.42) 58 8 50 1.11 (0.58-2.14) High irritant peaks 45 8 37 1.18 (0.64-2.18) 5 2 3 3.85 (1.32-11.21) Low risk asthma agents 957 132 825 0.88 (0.72-1.07) 389 64 325 1.31 (1.02-1.68) Exhaust 212 26 186 0.83 (0.57-1.20) 16 4 12 1.92 (0.82-4.48) Environmental tobacco smoke 11 2 9 0.85 (0.24-3.04) 35 5 30 0.99 (0.43-2.25) Possible irritants 418 55 363 0.82 (0.62-1.07) 146 27 119 1.49 (1.04-2.14) Low risk antigens 450 75 375 1.07 (0.84-1.35) 347 56 291 1.27 (0.98-1.66)

NE, not estimable due to small numbers of cases. *According to published matrix33. Participants can be categorised in more than one exposure category, therefore, numbers exceed 100%

35

Supplementary table 4. Prevalence ratios (PR) with 95% Confidence Interval (95% CI) for associations between occupational exposure groups and doctor-diagnosed adult-onset asthma by gender adjusted for county, age, and smoking status Men Women Exposures grouped according to Asthma-specific Job Exposure Matrix*

Number total

Number doctor diagnosed adult-onset asthma

Number no doctor diagnosed adult-onset asthma

PR (95% CI) Numbertotal

Number doctor diagnosed adult-onset asthma

Number no doctor diagnosed adult-onset asthma

PR (95% CI)

Not exposed (reference) 1,498 43 1,455 1.00 2,174 99 2,075 1.00 High risk asthma agents 482 14 468 1.05 (0.58-1.90) 827 42 785 1.14 (0.80-1.63) HMW agents 153 3 150 0.72 (0.23-2.31) 709 36 673 1.14 (0.78-1.67) Animals 33 1 32 0.99 (0.14-6.90) 28 2 26 1.58 (0.40-6.15) Fish 11 1 10 2.78 (0.43-18.11) 5 0 5 NE Flour 17 1 16 2.24 (0.32-15.87) 4 0 4 NE Plants 1 0 1 NE 1 0 1 NE Mites 1 0 1 NE 20 2 18 2.57 (0.67-9.90) Enzymes 16 1 15 2.69 (0.37-19.24) 4 0 4 NE Latex 42 0 42 NE 630 31 599 1.12 (0.76-1.67) Bioaerosols 71 1 70 0.52 (0.07-3.76) 23 3 20 3.08 (1.04-9.10) Drugs 3 0 3 NE 161 3 158 0.42 (0.14-1.32) LMW agents 272 5 267 0.65 (0.26-1.62) 553 31 522 1.28 (0.86-1.90) Reactive chemicals 82 0 82 NE 525 30 495 1.31 (0.88-1.96) Isocyanate 25 0 25 NE 8 0 8 NE Cleaning agents 40 0 40 NE 438 26 412 1.37 (0.90-2.10) Wood dust 0 0 0 NE 0 0 0 NE Metals 194 5 189 0.90 (0.36-2.24) 60 3 57 1.07 (0.35-3.29) Mixed environments 235 9 226 1.47 (0.72-2.99) 77 7 70 2.08 (1,00-4.36) Metal working fluids 81 3 78 1.41 (0.44-4.52) 8 1 7 3.04 (0.48-19.23) Textile 4 0 4 NE 11 0 11 NE Agricultural antigens 150 6 144 1.62 (0.70-3.77) 58 6 52 2.32 (1.06-5.10) High irritant peaks 45 3 42 2.13 (0.68-6.62) 5 0 5 NE Low risk asthma agents 957 28 929 1.04 (0.65-1.66) 389 21 368 1.26 (0.80-2.00) Exhaust 212 8 204 1.41 (0.67-2.95) 16 1 15 1.51 (0.22-10.40) Environmental tobacco smoke 11 0 11 NE 35 0 35 NE Possible irritants 418 8 410 0.68 (0.32-1.43) 146 9 137 1.44 (0.74-2.80) Low risk antigens 450 15 435 1.17 (0.66-2.08) 347 18 329 1.21 (0.74-1.98)

NE, not estimable due to small numbers of cases. *According to published matrix 33. Participants can be categorised in more than one exposure category, therefore, numbers exceed 100%

36