970638920773Int J Tuberc Lung DisInt. J. Tuberc. Lung Dis.The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease1027-37191815-792025574926458024110.5588/ijtld.14.0132HHSPA714307ArticleOccupational exposures associated with severe exacerbation of asthmaHennebergerP. K.*LiangX.*LillienbergL.Dahlman-HöglundA.TorénK.AnderssonE.Division of Respiratory Disease Studies, National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, Morgantown, West Virginia, USADepartment of Occupational and Environmental Medicine, Sahlgrenska University Hospital, Göteborg, SwedenCorrespondence to: P K Henneberger, Division of Respiratory Disease Studies, National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, 1095 Willowdale Road, MS H2800, Morgantown, WV 26505, USA. Tel: (+l) 304 285 6161. Fax: (+1) 304 285 5820. pkh0@cdc.gov1292015220152392015192244250SUMMARYBACKGROUND

The exacerbation of asthma by workplace conditions is common, but little is known about which agents pose a risk.

OBJECTIVE

We used data from an existing survey of adults with asthma to identify occupational exposures associated with severe exacerbation of asthma.

DESIGN

Questionnaires were completed by 557 working adults with asthma. Severe exacerbation of asthma in the past 12 months was defined as asthma-related hospitalization, or reports of both unplanned asthma care and treatment with a short course of oral corticosteroids. Occupational exposures for the same time period were assessed using an asthma-specific job exposure matrix. We modeled severe exacerbation to yield prevalence ratios (PRs) for exposures while controlling for potential confounders.

RESULTS

A total of 164 participants (29%) were positive for severe exacerbation, and 227 (40.8%) were assessed as being exposed to asthma agents at work. Elevated PRs were observed for several specific agents, notably the irritant subcategories of environmental tobacco smoke (PR 1.84, 95%CI 1.34–2.51) among all participants, inorganic dusts (PR 2.53, 95%CI 1.37– 4.67) among men, and the low molecular weight subcategory of other highly reactive agents (PR 1.97, 95%CI 1.08–3.60) among women.

CONCLUSION

Among working adults with asthma, severe exacerbation was associated with several occupational agents.

work-exacerbated asthmajob-exposure matrixoccupational epidemiology

WORK-RELATED ASTHMA comprises occupational asthma (OA) that is caused by conditions at work and work-exacerbated asthma (WEA), in which existing asthma is made worse by workplace conditions.1 Work-related asthma is common, with work accounting for an average 16.9% of all new adult-onset asthma cases,2 and with WEA detected in an estimated 21.5% of working adults with asthma.3 While substantial costs and lost work time are associated with both WEA and OA,4,5 WEA has received less attention in terms of research and prevention efforts.1 WEA is potentially preventable by minimizing harmful workplace exposures,3 but a better understanding of which agents pose a risk is needed to guide prevention efforts. While surveillance in the United States indicates that WEA is frequently caused by irritants, low molecular weight (e.g., acids, bases, aldehydes) and high molecular weight (e.g., latex) agents are also implicated.6 The absence of denominators (i.e., the number of workers exposed) prohibits using surveillance data to estimate agent-specific risk.

We used data from a previous study to identify occupational risk factors for exacerbation of asthma.7 In previous analyses of data from this study, we determined WEA status on a case-by-case basis. However, for the current investigation, we modeled severe exacerbation of asthma and determined which occupational exposures were associated with it, while simultaneously controlling for potential confounders. We assessed occupational exposures using a new asthma-specific job-exposure matrix (JEM).8

STUDY POPULATION AND METHODS

The details of the study are presented elsewhere.9 The protocol was approved by the Human Subjects Review Board of the National Institute for Occupational Safety and Health (NIOSH, Washington DC, USA), and all participants provided informed consent. Study participants were patients at a health maintenance organization (HMO) in the state of Massachusetts, USA. The HMO electronic medical records were reviewed during 2000–2001 to identify cases of asthma aged 18–44 years. Invitees were required to have active asthma as indicated by a recorded asthma diagnosis and treatment for asthma in the past 12 months. With the goal of studying exacerbation rather than onset of asthma during the 12 months before interview, we reviewed paper medical records to confirm that asthma onset was at least 1 year before enrollment, and excluded those who reported their first asthma attack as occurring <14 months before interview. A total of 598 (61%) of the 978 invitees completed a telephone questionnaire during 2001–2002; 41 were excluded because they had been unemployed in the previous 12 months, leaving 557 participants for the current analysis.

Responses to questionnaire items provided information on occupations, severe exacerbation of asthma, and demographics. Each job worked in the past 12 months was coded using the 1988 International Standard Classification of Occupations (ISCO-88).10 We assessed occupational exposures with a new asthma-specific JEM initially intended for use in Northern Europe, called the N-JEM.8 Development of the N-JEM followed the same principles used to create an earlier asthma JEM.11,12 Two industrial hygienists assessed each occupation as exposed, not exposed, or uncertain/low exposed. The ‘exposed’ label was assigned to occupations judged as having at least half of the workers with a high probability of exposure. The raters judged exposure (no/yes) for each of several subcategories in four major categories of agents: high molecular weight (HMW), low molecular weight (LMW), irritants (IRR), and accidental peak exposures to irritants (PEAKS). Assessments were then discussed with two occupational medicine specialists until consensus was reached.8

We defined severe exacerbation of asthma based on self reports if in the past 12 months the participant had either been hospitalized for asthma, or received both unplanned care for an acute asthma attack (at a doctor's office, urgent care facility, or emergency department) and a short course of oral corticosteroids for asthma. This definition is consistent with recent recommendations for defining severe exacerbation of asthma.13,14

As the outcome was relatively common, we calculated prevalence ratios (PRs) rather than odds ratios,15,16 using a Cox regression model with robust variance17 and a constant follow-up time.18 We fitted a base model for severe exacerbation using the following candidate variables: sex (female vs. male), age in years (30–38 and 39–44 vs. 18–29), highest education (college degree or more vs. some college or less), race (non-white vs. white), cigarette smoking (ever vs. never), age at asthma onset (≥18 vs. <18 years), history of allergies based on whether a doctor had ever said the participant had hay fever or skin allergies (yes vs. no), and asthma severity based on review of the medical records (persistent vs. intermittent).19 We used forward selection and backward elimination, with P ≤ 0.15 as the criterion for retention in the model. Occupational exposure variables developed using the N-JEM were then added to the base model, as detailed in the Results. Statistical tests were considered significant if P ≤ 0.05. All data analyses were conducted using SAS® software version 9.2 (Statistical Analysis System, Cary, NC, USA).

RESULTS

The 557 participants included approximately twice as many women (68%) as men (32%), and were overwhelmingly White (94%). The median age was 34 years, the median age at asthma onset was 13, and 345 (62%) participants had experienced onset before the age of 18 years; 40% were smokers (18% current, 22% former), 74% had allergies, and 37% had completed at least college education. The distribution by level of asthma severity was 31% mild intermittent and 69% mild/moderate/severe persistent. The group had worked 771 jobs in the 12 months before interview, with two thirds (n = 377, 68%) working one job, one fourth (n = 151, 27%) two jobs, and only a few three (n = 25), four (n = 3), or five (n = 1) jobs. The participants worked primarily in white collar and service jobs (n = 480, 86%, in ISCO-88 Groups 1–5) and infrequently in blue collar jobs (n = 90, 16%, in ISCO-88 Groups 6–9). Women were more likely than men to have jobs as technicians and associate professionals and service workers, but were less likely to be employed in craft trades and as plant and machine operators and assemblers.

Approximately two in every five participants (n = 227, 40.8%) were assessed by the N-JEM to have experienced occupational exposure to asthma-related agents (i.e., HMW, LMW, IRR, and PEAKS) in the past 12 months (Table 1). IRR was the most common exposure, with about one fourth of all participants (27.1%), and environmental tobacco smoke (ETS) was the most common IRR subcategory (11.3%). The other major agent categories in decreasing frequency were HMW (13.8%), LMW (4.8%), and PEAKS (1.1%). More men than women were exposed to asthma agents (48.3% vs. 37.1%, P = 0.02); men were more likely to have IRR and LMW exposures, and less likely to be exposed to HMW agents (Table 1). Many exposure subcategories had small sample sizes when the data were subdivided by sex.

Severe exacerbation was reported by 164 participants (29%). It was more common among women than men and those with persistent than intermittent severity, and was less common among the oldest participants (Table 2). Severe exacerbation showed little variation in frequency by education, race, smoking status, age at asthma onset, and history of allergies. The crude frequency was greater among participants with occupational exposure than among those with no exposure (35% vs. 25%, Table 3). By occupational exposure subcategories, crude values were notably high for the LMW agents epoxy (40%), other highly reactive agents (50%), and isocyanates (50%), and the IRR agents inorganic dusts (52%), metalworking fluids (42%), and ETS (48%).

The base regression model for severe exacerbation included covariates for sex, age (39–44 years vs. other ages), and asthma severity (persistent vs. intermittent). Participants with jobs that had no asthma-related exposures populated the common reference category for all occupational covariates. We did not estimate PRs for exposure categories with <5 exposed participants or no severe exacerbation cases. The PR for any exposure to asthma-related agents (other than uncertain or low exposure) was 1.43 (95% confidence interval [CI] 1.10–1.84, P = 0.007) (model not shown). We fitted separate models for each exposure category and subcategory (Table 4). All PRs for major exposure categories exceeded 1 and were statistically significant for LMW agents, IRR agents, and PEAKS. As all six participants with PEAKS also had IRR exposure and three had LMW exposure, it was impossible to evaluate the effect of PEAKS separately. With LMW and IRR in the same model (not shown), the PR (1.58, 95%CI 1.20–2.08, P=0.001) for IRR was similar to that observed when this exposure was in a model by itself, but the PR for LMW agents was no longer statistically significant (1.47, 95%CI 0.94–2.29, P = 0.09).

An association with severe¼exacerbation of asthma was observed for seven exposure subcategories (Table 4) when each agent was included in a regression model with the potential confounders: the LMW agents epoxy (PR 2.50, P = 0.046), other highly reactive agents (PR 1.93, P = 0.03), and isocyanates (PR 3.11, P = 0.001); and the IRR agents inorganic dusts (PR 3.61, P < 0.0001), metalworking fluids (PR 2.84, P = 0.005), combustion particles (PR 1.52, P = 0.07), and ETS (PR 1.88, P = 0.0001). Many participants experienced concurrent exposures and, in particular, exposure to inorganic dusts was related to all other implicated exposures except ETS. Epoxy and isocyanates overlapped so completely with inorganic dusts that it was impossible to test for their independent effects, while inorganic dusts had an effect that was separate from these two exposures (data not shown). In a model with the other five exposures, both inorganic dusts (PR 2.41) and ETS (PR 1.84) had strong positive effects (P<0.05), while the PR for other highly reactive agents was elevated but not statistically significant (PR 1.65, P = 0.08; Model A, Table 5).

Small sample sizes limited the exposure subcategories that we could examine separately for men and women. The model for men (Model B, Table 5) included covariates for four of the IRR agents implicated above, but excluded other highly reactive agents because only four men were exposed. The statistically significantly positive findings for men were inorganic dusts and ETS, similar to the findings for all participants. Women had sufficient sample sizes for only three of the seven exposures implicated, and this did not include inorganic dusts. The regression model specific to women (Model C, Table 5) had elevated PRs for other highly reactive agents (PR 1.97, P = 0.03) and ETS (PR 1.50, P = 0.03).

DISCUSSION

IRRs had the strongest association with severe exacerbation of asthma, and particularly the exposure subcategories of inorganic dusts and ETS. Based on surveillance conducted in the United States, mineral and inorganic dusts are frequently identified as the putative cause of WEA.6,20 Exposure to ETS was the most common occupational exposure among the 16 exposure subcategories, and has frequently been reported as a cause of WEA in other studies.6,2124 Cigarette smoking and ETS in the workplace are less common now than when the interviews were conducted in 2001–2002.2528 Specifically, for the state of Massachusetts (the site of the current study), the 2010 Behavioral Risk Factor Surveillance System survey found that 5.4% of non-smoking adults reported being exposed to secondhand smoke at work during the past week, down from 8% in 2003.29 However, the 2010 prevalence indicates that workplace ETS is still common, exceeding the prevalence of 13 of the 16 occupational exposure subcategories in the current investigation. The LMW subcategory of other highly reactive agents was a risk factor for severe exacerbation among women, and this category includes chemicals such as acids with irritant properties that may have contributed to exacerbation.

This investigation had several strengths: it was conducted in an HMO, a quasi-population-based setting, and participants were employed in different occupations and industries. Furthermore, by using regression to model severe exacerbation of asthma, we were able to control for potential confounders while estimating the strength of the association between the outcome and occupational exposures. The asthma-specific N-JEM has been used successfully in another study of asthma,8 and provided objective assessments of occupational exposures, thus avoiding the bias that has been observed with self-reported exposures.30 In addition, the N-JEM assigned participants to more specific subcategories of occupational exposures than was accomplished in other population-based studies that used either self reports and expert evaluation of workplace exposures31 or another JEM9 when investigating exacerbation of asthma.

The current study had several limitations. As the N-JEM was developed with a focus on the onset of asthma rather than exacerbation, exposures relevant to the latter but not the former may have been missed. When we limited the analysis to the 476 subjects with asthma onset before starting jobs held in the past 12 months, our findings remained unchanged (data not shown), suggesting that the results were not driven by exposure related to asthma onset. While the same jobs in northern Europe and the United States generally have similar occupational exposures, differences may have resulted in exposure misclassification.

Participants were required to have evidence in their medical records of treatment for asthma in the past 12 months. Very mild cases were therefore not included, and the findings are not necessarily relevant to them. While the participants’ occupational exposures and severe exacerbation events occurred in the same 12-month period, it is uncertain whether exposure always preceded the outcome. This probably introduced non-differential misclassification of dichotomous exposure variables, thus biasing exposure PRs to the null.32 As the maximum age at enrollment was 44 years, the results may not be relevant to older asthma patients. Excluding older adults probably did avoid confusing asthma with other respiratory conditions, notably chronic obstructive pulmonary disease.

Concurrent occupational exposures and small sample sizes limited our ability to test for exposure-response relationships. An extreme overlap with other exposures meant we were unable to assess the independent effects of PEAKS, epoxy, and isocyanates. Sample sizes were a particular problem when we divided the study group by sex. The small number of participants who had worked in manual labor jobs is indicative of Massachusetts, USA, which has fewer blue collar workers than most other states.33 At the same time, the pattern of more white vs. blue collar jobs is indicative of the entire United States, and current findings are relevant beyond Massachusetts.

CONCLUSION

The current study identified several occupational exposures associated with exacerbation of asthma, providing an additional level of detail that was not available in other population-based studies that used a risk-set approach. The results suggest that minimizing exposure to the implicated agents might reduce the frequency of WEA. Additional studies with larger sample sizes are needed to refine our understanding of occupational risk factors for exacerbation of asthma, and to assess the impact of different interventions.

Acknowledgements

The authors thank the adults with asthma who provided the data used in this article; and collaborators D Milton, R Boylstein, S Derk, P Preusse, and S Sama who helped with planning and implementing the original study.

Conflicts of interest: none declared.

Disclaimer: the findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Institute for Occupational Safety and Health, Washington DC, USA.

ReferencesTarloSMBalmesJBalkissoonRDiagnosis and management of work-related asthma: American College Of Chest Physicians Consensus Statement.Chest20081343 SupplS1S41TorenKBlancPDAsthma caused by occupational exposures is common—a systematic analysis of estimates of the population-attributable fraction.BMC Pulm Med20099719178702HennebergerPKRedlichCACallahanDBAn official American Thoracic Society Statement: work-exacerbated asthma.Am J Respir Crit Care Med201118436837821804122VandenplasOTorenKBlancPDHealth and socioeconomic impact of work-related asthma.Eur Respir J20032268969714582924VandenplasOHennebergerPKSocio-economic outcomes in work-exacerbated asthma.Curr Opin Allergy Clin Immunol2007723624117489041Centers for Disease Control and PreventionWork-related asthma: most frequently reported agents associated with work-related asthma cases by asthma classification, 1993–20062012CDCAtlanta, GA, USAhttp://www2a.cdc.gov/drds/WorldReportData/FigureTableDetails.asp?FigureTableID=2611&GroupRefNumber=T09-05A.November 2014HennebergerPKDerkSJSamaSRThe frequency of workplace exacerbation among health maintenance organisation members with asthma.J Occup Environ Med200663551557LillienbergLAnderssonEJansonCOccupational exposure and new-onset asthma in a population-based study in northern Europe (RHINE).Ann Occup Hyg20125748249223204511HennebergerPKMirabelliMCKogevinasMThe occupational contribution to severe exacerbation of asthma.Eur Respir J20103674375020351033International Labour OrganizationInternational Standard Classification of Occupations ISCO 881991ILOGeneva, SwitzerlandKennedySMLe MoualNChoudatDKauffmannFDevelopment of an asthma specific job exposure matrix and its application in the epidemiological study of genetics and environment in asthma (EGEA).Occup Environ Med20005763564110935945KogevinasMZockJPJarvisDExposure to substances in the workplace and new-onset asthma: an international prospective population-based study (ECRHS-II).Lancet200737033634117662882FuhlbriggeAPedenDApterAJAsthma outcomes: exacerbations.J Allergy Clin Immunol2012129SupplS34S4822386508ReddelHKTaylorDRBatemanEDAn Official American Thoracic Society/European Respiratory Society Statement: asthma control and exacerbations standardising endpoints for clinical asthma trials and clinical practice.Am J Respir Crit Care Med2009180599919535666LeeJTanCSChiaKSA practical guide for multivariate analysis of dichotomous outcomes.Ann Acad Med Singapore20093871471919736577ThompsonMLMyersJEKriebelDPrevalence odds ratio or prevalence ratio in the analysis of cross-sectional data: what is to be done!J Occup Environ Med199855272277LinDYWeiLJThe robust inference for the proportional hazards model.J Am Stat Assoc19898410741078BarrosAJHirakataVNAlternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio.BMC Med Res Methodol200332114567763Global Initiative for AsthmaGlobal strategy for asthma management and prevention.National Heart, Lung and Blood Insitute/World Health Organization workshop report1995NHLBIBethesda, MD, USAAndersonNJReeb-WhitakerCKBonautoDKRauserEWork-related asthma in Washington state.J Asthma20114877378221851158GassertTHHuHKelseyKTChristianiDCLong-term health and employment outcomes of occupational asthma and their determinants.J Occup Environ Med1998404814919604186MancusoCARinconMCharlsonMEAdverse work outcomes and events attributed to asthma.Am J Ind Med20034423624512929143TarloSMLeungKBroderISilvermanFHolnessDLAsthmatic subjects symptomatically worse at work: prevalence and characterization among a general asthma clinic population.Chest20001181309131411083679TarloSMLissGCoreyPBroderIA workers’ compensation claim population for occupational asthma. Comparison of subgroups.Chest19951076346417874929BohacDLHewettMJKapphahnKIGrimsrudDTApteMGGundelLAChange in indoor particle levels after a smoking ban in Minnesota Bars and restaurants.Am J Prev Med201039Suppl 1S3S921074674GoodmanPAgnewMMcCaffreyMPaulGClancyLEffects of the Irish smoking ban on respiratory health of bar workers and air quality in Dublin pubs.Am J Respir Crit Care Med200717584084517204724MenziesDNairAWilliamsonPThe impact of a legislative ban on smoking in public places on the quality of health, pulmonary function, and inflammation of bar-workers in Scotland.Thorax200661Suppl 2ii12MenziesDNairAWilliamson PARespiratory symptoms, pulmonary function, and markers of inflammation among bar workers before and after a legislative ban on smoking in public places.JAMA20062961742174817032987Centers for Disease Control and PreventionState spotlight: Massachusetts. BRFSS Facts and News2014National Center for Chronic Disease Prevention and Health PromotionAtlanta, GA, USAde VochtFZockJPKromhoutHComparison of self-reported occupational exposure with a job exposure matrix in an international community-based study on asthma.Am J Ind Med20054743444215828067SaarinenKKarjalainenAMartikainenRPrevalence of work-aggravated symptoms in clinically established asthma.Eur Respir J20032230530912952265DosemeciMWacholderSLubin JHDoes nondifferential misclassification of exposure always bias a true effect toward the null value.Am J Epidemiol19901327467482403115Kaiser Family FoundationWorkers by occupational category: states (2010–2011), U.S. (2011)2013KFFMenlo Park, CA, USAhttp://www.statehealthfacts.org/comparemaptable.jsp?typ=2&ind=748&cat=1&sub=5&sortc=2&o=a.November 2014

Frequency of asthma-related occupational exposures as assessed by the N-JEM for the 771 jobs worked by 557 adults with asthma

Occupational exposure assessed by the N-JEMAll(n = 557)n (%)Men(n = 180)n (%)Women(n = 337)n (%)
Any exposure to asthma-related agents (HMW+LMW+IRR+PEAKS)*227 (40.8)87 (48.3)140 (37.1)
HMW agents77 (13.8)15 (8.3)62 (16.4)
    Animal antigens, mixed agricultural agents (animals)4 (0.7)04 (1.1)
    Flour and plant antigens, mixed agricultural agents (not animals)9 (1.6)6 (3.3)3 (0.8)
    Mite and insect antigens, enzymes, molds, bioaerosols9 (1.6)7 (3.9)2 (0.5)
    Latex56 (10.1)3 (1.7)53 (14.1)
    Pharmaceutical products16 (2.9)0 (0)16 (4.2)
LMW agents27 (4.8)16 (8.9)11 (2.9)
    Highly reactive agents, acrylates9 (1.6)5 (2.8)4 (1.1)
    Highly reactive agents, epoxy10 (1.8)8 (4.4)2 (0.5)
    Other highly reactive agents, such as amines, aldehydes, acids, anhydrides, chromates, curing agents, reactive gases and dyes10 (1.8)4 (2.2)6 (1.6)
    Highly reactive chemicals, isocyanates12 (2.2)10 (5.6)2 (0.5)
IRR agents151 (27.1)78 (43.3)73 (19.4)
    Cleaning agents12 (2.2)6 (3.3)6 (1.6)
    Organic dust, textile industry1 (0.2)1 (0.6)0
    Organic dust, wood or paper16 (2.9)14 (7.8)2 (0.5)
    Inorganic dusts and fumes, mining and building construction workers, and others exposed to inorganic dusts21 (3.8)19 (10.6)2 (0.5)
    Metalworking fluids12 (2.2)11 (6.1)1 (0.3)
    Combustion particles/fumes: vehicle/motor exhaust47 (8.4)25 (13.9)22 (5.8)
    High probability of exposure to ETS63 (11.3)20 (11.1)43 (11.4)
PEAKS6 (1.1)5 (2.8)1 (0.3)
Uncertain or low exposure*31 (5.6)5 (2.8)26 (6.9)
Unexposed303 (54.4)90 (50.0)213 (56.5)

As four of the 227 participants with HMW, LMW, IRR, or PEAKS exposure had another job with uncertain or low exposure, they were also counted among the 31 in this other category.

N-JEM = asthma-specific job-exposure matrix; HMW = high molecular weight; LMW = low molecular weight; IRR = irritant; PEAKS = accidental peak exposures to irritants; ETS = environmental tobacco smoke.

Frequency of severe exacerbation among 557 working adults with asthma by selected characteristics

CharacteristicN in categorySevere exacerbation
n (%)P value*
Sex0.007
    Male18039 (22)
    Female377125 (33)
Age, years
    18–2920269 (34)Reference
    30–3818451 (28)0.21
    39–4417144 (26)0.10
Highest educational level
    Some college or less350107 (31)Reference
    College degree or more20757 (28)0.51
Race0.59
    White524153 (29)
    Non-White3111 (35)
Cigarette smoking status
    Never33496 (29)Reference
    Former12337 (30)0.87
    Current10031 (31)0.76
Age at asthma onset, years0.56
    <1834598 (28)
    ≥1821266 (31)
History of allergies0.64
    No14746 (31)
    Yes410118 (29)
Asthma severity
    Mild intermittent17135 (20)Reference
    Mild persistent17059 (35)0.005
    Moderate/severe persistent21670 (32)0.01

Based on continuity-corrected χ2 statistic.

Missing for two participants.

Doctor told participant s/he had hay fever or skin allergies.

Frequency of severe exacerbation by different occupational exposures as assessed using the N-JEM for the 771 jobs worked by 557 adults with asthma

Occupational exposure assessed by the N-JEMAlln/N (%)Menn/N (%)Womenn/N (%)
Any exposure to asthma-related agents (HMW+LMW+IRR+PEAKS)*79/227 (35)27/87 (31)52/140 (37)
HMW agents23/77 (30)3/15 (20)20/62 (32)
    Animal antigens, mixed agricultural agents (animals)1/4 (25)01/4 (25)
    Flour and plant antigens, mixed agricultural agents (not animals)1/9 (11)0/6 (0)1/3 (33)
    Mite and insect antigens, enzymes, molds, bioaerosols2/9 (22)1/7 (14)1/2 (50)
    Latex20/56 (36)2/3 (67)18/53 (34)
    Pharmaceutical products5/16 (31)05/16 (31)
LMW agents11/27 (41)6/16 (38)5/11 (45)
    Highly reactive agents, acrylates2/9 (22)2/5 (40)0/4 (0)
    Highly reactive agents, epoxy4/10 (40)4/8 (50)0/2 (0)
    Other highly reactive agents, such as amines, aldehydes, acids, anhydrides, chromates, curing agents, reactive gases and dyes5/10 (50)1/4 (25)4/6 (67)
    Highly reactive chemicals, isocyanates6/12 (50)5/10 (50)1/2 (50)
IRR agents55/151 (36)25/78 (32)30/73 (41)
    Cleaning agents2/12 (17)0/6 (0)2/6 (33)
    Organic dust, textile industry0/1 (0)0/1 (0)0
    Organic dust, wood or paper3/16 (19)3/14 (21)0/2 (0)
    Inorganic dusts and fumes, mining and building construction workers, and others exposed to inorganic dusts11/21 (52)9/19 (47)2/2 (100)
    Metalworking fluids5/12 (42)4/11 (36)1/1 (100)
    Combustion particles/fumes: vehicle/motor exhaust15/47 (32)7/25 (28)8/22 (36)
    High probability of exposure to ETS30/63 (48)10/20 (5020/43 (47)
PEAKS3/6 (50)2/5 (40)1/1 (100)
Uncertain or low exposure*9/31 (29)2/5 (40)7/26 (27)
Unexposed77/303 (25)11/90 (12)66/213 (31)

N-JEM = asthma-specific job-exposure matrix; HMW = high molecular weight; LMW = low molecular weight; IRR = irritant; PEAKS = accidental peak exposures to irritants; ETS = environmental tobacco smoke.

Prevalence ratios for occupational exposure from regression models of severe exacerbation of asthma

Occupational exposure assessed by the N-JEMPR (95%CI)
Unexposed: reference category*1.00
HMW agents1.09 (0.74–1.60)
    Mite and insect antigens, enzymes, molds, bioaerosols1.27 (0.40–4.07)
    Flour and plant antigens, mixed agricultural agents (not animals)0.65 (0.1 1–3.82)
    Latex1.21 (0.81–1.82)
    Pharmaceutical products1.01 (0.48–2.13)
LMW agents1.91 (1.19–3.09)
    Highly reactive agents, acrylates1.14 (0.29–4.45)
    Highly reactive agents, epoxy2.50 (1.02–6.14)
    Other highly reactive agents, e.g., amines, aldehydes, acids, anhydrides, chromates, curing agents, reactive gases, and dyes1.93 (1.09–3.43)
    Highly reactive chemicals, isocyanates3.11 (1.56–6.20)
IRR agents1.61 (1.22–2.12)
    Cleaning agents0.79 (0.24–2.69)
    Organic dust, wood or paper1.26 (0.42–3.79)
    Inorganic dusts and fumes, mining and building construction workers, and others exposed to inorganic dusts3.61 (2.18–5.97)
    Metalworking fluids2.84 (1.38–5.84)
    Combustion particles/fumes: vehicle/motor exhaust1.52 (0.96–2.39)§
    High probability of exposure to ETS1.88 (1.36–2.59)
Accidental peak exposures to irritants3.26 (1.46–7.29)
Uncertain or low exposure1.06 (0.60–1.89)

A separate regression model was fitted for each occupational exposure. The common reference category for occupational exposures comprised those who had jobs with no exposure to asthma-related agents. Each model also included three potential confounders: sex, the oldest of three age categories (39-44 years), and asthma severity (persistent vs. intermittent).

P < 0.01.

P ≤ 0.05.

0.05 < P < 0.10.

N-JEM = asthma-specific job-exposure matrix; PR = prevalence ratio; CI = confidence interval; HMW = high molecular weight; LMW = low molecular weight; IRR = irritant; ETS = environmental tobacco smoke.

PRs from regression models of severe exacerbation of asthma with several occupational exposure subcategories*

Occupational exposure assessed by the N-JEMPR (95%CI)P value
Model A: all participants
    LMW agent
        Other highly reactive agents1.65 (0.93–2.91)0.08
    IRR agents
        Inorganic dusts2.41 (1.46–3.99)0.0006
        Metalworking fluids1.34 (0.65–2.74)0.43
        Combustion particles1.24 (0.80–1.91)0.33
        High probability of exposure to ETS1.84 (1.34–2.51)0.0001
Model B: men
    IRR agents
        Inorganic dusts2.53 (1.37–4.67)0.003
        Metalworking fluids1.51 (0.60–3.79)0.38
        Combustion particles1.33 (0.63–2.80)0.45
        High probability of exposure to ETS3.25 (1.72–6.14)0.0003
Model C: women
    LMW agent
        Other highly reactive agents1.97 (1.08–3.60)0.03
    IRR agents
        Combustion particles1.16 (0.66–2.04)0.60
        High probability of exposure to ETS1.50 (1.04–2.17)0.03

The common reference category for occupational exposures comprised those who had jobs with no exposure to asthma-related agents. Each model also included three potential confounders: sex, the oldest of three age categories (39-44 years), and asthma severity (persistent vs. intermittent).

PR = prevalence ratio; N-JEM = asthma-specific job-exposure matrix; CI = confidence interval; LMW = low molecular weight; IRR = irritant; ETS = environmental tobacco smoke.