|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Occupational Medicine (Drs. Blanc and Israel), the Arthritis Research Group (Dr. Yelin), Department of Medicine, and the Cardiovascular Research Institute (Dr. Eisner), University of California at San Francisco, San Francisco, CA.
Correspondence to: Paul Blanc, MD, MSPH, FCCP, Division of Occupational and Environmental Medicine, UCSF Box 0924, 350 Parnassus Ave, No. 609, San Francisco, CA 94143-0924; e-mail: blancp{at}itsa.ucsf.edu
| Abstract |
|---|
|
|
|---|
Objective: The goal of this study was to estimate the proportion of adult asthma cases that can be attributed to occupational factors initiating new disease onset and exacerbating preexisting disease.
Methods: We performed a cross-sectional analysis of interview data for 150 adults with asthma recruited from a random sample of family practice specialists. We ascertained the asthma and work histories of the subjects and estimated the proportion with likely work-initiated asthma and work-related asthma recrudescence.
Results: Seventy-four subjects (49%) reported adult-onset asthma while employed; an additional 25 (17%) reported recrudescence of previously quiescent childhood-onset asthma during employment. Of those with new-onset asthma while employed, 15 (10% of the study group; 95% confidence interval, 5 to 15%) were employed in occupations at increased risk of occupational asthma initiation on the basis of an independent job scoring matrix. Of those with asthma recrudescence in adulthood, seven (5% of the study group; 95% confidence interval, 2 to 8%) were employed in occupations at increased risk of exposures aggravating asthma.
Conclusions: Among adults with asthma treated in general practice settings, > 1 in 10 patients has a work history strongly suggestive of a potential relationship between exposure and disease.
Key Words: asthma disability occupational disease work-related
| Introduction |
|---|
|
|
|---|
Adult asthma is a key sentinel condition for measuring occupational morbidity.9 A wide number of workplace exposures are known to initiate asthma.10 11 Moreover, other workplace factors can aggravate existing asthma or increase the probability of disability from this condition.12 We studied occupational factors among a patient panel of adults with asthma recruited from a random sample of family practitioners. Our goal was to estimate the proportion of patients at greatest likelihood of adult-onset asthma because of occupation and those with likely work-aggravated asthma, and the associated risk of work disability.
| Materials and Methods |
|---|
|
|
|---|
We attempted to recruit all 79 eligible family practitioners. Of these, 39 (49%) participated. Most nonparticipants cited lack of time as their reason for not participating. Because some physicians who agreed to participate did not ultimately contribute any participating subjects, the final number of participating family practice specialists was 34 (43% of the eligible pool).
Subject Recruitment
Each participating family practice specialist was asked to
maintain a registry of persons aged 18 to 50 years with outpatient
visits for asthma during a prospective 4-week period, with an
additional 4 weeks if subject recruitment numbers were less than five.
Participating physicians were asked to register patients with a
diagnosis of asthma on clinical grounds. Physicians were asked to
exclude patients who also had diagnoses of chronic bronchitis or
emphysema.
We did not instruct participating physicians to restrict recruitment on the basis of patient follow-up status (new or established patients were eligible) or on the basis of other clinical factors (for example, allergic history). The recruitment methods paralleled those we have used previously to recruit adults with asthma from pulmonary and allergy subspecialty practices.13 14
Each person registered by a participating physician was contacted by mail with a description of the study. Those agreeing to participate were telephoned by the study interviewer to arrange an initial telephone survey. There were 21 persons listed who declined participation or who could not be successfully contacted. We successfully recruited 150 of 171 subjects registered (88%). The median number of subjects contributed per physician was 4 (range, 1 to 10).
Interviews and Survey Instrument
All subject interviews were performed by a single trained survey
worker. The format was a structured, close-ended, computer-assisted
telephone interview lasting approximately 45 min. We used the same
structured baseline interview schedule previously administered to 601
subjects recruited from pulmonary and allergy
specialists.14
The interview covered asthma history
(medical history, symptoms, treatment), employment history,
comorbidity, general health, functional status, demographic variables,
cigarette smoking, and psychosocial variables. Subjects were informed
that the goal of the study was to ascertain the ways in which asthma
affects people's day-to-day lives. The protocol was approved by the
University of California San Francisco Committee on Human Research.
Asthma History and Severity
We ascertained the age of asthma onset with the following
question: "How old were you when you first had symptoms or an attack
of asthma?" We defined childhood-onset asthma as self-reported
occurrence of first symptoms or attack before 18 years of age. We also
asked the following question: "Would you say that your asthma `went
away' for one year or more or that you `outgrew it' only to have it
`come back' later?" We defined recrudescent asthma as childhood
disease that "went away" for at least 1 year before returning. We
quantified asthma severity using a previously validated
disease-specific severity-of-asthma scoring system.13
14
15
The score has a possible range from 0 to 28 with higher scores
reflecting more severe disease. It incorporates respiratory symptoms,
systemic corticosteroid use, other asthma medications, and past
hospitalizations for asthma. In addition to disease-specific severity,
we also quantified asthma-specific quality of life using the "Asthma
Quality of Life Questionnaire" of Marks et al.16
General
health status was measured with the physical and mental health
component summary score measures derived from the Medical Outcomes
Study Short-Form Health Survey (SF-36).17
Occupational History
We classified subjects on the basis of their history of ever
having had part-time or more salaried employment, by which we defined
any labor force participation. We then assessed the temporal
relationship between labor force participation and symptomatic asthma.
Some subjects gave no history of any labor force participation. The
remaining subjects, all of whom had at least some history of labor
force participation, fell into one of three groups. The first group
comprised subjects whose asthma symptoms occurred before employment.
This group includes persons with childhood-onset asthma and no
remission before adulthood, as well as adult-onset asthma when the
respondent was not a salaried employee at the time, for example a
student or housewife. The second group comprised those with full- or
part-time salaried employment at the time of their initial onset of
asthma. By definition, these would be persons with adult-onset asthma.
Finally, the third group comprised those with a recrudescence of
previously asymptomatic asthma while employed. By definition, these
would be persons with initial asthma onset in childhood followed by a
remission.
We used several different methods to identify subjects with potentially work-related asthma. First, on the basis of questionnaire responses to a specific list of workplace exposures, we identified those with reported job exposures to known irritants or sensitizers. These exposures included grain or flour dust, animal fur or dander, photographic chemicals, epoxies, polyurethanes, cotton dust, pharmaceuticals, sealants, refrigerants, irritant gases, fire smoke, and metal or other dusts. We have used this approach previously to identify high-risk exposures among a similar group of adults with asthma treated by subspecialists.18 We also elicited subject-reported regular heavy physical exertion on the job (either brief or sustained), as well as subject-reported changes or extremes in temperature at work. As a second method, we used occupation as a surrogate of exposure, with broadly defined categories as surrogates of risk on an a priori basis: technicians, service, and labor (manufacturing, agriculture, and other nonwhite-collar labor).18 In addition to self-reported specific exposures or broad job category as a surrogate of exposures, we also used a third measure of work-related asthma risk. We did this by assigning exposure probabilities on the basis of a job matrix scheme previously developed by the Occupational Health Surveillance and Evaluation Unit of the California Department of Health Services.19 This scoring system has two matrices: one for likelihood of exposures initiating asthma de novo and the second for the likelihood of exposures worsening preexisting asthma. In each case, scores range from 1 (high likelihood) to 4 (low likelihood). We defined an occupational score of <2.5 as high likelihood to either initiate or exacerbate asthma, depending on the matrix used.
For those with any history of labor force participation, we asked whether the subject had ever changed job or job duties because of asthma symptoms or whether the subject left work entirely and attributed this to asthma. We elicited from respondents whether they had ever received workers' compensation coverage for treatment of their asthma.
We did not ask subjects directly whether they attributed their asthma to occupational factors because we did not wish to potentially affect responses to other work-related questions by raising this issue. We did, however, elicit subject responses broadly measuring self-perceived disease origin. The question was asked as follows: "What do you believe is the underlying cause of your asthma? Please choose the best answer: (1) environment or allergic causes, (2) genetics or vulnerability, (3) psychological or emotional factors, (4) multiple factors, or (5) unknown or mysterious factors." By ascertaining self-perceived cause in this manner, we could assess potential reporting bias associated with positive responses to the specific job-exposure questions asked elsewhere in the interview.
Statistical Analysis
Interview data were exported into a compatible format (PC-SAS)
and analyzed with software (SAS; SAS Institute; Cary, NC). We
calculated descriptive frequencies with their 95% confidence intervals
(CIs) for labor force and work exposure variables. To calculate these
95% CIs, which give a measure of the observed proportion in terms of
sampling variability, we assumed a normal distribution, except when
there were
5 observations, for which we assumed a Poisson
distribution. For the former, the 95% CIs were calculated as a
function of the observed proportion (p), 1 - p, the sample size
(150), and the two-tailed 0.05 t value of 1.98; for the
latter, we used standard table of Poisson-distributed factors given the
observed number of subjects yielding the proportion in
question.20
We tested the differences in demographic and
clinical asthma variables between those with initial onset or
recrudescent asthma while employed and all others using the
t test,
2, or, for income, the
Wilcoxon rank test.
| Results |
|---|
|
|
|---|
|
|
|
As a measure of reporting validity, we directly compared job-associated exposures based on a subject's report of irritant or sensitizer job exposure (17 among those with new-onset asthma while employed and 8 among those with recrudescence, yielding 25 in total or 17% of the study group) or defined on the basis of an independent job matrix sore (15 among those with new asthma and 7 among those with recrudescence, yielding 22 or 15% of the group) (Table 3) . Assigned exposure based on job matrix attribution was strongly associated with a subject's self-report, both among the 74 with new-onset asthma (p < 0.001) and among the 25 with recrudescent disease (p < 0.001).
Of 150 subjects, 69 (46%) attributed the sole cause of their asthma to allergic or environmental factors. This self-attribution was not associated with subject-elicited responses linking the onset or recrudescence of asthma to employment status (n = 99) as compared with all others (n = 51) (45% vs 47%; p > 0.9). Within the group of 99 subjects with work-associated asthma, there was no association between an allergic or environmental self-attribution and subject-reported irritant or sensitizer job exposures (44% vs 46%; p > 0.9).
There were only two subjects who reported leaving work altogether because of their asthma and a third subject who reported an asthma-related change in work activities at the same job. Therefore, of 130 with any history of labor force participation, 3 (2.3%) reported some degree of asthma-related work disability. An additional three subjects, none of whom reported current work disability, reported receiving workers' compensation coverage for at least some of their asthma care. Of the 99 subjects who reported that their asthma began or recrudesced while employed, 13 (13%) were no longer employed at the time of the baseline interview.
| Discussion |
|---|
|
|
|---|
These data support and amplify our previous estimate of the frequency with which adult asthma morbidity may be attributable to workplace factors. Using a similar analytic strategy, we have previously estimated the attributable fraction of work-related asthma among a cohort of adults recruited from the practices of pulmonary and allergy subspecialists.18 In that cohort, 46% of the subjects had new adult-onset asthma while employed, and 13% reported exposure to asthma-associated sensitizers or irritants. These are similar to the estimates we report here for subjects recruited from general as opposed to subspecialty practices.
Similar potential limitations would apply to both the current and previous analyses. Although the asthma diagnosis was physician dependent, the data collected on age of asthma onset and its relationship to employment status were subject dependent. Recall bias may have affected ascertainment of exposures, although the proportion of those who believed that their asthma was caused by allergies or the environment did not differ by employment or by job exposure status. Moreover, self-reported exposures were consistent with exposure assigned independently through a job risk matrix. Although our earlier estimates could have been affected by subspecialty referral patterns, the use of subjects treated by physicians in primary care practices strengthens generalization to adults followed in other primary care settings. We cannot exclude, however, the potential for selection bias from the factors that may have been associated with willingness to participate among the physicians who joined the study, if these, in turn, are also linked to a systematically different patient population in terms of occupational risk.
Our estimates are consistent with a variety of analyses that have used heterogeneous approaches to estimating etiologic fraction or population-attributable risk percentage for occupationally related asthma among adults with asthma. Although these estimates vary widely, many fall within a 7 to 15% range.21 22 23 24 25 26 27 28 29 30 31 32 33 In general, estimates of work attribution have been limited to de novo asthma onset rather than work-related asthma exacerbation or recrudescence. In one relevant community-based study, 5 of 43 cases (12%) of recrudescent adult asthma were identified as likely to be work-related.34 Work as a factor in worsening asthma is particularly relevant to asthma-related work disability. In the current study, only 2 of 130 patients with past labor force participation attributed employment cessation to asthma. This is lower than the 7% disabled proportion we observed among our subspecialist cohort.11 Part of this difference may be consistent with the greater degree of illness severity in the subspecialist group.13 We cannot exclude the possibility that some subjects who ceased employment after the onset or recrudescence of asthma did so in part because of the condition without attributing this change in work status to reasons of health.
Although work-associated asthma may be common, there appear to be few obvious demographic or clinical features distinguishing those with a temporal association between their disease and employment status. These findings underscore the critical importance of a targeted occupational history in the evaluation of all adults with asthma. Without such an assessment, the link between exposure and new-onset or worsening disease cannot be established.35
| Footnotes |
|---|
Received for publication October 2, 1998. Accepted for publication December 15, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Shofer, B. M. Haus, and W. G. Kuschner Quality of occupational history assessments in working age adults with newly diagnosed asthma. Chest, August 1, 2006; 130(2): 455 - 462. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. H. Yen, E. H. Yelin, P. Katz, M. D. Eisner, and P. D. Blanc Perceived Neighborhood Problems and Quality of Life, Physical Functioning, and Depressive Symptoms Among Adults With Asthma Am J Public Health, May 1, 2006; 96(5): 873 - 879. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-J. A Weevers, A. J van der Beek, J. R Anema, G. van der Wal, and W. van Mechelen Work-related disease in general practice: a systematic review Fam. Pract., April 1, 2005; 22(2): 197 - 204. [Abstract] [Full Text] [PDF] |
||||
![]() |
American Thoracic Society Statement: Occupational Contribution to the Burden of Airway Disease Am. J. Respir. Crit. Care Med., March 1, 2003; 167(5): 787 - 797. [Full Text] [PDF] |
||||
![]() |
M. D. Eisner and P. D. Blanc Environmental Tobacco Smoke Exposure During Travel Among Adults With Asthma* Chest, September 1, 2002; 122(3): 826 - 828. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Tarlo, K. Leung, I. Broder, F. Silverman, and D. L. Holness Asthmatic Subjects Symptomatically Worse at Work : Prevalence and Characterization Among a General Asthma Clinic Population Chest, November 1, 2000; 118(5): 1309 - 1314. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Blanc Correction to the Scientific Record Chest, August 1, 2000; 118(2): 564 - 564. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |