Buyantseva, Larisa V. MD, MS; Liss, Gary M. MD, MS; Ribeiro, Marcos MD; Manno, Michael MSc; Luce, Carol E. BSc; Tarlo, Susan M. MB, BS
Occupational factors contributing to asthma have been estimated to account for up to 10% to 15% of all adult asthma cases in industrialized countries and therefore represent a significant public health problem.1,2 Occupational asthma (OA) is often caused by sensitizers at the workplace, so it is potentially preventable when the sensitizers are known and the modes of exposure are well understood. Diisocyanates (ISO) are the most commonly recognized causes of OA in most industrialized countries.3–6 In the Province of Ontario, Canada, the government mandates employers to follow a program that includes components of primary, secondary, and tertiary prevention of ISO OA, which was recently reviewed in detail.7 There is no such mandated program in Ontario for OA caused by other agents, that is, non-diisocyanate (N-ISO) OA. Nevertheless, all cases of OA are eligible for compensation through the workers’ compensation program, administered by the Ontario Workplace Safety and Insurance Board (WSIB), formerly known as the Ontario Workers’ Compensation Board. Detailed medical records collected in the course of claim evaluation, including those for OA claimants, are maintained by the WSIB. We have previously reviewed this information for asthma claims from 1980 to 1993 and identified a trend for reduction in ISO OA, as compared with N-ISO OA, following implementation of the diisocyanate surveillance program in Ontario, despite recognition that smaller companies may not have adhered to the program, and that other factors also could have contributed to the observed trend.8 The present study covers the more recent period, 1998 to 2002, and extends our previous investigation into changes in yearly rates of sensitizer-induced ISO and N-ISO OA, as reflected by compensation claims, to identify absolute and relative changes in ISO OA in Ontario. The secondary goal of our investigation was to examine the frequency of all identified sensitizers and occupations among the workers with OA.
MATERIALS AND METHODS
The study protocol was reviewed and approved by the Research Ethics Board of the University of Toronto. The WSIB occupational disease database was searched for claims that contained primary diseases that included asthma and/or reactive airways dysfunction syndrome (RADS). The other search criterion was WSIB recorded “date of accident”, which had to be between January, 1998, and December, 2002 (five consecutive years). The “date of accident” is the date that is identified in forms submitted to the WSIB as the date of exposure (in the case of a high accidental exposure leading to symptoms) or the date of onset of work-attributed symptoms. Those claims identified according to the earlier described criteria in the database constituted the initial evaluation sample eligible for abstraction. The final data extraction was performed by a single trained reviewer who abstracted information from each claim file in the WSIB database using the questionnaire form adapted from our previously published studies.4,8–10 When available, the following information was abstracted:
1. Demographics, including date of birth, gender, smoking status, work status, and occupation;
2. History of nasal symptoms (stuffy, runny, itchy nose, and sneezing) and lower respiratory symptoms (chest tightness, shortness of breath, wheezing, cough, and sputum) and their frequency;
3. Exposure history, including offending agent and duration of exposure;
4. History of previous asthma before starting the job for which the WSIB claim had been initiated;
5. Clinical diagnosis from the most specialized physician who had assessed the patient;
6. WSIB decision made by an adjudicator and/or a WSIB medical consultant;
7. Objective tests (spirometry, peak flow studies, methacholine challenge tests, skin tests to common allergens, and skin tests to workplace substances, if available). The tests were obtained by the claimants’ health care provider and were therefore not necessarily performed by a uniform method. Interpretation of those tests was previously described.4
Abstraction was started in December of 2005 and finalized in March of 2008. During the abstraction process, the claims were pooled for each study year chronologically, 1 year at a time. Any claims made from 1999 to 2002, which had not received a compensation decision by March, 2008, were not formally abstracted for inclusion in the detailed analysis. The reviewer collected information according to the questionnaire and categorized the claimants into five defined disease groups, as outlined in the Definitions section later. Any claimant who did not clearly fit into one of the five categories or had discrepancies between the WSIB and treating physicians’ diagnosis was reviewed by two physicians experienced in occupational lung diseases (S.M.T. and G.M.L.). The final categorization was reached by consensus of the two reviewing specialists.
The following definitions for possible conditions were assigned by the research team to each claimant.
For the purposes of this study, Sensitizer-induced OA was considered in cases with the following minimum criteria: (a) development of asthma during working life or recurrence of childhood asthma during working life; (b) presence of a known or presumed occupational sensitizer at the time of asthma onset/recurrence; (c) asthma symptoms had to subjectively improve when an affected individual was off work (on weekends and/or holidays); (d) objective support of asthma: testing had to demonstrate 12% or greater improvement in FEV1 (forced expiratory volume in the first second of expiration) after bronchodilator (and improvement of at least 180 mL) or a methacholine (or equivalent) PC20 of 8 mg/mL or less. Occupational asthma was divided into two subgroups: those who were exposed to diisocyanates, the (ISO) group, and those who were exposed to other agents: non-diisocyanates, the (N-ISO) group. Because the levels of objective evidence varied, three levels of further diagnostic certainty, as developed in our prior report,11 were applied in this study to those who had the above minimum criteria for OA, as follows (these were not WSIB categorizations).
Definite OA. At least one positive work-related test (ie, work-related changes on serial peak flow readings, at least 3-fold improvement in methacholine PC20 during a period away from work vs during a work period, positive skin test to a relevant sensitizer, or positive specific challenge with a workplace sensitizer). No negative work-related tests.
Probable OA. At least one positive work-related test, but there could be a negative response to another (eg, positive serial peak flow findings but no significant improvement in methacholine response).
Possible OA. One objective work-related test performed was negative, but others were either not performed or none were performed by the time of this review.
Occupational asthma caused by irritant exposures was not included in the main analyses of sensitizer-induced OA. It was subdivided into RADS and irritant-induced occupational asthma (IIA) as follows.
Reactive Airways Dysfunction Syndrome, a subset of OA, was assigned to the claim where new-onset asthma symptoms started within 24 hours after a single accident of high concentration irritant exposure and persisted for at least 3 months. Asthma had to be confirmed by objective testing and other lung disease excluded.12
Irritant-induced occupational asthma was assigned to a claim when new-onset asthma symptoms started within 24 to 48 hours of an accidental exposure to a high level of respiratory irritant. The symptoms had to persist for at least 6 to 12 weeks. There was no documented history of previous asthma or other chronic lung disease and there was demonstrated response to methacholine challenge test (PC20 ≤ 8 mg/mL when still symptomatic) and/or a ≥12% and 180-mL FEV1 response to a bronchodilator, but negative findings when symptoms had cleared would not exclude this categorization.10
The following other groupings were not included as OA.
Asthma-like reaction (ALR) was assigned to a claim when a one-time exposure of at least moderate severity occurred to an agent at work, causing asthma-type symptoms in an individual without a previous asthma history. The onset of symptoms had to occur within 24 hours of exposure and lasted less than 6 weeks with or without objective evidence of asthma or lasted for longer periods without objective evidence of asthma.
Work-exacerbated (aggravated) asthma for this study refers to allowed claimants where exacerbation of asthma symptoms occurred in a subject with preexisting asthma (present before the exposure for which the WSIB claim had been initiated). A small minority, less than 2%, were included in this category with concurrent onset of asthma that was worse at work with no presumed or identified sensitizer.
The WSIB file of each claimant was searched to identify the medical records for the first assessment by the most specialized physician seen by the claimant for work-related asthma symptoms. For this study, physicians who saw the patient in the emergency department or in family practice offices were considered less specialized in occupational lung diseases as than the pulmonologist, allergist, or occupational disease specialist. The greatest level of specialization was assigned to specialists with specific practice interest in work-related asthma. The first visit to the most specialized physician for the OA symptoms was designated the main medical assessment. This was typically not the first visit to a physician for the OA-related symptoms. The record of the main medical assessment was the source of historical information (such as onset of symptoms, possible sensitizers or agents at work, duration of symptoms, specific symptoms reported) for abstraction. Subsequent records by the most specialized physician were used to extract results of diagnostic evaluation and clinical course of each claimant. Reports in the file from the last visit to the same most specialized physician or another were used to abstract the outcome of OA symptoms and were designated the outcome evaluation following clinical diagnosis.
Duration of time off work is a specific term used by the WSIB to identify workers’ lost earning ability, which is expressed in the number of business days lost secondary to the illness. It was determined by WSIB adjudicators to calculate compensation. We extracted from the WSIB file the date when the worker began work with the implicated exposure agent (time point A), the date of symptom-onset (time point B), the date of the main medical assessment (time point C), the date of the removal from the implicated exposure agent (time point D), and the date of the final medical assessment (time point E) to calculate various time durations from those time points to specific assessment end-points. Those included the following:
1. Exposure before the start of symptoms (duration of time from point A to point B)
2. Exposure after onset of symptoms (duration of time from point B to point D)
3. Duration of symptoms before the main assessment (duration of time from point B to point C)
4. Duration from main medical assessment to the outcome evaluation (duration of time from point C to point E)
5. Duration of symptoms before the outcome evaluation (duration of time from point B to point E)
Data analyses were performed in SAS, version 9.1 (SAS Institute, Cary, NC). The outcome variables are expressed as crude rates of prevalence. Continuous variables are expressed as mean ± SD (median values), and categorical variables are expressed as frequencies or percentages. Comparisons of continuous and categorical variables were conducted via two-tailed t test (Wilcoxon ranked sum test for nonparametric analysis) and chi-square tests (Fisher exact test when appropriate), respectively.
Among the 1825 claims originally submitted for asthma during this time, 589 were abandoned when the claimant did not respond to requests for information, and 281 claims were denied as considered unlikely to be related to work. No further information is available on these 870 claims. There were 29 claims made with an accident date between 1998 and 2002 that had not received a compensation decision by January, 2006, when the study population was extracted from the WSIB database and were not formally abstracted (when the WSIB decision was subsequently reached, 11 claims were allowed as OA and 8 as work-exacerbated/aggravated asthma, but they are not included in this report). After excluding duplicate claims, there were 893 accepted claims that underwent file abstraction. This report analyzes the file information from the chart data that were available from the 99 categorized as sensitizer-induced OA. The remaining claims were categorized as follows: 12 for irritant-induced asthma, including RADS claims, 645 for work-exacerbated asthma, 110 ALRs, and 27 with insufficient information for definitive categorization.
Among the sensitizer-induced OA claims, a minority was attributable to ISO, 37 cases (37.4%) and 62 cases (62.6%) were N-ISO, resulting in an ISO/N-ISO ratio of 0.6 and average yearly allowed claims rate of 7.4 and 12.4, respectively (Fig. 1). Occupational asthma claims were classified as definite in 50% (15 ISO and 34 N-ISO claims), probable in 10% (5 ISO and 5N-ISO claims) and possible in 40% (17 ISO and 23 N-ISO claims). In our prior report spanning 1980 to 1993, 425 claims were in the ISO group and 419 were in the N-ISO, for the ratio of 1.01 and average yearly claims rates of 30.4 and 29.9, respectively. Those changes in relative distribution between ISO and N-ISO claims between the prior and the current observation periods were statistically significant (P = 0.0094).
Comparison of demographics and other characteristics between the ISO and N-ISO OA groups (Table 1) showed no significant differences, except for a greater percentage of males in the ISO group (86% vs 60%, P = 0.01). The mean time between the onset of symptoms and the final diagnosis was similar for both groups: 2.1 ± 1.9 years (median 1.1 years [IQ range, 2.3 years] in the ISO group and 2.4 ± 2.8 years (median 1.4 years [IQ range, 1.8 years] in the N-ISO group, P = 0.5). At the time of the main medical assessment, a similar fraction of workers remained exposed to the suspected culprit agent at work: 30% of the ISO group and 31% of the N-ISO group. The groups were also similar in composition of smokers, family history of asthma, and history of childhood asthma. There was no difference in specialization of physicians who took care of the two groups. A majority of OA cases were seen by a specialist, with respiratory physicians being the most frequent provider (86% in the ISO group vs 75% in the N-ISO group). Nonspecialists (primary care physicians) rarely took sole care of the OA cases (2.7% vs 6%, respectively). The groups did not vary in severity of pulmonary dysfunction, as indicated by their pulmonary function test results at the time of the most specialized medical assessment. Both groups showed on average mild airflow limitation and a considerable percentage of both groups were on inhaled steroids at the time of testing. Neither group showed significant changes in spirometry variables between initial and final assessments or in methacholine results off work initially and at final assessment.
Objective testing did not differ significantly between the ISO and N-ISO groups (Table 1). Nevertheless, of note, the median PC20 performed when at work in the ISO group was double that of the N-ISO group. Approximately half of allowed OA cases were assessed by measures of airway hyperresponsiveness, PC20 off work and a similar proportion while exposed to the work place agent(s). While half of the ISO group had a peak flow study interpreted, only a quarter underwent this test with interpretation in the N-ISO group. Analysis of paired PC20 values obtained at work versus off work was available in 16 ISO and 20 N-ISO cases and showed statistically significant difference (P = 0.0002 in both the groups). Considerably less than half had IgE antibody testing to the work agent in both the groups. Among the entire OA sample, only 9 (9%) had an occupational laboratory challenge, confirming the diagnosis in 3. Those 6 claims with a negative occupational laboratory challenge test had other findings that supported the diagnosis of OA and led to inclusion with our diagnostic criteria for this report. Four of them had positive methacholine changes at work versus off work (2 in this group also had positive peak flow changes and one also had positive skin testing). The remaining 2 had a positive methacholine response off work and positive peak flow changes with work.
In the ISO group, there were 22 (59.4%) cases with only one potential sensitizing exposure agent identified in the file, 13 (35.2%) cases with two possible implicated exposure agents (such as amines, epoxy compounds, acrylates), and 2 cases (5.4%) with three possible implicated exposure agents. In the N-ISO group, there were 51 (82.3%), 10 (16.1%), and 1 (1.6%), respectively. The type of ISO agent was specified in the file in only 16 cases (43%): toluene diisocyanate in 5, methylene diphenyl diisocyanate in 5, and hexamethylene diisocyanate in 6. The three main agents identified in the N-ISO group were epoxy resins (13%), flour (11%), and metal fumes/dust (10%). Spray painters (41%), production workers (38%), and motor vehicle (automotive) workers (6%) comprised the three main occupations encountered in the ISO claims (Table 2). The three most frequent occupations for the N-ISO group were production worker (49%), health care worker (10%), and food-related worker (7%).
There were no significant differences between the groups in the duration of exposure at work before the onset of asthma with a mean of approximately 7 years in each group (Table 3), the duration of exposure after the onset of symptoms with a mean of approximately 1 year, for the ISO and 0.7 years for the N-ISO group, and the duration of symptoms before the main specialist assessment, with a mean of 1.8 and 2.2 years, respectively (Table 3). After acceptance of the compensation claim, there was a similar period of time to the outcome assessment, performed to determine permanent impairment, with a mean approximately 1.6 and 1.4 years, respectively (Table 3), and there was no statistically significant difference in the fraction of cases who received a noneconomic loss award (ie, a disability award, which was generally determined following that assessment) for ongoing asthma (89% of the ISO group vs 85% of the N-ISO group). Work and unemployment status at the outcome evaluation did not significantly differ between the ISO and N-ISO groups. Approximately half of each group was unemployed at the time of their final assessment and the overall mean duration of time off work was not significantly different between the ISO (1.7 ± 1.9 years) and N-ISO (1.4 ± 1.6 years) groups. A majority in both groups reported at least an improvement in respiratory symptoms at the last assessment; and although not statistically significant, a greater proportion of the ISO group reported clearing of all of their symptoms (46% of the ISO group vs 27% of the N-ISO group). The methacholine PC20 at the time of the last assessment (mean 6.3 mg/mL in the ISO group and 9.1 mg/mL in the N-ISO group) were not significantly different per group and were similar to mean results when performed off work during the diagnostic investigations, and approximately 2- to 3-fold better than the PC20 in those assessed during a working period (Table 1).
This 5-year retrospective study of data from the WSIB extends our prior observations on OA in Ontario from ISO and N-ISO causes.4,5,9 WSIB adjudicates claims for occupational disease in Ontario including OA and is responsible for determination of the degree of workers’ permanent impairment from OA resulting from any of the occupational causes—ISO and N-ISO. The Ontario Ministry of Labour established a regulation in 1983 mandating a preventive program aimed at ISO, as discussed in detail in a recent review.7 It regulates the exposure limits for ISO (primary prevention) at the workplace and requires that employers offer medical surveillance (secondary prevention) to workers who may be exposed to ISO. All claimants with suspected OA should undergo systematic evaluation to determine their fitness to continue work with ISO. Nevertheless, in 1991, the mandatory participation of workers in the medical surveillance component of the surveillance program was phased out and became voluntary. Our current review offered an opportunity to retrospectively evaluate the effectiveness of the program over the time period of this study by comparing allowed compensation claims for ISO versus N-ISO OA. To this end, the relative frequency of annual compensated (allowed) claims of ISO OA versus N-ISO OA may reflect the ongoing program effectiveness.
In the prior report that spanned a 14-year period (1980 to 1993), our group determined that the yearly relative frequency of ISO OA allowed claims increased and from 1980 and for the 3 years between 1988 and 1990 was clearly higher than that of N-ISO OA, subsequently falling below the frequency of other OA allowed claims for the last 2 years.5 Initially it was around 50% (of all allowed OA claims) in 1980 and 1987, increasing to a maximum of 64% by 1988 after introduction of the medical surveillance program, and then falling to 29% in 1992 and 35% in 1993 (Fig. 1). The fall in allowed ISO claims in the last 2 years of the previous review period could have been due to effectiveness of the primary and secondary preventive measures of the program; however, other explanations are possible. Our current findings show that this drop in the relative yearly frequency of ISO OA (as a proportion of all OA claims) persisted as did a fall in the absolute number of ISO OA allowed claims (Fig. 1). This would support a hypothesis for a positive direct impact of the surveillance program on the proportion of ISO OA claim rates relative to all OA allowed claims. Another possible explanation may have been that there was reduced case identification with the change to voluntary participation in medical surveillance. Nevertheless, we did not find a longer time to diagnosis in the present cohort as might have been expected with this explanation (but the time to diagnosis has fallen over time in both ISO OA and N-ISO OA, likely reflecting changes in clinical practice).
Nevertheless, no less remarkable was the finding of decreased numbers of N-ISO claims during the 5-year study interval as compared with the earlier time period. Possible reasons for reduction in all OA allowed claims might include heightened OA awareness among workers, as well as among medical health care professionals. If this occurred, it potentially may have been a collateral benefit from the isocyanate surveillance program, or from other factors such as general worker and physician education programs perhaps enhanced by the publication and dissemination of guidelines and review articles on the recognition and diagnosis of OA in Canada.13 Other possible factors contributing to lower rates of allowed claims might include better occupational hygiene measures in workplaces, or potentially a reduction in jobs with exposure to sensitizers (eg, with a loss of manufacturing jobs in Southern Ontario following the 1994 North American Free Trade Agreement). It is also possible that the compensation system may have changed their criteria for acceptance of claims, but there has been no identified policy change to support this possibility. Unfortunately, we do not have comparable figures for rejected and abandoned claims during the previous period.
Comparisons of objective testing showed that performance of the methacholine challenge test at work (ie, during a working period) increased from 31% in 1980 to 1986 and 20% in 1987 to 1993 reports4,5 to 51% in the present (1998 to 2002) report. Nevertheless, performance of methacholine challenge test off work did not change considerably (58% vs 56% in the 1980 to 1986 study period and 56% in the 1987 to 1993 study period).4,5 This is consistent with the earlier time to assessment with specialists while patients were still at work, in keeping with Canadian guidelines.13 Also suggesting earlier suspicion and investigation of OA, in the current time period the mean time to the main medical assessment after the onset of symptoms was 1.8 years in the ISO group and 2.2 years in the N-ISO group, compared with 2.0 and 3.0 years, respectively, in our previous review of 1980 to 1993.5
Finally, it is possible that rather than a true reduction in OA there could be reduced recognition of work-relatedness, or reluctance to apply for compensation. Nevertheless, this is not suggested from the annual numbers of claims accepted for work-exacerbated asthma, which have risen over this time.4,14
In our earlier study, we offered several plausible explanations for the finding of a drop in ISO OA relative frequency in the last 2 years of the study.5 One explanation had to do with a possible transient left shift of the incidence curve due to earlier findings of cases. If true, it would be expected that the relative percentage of ISO out of the total OA claims would drift back toward 50% over subsequent years. Another plausible explanation for the relative decline in numbers of allowed ISO OA claims during the last 2 years of the 1992 to 1993 period was that it might have represented random yearly fluctuation—a statistical outlier. Our current report shows that this trend continued with overall relative frequency of 37.4% for ISO OA during the presently reported period (1998 to 2002), making it unlikely that the previous reduction was due to either the early case finding or the random data fluctuation.
In addition, the severity of ISO OA for that study period was milder and time to a specialist visit was shorter than that due to other causes.5 Our current investigation showed that the time to see a specialist (main medical assessment), the time to diagnosis, and severity of OA (Table 1) were not significantly different in the ISO and N-ISO groups. Nevertheless, claim numbers in both groups in the present review period were smaller, which may account for the lack of statistical significance. Of interest was the apparent reduction in time to the main medical assessment (1.1 ± 1.1 year) as compared with that in our prior report (mean of 2 to 3 years).5 Nevertheless, the 11 excluded cases without decision at the time of abstraction could have created this apparent shortening because the delay in decision likely reflected more complex cases. Those cases have recently received WSIB decisions and information obtained in this small group for the time to the main medical assessment shows it to be much longer, 4.5 ± 2.3 (median 4.6) years, supporting the likelihood of this explanation.
Among the ISO claimants, the proportion becoming asymptomatic at follow-up was considerably greater than that among the N-ISO claimants (although not reaching statistical significance). This finding and the somewhat greater initial PC20 among the ISO cases are also consistent with our previously published reports, indicating possible positive influence of the medical surveillance program with recognition of OA at more reversible stage.
Finally it is possible that the observed relative reduction in OA from ISO may be associated with changes in the formulations of ISO and the relative mix of monomers and oligomers, making them less likely to cause sensitizer-induced OA. This point may be particularly important in explaining why the relative frequency of ISO OA has remained below N-ISO since the early 1990s.
There is some evidence from England to suggest that relative and absolute number of ISO cases declined over recent time6,15 and a relative reduction in isocyanate OA has also been reported from Quebec during 1998 to 2001 in the absence of a medical surveillance program during those years.16
We acknowledge limitations of this report that include the retrospective nature of the study, lack of complete information on occupations, and exposures in the compensation file (potentially leading to misclassification of causative agent), lack of control over the WSIB adjudication practice, lack of denominator (overall number of subjects at risk or regional population), and inconsistent application of objective testing to the claimants. The latter limitation explains why we had to classify the cases into definite, possible, and probable subcategories. In addition, some of these factors potentially could have changed over time and might have contributed to changes in yearly accepted claims for isocyanate OA. Furthermore, we recognize that yearly incidence of compensation claims may be far lower than the true incidence of OA in respective populations.17 Nevertheless, the strengths of our study included the application of consistent abstraction methodology across our retrospective studies and the observation over a relatively long-time period. In view of the limited feasibility of randomized trials for preventing OA, this report provides some support for the value of governmental regulations.
In conclusion, the study suggests that there has been a reduction in absolute number of ISO and N-ISO OA allowed claims, with a somewhat greater relative decline of ISO OA claims. Our current observations support a notion that the surveillance program aimed at reducing ISO OA frequency had a positive impact. The overall reduction in frequency of compensated OA in Ontario is likely to have several contributing causes that could not be determined from this study.
The study was funded by a grant from the Research Advisory Council of the Ontario Workplace Safety and Insurance Board. We thank Ms Lee Vernich, Dalla Lana School of Public Health, University of Toronto, for additional data management and statistical support.
1. Blanc PD, Toren K. How much adult asthma can be attributed to occupational factors? Am J Med. 1999;107:580–587.
2. Balmes J, Becklake M, Blanc P, et al.. American Thoracic Society Statement: Occupational contribution to the burden of airway disease. Am J Respir Crit Care Med. 2003;167:787–797.
3. Meredith SK, Taylor VM, McDonald JC. Occupational respiratory disease in the United Kingdom 1989: a report to the British Thoracic Society and the Society of Occupational Medicine by the SWORD project group. Br J Ind Med. 1991;48:292–298.
4. Tarlo SM, Liss G, Corey P, Broder I. A workers’ compensation claim population for occupational asthma. Comparison of subgroups. Chest. 1995;107:634–641.
5. Tarlo SM, Liss GM, Yeung KS. Changes in rates and severity of compensation claims for asthma due to diisocyanates: a possible effect of medical surveillance measures. Occup Environ Med. 2002;59:58–62.
6. Bakerly ND, Moore VC, Vellore AD, Jaakkola MS, Robertson AS, Burge PS. Fifteen-year trends in occupational asthma: data from the Shield surveillance scheme. Occup Med (Lond). 2008;58:169–174.
7. Tarlo SM. Prevention of occupational asthma in Ontario. Can J Physiol Pharmacol. 2007;85:167–172.
8. Tarlo SM, Liss GM. Diisocyanate-induced asthma: diagnosis, prognosis, and effects of medical surveillance measures. Appl Occup Environ Hyg. 2002;17:902–908.
9. Tarlo SM, Banks D, Liss G, Broder I. Outcome determinants for isocyanate induced occupational asthma among compensation claimants. Occup Environ Med. 1997;54:756–761.
10. Chatkin JM, Tarlo SM, Liss G, Banks D, Broder I. The outcome of asthma related to workplace irritant exposures: a comparison of irritant-induced asthma and irritant aggravation of asthma. Chest. 1999;116:1780–1785.
11. Santos MS, Jung H, Peyrovi J, Lou W, Liss GM, Tarlo SM. Occupational asthma and work-exacerbated asthma: factors associated with time to diagnostic steps. Chest. 2007;131:1768–1775.
12. Brooks SM, Weiss MA, Bernstein IL. Reactive airways dysfunction syndrome (RADS). Persistent asthma syndrome after high level irritant exposures. Chest. 1985;88:376–384.
13. Tarlo SM, Boulet LP, Cartier A, et al.. Canadian Thoracic Society guidelines for occupational asthma. Can Respir J. 1998;5:289–300.
14. Buyantseva LV, Liss GM, Luce CE, Tarlo SM. Work-aggravated and occupational asthma in Ontario: relative frequency and exposures. Proc Am Thoracic Society, 2007;4:A804.
15. McDonald JC, Keynes HL, Meredith SK. Reported incidence of occupational asthma in the United Kingdom, 1989–97. Occup Environ Med. 2000;57:823–829.
16. Bernstein ILKH, Blanc P, Chan-Yeung M, Malo JL. Bernstein IL, Chan-Yeung M MJ-L, Bernstein D, eds. Medicolegal aspects, compensation aspects and evaluation of impairment/disability. In: Asthma in the Workplace and Related Conditions. 3rd ed. New York, NY: Taylor and Francis Group; 2006:319–335.
17. Cherry N, Beach J, Burstyn I, Fan X, Guo N, Kapur N. Data linkage to estimate the extent and distribution of occupational disease: new onset adult asthma in Alberta, Canada. Am J Ind Med. 2009;52:831–840.