Health surveillance in milling., baking and other food manufacturing operations five years' experience

Transcription

1 Occup. Med. Vol. 49, No. 3, pp , 999 Copyright O 999 Upplncott Williams & WBktns for SOM Printed In Great Britain. Al rights reserved /99 INTRODUCTION Health surveillance in milling., baking and other food manufacturing operations five years' experience T. A. Smith and J. Pattern Ranks Hovis McDougaU Limited, King Edward House, 7/3 King Edward Court, Windsor, Berkshire SL4 TJ, UK The objective of this study was to describe the incidence of allergic respiratory disease and its outcome in terms of symptoms and jobs,.across different flour-using industries. It uses the findings of a health surveillance programme in a large food organization over a five-year period. The population under surveillance consisted of 3,45 employees with exposure to ingredient dusts, of whom 4 were in flour milling,,65 in bread baking, 55 in cake baking and 85 in other flour-using operations. A total of 66 employees with either asthma or rhinitis symptoms attributable to sensitization to allergens in the workplace were identified. The majority of these (48/66) had become symptomatic prior to the commencement of the hearth surveillance programme in 993. The incidence rates (per million employees per year) for those who developed symptoms between 993 and 997 were 55 for flour milling,,94 for bread baking, for cake baking and 35 for other flour-using operations. The agent believed to be responsible for symptoms was most commonly grain dust in flour millers and fungal amytase in bread bakers. Wheat flour appeared to have a weaker sensitizing potential than these other two substances. In terms of outcome, at follow-up 8% of symptomatically sensitized employees had left the company. Two of the ex-employees retired through ill health due to occupational asthma. Of those still in employment, 63% described an improvement in symptoms, 3% were unchanged and 4% were worse than when first diagnosed. Over half the cases still in employment were continuing to work in the same job as at the time of diagnosis. Key words: Amylase; asthma; flour; rhinitis. Occup. Med. Vol. 49, 47-53, 999 Since 989 there has been a statutory requirement for UK employers to perform health surveillance on employees who are exposed to substances hazardous to health in the workplace. Employees in food manufacturing industries can be exposed to a number of potential respiratory sensitizers which are included within the definition of substances hazardous to health. In the flour manufacturing and flour-use industries the possible sensitizing agents include dust from the cleaning of grain, cereal allergens from a variety of different flours Correspondence and reprint requests to: Dr T. A. Smith, Ranks Hovls McDougall Limited, King Edward House, 7/3 King Edward Court, Windsor, Berkshire SL4 TJ, UK. Received 3 July 998; accepted in final form 7 October 998 and enzymes such as amylase and hemicellulase from fungal sources. Three issues concerning sensitization in these industries have been the source of considerable speculation and debate but remain largely unresolved: The relationship between exposure and sensitization. In particular, the relative importance of the different allergens in situations where exposure to more than one arises. The relative risk of sensitization in different manufacturing sectors. The outcome of those with allergic rhinitis or asthma in terms of their clinical progression and their socioeconomic circumstances. This paper describes the health surveillance findings, over a five-year period, in a large organization whose Downloaded from by guest on October 8

2 48 Occup. Mod. VW. 49, 999 interests include flour milling, bread baking, cake baking and diverse other manufacturing processes which use flour. It is important to point out that the data comes from a health surveillance programme and not a prospective study. Although the epidemiology associated with this type of work can never be perfect, it does give a useful insight into certain aspects of sensitization in the flour-using industries. MATERIALS AND METHODS The health surveillance groups The employees who were included in the health surveillance programme were those who, in the course of their normal work, could be exposed to dusts from any powdered ingredient. Identification of individual employees for inclusion was the responsibility of site management although guidance was given by occupational health professionals regarding at-risk groups. The types of substance exposure varied considerably between the different industrial sectors. These exposures and the approximate average number of employees in each sector over the five years of the study are detailed in Table. During the period of surveillance, new starters were screened prior to appointment. Those with a history of asthma over the previous five years were not taken on into jobs where there was any possible dust exposure. However, history of atopic conditions alone did not present a bar to employment. Health surveillance procedures Each employee was screened on appointment and then on an annual basis by questionnaire administered by an occupational health nurse. Those who gave a history of symptoms which arose solely at work or were worse at work were referred to one of four occupational physicians. The subsequent consultation consisted of a structured interview and skin prick testing. Where considered appropriate, serial peak flow readings were also performed. The interview covered occupational history in relation to ingredient exposure and enquiry regarding work-related or allergic respiratory symptoms. In particular, employees were asked about symptoms relating to possible rhinitis, conjunctivitis and asthma. The interview procedure covered the following points: episodes of wheezing or chest tightness; symptoms of shortness of breath or cough; symptoms of sneezing, runny nose or blocked nose; time of onset of symptoms; duration of symptoms; relationship of symptoms to work, i.e., whether worse at work or at home, whether only arising exclusively at work, whether connected with particular task or ingredient; if symptoms were associated with a particular task or use of a particular ingredient, the temporal relationship with dust exposure and the duration after exposure ceased; if symptoms were relatively persistent, if there was an improvement away from work during rest periods or on holiday and whether treatment had been taken for symptoms. The prime purpose of the consultation was to probe the relationship between exposure to dust and the pattern of symptoms in some detail. We believe that a structured interview, backed up by positive skin prick tests, is the most appropriate method to distinguish asthma or rhinitis due to ingredient sensitization from symptoms due to other causes. However, it has to be accepted that this form of data collection, based essentially on clinical acumen, can introduce bias. In order to minimize this, the participating occupational physicians used broadly agreed diagnostic criteria. In all cases, the diagnosis of occupational asthma or rhinitis required a history of work-related symptoms which were more than transient plus a positive skin prick test to at least one work-related allergen. Skin prick testing was performed against a standard set of solutions, namely: saline and histamine controls; commercially available common environmental allergens, i.e., house dust mite, cat fur and mixed grass pollens (Biodiagnostics Limited); work-related allergens at a concentration of mg.mt, i.e., wheat flour, soya flour, rice flour and Aspergillus-derived amylase (prepared from base ingredients by the National Heart and Lung Institute). The wheat flour used to make up the skin prick test solution was a standard bread-making variety with no fungal amylase added during milling. Skin prick tests were read after minutes and were considered to be positive if there was a wheal of at least 3 mm diameter. For the purpose of this study, atopy was defined by positive skin prick tests to one or more of the common environmental allergens, i.e., house dust mite, cat fur or mixed grass pollens. Downloaded from by guest on October 8 Table. Average numbers in each sector and the ingredient dusts to which they are exposed Manufacturing sector Flour milling Bread baking Cake baking Others Average number under surveillance 4, Ingredient dust exposures Grain, wheat flour Wheat flour, soya flour, rice flour, gluten, bread Improvers (containing fungal amylase and hemlcellulase), baking powder Wheat flour, soya flour, rice flour, sugar, powdered egg, baking powder Wheat flour, sugar, powdered egg, spices

3 T. A. Smith and J. Patton: Health surveillance In rnllng, baking and other food manufacturing 49 Follow-up of symptom progression and job outcome All employees who were in continuing employment were interviewed by occupational health nursing staff to determine whedier dieir symptoms had improved, remained the same or worsened since first diagnosed. They were also asked about their current job circumstances and those at the time of diagnosis. Personnel records of leavers were scrutinized to determine the reason why the individual had left employment. RESULTS Over the five year period of the health surveillance programme, 66 employees were identified who had symptoms which were considered attributable to ingredient sensitization. Of diese, 3 had asthmatic symptoms, with or without concomitant rhinitis and 36 had rhinitis alone. Agents responsible for allergic symptoms Although it can be difficult to ascribe allergic symptoms to a specific agent, particularly when exposure to more than one potential sensitizer occurs, it is usually possible to identify the most likely candidate from the history. Table shows die numbers of symptomatic employees in each manufacturing sector and the agent believed to be responsible for the symptoms. Presence or absence of specific IgE is valuable in providing confirmation of the likely allergen, as a back-up to die history. It was possible to identify specific IgE by skin prick testing in 6 of the 66 cases where either wheat flour, fungal amylase, hemicellulase or egg albumin were diought to be responsible for symptoms. No IgE confirmation was possible in the three cases due to grain or die single case due to mushroom spores, owing to lack of skin prick test solutions. Most of die symptomatic individuals (55/66) worked in large scale bread baking (i.e., plant bakeries). It is interesting to note tiiat die majority of these bread baking employees had allergic symptoms which were attributable to fungal amylase rather dian wheat flour. Fungal amylase is used as an additive bodi in bread improvers and in certain flours. However, the concentration in bread improvers is around times that in flour and so bread improvers constitute die most important source of exposure. Bread improvers are handled in a relatively small area of the bakery. Thus only a limited number of die workforce will be exposed to significant concentrations of airborne fungal amylase in die course of tiieir normal work. In contrast to fungal amylase, a much greater proportion of those witii ingredient dust exposure will be exposed to wheat flour. It is interesting to note that relatively few workers had symptomatic sensitization to flour, as compared widi fungal amylase. In addition, die small number of wheat flour cases was mirrored in all die manufacturing sectors. In particular it is worth commenting on die absence of symptomatically sensitized individuals in die cake baking group, despite significant flour dust exposures. Furthermore, amongst the flour millers, grain dust featured more prominendy than flour as die causative agent and was responsible for diree out of die four cases. Risk of symptomatic sensitization Aldiough 66 sensitized employees were identified during die five years of surveillance, a majority of these (48/66) had developed symptoms prior to die start of die programme in 993 and only 8 became symptomatic during die surveillance period. Of diese 8 new cases, die number which arose in each year since die start of 993 is shown in Table 3. Typically tiiere have been diree or four new cases of eitiier rhinitis or astiima in each of die five years. Again it is apparent diat diey were largely confined to bread baking operations. All die new cases of symptomatic sensitization in bread baking employees over die last five years were attributable to fungal amylase. The single new cases of sensitization in flour milling and odier food manufacture were due to grain dust and egg albumin, respectively. The 8 employees whose symptoms began after die onset of die healtii surveillance programme constitute an interesting subgroup, as we believe diat tiiey probably represent die sum total of all new cases of sensitization arising during die period. Over tiiis five-year period, no odier employees left die company on ill-healdi grounds widi occupational astiima and tiiere were no claims for astiima or rhinitis against die company from current or ex-employees who were not known to be sensitized. Although diese facts do not guarantee diat all sensitized employees were identified, it seems unlikely diat any were missed as a large majority of die employees are members of die pension scheme and would be likely to Downloaded from by guest on October 8 Table. Summary of all symptomatic cases Identified up to the end of 997 Activity Condition No. AgBnts responsible Rour milling Bread baking Cake baking Other activities Occupational asthma Allergic rhinitis Occupational asthma Allergic rhinitis Occupational asthma Allergic rhinitis Occupational asthma Allergic rhinitis Grain (), fungal amylase/flour () Grain () Fungal amylase (), hemlcellulase () Fungai amylase (3), flour () Flour (3), egg albumen (), mushroom spores () Gluten ()

4 5 Occup. Med. \tol. 49, 999 Table 3. New cases of symptomatic sensitization arising In each year since 993 Year Flour milling asthma rhinitis Bread baking asthma rhinitis asthma 3 rhinitis asthma 4 rhinitis asthma rhinitis asthma rhinitis Cake baking Other food manufacture asthma rhinitis seek an ill-health retirement pension if they were to develop asthma. Also, the trade unions in each of the manufacturing sectors have been active in seeking compensation for any employee who may develop possible occupational disease, with claims inevitably arising from a large number of those who were identified by the health surveillance programme. Given that the denominator population in the separate manufacturing sectors was known, it was possible to calculate the incidence of symptomatic sensitization for each of them. Taking only the cases whose symptoms began in the last five years, the incidence rate per million employees per year was,94 for bread baking, 5 for flour milling, 35 for other flour-using operations and for cake baking. Although the numbers were small, bread baking appeared to carry a five- to -fold greater risk than the other manufacturing sectors. Comparison of the incidence of allergic symptoms between bread baking and the other sectors using a x test showed a statistically significant difference (p -.). It is important to note that these incidence rates relate to the overall risk of developing symptomatic sensitization, i.e., both asthma and rhinitis. If the cases of rhinitis are excluded in order to estimate the incidence for asthma only, the equivalent rates for incidence of asthma alone are 485 for bread baking, 5 for flour milling, 35 for other flour-using operations and for cake baking. However, these rates need to be interpreted with caution since they are based on very small numbers, i.e., four in bread baking, one in flour milling, one in other flour-using operations and in cake baking. Prevalence of specific IgE antibodies As stated above, it was possible to identify specific IgE by skin prick testing in 6 of the 66 cases. The 55 symptomatic cases which arose in the bread baking group represent a large proportion of the total and have therefore been examined in greater detail regarding skin prick test findings. The prevalence of atopy in this group was around twice that expected in the general population (i.e., 65%). The prevalence of specific IgE to the workrelated allergens was greatest for fungal amylase (54/55 = 98%), followed by soya flour (3/55 = 4%), with Table 4. Latent period for development of symptoms Time (years) Mean Subgroup with onset of symptoms before 993 Subgroup with onset of symptoms after 993 No. No. No yrs yrs 7 39 Whole group yrs wheat flour least (/55 = 36%). It is important to note that although all of these individuals demonstrated allergic symptoms, presence of specific IgE was not necessarily associated with symptoms to that particular allergen. Nevertheless the prevalence of specific IgE probably gives some idea of the relative risk of sensitization to each of the allergens. Latent period Since both the date when ingredient dust exposure started and the date of onset of symptoms were known, it was possible to calculate the latent period for development of symptoms. For the whole group the mean time from start of exposure to onset of symptoms was 7. years. Separating those whose onset of symptoms developed before and after the beginning of the programme in 993, the mean latent period was shorter for those prior to 993 (i.e., 6.6 years) than those after (i.e., 8.8 years). However, statistical comparison of these two groups using a r-test showed that this difference was not significant (p =.). The distribution of the latent periods of the two groups is shown in Table 4. It is also worthy of comment that only three cases developed symptoms within the first year of exposure. All of these were in the group whose symptoms commenced prior to the start of the health surveillance programme. Symptom progression and job outcome Some information about ongoing health and job circumstances was available for all 66 cases. Twelve (8%) had left the company: two through ill health retirement as a result of their occupational asthma, three through voluntary redundancy and seven for other reasons (unconnected with their medical condition). Fifty-four employees (8%) were in continuing employment with the company. Data on current health and job circumstances were available for the 54 who were still in employment but it was not possible to determine the current health status of the ex-employees. Of the 54 who could be accounted for, 34 (63%) reported an improvement in their symptoms since the time of first diagnosis, 7 (3%) were unchanged and two (4%) had worsened. Both rhinitis and asthma cases showed an improvement in around two-thirds of all cases (see Table 5). However, the percentage of both rhinitis and asthma cases which 3 Downloaded from by guest on October 8

5 T. A. Smith and J. Patton: Health survellance In mffllng, bawng and other food manufacturing 5 Table 5. Symptom progression Diagnosis before 993 Diagnosis after 993 All cases Outcome Improved Unchanged Worse Total Rhinitis Asthma Rhinitis Asthma Rhinitis Asthma (58%) 7 (37%) (5%) (59%) 6 (35%) (6%) 9(69%) 4(%) (63%) 4(67%) 4(3%) (34%) 6(9%) (3%) (5%) had improved was greater for those cases diagnosed after 993 (rhinitis = 69%, asthma = %). The job outcome for those still in employment is shown in Table 6. In over half the cases, it was possible for the individual to continue working in their previous job. Perhaps it is somewhat surprising that this seemed to apply both to asthma and rhinitis cases, since ongoing asthmatic symptoms are quite disabling. For many of the employees, there were strong financial pressures to continue in the same job. From the medical standpoint, the desires of the individual were balanced against medical consequences but in some cases it was possible to keep employees in the same job providing they wore suitable respiratory protection. Mostly these were cases where symptoms were of long-standing (i.e., those whose onset pre-dated the start of the surveillance programme). For the cases of relatively recent onset which were diagnosed after 993, the individual's job circumstances were invariably changed in order to reduce or eliminate exposure to the allergen. This factor could, at least in part, account also for the better outcome in terms of symptoms for this subgroup. It is perhaps not surprising that when job outcome of those still in employment was examined against symptom progression, there was some evidence of a link between working elsewhere in the factory and improvement of symptoms (see Table 7). It is probably equally predictable that the only two cases who said their symptoms had worsened had not changed their job circumstances. However, somewhat against expectations, a substantial proportion, i.e., /54 (%), described an Table 6. Job outcome Individuals In continuing employment Job circumstances Rhinitis Asthma (n = 3) (n = ) Doing same job 8 (56%) (5%) Doing different Job In same area 4(3%) (5%) Working elsewhere in factory (3 %) (45%) improvement in symptoms even when they continued in the same job. DISCUSSION The findings from this study challenge the widely held view that wheat flour is the principal cause of allergic respiratory illness in the baking industry. Three points are of particular relevance to the argument Fungal amylase and not wheat flour is the agent which was most commonly identified as being responsible for allergic respiratory symptoms in the bread baking group. Specific IgE to fungal amylase was found in a larger number of bakers with allergic symptoms than was IgE to wheat flour allergens. The incidence of allergic symptoms was significantly greater in bread baking than in the other manufacturing sectors where wheat flour exposure is encountered. It is worth considering the possible validity of each of these points in turn. The first of the findings, i.e., fungal amylase is a more common cause of allergic symptoms than flour, has a fairly large subjective element as it is dependent upon the history of relationships between symptoms and exposure to particular ingredients. In practice, there are some cases where there is little doubt from the history that symptoms have a clear link with exposure to bread improvers and not to flour, or vice versa. However, in others it can be difficult to distinguish whether bread improver or flour exposure is responsible. In these cases, the presence or absence of specific IgE is an important factor in differentiation. Taking the second point, the greater prevalence of specific IgE to fungal amylase than to wheat flour, if taken in isolation, could simply be an index of the degree of exposure to the two substances. However, it is worth remembering that in this study the positive skin prick tests are found in symptomatic individuals. The fact that Downloaded from by guest on October 8 Table 7. Job outcome in relation to symptom progression Improved (34/54 = 63%; Unchanged (8/54 = 33%) Worse (/54 = 4%) Job circumstances Rhinitis Asthma Rhinitis Asthma Rhinitis Asthma Doing same job Doing different job in same area Working elsewhere in factory

6 5 Occup. Mod. Vol. 49, 999 specific IgE to fungal amylase is present alongside a history of symptoms associated with exposure to bread improver gives some credence to a possible causal relationship. In the past there has been a readiness to diagnose baker's asthma on the basis of history alone, even when specific IgE to wheat flour cannot be demonstrated. Gannon and Burge pointed out that measurement of specific IgE, in helping to confirm the diagnosis of occupational asthma, is under-utilized outside specialist centres. This may be a reflection of a lack of availability of test materials. Many physicians still do not routinely test for IgE to fungal amylase in suspected cases of baker's asthma but nevertheless diagnose occupational asthma to flour when serology for flour is negative. In some of these cases without IgE to wheat flour, it is possible that the agent responsible for symptoms may be fungal amylase. In this study we were able to identify specific IgE to either enzymes or wheat in all of the bread bakers with possible allergic symptoms. We would suggest that a diagnosis of allergic asthma or rhinitis in bread bakers who do not have IgE antibodies to wheat flour, soya or enzymes used in the baking process, is otherwise difficult to sustain. The differing incidence of allergic symptoms in each of the manufacturing sectors is an important point The significantly higher incidence in the bread baking group suggests that their risk of sensitization is greater than the other sectors. If this finding cannot be explained purely by chance, it must be either a function of variation in exposure levels or in sensitizing potential of the substances encountered. Although detail of exposure is outside the scope of this work, total inhalable dust levels and particle size distribution are broadly similar across the sectors. The main difference in risk would therefore seem to relate to the sensitizing potential of the agents met by each group. Whilst wheat flour may give a background level of risk across all the sectors, fungal amylase would seem to be responsible for the higher incidence of allergic problems in bread bakers. It is difficult to find published data with which to compare the overall risk of symptomatic sensitization. However, two other UK studies * 3 have estimated the incidence of baker's asthma at 445 and 334 per million employees per year. Superficially the incidence of baker's asthma as determined from our work appears broadly similar (i.e., 485 per million employees per year). However, the denominator populations are not directly comparable. The other two studies estimate incidence in all bakery workers, using demographic data. The demographic figures describing bakery workers will include a number who never come into contact with any ingredients during the course of their work, e.g., despatch, distribution and office personnel. In our study the denominator is confined just to employees who have exposure to ingredient dust and hence will be a smaller number. If the denominator was instead taken as the total employed in bread baking in the organization (i.e., 7,5), then the incidence would be 7 per million per year. Although this rate is lower than those from the other studies, there may not be a true difference as there was only a small number of new cases of asthma (four) in bread bakers over the five-year period. Alternatively the difference could be explained by different diagnostic criteria, since those in our study are quite tight, i.e., a clear history of consistent exposure-related symptoms plus a positive skin prick test However, it is also possible that measures to reduce dust exposure in the bakeries included in our study could have contributed to a genuinely lower rate. In the flour milling sector, the numbers with symptoms and the population exposed are both rather small, making it difficult to draw any firm conclusions. Nevertheless in 8% (4/5) of those identified with symptoms due to sensitization, the allergen was grain dust rather than flour. This suggests that the sensitizing potential of grain dust is probably greater than that of flour. Any research work concerning flour milling which aims to look at the risk of sensitization specifically to flour, should take this possible factor into consideration. The average latent period between onset of exposure and development of symptoms in these flour-using industries is quite long (overall mean = 7. years). Although it was not statistically significant, the latent period for those developing symptoms after 993 was longer than those developing symptoms before (i.e., 8.8 years vs. 6.6 years). If this is not a chance finding, it might reflect a change in exposure patterns, rather than a difference in the sensitizing potential of the allergens encountered. There is a belief that short term high exposures may be more important in producing sensitization than continuous lower levels. 4 The fact that effective dust control measures have been steadily introduced during the 99s may have lessened the risk of meeting a peak exposure event of sufficient magnitude to induce sensitization and hence be reflected in a longer mean latent period. It is perhaps significant that none of the cases identified in the last five years developed symptoms within the first year of employment. Apart from the issues relating to types of exposure which might give rise to sensitization, this has implications for the schedule of health surveillance. In the UK the current recommendation is to screen employees initially and then at intervals of six weeks, weeks and annually thereafter. 5 Our findings raise the question of whether the six and week screening procedures are necessary in industries with this type of ingredient dust exposure. Several authors have described health and job outcomes for occupational asthma, both in single industries and in the general population. 6 " 6 They describe variable rates of progression of symptoms with some individuals recovering, particularly if removed from exposure at an early stage. 4 In the largest study covering cases due to a variety of agents, 6 8% had recovered at follow-up. Nevertheless there is also evidence from other studies that some of those sensitized experience worsening symptoms even after cessation of exposure. 7 Typically around a half of those affected have to leave their job. 5 The progression of symptoms which we describe is generally comparable with the other studies. However, a larger proportion of the group in our study (8%) were Downloaded from by guest on October 8

7 T. A. Smith and J. Pattern: Health survebance In millng, baking and other food manufacturing 53 still with their original employer. Moreover, % were able to continue in the same job, albeit with controls to limit exposure to the allergen. Perhaps the better outcome which we describe may be related to the fact that these cases came from a large employer with an occupational health provision. Certainly other authors have suggested that access to occupational health services may affect the outcome of occupational asthma. 6 However, another explanation may lie in the nature of the allergen. It is possible that IgE-mediated respiratory disease due to high molecular weight allergens may have a more benign course than disease due to low molecular weight compounds. Irrespective of the reasons behind its natural history, it is our conclusion that baker's asthma does not necessarily have a pessimistic prognosis. REFERENCES. Health and Safety Executive. Regulation, General COSHH ACOP, Control of Substances Hazardous to Health Regulations. HSE, 994: Gannon PF, Burge PS. The SHIELD scheme in the West Midlands Region, United Kingdom. Br J Ind Med 993; 5: Meredith S. Reported incidence of occupational asthma in the United Kingdom, J Epidemiol Comm Health 993; 47: Morris L. Respiratory sensitisers. Controlling peak exposures. Tax Sub Bull 994; 4:. 5. Health and Safety Executive. Guidance Note MS5: Medical Aspects of Occupational Asthma. HMSO, 99: Burge PS. Occupational asthma in electronics workers caused by colophony fumes: follow-up of affected workers. Thorax 98; 37: Paggiaro PL, Loi AM, Rossi O, et al. Follow-up study of patients with respiratory disease due to toluene diisocyanate. Clin Allergy 984; 4: Hudson P, Cartier A, Pineau RT, et al. Follow-up of occupational asthma caused by crab and various agents. J Allergy Ctin Immunol 985; 76: Chan-Yeung M, MacLean L, Paggiaro PL. Follow-up study of 3 patients with occupational asthma caused by western red cedar. J Allergy Clin Immunol 987; 79: Lozewicz S, Assoufi BK, Hawkins R, Newman Taylor A. Outcome of asthma induced by isocyanates. BrJDis Chest 987; 8: 4-.. Mapp CE, Corona PC, de Marzo N, Fabbri L. Persistent asthma due to isocyanates. Am Rev Resp Dis 988; 37: Venables KM, Davison AG, Newman Taylor A. Consequences of occupational asthma. Respir Med 989; 83: Pisati G, Barufnni A, Zedda S. TDI induced asthma: outcome according to persistence or cessation of exposure. Br J Ind Med 993; 5: Gannon PFG,Weir DC, Robertson AS, Burge PS. Health, employment and financial outcomes in workers with occupational asthma. Br J Ind Med 993; 5: Cannon J, Cullinan P, Newman Taylor A. Consequences of occupational asthma. Br Med J 995; 3: Ross DJ, McDonald JC. Health and employment after a diagnosis of occupational asthma: a descriptive study. Occup Med 998; 48: 9-5. Downloaded from by guest on October 8

8 Downloaded from by guest on October 8