Banga, Amit MD; Reilly, Mary Jo MS; Rosenman, Kenneth D. MD
The process of joining pieces of metal at a point that has been made soft by heating is referred to as welding. Heating of the metal surfaces leads to the release of different types of fumes and particulates, which depend on the type of metal being welded, the welding electrode being used, and the welding technique. Welding is a common profession, with more than 400,000 people employed in the United States.1 More than 15,000 welders are employed in Michigan.2
Adverse health effects, both respiratory and nonrespiratory, are known to be associated with welding. In addition to causing acute and chronic upper and lower respiratory tract illnesses, welding may increase the risk of neurological symptoms, chronic renal failure, and malignancy.3
Work-related asthma (WRA) includes both occupational asthma (OA, new-onset asthma related to sensitizers or irritant work exposures) and work-exacerbated asthma (WEA, preexisting asthma that worsens on work exposures) and causes significant acute morbidity, long-term disability, as well as adverse social and economic impacts.4 It is estimated that 15% of adult asthmatic patients have OA,5 with the percentage increasing when WEA is included.6 The key management strategy of patients with WRA entails avoidance of occupational exposures since ongoing exposure leads to disease progression.
There have been several case reports of WRA in workers who have been welding on commonly used metals.7–10 Recent population-based studies have also shown that welding nearly doubles the risk of developing asthma.11,12 Earlier studies from both the United States13 and abroad14,15 have described the spectrum of respiratory symptoms, including asthma, among patients engaged in welding. Despite the well-documented risk of WRA among welders, there is a relative lack of awareness of the risks and the burden of illness and morbidity associated with welding. In the current study, we describe the characteristics of workers exposed to welding fumes in the State of Michigan who have been reported as WRA cases to the State-based surveillance system over the last two decades.
The State of Michigan instituted a surveillance program for WRA in 1988 with financial assistance from the National Institute for Occupational Safety and Health. The surveillance program, called the Sentinel Event Notification System for Occupational Risks, is a joint project of the Michigan Occupational Safety and Health Administration, the Michigan Department of Community Health, and Michigan State University. The goal of the surveillance program is the prevention of WRA. The report of an index patient is regarded as a sentinel health event to initiate a workplace investigation consisting of methods that may identify exposures that put other employees at risk or identify other employees who already have WRA.
The Division of Occupational and Environmental Medicine at the Michigan State University administers the WRA surveillance program for the State of Michigan. Reports of WRA are received from physicians (private practice and company physicians), hospitals, emergency departments, and the poison-control center. In addition, all claims for respiratory disease filed with the Michigan Workers’ Compensation Agency are reviewed.
Once these reports are received by the division, they are followed up with a telephonic interview with the subject by using a standardized questionnaire regarding the demographic profile, type of profession, smoking status, personal and family history of allergies, respiratory symptoms, and relation to work, as well as the diagnostic workup for asthma.16 Copies of pulmonary function test results are also obtained. Standard criteria are followed for the diagnosis of reversibility and bronchoprovocation testing.17 Workplace evaluation of the industries from where reports are received is conducted by the industrial hygienist. Results of the completed questionnaire and the medical records, including pulmonary function tests and workplace conditions, are reviewed by a physician board certified in both occupational and environmental medicine and internal medicine. Cases are then classified as either OA with known sensitizer, OA with unknown sensitizer, WEA, or reactive airways dysfunction syndrome (RADS).18 We used the Association of Occupational and Environmental Clinics exposure system to code agents to which individuals were exposed (http://www.aoec.org/aoeccode.htm).19
All cases reported as WRA during the years 1990 to 2007 were eligible for inclusion. We identified subjects who were exposed to welding fumes when their respiratory symptoms began. Among these, workers in whom the exposure to welding fumes was confirmed as responsible for WRA during the period 1990 to 2007 by the review of the questionnaire, workplace inspection, and medical records were included in the study. Workers could have been identified as welders or working around welding.
The study was approved by the institutional review board.
Analysis was done by using the statistical software SPSS version 11.0 (SPSS Inc, Chicago, IL). Continuous variables were described by using median with range and categorical variables by using proportions. For determining the predictors of abnormal lung functions, forced expiratory volume in one second to forced vital capacity ratio (FEV1/FVC ratio < 0.7) was taken as the dependent variable. Categorical variables (the Fisher exact test) and continuous variable (the Mann-Whitney U test) were initially compared in a univariate manner. Variables significant at P < 0.1 along with demographic parameters, personal and family history of allergy, and smoking were entered as covariates in the multivariate logistic regression analysis. Multiple equations were constructed to assess the independence of association between variables. Similar analytic technique was followed for the comparison of the profile of workers, with primary job description as welders, against others to determine risk factors for the development of WRA among “non-welders.” Statistical significance was considered at P < 0.05 (only two tailed).
A total of 2607 confirmed cases of WRA were identified in Michigan over the period 1990 to 2007. One hundred forty-two (5.4%) were the cases of welding-related WRA (age, 43.3 ± 11.4 years). Welding was the fifth leading cause of WRA. Frequency of reports per year largely remained unchanged (mean, 7.9 per year, Fig. 1). Majority workers were classified as OA with unknown sensitizer (n = 105, 73.9%), followed by WEA (n = 18, 12.7%), OA with known sensitizer (n = 13, 9.3%), and RADS (n = 6, 4.2%). All the subjects classified as OA with known sensitizer had exposure to stainless-steel welding.
Profile of the Study Group
Most subjects (n = 129, 90.8%) were interviewed within 1 year of being reported. Table 1 provides the details regarding the demographic and background data. Eighty-seven subjects (61.3%) had a history of smoking cigarettes, but only 20 (14.1%) were currently smoking. Only half reported using personal protective equipment. Thirty-two subjects (22.5%) had a prior history of asthma. Nevertheless, only 18 (60%) of 32 were classified as WEA since the rest had no respiratory symptoms or asthma medication use for at least 2 years before starting the job.
Among the asthma-related symptoms, shortness of breath was the most common (n = 132), followed by wheezing (n = 114) (Table 2). Wheezing was the most long-standing symptom. All subjects had at least two of the four symptoms and a majority had all four (n = 77). Commonly, symptoms worsened during work shifts (n = 104, 73.2%) and through the week (50%). Symptoms remitted away from work in 109 workers (76.7%). Twenty-one subjects (14.8%) reported a particular intense exposure after which their respiratory problems began.
Ninety-five (66.9%) subjects reported a doctor's diagnosis of asthma since they started working in the job. Only 58 (61.1%) of these subjects reported that a doctor told them that their asthma was work related.
Subjects had been in the company where respiratory symptoms began for a median period of 10 years (<1 to 47 years) and worked for 7 years (<1 to 42 years) before development of symptoms.
The industry and the primary occupation of the subjects are presented in Figs. 2 and 3. More than half (n = 72, 50.7%) were nonwelders but worked around welding fumes. On univariate analysis, “non-welders” developing WRA were more likely to be women and had higher number of years on the job. There was no difference in the frequency of smoking, personal or family history of allergy, or asthma between the two groups. On multivariate analysis, being a woman conferred an independent risk for the development of WRA among “non-welders” (adjusted odds ratio, 2.47; 95% confidence interval, 2.1 to 3.31; P = 0.02). Among the male workers, a fewer number of years on the job was independently associated with WRA in “welders” (adjusted odds ratio, 0.95; 95% confidence interval, 0.83 to 0.98; P = 0.02).
Almost half the respondents (n = 69, 43.7%) were aware of coworkers with similar symptoms. Forty-eight respondents (33.1%) were still exposed to the substance causing breathing problems, and all reported of having persistent symptoms. Figure 4 shows the status of the workers removed from the exposure.
Almost all workers had sought medical treatment since starting the job (136, 95.8%) and needed medications (133, 93.7%). Many had emergency room (ER) visits (n = 86, 60.6%) and required hospitalization (n = 50, 36.7%) for breathing problems. Median time between starting the job and the first ER visit (<1 to 38 years) as well as hospitalization (<1 to 41 years) was 2 years.
Spirometry had been performed for 106 subjects (74.6%). Pre-/postbronchodilator and methacholine challenge tests were done among 54 (50.1%) and 24 (22.6%) workers, respectively, and 65 (45.8%) subjects had either of the tests (Table 3). On univariate analysis, older age, longer time in the company and on the job, and having wheezing were associated with low FEV1/FVC ratio (<0.7). Nevertheless, on multivariate analysis, none of the covariates emerged as an independent predictor. Associations between spiro-metry abnormality and job and exposure and symptom durations were all lost after adjusting for age. Adjustment for smoking status did not affect the association.
The current study looked at the Michigan workers reported as WRA secondary to welding fume exposure. Although, the majo-rity of workers were from typical welding utilizing industries, 12% were from other industries, including rubber, electronic, and stone, where welding exposure might not be typically suspected (Fig. 2). Similarly, workers represented a wide array of job designations and only half identified themselves as welders (Fig. 3). The nonwelders were assisting the welders or were machine operators or engaged in ancillary processes such as assembling in area where welding was being done. In a previous study on WRA secondary to welding,20 24 of 34 workers were welders, with others being machine fitters or plumbers, etc. These findings highlight the importance of taking an occupational history that is more detailed than industry and job title alone. Welding may be the cause of WRA among nonwelders if they work in an environment where welding is done.
Literature on the frequency of WRA associated with welding exposure has been variable. This, to some extent, is due to markedly variable study designs and use of different surveillance protocols.12, 14, 21–23 Furthermore, the type of welding may also affect the likelihood of developing WRA.24 In a large population-based study,12 welding was shown to increase the risk of asthma by more than 30%. In another study, welding was the most common occupation associated with WRA and welders consisted of 9% of all workers reported as WRA.14 Surveillance data from the SWORD program in England had also identified welding as a common cause of WRA.21 A study from France found welding as the 9th leading occupational cause of WRA.22 The current study found welding to be among the top-five exposures responsible for WRA. Nevertheless, data from South Africa has been divergent, and welding was not identified as a cause among 195 cases of WRA.23 Welding was, however, implicated as the most common cause of RADS.23
Existing literature on the latency between exposure and development of WRA is variable. The risk of WRA, in general, is the highest soon after initial exposure, but the interval may vary from months to years.25 This is especially true with WRA related to welding in which longer latent periods have been reported.7,9,20 Consistent with this, duration between exposure and development of symptoms in the current study was generally long. Moreover, male workers who identified themselves as “non-welders” took longer to become symptomatic. These workers presumably had less-frequent and relatively milder exposure, which could explain the longer latent period. These findings are suggestive of repeated irritation as a biological mechanism for welding-related WRA.
Subjects had long duration of respiratory symptoms before being reported as WRA. Furthermore, nearly one third were not aware that they were diagnosed with asthma, and even less were informed that their symptoms could be work related, even though this information was in their medical record. Finally, a third of the workers were still working in the same work environment with ongoing symptoms. These workers suffered significant morbidity, often with multiple hospitalizations (n = 17). The workup of these subjects was often not comprehensive. Basic spirometry data were not available for nearly a quarter of the patients, with only a minority undergoing evaluation for hyperreactivity. Lack of appropriate workup for suspected WRA has been reported earlier as well.26 Whereas some of these findings may be related to a delay or underreporting of symptoms to health care providers, there appears to be a general lack of awareness among the average health care provider regarding welding as a cause of WRA and the critical role of avoidance of ongoing exposure. More aggressive efforts by clinicians to assess all adults with asthma for occupational exposures, including welding, need to be recommended.
An association was found between the duration of working in the company, on the current job, and some of the respiratory symptoms with abnormal lung functions, although these were lost after adjustment for age. None of the other variables, including smoking, had an effect on the association. These findings highlight the importance of age as a determinant of lung functions among welders. Age should be considered as an important criterion in the risk stratification of patients exposed to welding, and it may be worthwhile to monitor elderly workers more closely.
The current study has some limitations. The reporting physician made the diagnosis of WRA, and there remains a possibility that some of the subjects could have other workplace exposures that caused their WRA. Nevertheless, final determinations were made on the basis of a follow-up interview, workplace evaluation by industrial hygienist, and review of the collected data by a board-certified physician. Moreover, diagnosis of WRA and relation to welding were on the basis of earlier-made determinations and were not modified for the purpose of study. No specific antigen challenge testing was performed. Spirometry evaluations are often inadequate but represent the practice of medicine in the United States. Given this, there is a possibility of overdiagnosis. Nevertheless, on the contrary, underrecognition and underreporting of WRA, in general, continue to be a problem. We have earlier shown that the surveillance program in Michigan markedly underestimates the incidence of WRA.27
In conclusion, WRA secondary to welding occurs in workers from different industries while working in diverse jobs. Welding fumes are among the top-five workplace exposures associated with development of WRA among Michigan workers. A large majo-rity of subjects have chronic and long-standing symptoms. Use of personal protective equipment remains low and many workers remain in the same environment where symptoms began. Spirometry changes commonly occur, although several patients do not undergo appropriate evaluation. As reflected by frequent ER visits and hospitalizations, WRA among welders is associated with significant morbidity and high cost. Despite the extensive literature on WRA and its association with welding, there appears to be a lack of awareness among health care providers. Future programs aimed at increasing the awareness and compliance with reporting requirements of WRA and improving the working conditions where welding occurs should be instituted.
This study was funded by National Institute for Occupational Safety and Health (grant 5U60 OH008466).
1. US Bureau of Labor Statistics. Information on welding, soldering, and brazing workers. Occupational Outlook Handbook, 2010–11 Edition. Washington, DC: US Bureau of Labor Statistics. http://www.bls.gov/oco/ocos226.htm
. Accessed August 9, 2000.
3. Antonini JM. Health effects of welding. Crit Rev Toxicol. 2003;33:61–103.
4. Vandenplas O, Toren K, Blanc PD. Health and socioeconomic impact of work-related asthma. Eur Respir J. 2003;22:689–697.
5. Sama SR, Milton DK, Hunt PR, Houseman EA, Henneberger PK, Rosiello RA. Case-by-case assessment of adult-onset asthma attributable to occupational exposures among members of a health maintenance organization. J Occup Environ Med. 2006;48:400–407.
6. American Thoracic Society. Occupational contribution to the burden of airway disease. Am J Respir Crit Care Med. 2003;167:787–797.
7. Keskinen H, Kalliomäki PL, Alanko K. Occupational asthma due to stainless steel welding fumes. Clin Allergy. 1980;10:151–159.
8. Contreras GR, Chan-Yeung M. Bronchial reactions to exposure to welding fumes. Occup Environ Med. 1997;54:836–839.
9. Hannu T, Piipari R, Kasurinen H, Keskinen H, Tuppurainen M, Tuomi T. Occupational asthma due to manual metal-arc/welding of special stainless steels. Eur Respir J. 2005;26:736–739.
10. Muñoz X, Cruz MJ, Freixa A, Guardino X, Morell F. Occupational asthma caused by metal arc welding of iron. Respiration. 2009;78:455–459.
11. Karjalainen A, Kurppa K, Martikainen R, Karjalainen J, Klaukka T. Exploration of asthma risk by occupation—extended analysis of an incidence study of the Finnish population. Scand J Work Environ Health. 2002;28:49–57.
12. Kogevinas M, Antó JM, Sunyer J, Tobias A, Kromhout H, Burney P. Occupational asthma in Europe and other industrialised areas: a population-based study. European Community Respiratory Health Survey Study Group. Lancet. 1999;353:1750–1754.
13. Hammond SK, Gold E, Baker R, et al.. Respiratory health effects related to occupational spray painting and welding. J Occup Environ Med. 2005;47:728–739.
14. 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.
15. Lillienberg L, Zock JP, Kromhout H, et al.. A population-based study on welding exposures at work and respiratory symptoms. Ann Occup Hyg. 2008;52:107–115.
16. Rosenman KD, Reilly MJ, Schill DP, et al.. Cleaning products and work-related asthma. J Occup Environ Med. 2003;45:556–563.
17. Standardization of spirometry 1994 update American Thoracic Society. Am J Respir Crit Care Med. 1995;152:1107–1136.
18. Chan-Yeung M, Malo JL. Occupational asthma. N Engl J Med. 1995;333:107–112.
19. Hunting KL, McDonald SM. Development of a hierarchial exposure coding system for clinic-based surveillance of occupational disease and injury. Appl Occup Environ Hyg. 1995;10:317–322.
20. Hannu T, Piipari R, Tuppurainen M, Nordman H, Tuomi T. Occupational asthma caused by stainless steel welding fumes: a clinical study. Eur Respir J. 2007;29:85–90.
21. Meredith S. Reported incidence of occupational asthma in the United Kingdom, 1989–90. J Epidemiol Community Health. 1993;47:459–463.
22. Ameille J, Pauli G, Calastreng-Crinquand A, et al.. Reported incidence of occupational asthma in France, 1996–99: the ONAP programme. Occup Environ Med. 2003;60:136–141.
23. Hnizdo E, Esterhuizen TM, Rees D, Lalloo UG. Occupational asthma as identified by the Surveillance of Work-related and Occupational Respiratory Diseases programme in South Africa. Clin Exp Allergy. 2001;31:32–39.
24. Antonini JM, Lewis AB, Roberts JR, Whaley DA. Pulmonary effects of welding fumes: review of worker and experimental animal studies. Am J Ind Med. 2003;43:350–360.
25. Venables KM, Chan-Yeung M. Occupational asthma. Lancet. 1997;349:1465–1469.
26. Hendrick DJ. Recognition and surveillance of occupational asthma: a preventable illness with missed opportunities. Br Med Bull. 2010;95:175–192.
27. Rosenman KD, Reilly MJ, Kalinowski DJ. A state-based surveillance system for work-related asthma. J Occup Environ Med. 1997;39:415–425.