Since the early 1950s, it has been well established that smoking is the most important risk factor for lung cancer. 1–3 About 98% of male cases worldwide and 70–90% of European and American female cases report a history of smoking. 4 The population attributable risk percentage for smoking has been estimated to be between 80% and 95% in men 5–7 and about 80% in women. 8 Occupational exposures to carcinogens, indoor radon exposure, dietary habits, industrial air pollution, and environmental tobacco smoke (ETS) are other causes of lung cancer. 9 To examine the separate influence of these risk factors, it is important to control adequately for the impact of smoking. Rough or incomplete recording of the smoking history as well as inadequate consideration of tar contents, inhalation depth, time since smoking cessation, and other smoking behavior patterns have all been discussed as sources of incomplete control of confounding that could be responsible for incorrect risk estimates.
Restriction of the analysis to those cases and controls who never smoked should avoid such confounding effects. But, because of the small proportion of lung cancer cases who have never smoked, such restrictions tend to result in imprecise risk estimates. To avoid the problem of low power, an international multicenter case-control study of lung cancer in nonsmokers was initiated in 1988, coordinated by the International Agency for Research on Cancer (IARC). The main objective of the study was to investigate the association between ETS and lung cancer. Results on this exposure have been published elsewhere. 10,11 Here, we investigate the relation between occupational exposures and lung cancer in nonsmokers.
Subjects and Methods
Twelve study centers from seven European countries participated in the international case-control study of lung cancer in nonsmokers: three each from Germany and Italy; two from Portugal; and one each from Sweden, the United Kingdom, France, and Spain. Case and control subjects were enrolled between 1988 and 1994. The study design varied slightly across the centers. The most important difference was the selection of control subjects. Community-based controls were selected in six centers, hospital-based controls in five centers, and both community and hospital-based controls in one center. Hospital-based controls were selected from diseases not related to tobacco smoking. Because of differences in study-specific age restrictions, the upper age limit for inclusion in the current analysis was 75 years. We restricted the present study to nonsmoking cases and controls, defined as subjects who smoked fewer than 400 cigarettes during their lifetime. A common questionnaire was used in face-to-face interviews to gather information on relevant demographic variables, dietary habits, lifetime ETS exposure, and detailed low-level occasional smoking and occupational histories. More detailed information of this multicenter study may be obtained elsewhere. 10,11
For all cases and controls included in the present analysis, a lifelong occupational history was recorded, including job title and branch of industry, as well as the beginning and the end of each job period. Job titles were coded (masked relative to case/control status) according to the International Standard Classification of Occupations (ISCO). 12 The branch of industry was classified according to the International Standard Industrial Classification of All Economic Activities (ISIC). 13
To evaluate the total impact of occupational exposure on lung cancer among nonsmokers, we used the work of Ahrens and Merletti, 14 which translated occupations and industries that are known (list A) or suspected (list B) to be associated with lung cancer into the corresponding ISCO and/or ISIC codes (see Tables 2 and 3 for details). We then calculated effect estimates for subjects who worked for at least 6 months in a job belonging to one or both lists. To document the hierarchic order between the two lists, we subdivided exposed cases and controls into two groups: subjects exposed to list-A occupations (regardless of whether they were also exposed to list-B occupations) and subjects who were only exposed to list-B occupations. The referent category for these calculations was defined by cases and controls who never worked in a list-A or a list-B occupation. The association between time since first employment in list-A/-B occupations and lung cancer risk was analyzed by using the following cutpoints: nonexposed, >6 months to 29 years, 30–39 years, and 40+ years. The relationship between duration of employment in list-A and/or list-B occupations and lung cancer risk was investigated by means of the following categories: nonexposed, >6 months to 4 years, 5–14 years, and 15+ years. The choice of these cutpoints ensured an approximately equal distribution of exposed subjects in each category (tertile).
Investigation of the effect of specific occupations was generally restricted by small numbers, although we evaluate separately the effects of the three most common list-A occupations in men and the three most common list-B occupations in men and women. The odds ratios (ORs) associated with these specific occupations were computed in the following way: a case or a control was considered as exposed if he or she had worked for at least 6 months in that particular job. The nonexposed group consisted of subjects who had never worked (or had worked for less than 6 months) in that job, whatever their other exposures.
We used unconditional logistic regression to calculate gender-specific ORs and 95% confidence intervals (95% CIs). 15 Besides the exposure variable of interest, the basic regression model included terms for age and center. Further adjustment for occasional smoking (ever smoked occasionally, but fewer than 400 cigarettes) or other possible sources of confounding such as residence in urban or in rural areas, dietary habits, or ETS affected our risk estimates only to a negligible degree, so the ORs presented here do not include these adjustments.
We included 650 nonsmoking cases (96.5% with a histologically confirmed diagnosis) and 1,542 nonsmoking controls up to 75 years of age in the analyses. Table 1 shows the basic characteristics of the study subjects. The majority of study subjects were women, accounting for approximately 78% of cases and 66% of controls. The age distribution was similar among cases and controls; the mean age of male cases was 58 years, compared with 59 years among male controls, whereas in females the mean age was 62 for both cases and controls. In both genders, adenocarcinoma was the most common histologic type of lung cancer (51%). The response rates for the centers varied between 55% and 95%, with the exception of three centers (Germany 2, Germany 3, and Portugal 2), in which the response rate among control subjects was lower than 50%. A more detailed and center-specific description with regard to inclusion criteria, selection of controls, and diseases of hospital controls, as well as response rates of cases and controls for each center are presented in Boffetta et al.11
List A: Specific Occupations and Industries Known to Present an Excess Risk of Lung Cancer
Table 2 shows the number of exposed cases and controls according to occupations and industries known to present an excess risk of lung cancer. The three most common occupations among men in this group were shipyard/dockyard and railroad manufacture workers (combined OR = 1.30; 95% CI = 0.47–3.60); painters (OR = 1.84; 95% CI = 0.59–5.74); and workers in nonferrous metal basic industries, that is, smelting, alloying, refining, or rolling (OR = 1.68; 95% CI = 0.39–7.19). The majority of female cases and controls classified in list A also worked in these occupations, but the numbers involved were too small to produce reliable estimates of effect.
List B: Specific Occupations and Industries Suspected to Present an Excess Risk of Lung Cancer
Table 3 shows the number of cases and controls according to occupations and industries that are suspected to present an excess risk of lung cancer. The three most common occupations among men were carpenters and joiners (OR = 0.67; 95% CI = 0.29–1.54); bus/truck drivers or railroad workers, etc (OR = 1.29; 95% CI = 0.60–2.77); and mechanics and welders in motor vehicle manufacturing and repair (OR = 0.67; 95% CI = 0.22–2.07). Among females, other occupations and industries prevailed; most of the cases and controls worked as laundry workers and dry cleaners (OR = 1.83; 95% CI = 0.98–3.40), followed by various occupations in rubber manufacture (OR = 2.86; 95% CI = 1.00–8.20) and by ceramic, pottery, and glass workers (OR = 1.88; 95% CI = 0.59–6.05).
Ever-Exposure to Known or Suspected Risk Occupations
Among nonsmoking men, a total of 40 cases (28.4%) and 165 controls (31.1%) had ever worked in a job for which a carcinogenic risk has been demonstrated or suspected, that is, belonging to list A or list B. For these subjects, we calculated an OR of 1.20 (95% CI = 0.76–1.92); see Table 4. Men in list-A occupations had an OR of 1.52 (95% CI = 0.78–2.97), but there was little evidence of an elevated risk for list-B occupations (OR = 1.05; 95% CI = 0.60–1.83). No dose-response relation was found between lung cancer risk and time since first employment and duration of employment in list-A/-B occupations.
Among nonsmoking women, 46 cases (9.0%) and 69 controls (6.8%) had ever worked in a job that belongs to either list (OR = 1.67; 95% CI = 1.10–2.52); see Table 4. Only 1% of cases and controls were exposed to list-A jobs (OR = 1.50; 95% CI = 0.49–4.53), and the majority of exposed women worked in list-B occupations (OR = 1.69; 95% CI = 1.09–2.63). An inverse association was found with regard to time since first employment in list-B occupations; the highest risk estimate was calculated for cases and controls with less than 30 years since first employment (19 cases, 16 controls; OR = 2.92; 95% CI = 1.42–1.41). A moderately elevated risk was found for those subjects first employed 30–39 years earlier (12 cases, 18 controls; OR = 1.41; 95% CI = 0.65–3.05). Forty years and more since first employment was associated with only a slightly elevated risk (10 cases, 25 controls; OR = 1.11; 95% CI = 0.51–2.43). With regard to duration of exposure to list-B occupations, we calculated approximately twofold elevated risk estimates for the longest duration category (15+ years; OR = 1.99; 95% CI = 0.88–4.50), as well as for the middle one (5–14 years; OR = 2.18; 95% CI = 1.08–4.43). Women with less than 5 years of employment had an only slightly elevated risk of lung cancer (OR = 1.16; 95% CI = 0.56–2.41).
In the current analysis, we used two lists of occupations and industries that are known (list A) or suspected (list B) to be associated with lung cancer, established on the basis of previous epidemiologic studies and first published in 1983. 16 Since then, these lists have been used for risk estimation in several studies on occupational cancer. 17–19 In 1995 both lists were updated 20 in light of recent evaluations of occupational exposures by the IARC. With regard to occupations and industries known and suspected to be associated specifically with lung cancer, both lists were translated into codes of the ISCO and the ISIC by Ahrens and Merletti. 14
Some limitations of this study should be considered in evaluating our results. One methodologic concern is that of misclassification of nonsmoking status. In our study, 164 cases and 438 controls reported ever-consumption of fewer than 400 cigarettes (“occasional smokers”). Misclassification of smoking status is more likely to be present among such very light smokers than among nonsmokers. Exclusion of these occasional smokers had only minor consequences on the results; the ORs for ever working in a list-A occupation were 1.61 (95% CI = 0.70–3.72) in men and 1.48 (95% CI = 0.43–5.11) in women; for ever working in a list-B occupation (but not in a list-A occupation), the corresponding ORs were 1.35 (95% CI = 0.67–2.74) in men and 1.63 (95% CI = 0.98–2.70) in women.
We also controlled the main analyses of this investigation for spousal and work-place ETS but did not find any appreciable difference in effect.
Another possible source of bias was the different response rates between the centers, particularly with regard to low response rates among controls being observed in Germany 2, Germany 3, and Portugal (38–47%). An analysis conducted separately after excluding these three centers leads to results nearly identical to those presented in Table 4.
Moreover, we checked the influence of the different criteria for selection of control subjects between the centers. In the present study, four of the centers recruited hospital controls and seven of the centers recruited population controls (control subjects were both hospital and community based in the center from the United Kingdom). Because cases (as well as hospital controls) are aware of their disease status, community-based controls are often discussed as a potential source for recall bias resulting in a differential misclassification of exposure. We addressed this issue by comparing the results from subsets of centers defined according to their criteria for selection of control subjects and found only small differences, that is, there was no evidence that elevated risks were restricted to the centers with population controls.
The major strength of the current study is its size; as far as we know, it is the largest investigation of occupational risk factors among nonsmoking lung cancer cases and controls to date. We relied on only self-reported employment data; no data were obtained from next of kin or other surrogates. The recall of job title and industry in retrospective case-control studies is known to be a highly valid source of information 21,22 and provides a solid basis for a broad classification of subjects as exposed or nonexposed to occupational risk factors. Thus, definitions of high-risk occupations based on job titles and industry codes generally provide a useful approach to the analysis of occupational causes of cancer in epidemiologic studies. 14 High-risk occupations are in some cases too specific, however, and the closest possible fit by means of ISCO and ISIC classifications is too broad (for example, Agriculture/Insecticide application). As a result, a small number of occupations were not included in the lists. It is therefore important to bear in mind that the method we used here to identify risks due to occupational exposures has been optimized more for its specificity than for its sensitivity, by excluding an occupation or industry if the closest possible fit of codes in the ISIC or ISCO classifications was considered to be too broad.
Approximately 10% of all nonsmoking male cases and controls in this study were ever engaged in a list-A occupation. The majority of them worked as painters or railroad manufacture workers, or as workers in shipbuilding, nonferrous basic industries, or mining and quarrying. For each of these individual occupations, there were elevated effect estimates confirming the known associations with lung cancer. Despite the large original database of 12 case-control studies, the restriction to nonsmokers led to a relatively small number of male cases and imprecise effect estimates. The total number of females was higher, but the number of exposed subjects was too small (<1%) to detect precise estimates of effect.
Employment in a job suspected to be associated with excess risk of lung cancer (list B) was more common. Nevertheless, among men, the overall effect estimate for list-B occupations was only slightly elevated. Sizeable numbers of exposed cases and controls were transport workers (bus/truck drivers, etc), carpenters and joiners, and mechanics and welders in motor vehicle manufacturing and repair. Among females, we calculated an elevated OR in those who were ever employed in a list-B occupation. Most had worked in the rubber industry or as laundry and dry cleaners. The effect estimates for the latter occupations were 2.86 (95% CI = 1.00–8.20) and 1.83 (95% CI = 0.98–3.40).
A number of these elevated effect estimates in men and women can be at least partly attributed to occupational exposure to asbestos (for example, shipbuilding, metal production, or occupations in rubber manufacture). Other possible exposures responsible for causing lung cancer are welding fumes 23 or diesel emissions 24 that could contribute to the elevated ORs for occupations in shipbuilding and transport. A detailed review of epidemiologic studies with regard to occupational exposures, especially among nonsmokers, was recently published by Brownson et al.25
In the present study, perhaps the most interesting finding is the elevated risk among female laundry workers and dry cleaners. The most likely cause for an excess risk for this kind of work is thought to be exposure to perchlorethylene and other dry cleaning solvents. An increased risk of malignant neoplasms (primarily from an excess of lung and cervical cancer) among laundry and dry cleaning workers has been previously described. 26 Since then, several other studies have found elevated risks, especially for lung cancer, 27–29 but also for other sites (bladder, liver, esophageal, and renal cell cancer). An almost twofold risk among female laundry workers and dry cleaners was also reported in a recently published German case-control study 30 (OR = 1.9; 95% CI = 0.91–3.93) after adjustment for smoking (the nonsmoking subjects are also part of the present study); and a similar result was reported by Brownson et al31 in a U.S. study of nonsmoking women [OR = 2.1, (95% CI = 1.2–3.7) among females who were ever employed in the dry cleaning business]. A detailed summary of epidemiologic studies, especially in dry cleaning workers, may be found in an IARC monograph. 32
Within the group of subjects exposed to either list-A or -B occupations, our study demonstrates that women are more likely than men to be exposed to occupations that are suspected, rather than confirmed, as being associated with increased risk of developing lung cancer. Knowledge on occupational lung carcinogens, and presumably on occupational carcinogens in general, however, is mostly related to agents to which mainly men are exposed. Thus, little is known about carcinogenic occupational exposures among women.
This study was partially supported by the following grants: European Commission DG-XII (Contract EV5V-CT9:4-0555) for the coordination; in France, Association pour la Recherche sur le Cancer, European Commission (90CVV01018), and Caisse Nationale d’Assurance Maladie des Travailleurs Sociaux; in Germany 1, the Federal Ministry for Education, Science, Research and Technology (Grant 01 HK 546) and the Federal Ministry of Work and Social Affairs (Grant IIIb 7-27/13); in Germany 2, the Federal Office of Radiation Protection, Salzgitter (Grant St Sch 1066, 4047, 4074/1); in Germany 3, the Federal Office of Radiation Protection, Salzgitter (Grant No. St Sch 4006, 4112); in Italy 1, Italian Ministry of University and Scientific amd Technological Research (MURST), the Italian Association for Cancer Research, and the Regione Piemonte-Ricerca Finalizzata; in Italy 2, the National Research Council (Contract 91.00327.CT04) and the Italian Association for Cancer Research; in Portugal 1, Junta Nacional de Investigação Científica e Tecnológica (Contract PMCT/C/SAU/815.90); in Portugal 2, Comissãp de Fomento de Investigação em Cuidados de Saúde; in Spain, the Spanish Ministry of Health (Ref. No. 89002300); in Sweden, the Swedish Match (8913/9004/9109/9217) and the Swedish Environmental Protection Agency (53:30071-1); and in the United Kingdom, the Imperial Cancer Research Fund, the Department of Health, the Department of the Environment, and the European Commission.
We thank the following persons for their contributions to the study in the different centers. France (Paris): C. Schrameck, J. Azorin, P. Baldeyrou, J. P. Battesti, J. Benoit, J. Bignon, F. Blanchon, J. J. Bonerandi, C. Brambilla, J. L. Breau, J. M. Brechot, J. Briere, J. M. Brisset, C. Buffet, G. Cathelineau, Y. Chapuis, J. P. Chevrel, J. P. Droz, L. Dubertret, F. Dubois, F. Fekete, P. Girard, C. Gisselbrecht, D. Grunenwald, A. Hirsch, V. Izrael, A. Jardin, J. P. Kleisbauer, J. A. Krivitzky, A. Laugier, T. Le Chevallier, F. Mal, F. Mazas, C. Menkes, F. Mignon, B. Milleron, E. Modigliani, R. Modigliani, P. Morel, J. F. Muir, J. Y. Nordin, J. Paolaggi, J. B. Paolaggi, R. Pariente, S. Pretet, A. Puissant, J. Roussy, J. C. Sarles, B. Sastre, J. Sauvaget, G. Schaison, P. Testas, J. Tredaniel, and D. Valeyre. Germany 1: W. Hartmann and K. Eberhardt in Bremen and K.-M. Müller in Bochum. Germany 2: G. Dingerkus in Wuppertal, M. Gerken in Donaustauf, and K.-M. Müller in Bochum. Germany 3: J. Heinrich and G. Wölke in Erfurt. Italy 1 (Turin): S. Massacesi, G. Peyrano, M. Tedeschi, and M. Artom. Italy 3 (Rome): F. Anatra and T. Trequattrini. Portugal 1 (Lisbon): J. Maçanita, E. Teixeira, R. Sotto-Mayor, M. Cancela de Abreu, H. Lucas, M. J. Melo, and M. M. Cristóvão. Portugal 2: R. Nêveda in Viana do Castelo, J. M. Calheiros in Porto, and F. F. Rodrigues in Vila Nova de Gaia. Spain: C. Pallarès, X. Fabregat, J. Roselló, J. Estapé, J. Planas, N. Malats, and I. Machengs in Barcelona; A. Barnadas and M. G. Esteve in Badalona; I. Martínez-Ballarin in L’Hospitalet de Llobregat; and A. Badia in Mataró. Sweden (Stockholm): V. Agrenius, K. Svartengren, and C. Svensson. United Kingdom (Oxford): R. Doll, P. Silcocks, and B. Thakrar.
1. Levin ML, Goldstein H, Gerhardt PR. Cancer and tobacco smoking: a preliminary report. J Am Med Assoc 1950; 143: 336–338.
2. Wynder EL, Graham EA. Tobacco smoking as a possible etiological factor in bronchiogenic carcinoma. J Am Med Assoc 1950; 143: 329–336.
3. Doll R, Hill AB. A study of the aetiology of carcinoma of the lung. Br Med J 1952; 2: 1271–1286.
4. Koo LC, Ho JHC. Worldwide Epidemiological Patterns of Lung Cancer in Nonsmokers
. Int J Epidemiol 1990; 19 (suppl 1): S14–S23.
5. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. Oxford: Oxford University Press, 1981.
6. Parkin DM, Pisani P, Lopez AD, Masuyer E. At least one in seven cases of cancer is caused by smoking: global estimates for 1985. Int J Cancer 1994; 59: 494–504.
7. Siemiatycki J, Krewski D, Franco E, Kaiserman M. Associations between cigarette smoking and each of 21 types of cancer: a multi-site case-control study
. Int J Epidemiol 1995; 24: 504–514.
8. Ernster VL. Female lung cancer. Annu Rev Public Health 1996; 17: 97–114.
9. Samet JM, ed. Epidemiology of Lung Cancer. New York: Marcel Dekker, 1994.
10. Boffetta P, Agudo A, Ahrens W, Benhamou E, Benhamou S, Darby SC, Ferro G, Fortes C, Gonzales CA, Jöckel KH, Krauss M, Kreienbrock L, Kreuzer M, Mendes A, Merletti F, Nyberg F, Pershagen G, Pohlabeln H, Riboli E, Schmid G, Simonato L, Trédaniel J, Whitley E, Wichmann HE, Winck C, Zambon P, Saracci R. Multicenter case-control study
of exposure to environmental tobacco smoke and lung cancer in Europe. J Natl Cancer Inst 1998; 90: 1440–1450.
11. Boffetta P, Agudo A, Ahrens W, Benhamou E, Benhamou S, Darby SC, Ferro G, Fortes C, Gonzales CA, Jöckel KH, Krauss M, Kreienbrock L, Kreuzer M, Mendes A, Merletti F, Nyberg F, Pershagen G, Pohlabeln H, Riboli E, Schmid G, Simonato L, Trédaniel J, Whitley E, Wichmann HE, Winck C, Zambon P, Saracci R. European Multicenter Case-Control Study
of Lung Cancer in Nonsmokers
: Detailed Results on Exposure to Environmental Tobacco Smoke. International Agency for Research on Cancer Technical Report No. 33. Lyon: International Agency for Research on Cancer, 1998.
12. International Labour Office. International Standard Classifications of Occupations, 2nd ed. Geneva: International Labour Office Publications, 1968.
13. United Nations Publications. International Standard Industrial Classification of All Economic Activities (ISIC). New York: Publishing Service United Nations, 1971.
14. Ahrens W, Merletti F. A standard tool for the analysis of occupational lung cancer in epidemiologic studies. Int J Occup Environ Health 1998; 4: 236–240.
15. SAS Institute Inc. SAS/STAT®
User’s Guide, Version 6. 4th ed. vol. 2. Cary, NC: SAS Institute, 1989;1071–1126.
16. Simonato L, Saracci R. Cancer, occupational. In: Parmeggiani L, ed. Encyclopedia of Occupational Health and Safety, International Labour Office, Geneva 1983;369–375.
17. Simonato L, Vineis P, Fletcher AC. Estimates of the proportion of lung cancer attributable to occupational exposure. Carcinogenesis 1988; 9: 1159–1165.
18. Vineis P, Simonato L. Proportion of lung and bladder cancers in males resulting from occupation
: a systematic approach. Arch Environ Health 1991; 46: 6–15.
19. Ahrens W, Jöckel KH, Patzak W, Elsner G. Alcohol, smoking, and occupational factors in cancer of the larynx: a case-control study
. Am J Ind Med 1991; 20: 477–493.
20. Boffetta P, Kogevinas M, Simonato L, Wilbourn J, Saracci R. Current perspectives on occupational cancer risks. Int J Occup Environ Health 1995; 1: 315–325.
21. McGuire V, Nelson LM, Koepsell TD, Checkoway H, Longstreth WT. Assessment of occupational exposures in community based case-control studies. Annu Rev Public Health 1998; 19: 35–53.
22. Ahrens W. Retrospective assessment of occupational exposure in case-control studies. In: Wichmann HE, Jöckel KH, Robra BP, eds. Fortschritte in der Epidemiologie. Landsberg: Ecomed-Verlagsgesellschaft, 1999.
23. Jöckel KH, Ahrens W, Pohlabeln H, Bolm-Audorff U, Müller KM. Lung cancer risk from welding: results from a case-control study
in Germany. Am J Ind Med 1998; 33: 313–320.
24. Brüske-Hohlfeld I, Möhner M, Ahrens W, Pohlabeln H, Heinrich J, Kreuzer M, Jöckel KH, Wichmann HE. Lung cancer risk in male workers occupationally exposed to diesel motor emissions in Germany. Am J Ind Med 1999; 36: 405–414.
25. Brownson RC, Alavanja MCR, Caporaso N, Simoes EJ, Chang JC. Epidemiology and prevention of lung cancer in nonsmokers
. Epidemiol Rev 1999; 20: 218–236.
26. Blair A, Decoufle P, Grauman D. Causes of death among laundry and dry cleaning workers. Am J Public Health 1979; 69: 508–511.
27. Blair A, Stewart PA, Tolbert PE, Grauman D, Moran FX, Vaught J, Rayner J. Cancer and other causes of death among a cohort of dry cleaners. Br J Ind Med 1990; 47: 162–168.
28. Ruder AM, Ward EM, Brown DP. Cancer mortality in female and male dry cleaning workers. J Occup Med 1994; 36: 867–874.
29. Walker JT, Burnett CA, Lalich NR, Sestito JP, Halperin WE. Cancer mortality among laundry and dry cleaning workers. Am J Ind Med 1997; 32: 614–619.
30. Jahn I, Ahrens W, Brüske-Hohlfeld I, Kreuzer M, Möhner M, Pohlabeln H, Wichmann HE, Jöckel KH. Occupational risk factors for lung cancer in women: results of a case-control study
in Germany. Am J Ind Med 1999; 36: 90–100.
31. Brownson RC, Alavanja MCR, Chang JC. Occupational risk factors for lung cancer among nonsmoking women: a case-control study
in Missouri (United States). Cancer Causes Control 1993; 4: 449–454.
32. International Agency for Research on Cancer. IARC Monographs on the Evaluation of the Carcinogenic Risk to Humans. vol. 63. Dry Cleaning, Some Chlorinated Solvents and Other Industrial Chemicals. Lyon: International Agency for Research on Cancer, 1995.